IN THE CIRCUIT COURT FOR
Janet Watkins
PDF
AMDAC-20190807-Transcript 1
1
FOOD AND DRUG ADMINISTRATION
2
CENTER FOR DRUG EVALUATION AND RESEARCH
3
4
5
6
ANTIMICROBIAL DRUGS ADVISORY COMMITTEE MEETING
7
(AMDAC)
8
9
10
Wednesday, August 7, 2019
11
8:30 a.m. to 4:54 p.m.
12
13
14
15
16
FDA White Oak Campus
17
White Oak Conference Center
18
Building 31, The Great Room
19
10903 New Hampshire Avenue
20
Silver Spring, Maryland
21
22
A Matter of Record (301) 890-4188
2
1
Meeting Roster
2
DESIGNATED FEDERAL OFFICER (Non-Voting)
3
Lauren Tesh Hotaki, PharmD, BCPS, BCIDP
4
Division of Advisory Committee and Consultant
5
Management
6
Office of Executive Programs, CDER, FDA
7
8
ANTIMICROBIAL DRUGS ADVISORY COMMITTEE MEMBERS
9
(Voting)
10
Lindsey R. Baden, MD
11
(Chairperson)
12
Director of Clinical Research
13
Division of Infectious Diseases
14
Brigham and Women's Hospital
15
Director, Infectious Disease Service
16
Dana-Farber Cancer Institute
17
Associate Professor, Harvard Medical School
18
Boston, Massachusetts
19
20
21
22
A Matter of Record (301) 890-4188
3
1
CAPT Timothy H. Burgess, MD, MPH, FACP
2
Director
3
Infectious Disease Clinical Research Program
4
Preventative Medicine & Biostatistics
5
Uniformed Services University of the Health
6
Sciences
7
Bethesda, Maryland
8
9
Michael Green, MD, MPH
10
Professor of Pediatrics, Surgery and Clinical &
11
Translational Science
12
University of Pittsburgh School of Medicine
13
Division of Infectious Diseases
14
Director, Antimicrobial Stewardship & Infection
15
Prevention
16
Co-Director, Transplant Infectious Diseases
17
Children's Hospital of Pittsburgh
18
Pittsburgh, Pennsylvania
19
20
21
22
A Matter of Record (301) 890-4188
4
1
Barbara M. Gripshover, MD
2
Associate Professor of Medicine
3
University Hospitals Cleveland Medical Center
4
Case Western Reserve University
5
Division of Infectious Diseases and HIV Medicine
6
Cleveland, Ohio
7
8
Jennifer Le, PharmD, MAS
9
Professor of Clinical Pharmacy
10
University of California, San Diego
11
Skaggs School of Pharmacy and Pharmaceutical
12
Sciences
13
La Jolla, California
14
15
Ighovwerha Ofotokun, MD, MSc
16
Professor of Medicine
17
Division of Infectious Diseases
18
Department of Medicine
19
Emory University School of Medicine
20
Atlanta, Georgia
21
22
A Matter of Record (301) 890-4188
5
1
George K. Siberry, MD, MPH
2
Medical Officer
3
Division of Prevention, Care & Treatment
4
Office of HIV/AIDS (U.S. President's Emergency Plan
5
for AIDS Relief)
6
U.S. Agency for International Development
7
Arlington, Virginia
8
9
Sankar Swaminathan, MD
10
Professor and Chief
11
Division of Infectious Diseases
12
Department of Internal Medicine
13
University of Utah School of Medicine
14
Salt Lake City, Utah
15
16
Roblena E. Walker, PhD
17
(Consumer Representative)
18
Chief Executive Officer
19
EMAGAHA, INC.
20
Mableton, Georgia
21
22
A Matter of Record (301) 890-4188
6
1
Peter Joseph Weina, PhD, MD
2
Colonel, Medical Corps, US Army
3
Branch Chief
4
Research Regulatory Oversight Office
5
Office of the Under Secretary of Defense
6
(Personnel and Readiness)
7
Defense Health Headquarters
8
Falls Church, Virginia
9
10
TEMPORARY MEMBERS (Voting)
11
Laura W. Cheever, MD, ScM
12
Associate Administrator, HIV/AIDS Bureau
13
Health Resources and Serviced Administration
14
U.S. Department of Health and Human Services
15
Rockville, Maryland
16
17
Demetre C. Daskalakis, MD, MPH
18
Deputy Commissioner
19
Division of Disease Control
20
New York City Department of Health and Mental
21
Hygiene
22
New York, New York
A Matter of Record (301) 890-4188
7
1
Lori E. Dodd, PhD
2
Mathematical Statistician
3
Biostatistics Research Branch
4
Division of Clinical Research
5
National Institute of Allergy and
6
Infectious Diseases (NIAID)
7
National Institutes of Health (NIH)
8
Bethesda, Maryland
9
10
Thomas P. Giordano, MD, MPH
11
Professor and Chief
12
MD Anderson Foundation Chair Section of Infectious
13
Disease Department of Medicine
14
Baylor College of Medicine
15
Research Scientist
16
Michael E. DeBakey Veterans Affairs (VA) Medical
17
Center
18
Houston, Texas
19
20
21
22
A Matter of Record (301) 890-4188
8
1
Matthew Bidwell Goetz, MD
2
Chief, Infectious Diseases
3
VA Greater Los Angeles Healthcare System
4
Professor of Clinical Medicine
5
David Geffen School of Medicine at UCLA
6
Los Angeles, California
7
8
Patricia Lupole (via phone)
9
(Patient Representative)
10
Norfolk, Virginia
11
12
Sarah W. Read, MD, MHS
13
Deputy Director, Division of AIDS
14
NIAID, NIH
15
Rockville, Maryland
16
17
18
19
20
21
22
A Matter of Record (301) 890-4188
9
1
Dawn K. Smith, MD, MS, MPH
2
Biomedical Interventions Implementation Activity
3
Lead
4
Epidemiology Branch, Division of HIV/AIDS
5
Prevention
6
National Center for HIV, Viral Hepatitis, STD, and
7
Tuberculosis Prevention
8
Centers for Disease Control and Prevention
9
Atlanta, Georgia
10
11
ACTING INDUSTRY REPRESENTATIVE TO THE COMMITTEE
12
(Non-Voting)
13
Walid M. Awni, PhD
14
(Acting Industry Representative)
15
Awni BioPharmaceutical Consulting, LLC
16
Riverwoods, Illinois
17
18
FDA PARTICIPANTS (Non-Voting)
19
John Farley, MD, MPH
20
Deputy Director
21
Office of Antimicrobial Products (OAP)
22
Office of New Drugs (OND), CDER, FDA
A Matter of Record (301) 890-4188
10
1
Debra Birnkrant, MD
2
Director
3
Division of Antiviral Products (DAVP)
4
OAP, OND, CDER, FDA
5
6
Jeffrey Murray, MD, MPH
7
Deputy Director
8
DAVP, OAP, OND, CDER, FDA
9
10
Wendy Carter, DO
11
Medical Officer Team Leader
12
DAVP, OAP, OND, CDER, FDA
13
14
Peter Miele, MD
15
Medical Officer
16
DAVP, OAP, OND, CDER, FDA
17
18
Jenny H. Zheng, PhD
19
Clinical Pharmacology Reviewer
20
Division of Clinical Pharmacology IV
21
Office of Clinical Pharmacology
22
Office of Translational Sciences, CDER, FDA
A Matter of Record (301) 890-4188
11
1
C O N T E N T S
2
AGENDA ITEM
PAGE
3
Call to Order and Introduction of Committee
4
Lindsey Baden, MD
13
5
Conflict of Interest Statement
6
Lauren Tesh Hotaki, PharmD, BCPS, BCIDP
19
7
FDA Opening Remarks
8
Jeffrey Murray, MD, MPH
24
9
Applicant Presentation - Gilead Sciences, Inc.
10
Introduction
11
Diana Brainard, MD
29
12
DISCOVER Study Design, Treatment
13
Population, and Efficacy Results
14
Scott McCallister, MD
44
15
DISCOVER Safety and Extrapolations
16
Moupali Das, MD, MPH
61
17
Clinical Context
18
Richard Elion, MD
81
19
Clarifying Questions
99
20
21
22
A Matter of Record (301) 890-4188
1
C O N T E N T S (continued)
2
AGENDA ITEM
3
FDA Presentation
4
NDA 208215/S12 - Descovy PrEP
5
Peter Miele, MD
6
Clarifying Questions
7
Open Public Hearing
8
Clarifying Questions (continued)
9
Questions to the Committee and Discussion
10
Adjournment
11
12
13
14
15
16
17
18
19
20
21
22
12
PAGE
129 156 188 229 282 349
A Matter of Record (301) 890-4188
13
1
P R O C E E D I N G S
2
(8:30 a.m.)
3
Call to Order
4
Introduction of Committee
5
DR. BADEN: It's 8:30. Good morning. I
6
would first like to remind everyone to please
7
silence your cell phones, smartphones, and any
8
other devices if you have not already done so. I
9
would also like to identify the FDA press contacts,
10
Alison Hunt and Charles Kohler. If you're present,
11
please stand. They're in the back. If there are
12
questions for the press, please address them to
13
Alison and Charlie.
14
My name is Lindsey Baden. I will be
15
chairing today's meeting. I will now call the
16
Antimicrobial Drugs Advisory Committee to order.
17
We'll start by going around the table and
18
introducing ourselves. We'll start with the FDA to
19
my left and go around the table.
20
DR. FARLEY: Good morning. John Farley,
21
deputy director of the Office of Antimicrobial
22
Products, CDER, FDA.
A Matter of Record (301) 890-4188
14
1
DR. BIRNKRANT: Debbie Birnkrant, director,
2
Division of Antiviral Products, CDER, FDA.
3
DR. MURRAY: Jeff Murray, deputy, Division
4
of Antiviral Products, CDER, FDA.
5
DR. CARTER: Wendy Carter, clinical team
6
leader, Division of Antiviral Products, CDER, FDA.
7
DR. MIELE: Pete Miele, medical officer,
8
Division of Antiviral Products, CDER, FDA.
9
DR. ZHENG: Jenny Zheng, clinical
10
pharmacology reviewer for antiviral products, CDER,
11
FDA.
12
DR. CHEEVER: Hi. I'm Laura Cheever from
13
the HIV/AIDS Bureau at the Health Resources and
14
Services Administration.
15
DR. SWAMINATHAN: I'm Shankar Swaminathan,
16
infectious diseases division chief at the
17
University of Utah.
18
DR. SIBERRY: George Siberry, Office of
19
HIV/AIDS, Global Health Bureau, USAID.
20
DR. GRIPSHOVER: Barbara Gripshover from
21
University Hospitals Cleveland, Case Western
22
Reserve University, adult infectious disease.
A Matter of Record (301) 890-4188
15
1
DR. GREEN: Michael Green, University of
2
Pittsburgh School of Medicine, Children's Hospital
3
Pittsburgh, pediatric infectious diseases.
4
DR. WEINA: Peter Weina, adult infectious
5
diseases, Research Regulatory Oversight Office,
6
Defense Health Headquarters.
7
DR. HOTAKI: Lauren Hotaki, designated
8
federal officer.
9
DR. BADEN: Lindsey Baden, adult infectious
10
diseases, Brigham and Women's Hospital, Dana Farber
11
Cancer Institute, Harvard Medical School, Boston
12
Mass.
13
DR. OFOTOKUN: Igho Ofotokun, adult
14
infectious diseases, Emory University, Atlanta,
15
Georgia.
16
DR. BURGESS: Tim Burgess, adult infectious
17
diseases. I'm director of DoD's Infectious Disease
18
Clinical Research Program at Uniform Services
19
University.
20
DR. LE: Jennifer Le, professor of pharmacy
21
at UC San Diego, pediatric infectious diseases.
22
DR. WALKER: Good morning. Dr. Roblena
A Matter of Record (301) 890-4188
16
1
Walker, EMAGAHA, Inc., Atlanta, Georgia, consumer
2
representative.
3
DR. GIORDANO: Tom Giordano, adult
4
infectious disease, Baylor College of Medicine and
5
the Michael E. DeBakey VA Medical Center, Houston,
6
Texas.
7
DR. DASKALAKIS: Demetre Daskalakis, adult
8
infectious diseases, and also deputy commissioner
9
for disease control at the New York City Department
10
of Health and Mental Hygiene.
11
DR. READ: Sarah Read, deputy director of
12
the Division of AIDS at the National Institute of
13
Allergy and Infectious Diseases.
14
DR. SMITH: Hi. Dawn Smith, medical
15
epidemiologist, Centers for Disease Control and
16
Prevention.
17
DR. GOETZ: Matthew Goetz, VA Greater Los
18
Angeles Healthcare System, David Geffen School of
19
Medicine, adult infectious diseases.
20
DR. AWNI: Walid Awni, retired. I retired
21
from AbbVie last year as vice president of clinical
22
pharmacology and pharmacometrics. I'm the acting
A Matter of Record (301) 890-4188
17
1
industry representative.
2
DR. BADEN: Thank you. Dr. Dodd?
3
DR. DODD: Dr. Dodd, biostatistician at
4
National Institute of Allergy and Infectious
5
Diseases.
6
DR. BADEN: I think we may have someone on
7
the phone. Ms. Lupole?
8
MS. LUPOLE: Yes, sir. Good morning
9
[inaudible -feedback].
10
DR. BADEN: You have a bit of feedback.
11
MS. LUPOLE: I'm sorry. Can you hear me
12
now?
13
DR. BADEN: Yes, we can.
14
MS. LUPOLE: Patricia Lupole, patient
15
representative.
16
DR. BADEN: Thank you.
17
For topics such as those being discussed at
18
today's meeting, there are often a variety of
19
opinions, some of which are quite strongly held.
20
Our goal is that today's meeting will be a fair and
21
open forum for discussion of these issues and that
22
individuals can express their views without
A Matter of Record (301) 890-4188
18
1
interruption. Thus, as a gentle reminder,
2
individuals will be allowed to speak into the
3
record only if recognized by the chairperson. We
4
look forward to a productive meeting.
5
In the spirit of the Federal Advisory
6
Committee Act and the Government in the Sunshine
7
Act, we ask that the advisory committee members
8
take care that their conversations about the topic
9
at hand take place in the open forum of the
10
meeting.
11
We are aware that members of the media are
12
anxious to speak with the FDA about these
13
proceedings. However, FDA will refrain from
14
discussing the details of this meeting with the
15
media until its conclusion. Also, the committee is
16
reminded to please refrain from discussing the
17
meeting topic during breaks or lunch. Thank you.
18
I thank everyone for making the time to be
19
here to participate in this discussion. We know
20
how busy everyone is.
21
I will ask Dr. Lauren Hotaki to read the
22
Conflict of Interest Statement for the meeting.
A Matter of Record (301) 890-4188
19
1
Conflict of Interest Statement
2
DR. HOTAKI: The Food and Drug
3
Administration is convening today's meeting of the
4
Antimicrobial Drugs Advisory Committee under the
5
authority of the Federal Advisory Committee Act of
6
1972. With the exception of the industry
7
representative, all members and temporary voting
8
members of the committee are special government
9
employees or regular federal employees from other
10
agencies and are subject to federal conflict of
11
interest laws and regulations.
12
The following information on the status of
13
this committee's compliance with federal ethics and
14
conflict of interest laws, covered by but not
15
limited to those found at 18 U.S.C. Section 208, is
16
being provided to participants in today's meeting
17
and to the public.
18
FDA has determined that members and
19
temporary voting members of this committee are in
20
compliance with federal ethics and conflict of
21
interest laws. Under 18 U.S.C. Section 208,
22
Congress has authorized the FDA to grant waivers to
A Matter of Record (301) 890-4188
20
1
special government employees and regular federal
2
employees who have potential financial conflicts
3
when it is determined that the agency's need for a
4
special government employee's services outweighs
5
his or her potential financial conflict of
6
interest, or when the interest of a regular federal
7
employee is not so substantial as to be deemed
8
likely to affect the integrity of the services
9
which the government may expect from the employee.
10
Related to the discussion of today's
11
meeting, members and temporary voting members of
12
this committee have been screened for potential
13
financial conflicts of interest of their own as
14
well as those imputed to them, including those of
15
their spouses or minor children, and, for purposes
16
of 18 U.S.C. Section USC Section 208, their
17
employers. These interests may include
18
investments; consulting; expert witness testimony;
19
contracts, grants, CRADAs; teaching, speaking,
20
writing; patents and royalties; and primary
21
employment.
22
Today's agenda involves discussion of
A Matter of Record (301) 890-4188
21
1
supplemental new drug application 208215,
2
supplement 12, DESCOVY, emtricitabine
3
200 milligrams and tenofovir alafenamide
4
25 milligrams submitted by Gilead Sciences, Inc.,
5
proposed for pre-exposure prophylaxis to reduce the
6
risk of sexually acquired HIV-1 infection among
7
individuals who are HIV negative and at risk for
8
HIV. This is a particular matters meeting during
9
which specific matters related to Gilead's sNDA
10
will be discussed.
11
Based on the agenda for today's meeting and
12
all financial interests reported by the committee
13
members and temporary voting members, conflict of
14
interest waivers have been issued in accordance
15
with 18 U.S.C. Section 208(b)(3) to Dr. Lindsey
16
Baden and Dr. Barbara Gripshover.
17
Dr. Baden's waiver addresses his employer's
18
current research contract for related study by a
19
competing firm for which his employer receives
20
between $0 and $50,000 annually. Dr. Baden's waiver
21
also addresses his employer's current research
22
contract for related studies through the HIV
A Matter of Record (301) 890-4188
22
1
Vaccine Trials Network sponsored by the National
2
Institute of Allergy and Infectious Diseases of the
3
National Institutes of Health and a competing firm,
4
for which his employer receives $1.5 to $2.5
5
million annually.
6
Dr. Gripshover's waiver addresses her
7
employer's current research contracts for four
8
related studies involving competing/affected
9
products for which her employer receives between
10
$50,001 to $100,000 annually for two studies, and
11
between $300,000 to $400,000 annually, and to $0 to
12
$50,000 annually for the other two studies.
13
The waivers allow these individuals to
14
participate fully in today's deliberations. FDA's
15
reasons for issuing the waivers are described in
16
the waiver documents, which are posted at the FDA's
17
website. Copies of the waivers may also be
18
obtained by submitting a written request to the
19
agency's Freedom of Information Division,
20
5630 Fishers Lane, Room 1035, Rockville, Maryland,
21
20857, or requests may be sent via fax to 301-827-
22
9267.
A Matter of Record (301) 890-4188
23
1
To ensure transparency, we encourage all
2
standing committee members and temporary voting
3
members to disclose any public statements that they
4
have made concerning the product at issue. With
5
respect to FDA's invited industry representative,
6
we would like to disclose the Dr. Walid Awni is
7
participating in this meeting as a nonvoting
8
industry representative, acting on behalf of
9
regulated industry. Dr. Awni's role at this meeting
10
is to represent industry in general and not any
11
particular company. Dr. Awni is an independent
12
pharmaceutical consultant.
13
We'd like to remind members and temporary
14
voting members that if the discussions involved any
15
other products or firms not already on the agenda
16
for which an FDA participant has a personal or
17
imputed financial interest, the participants need
18
to exclude themselves from such involvement, and
19
their exclusion will be noted for the record. FDA
20
encourages all other participants to advise the
21
committee of any financial relationships that they
22
may have with the firm at issue. Thank you.
A Matter of Record (301) 890-4188
24
1
DR. BADEN: Thank you.
2
We will proceed with the FDA opening remarks
3
from Dr. Murray.
4
FDA Opening Remarks - Jeffrey Murray
5
DR. MURRAY: Good morning. The Division of
6
Antiviral Products extends its warm welcome to the
7
committee and to the audience to discuss a new
8
supplemental application for Descovy, for the
9
prevention of sexually-acquired HIV infection, and
10
we're happy to be talking about expanding the HIV
11
prevention armamentarium today.
12
Some of you on the panel, and perhaps in the
13
audience, may have been here in 2012, in this very
14
room -- I know I was -- when the advisory committee
15
voted on whether Truvada should be approved for
16
PrEP. As you recall, the committee voted yes, and
17
Truvada became the first U.S. approved product for
18
PrEP, and really the first product for HIV PrEP for
19
sexually-acquired HIV infection anywhere in the
20
world. Fast forward to seven years, and that
21
brings us to today's topics.
22
There are similarities and differences
A Matter of Record (301) 890-4188
25
1
between these two products. Both our fixed-dose
2
combinations that contain emtricitabine. Both
3
contain a prodrug of tenofovir with the same active
4
metabolite. Both are approved for HIV treatment,
5
both of the products. Tenofovir components are
6
also approved as single agents for the treatment of
7
chronic hepatitis B.
8
However, there are also differences,
9
specifically as they relate to the bioavailability
10
of tenofovir as delivered by these two different
11
prodrugs in these fixed-dose combination. Descovy
12
delivers lower levels of plasma tenofovir in tissue
13
and organs and higher levels of intracellular
14
tenofovir diphosphate of active metabolite.
15
As I said, there are also differences in
16
tissue and organ distribution. This results in
17
somewhat a different safety profile, but did not
18
result in efficacy differences for HIV treatment.
19
The other differences, Truvada is already approved
20
for prevention and Descovy is not, and that's a
21
topic for today.
22
To support the prevention indication of
A Matter of Record (301) 890-4188
26
1
Truvada, I remind you the applicant submitted two
2
clinical trials, a clinical trial in MSM
3
transgender women, iPrEx, and a trial in discordant
4
heterosexual couples, Partners PrEP, which allowed
5
for a broad indication among at-risk populations.
6
Today we're dealing with one clinical trial.
7
What are some of the regulatory
8
considerations about the basis of supporting an
9
approval? Really, the number of clinical trials
10
needed to support an approval depends on the
11
regulatory situation.
12
For a new molecule entering the market,
13
generally two or more drugs are expected -- or two
14
or more trials are expected. However, for a new
15
and related indication for a previously approved
16
drug, often only one trial is needed to support
17
approval.
18
Likewise, if there's a new dosing schedule,
19
say twice daily to once daily, and you can't make a
20
pharmacokinetic link, that's also been supported by
21
one trial, or for a new population where PK is
22
different, usually we rely on one clinical trial.
A Matter of Record (301) 890-4188
27
1
For Descovy, FDA's initial drug development
2
advice was that clinical trials should be conducted
3
in the relevant population, and that a PK link
4
alone would not be possible. So for this
5
application, as I said, we have one trial in MSM
6
and transgender women but none in cisgender women
7
at risk.
8
The primary issue for today and what you'll
9
be asked in the question is given the uncertainty
10
around the protective correlate, can extrapolation
11
be used to further expand the indicated population?
12
With that being said, this application is a
13
special case in the development of drugs for HIV
14
prevention, and with that, I have the following
15
caveat that the approach for Descovy may not apply
16
to future new molecular entities because in this
17
case, a prodrug, tenofovir for PrEP, has already
18
been approved, and there is a possibility that data
19
within and external to the Descovy program can be
20
leveraged. So we ask the committee today for their
21
advice on how this data can be best leveraged.
22
Thank you.
A Matter of Record (301) 890-4188
28
1
DR. BADEN: Thank you. We'll now move on to
2
the applicant presentations.
3
Both the FDA and the public believe in a
4
transparent process for information gathering and
5
decision making. To ensure such transparency at
6
the advisory committee meeting, FDA believes that
7
it is important to understand the context of an
8
individual's presentation.
9
For this reason, FDA encourages all
10
participants, including the applicant's
11
non-employee presenters, to advise the committee of
12
any financial relationships they may have with the
13
applicants, such as consulting fees, travel
14
expenses, honoraria, an interest in a sponsor,
15
including equity interests and those based upon the
16
outcome of the meeting.
17
Likewise, FDA encourages you at the
18
beginning of your presentation to advise the
19
committee if you do not have any such financial
20
relationships. If you choose not to address this
21
issue of financial relationships at the beginning
22
of your presentation, it will not preclude you from
A Matter of Record (301) 890-4188
29
1
speaking.
2
We'll now proceed with Gilead's
3
presentations. Dr. Brainard?
4
Applicant Presentation - Diana Brainard
5
DR. BRAINARD: Good morning. Ending the HIV
6
epidemic requires not just highly effective
7
treatments for people who have already been
8
infected, but additional options for preventing new
9
infections.
10
My name is Diana Brainard, and I lead the
11
HIV and emerging viruses group at Gilead Sciences.
12
I am an infectious diseases physician and have
13
worked as a clinician and scientist in both the
14
U.S. and Africa to care for people living with HIV
15
and tackle the epidemic. It is a pleasure and
16
honor to be here today to work with this committee
17
to bring forward another HIV prevention option that
18
will help us achieve our shared goal of HIV
19
elimination.
20
Seven years ago, as Dr. Murray mentioned,
21
Truvada was approved to prevent sexually-acquired
22
HIV infection and remains today the only approved
A Matter of Record (301) 890-4188
30
1
therapy for HIV pre-exposure prophylaxis or PrEP.
2
Truvada is the fixed-dose combination of 2 HIV
3
reverse transcriptase inhibitors, emtricitabine and
4
tenofovir disoproxil fumarate. Truvada is approved
5
as part of a complete regimen for the treatment of
6
HIV in adults and adolescents, as well as for PrEP.
7
Tenofovir disoproxil fumarate is also
8
approved as a single agent for the treatment of
9
chronic hepatitis B. Descovy is the fixed-dose
10
combination tablet of emtricitabine and tenofovir
11
alafenamide. It is approved for HIV treatment, and
12
tenofovir alafenamide is approved as a single agent
13
for treatment of chronic hepatitis B. Descovy is
14
not approved for PrEP.
15
We are proposing an indication for Descovy
16
for PrEP in adults and adolescents based on the
17
data we are discussing today. Tenofovir disoproxil
18
fumarate and tenofovir alafenamide are both
19
prodrugs of tenofovir, but they have markedly
20
different metabolism. Tenofovir alafenamide, or
21
TAF, is dosed at one 12th that of tenofovir
22
disoproxil fumarate, or TDF, because of the
A Matter of Record (301) 890-4188
31
1
difference in half-life.
2
TDF is rapidly converted to tenofovir, or
3
TFV, resulting in high plasma tenofovir levels
4
which have direct and indirect adverse effects on
5
kidney and bone. The half-life of TAF is 75 times
6
longer than that of TDF, which results in
7
90 percent lower plasma tenofovir levels.
8
Descovy's lower levels of circulating tenofovir
9
translate to fewer clinically relevant adverse
10
renal and bone effects.
11
The longer half-life of TAF also allows it
12
more time to enter peripheral blood mononuclear
13
cells. Intracellularly, TAF is metabolized to the
14
active metabolite, tenofovir diphosphate, or
15
TFV-DP, where it achieves 4 to 7-fold higher levels
16
of tenofovir diphosphate than those achieved by
17
TDF.
18
For both TAF and TDF, tenofovir diphosphate
19
within PBMCs and specifically CD-4 positive
20
T cells, is responsible for the inhibition of HIV
21
replication, which leads to protection against HIV
22
infection in the setting PrEP, as well as viral
A Matter of Record (301) 890-4188
32
1
suppression in the case of HIV treatments.
2
Tenofovir diphosphate is an adenosine analog
3
that inhibits the enzyme HIV reverse transcriptase,
4
which transcribes HIV RNA into proviral DNA. This
5
mechanism of action is the same for both the
6
prevention of HIV acquisition as well as for
7
suppression of viremia in the setting of treatment,
8
and the level of intracellular tenofovir
9
diphosphate correlates with antiviral activity.
10
Pharmacokinetic differences between TAF and
11
TDF result not only in higher levels of tenofovir
12
diphosphate with TAF versus TDF, but also a faster
13
rise of tenofovir diphosphate levels within PBMCs,
14
including the target cells for HIV replication,
15
CD-4 positive T cells. After a single dose,
16
Descovy achieves intracellular tenofovir
17
diphosphate levels above 40 femtomoles per million
18
cells within 2 hours.
19
This threshold is relevant for PrEP based on
20
its correlation using Truvada clinical data from a
21
trial in men who have sex with men, with a
22
90 percent reduction in risk of HIV acquisition as
A Matter of Record (301) 890-4188
33
1
compared to placebo.
2
In contrast, Truvada takes approximately
3
3 days for the mean level to reach this EC90, and
4
steady state levels remain lower than those for
5
Descovy. Once steady state is achieved with either
6
Descovy or Truvada, if drug is stopped, tenofovir
7
diphosphate levels start to decline at a similar
8
rate. However, since levels are so much higher
9
with Descovy as compared to Truvada, they remain
10
above this EC90 for 16 days with Descovy compared
11
with 10 days for Truvada.
12
These pharmacokinetic advantages of Descovy
13
suggested to us that Descovy could be highly
14
effective for PrEP, and the safety advantages
15
observed in people living with HIV taking Descovy
16
could also be realized among those at risk for HIV
17
infection.
18
In 2015, when the DISCOVER study design was
19
coming together, there was uncertainty around
20
whether drug levels in the genital tract or
21
peripheral blood mononuclear cells best correlated
22
with protection against HIV. The higher levels of
A Matter of Record (301) 890-4188
34
1
Tenofovir diphosphate in PBMCs with Descovy could
2
potentially confer an efficacy advantage and offer
3
a more forgiving regimen for PrEP, provided these
4
levels correlated with protection. However, if
5
genital tract tissue levels drive efficacy, Descovy
6
could be less effective for prevention.
7
Data from healthy volunteers show that
8
rectal tissue levels are 10-fold lower following
9
Descovy administration compared to Truvada. Our
10
hypothesis was that prevention efficacy for oral
11
drugs would be best measured by peripheral blood
12
mononuclear cell drug levels rather than tissue
13
homogenate levels, and that, therefore, Descovy
14
would be at least as efficacious as Truvada in
15
spite of this difference in rectal tissue levels.
16
This hypothesis was based on advances in the
17
understanding of mucosal transmission of HIV. HIV
18
must first breach the epithelium to reach the
19
subepithelium, and it is generally believed that a
20
single cell first becomes infected and initiates
21
subsequent events.
22
Chemokines, primarily secreted by
A Matter of Record (301) 890-4188
35
1
plasmacytoid dendritic cells, attract PBMCs.
2
Specifically, CD-4 positive and CD-8 positive
3
T cells from the pool of PBMCs traffic from the
4
circulation to the tissue. This then results in a
5
small founder population of initially infected CD-4
6
T cells located in the subepithelium.
7
The recruitment of target cells for HIV, the
8
CD-4 T cells, from the periphery is critical in
9
order for systemic infection to occur.
10
Dissemination of these recruited and now infected
11
CD-4 positive T cells occurs as they enter the
12
lymphatic system to travel to regional lymph nodes
13
and spread throughout the body.
14
Protection against systemic HIV infection
15
can occur via both topical and systemic modalities.
16
Topical antiretrovirals, such as investigational
17
tenofovir gel, allow for the diffusion of drug into
18
tissues within the genital tract. Efficacy depends
19
on reaching therapeutic levels intracellularly
20
within the local CD-4 positive T cells. These
21
methods have generally proven less effective than
22
PrEP delivered systemically.
A Matter of Record (301) 890-4188
36
1
Truvada and Descovy distribute widely
2
throughout the body and can offer a greater degree
3
of protection. They can both reach the genital
4
tissues through the blood supply, where they can
5
then access resident lymphocytes.
6
Importantly as well, PBMCs that contain the
7
active metabolite of both Truvada and Descovy,
8
tenofovir diphosphate also can reach the genital
9
tract and can be among the cells recruited to the
10
site of initial infection, as well as into the
11
regional draining lymph nodes so as to prevent
12
systemic infection.
13
While there has been no clear evidence of a
14
correlation of preventive efficacy of Truvada for
15
PrEP with tissue levels, efficacy strongly
16
correlates with drug levels of tenofovir
17
diphosphate within PBMCs.
18
A subset of participants in the IPREX study,
19
in men who have sex with men of Truvada versus
20
placebo, had tenofovir diphosphate levels assessed
21
in PBMCs. Because of the wide range of adherence
22
and that trial and the placebo arm to which a
A Matter of Record (301) 890-4188
37
1
direct comparison could be made, it was possible to
2
construct a relationship between PBMC tenofovir
3
diphosphate levels and risk production with respect
4
to HIV incidence.
5
It was with these data that Dr. Anderson
6
established the correlate of protection for
7
90 percent risk reduction for tenofovir diphosphate
8
levels at 40 femtomoles per million PBMCs and
9
showed that there's a range of protection above and
10
below that level. These data are now well
11
recognized and have been cited in the most recent
12
CDC PrEP guidance issued on July 18th of this year.
13
When the DISCOVER study was being designed,
14
the scientific and clinical understanding of HIV
15
prevention was less mature. At that time, Truvada
16
for PrEP was only approved in adults. Descovy was
17
under review by the FDA for the treatment of HIV.
18
We knew that rectal tissue levels with Descovy were
19
10-fold lower than those achieved with Truvada.
20
If the primary driver for prevention
21
efficacy, with orally administered tenofovir
22
prodrugs is local tissue drug levels, then Truvada
A Matter of Record (301) 890-4188
38
1
should be better than Descovy at preventing HIV
2
infection. However, if obtaining high levels in
3
PBMCs is what's important, then Descovy should be
4
at least as effective as Truvada.
5
This question was addressed in the phase 3
6
DISCOVER trial. The DISCOVER trial, an
7
international phase 3 study, was conducted to
8
assess the safety and efficacy of Descovy for HIV
9
prevention. This was a double-blind, active
10
comparator, noninferiority trial, comparing Descovy
11
to the standard of care for prevention, Truvada.
12
The study enrolled over 5,000 cismen and
13
transgender women who have sex with men. The trial
14
was designed and conducted in close collaboration
15
with FDA and the community. Importantly, the study
16
met its primary endpoint, demonstrating
17
noninferiority of Descovy to Truvada for the
18
prevention of HIV infection.
19
Among individuals randomized to Descovy, 7
20
acquired HIV infection for an incidence rate of
21
0.16 per 100 person-years. In the Truvada group,
22
there were 15 infections resulting in an incidence
A Matter of Record (301) 890-4188
39
1
rate of 0.34 per 100 person-years. The incident
2
rate ratio, the prespecified method for determining
3
noninferiority, was 0.47 with an upper bound of the
4
confidence interval less than the prespecified
5
margin of 1.62.
6
Additionally, the prespecified
7
alpha-controlled secondary safety endpoints were
8
met, demonstrating superiority of Descovy to
9
Truvada with respect to markers of bone and renal
10
toxicity. Collectively, these data demonstrate
11
that Descovy is highly effective at preventing HIV
12
acquisition and demonstrates safety benefits over
13
Truvada.
14
What we know now is that both Truvada and
15
Descovy are highly effective for PrEP if taken.
16
Adherence is the key determinant of efficacy. A
17
correlate of protection has been established for
18
tenofovir diphosphate levels in PBMCs. The
19
DISCOVER trial confirms that 10-fold lower rectal
20
levels of tenofovir diphosphate with Descovy versus
21
Truvada are not relevant for HIV protection, and
22
that the 7-fold higher tenofovir diphosphate levels
A Matter of Record (301) 890-4188
40
1
with Descovy versus Truvada might confer a
2
potential efficacy advantage for Descovy.
3
These results support the conclusion that
4
PBMC drug levels drive the efficacy of orally
5
administered tenofovir prodrugs. This finding is
6
an important consideration for the extrapolation of
7
the DISCOVER results from cismen and transgender
8
women to ciswomen.
9
To date, clinical trials in women have had
10
heterogeneous efficacy results reflecting highly
11
variable adherence. Data from clinical trials
12
demonstrate that when controlling for adherence,
13
Truvada is equally efficacious in women and men.
14
There is a biologic rationale for this finding.
15
The biology of HIV as well as the
16
intracellular antiviral activity of tenofovir
17
diphosphate are independent of gender. HIV
18
replicates within CD-4 positive lymphocytes, which
19
must be recruited to the site of initial infection
20
in order to successfully lead to systemic
21
transmission. Adequate drug levels within these
22
recruited cells are necessary and sufficient to
A Matter of Record (301) 890-4188
41
1
mediate protection against HIV infection.
2
Multiple lines of evidence support bridging
3
the efficacy results for Descovy for PrEP from the
4
men and transgender women in DISCOVER to ciswomen.
5
In the setting of HIV treatment, the efficacy and
6
safety of Descovy-based therapy have been well
7
established in over 2000 women and are comparable
8
to results in men.
9
Descovy and Truvada both inhibit HIV
10
replication in CD-4 T cells through the same active
11
metabolite, tenofovir diphosphate. Extensive
12
pharmacology assessments have demonstrated that the
13
levels of tenofovir diphosphate are similar
14
irrespective of HIV status or gender. Taken
15
together, these data support the use of Descovy for
16
HIV prevention in women.
17
There is similar support for the
18
extrapolation to adolescence. Descovy and the
19
three other Descovy-containing single-tablet
20
regimens are all indicated for HIV treatment in
21
adolescence based on the safety and efficacy
22
established in this group. HIV behaves similarly
A Matter of Record (301) 890-4188
42
1
independent of age, and therefore the extension of
2
safety and efficacy of Descovy for PrEP to
3
adolescents can be based on similar pharmacokinetic
4
exposures to Descovy between the DISCOVER study
5
participants and adolescents, as well as the
6
similar mechanism of action of these drugs. We
7
also know that HIV infection status has no relevant
8
impact on these parameters. Taken together, these
9
data support the use of Descovy for HIV prevention
10
in adolescents.
11
Based on the data from the DISCOVER study,
12
the established safety and efficacy of Descovy for
13
HIV treatment across men, women, and adolescents
14
and pharmacokinetic bridging, the following
15
additional indication is proposed for Descovy.
16
Descovy is indicated for pre-exposure prophylaxis
17
to reduce the risk of sexually-acquired HIV in
18
at-risk adults and adolescents weighing at least
19
35 kilograms.
20
You will next hear from Dr. Scott
21
McCallister, who will provide an overview of the
22
phase 3 DISCOVER trial and the efficacy results.
A Matter of Record (301) 890-4188
43
1
Then Dr. Moupali Das will present the safety of
2
Descovy for PrEP, as well as the basis for its use
3
for HIV prevention in women and adolescents.
4
We're honored to have Dr. Rick Elion with us
5
today. Dr. Elion has a long-standing history of
6
providing HIV treatment and prevention services to
7
individuals in the D.C. area for marginalized
8
communities, and he'll provide clinical context for
9
the results of the DISCOVER trial. I will then
10
return to lead our responses to questions. We have
11
Gilead team members spanning multiple disciplines
12
to address these questions.
13
In addition, we're pleased that Dr. Peter
14
Anderson is here today to address questions around
15
adherence. Dr. Anderson is a professor of
16
pharmaceutical sciences at the University of
17
Colorado. His laboratory specializes in the
18
assessment of drug levels in dried blood spots, and
19
they performed nearly 4,000 dried blood spot
20
analyses in DISCOVER as part of our adherence
21
assessments.
22
I'd like to now welcome Scott McAllister to
A Matter of Record (301) 890-4188
44
1
the lectern.
2
Applicant Presentation - Scott McCallister
3
DR. McCALLISTER: Thank you, Diana, and good
4
morning, everyone. I'm also an infectious disease
5
specialist with a long history of clinical patient
6
care and clinical research in HIV. In this
7
section, I'll describe the DISCOVER study design,
8
the treatment population, and the efficacy results.
9
DISCOVER is an ongoing, randomized,
10
double-blind, noninferiority trial that enrolled
11
both cismen and transgender women who have sex with
12
men. As shown at the top, participants were
13
randomized 1-to-1 to either Descovy or Truvada
14
daily. Each of the nearly 2700 MSM or transgender
15
women received 1 tablet of active drug and 1 dummy
16
placebo tablet.
17
The primary efficacy endpoint analysis was
18
time based and was conducted when all participants
19
completed 48 weeks in the study and half had
20
completed 96 weeks. The primary endpoint was the
21
HIV incidence rate per 100 person-years on study.
22
The study was blinded to investigators and study
A Matter of Record (301) 890-4188
45
1
participants until the final person enrolled
2
completed 96 weeks. At the next scheduled visit,
3
individual participants are unblinded and offered a
4
switch to open-labeled Descovy. Unblinding is
5
currently ongoing and not yet complete.
6
Eligibility criteria were designed to ensure
7
that the study enrolled a population at high risk
8
of HIV infection. All participants were required
9
to have at least 1 of the 2 following sexual risk
10
criteria: two or more episodes of condomless anal
11
sex with more than one unique partner in the 12
12
weeks before enrollment or a diagnosis of either
13
rectal gonorrhea, rectal chlamydia, or syphilis in
14
the 24 weeks before enrollment. All needed to be
15
HIV and hepatitis B negative. Prior or current use
16
of PrEP was permitted, and no washout of PrEP drugs
17
was required.
18
DISCOVER sites were mostly urban. They were
19
specifically chosen to be in locations with a high
20
background HIV incidence, and all sites were
21
required to be able to enroll people with
22
significant sexual risk for HIV acquisition. We
A Matter of Record (301) 890-4188
46
1
also selected sites with the cultural competence to
2
enroll and retain people of color and transgender
3
women.
4
Ultimately, DISCOVER included 94 sites in 11
5
countries in North America and Western Europe.
6
Some were hospitals, some private practices, and
7
some local sexually transmitted infection clinics.
8
Each site was responsible to determine the best
9
recruitment practice within their own community.
10
All of those who met eligibility criteria were
11
allowed to enroll. It was our goal to allow each
12
person who knew themselves to be at risk of our HIV
13
infection to participate.
14
When we designed DISCOVER, we consulted with
15
investigators from the prior PrEP trials in MSMs.
16
We wanted to ensure that the study included the
17
right design elements, comparable sexual risk
18
eligibility criteria, optimal HIV testing, and STI
19
testing, so that our study would yield reliable
20
results. We also discussed design issues with both
21
site investigators and with community members in
22
North America and Europe to ensure that the study
A Matter of Record (301) 890-4188
47
1
was practical and aligned with existing clinical
2
practice.
3
Community members encouraged us to establish
4
advisory boards for ongoing dialogue with them. As
5
a result, three community advisory boards were set
6
up, one that was DISCOVER specific and two that
7
dealt with broader HIV issues in North America and
8
the EU. We drew valuable input from these
9
interactions during the trial, during the design
10
phase, during recruitment, and during study
11
conduct.
12
The primary efficacy endpoint was based on
13
the number of HIV infections diagnosed in DISCOVER
14
divided by person-years of exposure in the study.
15
Noninferiority of Descovy to Truvada was assessed
16
by an incidence rate ratio in which the HIV
17
incidence rate in the Descovy arm was divided by
18
the rate in the Truvada arm.
19
We derived the noninferiority margin of 1.62
20
by pooling the incidence rates in the Truvada arms
21
of the three prior randomized controlled trials in
22
MSMs: iPrEx, PROUD, and IPERGAY. If the upper
A Matter of Record (301) 890-4188
48
1
bound around the confidence interval of the
2
incidence rate ratio in DISCOVER was less than
3
1.62, Descovy would be noninferior to Truvada.
4
The incidence rate ratio analysis of the
5
primary endpoint was a robust means of evaluating
6
the effectiveness of Descovy, ensuring that the
7
result was due to the drugs used in the study and
8
that the treatment population was at sufficient
9
risk of HIV.
10
At each visit, we assessed general safety,
11
including graded adverse events, adverse events
12
leading to discontinuation, serious adverse events,
13
and general safety labs. There was renal lab
14
testing at each visit, bone mineral density testing
15
every 48 weeks, and sexually transmitted infection
16
testing from 3 anatomic sites also at each visit.
17
We used an analysis cascade for 6
18
prespecified secondary safety endpoints, where
19
previous data suggested a possible difference
20
between the arms due to lower levels of plasma TFV
21
in those on TAF. The safety's cascade began with
22
changes from baseline in bone mineral density, or
A Matter of Record (301) 890-4188
49
1
BMD, at both the hip and spine.
2
If there were significant differences
3
favoring Descovy on each of these, we then
4
evaluated spillage of the specific proximal renal
5
tubular proteins associated with plasma TFV, the
6
beta-2 microglobulin, and retinol binding protein
7
to creatinine ratios. Then with continued
8
significant differences favoring Descovy, we moved
9
to evaluate glomerular function with general urine
10
proteins, serum creatinine, and estimated
11
glomerular filtration rate.
12
All participants completed confidential
13
questionnaires on an iPad at the screening visit
14
and at all study visits. These questions inquired
15
about the sexual behavior of each participant since
16
their last visit, including the number of partners,
17
the type of sex, the frequency of sex, condom use
18
habits, and about recent study drug adherence.
19
At each visit, all participants received HIV
20
risk reduction education, adherence support, and
21
condoms and lubricant from site staff. In
22
addition, opt-in/opt-out text messaging could be
A Matter of Record (301) 890-4188
50
1
used to remind the individual to take their study
2
meds daily with the actual words used in the text
3
chosen by sites and participants.
4
Adherence is a critical determinant of PrEP
5
efficacy, so we measured it in multiple ways. We
6
employed two subjective tests, the confidential
7
iPad-based questionnaires and counts from returned
8
pill bottles, both at each visit. We used one
9
objective test, a dried blood spot collection to
10
evaluate TFV diphosphate levels in red blood cells
11
also at each visit. Dried blood spots provided
12
validated analysis of chronic adherence over the
13
8 weeks prior to the collection date, and we looked
14
at a randomly selected subset of 540 participants,
15
about 10 percent of the DISCOVER population.
16
In addition to the randomly selected subset,
17
we also analyzed dried blood spots in a case
18
control analysis of those diagnosed with HIV in
19
DISCOVER with matching controls for each. In our
20
case control study, we compared every individual
21
diagnosed with HIV on study and matched them with
22
5 uninfected controls. The matched controls were
A Matter of Record (301) 890-4188
51
1
specifically chosen to be geographically linked, to
2
have similar time on the study drugs in DISCOVER,
3
and to have comparable sexual exposure as evidenced
4
by the on-study diagnosis of a rectal STI.
5
From the group of uninfected study
6
participants who were a match for each case, 5 were
7
randomly selected. Once all controls were
8
selected, dried blood spot analyses of the TFV
9
diphosphate level in red blood cells were tested on
10
the date of the HIV diagnosis and also on one visit
11
prior.
12
More than 5800 people were screened for
13
DISCOVER; 364 did not meet eligibility criteria,
14
including 49 who tested HIV positive; 5,399 were
15
randomized but 6 in each arm were not treated.
16
leaving 2694 treated in the Descovy arm and 2693
17
treated in the Truvada arm.
18
The full analysis set included 535,335
19
participants who were randomized, treated, and had
20
any post-baseline data. Of those who were
21
randomized and treated in the study, the median age
22
was 34, 12 percent of the population or emerging
A Matter of Record (301) 890-4188
52
1
adults below age 25 and not yet at peak bone mass.
2
In the ratio breakdown, across the 11 North
3
American and European countries, 84 percent
4
self-identified as white and 9 percent as black;
5
25 percent reported being of Hispanic or Latinx
6
ethnicity; 74 participants, or 1 to 2 percent of
7
the population, self-identified as a transgender
8
woman. From responses on the confidential
9
questionnaire, the self-reported sexual orientation
10
was gay or homosexual in 91 to 92 percent, bisexual
11
in 6 to 8 percent, and heterosexual in 1 percent.
12
Baseline sexual behavior data from the
13
confidential questionnaire showed that the
14
treatment population was at significant risk of HIV
15
infection. 58 to 60 percent had at least
16
2 condomless receptive anal sex partners in the 12
17
weeks prior to study entry; 9 to 13 percent
18
reported rectal gonorrhea, rectal chlamydia or
19
syphilis in the 24 weeks before study entry.
20
Two-thirds of DISCOVER participants had used
21
recreational drugs, and nearly a quarter reported
22
binge drinking, defined as 6 or more drinks on at
A Matter of Record (301) 890-4188
53
1
least one occasion and occurring at least monthly.
2
A total of 23 percent had used Truvada for PrEP in
3
the past, and 16 to 17 percent were on it at study
4
entry. While on study, DISCOVER participants
5
maintained this high level of sexual behavior
6
throughout all visits.
7
Participants averaged just under
8
4 condomless receptive anal sex partners at
9
baseline and continuing throughout the study,
10
similar between the arms. They also had high rates
11
of sexually transmitted infections; 57 percent of
12
those on the study were diagnosed with gonorrhea or
13
chlamydia from at least 1 of the 3 anatomic sites
14
tested. And including syphilis, the overall rate
15
on study for any one of these STIs range from 139
16
to one 145 per 100 person-years in DISCOVER.
17
Overall, 42 percent of participants had a
18
rectal STI on the study, most likely due to
19
condomless receptive anal sex, and 16 percent had a
20
urethral STI associated with condomless insertive
21
sex.
22
At the time of the primary endpoint
A Matter of Record (301) 890-4188
54
1
analysis, 16 to 17 percent of participants
2
discontinued study drug in DISCOVER. The most
3
common reasons for discontinuation from study drug
4
were participant decision or lost to follow 6 to
5
7 percent each. Only 1 to 2 percent of study
6
participants discontinued drug due to an adverse
7
event, and the other reasons for discontinuation
8
were less than 1 percent each.
9
As Diana described, the study met its
10
primary efficacy endpoint for noninferiority. In
11
over 8700 person-years on study across the 2 arms,
12
a total of 22 HIV infections were diagnosed; 7 in
13
the Descovy arm, 15 in the Truvada arm,
14
corresponding to HIV incidence rates of 0.16 and
15
0.34 per hundred person-years, respectively. The
16
rate ratio, where 0.16 is divided by 0.34, is 0.47.
17
For the primary endpoint analysis, the rate
18
ratio of 0.47 represents a 53 percent reduction in
19
HIV incidence for the Descovy arm relative to the
20
Truvada arm. The upper bound of the confidence
21
interval around 0.47 is 1.15. This is lower than
22
the 1.62 prespecified noninferiority margin that's
A Matter of Record (301) 890-4188
55
1
establishing the noninferiority of Descovy to
2
Truvada for PrEP.
3
We categorized the 7 diagnoses in the
4
Descovy arm and the 15 in the Truvada arm based on
5
whether or not they occurred prior to study entry.
6
Evaluating all available data and prior to
7
unblinding, a 3-physician panel concluded that 5 of
8
the 22 HIV diagnoses most likely occurred prior to
9
DISCOVER study entry between the screening and the
10
randomization visits. The 5 with suspected
11
baseline infections are shown here in the black
12
section at the bottom of each bar. Just above are
13
the 17 individuals, 6 in Descovy and 11 in Truvada,
14
who acquired HIV while on study.
15
To better understand the impact that the
16
5 suspected baseline infections had on the primary
17
efficacy endpoint, we went on to conduct a
18
sensitivity analysis. By excluding the
19
5 individuals with suspected baseline infection, 1
20
in the Descovy arm, 4 in the Truvada arm, the
21
incidence rate ratio in this sensitivity analysis
22
is 0.55.
A Matter of Record (301) 890-4188
56
1
The confidence interval around it extends to
2
1.48, which is still below the prespecified 1.62
3
noninferiority margin. Therefore, even excluding
4
the suspected baseline infections in the
5
sensitivity analysis, the incidence rate in the
6
Descovy arm remained noninferior to the rate in the
7
Truvada arm.
8
We next looked at efficacy analysis by
9
baseline subgroups. In this forest plot, the HIV
10
incidence rates for the 2 arms are shown again at
11
the top and just left of center. The rate ratio
12
and surrounding 95 percent confidence interval are
13
shown at far right. The rows of the table provide
14
incidence rates for both demographic and baseline
15
risk behavior subgroups.
16
For each of these subgroups, the incidence
17
rates are low and consistent with the rates in the
18
overall study, and the incidence rate ratios
19
demonstrate that the effect of Descovy or Truvada
20
was consistent with the rate ratio in the overall
21
study across all demographic and baseline risk
22
behavior subgroups.
A Matter of Record (301) 890-4188
57
1
In this diagram are genotypic resistance
2
data of the 22 individuals diagnosed with HIV; 19
3
had samples that could be successfully amplified
4
and evaluated. Of these 19, only 4 had Gina
5
genotypic resistance detected to either of the
6
study drugs.
7
All 4 occurred in the Truvada arm. All 4
8
were M184 mutations consistent with resistance to
9
FTC, and all 4 occurred in those with a suspected
10
baseline infection. Each of the 4 individuals with
11
M184 detected were able to be successfully
12
suppressed on ART, 3 with a Descovy-based regimen.
13
Our analysis of subjective adherence
14
measures demonstrates that there was a very high
15
level of adherence across the arms. With
16
self-report from the confidential questionnaires,
17
about 80 percent reported that they took their
18
study meds more than 95 percent of the time across
19
all study visits and similar across the arms. With
20
pill counts from returned bottles of study drug,
21
about 70 percent appeared to be using their study
22
meds more than 95 percent of the time, also similar
A Matter of Record (301) 890-4188
58
1
across the arms.
2
The levels of TFV diphosphate in red blood
3
cells from the subset of dried blood spots tested
4
also demonstrate that there was a high level of
5
adherence in DISCOVER for both study arms. From
6
the nearly 4,000 dried blood spots tested in the
7
random subset, 80 to 90 percent had TFV diphosphate
8
levels in a range consistent with taking 4 or more
9
tablets per week for both arms.
10
In contrast, very few, just 5 to 9 percent
11
at any visit, had TFV diphosphate levels consistent
12
with taking less than 2 tablets per week. In the
13
case control analysis where the 22 HIV cases were
14
compared to HIV uninfected controls, the dried
15
blood spot data analysis there provides a clear
16
explanation for the difference between those with
17
HIV and their matched controls.
18
Low or no adherence was the most significant
19
risk factor associated with HIV in the study for
20
both arms. In the case control study, drug
21
adherence as measured in dried blood spots was
22
significantly lower among those who became infected
A Matter of Record (301) 890-4188
59
1
as compared to matched controls. Most cases had
2
TFV diphosphate levels in red blood cells
3
consistent with using study drug less than 2 doses
4
per week, while more than 90 percent of controls
5
had TFV diphosphate levels consistent with higher
6
levels of adherence.
7
Finally, let's move from the TFV diphosphate
8
levels in red blood cells, which provide us this
9
measure of adherence, over to the levels in PBMCs,
10
which provide a measure of efficacy. The
11
data from dried blood spots showed a high and
12
comparable level of adherence across both arms.
13
The levels of activated drug TFV diphosphate
14
in PBMCs, however, were not the same across the
15
arms. At week 4, once steady state was achieved,
16
the median TFV diphosphate level in PBMCs was
17
6-fold higher in the Descovy relative to the
18
Truvada arm; 404 femtomoles per million cells in
19
Descovy and 61 femtomoles per million cells in
20
Truvada.
21
The amount of activated drug in the PBMCs
22
seen in DISCOVER is consistent with established PK
A Matter of Record (301) 890-4188
60
1
data observed from multiple clinical studies with
2
TAF and TDF-based regimens in chronic HIV
3
treatment. Given that 40 femtomoles per million
4
cells represents the 90 percent effective
5
concentration, or EC90, of TFV diphosphate in PBMCs,
6
98 percent in the Descovy arm were above this EC90,
7
while only 68 percent in the Truvada arm had levels
8
above this mark.
9
In summary, DISCOVER was conducted in MSMs
10
and transgender women with a high baseline risk of
11
HIV infection that was consistent over the course
12
of the study. Over 8700 person-years, the HIV
13
incidence rates were very low and the
14
noninferiority of Descovy to Truvada for HIV
15
prevention was established. Low adherence was the
16
most significant risk factor associated with an HIV
17
diagnosis on study.
18
While M184 mutations occurred in the Truvada
19
arm, there was no resistance to study drugs
20
reported in the Descovy arm. TFV diphosphate
21
levels in PBMCs were over 6-fold higher in the
22
Descovy arm as compared to Truvada with a
A Matter of Record (301) 890-4188
61
1
significantly higher proportion above the EC90 for
2
HIV protection. This PK advantage represents a
3
potential clinical benefit of Descovy for PrEP.
4
Thank you for your attention. I'd now like
5
to turn our presentation over to my colleague,
6
Dr. Moupali Das, who will describe the DISCOVER
7
safety data, and she'll provide a description of
8
the PK bridging data in support of an indication in
9
ciswomen and adolescents.
10
Applicant Presentation - Moupali Das
11
DR. DAS: Good morning, everyone. My name
12
is Moupali Das, and I'm also an infectious disease
13
physician. My career has been devoted to helping
14
end the HIV epidemic by increasing virologic
15
suppression rates and PrEP uptake. For the last
16
six years, I've worked exclusively on clinical
17
trials comparing the efficacy and safety of the two
18
tenofovir prodrugs.
19
The DISCOVER trial is the largest individual
20
trial with a single variable comparison of TAF with
21
TDF. It offers a unique opportunity to compare the
22
safety of TAF with TDF in the absence of underlying
A Matter of Record (301) 890-4188
62
1
HIV or hep B infection and without any accompanying
2
third agents.
3
The safety and tolerability of Descovy and
4
TAF have been thoroughly established in HIV and hep
5
B treatment with over 26,000 person-years of
6
experience in clinical trials and over 1.6 million
7
person-years of clinical experience. Descovy has a
8
superior renal and bone safety profile compared
9
with Truvada due to the 90 percent lower plasma
10
tenofovir levels with TAF compared with TDF. Early
11
favorable changes in renal and bone safety
12
biomarkers correlate with fewer clinical renal and
13
bone adverse events over longer term follow-up.
14
The DISCOVER results are the first
15
demonstration that these well understood renal and
16
bone safety advantages of Descovy compared with
17
Truvada are also true for the HIV uninfected
18
population. There was a meeting exposure of 86 to
19
87 weeks in Descovy and Truvada. The bone mineral
20
substudy had 9 weeks of exposure. Both Descovy
21
and Truvada were safe and well tolerated.
22
The type, frequency, and severity of adverse
A Matter of Record (301) 890-4188
63
1
events were similar between the Descovy and Truvada
2
arms. Most adverse events were grade 1 or 2 in
3
severity. There was a low percentage of study drug
4
related serious adverse events or adverse events
5
leading to discontinuation in both Descovy and
6
Truvada.
7
During treatment, 1 person died in each arm.
8
The most common adverse events in the DISCOVER
9
trial were sexually transmitted infections; 6 of
10
the 9 most common AEs were bacterial sexually
11
transmitted infections or exposure to STIs. This
12
is in contrast with Descovy treatment trials and
13
may reflect increased STI screening in DISCOVER,
14
which happened at every visit, or differences in
15
sexual behavior among DISCOVER participants
16
compared to the treatment trial participants, or a
17
combination of both.
18
I will review the STI data for the next few
19
slides and then come back to the general safety
20
data. The rates of sexually transmitted infections
21
were high and persistent throughout the trial.
22
About 15 percent of participants had lab-diagnosed
A Matter of Record (301) 890-4188
64
1
gonorrhea or chlamydia at any of the 3 anatomic
2
sites at baseline and throughout the study. There
3
were no differences between Descovy and Truvada.
4
Two of the most common AEs were rectal
5
gonorrhea and chlamydia. Approximately 10 percent
6
of participants had rectal gonorrhea or chlamydia
7
at baseline, and this did not change during the
8
study. There were no differences between Descovy
9
and Truvada. The by-visit positivity rates reflect
10
high and persistent sexual behavior over the study
11
with the persistent rectal STI rates reflecting
12
continued high risk for HIV acquisition.
13
The most commonly prescribed medications in
14
the DISCOVER trial were also different from our HIV
15
treatment trials. Four of the seven most commonly
16
prescribed medications are antibiotics used to
17
treat sexually transmitted infections. More than
18
half of the participants received azithromycin or
19
ceftriaxone. There was a high burden of STIs,
20
including rectal STIs diagnosed and treated during
21
the study.
22
Returning back to general safety, the common
A Matter of Record (301) 890-4188
65
1
study drug related adverse events reflect the most
2
common adverse events in the Descovy treatment
3
trials. Twenty percent of participants in Descovy
4
and 23 percent in Truvada had study drug related
5
adverse events. Common related adverse events were
6
low in frequency and similar between arms. The
7
majority were mild GI events and headache.
8
Laboratory abnormalities were also uncommon in the
9
study. Grade 3 or higher lab abnormalities
10
occurred at a low frequency and none were
11
clinically significant.
12
In the HIV treatment trials and in the
13
Truvada adherence subset in the iPrEx trial, the
14
lipid-lowering effect of Truvada has been well
15
documented. In DISCOVER, Truvada was also
16
associated with a reduction in lipid parameters.
17
Total cholesterol, HDL, and LDL cholesterol all
18
declined. The magnitude of these declines is not
19
clinically significant, whereas total cholesterol,
20
LDL, and HDL levels were generally unchanged in
21
participants taking Descovy.
22
Importantly, these changes resulted in no
A Matter of Record (301) 890-4188
66
1
difference in the total cholesterol to HDL ratios
2
between arms, which is strongly associated with
3
cardiovascular risks. While both Descovy and
4
Truvada were safe and well tolerated, Descovy was
5
significantly superior to Truvada on all 6
6
prespecified renal and bone safety endpoints.
7
To assess the renal safety of Descovy
8
compared with Truvada, we reviewed cases of
9
proximal renal tubulopathy, including Fanconi
10
syndrome, as well as all renal adverse events
11
leading to discontinuation.
12
To specifically assess glomerular function,
13
we measured the prespecified renal safety endpoint
14
of serum creatinine and calculated the estimated
15
glomerular filtration rate using the
16
Cockcroft-Gault equation. We also evaluated total
17
urine proteinuria by dipstick and quantitative
18
proteinuria by the urine protein to creatinine
19
ratio or UPCR.
20
To evaluate proximal tubular function, we
21
looked at 2 urine tubular protein to creatinine
22
ratios. In DISCOVER, after 8600 person-years of
A Matter of Record (301) 890-4188
67
1
exposure to study drug, there were no cases of
2
proximal tubulopathy or Fanconi Syndrome on
3
Descovy. There was one case of Fanconi Syndrome on
4
Truvada. There were numerically fewer
5
discontinuations due to renal AEs on Descovy
6
compared to Truvada, 2 versus 6.
7
Descovy had significantly improved
8
glomerular function compared with Truvada. The
9
differences in eGFR with Descovy and Truvada were
10
apparent as early as week 4 and continued through
11
week 48, the prespecified time point for the
12
assessment of secondary safety endpoints. At
13
week 48, Descovy participants also had significant
14
and lower serum creatinine, the prespecified safety
15
endpoint.
16
Glomerular proteinuria was significantly
17
lower in Descovy compared with Truvada. At
18
week 48, 21 percent of participants on Descovy
19
compared with 24 percent on Truvada developed
20
dipstick proteinuria. Fewer participants on
21
Descovy, 1 percent, compared to Truvada, 2 percent,
22
developed clinically significant quantitative
A Matter of Record (301) 890-4188
68
1
proteinuria as defined by the national kidney
2
foundation as a urine protein to creatinine ratio
3
of greater than 200 milligrams per gram.
4
Descovy also had superior outcomes to
5
Truvada in the two markers of proximal tubular
6
proteinuria. Retinal binding protein and beta-2
7
microglobulin are two low molecular weight
8
proteins, which are freely filtered across the
9
glomerularis and reabsorbed at the proximal tubule.
10
Increased spillage of these proteins into the urine
11
is a marker of increased proximal tubular
12
dysfunction and is reflected in higher urine RBP to
13
creatinine and urine beta-2 microglobulin to
14
creatinine ratios.
15
On the left panel, the Truvada group had a
16
20 percent increase from baseline in tubular
17
proteinuria indicating increased proximal tubular
18
dysfunction, while Descovy remained stable. On the
19
right panel, the Truvada group had a 15 percent
20
increase in tubular proteinuria from baseline
21
indicating worsening of tubular function, while the
22
Descovy group had a 10 percent decline or
A Matter of Record (301) 890-4188
69
1
improvement in tubular proteinuria.
2
The superior renal safety of Descovy was
3
also demonstrated in participants who were on
4
Truvada at baseline who switched to Descovy
5
compared to those who remained on Truvada. The
6
DISCOVER trial included participants taking Truvada
7
for PrEP at baseline and did not require a washout
8
of Truvada. There were a large number, 905 people,
9
who were on Truvada at baseline.
10
We prespecified sensitivity analyses of the
11
participants on baseline Truvada for key renal and
12
bone safety endpoints. As in the overall DISCOVER
13
population, those on baseline Truvada who switched
14
to Descovy had improvements in renal function
15
compared to those who remained on Truvada. The
16
improvements in eGFR and those who switched to
17
Descovy were apparent as early as week 4 and
18
persisted through week 48.
19
The improvements with switching to Descovy
20
are also present in markers of proximal tubular
21
function. Those who switch to Descovy had
22
significant declines in tubular proteinuria
A Matter of Record (301) 890-4188
70
1
indicating improved tubular function, while those
2
who remained on Truvada had an 11 percent increase
3
in retinal binding protein to creatinine ratio on
4
the left and a stable beta-2 microglobulin to
5
creatinine ratio on the right.
6
These changes again became apparent as early
7
as week 4 and continued through week 48. Descovy
8
was superior to Truvada on all prespecified
9
biomarkers of renal function. This was
10
demonstrated in both the overall population as well
11
as in the Truvada switchers.
12
We evaluated bone safety with the bone
13
mineral density substudy. The median age of
14
DISCOVER participants was 34, so approximately half
15
of the participants were still building to peak
16
bone mass, which is achieved in the early to mid
17
30s.
18
Descovy participants had a statistically
19
significant increase in mean spine bone mineral
20
density of about 0.5 percent from baseline and
21
stable hip bone mineral density, whereas those on
22
Truvada had statistically significant declines of
A Matter of Record (301) 890-4188
71
1
1 percent of both spine and hip bone mineral
2
density from baseline through week 48. Descovy was
3
statistically superior to Truvada in both
4
prespecified bone endpoints.
5
Using the T scores from the Baseline BMD
6
assessment, participants were classified into the
7
clinically relevant categories of normal bone
8
mineral density, osteopenia, and osteoporosis. At
9
baseline, 27 to 29 percent of participants had
10
either spine osteopenia or osteoporosis in the
11
Descovy and Truvada arms. After 48 weeks,
12
participants on Descovy had significantly less
13
osteopenia and osteoporosis than those on Truvada.
14
As Scott showed you, there were
15
6 prespecified secondary safety endpoints. Descovy
16
was superior to Truvada in all 6 prespecified,
17
alpha-controlled bone and renal safety endpoints at
18
the week 48 endpoint. We continue to follow
19
long-term renal and bone safety in the DISCOVER
20
study participants.
21
Both Descovy and Truvada were safe and well
22
tolerated. The rates of serious adverse events or
A Matter of Record (301) 890-4188
72
1
adverse events leading to discontinuation of study
2
drug were low and balanced between arms. The
3
magnitude of the differences in the early safety
4
endpoints between arms was similar to what is
5
observed in HIV and hep B treatment trials
6
comparing to Descovy to Truvada.
7
This large trial confirmed that the
8
well-established superior renal and bone safety
9
profile of Descovy to Truvada from HIV and hep B
10
treatment is also true in HIV prevention. The
11
safety benefits were seen in both the people
12
starting PrEP for the first time as well as those
13
switching from Truvada to Descovy.
14
This is a significant development from a
15
clinical perspective, so we can now offer a
16
similarly efficacious but safer drug as another
17
choice for HIV uninfected people who are simply at
18
risk for HIV acquisition. The efficacy and safety
19
of Descovy for ciswomen, cismen who have sex with
20
men, and adolescents can be inferred from DISCOVER.
21
The extensive clinical experience with
22
Descovy and Truvada for treatment and prevention
A Matter of Record (301) 890-4188
73
1
allows for the inference of efficacy and safety in
2
ciswomen and adolescents. We have over 15 million
3
person-years of clinical experience with Truvada
4
and TDF and 1.6 million person-years with Descovy
5
and TAF in HIV and hep B treatment.
6
We have over 108,000 person-years in Truvada
7
for PrEP and 6500 person-years for Descovy for PrEP
8
from the DISCOVER trial. Both Truvada and Descovy
9
are highly effective for treatment and prevention.
10
Efficacy is driven by tenofovir diphosphate in
11
peripheral blood mononuclear cells or PBMCs. In
12
contrast, safety is driven by plasma tenofovir.
13
The 90 percent lower plasma levels with Descovy
14
compared with Truvada is associated with an
15
improved bone and renal safety profile.
16
The PK of Descovy or Truvada is independent
17
of intrinsic and in extrinsic factors. This means
18
that the PK of plasma tenofovir and tenofovir
19
diphosphate in PBMCs is not affected by sex at
20
birth, current gender identity, or sexual
21
orientation. HIV infection status also does not
22
affect PK.
A Matter of Record (301) 890-4188
74
1
The active moiety for both Truvada and
2
Descovy associated with both HIV treatment and
3
prevention efficacy is tenofovir diphosphate in
4
PBMCs. Tenofovir diphosphate levels are comparable
5
with Descovy in the MSM and transwomen in DISCOVER
6
on the left and Descovy in ciswomen and cismen.
7
In contrast, the tenofovir diphosphate
8
levels are lower with Truvada on the right.
9
Tenofovir diphosphate levels are 4 to 7-fold higher
10
with Descovy than with Truvada, and this is
11
consistent with findings in prior trials. Efficacy
12
is high in both women and men on Descovy-based
13
regimens for HIV treatment as it is with Truvada.
14
Virologic suppression rates are similar on Descovy
15
and Truvada-based regimens for HIV treatment and
16
similar in women and men.
17
The key metabolite for both Truvada and
18
Descovy associated with safety is plasma tenofovir.
19
Plasma tenofovir with Descovy is similar in women
20
with HIV and in HIV uninfected female volunteers.
21
The PK is independent of HIV status. Plasma
22
tenofovir is 10-fold higher with Truvada shown here
A Matter of Record (301) 890-4188
75
1
in women with HIV on the right. Women with HIV
2
have improved renal safety on Descovy compared with
3
Truvada-containing regimens for HIV treatment.
4
In 519 women who were switched from Truvada
5
to Descovy or remained on Truvada, there were
6
significant improvements in both glomerular
7
function on the left and proximal tubular function
8
on the right through 96 weeks in the women switched
9
to Descovy. These renal improvements are
10
consistent with the DISCOVER results in those on
11
baseline Truvada who switched to Descovy.
12
Women on Truvada-containing regimens who
13
switched to Descovy also had clinically significant
14
improvements in osteopenia and osteoporosis within
15
48 weeks. At baseline, a third of women on
16
Truvada-based regimens for HIV treatment had
17
osteopenia or osteoporosis. Women who switched to
18
Descovy had less spine osteopenia and less
19
osteoporosis at week 48 compared to those who
20
continued on Truvada. These results were
21
statistically significant.
22
Descovy is also an efficacious and safe
A Matter of Record (301) 890-4188
76
1
treatment for HIV in adolescents. Descovy is
2
approved for HIV treatment in adolescents weighing
3
at least 35 kilograms in combination with third
4
agents and in 3 Descovy-containing, single-tablet
5
regimens. Descovy has similar renal and bone
6
safety benefits compared with Truvada in
7
adolescents with HIV. Truvada has been approved
8
for PrEP in adolescents weighing at least
9
35 kilograms since 2018, based on the extrapolation
10
of efficacy from adults.
11
Tenofovir diphosphates and PBMCs is the
12
active moiety associated with both HIV treatment
13
and prevention efficacy. Tenofovir diphosphate
14
levels are similar in adults in DISCOVER and in
15
adults and adolescents with HIV. As we saw before,
16
HIV infection status does not affect PK, so we
17
would expect similarly high levels in PBMCs in
18
adolescents if they were taking it for PrEP.
19
Fifty adolescents who initiated a
20
Descovy-containing regimen for HIV treatment were
21
also evaluated. The mean age in the study was 15
22
years and over half the participants were female.
A Matter of Record (301) 890-4188
77
1
Descovy was highly efficacious in adolescence for
2
HIV treatment. Efficacy was similar in adolescent
3
girls and boys.
4
Plasma tenofovir is associated with safety.
5
With Descovy, the plasma tenofovir levels in adults
6
without HIV and adults and adolescents with HIV are
7
similar across all three populations shown on the
8
left. Plasma tenofovir is 10-fold higher with
9
Truvada in adolescents with HIV, which is
10
consistent with results from prior studies.
11
Truvada has adverse effects on bone mineral
12
density in adolescents on Truvada for PrEP or HIV
13
treatment. Importantly, due to the 90 percent
14
lower plasma tenofovir concentrations with Descovy,
15
there is no such impact on bone growth.
16
Bone mineral density in the two key
17
pediatric metrics of total body less head and spine
18
increased through week 48 with a 0.19 percent
19
increase in total body and at 3.3 percent increase
20
in spine. Adolescent participants continue to
21
build bone mineral density similarly to an age-,
22
sex- and race-matched population.
A Matter of Record (301) 890-4188
78
1
Descovy is noninferior to Truvada in HIV
2
treatment and prevention efficacy. The tenofovir
3
diphosphate levels in PBMCs are comparable in the
4
men and transwomen in DISCOVER, in ciswomen, and in
5
adolescents. Descovy is superior to Truvada in
6
renal and bone safety.
7
Plasma tenofovir is 90 percent lower with
8
Descovy than Truvada and comparably low in
9
DISCOVER, ciswomen, and adolescents. The efficacy
10
and safety of Descovy for PrEP can be inferred for
11
ciswomen and adolescents. Taken together, the
12
comparable exposures and the extensive efficacy and
13
safety data support the purposefully inclusive
14
indication.
15
We have planned multiple effectiveness
16
studies of Descovy for PrEP in a diverse range of
17
populations, including ciswomen and adolescents.
18
We considered numerous approaches to studying the
19
efficacy of Descovy in women in 2015 when we were
20
designing the DISCOVER trial. Ciswomen were not
21
included in discover because the HIV incidence rate
22
in the sites where DISCOVER was conducted is about
A Matter of Record (301) 890-4188
79
1
13-lower in women with high risk for HIV compared
2
with the MSM with high risk in those locations.
3
With respect to doing a dedicated trial in
4
ciswomen, there are three generally accepted
5
approaches to randomized clinical trials for
6
efficacy. A placebo-controlled trial with Descovy
7
versus placebo is not ethical, as Truvada is
8
approved and highly effective for PrEP in adherent
9
women.
10
A superiority trial for Descovy over Truvada
11
was also not reasonable as both are oral daily
12
pills differentiated, primarily, although not
13
exclusively, on safety. Lastly, we considered a
14
noninferiority trial. Unlike DISCOVER where we
15
pulled treatment effects from three randomized
16
control trials with similar efficacy in MSM and
17
transwomen, the 5 randomized controlled trials in
18
women lacked a consistent treatment effect from
19
which we could conduct a defensible noninferiority
20
margin.
21
Using only the two trials with the highest
22
efficacy in women taking Truvada, we were able to
A Matter of Record (301) 890-4188
80
1
estimate that a noninferiority trial would require
2
enrollment of about 22,000 women in the high
3
incidence regions. This would require
4
approximately 8 to 10 years to conduct.
5
The design of the DISCOVER trial was not
6
amenable to the inclusion of adolescents. Truvada
7
was not approved for adolescents until 2018, so we
8
would be comparing this safety and efficacy of two
9
investigational agents. More importantly, we knew
10
from data with Truvada that adolescents require a
11
higher visit frequency to maintain adherence and
12
may benefit from age appropriate targeted
13
interventions to maximize recruitment and retention
14
in clinical trials.
15
The efficacy and safety of Descovy for PrEP
16
in women and adolescents can be inferred from the
17
totality of evidence for Truvada and Descovy for
18
HIV prevention and our extensive safety database
19
from HIV treatment. Clinical data are now needed
20
to inform providers and individuals at risk for HIV
21
regarding the clinical effectiveness of Descovy for
22
PrEP in ciswomen and adolescents.
A Matter of Record (301) 890-4188
81
1
We are dedicated to generating these data,
2
and we will be supporting a number of studies in
3
over 3400 ciswomen and adolescents in the United
4
States and in Africa. Key effectiveness research
5
questions include the evaluation of the safety of
6
Descovy for PrEP in pregnant and breastfeeding
7
women and how the improved safety tolerability and
8
smaller size of Descovy could improve PrEP uptake
9
and persistence.
10
We are strongly committed to understanding
11
how having an additional choice for PrEP with
12
Descovy, which has an improved renal and bone
13
safety profile and pharmacologic properties
14
consistent with an earlier and longer duration of
15
protection from HIV, can help address our shared
16
goals of increasing PrEP uptake and helping to end
17
the HIV epidemic.
18
Thank you. I'm now pleased to invite
19
Dr. Rick Elion to talk about the clinical impact of
20
the DISCOVER trial.
21
Applicant Presentation - Richard Elion
22
DR. ELION: Good morning. My name is
A Matter of Record (301) 890-4188
82
1
Dr. Rick Elion. My conflicts are I do research
2
currently with Gilead, ViiV, and Proteus. I'm a
3
member of an advisory panel for Gilead and ViiV.
4
I'm on the speakers bureau of Gilead, ViiV, and
5
Janssen. I have no stock or financial interest in
6
these proceedings.
7
I have been active in the care of HIV
8
patients since I left residency in 1983. I began
9
in Brooklyn and moved to the East Village in
10
Manhattan in 1985, at the time of the first HIV
11
test in April of that year, and I've been
12
continuously at the front lines of caring for HIV
13
patients and seeking solutions and improvements in
14
care for over 30 years. I've watched countless men
15
and women die in my first 10 years of practice.
16
Advances in treatment and now prevention
17
have transformed what was once a harrowing job to
18
one of immense satisfaction. I'm currently
19
director of research at the Washington Health
20
Institute that serves a low-income population in
21
the District and a clinical professor of medicine
22
at George Washington University. I've been the
A Matter of Record (301) 890-4188
83
1
director of research at Whitman Walker Health,
2
where we were a site for one of the first PrEP
3
demonstration projects, and I have supervised
4
hundreds of patients starting PrEP since 2013.
5
I also work at the Department of Health in
6
Washington, D.C. in the Wellness Program, which is
7
the Center for Caring for Those with Sexually
8
Transmitted Infections and providing PrEP and
9
research and methods to improve HIV prevention in
10
the District. I also continue to follow patients
11
at the Washington Health Institute, some of which I
12
have cared for, for 20 years or longer. I'm
13
grateful and honored to have the chance to share my
14
perspective on why Descovy is an important addition
15
to our prevention toolbox.
16
It's estimated that approximately 90 percent
17
of all new infections are coming from people who
18
either don't know their diagnosis, or they've been
19
diagnosed and are not engaged in care, or not
20
virologically suppressed. While treatment as
21
prevention is very successful at preventing new
22
infections, the bulk of new infections will not be
A Matter of Record (301) 890-4188
84
1
prevented with just treatment alone.
2
We need multiple options for patients to
3
choose their ideal method of prevention, as when
4
consumers have more choices, it leads to greater
5
engagement. We will not likely have a vaccine soon
6
to protect uninfected individuals, so PrEP is
7
critical to help us cut down the rate of new
8
infections.
9
Prevention is much broader than just a pill
10
to protect against HIV. They have a variety of
11
options, which allow each person to choose what's
12
right for them. Choice is critical for patients,
13
as they're much more comfortable and committed to
14
the choices that they make rather than being told
15
what to do.
16
Biomedical interventions including treatment
17
as prevention and PrEP are among the most useful.
18
Treatment as prevention has been fundamental in
19
helping the decline of 18 percent of new cases in
20
the United States. It's the future synergies of
21
these approaches for treatment for HIV infected
22
individuals and PrEP for uninfected individuals.
A Matter of Record (301) 890-4188
85
1
The analysis shown here looked at the change
2
in new diagnoses of HIV over a 5-year period in
3
states grouped by PrEP use. This data is
4
controlled for rates of virologic suppression and
5
in states with low PrEP use as defined as 3 percent
6
on PrEP. You can see about a 1 percent increase in
7
annual HIV diagnosis.
8
This can be compared to the high PrEP use
9
group with 11 percent on PrEP where you see almost
10
a 5 percent reduction in new cases. These rates of
11
PrEP utilization are still quite low considering
12
the risk profiles of patients in these communities
13
and could be greater if there were greater adoption
14
of PrEP.
15
This CDC report makes clear that communities
16
at need are receiving PrEP. This slide
17
demonstrates, however, the imbalance between the
18
potential need for PrEP in certain communities and
19
the actual use of PrEP. Dawn Smith, who's here
20
this morning, and colleagues reported on the
21
disparities between the potential members of
22
various populations that would qualify for PrEP
A Matter of Record (301) 890-4188
86
1
versus those who are using PrEP.
2
These differences are staggering in the
3
communities in need as can be seen in this life.
4
Gross differences exist but only a fraction of
5
these communities are benefiting from PrEP, ranging
6
from racial disparities for blacks and Hispanic to
7
women as well. Though not shown here, certainly
8
adolescents , those from 15 to 25 years of age, who
9
are sexually active are also facing these same
10
disparities.
11
We know that Truvada for PrEP is efficacious
12
in adolescents with adequate adherence. DISCOVER
13
demonstrated efficacy in cismen and transgender
14
women. Descovy pharmacology is consistent across
15
different ages. The PK profile of Descovy could
16
have some advantages for adolescents as reflected
17
by the higher exposures that are achieved and the
18
higher intracellular drug levels that stay elevated
19
longer after missed doses.
20
This notion of forgiveness of suboptimal
21
adherence is critical for this population. Aside
22
from the potential benefits of a better PK profile,
A Matter of Record (301) 890-4188
87
1
Descovy will be a safer medication for a population
2
that's actively building bone mass into their early
3
30's. They are depositing bone as part of their
4
normal growth, and the bone mineral density loss
5
with Truvada could potentially have a lasting
6
effect on adolescents, who after using Truvada may
7
not ever reach peak bone mass.
8
In this slide, we use the notion of a Z
9
score. A Z score compares your bone density to the
10
average values for a person of your same age and
11
gender. A low Z score below 2 is a warning sign
12
that you have less bone mass and/or may be losing
13
bone more rapidly than expected for someone of your
14
age.
15
Consideration of Z scores, which
16
standardized bone mineral density for age, race and
17
sex, is most important during adolescence when bone
18
mineral density variability increases. Z scores
19
are stable over at least three years during periods
20
of rapid bone accrual. Persistent Z score decline,
21
which you can see here, after stopping PrEP,
22
especially in the younger participants, is a
A Matter of Record (301) 890-4188
88
1
concerning finding of these analysis.
2
The Z scores for individuals on Truvada for
3
PrEP declined for both spine, hip, and total body
4
during the 48-week period of PrEP, and then recover
5
in a comparable time period but not back to
6
baseline after at least 48 weeks of observation.
7
The clinical significance of this is
8
two-fold. The first is that the bone mineral
9
density based on the Z score does not recover to
10
baseline 48 weeks after PrEP has been discontinued,
11
suggesting that there is some insult to bone
12
deposition that does not fully recover. Second,
13
the insult is worse for those 15 to 19 who are more
14
actively depositing bone.
15
TAF on the other hand has not been shown to
16
have this impact on bone as reflected in
17
HIV-positive adolescents on treatment. It is a
18
better choice, therefore, for adolescents.
19
Offering them a medication that will allow their
20
bones to grow in a normal fashion is an important
21
consideration in selecting the best medicine for
22
HIV prevention for adolescents.
A Matter of Record (301) 890-4188
89
1
It's equally important for women. I
2
previously mentioned the PK advantage for
3
adolescents. The PK data that were presented
4
earlier reflect a higher intracellular level of TAF
5
in the PBMCs, and hence, a longer period that these
6
levels stay above the threshold of efficacy, 16
7
days versus 10 days. This represents possibly a
8
significant difference between TAF and TDF if we
9
can accept the assumption that PBMCs have higher
10
drug levels and had been correlated with levels of
11
protection. This knowledge about HIV prevention
12
continues to evolve.
13
The DISCOVER study has demonstrated that the
14
failure of TAF to be present in higher tissue
15
levels for men was not a detriment to the overall
16
efficacy. This efficacy was likely driven by the
17
higher drug levels in the various components of the
18
PBMCs. The higher levels and the subsequent longer
19
time, until those levels fall below the threshold
20
of effectiveness, could portend a more forgiving
21
regimen and merit further study.
22
But forgiveness of missed or late doses
A Matter of Record (301) 890-4188
90
1
could be a significant advantage for TAF/FTC and
2
provide a clinical advantage for those who don't
3
take their pills every day. This would be very
4
important for both adolescents and for women who
5
historically have had a greater percentage of
6
failures of PrEP due to poor adherence.
7
PrEP with Truvada has already been approved
8
for women based on the studies seen in this slide
9
that show comparable efficacy to men when adequate
10
adherence is maintained. VOICE and FEM-PrEP show
11
poor efficacy when adherence was equally deficient.
12
Women make up approximately 19 percent of new
13
infections and have a great deal of unsafe exposure
14
through sexual contact.
15
Ninety-three percent of HIV negative women
16
reported having vaginal sex without a condom and
17
26 percent reported having anal sex.. Women
18
therefore make up an important part of the
19
population for controlling HIV infection and are
20
underrepresented in their low use of PrEP. As of
21
2015, only 2 percent of women who were eligible for
22
PrEP were on PrEP.
A Matter of Record (301) 890-4188
91
1
I work with colleagues at the Washington
2
Hospital Center in the Department of Health in the
3
District of Columbia on engaging women in PrEP in
4
our STI program, known as the Wellness Clinic, and
5
the Family Planning Program at Washington Hospital
6
Center. We have screened nearly a thousand women
7
who are at high risk through their sexual
8
practices, and only less than 1 percent have opted
9
to initiate PrEP despite a vigorous program using
10
videos, peer counseling, free medications, and peer
11
support. Despite our numerous efforts to explain
12
and encourage the adoption of more tools for HIV
13
prevention for women, we have been lagging behind.
14
This data of HIV incidence from
15
demonstration projects show the comparable rates of
16
new infections in men and women. There should be
17
little doubt that PrEP with Truvada works equally
18
well in men and women when adherence is
19
appropriate. These low rates of seroconversion
20
demonstrate real-world efficacy. There's no data
21
to suggest that this efficacy is different between
22
genders when adherence is similar.
A Matter of Record (301) 890-4188
92
1
These data from real-world experiences
2
support the fact that women benefit in the same
3
fashion as men with PrEP, with Truvada in the real
4
world, not just in clinical trial settings.
5
However, as mentioned earlier, women have not
6
adopted PrEP in significant number.
7
The reasons for this are complex and involve
8
multiple issues. This study of African Americans,
9
both men and women, demonstrate some of these
10
challenges. Denial of risk is a critical issue for
11
women, as well as fear of side effects and trust
12
that these treatments will actually work.
13
Other surveys have pointed out that mistrust
14
of providers; stigma; fear of being ousted as
15
needing HIV protection; partner notification and
16
lack of support from one's partner; cost; and
17
access to medicine remain key drivers for women's
18
reluctance to start PrEP. There are multiple
19
reasons in the decision-making exercise by women in
20
adopting HIV prevention, so anything we can do to
21
lessen this burden might help women make a better
22
informed decision.
A Matter of Record (301) 890-4188
93
1
The safety of TAF versus TDF regarding bone
2
demineralization is clear. There are clinical
3
consequences as reflected in higher risk of
4
fracture in the treatment population on Truvada.
5
These differences in bone mineral density could
6
potentially be even more important in women.
7
Bone mineral density in women can be
8
impacted by age and hormonal status. Approximately
9
1 in 2 women over the age of 50 will break a bone
10
because of osteoporosis. A woman's risk of
11
breaking a hip is equal to her combined risk of
12
breast, uterine, and ovarian cancer. HIV-positive
13
women have a 50 percent higher risk of osteopenia
14
at age 50 than men at a comparable age.
15
Descovy is a safer alternative for women
16
than Truvada who are at any risk for issues related
17
to bone health. Certainly in treatment settings,
18
I've switched patients on Truvada to TAF-containing
19
regimens to alleviate any concerns I would have or
20
they would have about their bone health.
21
We know that Truvada for PrEP is equally
22
effective in men and women. We know that PBMC
A Matter of Record (301) 890-4188
94
1
levels have been associated with efficacy for both
2
HIV treatment and HIV prevention. There is no
3
correlate of protection established for tissue
4
levels, and only modest efficacy for topical
5
regimens that may provide lower levels in PBMCs, as
6
been mentioned already.
7
The low TFV tissue levels in men were not
8
predictive of the success seen in DISCOVER and
9
supports the notion that PBMC levels provide a
10
correlate of protection. PBMC drug levels are
11
similar for men and women and higher for Descovy in
12
both men and women. And since the PBMC compartment
13
is likely a significant predictor of efficacy, then
14
women would have similar levels of protection as
15
their male counterparts.
16
The key issues at this hearing are based on
17
the evaluation of Descovy for PrEP and whether the
18
indication should extend beyond the participants in
19
the DISCOVER trial. There's a history of
20
conflicting data regarding the role of tissue
21
levels in protecting against HIV acquisition versus
22
the role of systemic protection through components
A Matter of Record (301) 890-4188
95
1
of the blood. This is obviously one of the key
2
questions before the committee today, establishing
3
efficacy and safety in HIV prevention for all
4
communities.
5
The data from the DISCOVER trial
6
demonstrated that systemic protection drives
7
efficacy in MSM and transgender women. There is an
8
established correlate of protection for the levels
9
of drug in PBMCs and iPrEx, and we see higher
10
levels in PBMCs with Descovy and lower levels in
11
rectal tissue. Yet, the point estimate in DISCOVER
12
showed a lower rate of infection on Descovy.
13
We do know that Descovy was safer in men and
14
transgender women, and we do know that there are
15
improvements in safety with TAF versus TD as well,
16
and we can debate whether these improvements will
17
convey a significant clinical benefit. I believe
18
they do convey a significant benefit from a safety
19
perspective for both adolescents and women, as
20
based on the data that has been shown today.
21
The differences in PK between TAF and TDF
22
could be a significant differentiator between the
A Matter of Record (301) 890-4188
96
1
two options for PrEP, and the improved profile seen
2
with TAF might improve outcomes by mitigating
3
suboptimal adherence. This evidence supports the
4
extension of the efficacy data from DISCOVER to
5
women and to adolescents.
6
Further, at the end of the day, as a
7
clinician, counseling men and women about the need
8
for HIV prevention and PrEP, I can't stress enough
9
how detrimental it would be to give men a choice of
10
a safer medicine but not offer the same choice to
11
ciswomen.
12
Ciswomen should have the same options that
13
would be available to cismen and transwomen. They
14
should not have to wait for a separate study to
15
prove efficacy and safety in ciswomen that can take
16
at least four years to result in approval for TAF
17
or PrEP. Please at least allow women to have that
18
choice. Let them decide if the safety advantages
19
outweigh the concerns about efficacy.
20
There may be different opinions about the
21
level of certainty that disclosure will work
22
equally well in both sexes, and there is reasonable
A Matter of Record (301) 890-4188
97
1
certainty, based on the success with Truvada for
2
HIV prevention, to extend this certainty from the
3
male and transgender women in DISCOVER to HIV
4
negative women and adolescents.
5
I am certain that if women and adolescents
6
don't have that choice and are told they can't use
7
a safer medicine that would have potentially been
8
approved for men, it will be seen as a signal of
9
uncertainty by these populations. They are likely
10
to feel left behind, and it will not lead to
11
increased engagement and use of HIV prevention. I
12
hope we don't give those vulnerable populations
13
that message and do allow them the option to choose
14
what will be best for them.
15
I have taught clinicians in Uganda, Kenya,
16
and Rwanda for the last five years, and these
17
clinicians also look to the U.S. and the FDA for
18
recommendations about indications for HIV
19
medications. Providing an effective and safe
20
medication for women in Africa, where women make up
21
over half the cases worldwide, is fundamentally
22
important, so these decisions today have
A Matter of Record (301) 890-4188
98
1
implications not just in the United States,
2
potentially.
3
For the last 30 years as a clinician, my
4
role has been to explain to patients their
5
therapeutic options and help them make the best
6
decisions for themselves. Please allow me to keep
7
doing my job of guiding and helping patients to
8
make the best decision and don't take the choice
9
out of our hands. The totality of the evidence and
10
the favorable benefit-risk profile of Descovy
11
support making this drug available to all those in
12
need. Thank you.
13
DR. BADEN: Thank you. I'd like to thank
14
the applicant for a very thorough presentation of a
15
tremendous amount of data.
16
Before we have clarifying questions to the
17
presenters, we'll take a 10-minute break. Panel
18
members, please remember there should be no
19
discussion of the meeting topic during the break
20
amongst yourselves or with any members of the
21
audience. We will resume at 10:25. Thank you.
22
(Whereupon, at 10:14 a.m., a recess was
A Matter of Record (301) 890-4188
99
1
taken.)
2
Clarifying Questions
3
DR. BADEN: We will now resume session. If
4
everyone can please take your seats, we have little
5
time and much ground to cover. We have about 50
6
minutes for clarifying questions for the
7
applicant's presentation. We may not complete all
8
of the clarifying questions, in which case we will
9
then resume after lunch, after we have a chance to
10
have the agency's presentations.
11
In the clarifying question process, for
12
those of you who are new to joining us, what I'd
13
like to do is to try to build on themes. I really
14
asked for committee members to please use the honor
15
system in how we do the clarifying question
16
process.
17
When we start, if you are interested in
18
asking a question, signal Lauren or I. We'll add
19
you to the list. If a question is asked and there
20
is a follow-on that builds on the theme, I would
21
very much like to build on the theme. Please take
22
your card, turn it on the side, and that will
A Matter of Record (301) 890-4188
100
1
indicate you want to build on the theme so that we
2
can have a series of questions on the same topic
3
and not be bouncing around with every other
4
question, going back and forth between topics.
5
I just asked the committee members to really
6
build on a theme and not have that -- a way so you
7
can ask another question faster. We'll try very
8
hard to get through all of the questions as quickly
9
as possible, but I would like to, as much as
10
possible, build on themes because I think that's
11
more efficient and more effective.
12
We will start with our member on the phone.
13
We'll start with the first clarifying question.
14
MS. LUPOLE: Yes, sir.
15
DR. BADEN: Please go ahead.
16
MS. LUPOLE: The question is, what, if any,
17
are the long-term effects on bone density and renal
18
[indiscernible] related to [indiscernible]
19
adherence and failure?
20
DR. BRAINARD: I'm going to ask Dr. Moupali
21
Das to come to the podium and speak to the longer
22
term bone and renal effects of Truvada and Descovy
A Matter of Record (301) 890-4188
101
1
use. I believe that was your question. We'll
2
start there. If you have an additional question,
3
we'll take it from there.
4
MS. LUPOLE: Yes, ma'am. Basically, the
5
failure to adhere and multiple incidences,
6
occurring multiple incidences.
7
DR. BRAINARD: The failure to what?
8
MS. LUPOLE: [Indiscernible - feedback]
9
DR. BRAINARD: The failure to adhere. Okay.
10
So we'll present the safety data first in
11
terms of the long-term effects, and then we'll talk
12
about adherence and how adherence is related to
13
efficacy.
14
DR. BADEN: And to our colleague on the
15
phone, if you can go on mute, as we're getting
16
feedback. Thank you.
17
DR. DAS: The bone mineral density
18
biomarker, and the renal tubular biomarkers, and
19
the glomerular function biomarkers chosen for
20
evaluation in DISCOVER are associated with
21
clinically meaningful differentiation between
22
Descovy and Truvada over the long term. First
A Matter of Record (301) 890-4188
102
1
we'll look at bone, and then we'll look at renal.
2
Slide 1 up, please. The early declines in
3
bone mineral density between weeks 24 and 48 widen
4
over time through 3 years of follow-up in a
5
representative example of hip BMD from 2 pooled
6
clinical trials of Descovy- and Truvada-based
7
regimen in HIV treatment.
8
On the right, you can see that that
9
separation in the BMD curves is associated with
10
clinically meaningful increase in discontinuations
11
due to bone adverse events. This is over a 3-year
12
time period. We have longer duration of data from
13
the HIV treatment literature.
14
Slide 2 up. This is an analysis from the
15
EuroSIDA cohort with 619 fractures in over 86,000
16
person-years of follow-up. What you see here in
17
the multivariate analysis of fracture risk, which
18
was adjusted for demographics, HIV-specific
19
variables, and comorbidities, is having ever been
20
on Truvada -- excuse me, TDF versus never being on
21
TDF was associated with a 40 percent increase of
22
fracture risk. Being currently on TDF versus never
A Matter of Record (301) 890-4188
103
1
being off of it was associated with a 25 percent
2
risk. You may ask what does this mean for people
3
on PrEP?
4
Slide 3 up. This is data from Chou, et al.,
5
published in JAMA earlier this year in terms of a
6
pooled meta-analysis of PrEP trials that was done
7
to support the U.S. Services Preventative Task
8
Force Recommendation with a grade A for PrEP for
9
HIV prevention. Here we see both TDF trials and
10
Truvada trials. The duration of follow-up in these
11
trials was significantly shorter than what we have
12
in the treatment literature, however, there was an
13
increased trend towards fractures in those
14
participants receiving TDF or Truvada.
15
Now we'll switch to renal discontinuations
16
in renal tubular biomarkers. Slide 1 up. A
17
similar pattern exists for the 2 tubular markers,
18
which are early markers of proximal tubular
19
dysfunction. There are early changes as early as
20
week 4, and the lines separate over time with
21
longer-term follow up through week 48 and week 96.
22
They continue to separate through week 144.
A Matter of Record (301) 890-4188
104
1
On the right-hand side, you can see that
2
there were no discontinuations on people on
3
Descovy-containing regimens, that the cumulative
4
effect of tenofovir toxicity is evidenced by
5
increasing adverse events from renal causes leading
6
to discontinuations in a step-wise fashion through
7
week 48 through week 144.
8
With respect to how this is relevant for
9
people with PrEP, we turn again to Chou, et al's
10
meta-analysis from JAMA earlier this year; slide 3
11
up. Here we see that either TDF or Truvada PrEP is
12
associated with an increased risk of renal AEs. In
13
summary, we chose these biomarkers because we were
14
aware of their association with clinical meaningful
15
differentiation in terms of renal and bone safety
16
over a longer term follow-up, and we continue to
17
follow participants in DISCOVER to follow them long
18
term.
19
DR. BRAINARD: With respect to the adherence
20
question, I'll say that there have been multiple
21
clinical trials, as well as real-world data sets,
22
demonstrating the close correlation between
A Matter of Record (301) 890-4188
105
1
efficacy to Truvada for PrEP and outcome.
2
Slide 1 up, please. This figure
3
demonstrates that across multiple different
4
clinical trials and real-world data sets, the
5
higher the adherence within the study or within the
6
subanalysis within the trial, looking at adherent
7
participants based on plasma tenofovir levels or
8
intracellular drug levels, the higher the efficacy
9
with respect to risk reduction for HIV acquisition.
10
DR. BADEN: If I may, on the
11
bone -- Ms. Lupole, do you have any follow-on
12
questions? If not, we'll have some follow-on in
13
the room.
14
(No response.)
15
DR. BADEN: Dr. Elion on slide 105 showed
16
TDF and adolescent bone development. Do you have
17
similar data for TAF and adolescent bone
18
development?
19
DR. BRAINARD: I'll ask Dr. Moupali Das to
20
review the data we have with TAF-containing
21
regimens in adolescents.
22
DR. DAS: First, I'll show you the data in
A Matter of Record (301) 890-4188
106
1
adolescents with HIV. Slide 1 up. Descovy or TAF
2
does not have the same impact on bone mineral
3
density as does Truvada. Here is bone safety in
4
adolescents with HIV. You can see both that the
5
spine and total body less head continued to
6
increase and grow, and that there are minimal
7
changes in the Z scores, which reflect age, race,
8
and gender-matched populations.
9
In the DISCOVER study, we included people
10
who were 18 and older. The age range was 18 to 76,
11
however, we did look at the bone mineral density in
12
participants stratified by age less than 25.
13
Slide 2 up. This comparative data between
14
Descovy and Truvada in the participants less than
15
25 years on the left with spine and in the middle
16
greater than 25 years with spine. You can see that
17
the participants on Descovy continue to have the
18
same amount of increase in bone mineral density,
19
whereas those on Truvada had significant declines,
20
which is particularly relevant for this population.
21
Similar trends were observed with hip in terms of
22
continued growth on Descovy or stable on Descovy
A Matter of Record (301) 890-4188
107
1
but declines on Truvada.
2
This is consistent with the findings in PrEP
3
that Dr. Elion showed.
4
DR. BADEN: Would these data -- compared to
5
age-matched controls, HIV uninfected, not on a
6
tenofovir compound, is the bone development on TAF
7
equal or is there a difference? Because comparing
8
it to tenofovir TDF, one may accept the decline and
9
say the decline is not as bad. How does it compare
10
to age-matched controls on none of these medicines?
11
DR. DAS: If we go back to slide 1, these
12
are participants with HIV, so there is that
13
consideration. But this is people who are on
14
Truvada, and you're asking about participants on
15
TAF.
16
DR. BADEN: Persons on TAF, and I'm
17
interested in the TAF comparison to bone
18
development in healthy age matched-controlled
19
children not on a tenofovir compound, so that the
20
bone development of a 20 year old on TAF and off
21
TAF is not different.
22
DR. DAS: Okay. In this slide, because we
A Matter of Record (301) 890-4188
108
1
don't have any TAF data in adolescents without
2
HIV -- this is TAF data in adolescents with
3
HIV -- the dotted lines are the Z scores, which are
4
matched by age, race and gender for the population.
5
You can see in the dotted lines between zero and
6
week 48, there's really no difference in the blue
7
dotted-line Z score, which is a spine Z score, and
8
the pink dotted-line Z score, which is total body
9
less head.
10
DR. BADEN: And those would then be
11
age-matched control unaffected?
12
DR. DAS: Yes.
13
DR. BADEN: That's what I thought you said.
14
I just wanted that crystal clear --
15
DR. DAS: I'm sorry I didn't clarify that,
16
yes.
17
DR. BADEN: -- that as best as we can tell,
18
there is no abrogation of normal bone development,
19
as best as you can tell, realizing they're HIV
20
infected; otherwise they wouldn't be on this for a
21
long term.
22
DR. DAS: Exactly. Thank you.
A Matter of Record (301) 890-4188
109
1
DR. BADEN: Thank you.
2
Did Dr. Goetz or Awni have a follow-on
3
question?
4
DR. GOETZ: My follow-up was related to the
5
adherence question rather than bone.
6
DR. BADEN: Please?
7
DR. GOETZ: The previous slide, the
8
backup 1232, showed the relationship between
9
adherence and efficacy. One of the considerations
10
is the relationship between adherence and efficacy
11
on Truvada-containing regimens, or TDF-containing
12
regimens, the same in women as in men. I think
13
that helps inform our thoughts as to where the
14
local tissue concentrations are important.
15
So looking at the VOICE/FEM-PrEP versus
16
iPrEP [ph] populations, I wonder if you can go into
17
more detail as to regards to how levels of
18
adherence, which then presumably correlate with
19
PBMC concentrations, correlate with levels of
20
protection in women versus men.
21
DR. BRAINARD: The levels of adherence
22
measured across these studies varied. They weren't
A Matter of Record (301) 890-4188
110
1
all uniformly assessing adherence through the same
2
mechanism. However, the association between
3
adherence and efficacy by these measures, whether
4
it was tenofovir plasma levels or tenofovir
5
diphosphate within the peripheral blood mononuclear
6
cells, that relationship held up across men and
7
women.
8
These studies shown here represent some of
9
the larger studies conducted to date. But since
10
Truvada for PrEP was approved seven years ago,
11
there's an even larger data set that has
12
accumulated.
13
Slide 1 up, please. The CDC recently
14
updated their website with the new data on efficacy
15
and analyses across all available data and
16
concluded that adherence is highly correlated with
17
outcomes for both men and women and that the
18
efficacy of Truvada for PrEP is estimated to be 99
19
percent for men and for women who are using Truvada
20
for PrEP consistently.
21
DR. GOETZ: If I can follow up on that, then
22
I guess my question goes into people who are
A Matter of Record (301) 890-4188
111
1
partially adherent. To the degree that it's
2
knowable -- these are hard questions -- is partial
3
adherence as effective in women as it is in men?
4
This indirectly gets at the question, is partial
5
adherence, we would think from the pharmacokinetic
6
data would lead to similar concentrations of TFV
7
diphosphate intracellularly and PBMCs, which may
8
not be sufficient.
9
Does it protect women as well as men?
10
Because if we get the same levels in PBMCs, they're
11
at the lower level, so there may be concerns about
12
tissue concentrations. So again, the question is,
13
does partial adherence protect women as equally to
14
men at thresholds? It's certainly going to be a
15
relationship between adherence and success.
16
DR. DAS: There was some recent data
17
presented just a few weeks ago at the IAS
18
conference from the HPTN 082 study, which was a
19
large study conducted in Africa in women. It was
20
looking at different interventions to increase
21
adherence. But in terms of the clinical outcomes
22
of that trial -- put slide 3 up, please -- among
A Matter of Record (301) 890-4188
112
1
the 400 Women age 16 to 25, who were enrolled in
2
this study, there were 4 infections. That gave an
3
overall incidence rate of 1 per 100 person-years.
4
They used dried blood spots, which the
5
methodology from the iPrEx study as well as from
6
DISCOVER trial, and they found that the infections
7
closely correlated with these measurements of
8
adherence using the dried blood spots, which is to
9
say that two of the infections occurred with no
10
detectable drug level and the other two occurred in
11
the setting of adherence consistent with less than
12
2 doses per week. This is suggests that in the
13
setting of low adherence, there is a similar
14
relationship.
15
DR. BADEN: Dr. Giordano?
16
DR. GIORDANO: The adherence question
17
relates to the PBMC question in my mind. The
18
argument is that you achieve high levels of the
19
tenofovir active component in PBMC, and that is a
20
correlate of protection. My understanding of the
21
Anderson et al data is that those data were
22
generated from people who were less adherent to
A Matter of Record (301) 890-4188
113
1
people who were more adherent to the tenofovir
2
drug.
3
So the PBMC data, are they not simply a
4
measure of adherence? And if you measured
5
tenofovir in hair or tenofovir in some other body
6
component, would you not arrive at the same
7
conclusion, that hair is a correlate of protection
8
for HIV prevention? So it gets to the strength of
9
those data, which are critical to this argument
10
that the company's making.
11
DR. BRAINARD: So tenofovir diphosphate,
12
when measured in red blood cells as is done with
13
the dried blood spot analysis, is a measurement of
14
adherence alone and can be analogous to measuring
15
tenofovir in hair levels or measuring plasma
16
tenofovir.
17
The advantage of the tenofovir diphosphate
18
in dried blood cells is that it allows for an
19
integrated assessment of efficacy over a longer
20
period of time, similar to a hemoglobin A1c.
21
Nevertheless, it doesn't speak itself to efficacy,
22
but we know that for tenofovir prodrugs orally
A Matter of Record (301) 890-4188
114
1
administered, the drug is acting within CD-4
2
positive T cells, which are a component of
3
peripheral blood mononuclear cells.
4
We also can draw the correlation between the
5
level of tenofovir diphosphate within the red blood
6
cells and what the corresponding level is within
7
PBMCs based on phase 1 studies in healthy
8
volunteers that Dr. Anderson did to validate that
9
analysis.
10
So the dried blood spot data is adherence
11
data. You can get thresholds of adherence, and
12
then based on the phase 1 studies done, where they
13
were able to match those adherence bands to exactly
14
how many doses were given per week, they can then
15
correlate that to be expected intracellular PBMC
16
levels, and we correlate that with efficacy because
17
that's where we know that the virus replicates. It
18
can only replicate in CD-4 cells
19
DR. GIORDANO: But that correlation with
20
efficacy is still fundamentally based on adherence,
21
as I understand it. How do we know that someone
22
with the exact same adherence -- let me ask it
A Matter of Record (301) 890-4188
115
1
differently. How do we know that the levels in
2
PBMC are the critical determinant of prevention,
3
not of treatment efficacy for someone who has HIV,
4
but of prevention?
That's what I'm not -- you
5
haven't really established that fact in my mind.
6
DR. BRAINARD: So we know that Truvada for
7
PrEP is highly an equally effective in men and
8
women and that adherence is the primary driver of
9
that efficacy. When we look at the vaginal and
10
rectal tissue levels of Truvada -- and I'll try to
11
show you that slide in one minute -- what we can
12
see is that the rectal levels of a tenofovir
13
diphosphate following Truvada use are 100-fold
14
higher than they are in vaginal tissue.
15
If genital tissue or rectal tissue -- and
16
I'll put slide 1 up please. These are the data I
17
just spoke to with the 100-fold higher rectal
18
tenofovir diphosphate levels as compared to vaginal
19
tenofovir diphosphate levels in the setting of
20
Truvada with healthy female volunteers.
21
So if genital tissue levels were driving
22
efficacy, then you wouldn't expect to see equal
A Matter of Record (301) 890-4188
116
1
efficacy in men and women. You'd see
2
disproportionate efficacy, but we don't. We also
3
know from the DISCOVER trial that Descovy is highly
4
effective at preventing HIV acquisition in men who
5
have sex with men and transgender women.
6
I'll put slide 2 up, please. These data add
7
in the Descovy data in the vaginal and rectal
8
compartments. What you can see is that for
9
Descovy, rectal compared to vaginal, rectal levels
10
are 10-fold higher. And within the rectal
11
compartment, Truvada achieves 10-fold higher levels
12
than Descovy.
13
So again, comparing Truvada to Descovy, if
14
rectal levels were driving efficacy, then you would
15
expect Truvada would be better than Descovy, but
16
that's not what we saw in the trial. We saw that
17
they were noninferior, and we saw that adherence
18
was the primary driver of efficacy.
19
So we know that we have these high rectal
20
levels. We know that vaginal levels with Truvada
21
are lower, but we know that Truvada is highly
22
effective in women who take the drug. Therefore,
A Matter of Record (301) 890-4188
117
1
that provides evidence, if you will, that the
2
active tenofovir diphosphate and the circulating
3
PBMCs is driving the efficacy and not the tissue
4
levels within the homogenate of the tissue.
5
DR. BADEN: We have several more follow-on
6
questions. Dr. Awni, did you have a follow on?
7
DR. AWNI: I actually thought it was a
8
follow-on, but related to the size, there were a
9
couple of statements saying the size of the two
10
tablets between the Truvada and Descovy, how much
11
difference in size? Size of the tablet could have
12
an impact on somebody taking it, easy to take it
13
somewhere else.
14
DR. BRAINARD: Yes. Descovy is
15
substantially smaller than Truvada because it's
16
such a smaller dose, 25 milligrams versus 300
17
milligrams.
18
Slide 1 up, please. Obviously not to size,
19
but you get the relative comparison between the
20
size of the two pills there. And size has been
21
cited in patient surveys as a factor that's been
22
seen as favorable.
A Matter of Record (301) 890-4188
118
1
DR. BADEN: Dr. Gripshover, we're still
2
doing follow-ons.
3
DR. GRIPSHOVER: My follow-on is back to the
4
PBMC question. I think the data that we've seen,
5
that you tried to show that it correlated with
6
efficacy, was based on the iPrEx study of men who
7
have sex with men, I think, and then we've seen it
8
in the DISCOVER trial with Descovy.
9
Do we have data from Partners PrEP, where we
10
did see Truvada being effective in women? Do we
11
know that that also works in women from any other
12
trials, or is it just in these two trials?
13
DR. BRAINARD: There are other trials
14
besides Partners PrEP that showed efficacy in
15
women. For example, TDF2 was a study conducted in
16
both men and women, and in the TDF2 study, if you
17
look at the as-treated population, which censors
18
people after they have no longer been taking drug
19
for at least 30 days, then the efficacy in men and
20
in women is comparable.
21
DR. GRIPSHOVER: Actually, my question was
22
do you have the T of the PMBC data in women in all
A Matter of Record (301) 890-4188
119
1
to know that that marker of efficacy works in those
2
trials? That was my question; not did it work.
3
Sorry.
4
DR. BRAINARD: So within Partners PrEP, we
5
had adherence, but I don't believe that there were
6
tenofovir diphosphate levels; it's plasma tenofovir
7
levels. So we can take a look and see if we can
8
find specific data regarding tenofovir diphosphate
9
levels in clinical studies within women.
10
But what I would say, I want to reemphasize
11
that the connection between adherence and the
12
tenofovir diphosphate within PBMCs is made based on
13
the validation of the dried blood spot assay. So
14
you get the dried blood spot assay -- and I might
15
have Dr. Anderson come up and speak to this just so
16
he can walk through how that connection between the
17
dried blood spot tenofovir diphosphate is then
18
related and corresponding to a PBMC level.
19
Because it seems like what we're talking
20
about is tenofovir diphosphate within the CD-4
21
cells, within the PBMCs driving efficacy, but you
22
don't actually measure the PBMCs during the study
A Matter of Record (301) 890-4188
120
1
for adherence; you measure the dried blood spot.
2
Then we know from the validated assays what that
3
level correlates to with respect to tenofovir
4
diphosphate and PBMCs. I'll also say that we know
5
across men, women, HIV infected, HIV uninfected,
6
that tenofovir diphosphate levels are consistent.
7
DR. ANDERSON: Good morning. I have
8
received grants and contracts from Gilead Sciences
9
paid to my institution, as well as some consulting
10
honoraria for my time here, but I do not have a
11
financial interest in the outcome of this meeting
12
or in the company.
13
So I wanted to explain the relationship
14
between DBS concentrations in efficacy as well as
15
PBMC concentrations and efficacy. If I can just
16
have slide 2 up, please.
17
First, I want to start with DBS. DBS was
18
used in the DISCOVER study as an adherence
19
biomarker. What we measured is intracellular
20
tenofovir diphosphate in red cells. And the reason
21
is the half-life in the red cell is 17 to 20 days.
22
That means the concentration will be proportional
A Matter of Record (301) 890-4188
121
1
to the exposure to the drug or adherence, so we'll
2
have a proportional relationship there. It's a
3
wonderful marker for assessing and quantifying
4
adherence.
5
If I could have slide 3 up, please. The way
6
that we operationalize this is dried blood spots
7
come into the lab. We take a punch from that spot,
8
so we normalize the amount that we assay. We then
9
use a validated method. We get a result, the
10
tenofovir diphosphate result. Then we have to
11
understand what that result means.
12
To do that, we conducted separate directly
13
observed dosing PK study in healthy volunteers, one
14
Descovy and one with Truvada, and gave them varying
15
adherence rates. Then we measured their
16
concentrations. And we could tell by then the
17
concentration, what adherence that person -- we
18
made that a standard curve relationship.
19
Then we wanted to know do these bands of
20
adherence relate with efficacy -- if I could have
21
slide 2 up, please -- and they do. These are DBS
22
results from the iPrEx open-label extension that
A Matter of Record (301) 890-4188
122
1
shows HIV incidence on the Y by the dried blood
2
spot level on the X-axis. Then the different
3
adherence bands you can see along the top, less
4
than 2 doses per week on average, 2 to 3 and
5
greater than 4.
6
People that had blood spot levels of greater
7
than 4, there were no infections in that group.
8
People who had blood spot levels of 2 to 3 had an
9
approximately 90 percent reduction in HIV incidence
10
relative to not being on PrEP. So if you hold
11
those dosing categories in mind, this is very
12
similar to what we see in PBMCs.
13
If I could have slide 2 up, please. Just
14
switching your mind now from dried blood spots to
15
PBMC intracellular tenofovir diphosphate, these are
16
the active sites now in peripheral blood
17
mononuclear cells. This is a case control from the
18
iPrEx randomized-controlled trial, and the
19
relationship between drug concentration in PBMC,
20
tenofovir diphosphate in PBMC, and efficacy
21
compared to placebo.
22
Now look at the bands along the top. These
A Matter of Record (301) 890-4188
123
1
are PBMC concentrations from a directly observed
2
dosing study showing 2 doses per week on that as
3
well as 4 or more doses per week. We saw
4
approximately the same efficacy relationship in the
5
PBMCs. Those that were in the roughly 2- to 3-dose
6
range had about a 90 percent reduction. Those in
7
the 4 or more had essentially a hundred percent
8
reduction in HIV incidence.
9
That is the connection that we make through
10
adherence. It's through adherence, and I think
11
that was the point that was brought up earlier, is
12
we're making a connection through adherence. The
13
difference with PBMCs is we know that's the active
14
site.
15
DR. BADEN: Dr. Swaminathan?
16
DR. SWAMINATHAN: So you've sort of drawn a
17
line between --
18
DR. BADEN: Please talk closer to your
19
microphone.
20
DR. SWAMINATHAN: You've sort of drawn a
21
line between the levels from the dried blood spot
22
test and PBMC levels, and the correlation with the
A Matter of Record (301) 890-4188
124
1
efficacy data. I guess the thesis is that because
2
PBMCs or T cells are the sites of replication of
3
the virus, and that TAF has good intracellular
4
levels in PBMCs, that the efficacy would be
5
expected to be as good or better.
6
But the connection between the PBMC levels
7
in previous PrEP trials depends on the
8
pharmacokinetics of Truvada. So whatever the
9
actual target cell that's necessary for effective
10
PrEP is, we know that it correlates with PBMC
11
levels; not that the PBMCs are the actual target
12
that makes PrEP efficacious.
13
How do we know that the correlation between
14
the PBMC levels and cell X is the same with regard
15
to the pharmacokinetics of these two different
16
drugs?
17
DR. BRAINARD: So we know cell X. Cell x is
18
a CD-4 positive T cell because that's the only cell
19
within which HIV will replicate in order to spread
20
infection. Now, that CD-4 positive T cell could be
21
within the tissues or it could be within the
22
peripheral blood mononuclear cells, recognizing
A Matter of Record (301) 890-4188
125
1
that these are not distinct subsets but that there
2
is circulation around the body and trafficking of
3
cells in and out of tissues.
4
I'll put slide 2 up, please, and then CC-15
5
and back, please. Just reviewing, again, how
6
infection is established, infection is initially
7
established, for mucosal transmission, by the
8
infection of a local cell. But in order for
9
infection to disseminate, two different things need
10
to happen. CD-4 T cells need to be recruited to
11
that site of initial infection so that a founder
12
population can be established, and then that
13
founder population needs to disseminate via the
14
lymphatic system.
15
We know from nonhuman primate studies that
16
these CD-4 T cells that form the founder population
17
are coming from the peripheral blood mononuclear
18
cells, and we know that when you're talking about
19
systemic oral drugs, that systemic drugs load both
20
peripheral blood mononuclear cells, but there are
21
also drugs circulating throughout the plasma.
22
Slide 2 up, please. Again, understanding
A Matter of Record (301) 890-4188
126
1
how that infection is occurring, with systemic
2
therapy or with oral tenofovir prodrugs, drug is
3
loaded within these peripheral blood mononuclear
4
cells, which are circulating around the body and
5
trafficking to different locations, including the
6
lymphatic tissue, and including to tissue,
7
particularly when there's a chemokinetic signal.
8
But also TAF, in the case of Descovy and plasma
9
tenofovir in the case of Truvada, are also
10
circulating throughout the blood and distributing
11
throughout the body. As those drugs get into the
12
tissues, they're also able to load resident cells
13
and provide protection that way.
14
So there are really two different ways that
15
systemic oral agents can provide protection against
16
HIV infection. Topical agents can also provide
17
protection, but the way they do that is not by
18
sitting on top of the mucosa but actually diffusing
19
into the subepithelium [ph]. Then in the case of
20
tenofovir gel, for example, they still have to get
21
inside that CD-4 T cell because that's the only
22
place that HIV replicates, and that's independent
A Matter of Record (301) 890-4188
127
1
of female, male, vagina, and rectum.
2
DR. SWAMINATHAN: I agree with most of what
3
you said, but I would just have to disagree with
4
this idea that there's this instantaneous dynamic
5
flux in equilibrium between peripheral blood
6
mononuclear cells and cervical or other submucosal
7
lymphoid populations.
8
The phenotype of resident memory T cells in
9
different tissues has been demonstrated to be
10
different, and there are high CCR5 resident
11
lymphocytes in vaginal tissue, for example, that
12
are different from rectal tissue, and certainly
13
different from PBMCs. Dendritic cells in the
14
vaginal tissue are also thought to play a role.
15
All I'm saying is that unless one actually
16
knows what the pharmacodynamics of these different
17
compounds, intracellularly, in different resident
18
populations, which is the population -- there's not
19
recruitment until after the virus gets there, and
20
there's infection locally of cells to come into the
21
site of infection. So the level of intracellular
22
drug in the lymphocytes that are going to be
A Matter of Record (301) 890-4188
128
1
infected immediately after exposure is what's
2
relevant, and no one has really been able to
3
measure, from what I understand.
4
DR. BRAINARD: I would like to -- oh, sorry.
5
DR. BADEN: Unfortunately, it's 11:07, and
6
we need to go to the agency's presentation. I
7
would have you respond, except this is a longer
8
discussion. So I think, as I anticipated, we
9
almost got through one question. So I will be very
10
interested, as all the committee members are, on
11
further discussion on this point. I think you
12
understand the key issue, and perhaps over lunch,
13
you'll further clarify how to educate us on the
14
rationale.
15
But we need to move to the agency's
16
presentation and clarifying questions to the
17
agency. And if there's time, we'll come back to
18
further clarifying questions to the applicant or
19
we'll do that after lunch.
20
DR. BRAINARD: Great. Thanks very much.
21
DR. BADEN: Thank you.
22
Dr. Miele, thank you for presenting the
A Matter of Record (301) 890-4188
129
1
agency's perspective on some of these key issues.
2
FDA Presentation - Peter Miele
3
DR. MIELE: Good morning. I am Peter Miele,
4
a medical officer in the Division of Antiviral
5
Products. I will be presenting on behalf of the
6
FDA review team for NDA 208215, supplement 12 for
7
Descovy, for pre-exposure prophylaxis or PrEP
8
indication. Here's my agenda.
9
I'll begin with a brief discussion of the
10
indication being proposed by this application, and
11
then move on to some context and background; in
12
particular, a brief discussion of the issues we've
13
been having here with regards to the potential role
14
of mucosal tissue drug concentrations in HIV
15
prevention.
16
I'll summarize the FDA findings from the
17
DISCOVER trial in men and transgender women who
18
have sex with men, and conclude with a discussion
19
of the extrapolation approach that's being proposed
20
in this application to support a PrEP indication in
21
cisgender women and the data that has been
22
submitted to support that approach.
A Matter of Record (301) 890-4188
130
1
As you know, this application proposes a new
2
indication for Descovy or F/TAF, which is
3
pre-exposure prophylaxis to reduce the risk of
4
sexually-acquired HIV-1 in at-risk adults and
5
adolescents weighing at least 35 kilograms.
6
To be clear, this indication would apply to
7
adult and adolescent men and transgender women who
8
have sex with men, men who have sex with women and
9
cisgender women who have sex with men. As such,
10
the proposed indication is similar to the currently
11
approved indication for PrEP for Truvada, which is
12
emtricitabine tenofovir disoproxil fumarate or
13
F/TDF, and the indication is listed for you here.
14
Now, as you know, the data that the agency
15
reviewed to support the PrEP indication for Truvada
16
consisted of data from a phase 3 double-blind,
17
placebo-controlled trial in MSM and transgender
18
women, or the iPrEx trial; as well as phase 3,
19
double-blind, placebo-controlled trial in adult
20
heterosexual men and women in HIV serodiscordant
21
relationships or the Partners PrEP trial; as well
22
as data from a phase 2 open-label trial in
A Matter of Record (301) 890-4188
131
1
adolescent MSM. The Adolescent Trial Network study
2
113.
3
As you've heard already, there are some
4
differences between TAF and TDF. Both drugs have
5
been approved for the treatment of HIV-1 and
6
chronic hepatitis B, but compared to
7
TDF 300 milligrams, oral administration of 25
8
milligrams of TAF results in a 4- to 7-fold higher
9
intracellular level of the active metabolite
10
tenofovir diphosphate in PBMCs, while also
11
resulting in 90 percent lower plasma levels of
12
tenofovir.
13
It's these differences in the plasma
14
exposure to tenofovir that may explain some of the
15
differences in a safety profile observed between
16
TAF and TDF, as if there's less circulating
17
tenofovir in plasma, there's a reduction in the
18
risk of off-target effects of tenofovir.
19
That said, there have been published
20
single-dose PK studies that suggest that
21
25 milligrams of oral TAF achieves lower tenofovir
22
and tenofovir diphosphate levels in rectal and
A Matter of Record (301) 890-4188
132
1
vaginal mucosal tissues as measured in homogenates
2
compared to oral 300 milligrams of TDF.
3
Why is this important? As you've heard in
4
the previous discussions, the relative importance
5
of mucosal tissue versus systemic drug
6
concentrations to PrEP efficacy is unknown.
7
Importantly, the minimum drug concentration in
8
mucosal tissues, if they are relevant, that would
9
be considered protective against HIV-1 infection is
10
also unknown.
11
But we have some indirect observations that
12
might support a role for mucosal tissue drug
13
concentrations in PrEP efficacy. For one, as has
14
been mentioned before, topical microbicide
15
experience suggests that vaginal mucosal tissue,
16
drug concentrations, at high enough levels and with
17
very limited systemic exposure can reduce the risk
18
of HIV-1 infection. And as you've heard, we know
19
that oral TDF dosing results in lower tenofovir
20
diphosphate exposure in vaginal tissue versus
21
rectal tissue.
22
It is this differential drug distribution
A Matter of Record (301) 890-4188
133
1
that has raised concerns that in combination with
2
poor adherence, it may have contributed to the
3
mixed efficacy results observed in PrEP clinical
4
trials of F/TDF in cisgender women as compared with
5
MSM. That's a controversial topic, and that has
6
been greatly debated over the last few years, but
7
the end result is that we're not entirely sure to
8
what extent this differential drug distribution may
9
have on the efficacy; or in other words, is
10
suboptimal adherence less forgiven in women than
11
men?
12
These concerns have practical implications.
13
The CDC PrEP guidelines, for example, acknowledge
14
the lack of scientific consensus on protective
15
contribution of drug exposure in specific body
16
tissues. The CDC addresses this issue by reporting
17
the time to achieve maximum intracellular
18
concentrations of tenofovir diphosphate in the
19
various compartments as based on PK studies. Some
20
state PrEP guidelines have followed suit in
21
recognizing the tissue differential also in
22
discussing the time to achieve protective
A Matter of Record (301) 890-4188
134
1
concentrations.
2
For example, the New York State PrEP
3
guidelines recommend a 7-day lead in of oral PrEP
4
use for protection with receptive anal sex. In
5
contrast, they recommend 20 days of daily PrEP use
6
for protection with receptive vaginal sex. Now,
7
these differences would not be necessary if there
8
was consensus that systemic PK was the prime
9
motivator or the prime driver for PrEP efficacy.
10
As we know, there are no gender differences with
11
respect to systemic PK for TDF.
12
So given that there is a lack of consensus
13
regarding the contribution of local tissue versus
14
systemic drug exposure to PrEP efficacy, and these
15
reports of lowered mucosal tissue tenofovir
16
diphosphate concentrations with oral TAF versus TDF
17
dosing, the agency determined that fully powered
18
clinical trials would be needed to support efficacy
19
of F/TAF for PrEP using F/TDF as the active
20
control.
21
Back to this application, the data that's
22
been submitted to support a PrEP indication for
A Matter of Record (301) 890-4188
135
1
Descovy consists of one phase 3 double-blind active
2
control clinical trial in MSM transgender women or
3
the DISCOVER trial. To support indication in
4
cisgender women and adolescents, an extrapolation
5
approach has been proposed. The FDA presentation
6
will focus on the extrapolation approach in
7
cisgender women.
8
The DISCOVER trial was designed as a
9
double-blind noninferiority trial of 5,000 subjects
10
randomized to F/TAF or F/TDF for at least 96 weeks.
11
Following the day 1 visits, subjects returned for
12
study visits at weeks 4 and 12 and then every
13
12 weeks. And at each follow-up visit, as you
14
heard, they received at risk reduction counseling
15
and adherence counseling, as well as STI screening
16
at all three anatomical sites, oral, rectal, and
17
urine.
18
The primary efficacy endpoint was the
19
incidence of HIV-1 infections per 100 person-years
20
when all subjects had reached a minimum of 48 weeks
21
of follow-up and at least 50 percent had reached
22
96 weeks of follow-up or permanently discontinued
A Matter of Record (301) 890-4188
136
1
from the trial.
2
For the relative risk analysis, a
3
noninferiority margin of 1.62 per 100 person-years
4
was derived based on historical data from three
5
clinical trials of F/TDF for PrEP and MSM, namely
6
the iPrEx, PROUD, and IPERGAY studies. Based on
7
equal weighing of the three trials, an HIV
8
incidence of 1.44 was assumed for the control arm
9
of F/TDF with a confidence interval of 2.64 and
10
9.7. Because the analysis for the DISCOVER trial
11
was a rate ratio, the square root of the lower
12
bound of this confidence interval provided the
13
noninferiority margins. So the square root of 2.64
14
is 1.62.
15
5,399 subjects were randomized in DISCOVER,
16
6 subjects per arm were randomized but not treated,
17
giving us a safety population of 5,387. The full
18
analysis set was used for the primary efficacy
19
analysis, and that consisted of subjects who were
20
randomized and treated HIV negative at baseline and
21
had at least one follow-up HIV test during the
22
trial, and that population was 5,335.
A Matter of Record (301) 890-4188
137
1
In the safety population, the baseline
2
characteristics and demographics were well balanced
3
between the two arms. As you've heard, the median
4
age was 34 years, and 99 percent of subjects were
5
MSM and 1 percent were transgender women; 84
6
percent of subjects were white, black, or mixed
7
black race made up 9 percent of subjects and
8
Hispanics made up 25 percent.
9
At baseline, 16 percent of subjects were
10
using Truvada for PrEP and 44 percent were
11
uncircumcised. The median duration of exposure was
12
86 weeks, and that was balanced between the two
13
arms. And as you've heard, adherence to study drug
14
was high by multiple measures in this trial.
15
For the primary efficacy analysis, a total
16
of 22 HIV infections were reported, 7 in the F/TAF
17
arm for an HIV infection rate of 0.16, and 15 in
18
the F/TDF arm for an HIV infection rate of 0.342.
19
The HIV infection rate ratio was 0.468 with the
20
confidence intervals shown here. Because the upper
21
bound of the confidence interval was below the
22
prespecified NI margin of 1.62, the DISCOVER trial
A Matter of Record (301) 890-4188
138
1
demonstrated noninferiority of F/TAF to F/TDF.
2
As an update, we received a report of an
3
additional HIV infection after the submission was
4
filed, one more HIV infection in the F/TAF group,
5
but this does not impact the primary efficacy
6
conclusion.
7
Data from the DISCOVER indicate that F/TAF
8
provides a similar level of protection as F/TDF
9
against rectal acquisition of HIV, but if we
10
consider other potential routes of HIV transmission
11
in men, such as penile HIV exposure, we do not have
12
any direct evidence to support the efficacy of
13
F/TAF for this relatively low-risk route of
14
transmission.
15
That said, we can assume that insertive sex
16
was occurring in the DISCOVER trial. At study
17
entry, subjects reported a mean of 4 unprotective
18
insertive anal intercourse partners in the 90 days
19
prior to screening, and during the trial, 16
20
percent of subjects had urethritis diagnosed with
21
gonorrhea or chlamydia likely from unprotected
22
insertive anal intercourse.
A Matter of Record (301) 890-4188
139
1
Thus, given the low rates of HIV infection
2
observed overall in the DISCOVER trial, it may be
3
reasonable to assume that men who practice
4
insertive sex were protected.
5
I'm now going to switch gears and talk about
6
safety as observed in the DISCOVER trial. Both
7
F/TAF and F/TDF were safe and well tolerated. We
8
observed no notable differences between the two
9
arms in the types, incidence, severity, or onset of
10
adverse events, or laboratory abnormalities.
11
As you've heard, the most common AEs were
12
sexually transmitted infections. If we exclude the
13
STIs and other infectious adverse events, the most
14
common AEs were diarrhea at 16 percent, nausea at 7
15
percent, headache at 7 percent, and fatigue at 6
16
percent, with comparable rates between the arms.
17
Six percent of subjects in the F/TAF arm
18
were considered serious adverse events and 5
19
percent of subjects in the F/TDF arm had serious
20
adverse events. The majority of these events were
21
not considered related to study drug. We also
22
observed low rates of adverse events leading to
A Matter of Record (301) 890-4188
140
1
drug discontinuation, 1 percent in the F/TAF arm
2
and 2 percent in the F/TDF arm.
3
The most common adverse events leading to
4
drug discontinuation were gastrointestinal
5
disorders, which led to drug withdrawal in less
6
than 1 percent of subjects in each arm. When we
7
looked at GI events overall, they tended to occur
8
in the first month of treatment, which is
9
consistent with the start-up syndrome described in
10
previous trials of F/TDF PrEP.
11
These issues did not seem to have a major
12
impact on body weight, however, there was a mean
13
increase of weight from baseline at week 48 of
14
1.1 kilograms for F/TAF and essentially no change
15
in weight for F/TDF.
16
Looking at renal safety, when we looked at
17
the mean absolute change in serum creatinine, there
18
was minimal change in either group at both weeks 48
19
or 96. The graph on the right shows the mean
20
change in estimated GFR from baseline. The blue
21
line shows the changes in the F/TAF group, which
22
essentially stayed pretty much consistent with
A Matter of Record (301) 890-4188
141
1
baseline, whereas in the F/TDF group, there was a 2
2
to 5 milliliter per minute decrease from baseline
3
over the course of the trial, which became apparent
4
as early as week 4.
5
The distribution of urine protein to
6
creatinine ratio, or UPCR categories, was a key
7
alpha protected safety endpoint in this study.
8
UPCR is generally regarded by the FDA Division of
9
Cardiovascular and Renal Products as a useful
10
laboratory assessment of proteinuria.
11
As shown here, the proportion of subjects
12
who had no significant proteinuria at baseline and
13
who then went on to develop significant proteinuria
14
at week 48 was low in both groups, but was higher
15
in the F/TDF group at 2 percent versus 1 percent in
16
the F/TAF group.
17
Conversely, the proportion of subjects who
18
had significant proteinuria at baseline, of which
19
there were only 25 per arm and who then had
20
improvement in their UPCR category, was higher in
21
the F/TAF group compared to the F/TDF group, at 57
22
versus 44 percent, respectively. These differences
A Matter of Record (301) 890-4188
142
1
were statistically significant at week 48, however,
2
the differences were not significant at week 96.
3
We also observed a greater frequency of
4
treatment emergent proteinuria by urine dipstick in
5
the F/TDF arm overall at 24 percent versus 21
6
percent per F/TAF. Most of these abnormalities,
7
however, were grade 1, and we saw no difference in
8
the frequency of grade 2 proteinuria between the
9
two arms.
10
Likewise, we saw very little differences in
11
the frequencies of graded treatment-emergent
12
laboratory abnormalities as they pertained to serum
13
creatinine, 2 percent for F/TDF and 1 percent for
14
F/TAF overall, and we saw no differences at all
15
between the two groups with respect to
16
hypophosphatemia regardless of severity.
17
Likewise, we saw very little difference
18
between the two groups in the frequency of
19
treatment-emergent adverse events related to renal
20
safety. There was one case of Fanconi syndrome
21
acquired in the F/TDF arm, but also a case of
22
glomerulonephropathy in the F/TAF arm, as well as a
A Matter of Record (301) 890-4188
143
1
case of nephrotic syndrome in the F/TAF arm.
2
The cases in the F/TAF were not considered
3
related to study drug, whereas the Fanconi syndrome
4
in the F/TDF arm was. But when we looked at other
5
adverse events as grouped by their MedDRA high
6
level terms for renal failure and impairment,
7
urinary abnormalities, electrolyte analyses, namely
8
blood phosphorous, decreased renal function, and
9
urinalysis not elsewhere classified, we saw no
10
differences between the groups in the reporting of
11
these adverse events. Likewise, there was very
12
little difference between the two groups in renal
13
adverse events that led to drug discontinuation,
14
although the numbers were very small.
15
In summary, for adverse events or graded
16
treatment-emergent laboratory abnormalities related
17
to renal function or safety, we observed no major
18
differences between the two groups in this
19
particular subject population. I would also remind
20
you that, as with Truvada, approved labeling for
21
Descovy still carries with it a warning for new
22
onset or worsening renal impairment.
A Matter of Record (301) 890-4188
144
1
Moving on to bone safety, mean percent
2
change from baseline at week 48 and hip and spine
3
bone mineral density were also key alpha-protected
4
safety endpoints. As shown on the table here,
5
there were differences between the two groups at
6
both hip and spine and at both weeks 48 and 96,
7
with essentially no great change in the F/TAF arm
8
but decreases of about a mean of 1 percent at each
9
time point, at each site for the F/TDF arm. These
10
differences were statistically significant at both
11
time points.
12
Consistent with other tenofovir-containing
13
product labeling, we conducted a categorical
14
analysis of the percent change in BMD from baseline
15
using the falling cutoffs, 7 percent change from
16
baseline for hip and 5 percent change from baseline
17
for spine, as these are cutoffs that the agency
18
considers clinically meaningful.
19
With regards to the hip, we saw absolutely
20
no difference between the two arms, whether in
21
decreases or increases. We also saw no difference
22
between F/TAF and F/TDF for decreases from baseline
A Matter of Record (301) 890-4188
145
1
in spinal BMD. However, there was a slight or
2
greater proportion of subjects in the F/TAF arm
3
that had a 5 percent or greater increase from
4
baseline in spine BMD at week 48.
5
The applicant has shown you results from a
6
categorical analysis regarding the change in BMD
7
clinical status from baseline to week 48 for the
8
spine. We concur that there was a greater
9
proportion of subjects in the F/TDF arm that had
10
worsening status at week 48 and conversely a
11
greater proportion of subjects in the F/TAF arm
12
that had greater improvement of their BMD status in
13
the hip at week 48.
14
However, when we did the same analysis for
15
the hip, we saw no differences in the proportion of
16
subjects at worsening status at week 48, and there
17
was actually a greater proportion of subjects in
18
the F/TDF arm that had improvement.
19
When we turn to adverse events as reported
20
in the DISCOVER trial, during the course of the
21
trial, we saw no differences between the two groups
22
with respect to fractures, most of which were
A Matter of Record (301) 890-4188
146
1
traumatic and occurred at a relatively low rate of
2
2 percent, or in pathological fractures, as well as
3
in reports of back pain, spinal pain, or bone pain.
4
Likewise, we saw no differences between the two
5
groups with respect to what the investigators
6
themselves reported as bone density decrease, bone
7
loss, osteopenia, osteoporosis, or
8
hypophosphatemia.
9
We looked also at the median change from
10
baseline in fasting serum lipids, and we noticed
11
that there was an overall trend to decrease in
12
fasting cholesterol and LDL in both arms, but the
13
magnitude of the decrease from baseline was greater
14
for F/TDF. We also noted that there was a slight
15
increase from baseline in fasting triglycerides
16
with F/TAF.
17
However, it's important to note that we saw
18
no differences in the median change from baseline
19
for the ratio of total cholesterol to HDL, either
20
within groups or between groups. That said, the
21
F/TAF group had consistently higher incidence of
22
graded laboratory abnormalities related to total
A Matter of Record (301) 890-4188
147
1
cholesterol, LDL, or triglycerides, across all
2
toxicity levels.
3
Lastly, we conducted a categorical analysis
4
of the shifts from baseline based on LDL categories
5
as adapted from the NIH's National Cholesterol
6
Education program. As shown here, we found that a
7
greater proportion of subjects in the F/TAF group
8
had worsening LDL category at week 48 compared to
9
the F/TDF group, 17 versus 10 percent. And
10
conversely, a greater proportion of subjects in the
11
F/TDF group had improvement in the LDL category
12
compared to F/TAF, 40 versus 28 percent,
13
respectively.
14
These findings did not translate into any
15
major differences between the two groups with
16
respect to adverse events, clinical adverse events,
17
such as the cerebrovascular or cardiovascular
18
events, which were very low in the trial anyway.
19
That said, while the proportion of subjects who
20
were on lipid-modifying agents at study entry was
21
balanced between the two arms, a slightly greater
22
proportion of subjects in the F/TAF arm initiated
A Matter of Record (301) 890-4188
148
1
these agents during the study at 2 percent versus
2
1 percent for the F/TDF arm.
3
In summary, again, both F/TAF and F/TDF were
4
both safe and well tolerated. Differences between
5
the groups were observed for various indices,
6
namely changes from baseline in renal biomarkers,
7
bone mineral density on DEXA scans, and fasting
8
serum lipids, consistent with previous trials that
9
compare TAF to TDF. In general, F/TAF and F/TDF
10
had similar adverse event profiles, including low
11
rates of serious adverse events or adverse events
12
leading to drug discontinuation.
13
I'll now turn to our discussion of the
14
indication for PrEP in cisgender women, but before
15
that, we acknowledged that conducting a trial in
16
women for a PrEP indication is challenging. As you
17
know, previous clinical trials in women have
18
demonstrated variable efficacy of oral F/TDF,
19
mostly driven by adherence it seems. As such, FDA
20
recommends superior designs whenever possible for
21
trials in women because determination of a
22
noninferiority margin is not readily feasible.
A Matter of Record (301) 890-4188
149
1
In this application, two extrapolation
2
strategies are proposed. One is the extrapolation
3
of F/TAF efficacy from MSM in the DISCOVER trial to
4
support indication in cisgender women. For this,
5
one must demonstrate comparable systemic exposures
6
between men and cisgender women, including
7
tenofovir and TAF concentrations in plasma, as well
8
as tenofovir diphosphate concentrations in the
9
PBMCs.
10
The second approach is to extrapolate
11
efficacy from F/TDF to support F/TAF in women. And
12
as you've heard, this approach makes use of a
13
published EC90 value of 40 femtomole per million
14
PBMCs as derived from the iPrEx trial of F/TDF in
15
MSM. For this approach, one must demonstrate
16
comparable or higher tenofovir diphosphate
17
concentrations in systemic PK but also in cervical
18
vaginal tissue with TAF relative to TDF.
19
With respect to the first approach, we don't
20
expect that there's going to be any clinically
21
relevant differences in the PK of emtricitabine, or
22
TAF, or PBMC-associated tenofovir diphosphate
A Matter of Record (301) 890-4188
150
1
between men and women. However, for reasons that
2
have been discussed already, matching systemic drug
3
exposures alone may not suffice because of the
4
unknown contribution of mucosal tissue
5
concentrations to PrEP efficacy.
6
For the second approach, where we tried to
7
extrapolate efficacy of F/TDF to support F/TAF,
8
while there's some overlap with the prior approach
9
regarding systemic drug exposures, for this
10
approach, the applicant has cited 40 femtomole of
11
tenofovir diphosphate per million PBMCs as a
12
threshold, or EC90 value, for PrEP efficacy.
13
The two things to consider here, as you've
14
heard, this threshold concentration was associated
15
with adherence to 3 to 4 doses of F/TDF per week,
16
specifically in MSM from the iPrEx trial. Also,
17
this concentration has not been validated as a PK
18
surrogate for tenofovir-based PrEP efficacy for all
19
populations.
20
The concern with relying on this PBMC
21
threshold concentration is that it may not
22
accurately reflect the drug concentrations at the
A Matter of Record (301) 890-4188
151
1
tissue level. For example, we know that with
2
F/TDF, with multiple dosing, we can achieve this
3
threshold concentration of 40 femtomoles per
4
million PBMCs in a matter of days, about 3 days.
5
And we know that this concentration in PBMCs
6
correlates to a rectal tissue concentration that is
7
greater than the 100 femtomoles per milligram or
8
significantly greater than the lower limit of
9
quantification.
10
In contrast, a single dose of F/TAF can
11
reach this PBMC concentration of 40 femtomoles in a
12
matter of hours, but the vaginal tissue
13
concentration can be reported as below the level of
14
quantification. In both scenarios, we have
15
achieved this threshold concentration that's being
16
proposed as a surrogate in PBMCs with diverse
17
results in relevant tissue concentrations.
18
The more conservative approach is to try to
19
match PK, both systemic and tissue, to support an
20
extrapolation of F/TDF efficacy in cisgender women
21
to F/TAF in the same population. For the systemic
22
part of this extrapolation approach, we already
A Matter of Record (301) 890-4188
152
1
know that TAF achieves higher levels of tenofovir
2
diphosphate in PBMCs, so we can check that off.
3
With regards to the tissue concentrations,
4
we know that single-dose TAF or TDF results in
5
concentrations that are mostly below the level of
6
quantification in tissue. What we don't know is
7
whether multiple dosing with TAF or TDF achieves
8
different results at the tissue level. And to that
9
end, data from an external study in healthy female
10
volunteers, study A15-137, was submitted to support
11
this latter part of the extrapolation.
12
This is the study design for study A15-137.
13
It was conducted in two parts, including a single
14
dose and multiple dose part where subjects were
15
treated for 14 days with F/TAF or F/TDF. We're
16
going to focus only on the approved doses for F/TAF
17
and F/TDF.
18
Multiple samples were collected for PK and
19
plasma, PBMC, rectal and cervical vaginal fluid, as
20
well as tissue biopsies, and we're going to focus
21
on the results for the rectal cervical and vaginal
22
tissue biopsies here, and in particular the
A Matter of Record (301) 890-4188
153
1
evaluations for tenofovir diphosphate
2
concentrations.
3
I will also note one thing about this
4
particular study design is that each woman
5
contributed cervical vaginal tissue samples at only
6
one given time point, and that's because tissue
7
samples were collected at different clinical sites
8
at different time points. Rectal tissues were
9
collected 4 hours post dose following 14-day
10
administration. Cervical vaginal tissues were
11
collected at 4 hours post-dose following
12
single-dose administration, as well as 4, 24, and
13
48 hours following 14-day administration.
14
The measurement in this study were tissue
15
homogenates, and assuming a tissue density of
16
1 gram per mL, final sample concentrations in the
17
lower limit of quantitation of 0.3 nanograms per mL
18
were converted to fmol/grams for tenofovir
19
diphosphate.
20
Here are the results that we obtained.
21
Following single-dose administration of F/TAF or
22
F/TDF, 83 percent of vaginal tissue samples were
A Matter of Record (301) 890-4188
154
1
below the lower limit of quantitation, or BLQ, at 4
2
hours. Following multiple doses of F/TAF or F/TDF,
3
a significant proportion of tissue PK samples were
4
also BLQ.
5
In vaginal tissues, tenofovir diphosphate
6
concentrations were higher for oral F/TAF dosing
7
compared to F/TDF only at 4 hours post-dose, but
8
they were mostly unquantifiable at 24 and 48 hours.
9
It is unclear if this isolated finding at 4 hours
10
translates to comparable or higher tenofovir
11
diphosphate concentrations in vaginal tissues
12
beyond 4 hours after multiple dose administration.
13
This table represents the results from the
14
multiple dose part of the study. It's a busy
15
study, so I'll try to walk you through it. If you
16
look at the first row, the 4-hour row and at the
17
column for F/TDF, you'll see that 62 percent of
18
vaginal tissue samples in the F/TDF arm were below
19
the level of quantification. In contrast, none of
20
the tissues in the F/TAF arm were BLQ.
21
Correspondingly, a median tenofovir diphosphate
22
concentration was calculated at 151 femtomoles per
A Matter of Record (301) 890-4188
155
1
milligram.
2
Similar results were seen with cervical
3
tissue biopsies, and as for the rectal tissue, the
4
results confirmed the previous reports that dosing
5
with oral F/TDF results in higher tenofovir
6
diphosphate concentrations in rectal tissue
7
compared to F/TAF dosing.
8
However, for 24 hours and 48 hours, the
9
majority of the tissue samples for the vaginal and
10
cervical tissues were below the level of
11
quantification, and we were not able to determine a
12
median tenofovir diphosphate level for these
13
tissues with any degree of confidence.
14
In conclusion, F/TAF and F/TDF afford
15
similar protection against sexual acquisition of
16
HIV-1 infection in MSM and transgender women at
17
substantial risk. Both F/TAF and F/TDF are safe
18
and well tolerated. F/TAF dosing results in
19
smaller changes or improvements from baseline in
20
biomarkers of proteinuria and bone mineral density
21
compared to F/TDF, but with less favorable lipid
22
changes. No major differences were noted with
A Matter of Record (301) 890-4188
156
1
respect to the side effect profile during the
2
course of this study.
3
However, clinical data regarding the use of
4
F/TAF for PrEP in cisgender women are lacking.
5
Robust tenofovir diphosphate concentration data in
6
the female genital tract are lacking. This
7
application proposes a PrEP indication in cisgender
8
women based on extrapolation of efficacy data via
9
tenofovir diphosphate concentrations and peripheral
10
blood mononuclear cells. However, the relative
11
importance of mucosal tissue versus systemic drug
12
concentrations to PrEP efficacy remains unknown.
13
That concludes my presentation, and I'll
14
take any clarifying questions from the committee.
15
Clarifying Questions
16
DR. BADEN: Thank you.
17
We will now take clarifying questions for
18
the agency's presentation, and I think
19
Dr. Daskalakis has the first question.
20
DR. DASKALAKIS: Peter, thanks for that
21
presentation. Just a question that may also
22
overlap with a question to the sponsor. You state
A Matter of Record (301) 890-4188
157
1
that there's evidence that TAF/FTC is effective in
2
preventing HIV in MSM and transgender women, but
3
we've never actually seen the transgender female
4
data broken out in any way.
5
Do you have a sense of what that really
6
looks like or is that a better question to defer to
7
the sponsor?
8
DR. MIELE: What I can say, and the
9
applicant is free to chime in, first, there was a
10
very small proportion of transgender women
11
enrolled. I believe about 30 percent dropped out
12
early during the course of the trial. None of the
13
HIV infections were seen in the transgender women.
14
Beyond that, I can't really say too much.
15
DR. DASKALAKIS: And a related question, any
16
pharmacokinetic data on tenofovir and TAF versus
17
TDF, versus Truvada, in regards to whether any of
18
those transwomen were using estrogen?
19
DR. MIELE: I will defer to the applicant as
20
to concomitant medications being used by the
21
transgender women in the study. We have seen
22
reports about the effect of PrEP with feminizing
A Matter of Record (301) 890-4188
158
1
hormone therapy, TAF, but no clinical drug
2
interaction studies have been conducted with TAF
3
and feminizing regimens.
4
I don't know if our clinical pharmacology
5
reviewer would like to discuss this topic further.
6
DR. ZHENG: Yes. There's no clinical drug
7
interaction study that's being conducted with TAF
8
and feminizing regimens, CYP-based drug
9
interactions are likely to be minimal. However,
10
some studies suggest that the estrogen can change
11
phosphorylation and the phosphorylation of
12
leukocytes and their analogs.
13
The study conducted with TDF in transgender
14
women receiving feminizing regimens, the paper
15
published recently by Dr. Cottrell, showing minimum
16
changes in plasma tenofovir concentrations for the
17
transgender woman and minimal changes in tenofovir
18
diphosphate concentration in PBMC for rectal tissue
19
as well.
20
The similar dATP to cisgender men, there was
21
significantly higher dATP in rectal tissues as
22
compared to ciswomen. We know that dATP may lower
A Matter of Record (301) 890-4188
159
1
effective concentration of tenofovir diphosphate in
2
rectal tissues. The sample size is very small, so
3
it seems like the study shows that the lower
4
concentration of tenofovir diphosphate is mostly
5
driven by the higher dATP levels in rectal tissue,
6
but we don't have any data for TAF. I don't know
7
if the sponsor has more to add.
8
DR. BADEN: One second, Dr. Daskalakis. I
9
would ask the applicant -- this is clarifying
10
questions to the agency. We realize the applicant
11
has important data in that space. So if you can
12
keep a list of these questions, then we will come
13
back and engage your data set after we clarify with
14
the agency.
15
Do you have further --
16
DR. DASKALAKIS: I'll hold for that.
17
DR. BADEN: No other follow-on. Then,
18
Dr. Green?
19
DR. GREEN: This is not a follow-on.
20
DR. BADEN: Correct.
21
DR. GREEN: Thank you. If we could see your
22
slide 46 again. I just want to make sure I
A Matter of Record (301) 890-4188
160
1
understand it. It was I think the last slide you
2
gave in your presentation, the one that you
3
described as complicated.
4
Now, my question is, simplistically for
5
percent BLQ, you want it to be lower because that's
6
the percentage, below the level that's quantified.
7
It's not really telling us what level is present;
8
it's just whether anything can be quantified or
9
not. And if I'm looking at this correctly -- I
10
just want to make sure I'm reading this
11
correctly -- I see that at 4 hours that F/TAF in
12
the vagina appears to be better than F/TDF, and in
13
cervical tissue, it appears to be better.
14
Then at 24 hours, F/TAF is not as good, but
15
neither one of them are very good in both
16
vaginal -- and it flips because it's better in
17
cervical, but neither one's very good. And at
18
48 hours, basically they all are not good. But
19
we're also demonstrating here that F/TDF doesn't
20
have that level of protection over time either.
21
So if mucosal levels are important, this
22
slide does not demonstrate that because you're not
A Matter of Record (301) 890-4188
161
1
seeing a difference that benefits F/TDF, which
2
already has a drug indication and has been shown to
3
be efficacious. Is that correct?
4
DR. MIELE: I don't know that we can say
5
that one is better than the other. The tissue
6
samples are just not quantifiable, so we can't
7
really say much of anything about that. But I
8
agree with you that the TDF samples were also not
9
showing much. And it may be an issue with the
10
assay. It may be the cutoffs that we used to
11
determine quantifiable.
12
Again, I don't know if the clin/pharm
13
reviewer has any other input here.
14
DR. ZHENG: Backup slide with the LLQ
15
question.
16
DR. HOTAKI: Can you tell me which number it
17
is?
18
DR. ZHENG: Slide 79. Oh, no, slide 80.
19
You can see the published study used
20
compatible units for the lower limit of
21
quantitations, and some are using a femtomole
22
sample or nanogram per mL, which makes it difficult
A Matter of Record (301) 890-4188
162
1
to compare assays sensitivities across studies.
2
In this, A15-137, we have a higher LLQ
3
compared to other study reported. The difference
4
in LLQ can be due to tissue biopsy size, and
5
because it's converted from the nanogram per mL to
6
femtomole per gram, so you have a smaller sample
7
size, it's more possible to have the LLQ values.
8
Also related to the sample storages,
9
stability, and some of those studies may not have
10
the long-term stability data, and also recovery
11
efficiencies, and also the assay sensitivity. So
12
it's probably related to, also, the assay
13
sensitivity, also has other issues.
14
DR. BADEN: Dr. Siberry, a follow-on?
15
DR. SIBERRY: Thanks very much. In thinking
16
about this problem of understanding whether the
17
drug levels in the genital compartment are the
18
actual proxy for protection, have you looked at the
19
trials where the only difference in treatment was
20
TAF and TDF in women to see if there was any
21
difference in plasma genital discordance and
22
suppression?
A Matter of Record (301) 890-4188
163
1
We know that some women who are on
2
suppressive therapy still can have HIV present in
3
the genital tract, and that may be an additional
4
way to get at a differential impact of TAF versus
5
TDF in the genital compartment.
6
Then just a follow-on, can you comment from
7
the agency perspective about how we should view the
8
appropriateness of an application for a drug that
9
intended all the while to have an indication in men
10
and women, coming in with clinical trial data only
11
for men with an expectation to apply to women? I'm
12
just concerned because often women are excluded
13
because of concerns about fetal safety and possible
14
pregnancy and other reasons, and this feels to me
15
like a potentially concerning precedent. Thanks.
16
DR. MIELE: To your first question, we have
17
not looked at that data. You're talking about an
18
HIV treatment. We have not looked at that data, at
19
least I'm not aware if the company has that
20
information, but we have not.
21
Second question, ideally, the agency would
22
like to see clinical trials in the populations for
A Matter of Record (301) 890-4188
164
1
which labeling is going to be indicated. We
2
recognize that conducting trials in women in the
3
current landscape with Truvada approved is
4
challenging. I think our first initial hurdle was
5
to agree on a trial, period; and that ultimately
6
was decided to be conducted in MSM and transgender
7
women.
8
Discussions about a trial in women were had
9
with the applicant. And again, we noticed that
10
there were challenges and difficulties, and the
11
agency itself was struggling with the appropriate
12
study design to recommend for this population. But
13
it was never really agreed upon that this
14
particular application, the way it appeared, would
15
support the indications that are being requested.
16
So at this point we're trying to work with
17
what was submitted to see if we can justify an
18
indication across the populations based on a study
19
that was conducted in one particular population.
20
But going forward, no; we're were not recommending
21
this particular approach. And again, as Dr. Murray
22
noted, this particular case is unique because we're
A Matter of Record (301) 890-4188
165
1
talking about basically two prodrugs or the same
2
drug, and we have an approved drug already for
3
Truvada in all those populations. But ordinarily
4
we would not rely on a single trial in one
5
population to support an indication across multiple
6
populations.
7
DR. BADEN: Dr. Siberry got two questions in
8
there, and I have two follow-ons, one for each of
9
his questions.
10
Given your review of the sum total of the
11
data, what is your impression, or the agency's
12
impression, of a marker of protection in the
13
vaginal compartment? Has that emerged or is that
14
still unclear given the state of the data?
15
DR. MIELE: I think that remains very much
16
unclear. I also want to emphasize that it's not a
17
one or other. It's not necessarily mucosal tissue
18
versus systemic. There may be a contribution of
19
both going on here, and that's the part we don't
20
understand.
21
If vaginal tissue concentrations are
22
relevant, are they acting as the primary line of
A Matter of Record (301) 890-4188
166
1
defense, and is systemic acting as a backup? I
2
think that was a theory that was floated by
3
Dr. Anderson actually. But we don't know at this
4
point. The only thing we can measure are these
5
tissue homogenates. Some studies have looked at
6
mononuclear cells within the vaginal tissues
7
themselves and have had mixed results with respect
8
to differences with the vaginal compartment and the
9
rectal compartment. I think the field itself, at
10
least to us, is a bit mixed or conflicted. So that
11
first question you asked is very much unclear in my
12
opinion.
13
DR. BADEN: Part of the challenge -- and
14
other members of the committee have mentioned this,
15
and I'll ask this of the applicant as well
16
later -- the vaginal compartment, there are
17
menstrual cycle issues, microbiome issues,
18
behavioral issues that are different than other
19
compartments, and it may be adherence or PBMC
20
concentration that may be all that matters, or
21
there may be an interaction with these other
22
factors.
A Matter of Record (301) 890-4188
167
1
Trying to understand from the data available
2
to determine if perhaps adherence is all we need,
3
which is in part what's being suggested, still I am
4
struck by VOICE and the other trials that did not
5
show even results in protection, although there are
6
explanations. It always worries me when there are
7
lots of explanations and were asked to embrace the
8
positive but not worry about the negative, and then
9
assume that it should just work the way we want it
10
to.
11
So I guess my question is, should we just
12
assume the vaginal compartment is an extension of
13
the systemic component, in this setting? And
14
obviously, this will be asked of the applicant
15
later, which is building on the conversation we've
16
been having for an hour or two.
17
DR. MIELE: I think it would be challenging
18
to do that given that we know that the
19
pharmacokinetics are very different between TAF and
20
TDF, and I think part of that difference that we
21
see systemically may be extending to the
22
compartments in question, for various reasons. TDF
A Matter of Record (301) 890-4188
168
1
itself may be cycling in the GI tract and achieve a
2
high protection in the rectal tissue, for example.
3
I don't know that we can confidently say that we
4
can extend the systemic to tell us what's going on
5
in the vaginal tissue.
6
DR. BADEN: My other follow-on I will take
7
in a minute. I will continue to follow on, on this
8
line of questioning.
9
Dr. Ofotokun?
10
DR. OFOTOKUN: Kind of along this line of
11
discussion was the significance of the time for
12
achieving protective concentration. That was
13
different for TDF, for different populations. It
14
was different for men and also different for women
15
from what we saw. And I think the guidelines vary
16
from state and different regions of the country
17
based on this time to achieve a protective
18
concentration for TDF.
19
Do we have a sense of that variability with
20
TAF?
21
DR. MIELE: I'll say this. I think the
22
guidelines are being conservative because of this
A Matter of Record (301) 890-4188
169
1
uncertainty around the role of tissue
2
concentration. I think CDC has presented the data
3
for prescribers to be aware of, and then some state
4
guidelines have pushed that even further into
5
actual prescribing recommendations.
6
Again, I think in the services trying to be
7
the most conservative, for TAF, we don't have any
8
information like that. It really depends on
9
whether you believe that systemic PK is the driver
10
of protection, in which case you probably don't
11
need this lead-in time. But if you believe at all
12
that tissue may be contributing to PrEP efficacy, I
13
don't think we have any data to even help us with
14
what's going on with TAF, at least in the vaginal
15
tissue.
16
DR. BADEN: Dr. Swaminathan?
17
DR. SWAMINATHAN: These are drugs that stop
18
viral replication; they're not disinfectants. So
19
the applicant's very valid points that it's T cells
20
that are the issue, is it really useful to look at
21
drug concentrations of homogenates of biopsies,
22
which are primarily everything but lymphocytes?
A Matter of Record (301) 890-4188
170
1
DR. MIELE: Well, in retrospect, probably
2
not. Going into this, what we had were these
3
single reports out there that were surprising I
4
think to the community that TAF was acting so
5
differently in local tissue compartments. And it's
6
probably what drove us to be conservative and
7
request a clinical trial to begin with.
8
If it had been established that systemic PK
9
were the main driver, we probably didn't need the
10
DISCOVER trial and 5,000 men. But there was a fair
11
amount of uncertainty, as I've tried to describe to
12
you, both in the literature and in the guidelines,
13
so the trial was conducted.
14
Now, I will say this. Granted, rectal
15
tissue concentrations with TAF are lower compared
16
to TDF, and the DISCOVER trial shows comparable
17
efficacy results regardless, but we don't know what
18
the minimum concentration would be. It may be that
19
whatever concentration is being achieved with TAF
20
and rectal tissue may suffice; we don't know.
21
But you're right. I don't know that tissue
22
homogenates is really the best measure to give us
A Matter of Record (301) 890-4188
171
1
valuable information at this point. I don't know.
2
DR. BADEN: Dr. Giordano?
3
DR. GIORDANO: Switching topics a little
4
bit, you mentioned the idea that penile
5
transmission or acquisition seemed reasonably
6
protected against here. Did the sponsor gather
7
data on types of sex? Is there any signal that the
8
acquisition was more likely in men who reported
9
anal receptive versus anal insertive, or was there
10
so much overlap between the two in any single
11
person that you can't distinguish that?
12
DR. MIELE: My impression is that within
13
each individual, there's a variety of sexual
14
practices such that we can't really decipher
15
whether there was a subgroup that was strictly
16
practicing insertive sex. I believe pretty much
17
all of the HIV seroconverters were practicing anal
18
receptive intercourse. But that said, some of them
19
also had reports of insertive sex in there.
20
There were individuals who reported -- for
21
the most part, insertive sex had showed up with
22
rectal STIs. And again, all these reports are
A Matter of Record (301) 890-4188
172
1
self-reported in the patient diaries. Like I said,
2
I don't think we have any direct evidence. Given
3
the low number of HIV infections, and the fact that
4
we presume that a lot of these individuals were
5
practicing insertive sex, that the protection
6
probably did confer to them as well.
7
DR. BADEN: Dr. Daskalakis, a follow-on?
8
DR. DASKALAKIS: Just a brief follow-up,
9
again, for clarification on this issue. If I
10
remember, I think about 60 something percent of the
11
folks in the DISCOVER trial were uncircumcised.
12
Any special circumcision signal
13
with seroconversion?
14
DR. MIELE: No. It was 44 percent.
15
DR. DASKALAKIS: Forty-four; sorry. I knew
16
it was high.
17
DR. MIELE: Yes. No, in terms of baseline
18
characteristics and HIV infection, we didn't see
19
any real correlation.
20
DR. DASKALAKIS: I do remember the
21
confidence interval for outside the U.S. was a lot
22
higher. Is circumcision at all involved in that?
A Matter of Record (301) 890-4188
173
1
DR. MIELE: I have to defer to our
2
statistician if you recall anything.
3
DR. BADEN: Please state your name at the
4
microphone.
5
DR. ZENG: Wen Zeng, statistical reviewer
6
for this NDA. For the subgroup analysis, I think
7
the sponsor already presented. There's no such
8
baseline characters that have a great impact on the
9
final result.
10
DR. BADEN: Follow-on? Not a follow-on, a
11
new topic.
12
Dr. Daskalakis, a new topic?
13
DR. DASKALAKIS: Yes, just a question about
14
a citation that you had in your briefing document
15
is a meta-analysis by Hale et al., that compares
16
TDF, that compares tenofovir, and Truvada versus
17
Descovy. Then subsequently, the safety data
18
presented talks about statistically significant
19
margins of safety.
20
Are any of these, from your perspective,
21
clinically significant?
22
DR. MIELE: I think in the clinical setting
A Matter of Record (301) 890-4188
174
1
probably not, but over the long term they might be.
2
I think the current average use of PrEP at this
3
point is 6 to 12 months.
4
DR. DASKALAKIS: Just again, a quick
5
follow-up on that.
6
DR. MIELE: Anyway, no. We didn't see
7
anything different between the two arms in terms of
8
clinical events.
9
DR. DASKALAKIS: If you stratify the bone
10
and kidney complications or adverse events by age,
11
is there anything that sort of flushes out in terms
12
of just being more common among older adults?
13
Because there are some 60 year olds and 50 year
14
olds in the study.
15
DR. MIELE: There were a small number of
16
older participants. We didn't see any differences
17
come up on either end of the age spectrum.
18
DR. DASKALAKIS: Great. Thank you.
19
DR. BADEN: Dr Goetz?
20
DR. GOETZ: My question relates to the
21
nature of risk in the patient population and the
22
expected rate of HIV in the patient population. I
A Matter of Record (301) 890-4188
175
1
know the study was projected at a rate of infection
2
that was somewhat higher. I wonder if someone from
3
the agency could run through the calculations that
4
predict what the expected rate of HIV infection was
5
in the absence of a prophylaxis.
6
Taking to the extreme, if the study is done
7
in a low-risk patient population, of course, no
8
infections are expected, and the two agents perform
9
similarly. So having confidence of the projections
10
of what the expected rate of infection is, I think
11
is an important consideration.
12
DR. MIELE: Do you mean without PrEP? This
13
wasn't a placebo-controlled trial, so to that end,
14
I think you're asking how would this compare. I
15
think the applicant did do a comparison to local
16
geographic areas in the U.S. based on
17
epidemiological data, looking at concurrent HIV
18
incidence in MSM not on PrEP. I think there were 4
19
to 5 incidents per 100 person-years. At least in
20
the U.S. population of MSM, in the geographical
21
areas where this study was conducted, the incidence
22
was much higher.
A Matter of Record (301) 890-4188
176
1
DR. BADEN: And the STI rate, how does that
2
influence your thinking of being in a high-risk
3
population?
4
DR. MIELE: Dr. Murray has published a meta-
5
analysis looking at various PrEP trials, and trying
6
to correlate the STI rate, at least for rectal
7
gonorrhea and what the predicted HIV incidence
8
would be, as you heard, there was actually a high
9
amount of STIs going on in this trial. And based
10
on the correlations that we've looked at, that
11
should have correlated to an HIV incidence, I
12
believe, of 6; so a much higher incidence.
13
DR. BADEN: Six what? Six of a thousand?
14
DR. MIELE: Six per hundred.
15
DR. BADEN: Okay. So 10-fold higher.
16
DR. MIELE: Much higher.
17
DR. BADEN: If no other follow-ons, then I
18
have another follow-on to Dr. Siberry's earlier
19
comment. The issue of a trial in cisgender women,
20
you mentioned that it was hard to come up with a
21
noninferiority margin. Can you help us understand,
22
does that mean it's not possible or how would you
A Matter of Record (301) 890-4188
177
1
frame a -- could that have been done or what might
2
it look like for us to understand the challenges in
3
a more fully powered trial?
4
DR. MIELE: The challenge is that we have
5
two trials that essentially showed no effect in
6
women, VOICE and FEM-PrEP. Then we have one trial,
7
the Partners PrEP trial. That did show
8
statistically significant protection, the TDF2
9
trial I don't think was powered for efficacy but
10
did show a point estimate that favored efficacy in
11
women.
12
So when you have such divergent results from
13
the historical trials, I think it becomes a
14
challenge to try to come up with an NI margin. I
15
don't know if the statisticians want to discuss
16
this any further, but that is basically -- the main
17
conundrum is that we would not be able to
18
adequately construct an NI margin with such
19
divergent variety in previous trials.
20
DR. BADEN: So in the MSM trans population,
21
where we have consistent results with Truvad, then
22
it's easier to design a trial that shows consistent
A Matter of Record (301) 890-4188
178
1
results.
2
DR. MIELE: Exactly.
3
DR. BADEN: And in cisgender women, where
4
the data are very uneven, it's difficult to have a
5
trial, but we should assume that it should work.
6
I'm just trying to follow the logic that's being
7
put before us.
8
DR. MIELE: I think a strict NI margin, a
9
noninferiority trial, would be difficult to
10
construct. That said, there might be other
11
possible study designs such as comparisons to local
12
HIV incidence in the population of study and the
13
community it studied. These are novel study
14
designs that we're grappling with ourselves in the
15
agency, given that we have a product on the market
16
that is highly effective.
17
DR. BADEN: Dr. Gripshover?
18
DR. GRIPSHOVER: What about a switch study?
19
Is that something that the FDA would consider it
20
would be appropriate? So if we have women who are
21
already taking Truvada for PrEP, would that be a
22
study design that could be considered flipping half
A Matter of Record (301) 890-4188
179
1
to TAF and going forward? You may not have a high
2
incidence rate, but at least you're comparing it to
3
an already approved drug.
4
I'm curious what the agency would think if
5
that's a study design that would work since you
6
don't have a noninferior number.
7
DR. MIELE: Yes. I don't know what the
8
comparison would be in a switch study other than
9
safety. We haven't really considered a switch
10
study for a registrational study.
11
DR. BADEN: Dr. Green?
12
DR. GREEN: This is a direct follow-on. You
13
can't easily come up with a strategy to give a
14
noninferiority study, but there's no reason to
15
presume that if you did a head to head, that it
16
would be superior. And it also seems like it would
17
be unethical to do a placebo study.
18
So you may be telling us that we're back to
19
the argument of the sponsor, that there's no
20
feasible way to assess it, so we have no choice but
21
to make a decision using extrapolation. I don't
22
know if that's your intent to say, but I'm hearing
A Matter of Record (301) 890-4188
180
1
that at least. Tell me why that's not true.
2
DR. MIELE: No. I think you've hit the nail
3
on the head. The question is how confident do we
4
feel that an extrapolation approach would be
5
reasonable to extend the indication. At this
6
point, we haven't really discussed any other
7
alternative study design, so that may still be on
8
the table. But where we are right now is you're
9
right. And it's not a question of whether we think
10
this is appropriate, but whether in the absence of
11
other supporting data, we feel confident that this
12
might work.
13
DR. BADEN: Dr. Ofotokun?
14
DR. OFOTOKUN: I don't seem to buy the
15
argument that we cannot construct an inferiority
16
margin around the data that we currently have for
17
women because we know from looking at the data, and
18
all the four studies in PrEP in women, the reason
19
that those studies, where efficacy was not
20
demonstrated was that way because of poor
21
adherence.
22
So in studies where women took those drugs,
A Matter of Record (301) 890-4188
181
1
the drug was effective, and we can construct a
2
margin around those studied. There's nothing that
3
says that you have to include all the studies that
4
have ever been done in order to construct an
5
inferiority margin in the study design. I remember
6
when the sponsor represented, they had a series of
7
planned studies in women. So how are they planning
8
to do that, if it's going to be impossible to have
9
what is a sample size calculation for studies in
10
women?
11
DR. MIELE: I'll answer your second part
12
first. We have not seen any of these proposals in
13
the agency that the applicant has proposed. We
14
have not seen any protocols. My understanding is
15
that these aren't going to be powered for efficacy
16
comparisons. They may be safety demonstration
17
projects.
18
To your first question, I think I'll defer
19
to our statistician colleague about constructing an
20
NI margin using just a select number of trials.
21
DR. BADEN: Please state your name.
22
DR. VALAPPIL: Yes. My name is Thamban
A Matter of Record (301) 890-4188
182
1
Valappil. I'm team leader for statistics. Based
2
on the noninferiority guidance document that has
3
been published, you need to have a clear evidence
4
of treatment effect historically, meaning
5
that -- especially for this population, there is no
6
treatment effect compared to placebo. Both the
7
studies have failed.
8
So unless you have a measurable treatment
9
effect based on historical trials, you won't be
10
able to construct a noninferiority margin. So the
11
compliance or the adherence cannot be adjusted to
12
be able to look at the margin if the plans have
13
already failed.
14
DR. BADEN: Dr. Smith, you have a follow-on?
15
DR. SMITH: Yes. There's a lot of work
16
going on to develop new agents for PrEP, for
17
pre-exposure prophylaxis, and it's not clear to me
18
whether you're saying that from now on, no studies
19
will be done in women because we can't define the
20
margin, and therefore we can't do a noninferiority
21
trial.
22
The current PrEP is so effective, I don't
A Matter of Record (301) 890-4188
183
1
understand what the implications of this decision
2
are for future trials. Given that, it may be
3
difficult to find a high incidence population of
4
women in the U.S. It's certainly not the case in
5
the developing world, and I think, in fact, in the
6
world at large, women are the largest number of new
7
infections.
8
So the need for effective prevention options
9
for women is even greater than for MSM, although
10
not in this country. So if we're saying from now
11
on that we'll do the studies in men and we'll do
12
some PK studies to extrapolate to women, that
13
doesn't sound like a good scientific approach. So
14
I'm trying to understand the boundaries that you're
15
drawing around this argument here and how that will
16
apply in the future.
17
DR. MIELE: Yes, Dawn, we're not saying that
18
at all. There is a path forward in terms of
19
superiority designs. A lot of new agents that are
20
being developed for PrEP are not necessarily
21
once-a-day pills. The challenge here is we have
22
two drugs that are very similar in terms of their
A Matter of Record (301) 890-4188
184
1
route of administration and their dosage.
2
If you're looking at long-acting agents, for
3
example, you can do a superiority trial, so that's
4
what we've been advocating. Again, like I said,
5
this is a particular circumstance here that is
6
complicated because we have two very similar
7
products. But no, this PK extrapolation that's
8
being proposed is not meant to be precedence
9
setting for future trials in women at all.
10
DR. BADEN: We do need to remember that the
11
business at hand is the current application. There
12
are broader questions that I think are appropriate
13
for us to highlight, as we've been doing, to set
14
the stage for data in the future that are needed to
15
make informed choices. So your points are very
16
well taken. I'm not sure we'll resolve agency
17
policy going forward, but I think the points have
18
been heard.
19
Dr. Goetz, did you have a follow-on?
20
(Dr. Goetz gestures no.)
21
DR. BADEN: Dr. Giordano?
22
DR. GIORDANO: Can you clarify from the
A Matter of Record (301) 890-4188
185
1
agency's perspective how much of a study needs to
2
be done in the U.S. versus abroad to achieve an
3
indication?
4
DR. MIELE: The guidance suggests we can
5
accept clinical data from foreign studies if the
6
sponsor has provided a justification or rationale
7
why that data are applicable to a U.S. population.
8
In this case, 60 percent of the subjects were in
9
the U.S., so I think we're covered.
10
I mean, it's not that preponderance of
11
foreign data here. But for PrEP in general, for
12
example, for women where most of these studies will
13
be conducted ex-U.S., the mechanism of transmission
14
of HIV and the mechanism of action for the drug
15
should be the same regardless of the geographical
16
populations.
17
DR. BADEN: Other questions for the agency
18
about their presentation and their analyses of the
19
data submitted?
20
(No response.)
21
DR. BADEN: If not, it is 12:20, and 12:25
22
is when we're supposed to take a break. I don't
A Matter of Record (301) 890-4188
186
1
think we have enough time to delve into another
2
line of questioning, but to the applicant, I think
3
you've heard issues around, and perhaps after we'll
4
do the open public session, and then we'll come
5
back to clarifying questions to the applicant.
6
Crisp data on the efficacy in trans and
7
crisp data on the insertive male partner, if you
8
have it, were some of the issues raised that I
9
think you have the data, and it would be just great
10
for the committee to see.
11
Comments around the design issue in the
12
cisgender female, which have come up, I think would
13
be very helpful for the committee to hear your
14
thoughts on that. You've touched on them, but I
15
think they're central to our discussion. Then
16
after lunch, we'll have the open public hearing
17
session, and then resume the discussion with the
18
applicant and the agency, but I think the agency
19
has finished their clarifying component.
20
So we will now take a break for lunch.
21
We'll reconvene again in this room at 1:30 sharp.
22
Please take any personal belongings you may want
A Matter of Record (301) 890-4188
187
1
with you at this time. Committee members, please
2
remember that there should be no discussion of the
3
meeting during lunch amongst yourselves, the press,
4
or any member of the audience. Thank you.
5
(Whereupon, at 12:20 p.m., a lunch recess
6
was taken.)
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
A Matter of Record (301) 890-4188
188
1
AFTERNOON SESSION
2
(1:30 p.m.)
3
Open Public Hearing
4
DR. BADEN: It is now 1:30, and we shall
5
resume. This is now the open public hearing part
6
of the meeting.
7
Both the FDA and the public believe in a
8
transparent process for information gathering and
9
decision making. To ensure such transparency at
10
the open public hearing session of the advisory
11
committee meeting, FDA believes that it is
12
important to understand the context of an
13
individual's presentation.
14
For this reason, FDA encourages you, the
15
open public hearing speaker, at the beginning of
16
your written or oral statement to advise the
17
committee of any financial relationship that you
18
may have related to the topic of the meeting.
19
Likewise, FDA encourages you at the
20
beginning of your statement to advise the committee
21
if you do not have any such financial
22
relationships. If you choose not to address this
A Matter of Record (301) 890-4188
189
1
issue of financial relationships at the beginning
2
of your statement, it will not preclude you from
3
speaking.
4
The FDA and this committee place great
5
importance in the open public hearing process. The
6
insights and comments provided can help the agency
7
and this committee in their consideration of the
8
issues before them. That said, in many instances
9
and for many topics, there will be a variety of
10
opinions.
11
One of our goals today is for this open
12
public hearing to be conducted in a fair and open
13
way, where every participant is listened to
14
carefully and treated with dignity, courtesy, and
15
respect. Therefore, please speak only when
16
recognized by the chairperson. Thank you for your
17
cooperation.
18
Will speaker number 1 step up to the podium
19
and introduce yourself? Please state your name and
20
any organization that you're representing for the
21
record.
22
DR. HALL: Christopher Hall, San Francisco
A Matter of Record (301) 890-4188
190
1
AIDS Foundation. Good afternoon. In my capacity
2
as vice president of medical affairs for the San
3
Francisco AIDS Foundation, I oversee the provision
4
of HIV pre-exposure prophylaxis through SFAF sexual
5
health clinic, serving populations at high risk for
6
HIV acquisition. Sorry. My disclosures are listed
7
in the previous slide. Thank you.
8
To date, these programs have prescribed
9
Truvada for PrEP to over 5,080 individuals. I will
10
present our position in support of the proposed
11
supplemental NDA by Gilead Sciences for the
12
fixed-dose combination of emtricitabine and
13
tenofovir alafenamide, which I will hereby refer to
14
as F/TAF for HIV PrEP.
15
We believe that FDA approval of F/TAF as an
16
additional PrEP option will expand the number of
17
individuals who will choose to use PrEP as an HIV
18
prevention method. And furthermore, F/TAF for PrEP
19
will allow centers like the ones we operate to
20
enroll more clients, especially those at higher
21
risk for HIV acquisition.
22
Before I continue, for transparency, I will
A Matter of Record (301) 890-4188
191
1
state that where SFAF has received programmatic
2
support from Gilead, accounting for less than 4
3
percent of its total annual revenues in the last
4
fiscal year, 3 percent of which are programmatic
5
and 1 percent are research-related, this statement
6
is derived in my own professional medical review in
7
the position of the foundation leadership, and has
8
in no way been influenced by Gilead or its staff.
9
SFAF enrolled 59 participants in the
10
DISCOVER trial through May 2017. We understand
11
that early results demonstrate a very low incidence
12
of HIV in both treatment arms, only 7 infections in
13
the F/TAF group and 15 in the Truvada group. Thus,
14
we do not draw the conclusion that F/TAF is better
15
than Truvada but agree that it is shown to be at
16
least as good as Truvada for preventing HIV.
17
In addition, preliminary DISCOVER data
18
suggests better renal and bone outcomes for those
19
participants on F/TAF, that we agree with the
20
interpretation that such improved outcomes are
21
likely marginal in significance. Yet, based on
22
remote and recent understanding of motivators for
A Matter of Record (301) 890-4188
192
1
PrEP engagement, we believe individuals at risk for
2
HIV, including those most at risk such as black and
3
Latino persons, may be more likely to engage with
4
PrEP, as more agents with improved side effect
5
profiles are available for approved use.
6
Additionally, F/TAF offers a PrEP
7
alternative for those who have compromised renal
8
function who cannot use Truvada for PrEP. As 12
9
percent of our PrEP patients are over 50 and thus
10
more likely to present with comorbidities,
11
including preexisting renal compromise or
12
osteopenia, an agent with a marginally better side
13
effect profile may promote engagement, and in fact
14
be meaningfully safer.
15
A strategic priority of the foundation is to
16
center its prep services on communities
17
disproportionately faced with alarming HIV
18
incidence rates, and this includes Black Americans.
19
Our recent efforts enrolling such persons on PrEP
20
have demonstrated the need for new approaches and
21
tools.
22
As black and African Americans face over
A Matter of Record (301) 890-4188
193
1
3 times the rate of kidney failure in the U.S.
2
compared to Caucasians, choice of a PrEP agent with
3
a marginally improved renal safety profile may
4
predispose engagement based on both real and
5
perceived advantages.
6
The approval of F/TAF may also enhance
7
programmatic capacity to provide expanded PrEP
8
services at CBOs like ours. PrEP service delivery
9
is affected by local and/or other structural
10
factors such as depicted hopefully here.
11
Innovation of an express model of STI
12
screening supporting clinical PrEP follow-ups, in
13
2018 facilitated sustained increases in our program
14
capacity and was associated with an approximate 30
15
percent increase in the number of active PrEP
16
patients.
17
Clinical management of PrEP patients using
18
Truvada requires renal function and monitoring,
19
including a baseline check and others every 3 to
20
6 months. In our setting, that includes confirming
21
creatinine elevations with a secondary point of
22
care assay, and in some cases scheduling earlier
A Matter of Record (301) 890-4188
194
1
and/or closer follow-up for those evidencing such
2
elevations. In the last year, for example, 1 in 15
3
clients required additional laboratory tests and/or
4
intensive follow-up while on Truvada.
5
With conventional use of F/TAF as an
6
antiretroviral, a lower threshold of diminished
7
renal function is tolerated before recommended
8
discontinuation. We believe that we can project a
9
more streamlined renal function monitoring
10
algorithm with use of F/TAF for PrEP, and in turn
11
an ability to follow more individuals on PrEP with
12
decreased laboratory expenditures, less intensive
13
lab monitoring, and fewer staff resources dedicated
14
to closer follow-up demanded by present use of
15
Truvada alone.
16
With all other circumstances held unchanged,
17
F/TAF, if approved for use and prescribed for a
18
portion of our PrEP patients, we project a future
19
internal PrEP capacity increased based on these
20
renal monitoring factors alone such that we can
21
enroll and follow an estimated 10 to 15 percent
22
more individuals on PrEP in the first year.
A Matter of Record (301) 890-4188
195
1
Features of U.S. PrEP programs vary, but the
2
foundation believes that the impact of F/TAF's
3
introduction as the second FDA-approved agent
4
indicated for PrEP will lead to meaningfully
5
enhanced capacity to reach more individuals in need
6
of this proven biomedical HIV prevention
7
intervention, especially African American, Latino
8
individuals served by the foundation, and
9
elsewhere. Thank you for your time and attention.
10
DR. BADEN: Thank you. Will speaker number
11
2 step up to the podium and introduce yourself?
12
Please state your name and any organization you're
13
representing for the record.
14
DR. FOX-RAWLINGS: Thank you for the
15
opportunity to speak today on behalf of the
16
National Center for Health Research. I am
17
Dr. Stephanie Fox-Rawlings. Our center analyzes
18
scientific and medical data to provide objective
19
health information to patients, health
20
professionals, and policymakers. We do not accept
21
funding from drug or medical device companies, so I
22
have no conflicts of interest.
A Matter of Record (301) 890-4188
196
1
A new treatment to prevent HIV infection
2
could be beneficial considering the safety concerns
3
of the currently available PrEP treatment.
4
However, Descovy would only provide benefit if it
5
is at least as effective and safe as Truvada for
6
each population for which it is indicated.
7
Otherwise, users could be at an unnecessarily
8
increased risk for HIV.
9
The DISCOVER trial found similar rates of
10
protection against HIV infection among participants
11
taking both drugs. While the trial seems well
12
designed to demonstrate comparable effectiveness of
13
these two drugs, it is still a single trial.
14
Replication is a key to scientific evidence, and
15
independent trials could result in different
16
infection rates due to differences in demographic
17
or treatment profiles of patients or other factors.
18
For example, study participants were more
19
likely to be white, older, and better educated than
20
the general U.S. population that is at risk for
21
HIV, which is the target audience for the drug.
22
While this population may be consistent with the
A Matter of Record (301) 890-4188
197
1
people who are currently using Truvada, there are
2
questions about the generalizability of the data to
3
the whole population who could consider using this
4
drug. It is important to study the general U.S.
5
population that is at risk for HIV.
6
The study also found improvements for
7
biomarkers related to kidney health and bone
8
density, suggesting that this was safer than
9
Truvada, however, this is only relevant if it
10
translates into clinically meaningful difference in
11
the number of adverse events related to kidneys or
12
bone fractures, which were similar in both
13
treatment groups in the clinical trial.
14
The trial suggests that the benefits
15
outweigh the risks for men who have sex with men.
16
However, the benefit-risk ratio was less clear for
17
transgender women. This is due in part to the
18
relatively low number of transgender women in the
19
trial, the high dropout rate, and the lack of
20
subgroup analysis. If FDA is considering approving
21
Descovy for transgender women, then the efficacy
22
and safety of the drug for transgender women should
A Matter of Record (301) 890-4188
198
1
be analyzed.
2
This is especially important given the
3
recent finding that feminizing hormone therapy can
4
interact with PrEP drugs. Similarly, there is
5
insufficient evidence that the drug is effective
6
and safe for PrEP for cisgendered women or
7
adolescents.
8
There are too many unanswered questions
9
regarding the levels of drug achieved, and relevant
10
tissues, and the amount needed in these tissues to
11
consider extrapolation for PrEP for use for
12
cisgendered women and girls.
13
Similarly, the benefits and the risks for
14
adolescent boys differ from that of men and should
15
be considered separately. Clinical trials
16
demonstrating effectiveness and safety for
17
cisgendered women and adolescents are needed if the
18
FDA is considering approval for them.
19
We understand the desire to provide a new
20
PrEP treatment indicated for a broad population,
21
especially when a new treatment may be expected to
22
have fewer risks for kidneys and bone density.
A Matter of Record (301) 890-4188
199
1
However, it is inappropriate and potentially
2
dangerous to approve this drug for subgroups of
3
patients that haven't been adequately studied. The
4
FDA law requires substantial evidence that benefits
5
outweigh the risks for each subpopulation, and the
6
new indication would include. Thank you.
7
DR. BADEN: Thank you. Will speaker number
8
3 step up to the podium and introduce yourself?
9
Please state your name and any organization you're
10
representing for the record.
11
MS. JOHNSON: Thank you so much. My name is
12
Jeremiah Johnson. I'm the HIV project director at
13
Treatment Action Group in New York. We appreciate
14
that this hearing is being held today. Considering
15
how centrally Gilead has controlled this entire
16
process around TAF development, all the way from
17
delaying it for a decade when they purported safety
18
and preventive benefits for this medication; all
19
the way to centrally controlling the DISCOVER trial
20
without adequate participation of community; all
21
the way to rushing us through this regulatory
22
process, we believe it is extremely important that
A Matter of Record (301) 890-4188
200
1
this regulatory agency and the community be given
2
time to have a transparent discussion about this,
3
and for us to take control of this process again,
4
and away from an applicant that has a vested
5
interest in maintaining a $2 billion a year market
6
in biomedical prevention in the U.S.
7
You may be familiar with TAG's work on
8
hepatitis C and tuberculosis as well. That's a
9
little bit about us. Within my 6 remaining
10
minutes, I'm going to go over three main points in
11
a small amount of time, so please pay attention.
12
To start, we're going to be talking about what
13
we've been talking about a lot here today in terms
14
of representation within DISCOVER and within the
15
broader body of evidence that we have as part of
16
this sNDA discussion today.
17
We have a number of concerns about Gilead's
18
active campaign against its own product, Truvada,
19
and what will be generic TDF/FTC PrEP within the
20
next year, and overstatement of efficacy and safety
21
benefits of Descovy compared to that, and have a
22
general discussion about the lack of transparency
A Matter of Record (301) 890-4188
201
1
in this whole process and a rushed process when
2
these involve important discussions that clearly
3
the community was not adequately consulted on early
4
on in the process.
5
I won't go into this slide too much.
6
Obviously, the trial participants within the
7
DISCOVER trial do not represent the broader
8
epidemic that we see in the United States and
9
around the world. With 84 percent of trial
10
participants being white and 99 percent being
11
cisgender men, and only 74 participants identifying
12
as transgender women, clearly we are not seeing a
13
body of evidence that is reflective of all
14
populations that need to be considered in terms of
15
efficacy, safety, and effectiveness for scale up of
16
a new prevention option.
17
Right now, you're hearing here at this
18
podium, and there's a lot of discussion online
19
right now, there's a lot of debate amongst
20
community advocates about what does this data mean
21
and what should we be advocating for considering
22
that we don't have sufficient data? And for no
A Matter of Record (301) 890-4188
202
1
good reason do we have insufficient data.
2
Some of us believe that we have to continue
3
to advocate for an indication for cisgender women
4
because we don't believe that the company sees it
5
as beneficial to their bottom line to do the
6
follow-up efficacy research in order to get an
7
indication for cisgender women. And if we miss out
8
on this opportunity, then there will be PrEPs for
9
different populations, and that's clearly a
10
problem.
11
But at the same time, we're concerned that
12
we're sending a message. Dr. Smith's comment
13
earlier was well taken that we're sending a message
14
that if we don't do adequate research within these
15
populations, that you don't have to do that, and
16
you can get a broader indication anyway, and that's
17
an enormous problem.
18
So it's up to the FDA here today and going
19
forward in this discussion whether you believe that
20
the information and the evidence presented thus far
21
is indicative of a broader indication or a narrow
22
indication. But what must be clear is that
A Matter of Record (301) 890-4188
203
1
cisgender women cannot be left behind; neither can
2
any of the populations that have been left behind
3
in this entire process.
4
If it is not approved with a broad
5
indication, there must be a guarantee and there
6
must be requirements that the company continues to
7
fund efficacy, effectiveness, and safety studies
8
within cisgender women so that they are not left
9
behind in this process.
10
If it is approved, it needs to be contingent
11
upon open-label studies that continue to give us
12
more information for that population. That is
13
essential, and it must happen, and this regulatory
14
body needs to make up for the deficit of actions
15
that took place earlier in this process. We also
16
need to see that for all of the communities that
17
are highly prioritized within our broader epidemic
18
work but were not prioritized within this research.
19
In terms of efficacy, DISCOVER was a
20
noninferiority trial. We are very aware that
21
Gilead is trying to paint Descovy as a superior
22
option in terms of efficacy. That is not borne out
A Matter of Record (301) 890-4188
204
1
in the evidence, and it stands to sabotage scaling
2
up of generic TDF/FTC next year, and to generally
3
sabotage Truvada's scale up for individuals who are
4
already stably avoiding HIV infection on that
5
regimen when there is no medical or efficacy
6
related reason for them to switch over.
7
They're also trying desperately to, even in
8
this room today, boldly assert that it is a safer
9
option when in fact we do not see clinically
10
different outcomes in the DISCOVER trial, and in
11
fact they continue to downplay statistically
12
significant increases in weight and challenging
13
issues around lipids that certainly indicate that
14
it is not necessarily a safer option.
15
It is important that this regulatory body
16
operate with the highest level of scrutiny with the
17
labeling, with the marketing, and with the
18
educational materials that come out of the
19
applicant should an indication be provided for
20
Descovy as PrEP.
21
Just quickly, these are slides that of
22
course Gilead will not be presenting today, but we
A Matter of Record (301) 890-4188
205
1
saw from the DISCOVER trial that there was a
2
statistically significant increase in weight in the
3
TAF arm. And of course in terms of treatment, we
4
are seeing a disproportionate impact on weight
5
within cisgender women and individuals of African
6
descent, which further stresses the need for
7
additional research in those populations
8
considering that the DISCOVER trial was not
9
representative of these populations.
10
Ultimately, we have to ask what is the rush
11
in this entire situation. There's not one
12
peer-reviewed publication that has come out of the
13
DISCOVER trial. There is little transparency.
14
Today, we are just starting to see some of the
15
information from the trial. And all of this is
16
coming after a decade of delaying TAF development.
17
We are frustrated as community members that Gilead
18
continues to centrally control this process, the
19
FDA does not, and that community has not been
20
adequately involved.
21
So going forward, this body needs to send a
22
clear signal that any manufacturer engaging in the
A Matter of Record (301) 890-4188
206
1
field of biomedical prevention research must do a
2
better job. They actually have to adhere to GPP;
3
they actually have to work with community from the
4
start; and they actually have to allow this
5
regulatory body to come up with a robust research
6
protocol that covers all populations and not just
7
their bottom line.
8
So with that, I will close in just saying
9
that we require robust postmarketing research
10
following this discussion today, that the labeling
11
and all materials need to be under high scrutiny,
12
and that we need a clear message from the FDA that
13
this process will go better in the future with
14
future provincial modalities. Thank you.
15
DR. BADEN: Thank you. Will speaker
16
number 4 step up to the podium and introduce
17
yourself. Please state your name and any
18
organization you're representing for the record.
19
MR. KRELLESTEIN: Hello. My name is James
20
Krellestein. I am a cofounder of the PrEP4All
21
collaboration. We are an all-volunteer group of
22
activists who are dedicated to ensuring universal
A Matter of Record (301) 890-4188
207
1
low-cost access to HIV-1 pre-exposure prophylaxis.
2
I just wanted to review, before we
3
begin -- next slide, please -- what Descovy is.
4
Descovy of course is a co-formulation of two
5
different drugs, tenofovir alafenamide fumarate,
6
which was first approved back in 2015 as part of
7
Genvoya, a fixed-dose, single-tablet regimen for
8
HIV treatment, and emtricitabine, which was first
9
FDA approved as Emtriva back in 2002.
10
One of the things that has been talked about
11
extensively today is that tenofovir alafenamide has
12
some alleged advantages over tenofovir disoproxil,
13
which is that the prodrug catabolism occurs
14
intracellularly rather than tenofovir disoproxil,
15
which is actually metabolized primarily
16
systemically, allowing lower plasmic exposures to
17
tenofovir but higher levels of tenofovir
18
diphosphate, the pharmacologically active
19
anabolite, compared to tenofovir disoproxil, at
20
least in PBMCs. This is alleged to convey certain
21
safety benefits compared to TDF.
22
I think that one of the things that's really
A Matter of Record (301) 890-4188
208
1
important, though, to realize in this entire
2
process is that TAF is not a new drug despite being
3
FDA approved in 2015. In fact, Gilead first filed
4
a patent application for TAF, claiming priority all
5
the way back to 2000, using of course its code name
6
at that time as GS7340. And of course, Gilead
7
scientists published a peer-reviewed scientific
8
journal in nucleosides, nucleotides, and nucleic
9
acids, all the way back in 2001 regarding the
10
metabolism of GS7340 now known as TAF.
11
I think the question that we should all be
12
asking ourselves is two things. First of all, why
13
in 2019 are we discussing an FDA application for
14
F/TAF as PrEP? And number two, why don't we have
15
better data on cisgender women?
16
Had F/TAF actually been developed when it
17
was supposed to be developed, we would have had an
18
F/TAF arm in iPrEx. We would have had an F/TAF arm
19
in Partners PrEP. We would have high-quality,
20
randomized-controlled evidence in all populations
21
that are being sought in indication for today.
22
So let's go through the development of TAF,
A Matter of Record (301) 890-4188
209
1
if you would. Gilead began early phase 1 and phase
2
2 trials back in 2001 and 2002 and filed an IND
3
with this agency for the development of TAF back in
4
January of 2002. But in October 23, 2004, they,
5
based on an internal business review, discontinued
6
development of TAF, only on October of 2010 to
7
restart development of TAF.
8
What was the reason for this stop-start
9
approach to drug development? You don't have to
10
look to me, you don't have to look to Jeremiah, and
11
you don't have to look to anyone in this room to
12
actually understand what Gilead was doing. You can
13
look to their former CEO, Dr. John Milligan, who
14
stated that "one of the reasons why we were
15
concerned about developing TAF was we were trying
16
to launch Truvada versus Epzicom at the time. And
17
to have our own study suggesting that TDF wasn't
18
the safest thing on the market, which it certainly
19
was at the time, it didn't seem like the best. It
20
didn't seem like we would have a mixed message."
21
That was in 2011.
22
So as someone who takes TDF every single
A Matter of Record (301) 890-4188
210
1
day, I have to say I'm highly disturbed by the
2
applicant's behavior to basically delay a drug that
3
they knew was going to present at least some safety
4
benefits compared to TDF, to protect their bottom
5
line rather than to protect the public health, and
6
I only wish that the FDA would share that very same
7
feeling. But instead, the FDA rewards Gilead's
8
decision to delay.
9
First of all, I was quite surprised by Dr
10
Murray's statement that TAF is similar to TDF, or
11
at least that it has the same active moiety as TDF
12
considering that the FDA had granted TAF new
13
chemical entity exclusivity in 2015, which prevents
14
any challenges through the patent paragraph 4
15
process until this year. It also recommended that
16
the USPTO give the maximum patent term adjustment
17
allowable under U.S. law, counting the entire
18
period of Gilead's delay as a testing phase. This
19
will prevent Americans from accessing generic
20
Descovy for an additional five years.
21
I think it is an extraordinarily disturbing
22
precedent that the Food and Drug Administration is
A Matter of Record (301) 890-4188
211
1
rewarding the decision of a corporation to delay
2
the development of a drug that it is now purporting
3
is safer than TDF/FTC.
4
As a final note, I will say that we are
5
placed today -- this entire committee is placed
6
today -- in an incredibly difficult position. We
7
are basically placed with a catch-22. Either on
8
one hand we deny the ability for Descovy or TAF/FTC
9
from getting a broad indication for cisgender women
10
and other populations, and despite having no
11
effective efficacy data for this population, or we
12
choose to deny the extension of that indication.
13
That, unfortunately in today's environment, would
14
basically work to deny women the choice to make the
15
decision of the drug that they would like to take.
16
I believe personally, and not representing
17
my organization, that the right choice is to extend
18
the indication of F/TAF to cisgender women. But I
19
have to admit that I am incredibly disturbed by the
20
precedent that that would set. We have to say,
21
today and more than two decades after AIDS
22
activists seized control of both the Food and Drug
A Matter of Record (301) 890-4188
212
1
Administration and the Centers for Disease Control
2
and Prevention, that cisgender women get HIV.
3
More than 50 percent of global infections
4
are in cisgender women, and the idea that an
5
applicant would decide not to basically provide
6
high-quality evidence supporting efficacy in this
7
population is disturbing. The fact that this
8
agency may be forced to grant a broad indication
9
with no efficacy data is also disturbing, and this
10
can never happen again.
11
I want to make that incredibly clear.
12
Cisgender women, transgender women, transgender
13
people, men who have sex with men, all of us
14
deserve -- that medical technologies that are
15
scaling up to fight one of the deadliest pandemics
16
of our time, they deserve high-quality evidence,
17
and we should never again encourage companies to
18
delay the development of innovative technologies,
19
and we should never allow them, once again, to not
20
provide high-quality evidence for these very
21
important technologies. Thank you.
22
DR. BADEN: Thank you. Will speaker number
A Matter of Record (301) 890-4188
213
1
5 step up to the podium and introduce yourself?
2
Please state your name and any organization you're
3
representing for the record.
4
DR. GIPSON: Hello. June Gipson, CEO of my
5
Brother's Keeper in Open Arms Health Care Center
6
located in Jackson, Mississippi. My Brother's
7
Keeper is a community-based organization with a
8
mission to reduce health disparities throughout the
9
United States by enhancing the health and wellbeing
10
of minorities and marginalized populations through
11
the leadership in public and community health
12
practices, collaboration, and partnerships.
13
We do it through an array of programs and
14
services, including our center for community-based
15
programs, and we also have a center for research
16
evaluation and policy change. One of our most
17
prominent centers is going to be Open Arms Health
18
Care Center. That's our primary healthcare clinic.
19
Open Arms Health Care Center is an
20
innovative, holistic, primary healthcare clinic
21
that offers preventive clinical mental health
22
services to underserved, uninsured,
A Matter of Record (301) 890-4188
214
1
underrepresented populations with an emphasis on
2
the LGBT population. We utilize a community-based
3
model that's a community health team led approach
4
to provide services to our clients that optimizes
5
their healthcare. We provide an array of services,
6
including women's health, family planning, men's
7
health, PrEP, HIV care, mental health, preventive
8
screenings, transportation, and emergency food
9
assistance.
10
When you look at our HIV and PrEP in
11
Mississippi, we've struggled, but we've been able
12
to accomplish some things. If you look at our
13
linkage for HIV testing and linkage to care, we're
14
either exceeding or we're meeting the national
15
goals. However, we continue to struggle in
16
retention and care and viral suppression.
17
When you look at our PrEP data, this is data
18
from Open Arms Health Care Center. This data is of
19
particular interest because we provide 75 percent
20
of all the PrEP in the state of Mississippi. As
21
you can tell, it's sparse. We are not hitting the
22
entire state. You have one red spot in the center
A Matter of Record (301) 890-4188
215
1
that really optimizes and says who we're reaching,
2
so we have more to do. There are multiple reasons
3
why we aren't able to do this in Mississippi.
4
We've done assessments with our patients and our
5
staff, and of course there's a lack of access.
6
That's a prevalent thing; it's around the country.
7
We also have a stigma. Stigma exists in
8
every form of HIV care that we provide. There's
9
also some other concerns that have come up
10
throughout our assessments with our patients. Not
11
only is there a low perception of risk -- and you
12
would think in Mississippi that that wouldn't be
13
the thing, but it is -- but there's also a concern
14
about side effects.
15
When you live in a state like Mississippi,
16
side effects are huge because we are already
17
existing with so many other negative health
18
outcomes. When we see the commercials that talk
19
about the benefits of the medication and how
20
wonderful they are, and that last 15 seconds when
21
they run through all of the side effects, that's
22
what we hear. And I hear it in particular when you
A Matter of Record (301) 890-4188
216
1
speak about Truvada because my dad is on dialysis.
2
My uncle is on dialysis. My cousin's on dialysis.
3
So we're living the side effects, and we need a
4
safer option.
5
We need a safer option for no other reason
6
that we have gone through enough. We have high
7
diabetes rates, high blood pressure rates, kidney
8
failure; you name it, we have it, and all of this
9
combined with HIV. We need Descovy for PrEP so we
10
can increase utilization. If we gave pills free to
11
everyone, that's access, but access is not
12
indicative of utilization.
13
So we need to have a safer option for our
14
community, and particularly with African Americans.
15
Again, we live these health disparities. We live
16
these side effects. And with women in particular
17
who may have a low perception of risk, it seems as
18
if we're taking Truvada, we're trading illnesses.
19
I get rid of one just to get a another? That's not
20
something that we want.
21
Particularly for adolescents, and I'm going
22
to include parents with our adolescents, parents
A Matter of Record (301) 890-4188
217
1
generally want the best for their children. And
2
they really don't want to think about their kids
3
having sex, but we know that that's a real thing.
4
But if they see the advertisements for Truvada, and
5
they see the side effects, it gives an impression
6
that they're going to expose their children to
7
something that's going to give them a lifelong
8
problem.
9
When you're in the state of Mississippi, you
10
see your family in a dialysis clinic. If you ever
11
have an opportunity come and visit, stop by a
12
dialysis clinic. You'll see that it's filled with
13
African Americans.
14
As it relates to adolescence, I actually
15
have a call tomorrow with a parent. Her
16
16-year-old son came to us. He tested positive for
17
syphilis and chlamydia. We put him on PrEP. She
18
called a month later wanting to take him off of
19
PrEP because she's so concerned with the health
20
issues and the side effects associated with PrEP.
21
Now, you and I, we understand the correlation
22
between syphilis and HIV, but that's not her
A Matter of Record (301) 890-4188
218
1
reality. That's not her perception. And we all
2
live within our perception because that's our true
3
reality. So I will like for Descovy to be approved
4
for PrEP for utilization broadly between African
5
American women and adolescents.
6
DR. BADEN: Thank you. Will speaker number
7
6 please step up to the podium and introduce
8
yourself? Please state your name and any
9
organization you're representing for the record.
10
MR. MYERS: Good afternoon. I'm Kirk Myers
11
of Abounding Prosperity in Dallas, Texas. I am the
12
founder and chief executive officer of an HIV and
13
AIDS prevention agency in Dallas, Texas. The
14
mission of my organization is to provide services
15
that address health, social, and economic
16
disparities among Black Americans with an emphasis
17
on the GBTQ community and their families.
18
I'm also a black man who has sex with men,
19
MSM, and who is living with HIV for over 26 years.
20
Through my lived experiences and managing my own
21
disease, and the leadership experience of managing
22
my agency, dedicated to decreasing new incidence of
A Matter of Record (301) 890-4188
219
1
HIV and AIDS via various prevention programs, I
2
know the delays and deliberations that are
3
surrounding the prompt approval of Descovy for the
4
proposed use of PrEP for black women, MSMs, and
5
trans individuals is out of sync with our
6
real-world reality.
7
For me, the simple language that best
8
captures the reality among my people, especially
9
those black women, MSMs, and trans individuals, is
10
overwhelmed by the social, economic, and health
11
disparities that they confront daily. So while
12
some people have privilege on their side for
13
time-consuming contemplation over the prompt
14
approval of Descovy for the proposed use of HIV as
15
PrEP, my community makes immediate choices on a
16
day-to-day basis that ultimately could result in
17
the acquisition or spread of HIV-AIDS.
18
Therefore, I urge the prompt approval of
19
Descovy for the proposed use of HIV PrEP because it
20
is right to give black women, MSMs, and trans
21
people the option to make a safer effective choice
22
on a daily basis to protect their lives as they go
A Matter of Record (301) 890-4188
220
1
about their business as usual. Whether their
2
business is at the level where I work as the CEO or
3
the street level of a sex worker, I will be
4
standing as an authentic voice to compel the
5
advisory community to consider the fact that I have
6
immediate access to those who would benefit from
7
Descovy for the proposed use of HIV prevention.
8
I have organized community forums, focus
9
groups, and one-on-one individual level
10
interventions to speak with authority that this
11
drug is wanted. The young women and gay men who
12
confide in me have expressed receptivity to a drug
13
that has the potential to protect them from HIV-
14
AIDS with lower side effects.
15
Finally, if anything is right at this
16
historical moment in HIV prevention efforts, it is
17
options to go beyond the past practice of
18
normalizing the majority and ignoring the pressing
19
needs of the minority. The right thing to do is to
20
empower black women, MSMs, and trans individuals
21
with the additional tools on a daily basis that are
22
purposefully designed to protect public health.
A Matter of Record (301) 890-4188
221
1
Without this option, expediency, desperation, and
2
ignorance will continue to drive up the statistics
3
of new incidence of HIV and AIDS
4
With all due respect, I am asking the
5
advisory committee members to join me in doing the
6
right thing and assist on the prompt approval of
7
Descovy for the proposed use of HIV-based
8
prevention on my intimate relationships with MSM
9
and transgender individuals, who expect me to speak
10
out and share our testimony. This is the right
11
step.
12
Furthermore, we implore that this drug be
13
approved not just in gay men and transwomen, but
14
women need this drug, and it will not be in the
15
interest of public health to have this drug
16
approved without including women, and to then be
17
further stigmatizing by being looked at as a gay
18
drug. Everyone deserves the same choices of
19
prevention options as the rest of us.
20
Now, as a black man living with HIV here in
21
America
22
for the past 26 years, there has been this divide
A Matter of Record (301) 890-4188
222
1
between black gay men and black women, and when we
2
look at our options, this is the best option for
3
all of us possible. I'm not a scientist. I didn't
4
have all the beautiful slides and all those things
5
to compel you to do anything, but I can tell you
6
from the grassroots level and at the street level
7
that this drug is needed. And again, if we only
8
approve it for one indication, it's going to create
9
further stigma that we do not need. Thank you.
10
DR. BADEN: Thank you. Will speaker number
11
7 step up to the podium and introduce yourself?
12
Please state your name and any organization you are
13
representing for the record.
14
MR. WARREN: Good afternoon. My name is
15
Mitchell Warren, and I'm the executive director of
16
AVAC, a New York-based global, nonprofit
17
organization focused on accelerating the
18
development and delivery of new prevention options.
19
We take no money from any pharmaceutical companies,
20
including from Gilead Sciences, although I should
21
note I was a member without any compensation of the
22
Independent Data Committee of the DISCOVER trial.
A Matter of Record (301) 890-4188
223
1
I stood here, as some of you in this room
2
did as well, seven years ago, and the task was
3
easy. The data was robust, the evidence was clear,
4
and I'm delighted that that committee then, and the
5
FDA shortly thereafter, followed the evidence and
6
approved TDF/FTC for oral PrEP for all populations.
7
I wished the task were as easy today.
8
There, while the evidence was clear, today we sit
9
in somewhat of an evidence-free zone, at least in
10
some areas as has been well discussed today. It's
11
a dynamic space and one that I hope we don't return
12
to ever again, and I have some thoughts about that
13
toward the end. But the data are the data, and we
14
must act on that most urgent data point presented,
15
and that is an epidemic that continues in multiple
16
places, in multiple populations. And what we do
17
today matters, not just in the United States, but
18
particularly for women at great risk of HIV
19
infection in Africa.
20
I recognize full well that that is outside
21
of the purview of the FDA and certainly of this
22
committee. Your job is to look at safety and
A Matter of Record (301) 890-4188
224
1
efficacy for the United States. That said,
2
decisions made in this room today, recommendations
3
made in this room, decisions made subsequently by
4
the FDA, will resonate and influence the global
5
response. And I realize that's a heavy burden, but
6
one that is real.
7
I'm going to take just a few minutes to go
8
through the two questions that you have on the
9
table, F/TAF for PrEP for men and transgender
10
women, first and foremost. It is very clear to me
11
and to AVAC, the organization I lead, that the data
12
presented in the application does indeed support a
13
noninferiority claim for F/TAF compared to F/TDF
14
for oral PrEP for gay men and transgender women.
15
I emphasize that as noninferiority. The
16
DISCOVER trial was set out to design for
17
noninferiority, and it certainly met that task. I
18
think that's a very important point not only as you
19
make your vote today in the committee, but as the
20
FDA works around the labeling with Gilead, that
21
this be very clearly registered as a noninferior
22
oral PrEP option.
A Matter of Record (301) 890-4188
225
1
Any claims of superiority I think are
2
unfounded. Yes, there's a different safety
3
profile, and we saw data today that made it seem
4
both safer on some level but concerns in others
5
with lipid and weight gain. But we need to be very
6
clear so there is no confusion to PrEP users today
7
on TDF/FTC, or PrEP users of tomorrow that we are
8
somehow promoting one as a safer and more effective
9
drug. This is a noninferior oral PrEP option, and I
10
support that wholeheartedly, but all labeling must
11
be consistent with that and be strongly enforced.
12
In terms of the second question you all will
13
consider, it's perhaps the more challenging in so
14
many respects, and that is F/TAF for cisgender
15
women. It is extremely unfortunate that similar
16
safety and efficacy data for F/TAF were not
17
collected in an efficacy trial. We can spend a lot
18
of our time Monday morning quarterbacking why that
19
was and why decisions were made.
20
I would argue that the best time to debate
21
that is not sitting in an FDA hearing to consider a
22
drug approval. Those should've been open
A Matter of Record (301) 890-4188
226
1
conversations we had with the company, with the
2
FDA, and with community groups far and wide to
3
discuss the best pathway for product development.
4
I do trust that that is the case, and I think
5
Dr. Murray in his introductory comment described
6
that for next generation new chemical entity PrEP
7
agents. That is not a change; that what we
8
discussed today is not creating a new status quo.
9
We do have to recognize that this is a tenofovir
10
prodrug and tenofovir-based prep, and I use my
11
comments in that regard.
12
If F/TAF is not extended to include
13
cisgender women, the one group, the only group that
14
will suffer and pay the price for that decision are
15
women at risk of HIV infection. The FDA won't
16
suffer, Gilead will not suffer, and other agencies
17
will not suffer. That said, we have to always be
18
clear that safety and efficacy matter.
19
Based on the data presented today, and I
20
should say in addition, not taking money from
21
pharmaceutical companies. I am not a statistician,
22
a trialist, an ethicist, or scientist of any type.
A Matter of Record (301) 890-4188
227
1
I'm an advocate. But I will say that based on the
2
data presented here, recognizing the systemic
3
levels as monitored in a range of studies both for
4
the safety study presented, although less robust
5
than we would like, as well as the treatment
6
studies, there's a very clear rationale for F/TAF
7
to work as well as F/TDF in women.
8
I believe that that is critical to approve.
9
That said, I believe that that PrEP indication
10
needs to come with an incredibly strong, robust,
11
and enforceable postmarketing surveillance,
12
research agenda, and a risk evaluation mitigation
13
strategy that makes it very clear that over the
14
next 12 to 24 months, Gilead will be responsible
15
for collecting, in collaboration with other
16
research groups, the relevant data for safety and
17
effectiveness.
18
As well discussed here, efficacy of PrEP in
19
women is hard to measure currently with oral prep;
20
not impossible, but hard to do. Let us focus on
21
effectiveness, and let us ensure that an
22
FDA-enforced postmarketing surveillance in REMs
A Matter of Record (301) 890-4188
228
1
ensures that we have that data.
2
I will say, too, in our work, in Africa
3
particularly, and has been reported in a number of
4
studies, including work that we have done, that
5
pill size does matter. It is one of the leading
6
reasons that women in programs in Africa talk about
7
not continuing with F/TDF. While again, I realize
8
Africa is not in the purview of this committee or
9
of the FDA, your decision will matter, and a
10
smaller drug, not necessarily safer or more
11
effective, but a smaller drug will be of enormous
12
benefit.
13
I want to emphasize again in closing that
14
the education prescriber information and supportive
15
materials that are part of any package going
16
forward with F/TAF need to be heavily monitored by
17
the FDA, and not just between the FDA and Gilead
18
but with community input; not tokenistically, but
19
in an active way to ensure that the language used
20
to describe this indication as a noninferior
21
product for all populations is clearly described,
22
clearly enforced, and robustly done.
A Matter of Record (301) 890-4188
229
1
So we do urge the committee to approve for
2
all populations F/TAF for PrEP. We do urge the
3
committee to consider the consequences of you
4
voting no, which would send a signal of delay and
5
distrust of the research community in a product
6
development, and at the same time committing
7
together that we are not changing the rules for
8
future products of new PrEP agents, that we ensure
9
that we have better conversations earlier in the
10
process so the products coming to this committee
11
and to the FDA are done with the most robust and
12
complete package possible. Thank you very much.
13
Clarifying Questions (continued)
14
DR. BADEN: Thank you.
15
Once again, the open public hearing speakers
16
have presented us with incredibly powerful insights
17
in the challenge at hand before us, and we thank
18
all of the speakers for sharing your thoughts and
19
convictions and insights in balancing this very
20
difficult problem.
21
The open public hearing portion of this
22
meeting is now concluded and we'll no longer take
A Matter of Record (301) 890-4188
230
1
comments from the audience. We'll now turn our
2
attention back to the business at hand, which is
3
evaluating the data presented before us, and we
4
will continue with our clarifying activities with
5
the applicant.
6
Prior to the applicant presenting some of
7
the follow-up, I just had one clarifying question
8
to the agency, which is adolescents, as I look at
9
all the materials we've received, seems to be
10
defined by a way to greater than or equal to 35
11
kilograms, and that is not how I've always thought
12
of adolescence. So I just want to know if there's
13
an age parameter there or simply a weight
14
parameter.
15
(Laughter.)
16
DR. BADEN: Is it 15 to 17 and of sufficient
17
weight or is it down to 10, or down to 5? I just
18
want to know are there any parameters around the
19
adolescent category?
20
DR. MURRAY: I think we're sticking with
21
weight. It becomes tricky to decide what age one
22
should be starting to use.
A Matter of Record (301) 890-4188
231
1
DR. BADEN: So could it be 10 then?
2
DR. MURRAY: Well --
3
DR. BADEN: If it's purely weight, then I
4
guess it could be a big 10 year old, could be 36
5
kilos.
6
DR. MURRAY: We know the safety and how to
7
dose down to 35 kilograms, and exactly when a
8
physician or a person who's of adolescent age
9
should consider it is probably up to them. And
10
when you put an age in there, it kind of boxes you
11
in, with a lot of respect.
12
DR. BADEN: Is Truvada 15? I thought
13
Truvada was 15 to 17, or is that purely weight
14
based? It's purely weight based.
15
DR. MURRAY: Weight based.
16
DR. BADEN: Okay. Well, thank you for
17
the -- I just wanted to make sure I was reading
18
weight as the determinant and not other factors.
19
So back to the applicant, who wanted to
20
clarify some of the concepts from this morning that
21
needed your input, and then we will come back to
22
the many questions we have on the list from the
A Matter of Record (301) 890-4188
232
1
panel members.
2
DR. BRAINARD: Thank you. We have four
3
clarifying answers to prior questions. The first
4
is around the data for Descovy in vaginal tissue,
5
and I'd like to just walk through what the
6
available data are, so everyone has a clear
7
understanding of the data in the literature.
8
Slide 1 up, please. There have been three
9
different studies of Descovy in vaginal tissue.
10
One was a single-dose study of Descovy, and in the
11
discussion of that manuscript, they compared those
12
results with another study that the same group had
13
conducted several years prior with Truvada.
14
The second study was done with a single dose
15
of Descovy and Truvada within the same study. The
16
third study was done by the same group and was
17
multiple doses of Descovy and Truvada looking at
18
vaginal tissue levels.
19
Slide 2 up, please. Here are the results
20
from those studies. In the first study, looking at
21
Descovy vaginal tissue levels following a single
22
dose of Descovy, the tissue levels of AUC was
A Matter of Record (301) 890-4188
233
1
reported as 132,098. This was compared in a
2
cross-study comparison to the Truvada levels, which
3
were noted to be 1.3 to 1.8-fold higher than those
4
with Descovy.
5
The second study that was done with a single
6
dose of Truvada or Descovy in the same trial
7
demonstrated that after 4 hours, all of the samples
8
with Truvada were below the limit of
9
quantification, and 69 percent of the samples with
10
Descovy were below the limit of quantification.
11
The conclusion from that was that multiple dose
12
data are needed.
13
In the setting of multiple doses, which is
14
indeed the more relevant setting to assess tissue
15
levels for a daily administered drug, 4 hours after
16
dosing of Descovy or Truvada, levels were 2.6-fold
17
higher with Descovy as compared to Truvada. FDA
18
presented these data in their presentation. At 24
19
hours and 48 hours after dosing stopped, there were
20
comparable and low levels between Descovy and
21
Truvada in the vaginal tissue.
22
I'd like to now put these vaginal tissue
A Matter of Record (301) 890-4188
234
1
data into context with what we know about the
2
rectal tissue data. Slide 1 up, please. The
3
vaginal tissue are just now a graphical
4
representation of the data I showed you at the
5
4-hour time point in the table, where you can see
6
that Descovy achieves slightly higher levels than
7
Truvada 4 hours after dosing.
8
As compared to rectal tissue levels, the
9
first thing to note is that Truvada achieves about
10
10-fold higher level than Descovy in the rectal
11
tissue. It's also relevant to note that the rectal
12
tissue levels with Truvada are somewhat of an
13
outlier as compared to the vaginal tissue levels
14
with both Descovy and Truvada, and the rectal
15
tissue with Descovy.
16
This has been hypothesized to be related to
17
the low bioavailability of Truvada, and the fact
18
that there may be drug delivered directly through
19
the GI tract to the rectal tissue with Truvada.
20
That's done so to a lesser extent with Descovy,
21
which has higher bioavailability. This is a
22
hypothesis without clinical or scientific proof.
A Matter of Record (301) 890-4188
235
1
Nevertheless, what we know about Truvada is
2
that despite the lower levels in the vaginal tissue
3
as compared to the rectal tissue levels with
4
Truvada, Truvada for PrEP is highly an equally
5
efficacious in men and women. So these lower
6
levels of vaginal tissue nevertheless correlate to
7
having efficacy in the setting of Truvada for PrEP
8
use in women.
9
Similarly, what we now know with the
10
DISCOVER trial is that despite having 10-fold lower
11
levels of tenofovir diphosphate in the rectal
12
tissue as compared to Truvada, both drugs
13
demonstrated that they were highly effective and
14
Descovy was noninferior to Truvada at preventing
15
HIV acquisition. These data contribute to the
16
increasing body of understanding that systemic drug
17
levels are what's driving efficacy, and efficacy is
18
not related to particularly homogenate tissue
19
levels.
20
I can keep going to the other issues or we
21
can stop for comments, Dr Baden.
22
DR. BADEN: Thank you. Your point's well
A Matter of Record (301) 890-4188
236
1
taken. Since this is such an important issue,
2
comments from the committee to better understand
3
these data since these bridging data are a critical
4
element.
5
DR. DODD: Lori Dodd, the statistician. One
6
of the concerns I have looking at these data is
7
they're extremely small numbers, and the box plots
8
you're seeing are really the interquartile range as
9
opposed to some confidence intervals. So I need
10
some help understanding how generalizable these
11
results are to the larger population. I'm unable
12
to do that based on the data presented.
13
DR. BRAINARD: In terms of the
14
generalizability, all tissue-level studies that
15
have been conducted have generally been in less
16
than 10 participants for group, occasionally
17
somewhere between 10 and 15. This is just related
18
to the invasive nature of conducting these studies
19
and the requirements for a biopsy.
20
In addition, we haven't seen any data
21
looking at tissue-level data in prevention studies
22
because, of course, taking biopsies in the setting
A Matter of Record (301) 890-4188
237
1
of individuals who are at risk for HIV infection
2
could actually increase their risk. So those data
3
are not available, nor are they likely to be
4
generated.
5
I would agree that the data are variable and
6
that there are not a large amount of data.
7
Nevertheless, when we think about what these data
8
mean in the setting of a high amount of clinical
9
data around the efficacy of Truvada for PrEP in
10
both men and women, those data can provide
11
reassurance.
12
DR. DODD: A little more clarity would help,
13
too, then. How is it that states are coming up
14
with guidelines on the amount of time needed to
15
obtain maximum intracellular concentrations and
16
pushing in that direction when we're only able to
17
get 14 participants from this study? I might also
18
ask if it would be appropriate to ask the agency to
19
comment on this as well.
20
DR. BADEN: Yes.
21
DR. BRAINARD: During the lunch break, we
22
tried to track down the data that actually were
A Matter of Record (301) 890-4188
238
1
behind the timeline recommendation around 20 days,
2
and the data really come back to very sparse tissue
3
data, and there are no data that connect to 20
4
days.
5
One study showed that when assessing vaginal
6
tissue levels and rectal tissue levels over time,
7
it seemed that at the 10-day time period, there
8
were stable and steady-state levels within the
9
rectal tissue obtained with Truvada, whereas in the
10
vaginal tissue, levels were still seen to be
11
increasing. It seems like that is the basis for
12
the extrapolation to 20 days required for
13
prevention. But this has never been validated, and
14
we don't know of any clinical data to speak to the
15
time to protection for women.
16
DR. BADEN: Would you like the agency to
17
comment if anyone is aware of the basis for those
18
recommendations?
19
DR. DODD: And also if they can comment on
20
their understanding of the uncertainty associated
21
with the concentrations in the tissues given the
22
small numbers.
A Matter of Record (301) 890-4188
239
1
DR. MIELE: Right. We actually have
2
concerns about the reliability of that data because
3
of the inconsistency and the small numbers, and the
4
differences in methodology. But that's all we have
5
right now, and the extrapolation approach is what's
6
being proposed.
7
As to the time to achieve protection, I
8
don't believe the CDC has a recommendation in that
9
regard. What they're stating is the time to
10
achieve maximum concentrations. Some state
11
guidelines have interpreted that to mean protective
12
levels, which kind of makes sense. But I agree
13
that I don't know that the data are very robust to
14
that extent, and these are very conservative
15
measures.
16
We have not introduced any of that to the
17
labeling, for example. We have not reviewed any of
18
that data because as it stands right now, PrEP is
19
meant to be used in combination with safer sex
20
practices, so it's sort of counter-productive or
21
counter-intuitive to suggest a lead-in period when
22
you could come off condoms for example.
A Matter of Record (301) 890-4188
240
1
So we have not entertained that and we have
2
not really reviewed that data, but I was pointing
3
it out that that concern about the differential
4
distribution is out there, and it's guiding some of
5
these recommendations that are being put out there
6
by states.
7
DR. BADEN: Do you have another follow-on,
8
Dr. Dodd?
9
DR. DODD: Just one final comment, and I
10
don't know if one of the statisticians who've
11
looked at the concentration data could comment.
12
But when I hear a number like a 10-fold increase in
13
the tissue concentrations. That can tend to stick
14
in everybody's mind, but we have to understand the
15
uncertainty associated with that.
16
Has the confidence interval been estimated
17
so that we don't get hung up on that number or
18
something like that? I think this actually is a
19
pretty important to point. I'll leave it at that,
20
but I just want to make that as a final point.
21
DR. ZHENG: This is Jenny from FDA. The
22
numbers we have for tissues normally were small
A Matter of Record (301) 890-4188
241
1
numbers and presented as median quartiles because a
2
lot of below limit of quantitation was observed in
3
those tissue concentrations, so there are some
4
limitations.
5
DR. BADEN: Dr. Ofotokun?
6
DR. OFOTOKUN: Mine is just a minor
7
clarification about the method. These tissue
8
concentrations are generated in the rectal and the
9
vagina. Can you confirm or clarify to me whether
10
the vagina data, is it a biopsy of the vagina
11
tissue, or is this CVL, or aspirate, or swab? How
12
were they -- I know they are different sometimes
13
when you look at those different compartments as
14
opposed to rectal drug concentration.
15
DR. BRAINARD: There are a range of
16
methodologies used for these compartments studies,
17
and cervical vaginal lavage is often a method where
18
tissue or cells are washed from the cervix, and
19
sometimes with or without scraping. The data that
20
I shared with you are biopsy data.
21
Slide 1 up, please. This slide provides a
22
very high level schematic of how these tissue
A Matter of Record (301) 890-4188
242
1
levels are measured. Whether it's in the rectum or
2
in the vagina, forceps are used to take biopsies.
3
Generally with rectal tissue, more biopsies are
4
taken than with vaginal sampling, where it's
5
generally limited to 1 to 2.
6
These biopsies consist predominantly of
7
epithelial cells and fibroblasts, which make up the
8
majority of the tissue and point to some of the
9
limitations of the sampling. Also contained within
10
that biopsy will be a variety of immune cells,
11
including relevant CD-4 T cells, but also
12
macrophages, B cells, neutrophils, NK cells, and
13
dendritic cells.
14
That tissue block is incubated with enzymes
15
in order to break up the cells because the
16
tenofovir diphosphate only exists inside cells. So
17
it's released from the cells through enzymes, and
18
then generally the amount of tenofovir diphosphate
19
is quantified using mass spec. So the total
20
tenofovir diphosphate level that is reported is the
21
tenofovir diphosphate across all of these different
22
cell types, recognizing that the predominant cells
A Matter of Record (301) 890-4188
243
1
that are contributing to these levels are
2
epithelial cells and fibroblasts.
3
It's been hypothesized that part of the
4
reason those vaginal tissue levels drop off at 24
5
hours and 48 hours, and why there are so many BLQ
6
measurements at those time periods is because
7
epithelial cells have a more rapid turnover, and
8
therefore tenofovir diphosphate within the
9
epithelial cells, which are representing a higher
10
proportion of contribution to the levels, are
11
turning over and are no longer having tenofovir
12
diphosphate at the 24-hour and 48-hour time point.
13
That's just a hypothesis.
14
DR. BADEN: Thank you very much. Please
15
continue with the other follow-ons from this
16
morning.
17
DR. BRAINARD: Dr. Daskalakis asked about
18
transgender women. There were 74 transgender women
19
enrolled in the DISCOVER trial. None of those
20
participants acquired HIV infection. There was
21
also a question about gender-affirming hormone use,
22
and 53 of the 74 transwomen reported using
A Matter of Record (301) 890-4188
244
1
gender-affirming hormones. We did look at the
2
subset of those participants who had PK sampling
3
down at the week 4 time point, and found that there
4
was no difference in tenofovir diphosphate levels
5
within that population.
6
Slide 2 up, please. This slide just shows
7
data on the 18 women who were part of the substudy
8
that had tenofovir diphosphate levels within PBMCs
9
measured at week 4. And you can see that there
10
trough concentration of tenofovir diphosphate is
11
similar to what was seen in the MSM population,
12
despite being on gender-affirming hormones.
13
DR. BADEN: Thank you. Any questions? I
14
think these are fairly clear. Thank you.
15
Continue.
16
DR. BRAINARD: The third issue was providing
17
some additional information about insertive anal
18
intercourse, and I'll ask Dr. Moupali Das to speak
19
to that.
20
DR. DAS: Just to remind everyone, the
21
eligibility criteria for the DISCOVER trial were in
22
two parts. The first piece was requiring two
A Matter of Record (301) 890-4188
245
1
episodes of condomless anal sex with more than one
2
unique partner in the past 12 weeks prior to
3
enrollment, or the second criteria was evidence of
4
rectal gonorrhea, rectal chlamydia or syphilis in
5
the past six months, past 24 weeks. A high
6
proportion of people in the study, as you saw,
7
reported condomless anal sex.
8
We're going to share the data with you of
9
the people reporting condomless insertive anal
10
intercourse in terms of number of partners at
11
screening prior to baseline. Slide 2 up. The mean
12
number of insertive anal intercourse partners was
13
4, which is the same as the report of condomless
14
anal intercourse partners. There were no
15
differences between arms. All the people who are
16
infected in this study, the 22 people who acquired
17
HIV, had data and biologic evidence of condomless
18
anal intercourse.
19
DR. BADEN: Just to clarify, so I'm
20
understanding these data and what's implied, do you
21
have data on men who were insertive but not
22
receptive? So purely insertive, and what degree of
A Matter of Record (301) 890-4188
246
1
transmission occurred in that population?
2
DR. BRAINARD: We don't have data on people
3
who were purely insertive, but the criteria for
4
eligibility in the study required evidence of
5
receptive anal intercourse that was unprotected.
6
There was no infections -- all the people who were
7
infected had evidence of receptive anal
8
intercourse.
9
DR. BADEN: Follow-on questions or
10
clarifications for this?
11
(No response.)
12
DR. BADEN: Okay. Please continue.
13
DR. BRAINARD: The last topic I'd like to
14
just proactively follow up is the question about
15
study design issues in ciswomen. As has been
16
pointed out by panelists, community members, and
17
FDA, there are challenges with conducting a
18
clinical trial in women, a superiority study for 2
19
oral drugs that are tenofovir prodrugs as
20
infeasible, and a placebo-controlled trial is not
21
going to be ethical given Truvada is effective in
22
women.
A Matter of Record (301) 890-4188
247
1
We talked a little bit about noninferiority
2
and the challenges around establishing a
3
noninferiority margin and FDA's perspective on the
4
inability to construct a noninferiority margin
5
because of the lack of consistency. We did look at
6
taking the effect from the two most effective
7
randomized clinical trials in women Partners PrEP,
8
which was one of the registrational studies for
9
Truvada, and then the Bangkok study, which was
10
actually a study in injection drug users, but, most
11
of the HIV acquisition in women was due to sexual
12
transmission.
13
So using the treatment effect from those two
14
studies, we calculated a noninferiority margin
15
using the same methodological approach we used for
16
DISCOVER, and came up with a sample size of 22,000
17
in a high-risk population. That would take 8 to 10
18
years to conduct, which was part of the reason that
19
we didn't initiate that study, particularly in the
20
setting of the ongoing DISCOVER study.
21
However, we also recognize there's been a
22
lot of discussion since 2015 about this conundrum
A Matter of Record (301) 890-4188
248
1
of what can be done to assess the efficacy of PrEP
2
in women, and also now where we are because of the
3
DISCOVER results and because there's highly
4
effective active comparator in Truvada, and now
5
Descovy, going forward in men as well.
6
Dr. Murray from FDA has been one of the
7
leaders in this area. We've been participating in
8
discussions as well as with PrEP experts, and
9
academics, and community members. There are some
10
novel trial design methodologies that don't fall
11
within the standard rubric, but I'm going to ask
12
Dr. Wulfsohn to discuss some of those approaches
13
from a statistical standpoint.
14
DR. WULFSOHN: Thank you. And just to
15
clarify, the 22,000 that Diana referred to would be
16
a study in Africa, so you're dealing with a high
17
incidence rate of 4 per hundred person-years. In
18
order to find the best noninferiority design, we
19
also selected 2 of the 5 women studies which had
20
the most benefit from Truvada. So we've cherry
21
picked the two studies to try and help us reduce
22
the sample size, and the lowest we can get it to is
A Matter of Record (301) 890-4188
249
1
22,000.
2
Now we're certainly open to more innovative
3
ideas. And fortunately for us, Jeff Murray gave a
4
great talk at IAS a week ago, and we're very
5
receptive to some of the ideas that Jeff proposed,
6
and I'd like to go through some of these in terms
7
of how a woman's study could look. All of these
8
are our proposals from Jeff, so I won't mention his
9
name anymore.
10
It was proposed that there should be at
11
least two placebo anchors in order to interpret a
12
woman study. The two that come to mind would be,
13
firstly, an epidemiologic assessment of the placebo
14
incidence. We would envisage a study in Africa
15
where that would be known based on current
16
epidemiologic data, what the incidence is in women
17
not on antiviral protection.
18
The second approach to estimate a placebo
19
incidence could be based on the screening period
20
from the study. Knowing how long it was from the
21
last test to beginning treatment, that being the
22
risk period, we could look at the subset of women
A Matter of Record (301) 890-4188
250
1
who are not on Truvada and assess the incidence.
2
And that would be a reference, placebo incidence.
3
The other thing that was proposed in order
4
to assess whether a PrEP drug is effective is that
5
it should lower the incidence by 5 to 10-fold. The
6
idea came from oral contraceptives, where oral
7
contraceptives actually lower the incidence about
8
40-fold, but we're trying to be realistic.
9
Just for reference, if you look at the
10
DISCOVER study where, granted, the adherence was
11
very high, we're estimating that Truvada lowered
12
the incidence by 10- to 20-fold based on two
13
different ways of estimating the placebo incidence,
14
and Descovy lowered the incidence by 20- to 40-
15
fold. So these are both effective agents.
16
I'd like to bring up slide number 1, which
17
is the noninferiority study we were talking about.
18
But just with reference to these five studies, the
19
two best studies, Partners PrEP and the Bangkok
20
study, which had the lowest risk ratio, in Partners
21
PrEP, we're lowering incidence 3-fold, and in
22
Bangkok, we're lowering incidence 5-fold. These
A Matter of Record (301) 890-4188
251
1
are just the point estimates.
2
I would also add that in our own
3
demonstration project where we've got a lot of
4
real-world data in women, our estimate is that
5
we're lowering incidence approximately 5-fold based
6
on observing an incidence of 0.8 per hundred
7
person-years, largely from cohorts in Africa where
8
you'd expect of incidence of 4 per hundred
9
person-years.
10
The 5-fold is as far as we are currently
11
getting with current adherence rates. It's
12
potentially possible to improve adherence and get
13
greater effect sizes, but clearly the metric for
14
what constitutes good enough efficacy will need to
15
be tailored to the population and adherence that
16
we're getting.
17
Another criteria, which is an extra
18
criteria, not a different option, is that the
19
incidence rate in the experimental arm should be no
20
more than 0.5 higher than the active control, which
21
would be Truvada, and that seems somewhat
22
reasonable.
A Matter of Record (301) 890-4188
252
1
Another separate approach was proposed, and
2
that is to look at the adherence subset of a study.
3
So in DISCOVER, if you look at the individuals who
4
are taking 2 or more tablets per week, we're only
5
seeing 2 infections, one in each arm. So we are
6
observing an incidence of less than 1 in a thousand
7
in adherence subjects. And it was proposed that
8
that threshold of 1 in a thousand is a reasonable
9
measure of what constitutes an effective agent.
10
These are all good ideas and very innovative
11
creative approaches that we can leverage and work
12
with the FDA on to try and design a woman's study
13
that answers the efficacy question, and we're
14
committed to doing this.
15
DR. BADEN: Thank you. If FEM-PrEP and
16
VOICE in the placebo groups had 5 per hundred
17
person-years, I'm having trouble understanding how
18
you come to a 22,000 person study, when if we look
19
at the DISCOVER, which had a 1 per hundred
20
person-years, you have a 5-fold increased event
21
rate, yet a 3-fold increase in sample size? I'm
22
having trouble understanding.
A Matter of Record (301) 890-4188
253
1
DR. WULFSOHN: Slide number 1 up. The main
2
thing that's driving up the sample size is the
3
weaker performance of Truvada in these women
4
studies. In the design of DISCOVER, we estimated,
5
based on the three historical controls, that we
6
would lower incidence 5-fold. Here, when you pool
7
these two best studies, you're lowering incidence
8
3-fold. So it becomes a lot harder to retain 50
9
percent of a weak effect.
10
DR. BADEN: I see your point. Still, I'm
11
concerned with the assumptions, but I see your
12
point.
13
Other questions? Please, Dr. Goetz?
14
DR. GOETZ: I want to come back to what you
15
just said, the weaker performance of Truvada in
16
these women. Are you stating that irrespective of
17
adherence?
18
DR. WULFSOHN: No.
19
DR. GOETZ: And it comes back to Dr. Baden's
20
question, then, as to why the sample size must be
21
so large. Are you projecting that the women you
22
will enroll will be non-adherent?
A Matter of Record (301) 890-4188
254
1
DR. WULFSOHN: Our interpretation of the
2
data is that adherence is a primary driver of
3
efficacy. As you've seen presented today, there
4
are several thousand women who've been, uh, given
5
PrEP and over a hundred thousand men who've been
6
given PrEP. And if you look at the literature,
7
there's a total of 6 case reports of individuals
8
getting infected while on treatment.
9
So it's highly unusual to get infected while
10
on adequate treatment or with adherence and that's
11
why the threshold for what constitutes good enough
12
is you need to have less than one in a thousand
13
individuals getting infected, because that's what
14
the current drugs can deliver.
15
DR. BADEN: Dr. Giordano?
16
DR. GIORDANO: But then, why did the
17
DISCOVER study work?
18
(Laughter.)
19
DR. BADEN: It's a circular problem we're
20
dealing with.
21
DR. GIORDANO: Because you expected a
22
10-fold higher rate of HIV than you saw in both
A Matter of Record (301) 890-4188
255
1
arms. The adherence was extremely high, higher
2
than was across the board in the previous studies,
3
and yet you ended up with a noninferior drug,
4
statistically proven noninferior drug in this
5
population. I don't get it.
6
DR. WULFSOHN: My understanding is that if
7
you're perfectly adherent to both Truvada or
8
Descovy, there's no advantage to one or the other
9
drug from an efficacy point of view. That's a
10
hypothesis. In the data from Truvada, and similar
11
for Descovy, we're seeing 1 and 2 and a half
12
thousand approximately infections in individuals
13
who are adherent; to find a DBS that's done every
14
3 months, showing adequate drug levels.
15
On the other end of the spectrum, if you
16
stop taking the drug completely, there's no
17
difference between what the drug can provide you
18
because it's not providing you any benefit. So the
19
benefits, to the extent there is a benefit, is in
20
the middle, the individuals who are not fully
21
adherent but are taking some drug, and the PK
22
properties of the drug lead us to believe that, at
A Matter of Record (301) 890-4188
256
1
least from the PBMC levels, that there could be an
2
advantage to the efficacy with Descovy.
3
These data are somewhat suggestive, an
4
unproven advantage. We don't have enough data to
5
say even in that subset there's an advantage, and
6
certainly the study overall hasn't shown
7
superiority, but that's a hypothesis that can be
8
tested in the future as well.
9
DR. BADEN: Dr. Walker, you had a question?
10
DR. WALKER: Yes, and it may not correlate
11
to the discussion that's going on at hand. And
12
forgive me if you have mentioned this. It's been a
13
lot of information that's been presented here. But
14
I just wanted to know, could you go back and let us
15
know some information about these baseline
16
demographics and exactly how the sites were
17
selected? I'm just curious to know, especially
18
within the U.S., knowing that HIV is not evenly
19
distributed amongst states and regions.
20
So I'm just curious to know how your sites
21
were selected, the 94 sites, and if you could just
22
kind of give me some details on the states, rural,
A Matter of Record (301) 890-4188
257
1
urban, and if that led to the disproportion of
2
African Americans in this study.
3
DR. BRAINARD: I'll ask Dr McCallister to
4
come and describe our site selection process.
5
While he's coming to the podium, I will say that
6
the DISCOVER study enrolled 9 percent overall
7
African American subjects. Within the U.S., that
8
proportion was 13 percent. As I believe was
9
pointed out earlier today, the population that we
10
enrolled into DISCOVER was largely reflective of
11
people taking PrEP today. It was not reflective of
12
the people who are at highest risk for new
13
infections right now.
14
Slide 2 up, please. This slide shows the
15
percentage of blacks and Hispanic and Latino
16
individuals enrolled in the DISCOVER trial on the
17
left as compared to the percentage of blacks and
18
Hispanic or Latino individuals taking PrEP today in
19
the U.S. As you can see, our proportions were
20
similar for DISCOVER for black participants, and we
21
enriched somewhat for participants who
22
self-identified as Hispanic or Latino.
A Matter of Record (301) 890-4188
258
1
I'll have Dr McCallister speak to the
2
efforts we took to enroll a diverse range of sites
3
in the study.
4
DR. McCALLISTER: We did specifically seek
5
out sites that were in high background HIV
6
incidence areas, in the U.S., Canada, and in
7
Europe. In so doing, we went to -- almost all of
8
them were urban centers, and they were in hospital,
9
in STI clinics, and health departments.
10
Within the U.S. population that wound up in
11
DISCOVER, we had a large percentage that were in
12
the northeast and southeast in particular. One of
13
the findings that has come out of our attempt to
14
understand what the background epidemiology was of
15
our sites, we used CDC data to get the HIV
16
incidence rate in these sites, and then compared it
17
to places where DISCOVER was conducted.
18
Could I get slide 1 up, please. These data
19
are incidence rates over time at 25 metropolitan
20
statistical areas inside the United States that
21
overlapped with DISCOVER sites. These are
22
incidence rates in MSMs in those locations who were
A Matter of Record (301) 890-4188
259
1
not using PrEP. What you see is over time, the
2
general incidence rate in both the DISCOVER sites
3
as well as non-DISCOVER sites has gone down a bit,
4
but the DISCOVER sites were in places where the
5
incidence was higher consistently over time.
6
DR. BADEN: Dr. Giordano, you had a follow-
7
on?
8
DR. GIORDANO: Yes. Can you clarify if that
9
comparison was adjusted for the racial and ethnic
10
distribution of the participants matched for what
11
the distribution is in the MSAs, weight sampling,
12
in essence?
13
DR. McCALLISTER: Right. These data on the
14
screen are all people at risk with a CDC indication
15
within these MSAs. However, when you do break it
16
down racially, the numbers are very close. They
17
range from 3.3 to 4.2.
18
DR. GIORDANO: I'm not sure I understand
19
that.
20
(Laughter.)
21
DR. GIORDANO: In other words, what I'm
22
asking is does this comparison, where you're saying
A Matter of Record (301) 890-4188
260
1
these are high-risk people in high-risk areas, yes,
2
they're in a high-risk area, but are they from a
3
racial and ethnic group, that is at high risk in
4
that area? So half of the HIV in Houston is in
5
African Americans right now. If you only enrolled
6
white people in Houston, you would get a lower rate
7
of HIV incidence than you would otherwise expect.
8
Does this adjust for that difference?
9
DR. McCALLISTER: It doesn't adjust -- it is
10
inclusive of all people in these MSAs.
11
DR. GIORDANO: So the answer is no.
12
DR. McCALLISTER: It's not adjusted --
13
DR. GIORDANO: Thank you.
14
DR. McCALLISTER: -- specifically just for
15
African Americans; that's correct. The rate in
16
African Americans and the rate in Caucasians from
17
these locations are very close to these numbers.
18
DR. GIORDANO: Am I being obtuse? Is he
19
being obtuse? We're not communicating.
20
DR. BADEN: The point has been made.
21
DR. GIORDANO: Okay. Thank you.
22
DR. BADEN: Dr. Le, did you have a
A Matter of Record (301) 890-4188
261
1
follow-on?
2
(Dr. Le gestures no.)
3
DR. BADEN: Okay. Dr. Goetz, a follow on?
4
DR. GOETZ: Yes. I'll try to follow up on
5
what I think is Dr. Giordano's question. You had
6
presented data on MSA, 1 metropolitan statistical
7
area -- I believe that's what the MSA is -- would
8
be Houston and another one would be Boston. The
9
MSAs from which you recruited patients on average
10
have higher rates of HIV acquisition than other
11
MSAs.
12
But I think Dr. Giordano's question is, or
13
my question is, the patients enrolled in the study,
14
though, are they representative of the ratio makeup
15
of that MSA, and thus it would be predicted to have
16
that higher rate, or were patients who were
17
enrolled in this study be from populations of lower
18
risk, which gets back to the whole question of
19
what's the risk of the population enrolled and
20
thus, the efficacy of the intervention?
21
DR. BRAINARD: I think what your question is
22
driving at is how confident can we be that we were
A Matter of Record (301) 890-4188
262
1
in the right population with high risk for HIV, and
2
to address that, I'll ask Dr. Wulfsohn to come to
3
the podium and speak to the two ways we tried to
4
estimate the putative or potential placebo rate to
5
understand whether we were in the right population.
6
DR. WULFSOHN: Just to answer the question
7
directly, we did an analysis where we forced the
8
racial makeup in the MSAs to match that in
9
DISCOVER, and the rate went down by 0.3. It was
10
3.8 overall in 2017, and 3.5 when you force it to
11
match the racial makeup in DISCOVER. And we have
12
looked at another method of assessing the placebo
13
rate, and that's using the rectal gonorrheal
14
approach, which I can show.
15
Slide 1 up. There have been 8 different
16
cohorts within controlled trials of placebo
17
control. Each of the 8 black dots on this graph
18
represents a placebo cohort. What's notable is
19
that the higher the rectal gonorrheal rate, the
20
higher the HIV incidence rate in these placebo
21
cohorts, and that's a linear relationship.
22
On this graph, we've also superimposed the
A Matter of Record (301) 890-4188
263
1
DISCOVER data, so just above 20 on the X-axis,
2
you'll see 2 little dots, and these represent data
3
from DISCOVER, gray for Truvada and blue for
4
Descovy. For both arms, we have the erectile
5
gonorrhea incidence, as well as the HIV incidence.
6
These 2 dots with the vertical confidence
7
intervals, which are hard to see because they're so
8
tight, are well below what the projected placebo
9
incidence would have been, and that gray area is
10
the 95 percent prediction interval around the
11
placebo rate.
12
DR. BADEN: Dr. Smith, did you have a
13
follow-on?
14
DR. SMITH: [Inaudible - off mic].
15
DR. BADEN: Microphone.
16
DR. SMITH: I had a question about the MSA
17
slide.
18
DR. BRAINARD: We'll pull that up for you;
19
just a sec.
20
Could we get the MSA slide, please?
21
DR. SMITH: Remind me the years in which the
22
DISCOVER trial was actually happening, '16 to '17
A Matter of Record (301) 890-4188
264
1
or '15 to '17?
2
DR. WULFSOHN: It started at the end of '16,
3
and it was largely in '17 as the bulk of the
4
follow-up.
5
DR. SMITH: So the incidence was falling in
6
both sets of communities before the start of the
7
study, and you really only have the last two time
8
points that are presumably related to the DISCOVER
9
trial?
10
DR. WULFSOHN: That's correct. If I could
11
have the slide on the fold increase relative to
12
placebo from Truvada, with the MSAs? When we
13
designed the study relative to the three historical
14
studies that were used for the design, we expected
15
Truvada to lower incidence 5-fold. So 1.44 was
16
expected for Truvada versus 6.96 in placebo from
17
the three studies.
18
Slide 1 up. When you look at the actual
19
data from DISCOVER, we're noticing that Truvada is
20
lowering the incidence by actually 8.6 fold if you
21
were to use the MSA data; that's this middle CDC
22
estimate of the placebo rate. The placebo rate we
A Matter of Record (301) 890-4188
265
1
estimated to be 3.83 during the duration of the
2
study versus the USA subset of DISCOVER, where the
3
observed Truvada incidence was 0.446.
4
Our active control was actually
5
substantially more active than we anticipated.
6
Using the rectal gonorrhea, it's actually 19-fold
7
reduction that we're seeing with our active
8
control, Truvada. So DISCOVER was actually a
9
better test of a new agent than we anticipated it
10
to be.
11
DR. SMITH: Okay.
12
DR. BADEN: Dr. Dodd?
13
DR. DODD: Yes. Was it really a better test
14
or was it just that the prevalence of circulating
15
HIV in the populations tested might have been
16
lower, and therefore exposure to HIV may have been
17
lower? I think you made the case that their
18
at-risk behavior was relatively high, but how do we
19
know that the gonorrhea curves that you showed and
20
the really low rates in the DISCOVER cohort weren't
21
just really because there was lower exposure to
22
HIV?
A Matter of Record (301) 890-4188
266
1
DR. WULFSOHN: If I could have the
2
gonorrheal slide back? If you look at the lower
3
bound of the interval around the projected -- slide
4
1 up. If you look at the lower bound of the
5
interval around the HIV incidence that we projected
6
for placebo, while we projected an incidence
7
slightly above 6, the lower bound is slightly above
8
3. So even in a conservative way of looking at
9
this, there is a big gap between how placebo would
10
have performed relative to how these two agents are
11
performing.
12
DR. BADEN: Thank you. We're close to our
13
break, but before we go to break, which I'll
14
shorten to 10 minutes, one last question on
15
ciswomen. Separate from this committee's
16
deliberation and the agency's action, what is your
17
commitment to studies in ciswomen in terms of
18
generating the data that are absent?
19
DR. BRAINARD: We're firmly committed to
20
generating data in women. As Moupali showed in her
21
presentation, we've got a number of studies that
22
we're supporting, that we're hoping to initiate
A Matter of Record (301) 890-4188
267
1
within the next year. These are not traditionally
2
powered for efficacy studies; these are clinical
3
effectiveness studies, and they are planned to be
4
conducted both in the U.S. as well as in high
5
incidence settings within Africa to demonstrate the
6
safety as well as the clinical efficacy across a
7
broad range of populations.
8
In addition, we are committed to generating
9
clinical data with Descovy for PrEP in women using
10
one of these novel approaches if we can come to an
11
agreement on what that approach should be.
12
Dr. Wulfsohn walked through some of the ideas.
13
We're in active discussions with investigators and
14
with experts on how to best get this done, and
15
we're committed to do it, and we're planning to
16
incorporate the feedback that we receive from FDA
17
and from the panelists into this decision.
18
DR. BADEN: So whether or not the indication
19
is granted, you will conduct studies in ciswomen to
20
determine the effectiveness.
21
DR. BRAINARD: Without a doubt.
22
DR. BADEN: And that's the hundred, in the
A Matter of Record (301) 890-4188
268
1
tens, hundreds, or thousands? I'm just looking for
2
a zip code.
3
DR. BRAINARD: The indication allows us to
4
go more broadly into clinical effectiveness
5
demonstration projects, so it clearly allows us to
6
get to a higher number and reach a higher number of
7
women more quickly because we have endorsement from
8
a regulatory agency that this drug is safe and
9
effective in the population.
10
If we don't have an indication, we're still
11
generating data in women, but the nature of that
12
type of data has to be restricted until we get the
13
endorsement from the regulatory bodies that we can
14
then go and do these demonstration projects. So
15
we're committed, we're going to generate data, and
16
I think that the proportion and maybe the velocity
17
of that data depends on where we land, but the
18
commitment is there, and it will happen. The time
19
period is it just depends.
20
DR. BADEN: Understood the constraints you
21
have to work under.
22
It's 3:04. We will take a break and resume
A Matter of Record (301) 890-4188
269
1
at 3:15 sharp.
2
(Whereupon, at 3:04 p.m., a recess was
3
taken.)
4
DR. BADEN: [Inaudible - mic off] -- before
5
that, we need to clarify as much as we can from the
6
applicant.
7
We have several committee members who still
8
have questions from this morning. I will ask the
9
committee members, as well as the applicant, to be
10
as pointed as possible in the question and the
11
response so that we can cover as much ground in the
12
next 15-20 minutes before we have to get to
13
discussion about the questions at hand.
14
I'm going to start with questions from this
15
morning. Dr. Daskalakis, you are on the list.
16
(Dr. Daskalakis gestures no.)
17
DR. BADEN: Thank you. Dr. Green?
18
DR. GREEN: Yes. Thank you. I have a
19
question that relates to the slide CC-50 from this
20
morning, which was the forest plot looking at the
21
different subgroups. I know there's been
22
conversation. I thought that adolescence was part
A Matter of Record (301) 890-4188
270
1
of the populations that you were contemplating,
2
including in your request for indication.
3
I wonder if you could explain, because the
4
closest thing we have to adolescence is the age
5
less than 25, and it's one of the only two data
6
points on this curve that show a favoring to TBD,
7
although not clinically significant, so maybe you
8
could just comment on that.
9
DR. BRAINARD: Yes. I'll first make the
10
point that there are wide confidence intervals
11
around this point estimate related to the
12
relatively small sample size as compared to the
13
entire study design. The incidence rates within
14
the population of participants who are less than 25
15
are higher than the overall incidence rates.
16
This is related to the relationship between
17
younger age and lower adherence, which has been
18
demonstrated in many PrEP studies and certainly
19
demonstrated in the adolescent ATN study with
20
Truvada, and reflects that those participants had
21
lower adherence in that age bracket. However, I
22
would point out that both Descovy and Truvada
A Matter of Record (301) 890-4188
271
1
nevertheless were highly effective and
2
substantially lowered the risk of HIV acquisition.
3
DR. BADEN: Thank you. A follow-on to that,
4
safety in the younger ones, I have learned that
5
adolescence is defined by weight of 35 kilograms.
6
The data you have on how low an age bound, you have
7
data. Do you have data on 10 year olds on Truvada,
8
12 year olds, 20 year olds? I just want to have
9
some sense of where we're are in the data-free zone
10
if we may be giving it to our 10 year olds.
11
DR. BRAINARD: We have Truvada, Descovy, and
12
then three other single-tablet regimens that
13
contain Descovy, as well as multiple regimens
14
containing Truvada, are indicated for adolescents
15
greater than or equal to 35 kilograms. And then we
16
also have indications in younger populations based
17
on the data that we've generated in treatment
18
trials.
19
So we have a fairly large body of evidence
20
to suggest that the Descovy-based therapy is safe
21
and well tolerated in these younger populations,
22
even extending less than 35 kilograms.
A Matter of Record (301) 890-4188
272
1
DR. BADEN: When you say younger, is that
2
from 0 to 10 years old in treatment?
3
DR. BRAINARD: In the setting of HIV
4
treatment, our youngest -- I think our lowest
5
weight indication is 25 kilograms.
6
DR. BADEN: Again, any labeling is going to
7
need to take into consideration the absence of
8
data.
9
DR. BRAINARD: I'm getting a signal that the
10
age cutoff is 6 --
11
DR. BADEN: Six.
12
DR. BRAINARD: -- so down to age 6.
13
DR. BADEN: Okay. So you have safety data
14
down to that obviously with indication.
15
Dr. Goetz, do you have a follow on?
16
DR. GOETZ: Not a follow-on.
17
DR. BADEN: So moving down, Dr. Gripshover,
18
you have a question from this morning.
19
DR. GRIPSHOVER: Actually, yes. I had one
20
question about weight, because we did see one slide
21
from the audience earlier, too, especially in the
22
HIV treatment world where being concerned with
A Matter of Record (301) 890-4188
273
1
obesity, they think some may be related to TAF,
2
some also the ACE inhibitors. I think in this
3
study they gained a kilogram in the men. It seems
4
that sometimes women gain more weight.
5
So I just wondered if we have any data maybe
6
in women on TAF outside of this if we're trying to
7
extrapolate this to a broader population of women.
8
DR. BRAINARD: I'll ask Dr. Das to come in
9
and discuss the weight gain in the DISCOVER study
10
and place it into context around what we know from
11
other PrEP trials. I'll also note that the data we
12
have from our HIV treatment setting suggests that
13
there are many factors associated with weight gain,
14
integrase inhibitor therapy being one of them, and
15
that's seen across different integrase inhibitors.
16
TAF in and of itself is not associated with
17
weight gain. What we see in the HIV treatment
18
space is that TDF is associated with a weight
19
suppressive effect, and when TDF is switched to
20
either a TAF-based regimen or a regimen without TAF
21
that doesn't contain TDF, that can be associated
22
with weight gain.
A Matter of Record (301) 890-4188
274
1
DR. DAS: The difference in weight in the
2
DISCOVER study was driven by the TDF weight
3
suppressive effect that Diana just discussed.
4
Slide 2 up, please. We've known about the
5
potential for Truvada to potentially suppress
6
weight since the iPrEx trial. The USPI has weight
7
loss as a known adverse drug reaction for Truvada,
8
based on the iPrEx trial.
9
On the left-hand side, you see placebo
10
across the top and Truvada across the bottom, and
11
you see that with iPrEx, there was a weight loss
12
through week 48 with Truvada and a weight gain on
13
placebo. This is in median percent changes in
14
weight. In DISCOVER, the Truvada arm looked very
15
similar to the iPrEx arm with initial weight loss
16
and a little bit of stabilization towards the end,
17
and the Descovy arm looked very similar to the
18
placebo.
19
The average placebo weight gain -- excuse
20
me. The average amount an American age 18 to 40
21
gains in a year is 1 kilogram, and the placebo
22
weight gain in the iPrEx trial and the weight gain
A Matter of Record (301) 890-4188
275
1
in the DISCOVER trial on the Descovy arm are
2
consistent with that. Further, if we look at
3
HPTN 077, at 41 weeks, the placebo arm also gained
4
about 1 kilogram. The cabo arm in that trial also
5
gained 1.1 kilogram.
6
So I think what we're seeing in trials that
7
compare TDF to TAF is the TDF weight suppression or
8
stabilization effect versus the release of that
9
effect in switch or the lack of that effect in the
10
TAF arm.
11
DR. BADEN: Thank you. Dr. Le?
12
DR. LE: Can you please go back to slide
13
CC-50 that you had earlier n in the subgroup
14
analysis of those less than 25 years? You alluded
15
to that this may have been where the incidence rate
16
is a little bit higher than the overall -- was
17
perhaps due to adherence as a reason for this.
18
What was the adherence for that group, and
19
was it similar to the treatment trials that you see
20
in adolescents? I'm trying to correlate this, for
21
younger people would we see the same trends?
22
DR. BRAINARD: I'll ask Dr. McCallister to
A Matter of Record (301) 890-4188
276
1
address the issue of adherence by age within the
2
DISCOVER trial, and I will say that, overall, we've
3
seen lower levels of adherence within studies of
4
PrEP in adolescents and younger individuals.
5
That was really one of the drivers for why
6
we didn't include adolescence in the DISCOVER
7
trial, was because of the data suggesting that they
8
really benefit from an increased visit frequency.
9
They're going to benefit from increased
10
interventions to improve adherence and have age
11
appropriate retention and recruitment methodology.
12
DR. McCALLISTER: Adherence in the
13
individuals less than age 25 was lower than in
14
those above age 25.
15
Could I get the slide 1 up please? This is
16
the pill count data that is broken down by less
17
than 25 years on the left, 25 to 50 in the middle,
18
and above age 50 on the right. These are, as you
19
can see, far lower for those less than age 25.
20
Another way of looking at it is through the
21
dried blood spot data, so slide 3 up, please, and
22
we really see the same pattern in the less than 25
A Matter of Record (301) 890-4188
277
1
using the TFV diphosphate levels in RBCs. There
2
were fewer of them in a range of 4 tablets per week
3
or higher. Of the 22 infections in DISCOVER, 7 of
4
them did occur in this group, and all 7 of them did
5
not have the detectable drug levels.
6
DR. BADEN: Thank you. Dr. Read from
7
earlier in the day.
8
DR. READ: Yes, my questions have already
9
been addressed. Thanks.
10
DR. BADEN: Dr. Giordano from earlier in the
11
day.
12
DR. GIORDANO: I have a question for the
13
agency. Is it within your -- two questions
14
actually for the agency. One is, is it within your
15
purview to say a registrational study should
16
include X proportion of people in Y category? In
17
other words, let's say a certain proportion are
18
black, African American, from U.S.. Is that
19
something you can say or is it really up to the
20
sponsor to design that?
21
DR. BIRNKRANT: We can make the
22
recommendation, but we wouldn't want to hold up a
A Matter of Record (301) 890-4188
278
1
trial or an approval if they didn't meet what the
2
suggested rate would have been in that certain
3
population.
4
DR. GIORDANO: Another question is, this
5
request to approve based on essentially drug level
6
extrapolation for women, do you have other examples
7
of when the agency has allowed that to happen? Can
8
you give us any guidance on when that's appropriate
9
or considered inappropriate at the agency's level
10
to help inform the committee?
11
DR. BIRNKRANT: I don't think we have any
12
other examples, based on --
13
DR. MURRAY: The tissue level, we
14
extrapolate efficacy for children all the time, and
15
we still get the safety data. So we've matched
16
efficacy in different populations based on systemic
17
PK. I don't think we've ever made a regulatory
18
approval decision based on a tissue a non-systemic
19
PK argument.
20
DR. BADEN: And presumably the prior
21
decision, you inferred the correlate of protection,
22
so to speak, an antibiotic level in blood, where
A Matter of Record (301) 890-4188
279
1
there's an understanding of what the protective
2
moiety supposedly is.
3
DR. MURRAY: We've always tried that, and
4
we've tried to match it to be as much
5
bioequivalent -- I use that term loosely -- to the
6
population that had the clinical data.
7
DR. BADEN: Yes. Dr. Goetz?
8
DR. GOETZ: That leads me to what I think I
9
can call a follow-up question. I want to go back
10
to one of the backup slides that was shown, which
11
showed correlation between dosage inferred from PBM
12
of red blood cell spots and protection. I think
13
that was BU461, is what I wrote down this morning,
14
and that was in the iPrEx study.
15
What I was interested in is trying to build
16
this bridge, which may or may not be buildable.
17
Are there similar data that are inferred based on
18
PBMCs or RBC studies in women that correlate the
19
same level of protection to 2 to 3 tablets per week
20
as being the cutpoint?
21
DR. BRAINARD: I'll ask Dr. Anderson to
22
address this question about the thresholds for
A Matter of Record (301) 890-4188
280
1
adherence for women and for men.
2
DR. ANDERSON: I would say not this level
3
and formal analysis in women. We just don't have
4
that yet. We do have, though, the -- if I can show
5
O82, perhaps. There's a very recent study
6
HPTN 082; it was in women.
7
Slide 2 up, please. This study is one of
8
the very few studies in women that have collected
9
dried blood spots or a marker where you can tell
10
different gradients of adherence, and this study
11
did actually collect those. They had 4 infections
12
in this study, and none of those infections
13
occurred at the middle or the high drug, the DBS
14
level. They all occurred at the low level.
15
DR. GOETZ: So aside from this sparse data
16
set, there are no data at your disposal that allow
17
us to map adherence -- a proxy for taking drug
18
based on a biological measure that correlates, in
19
some degree, with drug levels to protection in
20
women, and shows equivalence between the level of
21
protection that we expected in men with that level
22
and the level of protection demonstrated in women,
A Matter of Record (301) 890-4188
281
1
because that's the bridge that we're trying to
2
build, I think.
3
DR. ANDERSON: I think these results here on
4
the screen are consistent with what we saw. And
5
iPrEx OLE, it's a smaller data set, but it is
6
consistent; I would say that. And I think you had
7
something to add.
8
DR. GOETZ: Wide confidence interval.
9
DR. BRAINARD: I would also say that in the
10
Partners PrEP study, there was an assessment of
11
adherence based on tenofovir blood levels. And
12
unlike tenofovir diphosphate within red blood
13
cells, which is an integrated assessment of
14
adherence over 6 to 8 weeks, tenofovir plasma
15
levels reflect dosing within the last 4 days.
16
This is a measurement of adherence. It's
17
less precise, but it does offer an objective
18
assessment. And it is referenced in the new CDC
19
guidance as a meaningful assessment of what they
20
call recent PrEP use, which they correlate as
21
associated with a 90 percent protection for both
22
men and for women. Within that case-controlled
A Matter of Record (301) 890-4188
282
1
study, Partners PrEP, where they looked at both men
2
and women who had detectable tenofovir diphosphate
3
levels -- I'll put slide 3 up please -- the overall
4
efficacy was 92 percent, and in men, it was 89
5
percent, and in women, it was 94 percent.
6
So this represents a lower level of
7
adherence than, for example, we saw in the DISCOVER
8
trial. But nevertheless, it shows that there's no
9
difference between men and women.
10
DR. BADEN: Thank you.
11
We've made it through the list. Are there
12
other questions from the committee? We're not all
13
satisfied given the nature of the data, but are
14
there other questions that could help inform the
15
committee in our deliberations?
16
(No response.)
17
Questions to the Committee and Discussion
18
DR. BADEN: If not, we'll now proceed
19
with -- don't go yet to the questions to the
20
committee, but thank you. We'll now proceed with
21
the questions to the committee and panel
22
discussions. I'd like to remind the public
A Matter of Record (301) 890-4188
283
1
observers, while this meeting is open for public
2
observation, public attendees may not participate
3
except at the specific request of the panel.
4
I would like to thank Dr. Brainard and the
5
entire Gilead team for covering an incredible
6
amount of information. Given the size of the
7
problem, the amount of data could never approach
8
the magnitude of the problem. We were able to get
9
through I think about 15 percent of the slides you
10
had prepared. If I'm reading the lower right-hand
11
corner correctly, you have at least
12
1500-1600 slides. I think we got 150 to 200 of
13
them in front of us. So thank you for preparing
14
the information and sharing it with us.
15
Now we must turn to the questions at hand.
16
Before we move to the questions at hand, I have
17
some guidance I would like from the agency, and if
18
others have questions, let me know.
19
We're being asked -- and would be interested
20
in the agency's guidance, too -- particularly in
21
the cisgender women conundrum, I want to make sure
22
I understand the problem correctly. There are
A Matter of Record (301) 890-4188
284
1
multiple studies with Truvada. At least two showed
2
no benefit; two showed benefit. One of them led to
3
the indication. However, these data were not
4
strong enough to allow a determination of a study
5
design with a noninferiority margin, yet these data
6
are strong enough to guide us with a bridging study
7
to lead to an indication.
8
Is that the position we're sort of in as
9
we're reflecting on how to move forward with our
10
deliberations?
11
DR. MURRAY: That's correct. Even though we
12
had low efficacy in some studies, it was attributed
13
to low or no adherence, but we think if women are
14
adherent, that they should be 90 percent effective.
15
DR. BADEN: But that wasn't strong enough to
16
set a noninferiority margin so you could have a
17
female trial analogous to a male trial.
18
DR. MURRAY: Well, noninferiority studies
19
are tricky; rely on historical data you're supposed
20
to use as much as possible. I think the problem
21
with noninferiority studies is you need that
22
constancy assumption. You need to assume that what
A Matter of Record (301) 890-4188
285
1
you saw in the past is going to be repeated going
2
forward, and for studies in Africa, we're not sure
3
what the adherence rate is going to be, and then
4
that really hampers our ability to do a
5
noninferiority margin unless we did some novel way
6
of looking at noninferiority margins, which we
7
haven't done, Bayesian based on adherence and
8
things that we've really never looked at.
9
DR. BADEN: What tools do you have if broad
10
indications were given to mandate or require future
11
studies versus goodwill and intent to do future
12
studies?
13
DR. MURRAY: Well, obviously, I think this
14
would be a postmarketing commitment. Requirements
15
are for pediatric studies and for safety. This is
16
really expanding indications, so it would fall
17
under kind of a legal postmarketing commitment.
18
But those studies, particularly in the HIV arena,
19
are almost always completed, especially where
20
they're important, like this would be, to expand
21
the indication to women.
22
Anybody else want to comment on that?
A Matter of Record (301) 890-4188
286
1
(No response.)
2
DR. BADEN: Any other clarifying -- we have
3
to deliberate -- sorry. Dr. Green?
4
DR. GREEN: So again, I'm going to be the
5
pediatrician. On the packet that you have, it
6
looks like the application -- again, it does
7
include adolescents weighing at least 35 kilograms,
8
but I noticed that neither question 1 nor
9
question 2 addressed our opinion on adolescence.
10
So you're not interested in any opinions
11
from the committee on adolescents?
12
DR. MURRAY: Yes, we were planning to ask
13
that question. We were willing to extrapolate to
14
adolescents based on what's known for PK and safety
15
for the treatment and the fact that there's an
16
indication in adolescence for Truvada. We're
17
willing to kind of make that leap for the same
18
gender in adolescence, because it's the route of
19
transmission that we think could be the variable or
20
acquisition.
21
DR. BADEN: Dr. Giordano?
22
DR. GIORDANO: Does an indication have to
A Matter of Record (301) 890-4188
287
1
specify sex or can it specify behavior? So
2
approved for men who have sex with men or can it be
3
approved for men -- but does it have to be approved
4
for men? Do you see the distinction I'm making?
5
Is that something within the labeling options?
6
DR. MURRAY: It is. Are you talking about
7
MSM and heterosexual men, or those who have
8
insertive intercourse with women, or men who have
9
sex with men?
10
DR. GIORDANO: Before we get to that
11
discussion, because there's no -- essentially --
12
DR. MURRAY: Yes. It gets a little bit
13
tricky. But if the indication was limited just to
14
MSM, we'd really have to think about how the
15
indication would be worded for men in general.
16
DR. GIORDANO: Right.
17
DR. BADEN: Dr. Daskalakis?
18
DR. DASKALAKIS: Another labeling question.
19
On a label, are you able to say that this drug has
20
been studied in these populations; there's a
21
recommendation for use in another population, but
22
it's based on extrapolation? Is that something
A Matter of Record (301) 890-4188
288
1
that can be explicitly stated in the label?
2
DR. MURRAY: I think so. We do that, to a
3
certain extent, when they describe pediatric data.
4
DR. BADEN: Dr. Smith, do you have a
5
question?
6
DR. SMITH: Yes. Is it possible to discuss
7
the MSM indication separate from the transgender
8
women recommendation? Right now, they're in a
9
single statement, and I think I have questions
10
about one but not the others.
11
DR. MURRAY: Well, we didn't plan to have
12
the question answered that way, but you might have
13
that as a comment after your vote. But I think if
14
we're prepared to go ahead with MSM, the agency was
15
prepared to go ahead with the transgender women as
16
well, realizing that you're not going to be able to
17
do a powered study in transgendered women. There
18
were zero seroconversions out of 74 probably
19
indicative of some protection in and of itself in
20
the DISCOVER trial.
21
DR. BADEN: But Dr. Smith, you're getting at
22
just the power issue, given the population sizes.
A Matter of Record (301) 890-4188
289
1
DR. SMITH: Yes. I mean, if you look at the
2
iPrEx subset analysis that had 200 transgender
3
women defined slightly differently, there was no
4
evidence of the impact, statistically significant
5
evidence of protection.
6
So to me it's an extrapolation question. I
7
mean, they didn't include enough transgender women
8
in order to do a separate analysis, and now we're
9
asking to make an indication based on the fact that
10
it works for MSM. It was just my question.
11
DR. MURRAY: I think we're going to have it
12
voted on as a package deal, and then you can
13
explain why or why not you voted for it or not.
14
And if that's one of the issues, you can explain
15
that.
16
DR. BADEN: I think we'll vote on the
17
questions as written, but I think your point is the
18
guiding principle. We can then explain our
19
concerns or our reinforcements of how we look at
20
the different indications. After we vote, the
21
agency finds our comments even more helpful than
22
our vote. So it's very important that we'll vote,
A Matter of Record (301) 890-4188
290
1
which looks yes/no, but in reality, we can express
2
different elements that we find reassuring or
3
concerning where they should pay attention to.
4
I will be mindful of time, so we have about
5
50 minutes, and we all have been very energetic,
6
and it's a complex arena for all the reasons
7
discussed earlier.
8
Any other discussion amongst the committee
9
before we move to the vote? Are there any aspects
10
of the data or what would charged with that it
11
would be helpful to discuss or clarify?
12
(No response.)
13
DR. BADEN: If not, we can move to -- I can
14
read -- we will be using an electronic voting
15
system for this meeting. Once we begin the vote,
16
the buttons will start flashing. It is a new
17
system, so hopefully we won't get confused.
18
(Laughter.)
19
DR. BADEN: They'll continue to flash even
20
after you have entered your vote. Please press the
21
button firmly that corresponds to your vote. If
22
you're unsure of your vote or you wish to change
A Matter of Record (301) 890-4188
291
1
your vote, you may press the corresponding button
2
until the vote is closed.
3
After everyone has completed their vote, the
4
vote will be locked in. The vote will then be
5
displayed on the screen. The DFO will read the
6
vote from the screen into the record. Next, we'll
7
go around the room and each individual who voted
8
will state their name and vote into the record.
9
You can also state the reason why you voted as you
10
did if you want to. We'll continue in the same
11
manner until all the questions have been answered.
12
We will now move to the first question, and
13
I will ask if there are any questions about the
14
question before we vote. Has the applicant
15
provided substantial evidence of the safety and
16
effectiveness of Descovy for pre-exposure
17
prophylaxis, PrEP, to reduce the risk of
18
sexually-acquired HIV-1 infection in men and
19
transgender women who have sex with men?
20
If yes, provide your rationale. If no,
21
provide your rationale and list what additional
22
trials are needed. Please provide any additional
A Matter of Record (301) 890-4188
292
1
comments or thoughts on your vote. If yes, you can
2
still have a rationale about studies that are
3
needed.
4
So any questions about the question?
5
(No response.)
6
DR. BADEN: If not, then let's proceed to
7
voting.
8
(Voting.)
9
DR. BADEN: I assume the voting from our
10
online member is being handled. Okay, so that is
11
being handled. So I'll wait until you close
12
the --
13
DR. HOTAKI: For the record, the vote is 16
14
yes, two nos, zero abstentions, zero no votes.
15
DR. BADEN: We will now go around the room
16
and state your name and your vote into the record.
17
And if you have comments to the agency, please
18
share them. We'll start with Dr. Goetz.
19
DR. GOETZ: Thank you. Matthew Goetz. I
20
voted yes, that the DISCOVER trial supports the
21
approval of the Descovy, et cetera. I think the
22
word "support" is totally appropriate here because
A Matter of Record (301) 890-4188
293
1
it certainly supports the efficacy, and I use the
2
word "efficacy" appropriately as well.
3
I think what is really needed to enhance
4
this are the phase 4 trials to show the
5
effectiveness in real-world populations that span
6
transgender men -- I think I'm getting my
7
phraseology right here or I mean to you -- and also
8
in other populations and larger populations of
9
African American men, and populations where all
10
patients are to be fully adherent to PrEP.
11
The population that was tested here was
12
gratefully a highly adherent population, and we saw
13
a few infections. The real world I'm afraid
14
includes individuals who are less adherent, and
15
it's very important to demonstrate the
16
effectiveness in other populations that may face
17
challenges not seen in individuals who enrolled
18
here. Certainly, you want to see long-term safety
19
outcomes to see whether the biological signals that
20
favor TAF lead to clinical outcomes that are
21
favorable as well. I can go on, but I should leave
22
my panelists to say more.
A Matter of Record (301) 890-4188
294
1
DR. BADEN: Dr. Smith?
2
DR. SMITH: I voted yes because I think
3
there is substantial evidence to support an
4
indication for men who have sex with men. I am not
5
convinced that there's substantial evidence for
6
transgender women, and I think that additional
7
studies are going to be necessary, as my colleague
8
said, to understand how this is actually used in
9
populations that are at the highest risk of HIV
10
acquisition and who stand to benefit from it.
11
Adolescents, black men and women, and
12
transgender persons all have documented adherence
13
problems with Truvada, generally, and I think it
14
will be important to understand how TAF adds
15
protection or not in those populations.
16
DR. BADEN: Thank you. Dr. Read?
17
DR. READ: I voted yes, and my comments
18
largely have already been stated, but I think they
19
bear repeating. I think the data provided by the
20
applicant do support the safety and efficacy of
21
Descovy by demonstrating noninferiority to Truvada
22
in men who have sex with men and transgender women.
A Matter of Record (301) 890-4188
295
1
I think although there were a few infections
2
in the trial, the high rates of STI infections and
3
other indicators do support the high risk
4
characterization of the study population. And
5
further, Descovy appears to be safe as demonstrated
6
both in DISCOVER as well as the extensive treatment
7
experience in people living with HIV.
8
I do think that it has been stated
9
throughout the course of the day that the study
10
population enrolled in DISCOVER did not represent
11
the populations most at risk for HIV, and
12
therefore, if Descovy is approved for use in MSM
13
and transgender women, the applicant should be
14
required to collect postmarketing data on safety
15
and effectiveness in those underrepresented
16
populations, including transgender women, as has
17
just been stated, as well as people of color.
18
I think it's important, as was raised during
19
the public comment period, that the labeling and
20
advertising for Descovy, if approved, should only
21
speak to the noninferiority, not the superiority of
22
both the effectiveness as well as the safety of
A Matter of Record (301) 890-4188
296
1
Descovy. I think it's important to note that the
2
markers for kidney and bone toxicity were
3
biomarkers only and did not indicate a clinical
4
benefit. And I also think that it's important not
5
to disregard some of the potential negative adverse
6
events, including weight gain and lipids.
7
DR. DASKALAKIS: I'm Demetre Daskalakis. I
8
also voted yes. Mirroring some of the prior
9
comments, I think that the data presented in the
10
DISCOVER trial are very strong for supporting the
11
noninferiority of Descovy for pre-exposure
12
prophylaxis in men who have sex with men. I do
13
want to state again the importance of selling this
14
as a noninferiority both from efficacy and safety.
15
I think overselling the safety here could
16
create an environment where drug switches are done
17
in a way that don't reflect the data and may also
18
create significant disparities in various
19
populations of men who have sex with men.
20
My expectation of this approval is that it
21
should be marketed responsibly from the perspective
22
of not creating these disparities and having
A Matter of Record (301) 890-4188
297
1
Truvada be a drug for poor people and Descovy be a
2
drug for rich people, or for insured versus
3
uninsured. So I think it's really important that
4
we don't oversell the elements of noninferiority.
5
From the perspective of transgender
6
individuals, and I'm including transwomen and
7
transmen who have sex men on that list, I think
8
more data are necessary. I think that in the same
9
breath that we're going to probably discuss women,
10
we should also discuss transwomen and transmen and
11
the need and responsibility to actually get more
12
robust data.
13
Historically, the answer it's hard to do has
14
created a lot of disparity and mistrust of both
15
public health and research among transgender
16
individuals, so we need to work with strategies to
17
go beyond that rather than to stay with that.
18
Ultimately, then I think with the caveat of work to
19
do in the transgender population, I stand by my
20
vote of yes for noninferiority men who have sex
21
with men.
22
DR. GIORDANO: Tom Giordano. I voted yes
A Matter of Record (301) 890-4188
298
1
largely for the reasons that have already been
2
stated. I agree completely with the comments
3
already made. I will comment that I am not
4
convinced that we have enough data to say anything
5
about transgender women.
6
However, I did vote yes on that, including
7
that language, mainly because the biological
8
similarities is anal receptive sex primarily is the
9
risk factor. So I agree that there's sufficient
10
evidence, that that population probably would be
11
protected with this noninferior drug.
12
DR. DODD: Lori Dodd, and I voted no because
13
the question had the term "men and transgender
14
women," so my concern is really related to
15
transgender women. I agree with the comments said
16
previously, so I won't articulate further.
17
DR. BADEN: Dr. Walker?
18
DR. WALKER: Dr. Walker here. I voted no
19
for all the reasons that were expressed. According
20
to the CDC, more than 290,000 African Americans
21
with stage 3 HIV have died since the inception of
22
the HIV epidemic. As African Americans remain
A Matter of Record (301) 890-4188
299
1
disproportionately at risk for HIV, with gay and
2
bisexual men and heterosexual women being affected
3
more than any other race ethnicity, there was not
4
substantial or compelling evidence to indicate the
5
safety and effectiveness of Descovy for PrEP to
6
reduce HIV infection among this population. So
7
that's why I voted no.
8
As a public health researcher and a
9
community advocate, and an African American
10
heterosexual woman, I have alarming concerns
11
regarding the safety of Descovy, as well as the
12
sexual behaviors that will result from individuals
13
taking this drug. There was a lost opportunity to
14
provide data, substantial data, that is reflective
15
of the community in which its greatly impacted by
16
HIV. Furthermore, the data from the DISCOVER trial
17
failed to enough data on the prevention of HIV in
18
cisgendered women.
19
DR. BADEN: Thank you. Dr. Le?
20
DR. LE: I voted yes for this, for the
21
reason that the drug combination has demonstrated
22
noninferiority to Truvada and offers an alternative
A Matter of Record (301) 890-4188
300
1
for PrEP, which is critical in light of data
2
showing that only 7 percent of CDC's estimate of
3
1.1 million people in the United States with PrEP
4
indication actually received PrEP. Also, Descovy
5
may offer potential advantages of reduced bone and
6
renal toxicity. Despite voting yes, I do agree
7
that we need more information on transgendered
8
women.
9
DR. BADEN: Thank you. Dr. Burgess?
10
DR. BURGESS: Tim Burgess. I voted yes. I
11
think that data from the DISCOVER trial met the
12
noninferiority to Truvada, and just that, in men
13
who have sex with men.
14
DR. BADEN: Dr. Ofotokun?
15
DR. OFOTOKUN: Igho Ofotokun. I voted yes
16
for the same reasons that have been expressed by my
17
fellow committee members. I am convinced that
18
Descovy is noninferior to Truvada, and I think it
19
should be emphasized that this is a noninferiority
20
study.
21
Even though I voted yes, I am particularly
22
very concerned about the low number of non-white
A Matter of Record (301) 890-4188
301
1
participants in this study, and that should be
2
noted. I think if this moves forward, the agency
3
should strongly recommend a postmarketing study
4
that really include all this population, especially
5
men who have sex with men, black men who have sex
6
with men, who are most affected by this epidemic in
7
the U.S.
8
Again, I think, as has been expressed,
9
there's not enough transgender women to be able to
10
make a strong recommendation, but I believe, based
11
on the data, that it will be effective. And again,
12
this is another population that should be studied
13
should this approval move forward.
14
I think we should also emphasize the side
15
effects related to Descovy. It's sold as a safer
16
drug. I may be safer in some aspects, but there
17
are other aspects. For instance, the lipid profile
18
of Descovy is definitely something that should be
19
emphasized, and I am still concerned that the jury
20
is not yet out on the weight gain issue with TAF.
21
Thank you.
22
DR. BADEN: Thank you.
A Matter of Record (301) 890-4188
302
1
Lindsey Baden. I voted yes. We'll just
2
highlight some key issues. The continuum of the
3
body of evidence from the prior studies with
4
Truvada, with DISCOVER, it's a continuous set of
5
data that work well together and are very
6
reassuring that in MSM, it works very well.
7
I share the concerns in the population
8
studied, that's where we have the data.
9
Transgender were a very small subset, and then
10
other ethnic and racial backgrounds also have
11
limited representation, so that will just have to
12
be part of the consideration to grow the data set.
13
I think the weight and the lipids are not
14
trivial issues and can become significant over
15
years of treatment and perhaps consequence or not,
16
but that's where data and follow up will be
17
required.
18
Dr. Weina?
19
DR. WEINA: Peter Weina. I voted yes. I
20
believe there is substantial evidence of the safety
21
and effectiveness to reduce the risk of
22
sexually-acquired HIV in the indicated population.
A Matter of Record (301) 890-4188
303
1
Ignoring all the background politics, potential
2
gamesmanship, market pressures, whatever, this is
3
another approved product in our toolbox that gives
4
clinicians an option that we didn't previously
5
have.
6
While no package is ever ideal for all
7
potential patient populations, it actually was nice
8
to see a trial that was reasonably powered given
9
the targeted population, and in time when even more
10
is known about it, available to all patient
11
populations.
12
DR. BADEN: Dr. Green?
13
DR. GREEN: Michael Green. I voted yes. I
14
thought the data as presented clearly met the
15
criteria for noninferiority and have an equivalent,
16
if not superior, safety profile, though the impact
17
on lipid metabolism and weight gain might balance
18
out the bone density and renal benefits if they are
19
real. With the caveat that the study did not
20
include enough transgender women to allow subset
21
analysis, the study was generally well designed,
22
including a large cohort and robust follow-up.
A Matter of Record (301) 890-4188
304
1
If approved, the label should clearly
2
highlight the noninferiority performance of F/TAF
3
and not infer superiority. Safety claims should
4
highlight not only the potential benefits in terms
5
of renal and bone density but also the potential
6
increased risk related to lipid metabolism and
7
weight gain and obesity. Thank you.
8
DR. BADEN: Thank you. Not yet,
9
Dr. Gripshover. We have Dr. Lupole on the phone.
10
Do you have?
11
MS. LUPOLE: [Inaudible - distortion]
12
DR. BADEN: We're having trouble hearing
13
you. Now we can hear you.
14
MS. LUPOLE: All right. Can you hear me
15
now? [Inaudible - distortion].
16
DR. BADEN: That may not be working well.
17
Mute your computer while you speak is the advice
18
I'm given.
19
MS. LUPOLE: I'm sorry. What, sir?
20
DR. BADEN: That sounds good. What you just
21
did worked.
22
MS. LUPOLE: Okay, good. I voted yes. I
A Matter of Record (301) 890-4188
305
1
have concerns for transgender. I think more data
2
needs to be collected, but yes is the answer to the
3
question the way it was presented.
4
DR. BADEN: Thank you. Dr. Gripshover?
5
DR. GRIPSHOVER: I also voted yes. I
6
believe the DISCOVER trial showed the efficacy and
7
safety of Descovy to reduce the risk of HIV
8
acquisition in men. I think it's a little bit of a
9
stretch for transgender women, but I also agree
10
it's the same biologic, at least method, of
11
acquisition. But I think a small amount, but
12
statistically significant improvements in bone
13
mineral density and renal tubular function in TAF
14
versus TDF may be important in young adults
15
building bone and older ones losing it, or those
16
with other comorbidities and renal function.
17
However, for the vast majority of people,
18
TDF/FTC is safe, and I would not want those without
19
access to TAF due to geography or cost to forego
20
its benefit as PrEP, and I think we need to
21
emphasize that this was a noninferiority study.
22
DR. BADEN: Dr. Siberry?
A Matter of Record (301) 890-4188
306
1
DR. SIBERRY: George Siberry. I voted yes.
2
Like many before me, I think the trial adequately
3
provided evidence for a claim for noninferiority as
4
an alternative, both from an efficacy and a safe
5
clinically meaningful safety standpoint. I'd add
6
that the claim would include adolescence. I
7
strongly support the use of weight without age down
8
to 35 kilos and accept the ability to extrapolate
9
for adolescents in that claim. Thank you.
10
DR. BADEN: Dr. Swaminathan?
11
DR. SWAMINATHAN: Yes. I agree that there
12
was evidence of noninferiority as far as the
13
efficacy in MSM, but that the numbers were
14
insufficient to draw a clear conclusion about
15
transgender women.
16
Nevertheless, because I think the number of
17
variables that would have to be controlled for the
18
number of patients that would have been required to
19
be enrolled wouldn't really been feasible, and I
20
agree that it may have to depend on postmarketing
21
evaluations. But the way the question was phrased,
22
I agree that they did provide substantial evidence
A Matter of Record (301) 890-4188
307
1
of efficacy.
2
DR. BADEN: Dr. Cheever?
3
DR. CHEEVER: Laura Cheever. I voted yes,
4
and I think there is adequate evidence through the
5
DISCOVER trial for noninferiority. I am disturbed
6
that this far into the epidemic and this many
7
clinical trials, we still can't do trials in the
8
people most at risk in this country,
9
representatively, and that we really do need to be
10
looking at African Americans.
11
I echo other people talking about the lack
12
of transgender women really represented in this
13
trial. Once again, that needs to be looked at to
14
better understand the efficacy or noninferiority in
15
that population.
16
DR. BADEN: Thank you.
17
For question 1, it was 16 to 2, but even
18
those who voted no, there was a large consensus in
19
viewpoint that the data do support efficacy.
20
However, it's in the population studied, and there
21
was limited power in transgender and other key
22
populations, as well as some safety signal in
A Matter of Record (301) 890-4188
308
1
weight and lipids.
2
Those will all have to be carefully followed
3
and monitored in a postmarketing setting. The
4
database expanded in the key at-risk populations,
5
especially the transgender, and it's a
6
noninferiority, not superiority, on either of the
7
key issues.
8
Now we can move to question 2. Do the data
9
from the DISCOVER trial, in combination with the
10
available pharmacokinetic data and other previous
11
HIV-1 prevention trials with Truvada in cisgender
12
women, allow for the expansion of the DISCOVER PrEP
13
indication to include cisgender women?
14
If yes, please provide your rationale. If
15
no, please provide your rationale and list what
16
additional studies/trials are needed. Also comment
17
on the trial designs that would be adequate to
18
expand the indication. Please provide any
19
additional comments or thoughts on your vote.
20
Any questions about the question?
21
Dr. Siberry?
22
DR. SIBERRY: It's not simply asking whether
A Matter of Record (301) 890-4188
309
1
we would support expanding the indication, but
2
specifically saying do we think that the data are
3
the reason that we would expand it? Am I reading
4
that right? Because it's a little nuance there, I
5
think.
6
DR. BADEN: My read of this -- and the
7
agency can please correct me -- is do we believe
8
there are data establishing substantial efficacy
9
and safety in this population, which is cisgender
10
women, given the totality of the information
11
provided.
12
Is that the intent of the question?
13
DR. MURRAY: Yes.
14
DR. BADEN: Does that answer your question?
15
DR. SIBERRY: Yes.
16
DR. BADEN: If no other questions, then
17
let's vote.
18
(Voting.)
19
DR. HOTAKI: The online voter is being
20
handled; all done.
21
One more person needs to vote, so if
22
everyone can press theirs again.
A Matter of Record (301) 890-4188
310
1
DR. BADEN: Everyone repress your button.
2
(Pause.)
3
DR. HOTAKI: For the record, the vote is 8
4
yes, 10 no, zero abstention, zero no voting.
5
DR. BADEN: Please state your name and your
6
vote into the record. We'll start with
7
Dr. Cheever.
8
DR. CHEEVER: Laura Cheever. I voted no. I
9
really think that the company's demonstrated
10
difference in metabolism in TDF and TAF, and we
11
really do not know the protective factors for PrEP,
12
exactly how it works in the mechanisms. I know
13
that we do know that we have differences in
14
immunologic milieu between the vagina and the
15
rectal mucosa, so I have real concerns there about
16
what has been shown.
17
That said, I wanted to vote yes because the
18
thought of not having this indicated for women I
19
think will only further inhibit the implementation
20
of PrEP among women. So from a public health
21
perspective, I think there's probably more harm
22
than good not approving it for this indication, but
A Matter of Record (301) 890-4188
311
1
that wasn't the question that was asked.
2
So that's sort of how I split that. We've
3
talked about it all day long. The failure to
4
implement PrEP in women is huge. We keep glossing
5
over it and trying to just get to the adherent
6
women and throwing out all the rest, and I think
7
that is the wrong conversation to be having.
8
It's really about why women and why
9
transgender persons in youth and what we can do to
10
better get them to have protective effects of PrEP,
11
and whether that's different modalities or whatever
12
is part of that larger discussion that we weren't
13
having today.
14
DR. BADEN: Dr. Swaminathan?
15
DR. SWAMINATHAN: I voted no because -- I'll
16
just go through the reasons here. I think as far
17
as the question as to what the data allowed you to
18
conclude is what's key here. The cells that are
19
being infected in the vagina and cervix versus
20
those in the rectal or penile mucosa are not
21
clearly defined.
22
So although virus transcytosed in the
A Matter of Record (301) 890-4188
312
1
mucosal epithelium must infect dendritic or CD-4
2
cells resident in all tissues, the resident target
3
cell population at the time of exposure is the
4
local pool of T lymphocytes. This pool is known to
5
be relatively static, and more so in vaginal
6
tissues than in the GI tract. They are also
7
long-lived and replenished by local expansion.
8
Thus, the PK and PD in these lymphocytes may
9
not be the same as those in the peripheral blood or
10
other anatomic sites. We just do not know. Thus,
11
the relative efficacy of TAF and TDF may differ
12
between rectal and vaginal tissues and between MSM
13
and cisgender women.
14
Measurements of tissue drug levels in this
15
context is not particularly relevant. As unlike
16
with PBMCs, the levels are not being measured
17
primarily in cells that are infected but by rather
18
in bulk populations of extremely heterogeneous
19
cells from biopsies. And while there's evidence
20
that the safety profile of TAF may be superior,
21
particularly for long-term use, this has to be
22
balanced against the possibility of inferior
A Matter of Record (301) 890-4188
313
1
efficacy.
2
In this situation, a relative lack of
3
efficacy may translate into a currently incurable
4
infection. Thus, one has a potential choice
5
between long-term morbidity versus immediate risk.
6
Nevertheless, I do not believe that we can state
7
with scientific validity that TAF/FTC is as
8
effective as TDF/FTC in cisgender women for PrEP.
9
Extrapolation from one group to another is
10
defensible if there is no scientific reason to
11
believe that there could be pharmacokinetic or
12
pharmacodynamic differences between the two groups.
13
That is not the case here, and therefore the
14
basis for extrapolation from TDF to TAF in
15
cisgender women is not obvious. The absence of
16
actual clinical data in this group combined with
17
the potential difference in the site of exposure,
18
and other potential gender-based biological
19
co-factors, do not allow me to recommend labeling
20
this drug as effective in cisgender women.
21
I do not believe the drug should be approved
22
or labeled without adequate evidence merely because
A Matter of Record (301) 890-4188
314
1
doing the necessary clinical studies would be
2
challenging. The alternative is to potentially
3
expose segments of the population who are
4
underrepresented in studies to ineffective therapy.
5
DR. BADEN: Dr. Siberry?
6
DR. SIBERRY: George Siberry. I voted no.
7
I think that there's good evidence of a biologic
8
correlate of adherence. I remain unconvinced that
9
we have a good biologic correlate for protection.
10
For the reasons Dr. Swaminathan said, I think it is
11
inappropriate to extrapolate to women. However, I
12
feel like we have failed women by letting this
13
application come in without data from women to
14
begin with, and I fear we're failing them again by
15
having approval for use in men and not women.
16
That's why I asked for that clarifying
17
question about the question because I think these
18
are two different things, and I would be supportive
19
of an indication that includes women with a strong
20
postmarketing requirement for clinical evaluation
21
in women. Thank you.
22
DR. BADEN: Dr. Gripshover?
A Matter of Record (301) 890-4188
315
1
DR. GRIPSHOVER: Hi. Barb Gripshover. I
2
also voted no. I do not believe the data support
3
TAF/FTC efficacy as PrEP for women, as it's not
4
been studied in that population, and I don't think
5
it's clear that just the level of tenofovir
6
diphosphate in PBMCs is the sole determinant of
7
efficacy in women at risk for cervical vaginal
8
infections or the reasons the gentlemen have just
9
said.
10
I believe there is a large unmet need of
11
women at risk of acquiring HIV worldwide that
12
should be able to engaged in studies to answer this
13
question; maybe using matched geographic
14
demographic incidence rates as a control or
15
incidence in screening that has been suggested.
16
While I do not like the idea of approving a
17
drug for a single population, as it does look
18
effective in MSM, I also think we are obligated to
19
base our recommendations for use of a drug based on
20
data. Women in underserved populations deserve our
21
best efforts to make sure drugs are effective and
22
safe for them as well before we start recommending
A Matter of Record (301) 890-4188
316
1
it in lieu of one that has demonstrated safety and
2
efficacy.
3
So if the drug is approved for MSM, then I
4
would absolutely require a strict efficacy setting
5
in women as part of the agreement.
6
DR. BADEN: Wait, Dr. Green. Dr. Lupole?
7
(No response.)
8
DR. BADEN: You're on mute if you are
9
talking.
10
(No response.)
11
DR. BADEN: Okay. We may have lost the
12
connection. We can try to bring Dr. Lupole on.
13
MS. LUPOLE: Can you hear me now?
14
DR. BADEN: We can hear you now.
15
MS. LUPOLE: Okay. Sorry about all this.
16
I voted no as well. The lack of data, the
17
lack of study participants, conflicting data, it's
18
my recommendation that the trial for this drug in
19
cisgender women and juveniles be redesigned to
20
examine the impact because it's clear it's not been
21
presented to me that it would be safe and
22
effective. Thank you.
A Matter of Record (301) 890-4188
317
1
DR. BADEN: Thank you. Dr. Green?
2
DR. GREEN: Michael Green. I voted yes, but
3
I almost abstained and I almost voted no.
4
(Laughter.)
5
DR. GREEN: With regards to extension of
6
approval to cisgender women, the key concern of the
7
FDA appeared to have been relating to the tissue
8
level in vagina and cervix, and that those
9
associated with TAF were lower than TDF.
10
Therefore, the absence of a trial in cisgender
11
females directly to confirm efficacy, they're
12
asking us if we can extrapolate to extend approval
13
based on the DISCOVER population.
14
However, the data that was presented suggest
15
that TDF also has low levels, both at 4 four hours,
16
and at 24 hours, and 48 hours, and yet F/TDF
17
carries an approval in men, women, and adolescents
18
for pre-exposure prophylaxis against HIV and is
19
considered effective in these populations if those
20
taking it are compliant.
21
Accordingly, it's not clear that low tissue
22
levels had any impact on the effectiveness of
A Matter of Record (301) 890-4188
318
1
Truvada, and it seems unlikely, at least to me,
2
that it would for Descovy. Clearly, there is not a
3
concern that intracellular levels in PBMCs would be
4
different between men and women. We've also heard
5
the agency state that they feel challenged by
6
developing design for noninferiority studies, and
7
that there's no reason to expect a positive outcome
8
in a superiority trial, and that a comparison to
9
placebo would be unethical.
10
Given these issues, I felt it was
11
appropriate to include cisgender women in the
12
indication, especially given the equity issues that
13
have been discussed during this committee hearing.
14
Having said that, it would be important to mandate
15
postmarketing studies and this indication be
16
undertaken by the sponsor. And if these subsequent
17
studies did not bear out efficacy in cisgender
18
women, that the label be modified to reflect this
19
if not having the indication removed. Thank you.
20
DR. BADEN: Dr. Weina?
21
DR. WEINA: Peter Weina. I voted yes, but
22
the answer is really maybe. The reality is that we
A Matter of Record (301) 890-4188
319
1
really don't have a clue which is the appropriate
2
surrogate marker to use. Is it the tissue level?
3
Is it potentially PBMC levels? Is it adherence
4
that's the key? Or is it more likely something we
5
haven't even considered yet because we haven't
6
bothered to count it, and some revelation years
7
from now is finally going to give us that insight?
8
Right now, it seems like the surrogate
9
marker selected depends upon which opinion you'd
10
like to have supported, and the science behind it
11
is whichever you select, and that seems very
12
whimsical. So I reach back to the FDA's mission
13
statement, and the mission statement is to promote
14
and protect the public health by helping safe and
15
effective products reach the market in a timely
16
manner and monitor the products for continued
17
safety after they are in use.
18
This product is already out there for
19
treatment. It's already being demanded by patients
20
who are subjected to social media pressures, and
21
this is only going to accelerate. I have
22
absolutely no doubt that this is already being used
A Matter of Record (301) 890-4188
320
1
in cisgender women somewhere here in the United
2
States, and it's not being followed.
3
We should follow the FDA's mission statement
4
to get this to the market for the broadest
5
population possible and reasonable, and then
6
monitor the product for continued safety. Here of
7
course, I'm referring also to the efficacy because
8
if it doesn't work, then it's putting the users at
9
risk.
10
If approved for MSM and transgender women,
11
it's definitely going to be used either off label
12
or on label in adolescents and in cisgender women
13
just because of the perception of better safety.
14
We may as well carefully guide the postmarket
15
surveillance of this product and how well it works.
16
Clearly, we need carefully prescribed and intensive
17
postmarketing required trials.
18
DR. BADEN: Dr. Baden. I voted no. The
19
question was do we have substantial evidence of
20
safety and efficacy? There are no efficacy data
21
presented, and the historical efficacy data are too
22
strong to allow a placebo trial, but too weak to
A Matter of Record (301) 890-4188
321
1
allow a noninferiority margin. So one is choosing
2
which pieces of data to use to say that we cannot
3
study this population.
4
I share the open public hearing speakers, as
5
well as Dr. Siberry's comments we have failed
6
women. To be at this point and not have the data
7
to guide decision making is a shame on all of us.
8
I feel like Arrowsmith. "We are in a desperate
9
situation, therefore let's do something because we
10
can do something."
11
There are side effects to our interventions.
12
Our interventions are not benefit with no risk, and
13
the presumption that we can benefit and not have
14
risk is also shame on us. We need to generate some
15
data to guide the risk-benefit ratio, and the road
16
traveled for prevention in women is uneven with
17
high-quality large studies done. So for us to
18
presume that the good data are the ones we should
19
hang our hat on is presumptuous.
20
I think that given the mixed historical
21
data, the absence of data with this particular
22
agent, I cannot support an indication which has
A Matter of Record (301) 890-4188
322
1
efficacy. On the other hand, there should be a
2
mandated study. Whether it's mandated as part of
3
an approval or mandated in order to get approval,
4
both can be done, but it should be mandated.
5
I think once there's an approval, it's
6
impossible to undo even if there's no benefit
7
shown. If there's no approval, then the pressure
8
is to do the study, but then there are women at
9
risk who don't have opportunity to access this
10
medication. Hence, we have failed this population.
11
But I voted no because there were no data in the
12
population in question.
13
Dr. Ofotokun?
14
DR. OFOTOKUN: Igho Ofotokun. I voted yes.
15
Taking a look at the data as a whole, the Descovy
16
data and the historical data from Truvada, based on
17
data in HIV-infected individuals who are treated
18
with Descovy, I am convinced that the product is
19
just as safe in men and in women, and the big
20
question is that of the efficacy in ciswomen.
21
I tend to have some confidence in the
22
pharmacokinetic data and the correlate of Truvada
A Matter of Record (301) 890-4188
323
1
efficacy. The tenofovir diphosphate correlates
2
with protection, I seem to believe that that in
3
itself provides strong compelling data that TAF
4
would be just as efficacious in ciswomen.
5
I agree that it's a terrible failure that
6
the agency as well as the sponsor would come to
7
this committee with lack of data for women in this
8
hearing. I strongly believe, like others have
9
expressed, that there should be a mandated study to
10
look at women, ciswomen, either as part of the
11
approval process or before the approval of this
12
agent.
13
I also believe that approving Descovy for
14
PrEP in men who have sex with men alone would
15
create a two-tier system. It will just accentuate
16
this equity, the equity issue that already exist;
17
that either you're going to approve it for
18
indication for prep in men and women, or you're not
19
going to move forward with it.
20
I think creating a two-tier prevention
21
treatment will not be helpful, and we should remind
22
ourselves there are more women living with HIV in
A Matter of Record (301) 890-4188
324
1
the world than there are men, and that the risk of
2
new infection is significantly higher among women
3
if we look at this globally.
4
So I will stop there, and thank you.
5
DR. BADEN: Thank you. Dr. Burgess?
6
DR. BURGESS: Tim Burgess. I voted yes, but
7
as some others have said, I very nearly voted no
8
and very nearly abstained. I share concerns about
9
what we think we understand about the putative
10
mechanism of protection depending on route of
11
exposure.
12
My overarching concern was about the public
13
health impact of an indication in one population
14
and not in another population. Coupled with the
15
fairly compelling articulation of levels in PBMCs
16
as the primary, if not total component of the
17
likely mechanism of protection, led me to vote yes.
18
I, like others, articulate a strong recommendation
19
for, compelled postmarketing surveillance, focusing
20
on effectiveness in women.
21
DR. BADEN: Dr. Le?
22
DR. LE: My vote for approval in cisgender
A Matter of Record (301) 890-4188
325
1
woman was largely based on three factors: one,
2
data pertaining to vaginal tissue and PBMC as
3
presented earlier by Dr. Read; two, some safety
4
data but from other studies; and three, making this
5
drug combination available as an option for women,
6
not just for men, despite the lack of efficacy
7
data.
8
However, my vote for yes is contingent upon
9
full commitment from the applicant to incorporate a
10
robust package labeling, stating that efficacy and
11
effectiveness have not been established in
12
cisgender women with the use of this product, and
13
that vaginal tissue penetration was low, and that
14
the approval was based on extrapolation of existing
15
data in other populations.
16
Also, the applicant should commit to conduct
17
robust postmarketing studies to allow for us to
18
better understand efficacy and more effectiveness,
19
as well as incorporating safety monitoring for
20
weight gain, renal function, and on fasting plasma
21
lipid levels.
22
DR. BADEN: Thank you. Dr. Walker?
A Matter of Record (301) 890-4188
326
1
DR. WALKER: Dr. Roblena Walker. It was a
2
strong no for me, no wavering on the fence. I'm
3
almost highly appalled. There's about 8 women on
4
the committee, and that the agency and the
5
applicant would present insufficient data to
6
support the prevention of Descovy amongst
7
cisgendered women, or heterosexual women, or just
8
women in general, I was highly appalled that more
9
dedication and passion wasn't put into the study.
10
DR. BADEN: Dr. Dodd?
11
DR. DODD: So I voted no, and I was not on
12
the fence on this, unlike the last one. My concern
13
is about confusion or a lack of trust that might be
14
generated by an approval that wouldn't be supported
15
by strong science. We can't approve something just
16
because there's a need.
17
I also want to commend the agency for their
18
good discussion about surrogacy. I think this is
19
often a confusion in reviews of studies. There are
20
lots of reasons why a good correlate of protection
21
may fail as a surrogate endpoint for the clinical
22
benefit endpoint. In this case, the clinical
A Matter of Record (301) 890-4188
327
1
benefit endpoint is protection.
2
A correlate does not a surrogate make, and
3
we've seen data to support PBMCs as a good marker
4
of protection and women -- or we've not seen the
5
data; excuse me. And I thought the agency did a
6
good job of providing some reasonable arguments
7
about why PBMCs may not be a good marker of a
8
clinical benefit endpoint.
9
I think there probably should be both data
10
related to the biological mechanism supporting
11
additional surrogacy studies -- this looks like
12
more studies on tissue concentrations -- and
13
additionally, studies in ciswomen with an actual
14
clinical benefit endpoint of protection.
15
I'm not convinced that there's not a study
16
design out there that could be considered that
17
would support this. I don't know that it would
18
have to be something as large as a 20,000
19
participant study, but I think it's time to put
20
some creative heads together and think of some
21
feasible designs.
22
DR. BADEN: Dr. Giordano?
A Matter of Record (301) 890-4188
328
1
DR. GIORDANO: Tom Giordano. I voted no.
2
It pains me to say that. I really wanted to vote
3
yes because I believe there is the potential for
4
creating two systems, and one drug for the rich,
5
one for the poor, one for men, and one for women, I
6
think that's a horrible precedent.
7
Nonetheless, the FDA's approval, to me,
8
means we know this drug is safe and effective. I'm
9
convinced we know this drug is safe in women, no
10
doubt about that, but is it effective? That
11
remains a hypothesis. And given that there's
12
different biology involved between men and women
13
and the acquisition of HIV in men and women, I
14
think you need efficacy data, and it just boils
15
down to that for me.
16
I think that we're in this position is
17
absolutely horrible, but that's the position we're
18
in. So I don't envy the agency's ultimate
19
decision, but, to me, there is no way you can say
20
this drug has efficacy in cisgendered women. And
21
who's to blame for that? That's not my decision.
22
DR. BADEN: Dr. Daskalakis?
A Matter of Record (301) 890-4188
329
1
DR. DASKALAKIS: Demetre Daskalakis. I
2
voted yes. I think that there are a couple of
3
reasons. First, we have limited success with
4
topical agents that we know of to prevent HIV, and
5
I think we've seen data that intracellular levels
6
of drug seem to be protective, so that in
7
combination with what I thought was a pretty
8
convincing explanation for the role of a PBMC
9
level, intracellular level, and prevention made me
10
feel that I had enough evidence to recommend that
11
you consider approval for this drug for cisgender
12
women.
13
Now, I would put the caveat that labeling
14
would be very critical if it does come out like
15
this, so I think it would need to be an alternative
16
agent for women in certain clinical scenarios. And
17
I also think that it would be important to state
18
that there has not been an efficacy study done.
19
Now, from the safety perspective, I agree with what
20
everyone else has said, that I think safety has
21
been demonstrated by other studies, and that's not
22
really much of a debate.
A Matter of Record (301) 890-4188
330
1
I also want to say that I don't think the
2
approval of this drug would increase PrEP uptake
3
among women. So bottom line is that's not the
4
problem, at least in the U.S. The size of the pill
5
and the marginal improvement in bone and kidney
6
outcomes, not the problem.
7
The problem is that patients and providers
8
are unable to do appropriate assessment of who
9
needs PrEP, and I'm just concerned that creating
10
the tiered system that we may be creating if we
11
don't approve the drug for women will create even
12
more confusion with providers and poor advice to
13
their female patients who are considering PrEP, so
14
that makes me very concerned.
15
I think a mandatory study, no matter what,
16
whether it is after approval or preapproval,
17
requiring that is critical, and that needs to help
18
answer this question about intracellular level
19
versus mucosal level. So really good science that
20
looks at the role of mucosal levels of tenofovir in
21
women will be critical.
22
I also recommend thinking about coupling the
A Matter of Record (301) 890-4188
331
1
transgender female study with a women's study since
2
they are women, and that probably is a better way
3
to actually convince folks to enter the study. I
4
bet you one reason you can't recruit a transgender
5
study is because it says MSM, and that's not going
6
to work.
7
Another thing I just want to bring up
8
briefly is the precedent for extrapolating data.
9
We have U.S. preventative health services
10
recommendations that PrEP is an A recommendation,
11
and there is a line in there that says that
12
tenofovir could potentially be used as monotherapy
13
to prevent PrEP in women and heterosexual males and
14
females, and injection drug users.
15
I do not see us having a conversation about
16
using a generic, cheaper agent and extrapolating
17
that data to men who have sex with men, so we could
18
actually pour PrEP onto the entire country and be
19
less concerned about cost. So as we're having this
20
conversation about an expensive new drug, I would
21
encourage the agency to consider looking back at
22
the Bangkok PrEP study, at TDF2, at Partners PrEP,
A Matter of Record (301) 890-4188
332
1
and ask the question, should we be asking the same
2
thing about a drug that could cost as less as $5 a
3
month? Thank you.
4
DR. BADEN: Thank you. Dr. Read?
5
DR. READ: Sarah Read. I voted yes, but
6
with a lot of the same hesitations that have been
7
expressed by the other members who voted yes. Just
8
to be clear, I also agree that it's extremely
9
disappointing to be in a situation in which there
10
are no clinical efficacy data in cisgender women, a
11
population clearly in need of more effective
12
prevention choices and in whom much remains to be
13
learned regarding acceptability and preferences for
14
prevention choices. However, I felt in this case
15
that it was reasonable to extrapolate data from the
16
DISCOVER trials as well as previous prevention
17
trials with Truvada.
18
In terms of safety, although cisgender women
19
were not included in the DISCOVER trial, I think
20
it's reasonable to extrapolate safety from the
21
study participants, as well as the large experience
22
in treatment of women with HIV. And based on
A Matter of Record (301) 890-4188
333
1
treatment experience, I think that it's unlikely
2
that the safety profile will differ in cisgender
3
women relative to men.
4
In the absence of clinical efficacy data in
5
cisgender women and the question of the relevance
6
of PK in different compartments being not entirely
7
clear, extrapolation of the efficacy of cisgender
8
women is certainly not straightforward.
9
Although the collective data regarding PK
10
levels and correlation of clinical efficacy of oral
11
PrEP contain mixed results, I think it's reasonable
12
to extrapolate the clinical efficacy seen with
13
Truvada and Partners PrEP in cisgender women on the
14
basis of the data provided by the applicant,
15
indicating higher levels of TDF diphosphate in
16
PBMCs with F/TAF compared to Truvada.
17
PK data provided by the applicant on
18
cervical vaginal tissue levels, however, is less
19
clear given the number of samples that are
20
unevaluable. However, it's also unclear what
21
levels are required in this tissue. I therefore
22
think these data should largely be disregarded.
A Matter of Record (301) 890-4188
334
1
I think it would be problematic to approve
2
an indication in men who have sex with men alone
3
without including women. Such a limited indication
4
and subsequent delay in access for women for many
5
years would be untenable and an unfair situation.
6
Consideration, therefore, I think should be given
7
either to the approval with a broader indication to
8
include women or no approval at all until evidence
9
of adequate efficacy can be achieved in that
10
population.
11
If an indication to include cisgender women
12
is approved, like others, I recommend strongly that
13
the applicant be required to perform trials to
14
collect both safety and effectiveness data in this
15
population. Not only is the effectiveness
16
important, but also the safety profile in this
17
population needs to be further supported.
18
I think it's important that the company has
19
attested and pledged that they will perform these
20
trials, and I think it's up to the agency to
21
require them to do so.
22
DR. BADEN: Thank you. Dr. Smith?
A Matter of Record (301) 890-4188
335
1
DR. SMITH: I feel like we're moving
2
backwards from the 2012 meeting that approved
3
Truvada, in which there was a lot of discussion and
4
concern that we had data on African women and not
5
on African American women. And now we don't have
6
data on women at all. The decision has been made
7
that we'll do the trial in MSM, and then we'll
8
figure out what it means for women rather than
9
studying women themselves.
10
I find that bad science, and that's why I
11
voted no, but I also find it disrespectful and an
12
issue of sort of research equity. Women deserve
13
the same quality of data about the safety and
14
efficacy of the drugs that they're exposed to that
15
men get, and that's not the situation we find
16
ourselves in at the moment.
17
I also think that because we have Truvada
18
approved for women, we're not denying women access
19
to PrEP, and it's important to remember that. What
20
we are doing is saying that a second drug that is
21
similar in risk and benefit is available to one
22
population but not another, yet, based on the data
A Matter of Record (301) 890-4188
336
1
that we have. I think that's preferable to
2
approving it, doing an efficacy study and somebody
3
suggested maybe taking it back or modifying it if
4
it doesn't work out as well.
5
That's a recipe for disaster among the
6
African American community if we get ourselves into
7
a situation where we're approving something and
8
then saying, oh well, no, actually we weren't
9
right; that didn't work, so I wouldn't even think
10
about doing that.
11
I think the other thing is that even though
12
we think about the fact that it may be hard to
13
explain why this is for this group and not for that
14
group, if the proper studies are done in the short
15
term over the next three or four years to get the
16
kinds of data that is missing, then we'll be in a
17
position to say whatever is appropriate about
18
women.
19
I think we are going to increasingly in the
20
PrEP field have this situation of some things are
21
for some people and other things are for other
22
people. Whether that's the dapivirine ring, if
A Matter of Record (301) 890-4188
337
1
that becomes approved, that's surely not going to
2
be for all populations. I know we're nervous about
3
what that means when we suddenly have to start
4
making decisions, but I think this is not the
5
occasion in which that should overrule the absence
6
of data on efficacy for women as the basis for our
7
decision.
8
DR. BADEN: Dr. Goetz?
9
DR. GOETZ: Matthew Goetz. I did vote yes,
10
and I think I'm like the other 8 people who voted
11
yes. I do not hear a strong ringing endorsement
12
from anyone of strong data. I read the statement,
13
"allow for expansion" as a liberal statement,
14
"allow for expansion."
15
I thought critically about what we know
16
about surrogate markers, correlates of protection.
17
I think "correlate" is the right word in many
18
regards. The fact of the matter is that we will
19
need a phase 4 mandated clinical trial to
20
substantiate that this is I think an alternative,
21
and in any guidelines, documents, that are produced
22
by other societies, the strengths and weaknesses of
A Matter of Record (301) 890-4188
338
1
this, the conditional nature, and this is an
2
alternative needs to be very clear.
3
I felt very strongly that I'm not sure that
4
tissue markers are the surrogate either. As has
5
been pointed out by many individuals, when we
6
biopsy, first of all, we get very limited samples.
7
It's not robust. Secondly, the cells that we
8
sample are not likely the relevant cells. So we
9
either need robust data showing across levels of
10
different adherence, and we want everyone to be
11
adherent of course.
12
Inevitably, some people are going to be less
13
adherent, and we need to be able to correlate if
14
we're going to substantiate in any way. PBMCs show
15
that the correlate between PBMC and protection is
16
similar across all the relevant risk groups, and I
17
think that will go a long ways to demonstrating
18
what we need to show here.
19
Perhaps finally -- I can go on for a lot
20
long longer -- I think adherence is a crucial
21
measure. What we have in this drug and the study
22
we have in DISCOVER is a population that was
A Matter of Record (301) 890-4188
339
1
extraordinarily adherent. That's wonderful, but we
2
need to be clear that we want to really emphasize
3
adherence throughout. TAF may have a longer
4
half-life than plasma in cells, but that is not to
5
be taken as any opportunity to be less adherent;
6
phase 4 studies absolutely mandated.
7
DR. BADEN: Thank you. There you have it, 8
8
to 10 vote. The three key principles as I hear it,
9
because I will summarize the yes and the nos
10
together, the correlate is unclear and perceived
11
differently. The optics of approving for
12
population A but not population B has many
13
deleterious effects if done or not done. Everyone
14
agrees there needs to be actual data.
15
So then the challenge -- and I'll be
16
presumptuous, but I'll speak for the committee, and
17
to the agency, and to the applicant -- can you
18
please do the study as quickly as possible? And
19
it'd be designed -- I don't accept that it's too
20
hard, too big, too difficult.
21
There should be a way to do some type of
22
study systematically in a reasonable amount of time
A Matter of Record (301) 890-4188
340
1
if there's collective will to generate data
2
expeditiously, and that will be the best way to
3
minimize the optics of some of the concerns raised.
4
Many of us believe that this should work and will
5
work, but we cannot have belief guide policy or
6
regulatory pathway.
7
So that is the voting segment. We have run
8
15 minutes over. I would like to take 5 minutes to
9
discuss the last question, and that will be an open
10
discussion unless the agency advises me otherwise.
11
The open discussion is please discuss whether the
12
data from the DISCOVER trial are relevant to
13
at-risk men who practice insertive vaginal sex with
14
cisgender women.
15
I'll open the discussion and look for
16
disagreement or augmentation. There are many
17
aspects of insertive vaginal sex that have elements
18
that are analogous to MSM in the sense of the
19
biology of how the drug works and the nature of the
20
exposure. We weren't able to extract out MSM with
21
only insertive, but presumably there will be some
22
of that in the population -- it was a large
A Matter of Record (301) 890-4188
341
1
population -- and the biology in the prior
2
experience is such that I don't think it's
3
unreasonable to think that it's likely to work in
4
that population.
5
But I would like other comments from the
6
committee as to if others agree that it should
7
likely work in that population or if there are
8
concerns as to why it may not.
9
DR. SWAMINATHAN: Just to make clear, we're
10
talking about HIV, uninfected men having sex with a
11
discordant partner, female partner.
12
DR. BADEN: Yes, and circumcision has not
13
been addressed, but presumably the other preventive
14
strategies will be maximally encouraged.
15
Dr. Siberry?
16
DR. SIBERRY: I agree with your general view
17
that this can be extrapolated, but I do think that
18
the data should be looked at more carefully from
19
the DISCOVER trial. They enrolled people who had
20
condomless anal sex, not just condomless receptive
21
anal sex. So I think if they had collected
22
information about practices, you may be able to
A Matter of Record (301) 890-4188
342
1
segregate those practices, predominantly insertive
2
sex from those who didn't, look at it stratified by
3
condoms, and see if there was a difference in the
4
protective -- the levels of infection in the two
5
arms.
6
Granted, the overall infection risks are
7
probably lower in both arms of that group if you
8
limit it to those, but I think we should ask for
9
additional scrutiny of data.
10
DR. BADEN: And perhaps new data to actually
11
look at that population.
12
DR. SIBERRY: Yes.
13
DR. BADEN: Dr. Gripshover?
14
DR. GRIPSHOVER: I'm sorry. I didn't
15
realize there's still a vote, but I do think the
16
fact that 44 percent were umcircumcised means that
17
at least there was a group that had not yet even
18
used that other protective mechanism, so that's I
19
think helpful, too.
20
DR. BADEN: Dr. Weina?
21
DR. WEINA: Given the way the trial was
22
enrolled, the data's just not there. So I'm not
A Matter of Record (301) 890-4188
343
1
sure that you can actually extrapolate anything
2
from that trial.
3
DR. BADEN: So what is your view, then, on
4
the applicability to men who have vaginal sex?
5
DR. WEINA: I think it's just like -- well,
6
in my patient population, I have individuals that I
7
have in my patient population that are at high
8
risk, and in heterosexual relationships, and come
9
to me and are actually on Truvada for preventive
10
reasons, but the data is not really there to
11
support it. It just makes sense based upon the
12
data that is out there.
13
So there's an extrapolation because the
14
individual is at very high risk, and everything
15
that we can do to help prevent it is going to be
16
something that's worthwhile as long as they're
17
properly informed as to the risks associated with
18
taking the medication as well.
19
DR. BADEN: Well, let me push you a little
20
bit on that --
21
DR. WEINA: Sure.
22
DR. BADEN: -- in that if you have MSM who
A Matter of Record (301) 890-4188
344
1
have insertive and receptive, presumably the
2
insertive risk would be similar to the insertive
3
risk in non-anally receptive.
4
DR. WEINA: Agree.
5
DR. BADEN: So therefore, if data suggest
6
that it works in that population, even those not
7
specifically pulled out, that would be suggestive
8
that it is likely to work in that population.
9
DR. WEINA: so again, just like I was
10
talking about before, suggestive and correlates and
11
surrogate markers and everything else are --
12
DR. BADEN: Although, I think it's a little
13
different. These are human data --
14
DR. WEINA: True.
15
DR. BADEN: -- in men who are in study on
16
drug and not getting infected.
17
DR. WEINA: True.
18
DR. BADEN: This is not extrapolating from
19
assays that we're not completely sure what they
20
tell us with a correlate, that we're not sure what
21
it tells us in 5 people.
22
DR. WEINA: True. So given the potential
A Matter of Record (301) 890-4188
345
1
outcome of not putting the individual on Truvada,
2
when they come to me with exceedingly high risky
3
behavior with multiple unknown partners on a
4
regular basis, I inform them of the risks
5
associated with it and the potential benefits --
6
DR. BADEN: And the limitations of the data.
7
DR. WEINA: -- and allow them to make the
8
decision.
9
Dr. Daskalakis?
10
DR. DASKALAKIS: Just fusing this issue a
11
bit with the issue about the need for a study in
12
women, it seems as if there's a need for another
13
serodiscordant heterosexual study like a Partners
14
PrEP but that uses this drug.
15
DR. BADEN: Although part of a challenge
16
there is treatment as prevention.
17
DR. DASKALAKIS: Right, but still --
18
DR. BADEN: But still --
19
DR. DASKALAKIS: There's an environment
20
where it's still feasible with lower edge
21
retroviral uptake, so it wouldn't necessarily
22
launch that study --
A Matter of Record (301) 890-4188
346
1
DR. BADEN: But generating the data makes
2
sense.
3
DR. DASKALAKIS: But I think that there's
4
other parts of the world where you can have
5
serodiscordant couples and follow that, realizing
6
that there will be -- that the sample size will
7
probably have to be bigger and the effect may be
8
smaller.
9
DR. WEINA: But the point is that there are
10
parts of the world in which this could be done just
11
like we do malaria studies in other parts of the
12
world because we haven't got a whole lot of malaria
13
here in the United States to get new drugs
14
approved.
15
DR. BADEN: Dr. Swaminathan?
16
DR. SWAMINATHAN: I guess the difference to
17
me is that there's a little bit more that you can
18
extrapolate from. You have data in discordant
19
couples where the woman is positive that tenofovir
20
works, and MSM couples it works, and we have
21
evidence that TAF works in MSM couples.
22
So now you're just sort of bringing in the
A Matter of Record (301) 890-4188
347
1
fourth variable or a mix of those two variables to
2
say, well, TDF works for this situation and TAF
3
works for this situation, and TDF also works for
4
this situation, which is -- you can sort of
5
extrapolate a little bit more from that, that you
6
would expect the person who was protected by TDF in
7
the Partners study to be protected by TAF in the
8
future.
9
DR. BADEN: Dr. Goetz?
10
DR. GOETZ: I think another piece of
11
supporting evidence is the incidence of gonococcal
12
urethritis and other urethritis in the patient
13
population, which I think was 15 to 20 percent
14
thereabouts. So there was substantive exposure
15
to -- evidence of insertive practices.
16
Yet, if I recollect the data properly, all
17
the cases of infection were amongst those people
18
who clearly had receptive anal intercourse. The
19
presence of the urethritis I think is a strong
20
piece of evidence in favor of the fact that there
21
was risk in the patient population.
22
DR. BADEN: Dr. Smith?
A Matter of Record (301) 890-4188
348
1
(Dr. Smith gestures no.)
2
DR. BADEN: So I think that that touches on
3
a lot of the key issues around this question. Are
4
there any other issues the agency would like us to
5
address?
6
(No response.)
7
DR. BADEN: If not, then I would like to
8
thank the applicant for a tremendous amount of data
9
being presented and entertaining a lot of
10
discussion in a challenging area; the agency for
11
sharing your views in the challenge here; the panel
12
members for a robust, high energy day in covering a
13
lot of complex issues; and the public as well for
14
sharing your thoughts.
15
I'll see if the agency has any closing
16
remarks.
17
DR. BIRNKRANT: Well, I, too, on behalf of
18
our division and the agency, want to thank the
19
committee for their thoughtful discussion and
20
deliberations today. I also want to thank the
21
speakers who commented during the open public
22
hearing as well. I want to thank the company for
A Matter of Record (301) 890-4188
349
1
conducting the DISCOVER study and other pertinent
2
research and for committing earlier in the day to
3
conduct a trial, or trials, in women.
4
I also want to thank the trial participants
5
as well. Lastly, I'd like to thank our staff for
6
their dedication and diligence in conducting the
7
reviews and preparing for this committee. I want
8
to leave you with a couple thoughts before we end.
9
Our review of this application continues. We have
10
not made any final determinations as of today, and
11
your comments and the discussions will greatly
12
impact our final determination.
13
Lastly, I feel like we should dedicate our
14
collective efforts to ensuring the availability of
15
safe and effective medications for all populations
16
so that the next time we meet, we can definitively
17
state that the HIV incidence in the United States
18
has substantially declined in all populations, and
19
we are moving closer to defeating this epidemic.
20
Thank you very much.
21
Adjournment
22
DR. BADEN: Thank you, and I will now
A Matter of Record (301) 890-4188
350
1
adjourn the meeting. Safe travels.
2
(Whereupon, at 4:54 p.m., the meeting was
3
adjourned.)
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
A Matter of Record (301) 890-4188
https://www.convertapi.com