XIFIN Pay For Value White Paper

: Pdf Xifin Payforvaluewhitepaper XIFIN_PayForValueWhitePaper 08 2013 clpmag uploads wp-content

Open the PDF directly: View PDF .
Page Count: 28

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

WWW.XIFIN.COM

A Pay-For-Value, Data-Driven

Approach for the Coverage of

Innovative Genetic Tests

Co-Authors:

Paul Billings,

Chief Medical Ofcer, Life Technologies

Richard Ding,

Chief Executive Ofcer, bioTheranostics, Inc.

Lâle White,

Chief Executive Ofcer, XIFIN, Inc.

Rina Wolf,

Vice President of Commercialization Strategies,

Consulting & Industry Affairs, XIFIN, Inc.

David Lorber,

Director of Business Development, XIFIN, Inc.

WHITE PAPER

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

CONTENTS

Summary 1

Introduction 2

Centers for Medicare & Medicaid Services (CMS) Recognizes the Need for Change 2

The Rise of Coverage with Evidence Development (CED) 2

CED and Molecular Diagnostics 4

Molecular Diagnostics: Unique Considerations and Unique Opportunities 4

Critical Concepts for Coverage 7

Clinical Utility 7

Prognostic Rather Than Predictive Results 8

Intellectual Property: Business Constraints 9

Proposal: A Market-Driven, Self-Regulating Approach 10

A Self-Regulating System 10

Shared Risk 11

Data Exchange Supports an Outcomes-Driven Reimbursement Model 14

Take Advantage of the Shift Toward Pay for Value Instead of Fee for Service 18

Considerations for Reimbursement Rates 19

Conclusion 21

Acknowledgements 21

References 22

About the Authors 24

pg i

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 1

SUMMARY

The Coverage with Evidence Determination (CED) methodology is necessary

for treatments and appropriate for traditional diagnostics. However, when

applied to innovations in personalized medicine and molecular diagnostics,

CED (in its current form and with proposed modications) fails to capture the

benets of improved health care and decreased costs that are possible from

the new technology and the data generated from these services. Given the

speed of innovation in this important clinical area, CED methods may make new

tests obsolete before they are commercialized, hindering investment in their

development. Current registry systems have demonstrated promising concepts

for an alternative to CED, but have also uncovered additional challenges, as

would be expected by pioneering new ideas.

This paper proposes an alternative model for coverage determination that

builds upon programs like MolDx by Palmetto and capitalizes on the knowledge

gleaned from early registries, addresses the key underlying nancial motivations

that drive coverage determinations, and advocates for a shift to a Pay-For-Value

(PFV) model based on risk sharing between parties. This proposed model is

intended to move forward the discussion on a design for coverage determination

that encourages innovation in the agile environment of molecular diagnostics.

Changes are urgently needed to allow health care to meet the new demands

posed by increased medical personalization and for accountability of all parties

in the US healthcare system. A technology-enabled coverage determination

process presents a unique opportunity for a self-regulating system that inherently

optimizes patient outcomes, payor costs, and developer prots.

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 2

INTRODUCTION

Centers for Medicare & Medicaid Services (CMS) Recognizes the Need

for Change

One in four Americans receives benets through CMS.1 Thus CMS is in a unique

position to not only make coverage determinations and negotiate low pricing

for its beneciaries, but also to establish the coverage and pricing precedents

most other payors follow. The mission of CMS is to “ensure effective,

up-to-date health care coverage and to promote quality care for

beneciaries.”2 In 2000, CMS (then known as the Health Care

Financing Administration) needed to develop coverage policies for

procedures “where we believed that the enthusiasm of interested

parties was disproportionate to the persuasiveness of the then-current

evidence base.”3 Put more bluntly, diagnostic providers held out higher

hope for certain diagnostics and therapies and therefore recommended

their prescription even when there was no direct evidence of benet. In 2000,

CMS recognized several key shortcomings4 that are still at issue today:

1) For many emerging technologies, seemingly insufcient utilization

evidence had been collected to support a coverage policy; a clear

denition of “sufcient evidence” has yet to be seen in 2013.

2) In many cases CMS beneciaries could benet from early access to an

emerging technology, and by covering an emerging technology CMS

could itself expedite the collection of the necessary data to support a

coverage decision; this is still true in 2013.

3) Randomized Controlled Trials (RCTs), previously considered the gold

standard, are neither a good t, nor are feasible, for the rapid pace

of development of some medical innovations. Applied to molecular

diagnostics, it may increase the cost of development substantially. In

2013, retrospective data, not garnered from RCTs, is becoming more

acceptable as illustrated by a recent reference from Novitas Draft LCD

DL33138 (see Figure 1).5

The Rise of Coverage with Evidence Development (CED)

In 2006, the Coverage with Evidence Development (CED) program6 formalized

a control mechanism to allow CMS to provide cutting-edge health care coverage

CMS is one of the

largest purchasers

of health care in the

world, covering one

in four Americans

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

while still evaluating the clinical utility of new procedures. The CED process, with

its classically designed studies and data collection techniques, was created for

drugs and biologics and is not well suited to molecular diagnostics.

CED approaches t best in a medical system akin

to an assembly line stage-gate supply chain model:

Not only is this stage-gate process not optimal for

molecular diagnostics, but it negates (by providing

no incentive or collection mechanism for) the vast

amount of data the tests can produce and must

freely ow between the stages.

In early 2012, CMS solicited feedback about

updating the program and at about the same time

the President released the National Bioeconomy

Blueprint that specically stated, “Expanding the Coverage with Evidence

Development program to drive innovation: reimbursement for medical treatments

is a powerful driver of industry investment.”7 The draft guidance for CED8, the

review of which ended in January 2013, makes recommendations that are

based on the feedback from the existing CED program, and are in alignment

with the Blueprint. While these recommendations provide some minor tweaks to

the current CED program, they do not address the unique considerations and

opportunities of molecular diagnostics (outlined below). This results in molecular

diagnostics developers being saddled with untenable returns on investment

delaying potentially life-saving (and cost-saving) diagnostics.9

pg 3

Novitas Draft LCD

DL33138

Comparative cohort designs are critical for

assessing predictive ability, and randomization

is highly desirable, although nonconcurrent

randomized cohorts can speed the evaluation

process. The sufficiency of nonrandomized cohorts

must be very carefully scrutinized, on a case by

case basis. This is a reflection of the fact that

alternative approaches to more conventional

randomized prospective controlled trials—such as

prospective-retrospective study designs—may be

able to support predictive biomarker CVU, as long

as they are appropriately conducted. Per above,

there are currently NO standardized thresholds

and/or benchmarks for evaluating the CVU/medical

necessity of emerging biomarkers.

Figure 1: This dra LCD from Novitas indicates alternaves to

randomized controlled trials are viable, if appropriately conducted.5

01 Validation of

diagnostic

02 Clinical utility

03 Reimbursement

determination

04 If additional

indications, go

to stage 01

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 4

CED and Molecular Diagnostics

The CED process, in efforts to ensure correct utilization, can

unnecessarily stie innovation for smaller, venture-backed companies

that do not have the resources to survive the extended period of data-

gathering and potential non-coverage dictated by the CED process.

Given that a great deal of innovation in the molecular diagnostic

space is done by these venture-backed companies, the CED process

has the effect of dampening the entire molecular diagnostics industry.

CED—by design—can take years, and the diagnostics going through

the process may be saddled with an unrecoverable cost burden,

private insurer non-coverage, and be eclipsed by another technology

before reaching market (similar to why labs oppose the FDA process

in favor of the more agile Clinical Laboratory Improvement Act

(CLIA) process for Laboratory Developed Test (LDT) oversight). The

proposed CED process requires the services of a protocol “expert,”

such as the independent Center for Medical Technology Policy

(CMTP)11 further increasing the cost. With these burdens imposed by

CED protocols, the risk-adjusted ROI for developing new diagnostics

such as NexGen Sequencing may not be favorable to produce

desirable rapid development and deployment.

Finally, there is growing recognition of the value of longitudinal

studies, which can capture real world clinical practice, outside

of a sometimes “articial” study environment. Another system

(not CED) that embraces broader data collection and analysis

capabilities and delivers a revamped incentive system is needed.

It should be parsimonious and let research physicians explore how

new diagnostics can be most valuable and allow producers to be

innovative and agile.

Molecular Diagnostics: Unique Considerations and Unique

Opportunities

Genetic testing, molecular diagnostics, genomic testing, and

personalized medicine are fundamentally different from, and will

change, traditional diagnostics and medicine. The difference

Food and Drug

Administration Regulation

Could Add Additional

Burden

The role of the Food and Drug

Administration (FDA) should also be

considered. Currently through a trial

parallel review program, parts of the

clinical trial process can be run in

parallel with early validity studies.

While this may help accelerate the

FDA-approval path, it further codifies

a path that the majority of molecular

diagnostics do not currently follow.

Laboratory Developed Tests (LDTs),

subject to more industry appropriate,

agile regulations and oversight by

CLIA have, and for the foreseeable

future will (due to the structure of

the industry), provide a rich source

of new diagnostics. LDTs, regulated

under the Clinical Laboratory

Improvement Amendments (CLIA) of

1998, have historically been able to

deliver critical tests of public health

importance before FDA approval

is received. An LDT diagnostic for

HIV viral load testing was available

six years before an FDA approved

kit was available. Similarly, an LDT

diagnostic for KRAS to determine

which patients would respond to

cancer treatments was available

years before the FDA kit. A recent

ACLA petition stated that “FDA’s

regulation of LDTs as devices would

adversely affect patient care in the

US.”10

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 5

between traditional and molecular diagnostics is comparable to that between

devices and “smart” devices; a room “knows” when someone enters and turns

on the lights, a car knows when it is skidding and applies the anti-locking break

system. Very specic technology-provided information (motion detection, skid

recognition) alters outcomes in a manner completely impossible without the

added information. Similarly, very specic information provided by molecular

diagnostics for specic patients, can alter the course of medical treatment in a

manner completely impossible without the added information.

Molecular diagnostics presents unique nancial, timetable, and scientic

benets relative to other diagnostics, which are not leveraged by the CED

process:

1. Financial – molecular diagnostics target small markets with high

development costs

• Unit costs are higher – The development costs for molecular diagnostics

and treatments are commonly on par with, or higher than, non-molecular

tests and treatments, but generally are only given to a relatively small

number of patients exhibiting very specic indicators.

• Benetsaregreater – Molecular diagnostics can be more precise than

traditional diagnostics, allowing appropriate therapies to be delivered

sooner, resulting in better outcomes and less waste.

• Relative value is greater – As treatments and diagnostics evolve (Figure

2), their costs generally increase, as does the value of knowing when to

utilize a treatment. In other words, a $20 test to determine if $50 worth

Visual See a lesion/growth

See the bacteria

Stain for specific

structures

Micro-

scopic

Chemical

Marker

Evolution of Diagnostics

Genetic Identify underlying

genetic cause

Physical Remove lesion / growth

Tree bark, Penicillin

Broad-based chemical

agents

Natural

Synthetic

Evolution of Treatments

Genetic Drugs targeted at

specific mutations

Figure 2: Both diagnoscs and treatments connue to evolve to higher-cost, more complex protocols that also possess higher potenal value to

the paent.

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 6

of antibiotics should be prescribed is a benecial test; a $3,000 test to

determine if $100,000 worth of a cancer therapy should be prescribed is

a more benecial one. This results in the personalized test being more

valuable from a purely nancial perspective.

The CED process does not take into account the increased development costs

to the developer, the improved health outcomes for the patient, the potential

for waste reduction or cost control, or the decreased costs for payors. It

additionally assumes that costs will be offset in the long run when the diagnostic

is disseminated for usage by the general public. The entire point and value of

genetics is that they will never be broadly disseminated. Markets will be smaller

and even more personalized.

2. Timetable of molecular diagnostics – studies are slow, trials need not

be slow

• Conductingstudiesismoredifcult–Because molecular diagnostics

are narrowly targeted, the group of patients who have the correct set of

highly specic risk factors that indicate appropriate use of a diagnostic is

necessarily small. Identifying a patient population large enough to support

statistically signicant results can prove time-consuming and difcult, and

could delay access. The challenge of nding suitable patients increases

the value of data from each patient and speaks directly to the need for

increased data federation through electronic health records.

• Trial periods – While studies can be slow, trials, and the inception of

data gathering, can happen more quickly. It is benecial and preferred to

begin gathering data immediately rather than in a drawn out trial process.

Unlike treatment trials, diagnostic trials can, by appropriate weighting of the

results, present no health risk to patients. It is important to note that in the

absence of decision support tools to help physicians determine the weight

that should be applied to diagnostics results, the diagnostics may not be

used in an efcacious manner.12

The CED process makes no allowance for the rapidity with which molecular

diagnostics evolve or the dearth of test subjects who can contribute to the

knowledge pool (and the data these test subjects can provide).

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 7

3. Scientic–moleculardiagnosticsinvolveshugenumbersofvariables

Because of the virtually limitless combinations of genetic markers possible, it

is likely that during the course of evaluating a specic molecular diagnostic,

useful variations will be identied. These variations are far more numerous

than could be reasonably addressed by sequential independent studies.

The variations may:

• Increase the accuracy or specicity of a diagnostic

• Identify new applications in test subjects (e.g., identifying the presence of

HER2 in colon cancer patients who subsequently respond to Herceptin)

The CED process does not allow tests to evolve dynamically based on the

cornucopia of data that can be provided by slight biomarker variations, or

newly-discovered indications.

CRITICAL CONCEPTS FOR COVERAGE

Clinical Utility

The core element to determining appropriate coverage centers on clinical utility.

There is little dispute that for analytic or clinical validity, measurements should

be done accurately and be reproducible in varying

clinical environments.13 The disputes arise around

the concept of utility or clinical utility—what does it

mean? How is it measured? Who pays to measure

it? How is it used as both a yardstick and gate for

payment?

All parties have slightly varying denitions of utility;

utility can have different meanings when applied

to inherited conditions vs. reproductive decision-

making vs. public health, and there is uncertainty around when an appropriate

quantity of data has been accumulated to measure utility. In the context of CED,

the payors’ denition that a clinically useful result is one that alters the course of

treatment (e.g., identifying the right type of intervention or changes in monitoring

methods or frequency), is probably most relevant.

Measuring clinical utility, or determining when it needs to be measured

through the CED process, is not clearly dened. The historical code-stacking

A study done by McKinsey & Co for the Personalized

Medicine Coalition revealed that of $292B spent on

medications in 2008, approximately $145B went to

drugs that were ineffective for the patients who took

them. Further findings estimate the cost of adverse

drug events to range between $45 to $135B per

year. An estimated 25% of these costs could be

averted through the use of diagnostic tests for the

appropriate biomarkers.

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 8

reimbursement methodology (the use of which we are not advocating) largely

circumvented the need to dene clinical utility by allowing diagnostics, still lacking

complete clinical utility studies, to be reimbursed. In fact, under the code-stack

methodology, tests were frequently reimbursed before much, if any, clinical

utility data were generated—allowing information

to be collected after commercialization. Since

diagnostics are ordered by physicians, adoption

of new diagnostics is generally driven by expert

specialty physicians or key opinion leaders integral

to the research process and the generation of

data for more widespread publication. The end

of code-stacking at the beginning of 2013, and

the concomitant introduction of specic codes for

various diagnostics, resulted in limiting the scope

of use of some diagnostics (e.g., EGFR) and

requiring new tests to be coded as miscellaneous

requiring appeals. Without reimbursement for new

tests, physicians may be unnecessarily constrained

in using their best judgment for medical care.

Additionally, under the code-stacking model, slight

variations of tests could be used and reimbursed.

Code-stacking, while far from ideal, provided a mechanism by which patients

could benet from innovative diagnostics.

An increased reliance on cost-prohibitive, xed CED utilization studies, when

combined with the discontinuation of code-stacking provides impediments

to: investment in new molecular diagnostics (costs are higher and revenue is

delayed); the collection of data so new diagnostics can be approved; and most

importantly, providing the best care. Diagnostics that fulll unmet medical needs

are not properly incented or nanced.

Prognostic Rather Than Predictive Results

There is an important distinction between predictive and prognostic factors

in relation to the utilization of diagnostics.14 Prognostic factors speak to the

probable course and outcome of a disease, while predictive factors address

Clinical utility is in the eye of the

beholder

For patients, relief or cure is a shared goal with

the provider. But utility might be the end of wasteful

payments, palliative care provision, or estate

planning.

For providers, utility may be a simple change of

guidelines or atrophied standards of care but could

also mean changes in office or hospital workflows,

simpler and more targeted test ordering or less

iatrogenic illness.

For hospitals, utility is better Diagnosis related

group management and facility optimization as well

as potentials for marketing and differentiation.

For payors, utility is cost saving, improved

outcomes and more informed management of

groups.

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

the likely effectiveness of a therapy. The coverage decision for a diagnostic

should be based on the test’s ability to provide prognostic guidance that impacts

patient care, (a so called “actionable result”), not on the predicted (or realized)

effectiveness of the resultant therapy. The prognostic guidance is realized

as soon as the physician sees the results. The results could range from the

avoidance of an unnecessary therapeutic intervention to simply a change in

surveillance.15 This “instantaneous” value of diagnostics makes them ideal for a

PFV payment structure.

Due to the complexity of molecular diagnostic results, the determination of

“standard of care” and coverage should to some extent be based on the

decision support provided to the physician along with the result. For example,

if a physician decides on a course of action different than one that would have

been taken without the diagnostic, then the test demonstrates utility and the

diagnostic provider should be reimbursed. If the physician takes appropriate

action as a result of a diagnostic and its associate decision support, the physician

should be considered to have delivered the standard of care and should also be

reimbursed.

Intellectual Property: Business Constraints

Pharmaceuticals and other therapies generally enjoy a relatively long period

of post-approval patent protection, and consumption patterns that may last for

several years. The rate at which molecular diagnostics are evolving means that

the useful lifetime of diagnostic patents could be signicantly shorter than for

therapeutic patents. Further, molecular diagnostics are almost always deployed

as a single instance (i.e., one set of biomarkers). Even if a patent remains viable,

it is likely that an enhanced version of the diagnostic would become available

prior to the protocol endpoint. In other words, because molecular diagnostic

technology evolves so rapidly, patents for these diagnostics are much more

likely to be obsolete than are patents for medical devices or slower-moving

technologies.

pg 9

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

PROPOSAL: A MARKET-DRIVEN, SELF-REGULATING APPROACH

The alternative to CED proposed herein leverages advances in technology

to exchange information, and the vast amounts of data created by molecular

diagnostics, to allow for a more market-driven, self-regulating, PFV approach to

health care.

A Self-Regulating System

The driver for the interactions between the payor, diagnostic provider and the

physician/patient is reimbursement (in addition to the more altruistic desires to

improve outcomes). Reimbursement drives two opposite, but complementary

cycles (Figure 3).

Counter clockwise cycle:

• Patients and physicians want diagnostic

providers to develop tests that improve

treatment decisions.

• Payors want patients and physicians to

efciently use nascent diagnostics (i.e.,

correctly order and interpret results leading to

lower overall costs, less waste and improved

outcomes).

• Diagnostic providers want payors to provide

reimbursement for their innovations (a driver

for this cycle).

For the complementary clockwise cycle:

• Diagnostic providers want data supporting the

clinical utility of their new tests.

• Payors want decision support systems from

diagnostic providers to help physicians

correctly 1) order diagnostics, 2) interpret the

results, and 3) select the most appropriate

treatments.

• Physicians want payors to provide

reimbursement for efciently practicing

medicine (a driver for this cycle).

pg 10

Diagnostic Tests

Data

Physician/

Patient

Diagnostic

Provider

Eciency

Decision Support

Reimbursement

Reimbursement

Payor

Figure 3: Two cycles acvely linking payors, physicians, and providers to

eciently deliver health care.

The need for decision support systems to help

physicians improve use and interpretation of tests

was highlighted in a recent study of US medical

records by the Centre for Health Systems and Safety

Research, at the University of New South Wales. The

investigators found that doctors in the United States

fail to follow up as many as 62% of clinical pathology

laboratory tests and up to 35% of radiology reports.

This means that they are missing critical diagnoses.

In turn, this causes delays in treatments for many

conditions, including cancer.16

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

These complementary cycles provide a unique opportunity for a paradigm

shift in how molecular diagnostics are developed, approved, reimbursed, and

continually studied. Molecular diagnostics are particularly poorly suited to CED

and Fee-For-Service (FFS) reimbursement and are particularly well suited to an

alternative process and

pay-for-performance

reimbursement (Figure 4).

The key elements

that allow molecular

diagnostics to advance

from series of static

stages of development

to the more uid market-

driven cycle described in

Figure 4 are risk sharing

and data exchange.

The end result should

be a model that allows

the agile renement

of practice guidelines

that efciently optimize

healthcare.

Shared Risk

Risk sharing for medical

coverage is a relatively

new concept. After

Johnson & Johnson

received a coverage

rejection in 2006 for

its multiple myeloma agent, Velcade, from the UK’s National Institute of Health

& Clinical Excellence (NICE), it proposed the Velcade Response Scheme.17

This scheme essentially placed the payment burden back on the developer in

cases where the therapy was ineffective. About the same time, Genomic Health

pg 11

CED and FFS

Stages of Development

Design…likely to continu-

ally morph as more data

is collected

Approval…likely to be

required for many very

similar variations

Reimbursed or not, regard-

less of appropriateness of

the test

Reimbursed on a sliding

scale if treatment is altered.

Alternative Process and PFV

Reimbursement…values

are likely to be strongly

contested—high value

results

Fixed set of experiments

using a fixed protocol

Granted or not granted

based on the protocol

Let the biomarker set morph

and guide physicians in the

use of the evolving results

Remove the concept of

approval for molecular

diagnostics. The physician

makes informed decisions

about the appropriateness

of a diagnostic given the

indicators and available

data

Additional studies need to

be funded and undertaken

Incentivized to:

Have the diagnostic ordered

as much as possible

Not prescribe the diagnostic

because it might not alter

the course of treatment

Limit access to diagnostics;

difficulty in getting coverage

for their patients

The developer can promote

use of the diagnostic, by

assuming varying amounts

of the risk associated with

the cost

Incentivized to:

Have the diagnostic ordered

only where it might alter the

course of treatment

Have the diagnostic ordered

whenever the physician

believes it will alter

treatment

Prescribe any diagnostic

and provided with all

available data to make an

informed decision

Additional indications …

likely to be discovered as

more data is collected

Interested parties

Developers

Payors

Physicians

Figure 4: Comparing the status quo (CED and FFS) to an alternave process based on PFV from the perspecve

of development of a diagnosc and each of the interested pares.

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

launched OncotypeDX. This molecular diagnostic for the determination of

recurrent invasive breast cancer in estrogen receptor positive and lymph node

negative women, had signicant validation, but lacked clinical utility data. Lee

Newcomer at United Healthcare led an effort with Genomic Health to develop a

risk sharing model for OncotypeDX.

An important aspect of risk sharing as it applies to molecular diagnostics involves

managing risk through decision support tools (discussed below). Both the

ordering of diagnostics (e.g., when to use a molecular diagnostic, which specic

analytes should be tested) and the interpretation of results are complex. As part

of sharing risk, payors must collaborate with diagnostic providers to ensure they

can provide correct guidance for physicians so that the right tests are ordered at

the right time and so that the right treatment decisions result.

There are two important notes about risk sharing, a relatively labor-intensive

methodology. First, risk sharing is not being proposed as a primary, long-term

solution, but rather a model that allows clinical utility data to be accumulated

while providing appropriate remuneration in the dynamic molecular diagnostics

market. Second, to avoid the complexities of risk sharing but still allow for

some non-covered procedures, a pre-

authorization process is sometimes

used. While pre-authorization does

avoid some complexities it is highly

inefcient, time-consuming, and can

signicantly increase costs.18

As an example of how a risk sharing

model might work, assume a

hypothetical diagnostic, ABC-test

(Figure 5). The risk sharing model is based on the idea that the lab should

be paid (see the Considerations for Reimbursement section) if the diagnostic

produces actionable results. ABC-test has three indicators, each with equal

prognostic weight (row A). The lab, using its own data, expresses the risk it

is willing to take that the diagnostic will produce actionable results (row B). If

the lab is highly condent that the diagnostic will produce actionable results, it

pg 12

Indicators present

Lab’s confidence /

interest in results

Cost distribution if

treatment is not altered

Lab

Payor

Patient

1

5%

2

60%

3

90%

New

70%

100%

Negotiated Diagnostic Risk Sharing for ABC-test

The payor pays 100% if the diagnostic produces actionable results

—whatever treatment is provided is known to be the right one.

A

D

C

B

Figure 5: Hypothecal negoated diagnosc risk sharing. Provider expresses its condence

in its test, guiding coverage and cost distribuon.

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

could make this percentage high—effectively saying that if the diagnostic fails to

produce actionable results, the lab will absorb the cost. This condence level can

then be used by payors to make a coverage determination (row D).

For ABC-test:

• For one indication, the diagnostic provider has very low condence that the

diagnostic will alter treatment. It expresses this by indicating that if the test

does not alter the course of treatment, it will only absorb 5% of the cost—a

clear signal from the lab to the payor that the test should not be covered in

this case.

• The payor will likely follow the lab’s guidance and not provide coverage for

the test in this case.

• The physician and patient can then determine if they feel the test will be

benecial, knowing that the patient will likely be responsible for the bill.

• For three indications, the diagnostic provider is condent (90%) that the

diagnostic will alter the standard treatment. It expresses this by indicating

that if the test does not alter the course of treatment, it will absorb 90% of the

cost—a clear signal from the lab to the payor that the test should be covered.

• The payor will likely follow the lab’s guidance and provide coverage;

the lab has already placed a large bet on the test altering the course of

treatment.

• The physician and patient have high condence that the payor and

diagnostic provider believe the diagnostic is appropriate and one of them

will cover the cost.

• For a new indication, one for which the diagnostic provider thinks the test

might be useful but lacks clinical utility data; the lab can also place a bet. The

extent to which the lab wants the data and believes that its diagnostic can

alter treatment is expressed by the risk it is willing to take.

• The payor will likely follow the lab’s guidance. If the lab has high

condence in the diagnostic’s ability to alter treatment, the payor will

probably cover the diagnostic—either it alters treatment (good for the

payor; although it will pay for the diagnostic, it will get a better outcome) or

it does not (good for the payor, since the lab will absorb most of the cost).

• The physician and the patient have a condence level in the diagnostic

informed by the willingness of the lab and payor to cover the cost.

pg 13

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

• The lab must only be held accountable for these pricing mechanisms if

the physician has followed criteria that have been agreed to between the

payor and the lab for appropriate patient selection, timing of use of test

along care continuum, and treatment guidelines.

It is through this coverage determination that payors must incentivize providers

to design appropriate practice guidelines, and incentivize physicians to use new

and emerging technologies appropriately. In order for this balance to work fairly,

payors must:

• Compel test providers to develop appropriate protocols that clearly

demonstrate where and how these tests should be used in the continuum of

diagnosis and care;

• Work with providers to facilitate the educational process and to set policies

that encourage appropriate and effective use, and discourage less effective,

often older, modalities (e.g., limits on number of IHC stains prior to a

molecular alternative being mandated);

• Provide mechanisms to measure the effectiveness of these protocols. As an

example, Aetna is involved in a program with eviti Inc., to provide participating

Aetna oncologists in two states with access to eviti’s evidence-based decision

support tool.19

Together, payors, physicians, and laboratories hold the key to ensuring optimal

ordering of new diagnostics.

DATA EXCHANGE SUPPORTS AN OUTCOMES-DRIVEN

REIMBURSEMENT MODEL

Data exchange involves 1) the collection of data, 2) a repository in which the data

is federated, and 3) the expert analysis and formulation of the data into decision

support tools for physicians guiding the ordering of molecular diagnostics and the

interpretation of the results. The current regulatory environment forces diagnostic

providers to be passive purveyors of tests, not involved in the discipline of their

ordering. This forced separation results in the Ofce of the Inspector General

(OIG) preventing the free ow of data between the needed parties. The CED

process does not support, and by failing to create an enabling mechanism

effectively prohibits, this data exchange.

pg 14

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

Data Collection

The collection of data would need to include outcomes, phenotypes exhibited or

indicators observed, and specic biomarkers and methodologies. Outcomes, in

reference to diagnostics, are probably better referred to as actionable results.

Unlike therapeutics, where outcomes (quality of life, survival rate, etc.) may not be

known for years, the “outcomes” of most diagnostics are known within days—the

physician reads the report, and either takes the same action s/he would have

without the diagnostic, or takes a different action. An incentive system must be

put in place, potentially as part of the nal laboratory report, where reimbursement

is tied to improved stratication or diagnosis, prognosis, monitoring or predicting

response.

Data Repository

Currently, physicians struggle to keep abreast of the subtle differences between

similar diagnostics and then to ensure they are correctly interpreting and acting

on them. Developers attempt to collect data about the set of conditions that led

to the diagnostic’s prescription and the set of actions taken as a result of the

diagnostic. Payors struggle to understand the correct ordering of diagnostics

and if the diagnostic impacted the course of treatment. Driven by the need for

coverage, diagnostic providers are opting-in to Palmetto’s MolDx program (PTI

codes), the McKesson Z-Codes, and other registry systems. These systems

provide a promising beta test for how such registry systems might work (e.g.,

they provide the additional specicity that was lacking from code stacks), but they

also identied a new challenge by being so specic that they prevent variants

of the diagnostic from easily being ordered or reimbursed. If during the course

of use, a clinically useful biomarker variant is identied, current registry systems

set up the variant as a new diagnostic rather than creating a branch. Additionally,

the current systems may suffer from a perceived lack of independence needed

to be a universal, impartial “single source of truth.” This may result in some

diagnostic providers being somewhat reticent to contribute the desired level of

data needed to optimize healthcare decisions: 1) indications and decision support

leading to diagnostic ordering and 2) results and decision support to guide results

interpretation.

The National Institutes of Health (NIH) has created the Genetic Testing Registry

(GTR)20 and ClinVar21 to aggregate test-related information. GTR provides

pg 15

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

detailed information about the purpose of the test and the indications (primarily

diagnoses) leading to use. ClinVar aggregates information about sequence

variation and its relationship to health. Together, these two data resources can

provide the foundation for a collaborative, easy to use and impartial repository,

with the caveat that the data in any repository is inherently limited by the honesty

of the submitters. A solution to facilitate the free exchange of data may be

relatively close.

The GTR accumulates clinical utility statements and citations for individual

tests and curates professional practice guidelines, position statements, and

recommendations. A results database, maintained by the developer, could utilize

the GTR to refer to accessioned tests that explicitly describe the diagnostic.

With versioning, changes in biomarkers or methodologies—or other factors—

could be delineated clearly for an evolving test and would enable the reporting

of outcomes to continue. The GTR currently allows physicians to see a list of

molecular diagnostics; view information about validity, prociency testing, and

utility; understand the appropriate use cases (phenotypes/indications); and

access a wide variety of resources (relevant trials/studies, molecular details,

drug labeling related to companion diagnostics). Developers of diagnostics could

also be required to provide a feedback/reporting mechanism as a condition of

coverage. HIT vendors would utilize the public, open source data to provide

content for optimizing ordering and reporting through EMRs, CPOEs, PMS,

etc. The foundation for the proposed paradigm shift in diagnostic prescription,

development, and reimbursement is predicated upon the free exchange of

data. Physicians must provide data to developers so developers can provide

effective decision support tools that allow payors to know diagnostics are being

prescribed and interpreted correctly, and payors, in turn, will provide coverage

for the diagnostic the physician wants to prescribe. The payor’s role must shift

from simply paying the bills, to strategically investing in (covering the cost of)

diagnostics.

Decision Support

Diagnostic developers—labs—have the most complete information about

when to order their diagnostics and how the results should be interpreted. To

pg 16

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

ensure physicians have the best possible information, the burden of creating

and providing decision support should be placed with the diagnostic provider.

An appropriate risk sharing model can ensure that labs provide unbiased

information. Registry information will facilitate the timely renement of such

decision support and the eventual establishment of practice guidelines.

Figure 6 shows a highly simplied example of decision support for ordering the

hypothetical molecular diagnostic, ABC-test. The data in the table is a summary

of physician-reported utilization and treatment decisions. In this example there

are again three established indicators,

each with equal prognostic weight,

and one new, relatively unutilized

indicator. Additionally, treatment in this

case is assumed to be binary (altered

or unaltered), when in practice, there

is likely to be a sliding scale as to the

extent treatment has been altered.

These data, combined with the payors’

current coverage decision, allows the physician, with the patient, to make an

informed decision about the appropriateness of the diagnostic.

The second critical aspect to a decision support system for molecular diagnostics

is the results decision support. Figure 7 shows a highly simplied example of

decision support for interpreting the results from ABC-test. The data in the table

is a summary of physician reported results and treatment decisions. In this

example there are three sets

of biomarkers observed,

the number of cases each

set was observed and the

percentage of those cases

where treatment was altered.

Again, treatment is assumed

to be binary. Importantly, the

data also includes a physician retrospective—does the physician feel s/he made

the right decision. The higher percentage of physicians that feel they made the

pg 17

1

2

3

New

Total

100

200

390

10

700

25%

32%

80%

90%

410

Indicators Present # of Prescriptions Altered Treatment

Figure 6: ABC-Test decision support for ordering. Empirical data to guide physicians in

prescribing the diagnosc and to set payment raos.

50%

70%

90%

Total

300

250

150

700

Biomarkers

Matched

33%

68%

94%

410

(n=99) 50% Yes

(n=170) 80% Yes

(n=141) 90% Yes

(n=201) 50% Yes

(n=80) 30% Yes

(n=9) 20% Yes

Altered

Treatment

Physician Retrospective

Altered Unaltered

# of Patients

Figure 7: ABC-Test results decision support. Empirical data guides physician in selecng a treatment.

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

right decision based on the results of the diagnostic further validates the decision

support tool. This data quickly and easily provides physicians with the latest

utilization data.

Ideally, a data exchange system supporting data collection, curation, and

decision support could be implemented that would allow coverage decisions for

new molecular diagnostics with little or no regulation/oversight.

Take Advantage of the Shift Toward Pay for Value (PFV) Instead of Fee

For Service (FFS)

As shown in Figure 8, Fee for Service (FFS), the most prevalent reimbursement

model in the US healthcare system, is projected to rapidly give way to Pay for

Value22 (PFV). The change is being

driven by the exploding costs of

health care and is being enabled by

the relative ease of data sharing. An

FFS system is simpler than a PFV

system; under FFS, a service (checkup,

diagnostic, treatment) is provided and

payment is rendered, whereas PFV

requires a feedback loop to determine

payment. This feedback loop is a

challenge because it requires all

parties to use more data; physicians,

diagnostic providers, and payors need

to consider it.

The processes already in place for molecular diagnostics reimbursement allow

for a limited PFV system. As part of the negotiation between developers and

payors, conditions for reporting use patterns can be established directly, tying

coverage to correct use of the diagnostic. These conditions are almost always

part of the CED process while clinical utility is being demonstrated, but do not

have to be limited to that process; they can be required after clinical utility has

been established. Extending reporting after the demonstration of clinical utility

pg 18

Projected Decline in Fee-for-Service

Percent of Hospital-based Physicians

Figure 8: Fee-for-service reimbursement models are expected to be rapidly replaced

over the next decade by pay-for-value models.22

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

would allow for the collection of additional data about similar variations on

approved use conditions.

CONSIDERATIONS FOR REIMBURSEMENT RATES

In simple terms, there should be a sliding scale for payment responsibility. A

test that has little impact on the course of treatment, arguably has little value,

and should be reimbursed at a correspondingly low rate. Similarly, a test that

signicantly alters the course of treatment (compared

to without the test) has signicant value and should be

reimbursed at a correspondingly higher rate (Figure 9). Of

course payment for medical services is not this simple and

straightforward.

Assuming a philosophy of shared risk/payment for

molecular diagnostics based on the degree to which the

diagnostic altered the course of treatment, there is still the

core question, “What is an appropriate price for a test?” To

begin answering this question, look to the analyte being

measured (Figure 10).

If the key analytes or sequences being studied by a new diagnostic are

new, it is important to understand if there is a similar diagnostic for the same

indicated use (different gene, identical condition). If there is a similar diagnostic,

with less sensitivity or specicity for the identical condition (traditional PSA

vs. ultra-sensitive PSA, CTID vs. IHC), then the new diagnostic should be

reimbursed incrementally more than the existing, less sensitive, diagnostic. The

reimbursement uptick should be based on the relative superiority of the new

diagnostic over its predecessor. If the new analyte is equal to or no better than

the existing diagnostic they both should be reimbursed at a discount as they are

commodities.

If there is no similar diagnostic for the condition, then the impact on the standard

of care must be determined. The new diagnostic should be reimbursed at a

rate appropriate for the value it delivers. When the clinical utility is anticipated

but not proven, the diagnostic initially should be reimbursed at cost plus, until

pg 19

Figure 9: A sliding scale of cost responsibility based on

“aconability” of results.

Payor

Responsibility

Developer

Responsibility

% of Cost

Treatment as a result of diagnosis

Unchanged Somewhat

changed

Significantly

changed

100%

75%

50%

25%

0%

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

clinical utility and overall health economic value is determined. At this time the

test should be reimbursed based on its value. There are a number of established

methods for determining the value of specic diagnostics that should be used.23, 24

Finally, if the analyte being studied is not new, or clinical utility of the existing

diagnostic is not obvious or available, the reimbursement rate is essentially cost-

plus—there is no justication for anything else. If the clinical utility is known, then

a value-based price with cost consideration would be appropriate.

pg 20

New

Analyte?

Similar

Diagnostic?

Superiority

Over Standard?

Known

Care Impact?

Clinical

Utility?

Cost Plus

Not obvious

or available

Value-BasedValue-BasedDiscountPremium

Yes

Yes

Yes

No

Yes

Yes

No

No

Demonstrate

Clinical Utility

&

Health

Economics

Value-based

with Cost

Consideration

Figure 10: Proposed price determinaon decision methodology for an analyte.

Price Determinaon Decision Factors

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

CONCLUSION

For gradually evolving science, therapies and medical equipment that can

jeopardize the health and well-being of the patient, the more cautious and

lengthy process of a CED may provide an excellent scientic validation assuring

that payor resources are not wasted on ineffective treatments.25 However, in

the fast-paced world of molecular diagnostics, the very controls that make CED

so effective become a hindrance for the development of a diagnostic that can

protect against the misapplication of a therapy. The rigid testing protocols, long

testing periods, and the de facto use of CED as a means for private payors to

avoid coverage, all make the current and formally proposed changes to CED a

roadblock to the development of new molecular diagnostics. Just as is the case

with software development and analysis of big data, a new, faster, more dynamic

process is needed to ensure that CMS remains a leader in outcomes-based

medicine and fullls the meaning of the National Bioeconomy Blueprint.

A more efcient process providing the speed and agility demanded by patients

and physicians, the real-world utilization data needed by laboratories, and the

cost assuredness demanded by payors is possible for molecular diagnostics by

relying on:

• A risk-sharing, market-driven, pricing structure;

• The free exchange of data including collection, an unbiased repository, and

decision support; and

• The healthcare-wide push for pay for value business model.

ACKNOWLEDGEMENTS

We gratefully acknowledge Wendy Rubinstein, Director, NIH Genetic Testing

Registry, for reviewing the content related to NIH’s Genetic Testing Registry and

ClinVar.

© 2013, XIFIN, Inc. All rights reserved. XIFIN and the XIFIN logo,are registered trademarks of XIFIN, Inc. All

other marks are property of their respective owners. XMD-3034-13

pg 21

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

REFERENCES

1. “CMS Financial Report: Fiscal Year 2012” Centers for Medicare and Medicaid. November 15, 2012.

http://tinyurl.com/bc6qbpg, Accessed February 12, 2013.

2. U.S. Department of Health and Human Services. http://tinyurl.com/by7lbeq, Accessed February 12,

2013.

3. “Draft Guidance for the Public, Industry, and CMS Staff Coverage with Evidence Development in

the context of coverage decisions.” Centers for Medicare & Medicaid Services. November 29, 2012,

http://tinyurl.com/cyp3jkd, accessed January 28, 2013.

4. “Draft Guidance for the Public, Industry, and CMS Staff Coverage with Evidence Development in

the context of coverage decisions.” Centers for Medicare & Medicaid Services. November 29, 2012,

http://tinyurl.com/cyp3jkd, accessed January 28, 2013.

5. “Novitas LCD DL33138 – Biomarkers for Oncology” Novitas Solutions, January 29, 2013. http://

tinyurl.com/ae82wcd, accessed February 19, 2013.

6. “National Coverage Determinations with Data Collection as a Condition of Coverage: Coverage with

Evidence Development.” Centers for Medicare & Medicaid Services. July 12, 2006, http://tinyurl.com/

athoyo2, accessed January 28, 2013.

7. “National Bioeconomy Blueprint.” White House. April 26, 2012 http://tinyurl.com/cd9j485, accessed

January 28, 2013

8. “Draft Guidance for the Public, Industry, and CMS Staff Coverage with Evidence Development in

the context of coverage decisions.” Centers for Medicare & Medicaid Services. November 29, 2012,

http://tinyurl.com/cy3jkd, accessed January 28, 2013.

9. “Invention reinvented: McKinsey perspectives on pharmaceutical R&D 2010” McKinsey & Company,

p.92 http://tinyurl.com/nhqx9zj, accessed July 11, 2013

10. “Lab Group Submits Citizen Petition against FDA Regulation of Lab Developed Tests” GenomeWeb

June 5, 2013, http://tinyurl.com/km7ldnm, accessed June 6, 2013

11. “Evaluation of Clinical Validity and Clinical Utility of Actionable Molecular Diagnostic Tests in Adult

Oncology” Center for Medical Technology Policy, May 1, 2013. http://www.cmtpnet.org/effectiveness-

guidance-documents/molecular-diagnostics-egd/, accessed May 16, 2013

12. “The Gap in Medical Testing,” New York Times, the Editorial Board, July 7, 2013, http://tinyurl.com/

lgr9w6y, accessed July 11, 2013

13. “The Value of Laboratory Screening and Diagnostic Tests for Prevention and Health Care

Improvement” The Lewin Group, Inc. September 2009. http://tinyurl.com/k4af3tv, accessed March 4,

2013

14. Italiano, Antione. “Prognostic or Predictive? It’s Time to Get Back to Denitions!” Journal of Clinical

Oncology, Dec. 10, 2011. 29:35, p. 4718 http://jco.ascopubs.org/content/29/35/4718.1.full, accessed

March 2, 2013.

15. “PROPOSED/DRAFT Local Coverage Determination (LCD): Molecular Diagnostic Testing (DL33219)”

Centers for Medicare and Medicaid Services, http://tinyurl.com/nfyfvde, accessed July 1, 2013

16. Callen, J, et al. “Failure to Follow-Up Test Results for Ambulatory Patients: A Systematic Review”

Journal of General Internal Medicine, Dec 20, 2011. http://link.springer.com/article/10.1007/s11606-

011-1949-5/fulltext.html, accessed June 20, 2913

17. “Risk Sharing Schemes in Oncology: Future or Fad?” Oncology Business Review. March 2010, p. 16.

http://obroncology.com/documents/OBR_MAR10_RS(1).pdf, accessed June 24, 2013

18. “Payers Try New Approaches To Manage Molecular Diagnostics” Biotechnology Healthcare, Fall

2010; 7(3) p.26. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2957737/, accessed June, 24, 2013

pg 22

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

19. “Aetna, eviti Launch A Decision Support Program To Make Cancer Care More Effective And Efcient”

Aetna News Hub. August 28, 2012. http://tinyurl.com/o6vloz4, accessed June 24, 2013

20. Rubinstein, W. et al. “The NIH genetic testing registry: a new, centralized database of genetic tests

to enable access to comprehensive information and improve transparency” Nucleic Acids Research

2012; 41:D1 D925-D935 http://nar.oxfordjournals.org/content/41/D1/D925.full, accessed July 5, 2013

21. “ClinVar” National Center for Biotechnology Information, U.S. National Library of Medicine. http://www.

ncbi.nlm.nih.gov/clinvar, accessed July 5, 2013

22. “The View from Healthcare’s Front Lines: An Oliver Wyman CEO Survey” Oliver Wyman, http://

www.oliverwyman.com/media/The_View_from_Healthcares_Front_Lines_An_Oliver_Wyman_CEO_

Survey.pdf, accessed March 8, 2012

23. Di Ruffano, Lavinia, et al. “Assessing the value of diagnostic tests: a framework for designing and

evaluating trials” BMJ 2012; 344:e686 http://www.bmj.com/content/344/bmj.e686, accessed March 8,

2013

24. Moons, K, et al. “Quantifying the Added Value of a Diagnostic Test or Marker” Clinical Chemistry

2012; 182550 http://www.clinchem.org/content/58/10/1408.full.pdf+html, accessed March 8, 2013

25. Daniel, G.et al. “Coverage With Evidence Development for Medicare Beneciaries: Challenges

and Next Steps,” Journal of the American Medical Association, July 22, 2013. 173:14 p 1281 http://

archinte.jamanetwork.com/article.aspx?articleid=1685892, accessed July 22, 2013

pg 23

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 24

ABOUT THE AUTHORS

Paul Billings, CMO, Life Technologies

Board certied internist and clinical geneticist Dr. Paul R. Billings serves as

Chief Medical Ofcer at Life Technologies. Dr. Billings brings extensive expertise

and clinical experience in the areas of genomics and molecular medicine. Most

recently, he served as Director and Chief Scientic Ofcer of the Genomic

Medicine Institute at El Camino Hospital, the largest community hospital in the

Silicon Valley. He currently serves as a member of the United States Department

of Health and Human Services Secretary’s Advisory Committee on Genetics,

Health and Society, where he helps shape policy in the rapidly evolving eld of

genomic medicine. Dr. Billings has had a distinguished career as a physician

and researcher. He has been a founder or chief executive ofcer of companies

involved in genetic and diagnostic medicine, including GeneSage, Omicia

and CELLective Dx Corporation. Previously, he was senior vice president

for corporate development at Laboratory Corporation of America Holdings

(LabCorp). He is the author of nearly 200 publications and books on genomic

medicine. Dr. Billings holds an MD from Harvard Medical School and a PhD in

immunology, also from Harvard University.

Richard Ding, CEO, bioTheranostics

Richard Ding has over 20 years of management experience in the diagnostics

and pharmaceutical industry. As a member of the team at Myriad Genetics

that cloned BRCA1 gene, Richard Ding made contribution to the development

of BRAC tests. Prior to becoming CEO of bioTheranostics, Richard Ding had

taken various cross-functional positions at Eli Lilly and Company. More recently,

Richard serves on the Executive Committee at bioMérieux as Senior VP, Strategy

and Business Development. Major achievements include twelve M&A deals and

a number of companion diagnostics collaboration with major pharmaceutical

companies. Richard holds an MBA from the University of Utah and received his

undergraduate education and scientic training at Fudan University, China.

Lâle White, CEO, XIFIN, Inc.

Lâle White is a nationally recognized expert in the eld of medical nancial

management and regulatory compliance, with over 25 years of experience in

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests

information systems development and medical billing. She lectures extensively

on these topics and has consulted for major laboratories and laboratory

associations throughout the US. She worked with HCFA and the US Ofce of

the Inspector General to develop the rst OIG Model Compliance Program.

Ms. White was previously Vice President - Finance of Laboratory Corporation

of America (NYSE: LH), one of the largest clinical reference laboratories in the

US, and its predecessor National Health Laboratories (NYSE: NHLI), where she

led the software development of several accounts receivable, inventory, cost

accounting and nancial management systems for the laboratory industry. Ms.

White has a BA in nance and an MBA from Florida International University.

Rina Wolf, Vice President of Commercialization Strategies, Consulting &

Industry Affairs, XIFIN, Inc.

Rina Wolf is a nationally recognized expert in the eld of laboratory

commercialization and reimbursement, with over 20 years of experience in the

diagnostic laboratory industry, specializing in Molecular Diagnostic Laboratories.

She lectures extensively on these topics and has consulted for major laboratories

and laboratory associations throughout the US. She is a former president

and board member of the California Clinical Laboratory Association and is an

active participant with the American Clinical Laboratory Association and the

Personalized Medicine Coalition. Ms. Wolf also advises and presents to investor

audiences, recent speaking engagements include Piper Jaffray, Cowen Group

and Bloomberg’s G2 Intelligence Lab Investment Forum. Most recently Ms. Wolf

held the position of Vice President of Reimbursement and Regulatory Affairs at

Axial Biotech, Inc. where she was responsible for creating and implementing their

successful reimbursement strategies. Prior to joining Axial Biotech, Inc., Ms. Wolf

held executive positions in the area of commercialization and reimbursement at

RedPath Integrated Pathology, Inc., Genomic Health, Inc., and Esoterix (now

LabCorp). Ms. Wolf has a Bachelor of Arts degree from UCLA and a Masters of

HealthCare Administration.

David Lorber, Director of Business Development, XIFIN, Inc.

David Lorber is the Director of Business Development at XIFIN.

Before joining XIFIN, he was the Director of Global Sales Operations at Accelrys;

a leading scientic informatics company. He also held various marketing

pg 25

A Pay-For-Value, Data-Driven Approach for the Coverage of Innovative Genetic Tests pg 25

and product management roles at Accelrys, bringing new products to market

and redesigning the corporate pricing strategy. He began his career as a

computational chemist designing early-stage drug leads for oncology targets. He

holds a PhD in Computational Chemistry from Northwestern University and an

MBA from the Rady School of Management at UC San Diego.