Next-generation sequencing is used with increasing frequency to provide essential information about a patient’s diagnosis and treatment. In recent months, the U.S. Food and Drug Administration (FDA) has approved several new next-generation sequencing diagnostic tools, and the Centers for Medicare & Medicaid Services (CMS) has issued a positive national coverage decision.

Next-generation sequencing tests can provide data about hundreds of genomic alterations at once, thus ensuring greater sophistication of diagnosis and treatment. However, the types of evidence that result from the tests and the way they are communicated to interested parties are still open to debate.

Such debate was the order of the day at the Friends of Cancer Research (Friends) Blueprint for Breakthrough Forum: Research and Reimbursement in the Age of Precision Medicine held in Washington, DC, late this summer and cosponsored by Alexandria Real Estate Equities, Inc.

Panelists discussed the following three key questions about evidence development and reimbursement.

Minimum Core Data

What are the minimum core data that should be made publicly available for next-generation sequencing–based diagnostic tests, and what information do patients, providers, and payers need?

The panelists agreed that increasing sophistication of the technology makes it more difficult and more imperative to validate it and communicate performance specifications. Everyone involved must be confident that diagnostic tests are clinically valid and appropriate—but because of their complexity, it is increasingly difficult to communicate this information to those who need it. 

It is important to identify innovative methods and standards for data collection in the real world in order to assess the effect of [next-generation sequencing] testing on patients’ experiences.

— Jonathan Hirsch, MS

Tweet this quote

For example, many laboratories have begun reporting gene signatures relevant to oncology such as microsatellite instability and tumor mutational burden, both of which make validation more complex. Moreover, the detail required differs significantly depending on who is to receive it.

One solution might be for laboratories to provide test performance characteristics, along with results, in a standard format on a public database, thus allowing providers and patients to assess the advantages and disadvantages of various tests. Or a list of tests that meet certain analytic and clinical performance criteria (such as those that are clinically actionable such as BRAF, BRCA1/2, EGFR, ERBB2, and the like) could be made publicly available. 

Real-World Data Collection

What mechanisms can support collection of relevant data in the real world, and what standards are required?

The analytic and clinical validity of various diagnostic tests is critical not only to diagnosis and treatment, but to reimbursement as well. Panelists discussed ways to assess clinical utility and changes in patient outcomes without conducting full clinical trials.

They explored evidence from the real world, including vehicles and standards for data collection, identifying real-world endpoints that can establish clinical utility, defining a pathway to validate such endpoints, and establishing a framework for reimbursement.

Jonathan Hirsch, MS, Founder and President, Syapse, San Francisco, said that collecting real-world evidence can be practical, but data must be both retrospective and prospective. “It is important to identify innovative methods and standards for data collection in the real world in order to assess the effect of [next-generation sequencing] testing on patients’ experiences.” They might include:

• Disease characteristics such as primary cancer type, stage at diagnosis, status of metastasis, and prior treatment, because these parameters affect clinical endpoints
• Test characteristics such as vendor, type, genes and their variants, and genomic results
• Change in care as a result of test results, including intent to change treatment, actual change in treatment, and differences between them
• How change affects clinical outcome such as overall survival, progression-free survival, tolerability/toxicity, and objective response rate
• Nonclinical measures, such as patient-reported outcomes.

Data Reporting

How should the reporting of data be formatted to make it as informative as possible?

High-quality information regarding diagnostic tests is critical for providers and patients, but the data have to be understandable to the intended audience.

Providers need to know the expectations and capabilities of each test in order to support decision-making. Katherine Szarama, PhD, Social Science Research Analyst, CMS Center for Clinical Standards and Quality, said that CMS has deemed next-generation sequencing diagnostic tests to be reasonable and necessary, and the agency reimburses them as long as they are performed in a Clinical Laboratory Improvement Amendments (CLIA)-certified laboratory and if the following requirements are met:

• The patient has recurrent, relapsed, refractory, metastatic, or advanced of cancer.
• The same next-generation sequencing test has not been used previously for the same primary diagnosis.
• The patient has decided to seek further treatment such as chemotherapy.
• The next-generation sequencing test has FDA approval as a companion in vitro diagnostic, as well as approval for that patient’s cancer.

CMS has a program called Appropriate Use Criteria that can be adapted to communicate information about quality and appropriateness of diagnostic tests by incorporating it into an electronic decision-making mechanism used by providers. The Appropriate Use Criteria program takes into account accuracy, analytic sensitivity, precision, sample stability, clinical performance characteristics, and clinical utility.

Most patients don’t understand or want the specificity that providers need, but they still need assurance that a test is valid. A general grading scale could be devised and administered by the Appropriate Use Criteria program and conveyed to patients via their providers.

Additional Questions

In addition to the key questions discussed by the panelists, Friends has posed others that the organization deems essential to the issue of next-generation sequencing:

• What establishes confidence in a clinically well characterized biomarker with an existing companion diagnostic test? Is a clinical trial always necessary to determine this?
• Could clinical experience with a next-generation sequencing test be used to identify a targeted population? If so, what are the required data?
• Could a next-generation sequencing test be eligible for reimbursement without being developed as a companion to a drug, and if so, under what circumstances?
• When a new diagnostic/drug pair is approved for a new variant or new indication, what evidence should be required for both regulatory and reimbursement approval?
• Should “higher” levels of evidence support higher levels of reimbursement? What tiers should be established?
• What incentives or legal protections would promote data sharing? ■

DISCLOSURE: Mr. Hirsch is Founder and President of Syapse. Dr. Szarama reported no conflicts of interest.