Patricia Deverka, Principal Researcher, Health Care Group
American Institutes for Research, Chapel Hill, NC; Adjunct Associate Professor, Center for Pharmacogenomics & Individualized Therapy, University of North Carolina at Chapel Hill
Q: Paying for Precision Oncology – Who Decides?
A: There is tremendous enthusiasm for the scientific rationale and clinical promise of precision oncology amongst researchers, oncologists, industry and subgroups of cancer patients. Nearly three-quarters of the compounds in the oncology pipeline have the potential to become personalized medicines and over 90% of the $25 billion personalized medicine market in 2015 came from the sale of oncology drugs that required use of a companion diagnostic, such as Herceptin and Her2 testing. The typical companion diagnostic analyzes single gene mutations or abnormal gene expression profiles, in contrast to next generation tumor sequencing (NGTS), which assays genomic alterations in tens to hundreds of genes simultaneously. Tumor profiling using NGTS now occurs frequently at major cancer centers across the U.S., however reimbursement for both the test and the off-label use of targeted therapies is unpredictable and challenging for most molecular tumor board staff, oncologists and their patients. Without coverage by either private or public insurers, most patients will not have access to NGTS tests and targeted therapies.
Payers typically use a stepwise approach for coverage decision-making when evaluating the evidence supporting the technical and clinical aspects of new molecular diagnostic tests. Based on published studies, technology assessments and professional guidelines, payers assess:
1) analytic validity – whether a new test is accurate and reliable,
2) clinical validity – if the result is medically meaningful, and
3) clinical utility of the test – whether results affect clinical decisions and improve health outcomes.
Payers are one of key stakeholder groups that require evidence of clinical utility for decision-making, however clinical utility cannot be demonstrated until analytic and clinical validity are established. As compared to “ single test/single result” assays, each of the two validation steps are inherently more difficult for payers to assess with NGTS tests, due to the complexity of the technology, as well as variation in informatics analyses and procedures for interpretation and reporting of sequence variants. Economic considerations such as whether use of the test will lead to cost offsets (e.g., reductions in hospitalizations) are potential factors in test evaluations, but they also depend on evidence of clinical utility since changes in resource utilization must be linked to an alteration in patient outcomes, such as improved response or avoidance of side effects compared to an alternative approach.
When determining medical policy for an insured group, assessment of the evidence base typically leads to a determination of whether a test is “medically necessary” and therefore covered, or “experimental/investigational” and not covered. NGTS tests create particular challenges for payers given that they include both established and novel targets and test results are used to guide the use of off-label therapy, thereby challenging the standard approach to evidence evaluations. There is growing recognition that the traditional reliance on randomized controlled trials to demonstrate clinical utility is not feasible and that use of basket trials, observational studies, registries, n-of-1 studies and modeling may be required. These study methods are less familiar to payers and will require education and published examples. Nevertheless, more payers are acknowledging that flexibility in evidence requirements is required and oncology is one of the most likely areas for innovation.
However while many payers view NGTS clinical applications as a reasonable hypothesis, they still consider this approach to be investigational. What is needed now is stronger evidence to support the proof of concept of using NGTS to guide treatment decisions. One example is the National Cancer Institute’s basket trial, Molecular Analysis for Therapy Choice (NCI-MATCH), which pairs patients’ genomically profiled tumors with a treatment regardless of anatomical location. The study has recently expanded the number of targeted therapy treatment arms and is actively enrolling patients. In the meantime, the current model of health care delivery encourages standardization and use of treatment pathways as a mechanism to promote quality of care and reduce unnecessary costs. Thus the purported advantages of NGTS tests such as individual patient customization run counter to current system incentives.
Without convincing evidence that NGTS tests lead to better outcomes through better decision-making, payers will continue to struggle in the near-term to develop transparent, evidence-based affirmative coverage decisions. One example of how the Blue Cross Blue Shield Association is addressing the problem is a new subscription service called Evidence Street. Researchers in this group conduct technology assessments of new drugs, devices and tests, including molecular diagnostics. They work with test developers, commercial laboratories, drug companies and professional societies to ensure more consistent and transparent evidence standards for their technology assessments. Evidence Street does not make coverage decisions, but provides evidence to support Blues plans in their technology evaluations throughout the country.
So what is the path forward? There needs to be a combination of efforts from all stakeholders. Researchers in both for-profit and academic settings need to design and publish context-specific NGTS clinical utility studies, using a variety of appropriate methods. They should involve patients, caregivers and payers in study planning to ensure relevancy of results and use a range of outcome measures, including those developed by various research groups supported by NHGRI, such as the Clinical Sequencing Exploratory Network (CSER). Whenever possible studies should be conducted in practice-based research networks to reduce data collection costs and reflect usual care.
There are efforts underway to collect test use and patient outcomes data more systematically as part of multi-stakeholder supported registries (ASCO’s Targeted Agent and Profiling Utilization Registry; Molecular Evidence Development Consortium) and through third party oncology data aggregators (Syapse, Flatiron). Medicare’s MolDX (molecular diagnostics) program administered by Palmetto GBA allows the option of provisional coverage for tests under its Coverage with Data Development program which is used for promising tests that meet evidence standards for analytic and clinical validity but lack sufficient proof of clinical utility.
While NGTS tests are potentially disruptive technologies, if they are to be widely covered by private and public payers, stakeholders will need to demonstrate evidence of test accuracy, medical relevancy and ability to improve the process and/or outcomes of care. Assessment of clinical utility answers practical questions such as “Can and will we take action on the NGTS test results?” and “Does the outcome change in a way in which patients and other stakeholders find value relative to the outcome without the test?” Given the opportunity costs for patients, families and the health care system, we must work together to answer these questions.
Copyright: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.