Personalized Medicine Education and Advocacy

by Cynthia A. Bens, Senior Vice President, Public Policy, PMC

Cynthia A. Bens

On November 30, 2017, the U.S. Food and Drug Administration (FDA) and the Centers for Medicare & Medicaid Services (CMS) announced the approval and preliminary coverage of Foundation Medicine’s FoundationOne CDx. As part of its announcement, CMS released a draft proposed decision memo for the FoundationOne CDx and similar next-generation sequencing (NGS) tests for use in the diagnosis of advanced cancer.

The draft coverage proposal was scrutinized by numerous stakeholders during an extended public comment period that resulted in CMS receiving nearly 400 pages of feedback. Many of these interest groups, including the Personalized Medicine Coalition (PMC), requested clarification from CMS on different aspects of the proposed decision memo. PMC’s comments highlighted the need to extend coverage for NGS testing to earlier stages of cancer and make allowances for retesting during the course of a patient’s life. In addition, PMC called on CMS to expand full coverage beyond FDA-approved NGS tests and consult with stakeholders to define a rigorous but less burdensome pathway to coverage with evidence development (CED) for NGS tests that do not meet CMS’ proposed criteria for full coverage.

Fortunately, CMS addressed many of the community’s concerns in the resulting national coverage determination (NCD) that was released on March 16, 2018. The following is a summary of what we view as major changes from the draft NCD.

• Coverage granted for FDA-approved and FDA-cleared NGS-based in vitro companion diagnostic tests. FDA-cleared NGS tests for advanced cancer would have fallen under a CED program in the original program. The final NCD expands the scope of coverage to NGS tests that are cleared as in vitro companion diagnostics through FDA pathways, such as 510(K), in addition to tests that are FDA-approved as in vitro companion diagnostics. Because the final determination does not specify sample type, FDA-cleared or -approved NGS liquid biopsy tests that are in vitro companion diagnostics will be covered as they become available, provided that all of the required patient and test indications are met.
• NGS tests indicated for use outside of advanced cancer not at risk of immediate non-coverage as previously thought. Under the proposed NCD, Medicare coverage was uncertain for tests that did not meet the narrow categories of FDA-approved or -cleared tests, and even these tests would have been subject to CED. Coverage decisions for both NGS-based in vitro diagnostics and laboratory-developed tests (LDTs) in areas such as hereditary testing, screening, and other non-cancer conditions will remain with local Medicare Administrative Contractors (MACs). Under the final NCD, coverage for NGS-based tests in patients with cancer without FDA-approved or -cleared companion diagnostic indications may be handled through the local coverage determination process, subject to the restrictions set forth in the final NCD.
• CED dropped. CED was an important but problematic feature of CMS’ draft. In the proposed coverage decision, CED programs would have been required for all other NGS-based cancer tests that were not FDA-approved NGS-based in vitro companion diagnostic tests for advanced cancer. Further, CED for LDTs was limited only to those participating in an NIH-NCI National Clinical Trial Network. CED was not included in the final CMS determination, and instead, non-FDA-approved NGS tests will need to be evaluated by the MACs to determine coverage status.
• Additional cancer indications included for coverage and one-time testing limitation removed. The proposed decision memo only provides pathways to coverage for NGS testing for Medicare beneficiaries with recurrent, metastatic, and stage IV cancer. In addition to coverage for these patients, the final NCD also covers NGS testing for patients with either relapsed, refractory, or stage III cancer when the NGS test meets the diagnostic assay requirements for coverage. The final NCD precludes coverage for patients with stage II or earlier cancer. In the final NCD, CMS also expanded the frequency of testing allowed, from using the same diagnostic laboratory test once to using the same diagnostic laboratory test once for each new primary diagnosis of cancer. Repeat testing for the purposes of treatment monitoring is not included in the final NCD.

The NCD is a net positive for NGS testing, as it now guarantees Medicare coverage at a national level for FDA-approved or -cleared NGS-based companion in vitro diagnostic tests for advanced cancer. PMC applauds CMS for recognizing that NGS testing is a breakthrough technology that is critical to advancing personalized medicine. We will continue to work with policymakers to advance diagnostics regulations and legislation that we believe will further improve patient care.

The featured speaker at PMC’s Policy Committee Meeting on April 24 will address the scope of CMS’ final coverage determination and its implementation. Ongoing federal activities related to diagnostics will also be the focus of a BIO 2018 session moderated by PMC. The panel, “Regulatory Oversight of Personalized Medicine: Today and Tomorrow,” will take place in Boston, Massachusetts, on June 6, 2018.

by Daryl Pritchard, Ph.D., Senior Vice President, Science Policy, PMC

Daryl Pritchard, Ph.D.

The Personalized Medicine Coalition’s most recent white paper, Personalized Medicine and Value Assessment Frameworks: Context, Considerations, and Next Steps, contends that to improve clinical outcomes and facilitate more cost-effective health care, value assessment frameworks (VAFs) must incorporate the principles of personalized medicine. The report examines the characteristics of U.S.-centered VAFs and identifies how each integrates or, as in most cases, fails to integrate personalized medicine.

Health insurance companies have been conducting health technology assessments for years. As part of these processes they may or may not have considered value assessment reports from independent organizations.  Recent indications from both public and private payers, however, have suggested a greater interest in the use of VAFs for coverage and payment decision-making.  Health care delivery organizations and physician groups have also taken a greater interest in using frameworks to help quantify the value of various treatments and interventions. This has increased the likelihood that VAFs will impact access to care.

Several frameworks are already in use by different end-users. ICER’s framework, the Drug Abacus developed by Memorial Sloan Kettering, and frameworks developed by the American Society for Clinical Oncology (ASCO) and the National Comprehensive Cancer Network focus primarily on value to payers and providers at a health system level, while newer VAFs, including FasterCures’ Patient Perspective Value Framework and the Innovation and Value Initiative, were developed with patients in mind.

To facilitate access to treatment options of highest value at both the individual patient and health care system levels, however, VAFs must consider all of these perspectives. PMC’s report argues that to assess the value of new treatments in a way that meets the needs of patients and innovators as well as payers and providers, VAFs must integrate personalized medicine.

By identifying which medical treatments and procedures will work best for each patient, personalized medicine improves efficiency at both the individual and health system levels. But as they are currently designed, most VAFs involve static comparisons based on population health, and therefore neglect to adequately consider patient-level efficiencies learned through ongoing clinical practice. PMC’s report highlights five personalized medicine considerations — namely diagnostic testing, heterogeneity of treatment effects, treatment efficiency, individual values and circumstances, and emerging or evolving value elements, which, if accounted for in VAF methodologies, would help ensure that value is considered at both the health system and patient levels.

In a panel discussion to launch the report, Kristen Migliaccio-Walle, who was the report’s lead author from Xcenda, Dan Leonard, President of the National Pharmaceutical Council, Alan Balch, CEO of the National Patient Advocate Foundation, and Dana Wollins, Director of Health Policy at ASCO, agreed that alignment of VAFs with the goals and benefits of personalized medicine is critical to ensure informed decision making.

Other VAF developers concur. ICER’s President, Steven Pearson, for example, recently noted in an email to PMC that he is “convinced that we share the same vision: a health care system in which insurers and drug makers can continue the progress toward even more [personalized] tests and treatments while making them more affordable and more accessible for the patients who need them.”

The report concludes with five strategic recommendations for developers of VAFs:

1. Consider diagnostic testing as an explicit and integral part of the value assessment of treatment options where efficacy and/or safety information can be obtained;
2. Include a formal mechanism for consideration of heterogeneity of treatment response appropriately balanced with population-based considerations;
3. Develop methods for the consideration of emerging or evolving elements of value over time to fully account for emergent benefits at the health system and individual patient levels;
4. Ensure appropriate awareness and education about the use of value assessment frameworks in personalized medicine to reduce the risk of inappropriate restriction of reimbursement and/or access to individualized care; and
5. Consider the perspectives of all stakeholders, especially patients.

As VAFs have gained prominence and become a focus of discussion by decision-makers within the health care community, it has become more important than ever to realize that patient-level efficiencies convey real value at the health system level. Currently, the value assessment viewpoints of payers and providers are often not aligned with those of patients. Following the recommendations in PMC’s report will help bridge that gap.
Personalized Medicine and Value Assessment Frameworks: Context, Considerations, and Next Steps

Guest Blog
by Mark Gardner, CEO, OmniSeq

Mark Gardner

Immunotherapies, specifically checkpoint inhibitors (CPIs), are an exciting new class of drugs for advanced-stage cancer patients. Unlike patients treated with chemotherapy and targeted therapies, a significant percentage of patients receiving CPIs experience durable responses, with progression-free survival often measured in years.¹

Absent a more clearly defined target population, however, the incremental costs of CPIs may strain resources available for treating an aging population. Given the potential benefits of these drugs, CPIs cost over $100,000 per year, and these costs are rapidly adding up. The U.S. will likely spend more than $6 billion on CPIs in 2017, and an analysis by L.E.K. Consulting suggests that these costs may grow to exceed $16 billion by 2021. Unfortunately, CPIs do not work for approximately 60 – 80 percent of patients, depending on histology.²

In order to assist physicians with their decisions to treat patients with CPIs, there are several biomarkers in use today. Physicians may measure PD-L1 expression by immunohistochemistry (IHC) and potentially by fluorescent in situ hybridization (FISH) to find PD-L1 copy number gains. Recently, FDA also approved the use of pembrolizumab for patients with high microsatellite instability (MSI-H) regardless of histology, and it is likely that other measures of DNA instability and mismatch repair damage (such as high mutational burden) will gain traction as companion or complementary diagnostics.

Unfortunately, while biomarkers such as MSI-high^3-5 and PD-L1 copy number gain^6-8 are highly correlated with response, they are prevalent for only a small percentage of patients. In contrast, PD-L1 expression of greater than 1 percent of tumor cells is found in a high proportion of all patients, but meeting this cutoff is not highly correlated with response.

In July of 2017, scientists at Merck published a paper highlighting the promise of gene expression signatures related to the IFN pathway with promising results and an “area under the curve” (AUC) better than both mutational burden and PD-L1 IHC.⁹ Merck scientists have generally presented a case to use all three categories of biomarkers (PD-L1 expression, DNA damage and gene expression signatures) to best inform treatment decisions.

Given the imperfect state of biomarkers today, the best method to assist oncologists with their decisions to treat with CPIs is through a three-pronged comprehensive immune profile that assesses PD-L1 IHC/FISH, MSI/mutational burden and RNA-seq.

Clinicians should not settle, however, for identifying only the 20 – 40 percent of patients who will respond to CPIs alone. OmniSeq believes gene expression biomarkers can be used to fully assess the cancer adaptive immune response cycle, and patient response rates can be increased by identifying candidates for appropriate combination therapies. This is especially needed in the subset of over 1,000 total immunotherapy trials¹⁰ that are testing combinations of novel agents with existing drugs, such as anti-PD-1 drugs like pembrolizumab and nivolumab.

Indeed, as we interviewed physicians running clinical trials across the country, we did not observe an obvious logic as to how patients are being assigned to experimental combination immunotherapies. Given the complexity of the cancer immune cycle, patients should be guided toward trials primarily based on the biology at work in each patient’s tumor microenvironment. OmniSeq hypothesizes specifically that over-expression of targeted genes relative to a reference population is the most rational method for guiding patients onto combination therapy clinical trials.

Guided by these insights, OmniSeq is excited to participate in a new era of precision immunotherapy, and is actively seeking collaborators to generate data to test whether over-expression of a target in a tumor’s microenvironment represents the most logical method for assigning patients to combination immune therapy clinical trials. We believe this approach may shape the future of precision immunotherapy.

References

1. Harris, S. J., Brown, J., Lopez, J. & Yap, T. A. Immuno-oncology combinations: raising the tail of the survival curve. Cancer Biol. Med. 13, 171–93 (2016).
2. Grady, D. A Sickened Body as Cancer Weapon: Harnessing the Power of the Immune System. The New York Times CLXV, (2016).
3. Chalmers, Z. R. et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 9, 34 (2017).
4. Le, D. T. et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science 357, 409–413 (2017).
5. Cortes-Ciriano, I., Lee, S., Park, W.-Y., Kim, T.-M. & Park, P. J. A molecular portrait of microsatellite instability across multiple cancers. Nat. Commun. 8, 15180 (2017).
6. Inoue, Y., Osman, M., Suda, T. & Sugimura, H. PD-L1 copy number gains: a predictive biomarker for PD-1/PD-L1 blockade therapy? Transl. Cancer Res. 5, S199–S202 (2016).
7. Guo, L. et al. PD-L1 expression and CD274 gene alteration in triple-negative breast cancer: implication for prognostic biomarker. Springerplus 5, 805 (2016).
8. Inoue, Y. et al. Clinical significance of PD‑L1 and PD‑L2 copy number gains in non‑small‑cell lung cancer. Oncotarget 7, (2016).
9. Ayers, M. et al. IFN-γ-related mRNA profile predicts clinical response to PD-1 blockade. J. Clin. Invest. 127, 2930–2940 (2017).
10. Kolata, G. A Cancer Conundrum: Too Many Drug Trials, Too Few Patients. The New York Times 1 (2017).

Guest Blog
by Joydeep Goswami, President, Clinical Next-Generation Sequencing and Oncology, Thermo Fisher Scientific

Joydeep Goswami

FDA’s recent decision to grant premarket approval of the first next-generation sequencing (NGS)-based companion diagnostic for non-small cell lung cancer (NSCLC) marks an important milestone in precision medicine. Patients and their oncologists now have access to an in vitro diagnostic (IVD) that can help expedite the selection of targeted therapies in days, a key advancement considering traditional testing methods can take several weeks for an answer — time that many NSCLC patients simply don’t have.

While this is great news for one segment of the more than 1.6 million estimated new cases of cancer that will be diagnosed in 2017,¹ what options are available for the rest of the patients battling this disease and for which there is no approved IVD? There has been a lot of debate about laboratory-developed tests (LDTs) versus IVD testing, particularly as to whether to stop the use of LDTs or to regulate them. In the fast-moving field of oncology, where the power of understanding the disease and the selection of precision treatments are being improved rapidly by technologies like NGS, clinicians and oncologists want to be able to incorporate these latest advancements into their clinical diagnostic practices.

While IVD tests undoubtedly are the gold standard in terms of demonstrating clinical validity and utility, developing an IVD can take several years. Patients and doctors dealing with fast moving and fatal diseases like cancer do not have the luxury of waiting until an IVD test is developed and approved in order to have access to a viable test. On the other hand, LDTs allow clinicians and pathologists to validate a diagnostic much faster under the appropriate controls and quality guidelines from the Clinical Laboratory Improvement Amendments (CLIA) and the College of American Pathologists (CAP).

Although LDTs do not meet the same gold standard achieved by FDA-approved IVD tests, they do provide patients and clinicians with a much-needed option and quicker access to the latest advances in science while an equivalent IVD test is being developed. LDTs thus speed the path and utilization of the latest biological discoveries and technological innovations for the benefit of patients. Banning them or over-regulating them could have a negative impact both on innovation and, more importantly, patient access.

But there may be a solution that would enable patients to reap the benefits provided by both testing approaches. In one scenario, FDA could mandate labs to switch from an LDT to an equivalent IVD once one is approved. Alternatively, considering the much more rigorous validation process that an IVD product has to undergo, the Centers for Medicare & Medicaid Services and FDA could work together to encourage their use by setting reimbursement levels for IVDs higher than those associated with LDTs. Such measures could more effectively balance patient safety concerns with the right to benefit from cutting-edge science immediately.

1. American Cancer Society. Cancer Facts & Figures. 2017. https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-figures-2017.html. Accessed July 11, 2017.

by David Davenport, Office Administrator, Personalized Medicine Coalition

J. Craig Venter, Ph.D., Founder, President, CEO, J. Craig Venter Institute, delivers the opening keynote address at the BIO International Convention’s Personalized Medicine & Diagnostics Track last month.

“Health care today is reactive and costly … anything but personalized … but we are now entering a new era where health care is becoming proactive, preventive, highly personalized and most importantly predictive,” said J. Craig Venter, Ph.D., Founder, President, CEO, J. Craig Venter Institute, during his opening keynote at the Personalized Medicine and Diagnostics Track at the 2017 BIO International Convention in San Diego from June 21 – 22. The track, co-organized by PMC, brought together thought leaders to discuss breakthroughs in advancing personalized medicine. From those conversations several themes emerged:

Complex genetic data require a “knowledge network” to translate into personalized care.

During the session titled The Next Frontier: Navigating Clinical Adoption of Personalized Medicine, moderated by PMC Vice President for Science Policy Daryl Pritchard, Ph.D., panelists discussed how to accelerate the clinical adoption of innovative personalized therapies. Jennifer Levin Carter, M.D., Founder and Chief Medical Officer of N-of-One, a clinical diagnostic testing interpretation service company, explained that as data grows in complexity, there is a growing need for partnerships to efficiently analyze the data and develop effective targeted treatment plans. India Hook-Barnard, Ph.D., Director of Research Strategy, Associate Director of Precision Medicine, University of California, San Francisco (UCSF), agreed and discussed the need to build a “knowledge network” that can harness data and expertise to inform provider-patient decision-making.

Discussing how personalized medicine can be integrated into community health centers lacking large research budgets, Lynn Dressler, Dr.P.H., Director of Personalized Medicine and Pharmacogenomics at Mission Health Systems, a rural community health care delivery system in Asheville, North Carolina, discussed the need to better educate physicians and patients as well as the role that a knowledge network could play in providing easy and cost-effective access to diagnostic testing services.

Delivering personalized medicine requires innovative partnerships involving industry, IT companies, providers, payers and the government.

During It’s a Converging World: Innovative Partnerships and Precision Medicine, a panel moderated by Kristin Pothier, Global Head of Life Sciences Strategy, Ernst & Young, discussed the need for “open data” where improved patient care is the shared goal, and how public-private partnerships that address education, evidence development and access to care can help foster personalized medicine.

During a session titled Nevada as a New Model for Population Health Study, Nevada-based health system Renown Health outlined a study in which it partnered with genetic testing company 23andMe to examine whether free access to genetic testing changes participants’ practices in managing their own health and facilitates the utilization of personalized medicine.

In the era of personalized medicine, measuring and delivering value requires a paradigm shift from population-based to individual-based evidence.

Following a discussion on regulatory and reimbursement challenges moderated by Bruce Quinn, M.D., Ph.D., Principal, Bruce Quinn Associates, during which panelists called for the simplification of payment structures to be more consistent, more efficient and more connected to the patient market, a panel moderated by Jennifer Snow, Director of Health Policy at Xcenda, discussed how value assessment frameworks must adapt to consider the value of personalized medicine. During The Whole Picture: Consideration of Personalized Medicine in Value Assessment Frameworks, panelist Mitch Higashi, Ph.D., Vice President, Health Economics and Outcomes Research, U.S., Bristol-Myers Squibb, called for patient-centered definitions of value and advocated for the inclusion of predictive biomarkers in all value frameworks. Donna Cryer, J.D., President, CEO, Global Liver Institute, added that the “patient must be the ultimate ‘arbiter of value’” and urged “transparency” in how value assessment frameworks are used.

Noting that different assessment frameworks have different goals, Roger Longman, CEO, Real Endpoints, called for more dynamic frameworks that allow different stakeholders to “use the same criteria but weigh them differently.” The panel concluded that to advance personalized medicine, value frameworks must be meaningful, practical and predictive for patients; reflect evolving evidence needs like real-world evidence; and consider breakthrough payment structures like bundled payments.

From Promise to Practice: The Way Forward for Personalized Medicine

During the concluding session, Creating a Universal Biomarker Program, moderated by Ian Wright, Owner, Strategic Innovations LLC, on behalf of Cedars-Sinai Precision Health, panelists discussed how to make patients the point of reference for their own care, as opposed to being compared to the “normal” range of population averages in treatment decisions using biomarkers. The speakers concluded that moving in that direction requires providers to establish baselines for each patient, along with tools and metrics to facilitate the approach.

In the words of Donna Cryer, “personalized medicine is the definition of value for a patient.” With the ability to detect diseases before they even express themselves, the promise of personalized medicine has never been greater.

However, changing the health care system to improve patient access to valuable personalized medicines requires innovation and collaboration. As PMC President Edward Abrahams, Ph.D., said during his opening remarks for the track, that change “doesn’t come easily,” but “breakthrough” discussions like these continue to move us forward.

The complete track agenda can be downloaded here.