Authors from Foundation Medicine explain the regulatory path that led to approval of FoundationOne CDx.
Foundation Medicine's FoundationOne CDx was the second product to pursue FDA/CMS dual review, paving the way for comprehensive genomic profiling in advanced cancer patients.Keeping up with the ongoing changes in oncology is becoming a difficult task for clinicians and payers. New relevant biomarkers and biomarker-driven treatments are introduced each year, and many more are in late-stage development. For example, 8 new biomarker-driven oncology treatments were approved in 2017 alone.1 Although many patients have benefited from this revolution in precision medicine by using comprehensive genomic profiling (CGP) of their tumors to help direct therapy, many others have missed this opportunity by receiving conventional testing or, worse, by failing to receive any molecular testing.
CGP refers to next-generation sequencing (NGS)—based testing of tumors that has been optimized to identify all types of cancer-relevant molecular alterations and complex genomic signatures in known cancer-related genes in a single test, using complex (often proprietary) bioinformatics. There has been substantial debate of the value of CGP in both the clinical oncology and managed care communities. Regardless, the demand for the technology exists among patients, providers, and biopharmaceutical companies alike.2
A key rationale for using CGP is the well-documented problem of low molecular testing rates and slow adoption rates for new biomarkers.3,4 Patients who receive incomplete or partial testing per guidelines may miss the opportunity to receive potentially life-extending therapies that, for some patients, have been shown to improve quality of life compared with cytotoxic chemotherapy. Alternatively, even if single-assay tests are performed, there is a high likelihood of having insufficient tissue or there being a need for repeated invasive biopsies.5 One study looking at diagnosis patterns in non—small cell lung cancer found that only 8% of patients received testing for all guideline-recommended biomarkers prior to therapy.4 Not receiving targeted therapy resulted in poorer outcomes. This is a significant area of opportunity for quality improvement in patients with advanced cancers. Unfortunately, the few established quality measures related to biomarker testing seem to be written in the reverse order, measuring if a certain test was used for patients who had received a certain drug.6 The more relevant measure might be to assess if a patient with cancer about to receive anti-oncologic therapy had received complete testing.
A CGP approach with FoundationOne CDx offers a potential solution for slow testing adoption rates and tissue exhaustion, and a 1-stop diagnostic to best leverage the rapidly changing treatment landscape. By using a platform that can accommodate additional biomarkers and companion diagnostics, FoundationOne CDx is well suited to keep pace with precision oncology. Testing patients with a CGP approach improves quality of care and offers the opportunity for patients to receive an evidence-based therapy or enroll into a clinical trial, which can be life extending but is often biomarker driven. FoundationOne CDx provides a comprehensive profile of 324 genes and is suitable for use in all solid tumors.7 It encom- passes guideline-recommended genes for testing in solid tumors and has FDA approval as a companion diagnostic for 17 targeted therapies in 5 solid tumor types. In addition, FoundationOne CDx assesses complex genomic signatures to help inform immunotherapy decisions. A comprehensive genomic profile is essential for quantifying these genomic signatures.8,9 FoundationOne CDx is an efficient way of navigating all of these established and emerging biomarkers, potentially limiting the amount of tissue and time needed to assess for multiple clinically relevant alterations compared with sequential single-gene testing.
FDA's Risk-Based Regulation of Diagnostics
Foundation Medicine, Inc, was founded on the premise that no mechanism existed whereby clinicians, in the day-to-day practice of clinical oncology, could systematically obtain tumor information and efficiently interpret the specific molecular alterations associated with each patient’s disease.10 This mission was a driving force in its decision to participate in the parallel review process after being approached by the FDA and CMS.Laboratory developed tests (LDTs) traditionally have not been required to receive approval by the FDA prior to being used. Additionally, the FDA process for medical devices (including LDTs) is somewhat less well defined and slightly more complex compared with the well-established regulatory path for small molecules and biologic therapies. The 1976 Medical Device Amendment provided a comprehensive regulatory framework for medical devices and established a risk-based regulatory classification system, as described in Figure 1.11 Diagnostics are medical devices and are classified based on the risk posed to the patient using the device. The lowest risk devices are deemed Class I and are mostly exempt from any requirements prior to marketing within the United States. Examples of Class I devices include arm slings, latex examina- tion gloves, and most hearing aids.
Premarket notification or 510(k)
Most medical devices and diagnostics fall in the Class II category, and manufacturers of such devices are required to notify the FDA prior to marketing those devices via a 510(k) submission (premarket notification [PMN]).12 The 510(k) application allows the manufacturer to demonstrate that its device is “substantially equivalent” in terms of its intended use, safety, and effectiveness to an already legally marketed “predicate” medical device in the United States.
Class III is reserved for devices deemed high risk and subject to a premarket approval (PMA) procedure, like that for new drugs. By statute, the PMA process is reserved for medical devices that “support or sustain human life, are of substantial importance in preventing impairment of human health, or which present a potential, unreasonable risk of illness or injury.”13 For this reason, almost all companion diagnostics that may direct treatment are also categorized as Class III and require FDA approval, based on clinical experience, before a product can be marketed.
Path to Market for LDTs
Medicare Coverage Determination
FDA/CMS Parallel Review Process
Pre-market approval is the most involved and expensive process that a medical device manufacturer typically pursues. This type of approval is based on a determination by the FDA that the applicant has submitted sufficient valid scientific evidence to assure that the device is safe and effective for its intended use(s). For example, Figure 2 shows the types of evidence submitted by Foundation Medicine for FoundationOne CDx.Rigorous FDA evaluations, as shown in Figure 2, are in stark contrast to the path to market for LDTs, although many valuable diagnostic assays have come to market via the LDT pathway. To date, the FDA has exercised a policy of enforce- ment discretion wherein LDTs have not been required to seek authorization in advance of being clinically offered as a test. LDTs are regulated by CMS’ Clinical Laboratory Improvement Amendments (CLIA), which is administered through various accreditation bodies, including the College of American Pathologists. Additionally, various state agencies, including the New York State Department of Health, have their own requirements for laboratories accepting patient samples originating in their states. Although CLIA establishes quality standards for laboratories to ensure the accuracy, reliability, and timeliness of patients’ test results, they do not cover how to perform a pre-market review of analytic validation or require clinical validity data. This contrasts with the FDA review process for Class III PMAs where both analytic and clinical validity are evaluated. Moreover, whereas medical devices and diagnostics must register with the FDA and are also required to submit adverse events, recalls, and user complaints, LDTs have no such requirement. This can lead to lack of awareness of potential safety signals with a given LDT.To be covered by CMS, medical products must fall into one of the statutorily defined “benefit categories” and be “reasonable and necessary” for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member. To meet the “reasonable and necessary” standard, a product or service must improve health outcomes, be safe and effective, and not be deemed experimental or investigational. The reasonable and necessary provisions are not defined explicitly in regulation and remain at the discretion of Medicare. For Medicare, FDA-approved devices with therapeutic indica- tions are presumed to meet this definition unless directly addressed through a Local Coverage Determination (LCD) or National Coverage Determination (NCD). Public payers are often subject to requirements to develop their coverage policies in an open and transparent manner. For example, LCDs and NCDs must undergo opportunities for public comment and are open to more political scrutiny given the nature of publicly funded programs.14The FDA and CMS have clearly different objectives: safety and effectiveness of a device and whether the device is reasonable and necessary, respectively. Therefore, neither agency is usually influenced by the other’s findings or decisions, nor has there been coordination between the individual review processes. To support medical device innovation, the FDA and CMS entered a memorandum of understanding, which led to a pilot FDA/CMS parallel review process.15 The purpose of this program was mainly to minimize the time between regulatory authorization and reimbursement, an important barrier to patient access for the latest medical device innovation.16 Creating accelerated approval processes or breakthrough device designations would be insufficient if there was no comparable innovation in reimbursement policies. This is particularly unique to the medical device/diagnostic space in contrast to the oncology therapeutic space where FDA approval of a drug or biologic is sufficient for reimbursement and therefore access by the patient. Medical device/diagnostics must pursue reimbursement separately via an LCD or NCD following FDA review.17 The pilot project has since been made permanent.18
Careful evaluation of the National Coverage Analysis process and the resulting NCD can provide valuable insight. CMS assessed CGP technology through the parallel review process using the ACCE (analytical validity, clinical validity, clinical utility, and ethical, legal and social implications of genetic testing) model for assessment. For analytic validity and clinical validity, FDA approval of FoundationOne CDx signifies assurance of safety and effectiveness and that there is an acceptable evidence base for analytic and clinical validity requirements for coverage.19,20 A key area of focus for CMS in the parallel review was to evaluate clinical utility.20,21 When these processes are undertaken in parallel, the product’s complete clinical value can be assessed (Figure 3).
Paving The Way in Advanced Cancer: FoundationOne CDx
Through the parallel review process for FoundationOne CDx, CMS identified and reviewed over 250 studies to assess the clinical utility of NGS testing in advanced cancer, demonstrating the breadth of available evidence. The clinical outcome measures CMS considered included overall survival, progression-free survival, partial response, complete response, stable disease, time to progression, overall response rate, and time to treatment failure. Systematic evidence reviews and meta-analyses of clinical trial data provide a strong level of evidence, and observational and other study designs provide additional supportive evidence across a broad population of patients with cancer.20,22 Meta-analyses of data from therapeutic clinical trials, including randomized controlled trials, which have traditionally been the gold stan- dard in evidence generation; data from nonran- domized studies utilizing CGP; and real-world evidence support the use of a genomically guided treatment approach to improve outcomes.22-27 From this evidence review, CMS concluded that there is sufficient clinical utility demonstrating that FDA-approved and FDA-cleared laboratory in vitro companion diagnostic tests using NGS improve health outcomes for patients with advanced cancer when used by the treating physician and patient to guide selection of proven treatments.20 In addition, in the final NCD, CMS encouraged the community of stakeholders to continue to develop and publish evidence related to the meaningful endpoints assessed in the coverage analysis. Foundation Medicine is committed to publishing evidence as it is generated, as evidenced by more than 300 Foundation Medicine—authored publications.The parallel review process has served as an incentive for 2 LDTs to pursue regulatory approval: Exact Sciences’ Cologuard for colorectal cancer screening and, more recently, FoundationOne CDx.20,28 These have been characterized as “single- site PMAs” where both diagnostic tests are offered from a single laboratory. The NCD for NGS-based diagnostics provides confirmed national coverage for FDA-approved or -cleared diagnostics, while LDTs without FDA approval or clearance may still be reimbursed at the discretion of the local Medicare Administrative Contractors.
Translating Parallel Review to Commercial Payers
The parallel review process culminated in FDA approval of FoundationOne CDx and a CMS NCD that provided Medicare beneficiaries with advanced cancer broad access to CGP as a path to precision oncology treatment, as outlined by the coverage criteria in Table.20Although CMS coverage of NGS tests in advanced cancer does not automatically translate into commercial coverage, as reported in a recent survey of some major commercial payers,29 it would seem prudent for commercial payers to seize this moment and re-evaluate the role of CGP in providing important and tangible benefits for the sickest cancer patients regardless of age or insurer. The parallel review process provides the oppor- tunity to recognize and integrate the outcomes of the rigorous FDA regulatory and CMS coverage pathways in order to further pave the path forward for higher quality oncology care.
Commercial payers may be challenged to rethink the approach to coverage determination for non—Medicare Advantage members. With some exceptions, commercial policies have historically described CGP tests as investigational or experimental. These policies often cite lack of clinical utility as a reason for noncoverage; however, there are several considerations for determining clinical utility for genomic panel tests in oncology that differ from single gene testing, including:
To further validate this, a CGP approach has been integrated into many large innovative clinical trials, such as the American Society of Clinical Oncology’s TAPUR and Genentech’s MyPathways; into numerous clinical guidelines; and is used routinely across academic centers and advanced community practices. The evidence base supporting the clinical utility of this approach stems from data across multiple types of studies conducted in the oncology setting using biomarker-driven therapies. Collectively, these studies demonstrate that:
Thus, genomic testing has a significant role in fulfilling the promise of precision medicine in oncology to improve outcomes for patients, and the NCD is a landmark event for patients with advanced cancer.
Future of the Parallel Review Process in Precision Oncology
When it comes to broad commercial coverage, private payers are often faced with the challenge of addressing or creating a significant disparity in oncology care between patients with Medicare and other commercially covered patients with advanced cancers. Because the analytic validity, clinical validity, and clinical utility evidence reviewed by the FDA and CMS are not limited to patients ≥ 65 years, this same evidence base should be considered by commercial insurers. Cancer is not a disease affecting only the Medicare population. US epidemiology data (Figure 430) estimate the median age at cancer diagnosis to be 66 years, yet 46.8% of cancer diagnoses and 30.7% of cancer deaths occur before age 65.30 In addition, the proportion of cancer cases that are reported as advanced stage (have spread locally or distant) for the Medicare and non-Medicare population are nearly equal (45.3% for <65 years vs and 45% for >65 years). Coverage for commercial lives is imperative so that patients with advanced cancer are provided the greatest opportunity to receive genomically matched or biomarker-driven therapies regardless of age or insurer.Now that CMS has released the NCD that addresses all NGS-based oncology assays used in advanced cancer, and with the subsequent announcements by multiple CGP diagnostics providers, it is expected that more laboratory tests will undergo the FDA approval process via the PMA pathway.31 Although commercial payers are only directly affected by a new NCD if they offer Medicare Advantage plans, the data supporting the NCD will offer the chance for payers to consider the same information used by CMS to ultimately determine for themselves that outcomes are improved by CGP utilization. If CGP is only accessible to the Medicare population, there is a risk of creating additional disparity in the US healthcare ecosystem. Given the numerous new biomarkers, novel biomarker-driven treatments, tumor-agnostic biomarkers (ie, microsatellite instability), and genomic signatures that require a large panel test, there is no doubt that CGP is not only here to stay but will become mainstream.
In many cases, testing is treated passively by payers (eg, by only asking if a patient was tested if a biomarker-driven therapy is requested). However, CGP offers an opportunity for payers to proactively require that all appropriate patients undergo guideline-driven testing to improve quality of care and decrease waste by assuring the right patients receive the right treatment at the right time. Payers who embrace CGP could be afforded the opportunity to not only improve quality of care but also to better manage and predict cancer drug spend. It would not be hard to imagine a payer taking it one step further and requiring CGP for certain patients, which undoubtedly would increase the utilization of biomarker-driven treatments. The predictable increase in use of biomarker-driven treatments, for both on-label and off-label use, could allow for improved value-based agreements based on the volume of patients receiving CGP, thus off setting possible cost increases for payers while improving quality of care for patients.
Because FoundationOne CDx completed the parallel review process, the path to CMS coverage has been paved for future NGS tests. The broad nature of the NCD means other NGS test providers will only need to pursue companion diagnostic FDA approval or clearance status to obtain coverage from CMS. As outlined in the CMS NCD, the FDA assesses analytic validity and clinical validity as part of the approval process. FDA approval or clearance, therefore, assures market access to high-quality CGP assays. While the CMS NCD ensures access to these approved assays to qualifying Medicare beneficiaries with advanced cancer, the CGP access created by this parallel review will undoubtedly lead to more patients with advanced cancers receiving life-extending precision medicine based therapies. It is fully expected that the understanding of the clinical utility of CGP will continue to evolve, the evidence will continue to grow, and the parallel review process will provide a sound starting place for reasonable assessment of clinical utility for other technologies in this unique space. Lastly, commercial payers will have the opportunity to move from passive to active management of biomarker-driven therapies in a way they have not in the past. Ultimately these decisions will catalyze the paradigm shift for the treatment of advanced cancers as a disease of the genome at all levels of patient care, from the lab to the provider to the payer. If value is defined in part by the therapy’s efficacy, it is hard to see how we can achieve value in oncology without CGP.Author Information
All authors are affiliated with Foundation Medicine, Inc, Cambridge, Massachusetts. Lakshman Ramamurthy, 􏰌PhD􏰃, is the global regulatory lead;􏰂 K􏰙risti Ma􏰊well, MS, CGC, is a certified genetic counselor􏰂; Bethany Sawchyn, P􏰌harmD􏰃, is associate director for P􏰌ayer 􏰌Policy and Health Outcomes;􏰂 and Rachel Anhorn, 􏰌PharmD􏰃, is the director, 􏰌Payer 􏰌Policy and Health Outcomes.References: