Molecular diagnostic tests are increasingly being used to guide care for patients with cancer, as the understanding of how different biomarkers contribute to tumor growth improves, technologies allowing for the simultaneous evaluation of multiple tumor alterations become available, and a growing number of transformative biomarker-targeted therapies are developed.1 These targeted therapies have proved beneficial to many patients with cancer, and in certain cancer indications, including EGFR mutation–positive advanced non–small-cell lung cancer (NSCLC), these therapies have led to notable improvements in patient outcomes, such as 5-year survival rates and progression-free survival.2
Recognizing the growing evidence that supports the use of diagnostic tests to identify appropriate therapies for patients with advanced NSCLC, leading organizations, such as the National Comprehensive Cancer Network (NCCN), the American Society of Clinical Oncology, the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology, have established clinical guidelines that universally recommend diagnostic testing for most patients with advanced or metastatic NSCLC.3-5 These guidelines are intended for use by oncologists in the diagnosis and treatment decision-making processes.3-5
In addition, payers such as the Centers for Medicare & Medicaid Services (CMS), the largest payer in the US healthcare system, have recognized the importance of molecular diagnostic tests to their patient populations. In March 2018, CMS released a coverage determination in which US Food and Drug Administration (FDA)-approved or -cleared diagnostic tests that use next-generation sequencing technologies were deemed reasonable and necessary for patients with recurrent, relapsed, refractory, metastatic, or advanced stage III or IV cancer.6 This decision ensured consistent coverage policy of FDA-approved and FDA-cleared diagnostic tests for Medicare patients at the national level. Recent data also suggest that these multigene panels may be more cost-effective than single-marker genetic tests.7
Diagnostic test results must be accurate, reliable, and clinically meaningful for patients to reap the benefits of precision medicine and biomarker-targeted therapies. Currently, the performance metrics of diagnostics are evaluated under a bifurcated regulatory system. The FDA was given authority to regulate all in vitro diagnostics as medical devices under the 1976 Medical Device Amendments to the Federal Food, Drug, and Cosmetic Act.8 However, the agency has generally exercised discretion and has not enforced the device provisions of this legislation consistently on the subset of in vitro diagnostics (referred to as laboratory-developed tests); this is because, historically, laboratory-developed tests used relatively simple technology and have had limited availability.8,9
Instead, laboratory-developed tests have been regulated under the Clinical Laboratory Improvement Amendments (CLIA) program, which is overseen by CMS.10 In contrast to rigorous FDA review, which requires demonstrations of analytical and clinical validity, the CLIA certification primarily assesses a laboratory’s ability to properly conduct tests through protocol adherence and personnel qualification, but typically not the performance metrics of the test itself.10
Consequently, as the practice of medicine has evolved, and diagnostic tests have become increasingly complex and vital to clinical care, stakeholders such as patient advocacy organizations, congressional legislators, and the FDA have raised questions about the lack of standardized performance metrics for FDA-approved tests and for laboratory-developed tests, emphasizing the risk that this lack of standardization poses to patients.11
In fact, after identifying multiple problems with the performance of several high-risk laboratory-developed tests in 2010, the FDA announced its intent to reconsider its policy of enforcement discretion and took several subsequent actions. These actions included the development of draft guidance that outlined an approach to laboratory-developed test oversight in 2014, the publication of a report with 20 case studies that documented the public health impact of problematic laboratory-developed tests in 2015, and the release of a discussion paper on such tests in 2017.9,11,12
In December 2018, congressional leaders working with the FDA and other stakeholders, including clinical laboratories, diagnostic test manufacturers, trade associations, and patient advocacy groups, released draft legislation aimed at establishing a uniform regulatory framework for all diagnostic tests under the authority of the FDA, with the intent of reducing the lack of standardization and inconsistent regulatory requirements for FDA-approved tests and laboratory-developed tests.13
This draft legislation was updated to incorporate extensive comments from the FDA and the broader community and was introduced to the US House of Representatives in early 2020, but its legislative path forward remains unclear.14
As part of our 2015 national survey of oncologists who managed patients with advanced NSCLC, we explored the use of FDA-approved diagnostic tests and laboratory-developed tests.15 The findings from this original survey indicated that most patients with advanced NSCLC received testing for EGFR mutations and ALK rearrangements, with testing for EGFR more frequently performed in privately owned, academic, and community-based treatment settings using laboratory-developed tests, and testing for ALK evenly split between laboratory-developed tests and FDA-approved diagnostics across these settings. This utilization of different tests within the same treatment setting raised concerns that an unknown degree of variability could exist between tests with the same intended use.15
In the period since this original survey was conducted, tests have increased in complexity and have become more widely available, policy discussions have evolved, and major coverage determinations have been made; therefore, a new landscape analysis is warranted.
Our current study updates the original survey data15 by investigating the use of FDA-approved tests and laboratory-developed tests based on a 2019 national survey of oncologists who managed patients with advanced NSCLC. In addition, this study includes survey questions regarding the factors that influence the diagnostic test ordering and treatment decision-making practices of physicians who manage patients with advanced NSCLC under current regulatory and coverage frameworks.
The study sample was based on a national panel of oncologists maintained by M3 Global Research. This panel includes more than 1000 physicians and is broadly representative of all oncologists in the United States across the demographic dimensions of region and years in medical practice. Study invitations were sent to all oncologists within the M3 panel. Respondents qualified for this study if they managed at least 5 patients with advanced NSCLC per month and had personally diagnosed at least 1 patient with advanced NSCLC in the 12-month period before the survey fielding (March 2018-February 2019).
The study consisted of 2 parts: a short survey and a patient chart review. A total of 150 physicians completed the study, translating to a response rate of 14%. The participants were offered an industry-standard honorarium as compensation for their time to complete the study. The study was administered online and was fielded from March 1, 2019, to March 25, 2019.
A data collection instrument was developed to capture deidentified information on patients with stage IV NSCLC in the United States for use in the chart review portion of the study. Based on a similar instrument fielded in 2015, we made updates to reflect new treatment and diagnostic test options that were available at the time of the study.
Before completing the survey, responding physicians were instructed to choose between 4 and 8 patients with stage IV NSCLC from their list of active patient charts. To facilitate the selection of random charts, oncologists were instructed to identify patients based on the assignment of random letters to correspond with the first letter of the patients’ last names. The patient charts were required to have been active within the practice during the past 12 months.
Patient information, including age, weight, sex, ethnic origin, concomitant conditions, insurance type, smoking status, diagnosis year, genetic testing information, and all treatment lines, was recorded by physicians using the data collection instrument for each randomly selected patient chart. A total of 815 patient charts were included in the study, which reflected the practices of 150 responding oncologists who managed patients with advanced NSCLC.
All survey data, including the identities of the responding physicians, were analyzed in aggregate and were completely anonymized. When the specific genetic test types were unknown to the responding physician, consent was obtained to contact the affiliated hospital pathology laboratory. Follow-up phone calls were made in these instances to determine which genetic testing platform or external testing services a specific hospital uses. The data were analyzed across all patients, as well as across the histologic subtypes of advanced NSCLC.
Key patient demographics, such as patient insurance type and practice setting, were also analyzed for patterns in genetic testing. Patients were included in the analysis of the use of FDA-approved and laboratory-developed tests if they were diagnosed with advanced NSCLC after the first FDA approval of a diagnostic test for a given NSCLC-related mutation, and if their test type could be determined. Patients who were diagnosed with advanced NSCLC before the first FDA approval of a diagnostic test for a given NSCLC-related mutation and whose test type could not be determined were excluded from this analysis.
The analysis of insurance type was limited to patients with private insurance or with Medicare coverage. Patients with Medicaid, military insurance, self-insurance, no insurance, or unknown insurance types were excluded because of low sample sizes or because of the possibility of inherent confounding variables in these populations.
All subgroup differences in proportions were tested using a chi-square analysis. Post-hoc pairwise comparisons used a Bonferroni correction to reduce the risk for type 1 errors. The pairwise comparisons were tested for significance at the .05 level. All statistical analysis was performed using IBM SPSS Statistics Version 20.0 (IBM; Armonk, NY).
The study physician and patient characteristics are presented in Table 1. Our sample of physicians was broadly representative of oncology physicians across factors of interest such as years in practice and geographic region. In addition, the patient population included in our chart review was representative of the wider population of patients with advanced NSCLC across factors of interest such as age and histologic subtype.16
Biomarker testing rates among patients selected in the chart review portion of our study were examined. Of the 815 patients with advanced NSCLC in this study, 812 (99.6%) patients were tested for at least 1 mutation. A total of 669 (82%) patients were tested for EGFR mutations, 586 (72%) for ALK rearrangements, 298 (37%) for BRAF V600E mutations, and 380 (47%) for ROS1 rearrangements. Other biomarkers were tested at lower frequencies. Of the 601 patients who were not classified as having squamous-cell carcinoma, 203 (34%) were tested for all 4 of these mutations.
The use of FDA-approved diagnostic tests for EGFR mutations, ALK rearrangements, BRAF V600E mutations, and ROS1 rearrangements across treatment setting and insurance type is shown in Table 2. The differences in the use of FDA-approved diagnostic tests across treatment settings were significant (P <.05), indicating that the setting in which a patient receives treatment may influence whether he or she receives an FDA-approved test or a laboratory-developed test. Specifically, the difference in the use of FDA-approved tests for EGFR mutations between the academic (65%) and private (82%) practice settings was determined to be significant (P <.05).
The differences in the use of FDA-approved tests for BRAF V600E mutations between the academic (44%) and private (79%) practice settings were also significant (P <.05). The differences in the use of FDA-approved tests for ROS1 rearrangements between the community-based (18%) and private (44%) practice settings were significant as well (P <.05). No significant differences were observed between patients with private insurance and patients with Medicare coverage.
The receipt of an appropriate targeted therapy among patients who tested positive for an actionable biomarker was evaluated for EGFR mutations and ALK rearrangements across the treatment settings and insurance types. Therapies were deemed “appropriate” in this analysis if they were FDA-approved for advanced NSCLC, and if they targeted the biomarker(s) for which a patient tested positive.
Of the 669 patients who were tested for EGFR mutations, 223 (33%) tested positive for that mutation. In all, 185 (83%) of these patients received a targeted therapy indicated for patients with EGFR-positive advanced NSCLC. Similarly, of the 586 patients who were tested for ALK rearrangements, 72 (12%) tested positive for that mutation. A total of 60 (83%) of these patients received a targeted therapy indicated for patients with ALK-positive advanced NSCLC.
The differences in the use of an appropriate targeted therapy for patients testing positive for EGFR mutations between the academic (92%) and community-based (75%) practice settings were determined to be significant (P <.05). The difference in the use of an appropriate targeted therapy for patients testing positive for EGFR mutations between the academic (92%) and private (77%) practice settings was also significant (P <.05).
These significant differences indicate that a patient’s treatment setting may influence whether he or she receives a targeted treatment. The use of an appropriate targeted therapy for patients testing positive for ALK rearrangement was significant overall (P <.05), but there were no significant differences in the pair-wise comparisons. No significant differences were observed between patients with private insurance and those with Medicare coverage.
A total of 66 (44%) physicians who responded to the survey indicated that the availability of targeted therapies influences their decision to order diagnostic tests for patients with advanced NSCLC. In all, 98 (65%) physicians indicated that the patient’s type of insurance coverage factors into their decision to order diagnostic tests for patients with advanced NSCLC (Figure Part A).
A total of 107 (71%) respondents indicated that they were aware of the CMS next-generation sequencing coverage determination (Figure Part B), and 74 (69%) of those 107 physicians indicated that this determination increased the frequency of diagnostic tests that they order for patients with advanced NSCLC (Figure Part C). In all, 149 (99%) physicians indicated that they are confident that diagnostic tests yield high-quality data about their patients with advanced NSCLC.
A total of 69 physicians (45%) responded that in a case when a diagnostic test indicated a specific therapy for a patient with advanced NSCLC, cost issues or the patient’s insurance coverage could influence them not to prescribe the indicated therapy (Figure Part D). The full physician questionnaire can be found in the Appendix. Selected questionnaire responses are shown in the Figure.
Given the increasing importance of molecular diagnostic tests and targeted therapies in the diagnosis and treatment of patients with cancer, we examined the diagnostic test ordering and treatment decision-making practices of physicians who are managing patients with advanced NSCLC under the current federal regulatory and healthcare coverage frameworks. Our analysis of 150 physician survey responses and 815 patient records produced 3 key findings.
First, our findings indicate that diagnostic testing has become routine in the diagnosis of and treatment decision-making for patients with advanced NSCLC. The survey respondents reported that almost all patients included in the chart review received at least 1 diagnostic test, which is consistent with previous findings in the literature that indicate high rates of diagnostic testing in patients with lung cancer.17-20
In line with evidence-based clinical practice guidelines issued by the NCCN,5 we observed that most patients were tested for EGFR mutations and ALK rearrangements, and that the testing rates differed across the histologic subtypes. Although most patients were tested for these individual genetic markers, few patients received the NCCN-recommended comprehensive biomarker testing for their subtypes,5 indicating that barriers to comprehensive biomarker testing may still exist.
In addition, our survey highlights that frameworks for the appropriate insurance coverage of diagnostic tests and targeted therapies are important to the access of care for patients with advanced NSCLC. The survey respondents indicated that insurance coverage influences their decision to order diagnostic tests for patients with advanced NSCLC. Furthermore, insurance coverage and cost were the most frequently reported factors that influenced a physician not to prescribe an indicated therapy.
Although we did not observe significant differences in the ordering of diagnostic tests and in the prescription of targeted therapies across the insurance types it is likely because, as a result of sample size concerns, our analyses were limited to patients with private insurance and Medicare, and the value of diagnostic tests and targeted therapies is widely acknowledged by these payers. Further examination of the impact of insurance type on a patient’s access to diagnostic tests and targeted therapies is warranted, and if inequities in access to care are observed, intervention may be required at the societal and governmental levels.
Finally, we observed a shift toward the use of FDA-approved diagnostic tests since the publication of our 2015 survey,15 but the magnitude of this shift differed across the treatment settings. In our 2015 survey, only 13% of patients tested for EGFR and 51% of patients tested for ALK received FDA-approved diagnostic tests,15 compared with 73% and 70%, respectively, in our current survey (Table 2). This shift may indicate that physicians and, ultimately, pathologists value FDA regulation as well as newly approved technologies, such as next-generation sequencing panels, which were approved for use in this population after 2015.
The differences in the magnitude of this shift across the treatment settings may reflect, in part, that physicians in the academic setting preferentially use tests developed by their individual institutions. Our results also suggest that the overall increase in the use of FDA-approved tests may be explained by the increasing number of diagnostic assays approved by the FDA (eg, 2 tests approved in 2015 for EGFR vs 4 in 2019; 1 test approved in 2015 for ALK vs 3 in 2019) and the length of time since the initial FDA approval of a diagnostic test for a given mutation (eg, the first EGFR mutation test was approved in 2013; in 2015, that test had been on the market for 2 years vs 6 years in 2019).
Policy Implications and Recommendations
Based on the results of this 2-part study consisting of a survey and chart review, we recommend 2 policy changes to ensure patient access to high-quality, well-validated diagnostic tests and to their indicated biomarker-targeted therapies.
First, we recommend that as legislators and other stakeholders continue to work toward improving the current regulatory system for the benefit of patients, they prioritize the development of a predictable regulatory framework that fosters and encourages innovation while maintaining uniform oversight. Although we observed a shift toward the use of FDA-approved tests in our survey, a significant number of patients with advanced NSCLC still received molecular assessments that are subject to regulatory requirements different from those pursuing FDA premarket review.
However, this study did not seek to address the relative quality of laboratory-developed tests and FDA-approved diagnostic tests, and further research on potential variability in performance metrics and comparative outcomes is warranted. An improved framework should maintain the FDA’s standards for analytical and clinical validity but not impose an excessive burden on stakeholders involved in diagnostic test innovation, such as academic laboratories that have voiced concerns that seeking FDA approval would be onerous and expensive.
Second, we recommend that as payers consider future coverage decisions, in the absence of a uniform regulatory framework, they develop a minimum set of performance characteristics necessary to support determinations of coverage for diagnostic tests. Because diagnostic test results are frequently used in treatment decision-making processes, it is critical that accurate results are produced to prevent patients from being exposed to nonefficacious treatments and the unnecessary toxicities that would result from inappropriate identification of candidates for treatment.
As our survey demonstrated, physician decision-making is affected by insurance coverage, and the latest shifts toward the use of FDA-approved diagnostic tests coincided with the recent CMS coverage determination, which indicates that coverage frameworks may have the potential to alter physicians’ prescribing patterns. Furthermore, the continued coverage of biomarker-targeted therapies and mechanisms for rapidly incorporating new diagnostic test and drug approvals into coverage frameworks are essential to preserve patient access.21
This study has several limitations. First, this survey focused on oncologists, not pathologists. Although pathologists may have access to more diagnostic test–related information, oncologists were more appropriate for this study because our goals were to evaluate the use of diagnostic tests for patients with advanced NSCLC and to identify factors influencing physician decision-making under existing regulatory and coverage frameworks.
In addition, our study was not designed to address the comparative outcomes of patients who were tested with laboratory-developed tests versus FDA-approved tests.
Furthermore, as with most surveys, the potential for response bias and for nonresponding physicians bias exists.
The potential impact of confounders, such as a patient’s inability to receive a diagnostic test because of insufficient tissue, is unknown.
Finally, a portion of the patient records (and associated pathology reports) did not include information on the type of test used to detect lung cancer mutations, even after follow-up phone calls, and had to be excluded from further analysis, including 295 (44%) patients who were tested for EGFR mutations, 272 (46%) patients who were tested for ALK rearrangements, 169 (57%) patients tested for BRAF V600E mutation, and 217 (57%) patients tested for ROS1 rearrangements.
Molecular diagnostic tests and biomarker targeted therapies are routinely used in oncology care and will continue to drive the concept of precision medicine forward. Our study presents novel survey and chart review data that illustrate the routine use of diagnostic tests in the treatment of patients with advanced NSCLC and demonstrate an increase in the use of FDA-approved diagnostic tests. We also identify factors, such as insurance coverage and cost, that influence physicians’ diagnosis and treatment decision-making processes. Together, these findings illustrate that optimized regulatory and coverage frameworks are critical to an oncology patient’s access to care.
Author Disclosure Statement
Ms Wempe, Dr Stewart, Dr Glass, Dr Lasiter, Dr Vega, Dr Allen, Dr Sigal, and Ms Ramamurthy have no conflicts of interest to report.
- Berberabe T. With biomarker expansion, combinations and personalized medicine continue to rise. Target Ther Oncol. July 5, 2019. www.targetedonc.com/publications/targeted-therapy-news/2019/July-2019/with-biomarker-expansion-combinations-and-personalized-medicine-continue-to-rise. Accessed September 6, 2019.
- Lin JJ, Cardarella S, Lydon CA, et al. Five-year survival in EGFR-mutant metastatic lung adenocarcinoma treated with EGFR-TKIs. J Thorac Oncol. 2016;11:556-565.
- Kalemkerian GP, Narula N, Kennedy EB, et al. Molecular testing guideline for the selection of patients with lung cancer for treatment with targeted tyrosine kinase inhibitors: American Society of Clinical Oncology endorsement of the College of American Pathologists/International Association for the Study of Lung Cancer/Association for Molecular Pathology clinical practice guideline update. J Clin Oncol. 2018;36:911-919.
- Lindeman NI, Cagle PT, Aisner DL, et al. Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology. Arch Pathol Lab Med. 2018;142:321-346.
- National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines): Non-Small Cell Lung Cancer: NCCN Evidence Blocks. Version 7.2019. August 30, 2019. www.nccn.org/professionals/physician_gls/pdf/nscl_blocks.pdf. Accessed September 6, 2019.
- Centers for Medicare & Medicaid Services. Decision memo for next generation sequencing (NGS) for Medicare beneficiaries with advanced cancer (CAG-00450N). March 16, 2018. www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?NCAId=290. Accessed September 6, 2019.
- Steuten L, Goulart B, Meropol NJ, et al. Cost effectiveness of multigene panel sequencing for patients with advanced non–small-cell lung cancer. JCO Clin Cancer Inform. 2019;3:1-10. doi.org/10.1200/CCI.19.00002.
- Federal Food, Drug, and Cosmetic Act, 21 USC §321 (2013). www.govinfo.gov/content/pkg/USCODE-2013-title21/html/USCODE-2013-title21-chap9-subchapII-sec321.htm. Accessed September 6, 2019.
- US Food and Drug Administration. The public health evidence for FDA oversight of laboratory developed tests: 20 case studies. November 16, 2015. http://wayback.archive-it.org/7993/20171115144712/https://www.fda.gov/downloads/AboutFDA/ReportsManualsForms/Reports/UCM472777.pdf. Accessed September 6, 2019.
- The Clinical Laboratory Improvement Amendments. J Natl Cancer Inst. 2004;96:501. doi.org/10.1093/jnci/96.7.501.
- US Food and Drug Administration. Discussion paper on laboratory developed tests (LDTs). January 13, 2017. www.fda.gov/media/102367/download. Accessed September 6, 2019.
- US Food and Drug Administration. Draft guidance for industry, Food and Drug Administration staff, and clinical laboratories: framework for regulatory oversight of laboratory developed tests (LDTs). October 3, 2014. www.fda.gov/media/89841/download. Accessed September 6, 2019.
- Verifying Accurate Leading-edge IVCT Development Act of 2018. Preliminary discussion draft. December 6, 2018. https://degette.house.gov/sites/degette.house.gov/files/valid_act_discussion_draft_12.6.18.pdf. Accessed September 6, 2019.
- Verifying Accurate Leading-edge IVCT Development (VALID) Act of 2020. H.R. 6102. 116th Congress. March 5, 2020. www.congress.gov/116/bills/hr6102/BILLS-116hr6102ih.pdf. Accessed May 21, 2020.
- Audibert CM, Shea MB, Glass DJ, et al. Use of FDA-approved and laboratory-developed tests in advanced non–small cell lung cancer: results of a retrospective market analysis. Personalized Med Onc. 2016;5(7):278-284.
- Stewart M, Norden AD, Dreyer N, et al. An exploratory analysis of real-world end points for assessing outcomes among immunotherapy-treated patients with advanced non–small-cell lung cancer. JCO Clin Cancer Inform. 2019;3:1-15. doi.org/10.1200/CCI.18.00155.
- Mason C, Ellis PG, Lokay K, et al. Patterns of biomarker testing rates and appropriate use of targeted therapy in the first-line, metastatic non-small cell lung cancer treatment setting. J Clin Pathw. 2018;4:49-54.
- Lim C, Tsao MS, Le LW, et al. Biomarker testing and time to treatment decision in patients with advanced nonsmall-cell lung cancer. Ann Oncol. 2015;26:1415-1421.
- Schink JC, Trosman JR, Weldon CB, et al. Biomarker testing for breast, lung, and gastroesophageal cancers at NCI designated cancer centers. J Natl Cancer Inst. 2014;106:dju256. doi.org/10.1093/jnci/dju256.
- Lee DH, Tsao MS, Kambartel KO, et al. Molecular testing and treatment patterns for patients with advanced non-small cell lung cancer: PIvOTAL observational study. PLoS One. 2018;13:e0202865. doi.org/10.1371/journal.pone.0202865.
- Alexandria Real Estate Equities; Friends of Cancer Research. Blueprint for breakthrough forum: research and reimbursement in the age of precision medicine. September 2018. www.focr.org/sites/default/files/pdf/Friends Alexandria Blueprint White Paper_October.pdf. Accessed September 6, 2019.