Regulation of Laboratory-Developed Tests in Preventive Oncology: Emerging Needs and Opportunities

Abstract

Cancer predictive or diagnostic assays, offered as Laboratory-Developed Tests (LDTs), have been subject to regulatory authority and enforcement discretion by the US Food and Drug Administration. Many LDTs enter the market without US Food and Drug Administration or any regulatory review. The Centers for Medicare & Medicaid Services under the Clinical Laboratory Improvement Amendments focuses on analytic performance, but has limited oversight of the quality or utility of LDTs, including whether patients have been harmed as a result of their use. Increasingly, LDTs for cancer risk or early detection have been marketed directly to consumers, with many LDT developers depicting these tests, requested by patients but ordered by personal or company-associated physicians, as procedures falling under the practice of medicine. This patchwork of regulation and enforcement uncertainty regarding LDTs and public concerns about accuracy of tests given emergency authorization during the COVID-19 pandemic led to the Verifying Accurate Leading-edge IVCT (in vitro clinical test) Development Act of 2021. This pending federal legislation represents an opportunity to harmonize regulatory policies and address growing concerns over quality, utility, and safety of LDTs for cancer genomics, including tests marketed directly to consumers. We review here questions regarding the potential benefits and harms of some cancer-related LDTs for cancer risk and presymptomatic molecular diagnosis, increasingly marketed to oncologists or directly to the worried well. We offer specific proposals to strengthen oversight of the accuracy and clinical utility of cancer genetic testing to ensure public safety.

Genomic advances have facilitated hereditary cancer risk prediction and improved molecular diagnostics,¹ resulting in a panoply of tests for inherited disease risk and presymptomatic disease detection (Table 1).² From the regulatory perspective, cancer predictive or diagnostic assays are considered Laboratory-Developed Tests (LDTs). Currently, LDTs, provided by an estimated 12,000 laboratories, reach millions and are increasingly offered directly to consumers.³ With increasing test complexity, the US Food and Drug Administration (FDA) has begun to provide an expedited de novo process for low-risk to moderate-risk devices, exercising regulatory authority and enforcement discretion over the safety and effectiveness of LDTs. Although the FDA approves or clears many tests before they reach patients, a large number of LDTs enter the market without FDA or any regulatory review because they are created and used in the same facility.³ The laboratories themselves are principally regulated by the Centers for Medicare & Medicaid Services (CMS) under the Clinical Laboratory Improvement Amendments (CLIA). CMS has limited oversight of the quality, reliability, or usefulness of LDTs, including whether patients have been harmed as a result of their use; CLIA focus on analytical validity of tests during inspections that may occur up to two years after an LDT is first performed.³

TABLE 1.

Consumer-Targeted Companies With Health-Related Genetic Testing Products Classified by Indication, Physician Involvement in the Ordering Process, and Potential for Submission of Raw Genomic Data

CONTEXT

• Key Objective

• How can more comprehensive oversight of Laboratory-Developed Tests (LDTs) improve safety of genomic risk assessment and molecular diagnosis of cancer?

• Knowledge Generated

• There has been a rapid increase in consumer-initiated cancer-related LDTs, which may not be subject to review by the US Food and Drug Administration. The existing patchwork of regulatory oversight of LDTs has raised concerns of harms as well as limited benefits and has led to pending federal legislation, the Verifying Accurate Leading-edge IVCT (in vitro clinical test) Development (VALID) Act of 2021.

• Relevance

• The VALID Act and similar measures can play an important role in enabling clinicians to assess the clinical utility and safety of tests for inherited cancer risk and presymptomatic liquid biopsies marketed to consumers and cancer care providers.

Recently, marketing of consumer genomic testing (CGT) has focused on patients and the worried well.^4-7 The CGT paradigm extends direct-to-consumer provision of testing to a model where an individual is encouraged to request a specific test from their personal physician or from a health provider employed or contracted by the testing laboratory.⁵ Although FDA considers LDTs as medical devices under its jurisdiction, many LDT developers using the CGT paradigm depict these tests, ordered by physicians, as procedures falling under the practice of medicine. The FDA does not regulate procedures falling under the practice of medicine.

This patchwork of regulation and enforcement uncertainty regarding LDTs and public concerns about accuracy of LDTs given emergency authorization during the COVID-19 pandemic, as well as the disgraced laboratory startup Theranos, constitute factors driving comprehensive federal legislation to regulate LDTs.^3,8 We review here recent concerns over the accuracy, utility, and validity of some cancer-related LDTs for cancer risk and early diagnosis, increasingly marketed to oncologists or directly to patients and the worried well. We document both risks and harms as well as benefits and offer specific proposals to strengthen oversight of both analytic and clinical validities of cancer genetic testing to ensure public safety. We conclude that amendment and passage of a pending piece of federal legislation, the Verifying Accurate Leading-edge IVCT (in vitro clinical test) Development Act of 2021, offers an opportunity to harmonize regulatory policies and address growing concerns over quality, utility, and safety of LDTs for cancer genomics, including tests marketed directly to consumers.

CURRENT LANDSCAPE OF LDTs IN PREVENTIVE ONCOLOGY

In addition to a public website of commercial laboratories that provides tests for panels of cancer-predisposing genes,⁹ an industry website directed to consumers lists more than 120 companies offering ancestry and/or genetic testing services to consumers (Table 1).¹⁰ Of five major ancestry companies that claim to hold DNA samples on more than 26 million individuals,¹¹ two offer an option for medical tests, one of which remains the only FDA-authorized health report. Of 96 entities listed with sufficient information to evaluate, 85 use a CGT model requiring cash or credit card payment without mention of acceptance of governmental or private insurance. As shown in Table 1, of the tests listed on this website, with the exception of a broad category including nutrigenomics, the highest number of tests reported was for the category of cancer-related testing. In addition, more than a dozen laboratories now offer liquid biopsy tests for circulating tumor DNA (ctDNA) or cell-free DNA for tumor detection, monitoring, or treatment selection using samples of blood, stool, or other fluids, with differences in read depth, regions covered, limits of detection, and methodologies (Appendix Table A2, online only). Marketing of at least one ctDNA assay is following the consumer-initiated, physician-ordered, direct pay model.¹² In 2021, a commercial laboratory announced a collaboration with a large diagnostic company facilitating sample collection for a not yet FDA-approved ctDNA test.¹³

In addition to providing reports and analyses, some CGT companies provide personalized raw data for consumers to download; as shown in Table 1, more than a dozen companies offer reports derived from input of these raw data. Such reports bear on characteristics such as fitness and cancer and heart disease risks. It is estimated that up to 62% of consumers use third-party applications to interpret their raw data and health information,¹⁴ with 40% of genetic variations found and then sent for clinical confirmation resulting in false positives.¹⁵

REGULATORY CONTEXT

The hallmark illustration of the loophole in FDA regulation of LDTs was the Theranos case, in which a company marketed what was later found to be a fraudulent technology that purportedly allowed thousands of multiplexed tests.¹⁶ Establishing important precedents during the pandemic, the FDA gave Emergency Use Authorization to at-home self-collection COVID-19 tests^17-19 and the US Department of Health and Human Services (HHS) determined that the FDA would not require premarket review of LDTs, including but not limited to COVID-19 tests.^20,21 These regulatory exceptions were also applied to cancer genomic testing. In November 2021, HHS reinstated a requirement for FDA Emergency Use Authorization approval for COVID-19 LDTs. Non–COVID-19 LDTs remained under FDA's enforcement discretion, whereby FDA reserves the right to take action when harm occurs or may occur.²² The regulatory and enforcement uncertainty regarding LDTs and public concerns about COVID-19 laboratory testing and the Theranos case have increased the likelihood of comprehensive LDT legislation this year.^3,8,23 Such legislation, introduced in a bipartisan and bicameral manner in the current session of Congress, is represented by the Verifying Accurate Leading-edge IVCT (in vitro clinical test) Development Act of 2021 (Appendix Table A1, online only). At the same time, legislation introduced in the House (H.R. 8845) and Senate (S5051), the Multi-Cancer Early Detection Screening Coverage Act, seeks to mandate Medicare coverage of molecular early cancer detection assays.²⁴

MEDICAL CONTEXT FOR REGULATORY OVERSIGHT

In vitro diagnostic tests (IVDTs) use biospecimens to determine the presence or risk of certain diseases. Analytic validity refers to whether an IVDT can measure what it purports to measure (eg, DNA sequence changes), whereas clinical validity is the ability of an IVDT to measure a medical condition or predisposition. Clinical validity can be quantified by the sensitivity, specificity, and predictive value of an IVDT, whereas the most important parameter to the clinician is clinical utility, namely, whether the IVDT and any subsequent interventions lead to an improved health outcome among people with a positive test result.²⁵ Laboratory-developed in vitro tests (LDTs or home brew tests) were not historically regulated by the FDA; under the Medical Device Amendments of 1976,²⁶ certain LDTs have been exempt,²⁷ with the agency using enforcement discretion when there is alleged harm to consumers. According to the FDA, the LDT exemption policy did not routinely apply to direct-to-consumer genetic tests,²⁸ which it can regulate, pursuant to its enforcement discretion, as medical devices.²⁹ Only laboratories certified by the CLIA as being high-complexity can develop and deploy LDTs, with an estimated 12,000 of 267,000 such laboratories registered with CMS as of March 2020.³ Many of these laboratories process thousands of patient samples per day, and because these LDTs are not centrally registered, the number of such tests on the market or their performance as compared with FDA-reviewed diagnostics is not known. As noted in a recent Pew Report, when the FDA was originally granted oversight of medical devices, most LDTs served a limited number of patients living near the laboratories that developed them. Today, LDTs reach millions of people and increasingly, these tests are being offered directly to consumers.³ However, the FDA has reviewed very few of the LDTs offered via the CGT paradigm³⁰; the remaining are unapproved LDTs.

Widespread consumer access to genomic testing has potential benefits such as increased testing accessibility, affordability, and consumer health empowerment⁴ and concerns including inconclusive results, false negatives and positives, poor interpretation, and therefore potential for inappropriate medical management.³¹ The FDA has pursued divergent approaches to regulate the CGT industry (Fig 1).³² Perhaps because of deficiencies of the existing review processes³³ or deregulatory trends, the FDA has embraced flexible regulatory paradigms for tests offered through CGT.³³ To reduce the burden on test developers, the FDA accredited the New York State Department of Health as a third-party reviewer of certain LDTs.^34,35 These third-party accreditation programs do not currently include most genetic tests,³⁶ but entities like the New York Department of Public Health are accredited to review 510(k) submissions of next-generation sequencing–based tumor profiling tests.³⁵

FIG 1.

Timeline of FDA authorization of direct-to-consumer and COVID-19 testing. ACLA, American Clinical Laboratory Association; DAIA, Diagnostic Accuracy and Innovation Act; DTC, direct-to-consumer; FDA, US Food and Drug Administration; HHS, Department of Health and Human Services; IVCT, in vitro clinical test; LDT, laboratory-developed test; PGx, pharmacogenomic; VALID, Verifying Accurate, Leading-edge IVCT Development.

Although genetic testing has improved disease prevention and management, most notably for cancer,³⁷ substantial challenges remain, including the interpretation of results of genomic variants of uncertain significance, incomplete genetic knowledge of health care providers, and disparities in access to personalized genomic services.¹

SAFETY CONCERNS

Safety concerns of nonapproved LDTs and consumer-marketed testing relate to false-positive and false-negative results because of analytic or interpretative error, incomplete genetic assessment, failure to communicate results or take medical action, and untoward psychosocial effects.^38-42 As documented in Appendix Table A3, online only, we and others have observed cases where we were unable to confirm results of consumer-initiated tests for cancer predisposition (eg, a case of Peutz-Jeghers syndrome), errant interpretation (eg, a patient with a CHEK2 mutation told that she could have Li-Fraumeni syndrome), and severe psychological sequelae because of absent counseling (eg, a 14-year-old offered testing by her parents for recreational purposes). Consumer risks may also result from pursuing inappropriate medical interventions on the basis of results of third-party raw genomic data interpretation companies, which are largely unvalidated (Appendix Table A3).⁴³ At present, almost half of consumers seeking clinical confirmation of their genetic testing results from raw genomic data interpretation services could not confirm results, wasting health care dollars.¹⁵ We and others have seen in consultation numerous cases where results of cancer risk testing from CGT companies or those that interpret raw data from self-directed DNA sequencing were not confirmed, and some led to adverse sequelae including scheduling of unnecessary prophylactic surgeries, false reassurance, or undue anxiety (Appendix Table A3).⁴⁴ We are conducting an ongoing survey of genetic counselors in the United States seeking to document the occurrence of cases where tests by CGT or raw data interpretation were not confirmed or led to adverse outcomes.⁴⁵

Adding to safety concerns, consumer-marketed genetic testing has been shown to be associated with failure to take follow-up action.^38,39,46 Only 27% of patients in one study shared their genetic results with primary care providers³⁸; this may lead to misinterpretation of genetic tests and false reassurance or undue anxiety.^43,44,46-49

Finally, safety concerns also stem from limited FDA oversight and inconsistent reporting of metrics for clinical validity. For example, increasingly marketed as consumer-initiated LDTs, a recently developed ctDNA test for cancer early detection is not FDA-approved. Overall, the sensitivities and specificities of current ctDNA tests vary substantially with lowest sensitivities for some tests observed for early stages of disease (Appendix Table A2) and with false-positive rates of some ctDNA presymptomatic tests recently marketed in the United States ranging substantially depending on how the calculation is made.^50-56 For one methylation-based ctDNA LDT, the positive predictive value, the proportion of actual cancers found after an abnormal test in an asymptomatic population, ranged from < 10% to 45%, depending on how the calculation was performed.⁵⁵ These inconsistent metrics are due to differing methodologic assumptions regarding incident rather than prevalent cancer rates in asymptomatic populations, lower test sensitivities in earlier-stage disease, and the presence of noncancer conditions affecting specificity,^53-56 underscoring the need for clinicians and regulators to have access to reproducible metrics of clinical validity. For presymptomatic cancer LDTs, a predictive value of a positive test that is < 50% will have substantial safety implications in terms of risks of resulting invasive diagnostic procedures. Indeed, the number of false positives for one recently marketed ctDNA LDT was noted to be higher than the number of true positives, with many true positives being lymphoid cancers for which early disease detection is of unproven efficacy.⁵⁷ For other ctDNA methodologies, inconsistencies have been noted between orthogonal approaches.⁵⁸

STRENGTHENING GENETIC TESTING REGULATORY FRAMEWORKS TO PROTECT PUBLIC HEALTH

Drafted well before the COVID-19 pandemic–focused concern on FDA oversight of LDTs, the Verifying Accurate Leading-edge IVCT Development (VALID) Act of 2021 was reintroduced into both the House and Senate by a bipartisan group including US Senators Michael Bennet (D-CO) and Richard Burr (R-NC). The Act seeks to modernize regulatory oversight of LDTs by creating a single, diagnostics-specific, regulatory framework under the authority of the FDA (Appendix Table A1).

The VALID Act proposes a new tiered, risk-based system for the regulation of IVCTs, which includes LDTs, that resembles the traditional approach to regulating medical devices.⁵⁹ For high-risk IVCTs, there is an FDA preapproval review, from which low-risk diagnostics are exempt, with a middle tier requiring approval, but without meeting the more stringent requirements of high-risk diagnostics. The Act also allows for grandfathering status for qualifying LDTs that were offered for clinical use before enactment of the legislation and a Technology Certification program for marketing authorization on the basis of documentation of methods, procedures for test development, validation and maintenance, and clinical and nonclinical data used in designing the test (Appendix Table A1). One opportunity to pass the VALID Act in 2022 will be as an attachment to the reauthorization of the Medical Device User Fee Amendments (MDUFA), which funds almost half of the FDA's annual budget.⁸

The approach of the VALID Act would build on the FDA's risk-based classifications for specific genetic associations, modeled on the framework for analyte-specific reagents (ASRs).⁶⁰ Those ASRs that are Class II or Class III would have additional requirements for their safe and effective use. The FDA's recent regulation of pharmacogenomic (PGx) testing is illustrative of the use of risk-based approaches to labeling and regulation. As the FDA authorized the first and only CGT PGx test, it issued a simultaneous warning letter cautioning the public against adjusting the dose or stopping medication on the basis of PGx tests with unapproved claims.²⁹ The FDA subsequently demanded another testing company to stop offering physician-ordered PGx tests that lacked evidence of clinical validity, while at the same time, a PGx testing company provided limited consumer access to drug-specific information.²⁹ In February 2020, the FDA published a list of pharmacogenetic associations that it believes to be supported by evidence.⁶¹

A counterproposal to the VALID Act, supported by some testing laboratories, the Verified Innovative Testing in American Laboratories Act, would maintain LDTs under CLIA regulation with updates to CLIA to account for modern tests.⁶² At the same time, new legislation was introduced into the House (H.R. 8845) and Senate (S5051), the Multi-Cancer Early Detection Screening Coverage Act, to ensure timely Medicare coverage of molecular early cancer detection assays once they are FDA-approved.²⁴ Such legislation is unprecedented as it would bypass the evidentiary review by bodies such as the US Preventive Services Task Force and professional organizations.

STRENGTHENING THE FDA's RISK-BASED REGULATION

As a matter of public health policy, it seems prudent that risk-based regulatory approaches should apply to all LDTs regardless of whether a test is consumer-initiated or physician-ordered and whether the health-related report is derived from data from an in-house in vitro diagnostic test or raw data from another source. Laboratories marketing consumer-requested germline or somatic (ctDNA) assays should not be permitted to evade review simply because tests are ordered by a prescribing doctor or subcontracted to companies that hire physicians to process CGT requests. In New York, for example, tests ordered by physicians employed by the laboratory performing the test are illegal (Section 238 of Public Health Law) and physicians may not order tests that are not approved by the New York State Department of Public Health. A risk-based approach to oversee all LDTs would serve to avoid regulatory end runs⁶³ by CGT marketers.

STRENGTHENING THE VALID ACT TO ADDRESS SAFETY CONCERNS OF LDTs

As summarized in Table 2, the VALID Act offers an opportunity for cancer genomic tests to be placed in a high-risk LDT category to ensure that there is premarket review of analytic and clinical validities of tests that will determine medical interventions. Congress should amend the Act to strengthen postmarket protections and reinforce the requirement that health care workers report to FDA cases of test-related patient harm and define specific mandates for FDA collection of user fees from test developers to fund oversight activities.⁶⁴ There should be uniform requirements of provision to consumers of counseling resources before inherited risk testing and assurance of access to appropriate medical follow-up after testing. Rules endorsed by professional societies pertaining to testing vulnerable populations (eg, young children who would not benefit from testing at early age) should be incorporated into FDA guidelines. Although the 2016 21st Century Cures Act excluded certain medical software from regulation as medical devices,⁶⁵ the VALID Act should clarify that software used to analyze raw consumer–entered genomic data to generate reports that bear on health (eg, cancer-causing mutations) fall within the purview of FDA,⁶⁶ regardless of whether the reports are generated for profit or not for profit.⁶⁷

TABLE 2.

Potential Enhancements of Regulatory Oversight of Consumer-Initiated Genomic Testing for Cancer Detection and Risk Assessment

The VALID Act can also address many of the safety concerns stemming from newly developed LDTs including ctDNA assays. It is critical that H.R. 8845 is harmonized with the VALID Act to ensure full premarket FDA review so that ctDNA tests meet benchmarks of other diagnostic tests. The adoption of proposed congressional mandates for CMS coverage of ctDNA would bypass current evidentiary review, increase costs because of false-positive tests, and potentially increase health disparities. It has been shown that even after a genetic test has been performed, there remain substantial disparities in access to and adoption of preventive surgery or radiographic screening, as well as genetically targeted therapies, in historically underserved groups.⁶⁸ ctDNA LDTs should also be classified as high risk, and manufacturers should be required to submit to the FDA-standardized reporting of metrics for clinical validity including sensitivity, specificity, false-positive and false-negative rates, and positive predictive value. As with physician-ordered PGx tests, the FDA should require that manufacturers of ctDNA LDTs demonstrate clear evidence of clinical validity of tests.

Perhaps the most effective strategy to ensure the quality and safety of genetic testing services has been the effort to promulgate a proficiency testing (PT) of laboratories providing LDTs. An example is the PT program for inborn errors of metabolism, involving more than 150 laboratories.⁶⁹ Currently, all laboratories certified under CLIA, or its main accrediting bodies, the New York State Department of Health, the College of Pathology, or the Joint Commission, require PT testing.⁷⁰ A regulatory requirement for PT should be reinforced by the VALID Act for all laboratories that offer consumer-initiated germline or ctDNA assays. Laboratories should not be permitted to declare that their tests are for educational purposes only, hence not subject to oversight, if the laboratories suggest to consumers that results be discussed with health care providers as they may generate clinically actionable findings. The Act should also provide consumers with information regarding professional training (eg, appropriate board certification) of geneticists or oncologists ordering and interpreting cancer genetic tests and guiding their medical care, who should have fiduciary responsibilities to the patient or consumer and not to the testing laboratory.

CLINICAL VALIDITY AND FDA PARTNERSHIPS

In view of its finite resources, the FDA has recognized the genetic variant information in the Clinical Genome Resource (ClinGen) consortium's ClinGen Expert Curated Human Genetic Data as a source of scientific evidence that can be used to support clinical validity in premarket submissions.⁷¹ Such recognition by the FDA increases safety and availability of information on clinical validity and the pathogenicity of individual genomic variants, made possible by more than 700 ClinGen stakeholders aiming to standardize clinical annotation and interpretation of genetic data.⁷² Requirements for clinical validity should be built into FDA approval of high-risk LDTs; these considerations have been applied by the Federal Trade Commission for CGT companies, making unsupported nutrigenetic and dermagenetic claims of clinical utility for products in the absence of randomized clinical trials.⁷³

ClinGen and its variant database partner, ClinVar, play a vital role in curating genomic information to support genomic medicine and research.⁷⁴ However, this database relies on voluntary contribution of data.⁷⁵ The FDA can play a stronger role in encouraging such participation; indeed, it has already expressed its need to request raw data to assess analytical and clinical validities of high-risk IVCTs.⁷⁶ The FDA can also require disclosure of proprietary information to the agency, providing incentives to CGT companies like those contained in the Hatch-Waxman Act.⁷⁷ As a step in this direction, the FDA is a sponsor of an international data set of genetic variant information for the BRCA genes built by the Global Alliance for Genomics and Health.⁷⁸ In addition, with HHS, the FDA proposed a Comprehensive Test Information System in its initial comments on the VALID Act.⁷⁹

CONSUMER ACTIVISM

Another driving impetus for regulatory protection relates to privacy considerations resulting from immense private genetic data sets. The DNA of a projected 60 million Americans is expected to be in the possession of commercial laboratories by 2025.⁸⁰ Genomic data, without voluntary release, have already been used for forensic purposes, and there is a movement to create large genetic databases to aid law enforcement.⁸¹ The Health Insurance Portability and Accountability Act was modified in 2013 to safeguard privacy and genetic information, but it is unclear if this regulation applies to CGT results if these are used for a medical purpose.^40,77,82 In the realm of genetic privacy, consumer activism will be required to motivate regulatory action and inquiry, as is occurring now in the debate over private and sensitive information sold by social media for commercial purposes.⁸³

In conclusion, oncologists and other health practitioners are faced with an increasing number of LDTs to guide preventive, diagnostic, and therapeutic aspects of care. Genomic tests constitute an important subset of biomarkers for cancer risk, diagnosis, and prognosis. To safely and responsibly use new technologies, clinicians ultimately rely on shared performance metrics for these biomarkers, including clinical validity and, most importantly, the clinical utility or actionability of the LDT result. For genomic LDTs, in the absence of uniform regulatory oversight and transparent communication of clinical utility metrics by commercial laboratories, practitioners can rely on guidance of expert bodies such as the National Comprehensive Cancer Centers, the US Prevention Services Task Force, and federally supported efforts such as the ClinGen consortium.^84-86

More widespread accessibility to genomic tests, facilitated by the direct-to-consumer model, has potential benefits: it can increase consumer awareness of the importance of genetics and preventive health, improve access to genetic information not provided by clinicians or insurance companies, lower costs, enhance the convenience of sending of samples from home, and make data available to companies for research. Potential drawbacks of consumer-initiated genetic testing include questionable reliability of results, failure to take into account other cancer risk factors, incomplete translation of test results to preventive action without input by clinicians, and concerns about security and privacy of consumers' genetic information.³¹ On a societal level, commercial profitability of a new LDT is determined by the sheer volume of tests performed, whether positive or negative. In some cases, CGT approaches could be cost-effective when comparing the costs of the test and resulting interventions relative to the amount of benefit that they yield. Such would be the case for widespread testing, for example, of founder mutations of susceptibility genes for breast, ovarian, prostate, and colon cancers in defined populations. However, cost-effectiveness also takes into account the burden of false negatives and positives, including costs of un-needed diagnostic procedures, or missed diagnoses because of false-negative tests. Randomized cohort studies, particularly for emerging liquid biopsy tests, may be required to provide the evidence base needed to derive both medical efficacy and societal cost-effectiveness compared with other strategies of cancer screening. In the absence of these data and uniform regulatory oversight of CGT, consumers and clinicians will need to assess test results in the context of individual personal and family histories of cancer and other risk factors and existing professional guidelines.

Although clinicians can be assured that the majority of testing laboratories are committed to the highest quality and meet current regulatory requirements, safety concerns resulting from the recent proliferation of consumer-initiated cancer genomic testing create a need for the FDA to assume full oversight of LDTs for cancer and other diseases. The VALID Act's requirement of the FDA to use a risk-based approach to regulate quality and safety of LDTs seems a necessary and important first step. The Act should be amended so that self-directed genetic tests and commercial laboratories that produce personalized health reports on the basis of genomic data have the same regulatory oversight as other high-risk tests. Barriers should be put in place to avoid regulatory end runs⁶³ by companies hiring physicians to order genetic tests. Passage of an amended VALID Act will uphold the FDA's mission to protect public health by ensuring that laboratory-developed presymptomatic and diagnostic genetic tests marketed to consumers are safe and effectively used to guide medical care and disease prevention.

APPENDIX

TABLE A1.

Summary of VALID Act⁸⁷

TABLE A2.

Comparison of Selected Commercial Laboratories Offering Molecular Detection of Cancer From Liquid Biopsy Specimens

TABLE A3.

Selected Literature and Case Reports Following Self-Directed Genetic Testing, Grouped by Challenges Traditionally Addressed by Genetic Counseling¹⁰⁶

AUTHOR CONTRIBUTIONS

Conception and design: Kenneth Offit, Catherine M. Sharkey, Xiaohan Wu, Jada G. Hamilton, Michael F. Walsh, Steven M. Lipkin, Michele Caggana

Administrative support: Kenneth Offit, Sita Dandiker

Provision of study materials or patients: Dina Green, Magan Trottier

Collection and assembly of data: Dina Green, Xiaohan Wu, Magan Trottier, Michael F. Walsh, Sita Dandiker, Sami Belhadj, Thelma Alessandra Sugrañes

Data analysis and interpretation: Kenneth Offit, Catherine M. Sharkey, Dina Green, Xiaohan Wu, Michael F. Walsh, Sita Dandiker, Zsofia K. Stadler

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Regulation of Laboratory-Developed Tests in Preventive Oncology: Emerging Needs and Opportunities

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).