Ready, Fire, Aim: Addressing Issues Associated With Multigene Panel Testing for Cancer Susceptibility

September 15, 2016

Panel testing has important advantages but is being misused due to payer constraints and laboratory marketing pressures. Much testing is haphazard and results in utilization of limited genetic counseling resources in the discussion of variants of uncertain significance and low-penetrance gene mutations.

Genetic testing for hereditary predisposition to cancer is recognized as a standard aspect of oncology practice.[1] Identification of a pathogenic mutation in a cancer susceptibility gene enables an individual and his or her family to take proactive measures that may lead to the early detection of cancer, or, as in the case of prophylactic surgery, that may significantly reduce the risk of developing cancer.[2] In this month’s issue of ONCOLOGY, Plichta et al provide a comprehensive and thoughtful review of the current state of genetic testing for hereditary cancer susceptibility, and highlight the challenges that accompany multigene panel testing (“panel testing”).[3]

In clinical practice, panel testing may be ordered for patients who meet clinical guidelines for a single syndrome, such as hereditary breast and ovarian cancer syndrome. The objective in ordering a panel test is to identify a causative mutation by including other candidate genes.[4] Most health plans do not reimburse for panel testing, but the testing laboratories do not bill for unreimbursed costs as long as the core testing (such as testing for BRCA1 and BRCA2) is a covered service. An increasing number of patients who have undergone panel testing through an outside provider are being referred to surgical oncologists for prophylactic surgery and/or to medical genetics specialists for posttest genetic counseling. Of particular concern is the referral of patients for prophylactic surgery based on a misunderstanding of genetic test results. The results in question are often variants of uncertain significance, but increasingly they include pathogenic mutations in moderate- or low-penetrance genes for which there is no consensus or recommendation by the National Comprehensive Cancer Network (NCCN) or other professional organizations for risk-reducing surgery. These “near misses” are disturbing and validate the admonitions of the NCCN et al that genetic testing is best performed in the context of pretest genetic counseling, particularly when panel testing is being considered.[5] Nevertheless, the number of panel tests ordered continues to grow, based on the misconception that “one size fits all.” This flies in the face of the basic tenets of clinical judgment and precision medicine.

There are several reasons for the rapid adoption of panel testing. In part it has resulted from a highly competitive marketing environment among testing laboratories. Other factors include the reluctance of most health plans to affirmatively cover testing for genes such as PALB2, and the continued high price of genetic testing, despite the cost efficiencies of next-generation sequencing. For example, despite evidence that PALB2 mutations confer a 41% to 67% breast cancer risk in high-risk families, testing for PALB2 is not covered by most health plans.[6] A PALB2 mutation is therefore typically identified through a panel test. Thus, even among genetics professionals, testing decisions are often influenced by insurance considerations rather than clinical assessment and differential diagnoses.

Health plans direct providers to use specific laboratories that may not offer panel options that are the best fit for a particular patient’s personal/family history or that can accommodate their preference to test only for clinically actionable genes. When compelled to select a large panel that includes genes for which there is no suspicion, we are too often confronted with results that require a complex discussion with the patient and with information that is confusing or unnecessarily worrisome to family members at risk for the mutation.[7] Two examples of this include the p.I1307K mutation in APC, which is a low-penetrance variant common in persons of Ashkenazi Jewish heritage, and suspected somatic mosaicism in TP53, which is being found in older patients who have received chemotherapy.[7,8] Panel testing for recessive disorders in the prenatal setting is also identifying questionable mutations (eg, in the fumarate hydratase gene [FH]), prompting referrals for cancer genetic counseling.[9] The clinical implications of even some common mutations are ambiguous and vexing for the patient and family. Additionally, the screening costs for what may be a negligible risk (if any) is unlikely to be a wise use of healthcare dollars.

We concur with the authors that the protracted testing odysseys of the prepanel era were inefficient, stressful for the patient, and often uninformative.[3] Advocates of broad testing may properly assert that unexpected mutations found in clinically actionable genes such as MSH6 and PMS2 justify large panel testing. However, the finding of an MSH6 mutation in a breast cancer patient being tested for surgical decision purposes provides no useful information for the imminent decision and creates additional concerns for the patient/family. This incidental finding could have been identified in reflex testing performed after the patient’s surgery. In such a situation, there is also a temptation to attribute the breast cancer diagnosis to the MSH6 mutation, but this would be conjecture in the absence of an abnormal tumor immunohistochemistry result. Furthermore, first-degree family members who test negative for the MSH6 mutation might incorrectly conclude that they are at population risk of breast cancer, when their residual lifetime risk might be twofold and should be managed as such.

So, how might we reign in testing for cancer susceptibility without sacrificing important opportunities to identify carriers and reduce cancer incidence? Complex solutions involving detailed capture of medical/family history and algorithms for test selection (as proposed by the authors) will undoubtedly provide the best opportunity for genetic test selection with optimal sensitivity and specificity.[3] Unfortunately, the development and implementation of such solutions is likely several years away. We propose three interim steps that would address the issue of unfocused testing and improve identification of individuals at risk for hereditary cancer predisposition:

1) Update payer guidelines. Payers should cover panel testing based on published data.[10] This would eliminate the practice of selecting broad panels as an initial test, thereby reducing the number of variants identified. Genetic counselors and providers would then order reflex testing on an “as-needed” basis.

2) Improve laboratory options. Certain large laboratories only offer fixed panels. More laboratories should offer customized panel options, allowing for selection of suspect genes and omission of genes that are not suspect. This would facilitate more refined testing based on medical/family history and selection of clinically actionable genes. In addition, laboratories should offer the option of reflex testing within a designated time frame at no additional cost.

3) Improve access to variant interpretation. All Clinical Laboratory Improvement Amendments laboratories should submit variant data to public databases. Discordant interpretations of variants by individual laboratories should be eliminated to the greatest degree possible.

In summary, panel testing has important advantages but is being misused due to payer constraints and laboratory marketing pressures. Much testing is haphazard and results in utilization of limited genetic counseling resources in the discussion of variants of uncertain significance and low-penetrance gene mutations. This time would be better spent identifying clinically actionable genes. As the authors astutely point out, panel testing is no different than population screening in many cases.[3] Initiating controlled population screening by offering coverage for the three Ashkenazi Jewish founder mutations in BRCA1 and BRCA2 seems an obvious first step in maximizing cancer risk reduction through genetic testing.

Financial Disclosure:The authors have no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.

References:

1. Garber JE, Offit K. Hereditary cancer predisposition syndromes. J Clin Oncol. 2005;23:276-92.

2. Chai X, Friebel TM, Singer CF, et al. Use of risk-reducing surgeries in a prospective cohort of 1,499 BRCA1 and BRCA2 mutation carriers. Breast Cancer Res Treat. 2014;148:397-406.

3. Plichta JK, Griffin M, Thakuria J, Hughes KS. What’s new in genetic testing for cancer susceptibility? Oncology (Williston Park). 2016;30:787-99.

4. Walsh T, Casadei S, Lee MK, et al. Mutations in 12 genes for inherited ovarian, fallopian tube and peritoneal carcinoma identified by massively parallel sequencing. Proc Natl Acad Sci USA. 2011;108:18032-7.

5. Daly MB, Pilarski R, Axilbund JE, et al. Genetic/familial high-risk assessment: breast and ovarian. Version 2.2015. J Natl Compr Canc Netw. 2016;14:153-62.

6. Antoniou AC, Casadei S, Heikkinen T, et al. Breast cancer risk in families with mutations in PALB2. N Engl J Med. 2014;371:497-506.

7. Slavin TP, Niell-Swiller M, Solomon I, et al. Clinical application of multigene panels: challenges of next-generation counseling and cancer risk management. Front Oncol. 2015;5:208.

8. Boursi B, Sella T, Liberman E, et al. The APC p.I1307K polymorphism is a significant risk factor for CRC in average risk Ashkenazi Jews. Eur J Cancer. 2013;49:3680-5.

9. Alam NA, Rowan AJ, Wortham NC, et al. Genetic and functional analyses of FH mutations in multiple cutaneous and uterine leiomyomatosis, hereditary leiomyomatosis and renal cancer, and fumarate hydratase deficiency. Hum Mol Genet. 2003;12:1241-52.

10. LaDuca H, Stuenkel AJ, Dolinsky JS, et al. Utilization of multigene panels in hereditary cancer predisposition testing: analysis of more than 2,000 patients. Genet Med. 2014;16:830-7.