The Dividing Line for Germline Mutation Testing Is Often Arbitrary

Maurie Markman, MD

It is difficult to overstate the impact on oncology practice of molecular evaluation of mutations and unique variants within the germline as well as the identification of somatic abnormalities within the cancer itself, such as specific mutations and recombination events.

Although hereditary cancer risk was defined solely by family history in the not-so-distant past, today increasingly robust data may help define an individual’s heightened lifetime risk based on the presence of specific molecular findings within the germline. Further, where the actual risk associated with a newly discovered or particularly uncommon variant is presently unknown, prospective collection of data related to current and future clinical events can answer the question about the degree of cancer risk (whether it is high, equivalent to the population not bearing the finding, etc) when an adequate number of individuals have been followed over a sufficient period of time.

In addition, mutations and other molecular abnormalities discovered within the tumor cell (or in the germline) may identify unique cancer driver mutations (eg, BRCA, BRAF) where specific interventions targeting these events may substantially favorably influence the natural history of the malignancy. Alternatively, the presence of a particular molecular finding may identify a phenotype that is, unfortunately, highly unlikely to respond to a given therapeutic strategy, such as monoclonal antibody—directed therapy against EGFR in KRAS mutation—positive metastatic colon cancer.

However, a discussion of the impact of this discovery process, which results in so-called evidence-based recommendations that form the future standard of care in a given clinical setting, would be incomplete if one failed to acknowledge the often rather arbitrary dividing line between what is determined to have achieved a risk status meriting testing versus what has failed to attain this recognition.

BRCA Mutations Serve as Examples

This divide is illustrated in the following 2 examples, chosen from a number of possible recent instances, of the current state of affairs in the complex and potentially problematic arena of defining the clinical importance of germline genetic assessments of cancer risk.

In a recent pronouncement of revised recommendations from the US Preventive Services Task Force related to “risk assessment for BRCA-related cancer,” the panel concluded that “for women whose personal or family history or ancestry is not associated with an increased risk for harmful mutations in the BRCA1/2 genes, there is adequate evidence that the benefits of risk assessment, genetic counseling, genetic testing, and interventions are small or none [italics added].”¹ In the discussion accompanying this detailed and thoughtful effort, the panel highlighted the observation that the “risk” of discovering a BRCA mutation based on the presence of a family history of breast or ovarian cancer was 13.6% and 7.9% for BRCA1 and BRCA2 mutations, respectively.

Although it is most unlikely that anyone would debate the clinical relevance of this incidence level of germline mutations within BRCA, it is still a minority (<25%) of individuals even with this positive family history who will be found to possess 1 of these specific genetic abnormalities. However, as noted in the manuscript, for women within the general population with no known prior family history of these specific malignancies the estimated prevalence of 1 of these genes was 0.2% to 0.3%.¹ Although it would certainly be reasonable to characterize the risk as relatively “small” for this population, the addition of the term “none” to this description of risk is a conclusion that should be challenged.

The intent in raising this issue is not to argue against the potential requirement of a governmental agency to make a specific policy decision, particularly one with considerable financial implications, based on epidemiological data (eg, testing of all adult women for the presence of BRCA mutations versus a more limited population subset), but rather to highlight from a clinical perspective the somewhat arbitrary nature of this decision.

Within the “0.2% or 0.3% of women” within the population who possess a germline BRCA mutation are those who for a variety of possible reasons, such as a lack of awareness of their family history or a limited number of females within their family unit, do not know they may face an elevated risk of breast or ovarian cancers. However, their risk for the development of 1 or both of these 2 serious malignancies is as high as women who do know their family history. Alternatively, of course, the germline of these women may contain a new pathologic mutation within BRCA.

The second example serves as a useful extension of the concern about the well-intentioned but still arbitrary nature of determining cancer risk assessments. Within the population of women who develop ovarian cancer, it is well recognized today that the results of BRCA germline testing are potentially clinically relevant, not only for possible therapeutic options such as the use of PARP inhibitors but also for counseling, screening, and prophylactic surgical approaches for affected family members. However, even within this “high-risk population” only a minority of patients (<20%-25%) will be found to have a germline mutation.

In an effort to be more focused regarding who might reasonably be recommended for germline BRCA testing, investigators examined a large population of patients with ovarian cancer (N = 2192) and found a substantial subset (66%) with a “particularly low risk (3.5%) to carry a BRCA mutation” versus a smaller patient group (24%) with “a probability higher than 17% to carry a pathogenic mutation.”² Not surprisingly, factors influencing this analysis included patient age at diagnosis and additional family history of breast or ovarian cancer.

Although this is an interesting analysis, the question that should be considered is whether a policy—whether originating from a governmental agency, a payer, or a practice guideline—would be clinically or ethically acceptable if it excludes from routine BRCA testing women with ovarian cancer who have a relatively “low risk” (3.5%) of possessing this potentially life-changing germline abnormality.

References

1. US Preventive Services Task Force, Owens DK, Davidson KW et al. Risk assessment, genetic counseling, and genetic testing for BRCA-related cancer: US Preventive Services Task Force Recommendations Statement. JAMA. 2019;322(7):652-665. doi: 10.1001/jama.2019.10987.
2. Paradiso AV, Digennaro M, Patruno M, De Summa S, Tommasi S, Berindan-Neagoe I. BRCA germline mutation test for all women with ovarian cancer? BMC Cancer. 2019;19(1):641. doi: 10.1186/s12885-019-5829-4.