When to Suspect Germline Abnormality From Genetic Testing Results

When to Suspect Germline Abnormality From Genetic Testing Results

In the following article, W. Michael Korn, MD, and Robert Nussbaum, MD, answer a question posed by an attendee during a 2015 Best of ASCO® Meeting. Dr. Korn is a professor of medicine in the Divisions of Gastroenterology and Medical Oncology at University of California, San Francisco, and Dr. Nussbaum is a professor in the Department of Medicine and chief of the Division of Medical Genetics at University of California, San Francisco.

Dr. W. Michael Korn

Question: When should you suspect a germline abnormality when genetic testing results come back?

Answer: This is an important question highlighting the clinical and ethical challenges associated with the increasing availability of next-generation DNA sequencing technology and the current lack of evidence-based guidelines.

A recent study examined the rates of cancer-predisposing mutations in matched tumor/normal DNA samples using a 341-gene panel, of which 187 genes have been implicated in hereditary cancer predisposition syndromes.1 The study demonstrated that 12.6% of patients undergoing tumor sequencing were found to carry germline mutations in one of these 187 genes, such as inactivating mutations of TP53 (Li-Fraumeni syndrome).1 In addition, some germline variants are associated with risk of adverse effects of therapy. Thus, even though next-generation DNA sequence analysis of tumor DNA might be aimed at discovering therapeutically targetable alterations, incidental discovery of germline abnormalities has to be anticipated and met with an appropriate response that assures optimal care for patients and their families.

Dr. Robert Nussbaum
The list of germline gene alterations associated with increased cancer risk is rapidly growing and currently includes close to 100 genes,1,2 many of which are included in commercial assays analyzing tumor DNA that do not test  germline DNA directly. Although mutations in some of these genes, such as BRCA1 or MSH2, are known to have high penetrance and can be associated with a typical family history, many of these show moderate penetrance and might be present in a patient’s germline in the absence of a high cancer burden in the family. Furthermore, germline mutations in cancer susceptibility genes are frequently associated with unexpected phenotypes, for example, the finding of BAP1 mutations (frequently associated with melanoma) in hepatocellular carcinoma.1 Thus, when interpreting results obtained with assays that analyze tumor DNA only, the high frequency of  germline mutations in patients with advanced cancers, as well as the limitations of family history and tumor classification, need to be taken into account.

In some cases, the nature of the somatic alteration might hold some clues. For example, the finding of a recurrent founder mutation in BRCA1 is strongly suggestive of germline abnormalities. Beyond such well-documented findings, any pathogenic or likely pathogenic mutation (according to the American College of Medical Genetics and Genomics [ACMG] pathogenicity criteria3) involving known cancer-predisposition genes found in tumor DNA should raise the suspicion of a  germline aberration. The ACMG has defined a minimum list of genes in which there is both disease susceptibility and clinical actionability. The list includes 26 cancer-related genes and recommends that laboratories report incidentally found mutations in these genes if discovered during whole-exome sequencing.4 Although the ACMG list was not designed to apply to tumor-only DNA analysis, the underlying principle still applies; medically relevant mutations found incidental to the primary purpose for the test (in this case the discovery of therapeutically targetable alterations) should not be withheld from patients and the ordering clinician.

As outlined in detail by the Clinical Sequencing Exploratory Research Consortium Tumor Working Group,5 physicians and their patients using tumor-only DNA sequencing panels must be aware that although such assays are aimed at understanding tumors and their vulnerabilities, they also provide a window, albeit of reduced sensitivity and specificity, into the patient’s germline. In contrast, assays that analyze tumor and normal DNA in parallel (whole-exome sequencing assays or gene panels, such as MSK-IMPACT and the UCSF 500 sequencing panels) provide fundamental advantages over tumor-only DNA tests. Firstly, nonpathogenic sequence alterations present in tumor DNA can be rapidly identified through comparison with the patient’s germline DNA sequence, thus substantially reducing the rate of false-positives.2 Secondly, categorization of mutations in cancer susceptibility as somatic versus germline is straightforward.

Therefore, we strongly recommend routine use of tumor-normal DNA sequencing assays in clinical cancer care. However, as with all germline genetic testing, information on possible consequences of discovery of germline variants through paired germline–tumor testing has implications not only for the patient but also for the patient’s relatives and should be provided after appropriate pretest counseling. Information on possible consequences of discovery of germline variants in tumor DNA should be provided through a process that allows patients to express their preferences regarding communication of such findings.