Many challenges remain to the optimal interpretation of genetic test results and the effective translation of these data into improved diagnostic, prognostic, and therapeutic capabilities.
During the past several decades, genetic testing for germline and somatic mutations has been incorporated into routine clinical practice for many different cancer types. Germline mutations, contained within the heritable genome, and somatic mutations, acquired de novo by cancer cells, have historically been considered as separate entities. Until recently, each had unique clinical applications and implications for patient care.
As the use of next-generation sequencing (NGS) has become more widespread, the lines between the 2 types of mutations have increasingly become blurred. Researchers are seeking to make sense of unprecedented amounts of genetic information; for example, investigators conducting a pan-cancer analysis across 12 cancer types identified distinct patterns of aberrations in 51 genes due to germline and somatic mutations.1
Many challenges remain to the optimal interpretation of genetic test results and the effective translation of these data into improved diagnostic, prognostic, and therapeutic capabilities. Nevertheless, there are signs of change. In December 2016, the FDA approved the PARP inhibitor rucaparib (Rubraca) for the treatment of patients with BRCA-mutated ovarian cancer, regardless of whether the alterations stemmed from germline or somatic changes.2For decades, the prevailing paradigm in oncology research has been that cancer is driven predominantly by the accumulation of mutations in the genetic material as a result of DNA-damaging chemical or environmental assaults on the cell. Anywhere from 2 to 8, potentially as many as 20 mutations, are required for the development of a cancer.3,4
In most cases, cancers are sporadic, driven by acquired somatic mutations. These mutations are unique to tumor cells and occur in genes encoding proteins that play a central role in the hallmark processes that dictate malignant growth. Sometimes, however, the mutations found in cancer cells are present in all cells of the body and represent genetic variations inherited from a parent. These are germline mutations that were present in the germ cells that created the sperm or egg from which the fetus developed.
Germline mutations increase the likelihood that an individual will develop cancer in his or her lifetime. Since most cancers result from multiple “hits” to a gene or to several different types of genes, individuals with germline mutations already have the first hit present from birth. In approximately 5% to 10% of cases, hereditary cancer ensues and family members can develop the same type of cancer or multiple types of primary cancer. An additional 10% to 15% of cases are referred to as familial, in which an inherited gene is suspected but has not been identified (Figure).5
In some genes, germline mutations have been linked to the development of certain types of cancers and form the genetic basis of several wellknown hereditary cancer syndromes (Table 1).6
The best characterized of these likely are hereditary breast and ovarian cancers (HBOCs), caused by germline mutations in the breast cancer susceptibility genes BRCA1 and BRCA2. It is estimated that approximately 1% of the general population carries germline BRCA1/2 mutations. Their presence increases the risk of developing ovarian cancer from about 1.3% to 39% (BRCA1) and 11% to 17% (BRCA2); for breast cancer, the increased risk ranges from 12% to 55% or 65% (BRCA1) and up to 45% (BRCA2).6-8 Germline BRCA1/2 mutations have also been implicated in several other tumor types, including prostate and pancreatic cancer.
Another well-known hereditary cancer syndrome is Lynch syndrome, in which mutations in several DNA repair genes increase the lifetime risk of colorectal cancer (CRC) from 6% to 80%, in addition to conferring an increased risk of numerous other malignancies, such as endometrial, small intestine, and liver cancers.9Testing for somatic and germline mutations has historically been performed using different samples and methods, and for different reasons. Germline mutation testing is typically performed using a blood sample and the results are combined with information regarding family history and lifestyle factors to determine a patient’s risk of getting a particular type of cancer.
The risk is quantified by describing the penetrance of a mutation, or the proportion of individuals carrying the mutation who would likely develop cancer. Highly penetrant genes confer a relative increase in risk of more than 5 to 10 times that of the general population, moderately penetrant genes confer 2 to 5 times the risk, and low-penetrant genes less than 2 times the risk.10
Genetic testing for germline mutations has become standard practice for several tumor types, fueled in part by an increase in commercial interest in developing tests following the Supreme Court decision in 2013 that Myriad Genetics could not patent “naturally occurring genes.”
The poster child for germline mutations is BRCA1/2 gene testing in HBOCs. Typically, only individuals who meet certain criteria, such as a family history of breast cancer, are referred for testing. Mathematical and predictive models have been developed to assist in identifying patients who should be tested, but recognizing the clinical “red flags” of hereditary cancers can be difficult and rates of referral for testing are suboptimal.11-13
There are several outcomes of germline testing: a positive result indicates that a known pathogenic variant was identified and a negative result means that none was found. A problematic third outcome is a variant of unknown significance (VUS), wherein an abnormality is found, but its association with cancer risk is unknown. Reported VUS rates vary according to the test used and large collaborative efforts are underway to catalog and evaluate unknown variants uncovered during routine testing, in order to reduce those rates further.14,15
Several of the germline mutation assays currently approved by the FDA are designed to test for the presence of mutations in a single gene. In this case, a negative test result can also be uninformative if patients have a family history of cancer, since it could simply mean that a germline mutation in a different gene is behind a genetic predisposition.
In these instances, it is becoming increasingly common for oncologists to pursue second-line testing for a panel of genes. A plethora of multigene testing panels are now commercially available. In addition to establishing a risk of inherited cancer, a positive test result has several therapeutic implications. Patients who do not have cancer can choose to be closely monitored for the development of tumors or to undergo prophylactic surgery (eg, mastectomy or salpingo-oophorectomy for HBOCs, near-total colectomy for CRC).
For individuals who already have cancer, a positive result can also inform treatment choices. BRCA1/2-mutant breast and ovarian cancers have been shown to demonstrate improved response to chemotherapy and targeted therapy with PARP inhibitors.
Increasingly, experts are in favor of broadening the scope of germline mutation testing in wellestablished cancer settings, with some research leaders advocating universal testing of all patients.16 A growing number of studies have demonstrated that pathogenic germline BRCA1/2 mutations occur in a significant proportion of women who do not have a strong family history of breast and ovarian cancers or other cancers linked to these mutations. Evidence of mismatch repair deficiency in patients with endometrial and CRCs has prompted screening of an increasing number of patients for Lynch syndrome. In both cases, universal testing has been shown to be cost effective, although significant barriers to its widespread implementation still exist.17As far as somatic mutations are concerned, the primary goal of testing has been to guide therapeutic decision making. Successful development of imatinib (Gleevec), dubbed the “magic bullet” when it was introduced in 2001 as a targeted therapy for the treatment of patients with chronic myeloid leukemia with the BCR-ABL gene fusion, ushered in the era of targeted cancer therapies.
In the intervening years, clinicians have amassed a small arsenal of targeted drugs to counteract common genetic abnormalities that has translated into improved clinical outcomes for patients. Prime examples are the development of small molecule tyrosine kinase inhibitors (TKIs) for patients with somatic mutations in the tyrosine kinase receptor genes. Epidermal growth factor receptor (EGFR) inhibitors for patients with non—small cell lung cancer (NSCLC) whose tumors harbor somatic mutations in the EGFR gene and BRAF inhibitors for patients with melanoma with somatic BRAF mutations have revolutionized treatment paradigms in those malignancies.
The FDA has approved assays to test for the presence of BRAF, BRCA1/2, EGFR, FLT3, and KRAS mutations and ROS1 fusions as companion diagnostics required for the use of specific targeted therapies (Table 218).Other companion diagnostics analyze the impact of genetic aberrations through gene amplification or protein expression testing. Additionally, the agency has approved the use of several nucleic acid—based tests and multigene sequencing panels to aid in clinical decision making.
Reliance on archival formalin-fixed tumor tissue biopsies for somatic mutation analysis can be problematic. Often biopsied tissue is not available, and even when it is it can be highly variable in terms of quality and quantity. In recent years, simpler, more cost-effective, and less invasive tests have been developed. So-called “liquid biopsies” seek out cellfree tumor DNA within a patient’s blood for analysis.
Researchers are still refining the use of liquid biopsies, but commercial interest has been piqued and more than 35 companies within the United States alone are developing them. The Cobas EGFR Mutation Test v2 recently became the first liquid biopsy-based companion diagnostic to be approved by the FDA.19,20Applying NGS to germline mutation testing is challenging because it can leave clinicians with a massive amount of data that are difficult to interpret. Disclosure of that data can have potential ethical and legal ramifications and consensus guidelines for genetic counseling and clinical application of the results following NGS currently are lacking. Furthermore, the likelihood of detecting VUS increases with the number of genes that are tested.
Nevertheless, implementing NGS for the identification of somatic mutations offers the potential to rapidly identify all the mutations in a tumor and match them with the expanding arsenal of targeted therapies. In order to distinguish true somatic mutations, the tumor sequence is compared with the background, germline genome, with the assumption that germline variants are probably not related to cancer.
However, in order to save time and money, a comparison of mutations in the tumor with the germline often is not conducted. When it is performed, such testing raises major ethical concerns because some of these germline variants may in fact be related to the development of cancer. Thus, genome sequencing has the potential to reveal so-called “incidental” findings of germline variants that predict increased cancer risk, which was not the intention of the original mutation test. This has implications both for the patient being tested and their family members who could unknowingly be carriers.
This “incidentalome,” as this development is sometimes called, is now increasingly being recognized as a significant challenge. Numerous study results suggest that the incidence of germline mutations in advanced cancers is more common than previously thought. One study assessed 815 tumor-normal paired samples from patients with 15 tumor types using either exome sequencing or a 111-gene panel.
In addition to the fact that tumor-only testing misinterpreted germline mutations as somatic in 31% of cases with the gene panel and 65% of cases with exome sequencing, investigators found that 3% of patients with suspected somatic changes harbored germline alterations in cancer-predisposing genes.21 This finding not only illustrates the importance of performing matched tumor-normal sequencing, but also the small but significant percentage of cases where patients undergoing somatic mutation testing through tumor sequencing will have clinically relevant germline mutations.
The American College of Medical Genetics and Genomics has released a policy statement on genetic sequencing that emphasizes the importance of disclosing the possibility of incidental findings to patients before they undergo testing. The group has also developed a list of 56 mutant genes that have the potential to lead to severe outcomes but that are medically actionable. They recommend that if mutations in any of these genes are identified, they should be reported to the patient, regardless of the patient’s age.16,22,23Until recently, germline and somatic mutations have been largely considered as 2 separate clinical entities, with clear-cut differences in the information they provide. That view is beginning to change in the era of genome sequencing.
A broadening understanding of the tumor genome has revealed some unexpected findings. Somatic mutations in genes previously studied mostly in the germline context and germline mutations originally in genes more widely appreciated in a somatic role have been uncovered.
In fact, an increasing number of genes that can undergo either a somatic mutation or be associated with germline cancer susceptibility syndromes have been identified. Somatic mutations in the BRCA1/2 genes have been shown to account for a significant proportion of the BRCA1/2 mutations found in ovarian and breast cancers. Further, the reported rates of somatic BRCA1/2 mutations vary among studies, but they appear to be present in 4% to 7% of patients with ovarian cancer and more recent study results suggest they may also occur in around 3% of breast cancers.17,24
The tumor phenotype associated with both germline and somatic mutations appears to be similar, although germline mutations tend to be found in patients who are younger at diagnosis. Although the full clinical relevance of somatic BRCA1/2 mutations is still being unraveled, they seem to convey sensitivity to platinum-based chemotherapy and PARP inhibitors in a manner similar to their germline counterparts, with an improved prognosis for patients with ovarian cancer.17,24
The PARP inhibitor rucaparib is the first member of its drug class to be approved for use in those with either germline or somatic mutations. It was approved in conjunction with the first FDA-approved NGS-based companion diagnostic, FoundationFocus CDxBRCA.2 Meanwhile, germline mutations in the EGFR gene were discovered in 2005 and have been linked to inherited susceptibility to lung cancer.
Several studies have reported that relatives with multigenerational EGFR-mutant NSCLC harbor activating mutations in EGFR in combination with germline EGFR T790M mutations, a variant that, in its somatic form, is commonly associated with acquired resistance to EGFR TKIs in lung cancer. A second, rarer EGFR mutation associated with inherited risk of lung cancer is EGFR V843I.25