Precision Oncology News and My Gene Counsel have partnered to produce the “Genetic Testing Challenges in Oncology” series to highlight real-world issues that genetics experts and medical professionals are encountering as genetic tests are increasingly used in cancer care. Experts submit anonymized case reports to My Gene Counsel, and based on the details in these reports, Precision Oncology News writes a feature that describes the case history, the challenges encountered by professionals in dealing with the case, and strategies they used in response to challenges or errors. The features also include a discussion with My Gene Counsel genetic counseling experts on better approaches that could be considered if similar cases are encountered in the future. In publishing this series, our aim is to educate experts in the field and foster discussion. If you would like to submit a case report, please email email@example.com.
A patient with Cowden syndrome developed a neuroendocrine tumor. Cowden syndrome is a hereditary condition that causes non-cancerous tumor-like growths in the skin and mucous membranes. It can also increase the risk for certain cancers, such as melanoma, breast, thyroid, and endometrial cancer. However, neuroendocrine tumors are not typically associated with this syndrome. As part of the Cowden syndrome diagnosis, earlier germline genetic testing had revealed this patient had inherited a point mutation in the PTEN gene. This mutation is well known in the small Cowden syndrome community and reported in the medical literature multiple times in families with this syndrome.
Most cases of Cowden syndrome are due to PTEN mutations, though point mutations are not typically classified as pathogenic. However, molecular tests have shown that this particular point mutation results in the loss of key PTEN functions. To further research the molecular features of the neuroendocrine tumor, the oncologist, with the patient’s permission, sent a sample from the tumor for sequencing to a commercial testing lab that does somatic testing. When the results came back, the oncologist took note of the fact that the lab did not list the germline PTEN mutation among the pathogenic variants in the report. But the lab did list this PTEN mutation later in the report as a variant of unknown significance (VUS), indicating inaccurately that the variant’s clinical significance was unknown.
A woman developed metastatic melanoma, and her oncologist suspected that it may be due to a familial BRCA2 mutation. The patient’s mother had breast cancer and harbored a synonymous BRCA2 mutation in the coding region of the gene.
Synonymous mutations are single-base changes that often don’t impact the gene’s protein product. However, some synonymous mutations are not benign and can impact how DNA is cut and spliced together to make RNA. This, in turn, can disrupt the protein translation process and cause increased risk of cancer. The BRCA2 mutation the mother has is known to destroy a splice donor site and causes an exon to be skipped. It is described in the literature as an example of a seemingly “silent” mutation that is actually pathogenic. In the publicly accessible variant database ClinVar, this variant is listed as pathogenic in all 21 submissions.
The oncologist suspected that the metastatic melanoma patient had inherited the same BRCA2 mutation from her mother. Melanomas have been linked to BRCA2 mutation carriers in studies. The doctor sent a sample of the daughter’s skin cancer to a lab for somatic sequencing in the hopes of identifying additional treatment options. When the test results came back, the BRCA2 mutation was not reported.
How were these cases resolved?
Since the woman’s PTEN mutation associated with Cowden’s syndrome was already known, the patient’s care was not impacted by the fact that the lab classified the PTEN mutation as a VUS. However, for the oncologist, who has a special interest in genetics, this is an important example of the differences between labs that specialize in germline versus somatic testing.
While a lab specializing in germline testing would have been able to correctly identify a well-known PTEN mutation associated with Cowden syndrome, a somatic testing lab, focused on identifying mutations in the tumor that can be targeted by treatments, may not use the same information sources to interpret variants. This can lead to mistakes. In this case, it appears that when the somatic testing lab detected a PTEN point mutation and considered the fact that point mutations are not typically pathogenic, they did not investigate further and called it a VUS.
When the melanoma patient’s test report came back without any reference to the BRCA2 mutation, the oncologist contacted the lab and explained about the specific BRCA2 mutation harbored by her mother with breast cancer. The lab went back into the case and saw that the test did in fact detect the mutation. However, the lab did not report it because this “silent” but pathogenic BRCA2 mutation was not listed in its database. Since the somatic variant database and classification process at this lab are most likely optimized for somatic variant classification, the lab would not rely on other sources utilized in a germline setting, such as the documentation of specific exceptions from the medical literature. The daughter subsequently received germline testing, which confirmed that she had inherited the BRCA2 mutation. This finding did not change her treatment strategy, but it is important information for her future cancer risk and family planning.
What could have gone wrong?
Both cases illustrate the different goals, methods, variant classification, and reporting practices of labs testing for germline mutations that are inherited in all the cells of the body and labs testing for somatic mutations that are acquired in certain cells of the body and cannot be inherited. Germline testing is used to characterize cancer risk for patients and their relatives, but increasingly, germline mutations in certain genes, such as BRCA1/2, can also inform therapeutic strategy (e.g. PARP inhibitors). The goal of somatic testing, meanwhile, is to detect the molecular perturbations driving a patient’s tumor to grow and spread, and identify the drugs that can target and shut down those mechanisms.
However, somatic mutations that can be therapeutically targeted are typically not inherited, meaning found in the germline, and conversely, a pathogenic germline variant known to increase cancer risk may not have any therapeutic relevance. As such, lab processes to detect and clinically classify these different types of variants are not the same and require different types of expertise.
Moreover, when oncologists are ordering tumor testing, they may be primarily concerned with identifying mutations that can inform their patients’ treatments, and germline mutations associated with future cancer risk may not be a priority. Even if such information were important to the oncologists, they may not realize that a lab focused on somatic testing may not be equipped to flag germline findings.
“The oncologist in these two cases really understood the nuances of genetics and had both treatment implications and hereditary cancer predisposition front-of-mind, and that’s probably why nothing went wrong,” said Meagan Farmer, a genetic counselor and genetic clinical operations director at My Gene Counsel. “The average oncologist, though, may not.”
That could have resulted in critical missed opportunities in cancer screening and risk reduction for the patients and their relatives. What if, for example, in the first case, the PTEN point mutation associated with Cowden syndrome had not been known, and the oncologist had just ordered the somatic test? The oncologist would not have learned about the PTEN mutation because the variant was listed much later in the report and misclassified as a VUS. Some oncologists with expertise in precision oncology delve into the VUS section in these lengthy NGS tumor test reports, but many doctors don’t pay close attention to that section, Farmer said.
In the second case, it was important that the oncologist knew about the BRCA2 variant in the mother and contacted the lab to check whether it didn’t detect the mutation in the daughter at all, or detected it but didn’t report it. In the end, although the BRCA2 mutation didn’t change the melanoma treatment options for the daughter, given the heightened risk for breast and ovarian cancer associated with such mutations, the knowledge of this mutation is important for her family planning and future cancer risk screening.
“Someone might wonder, ‘Why would anyone care or think of looking for a germline mutation in a somatic test report?'” noted Farmer. “I can’t recall how many times I’ve been in a molecular tumor board, and someone suggested that a patient undergo germline testing, and the treating oncologist said, ‘Oh, we already tested the tumor and we didn’t find anything.'”
Certain genes tend to be commonly mutated in the tumor but rarely in the germline, while others are more likely to be altered in the germline and rarely acquired just in the tumor. As such, blood or normal tissue samples are needed for the determination of germline mutations, which cannot be definitively detected from the assessment of tumor tissue only. However, since germline mutations occur in every cell in the body, including tumor cells, next-generation sequencing tests that assess tumor samples for alterations in dozens to hundreds of genes are increasingly identifying mutations that may be germline in origin.
For example, if a somatic testing lab detects a BRCA2 mutation in the tumor of a patient, there is a high likelihood that it is a germline mutation, and professional practice guidelines recommend that patients receive confirmatory germline testing. However, many labs don’t report these presumptive germline variants or report only a subset of them due to increased costs, limited in-house expertise, and patient consent issues.
How can challenges like this be avoided?
Although currently, there are no standardized policies for how labs should deal with germline findings from tumor testing, increasingly genetics experts are of the view that it is unethical for labs to suppress findings that could be potentially clinically significant for patients and their families. By the same token, expert bodies are also advising oncologists to consider the possibility of an inherited cancer predisposition in their patients, and to not ignore germline findings reported within tumor sequencing reports.
In a recent points to consider document, the American College of Medical Genetics and Genomics (ACMG) told genetic testing labs to be transparent about their ability to detect germline variants from tumor testing, as well as their reporting practices. Additionally, the ACMG told clinicians to take the opportunity when ordering tumor testing to evaluate the patient for clinical signs of an underlying hereditary cancer syndrome that may require germline testing.
When deciding whether to order somatic or germline testing, or both, oncologists should be aware of and consider the benefits, limits, and possible outcomes with these types of tests. If the doctor orders somatic testing first, and this doesn’t flag any clinically significant potential germline findings, but there is still a suspicion that this patient may have an underlying hereditary cancer, then germline testing should be ordered, Farmer said.
Even if the doctor doesn’t believe a patient’s clinical scenario warrants germline testing, “these things often get identified incidentally,” Farmer said. “Especially with some of the more moderate-risk cancer genes, it may not be uncommon to not have a striking personal or family history of cancer.”
When tumor testing unexpectedly identifies presumed germline variants that may be clinically important, these findings may be confusing for the patient. Farmer noted that doctors can prepare patients for the possibility of germline results in the pre-test conversation and connect them to genetic counselors in the post-test setting if there are clinically relevant findings. “It will become really important from an informed consent and liability standpoint to tell patients that germline variants could be found from tumor testing,” she said.