Chicago, IL—Tumors with high microsatellite instability (MSI-H) are predictive of Lynch syndrome. The recent approval of the first immunotherapies for use in patients with MSI-H or with mismatch repair-deficiency (dMMR) in solid tumors is now leading to routine testing for MSI-H or dMMR in patients with advanced solid tumors.
In a genomic study of more than 15,000 tumor samples, 16% of individuals with MSI-H tumors were later found to have Lynch syndrome. Zsofia Kinga Stadler, MD, Clinic Director, Clinical Genetics Service, Memorial Sloan Kettering Cancer Center, New York City, presented the results at a press briefing at ASCO 2018.
MSI is a genomic tumor marker (or biomarker) characterized by the presence of multiple mutations caused by a variance in the length of short DNA sequence repeats in tumors compared with normal tissue. MSI-H is a hallmark of Lynch syndrome–associated tumors, said Dr Stadler, who presented the study data at an ASCO 2018 press briefing.
In addition to MSI-H being a marker for Lynch syndrome, it is now also used as a marker of response to immunotherapy. In 2017, the FDA approved pembrolizumab (Keytruda) for use in all patients with MSI-H and advanced solid tumors, “leading to a surge in MSI testing of metastatic cancers,” Dr Stadler said.
Lynch Syndrome
Lynch syndrome is caused by germline mutations in the DNA mismatch repair genes MLH1, MSH2, MSH6, PMS2, and EPCAM. This syndrome is responsible for approximately 3% of all colorectal cancers and 6% of endometrial cancers.
“Because of this relatively high prevalence, universal tumor screening of all colon and endometrial cancers is now recommended for markers of Lynch syndrome, and this is usually done either via MSI analysis or immunohistochemical [IHC] staining for the DNA MMR proteins,” Dr Stadler said. In addition, a second immunotherapy, nivolumab (Opdivo) was also approved in 2017 for use in patients with dMMR advanced colorectal cancer.
Patients with tumors suggestive of Lynch syndrome should therefore have counseling and genetic testing for the syndrome. Patients diagnosed with Lynch syndrome would benefit from enhanced cancer surveillance and possibly risk-reducing surgery, such as removal of the ovaries and uterus for the prevention of ovarian and endometrial cancers.
Prevalence of MSI-H and dMMR Mutations
Dr Stadler’s group studied the prevalence of germline mutations in the DNA MMR genes diagnostic of Lynch syndrome across all MSI-H tumors, regardless of cancer type. The study included 15,045 tumors representing more than 50 cancer types. Tumors were analyzed by MSK-IMPACT, an FDA-approved next-generation sequencing platform that includes >400 cancer-associated genes using tumor-normal samples. MSK-IMPACT also incorporates an MSI stratification algorithm that stratifies tumors as being MSI-H, MSI-indeterminate (MSI-I), or microsatellite stable (MSS).
As expected, the highest level of MSI-H was seen in small bowel cancer (25%), followed by endometrial (16%), colorectal (14%), and gastric (6%) cancers. High-frequency MSI was also seen in other solid tumors.
Overall, 16.3% of MSI-H tumors harbored Lynch syndrome versus 1.9% of MSI-I and 0.3% of MSS tumors. Additional tumor evaluation that included IHC staining for dMMR genes and tumor signatures in patients with Lynch syndrome corroborated the finding that MSI-H and MSI-I tumors were caused by Lynch syndrome.
Patients with Lynch syndrome who had MSS tumors had tumor signatures suggesting that the Lynch syndrome did not cause these cancers. The prevalence of Lynch syndrome in the MSS cohort was 0.3%, a percentage that is equivalent to the prevalence of Lynch syndrome in the general at-large population, noted Dr Stadler.
Lynch Syndrome Linked to Many Cancer Types
The study findings support the conclusion that the spectrum of cancers associated with Lynch syndrome is much broader than previously thought, because 50% of patients with Lynch syndrome and MSI-H or MSI-I tumors had cancers other than colorectal and endometrial.
“Lynch syndrome was recognized in tumors rarely or not previously associated with Lynch, including prostate, sarcoma, mesothelioma, adrenocortical carcinoma, and even in young patients with ovarian germ-cell tumors,” she said. “Forty-five percent of these noncolon, nonendometrial tumors did not meet clinical testing criteria for Lynch syndrome and would not have otherwise had Lynch syndrome testing.”
The findings also suggest that an MSI-H or dMMR tumor signature, regardless of cancer type or family history, should prompt genetic testing to evaluate for Lynch syndrome.