Lynch syndrome is common among cancer patients with tumours with high microsatellite instability

A large genomic study of more than 15,000 tumour samples showed that people with tumours that have high microsatellite instability (MSI-H) are more likely to have Lynch syndrome, a hereditary condition that increases a person’s risk of developing many different types of cancer. In fact, among patients with MSI-H tumours, 16% were subsequently found to have Lynch syndrome. These findings suggest that all patients with MSI-H tumours should be tested for Lynch syndrome, regardless of cancer type or family or personal history of cancer. Diagnosing Lynch syndrome gives the opportunity to help the cancer patient, but also at-risk family members, as their cancer risk can be lowered through increased cancer surveillance and preventive surgery.

It is estimated that about 1 in 300 (0.3%) people in the general population has Lynch syndrome. The most common cancers associated with Lynch syndrome are colorectal and endometrial cancers, but people with Lynch syndrome also have a higher risk of developing other gastrointestinal, ovarian, brain, and skin cancers. The hallmark of Lynch syndrome-associated tumours is MSI-H. MSI is a genomic marker that indicates a defect in a cell’s ability to repair damaged DNA, resulting in the accumulation of mutations. Traditionally, MSI testing has been performed on colorectal and endometrial cancers as an initial screening test to identify those patients who may be at risk for having Lynch syndrome. Since the FDA approved the immunotherapy pembrolizumab in 2017 for use in all MSI-H tumours, regardless of tumour type, MSI testing of tumours has become broadly used to identify patients who may benefit from pembrolizumab.

In the study at hand, over 15.000 tumour samples were collected from patients with more than 50 different types of advanced cancer and were evaluated with a comprehensive genomic test called MSK-IMPACT. This test uses next-generation sequencing (NGS) to look for mutations in hundreds of cancer-related genes, as well as other molecular changes, including MSI. In addition to this, blood samples from study participants were tested for inherited mutations in genes involved in DNA repair and who are known to cause Lynch syndrome (i.e. MLH1, MSH2, MSH6, PMS2 and EPCAM).

Based on the results of the genomic analysis, the tumour samples were classified into 3 groups: MSI-stable (MSS, no MSI instability found), MSI-indeterminate (MSI-I, moderate level of MSI), and MSI-H. The vast majority (93.2%) of tumours were found to be MSS; 4.6% were MSI-I and 2.2% were found to be MSI-H. Inherited mutations in Lynch syndrome-associated genes were found in 16% of people with MSI-H tumours, compared to 1.9% of those with MSI-I tumours and only 0.3% of those with MSS tumours (p<0.0001). As expected, about 25% of the 1,025 MSI-H/MSI-I tumours were colorectal or endometrial cancers. However, nearly 50% of patients with MSI-H/MSI-I tumours who were identified as having Lynch syndrome had cancer types not previously, or rarely, linked to the syndrome. This included mesothelioma, sarcoma, adrenocortical cancer, melanoma, prostate, and ovarian germ cell cancer. Of these patients, 45% did not meet Lynch syndrome genetic testing criteria based on family or personal cancer history. This suggests that Lynch syndrome is linked to a broader spectrum of cancers than previously thought and indicates that MSI-H/MMR-D is predictive of Lynch syndrome, regardless of the cancer type. In the final step of the study, 57 MSI-I/MSI-H tumour samples were also tested for abnormal DNA repair proteins. MMR deficiency was found in nearly all (98.3%) of those tumours.

In summary, these findings suggest that if either MSI-H or MMR-D is found in the tumour, hereditary genetic testing for Lynch syndrome should be performed. This is important as the chance of developing certain cancers linked to Lynch syndrome can be lowered through frequent screening (e.g., yearly colonoscopy and endoscopy for gastrointestinal cancers) and preventive surgery (e.g., removal of the uterus and ovaries for gynaecologic cancers).

Reference

Latham Schwark et al. Pan-cancer Microsatellite Instability Predicts for Presence of Lynch Syndrome. Presented at ASCO 2018; Abstract LBA1509.