Questions Remain in Identifying Mutations in Non-Lynch Syndrome Patients With Colorectal Cancer

It is unclear which strategy is optimal for determining the prevalence of germline cancer susceptibility gene mutations in patients with colorectal cancer (CRC), according to an analysis presented at 2016 Annual Meeting of the American Society of Clinical Oncology (ASCO), a gathering of over 30,000 oncology professionals in Chicago.

Inherited factors are important determinants of the risk for CRC. However, current estimates of the prevalence of inherited CRC syndromes have been determined by testing for syndrome-specific genes in patients with high-risk features, such as a personal or family history of CRC or the known presence of CRC-associated genes.

The aim of this study, presented by Matthew B. Yurgelun, physician, instructor in Medicine, Harvard Medical School, Dana-Farber Cancer Institute, was to determine the prevalence of pathogenic germline mutations in cancer risk genes in patients with CRC who were not selected for the presence of high-risk indications. Individuals seen at Dana-Farber Cancer Institute for a CRC diagnosis from December 2008 through March 2014 were eligible for the study. Of 1,100 patients screened, 1,059 participated in the study.

Testing was done on blood samples at a CLIA-approved commercial laboratory using a 25-gene hereditary cancer panel that included high-penetrance susceptibility genes (e.g., those associated with Lynch syndrome, adenomatous polyposis syndromes and hereditary breast and ovarian cancer [BRCA1 and BRCA2]), as well as genes of moderate penetrance both linked and not linked to CRC risk.

Data collected from study participants included results of microsatellite instability (MSI) and mismatch-repair protein immunohistochemistry (MMR IHC), which is used to screen for Lynch syndrome. PREMM1,2,6 scores were used to generate a numerical estimate of the likelihood of an underlying Lynch syndrome mutation.

Patients were 56 percent male, with a mean age of 55.7 years at diagnosis. The median PREMM1,2,6 score was 3.93 percent; 39 percent of patients had a PREMM1,2,6 score of at least 5 percent, which is the cut-off value for recommended testing for Lynch syndrome.

Germline testing results revealed 106 patients (10 percent) with at least one pathogenic mutation. Of the 5.2 percent of patients with high-penetrance gene mutations, 3.1 percent had Lynch syndrome. The most common non-Lynch, high-penetrance mutation was BRCA1/2, which occurred in 1 percent of the patient population, which is higher than the 0.2 percent to 0.3 percent estimated prevalence in the general population.

Although this study included a large group of patients who were not selected for high-risk features, it was clinic-based; personal and family history were extensive, but were not verified beyond a review of the medical record. Tumor testing data were incomplete for some patients, and enrollment of many occurred before universal MMR IHC testing. Additional limitations were a lack of data on Ashkenazi ancestry, polyp number and histology.

Multigene germline genetic testing in this study revealed an additional 7 percent inherited non-Lynch genes associated with CRC risk that were not predicted by personal or family history. Many of the high- and moderate-penetrance genes, whether linked to CRC risk or not, have National Comprehensive Cancer Network—recommended screening or risk-reduction interventions. Early identification of these genes would be required to implement these interventions.

Although MSI/MMR IHC and PREMM1,2,6 reliably identify the presence of Lynch syndrome in patients with CRC, Yurgelun concluded that the spectrum of genetic factors in CRC is more diverse than has been appreciated; personal and family history are not reliable predictors for these factors.

A participant at the session mentioned observing an excess of CRC in women below the age of 50 who have BRCA1 or BRCA2 mutations, and wondered if men with those mutations might also have a higher risk of CRC. He asked whether these individuals should be screened for CRC, and if so, how often and at what age.

Yurgelun answered, “It is not a question that I have an obvious or a direct answer to. I agree that data such as ours today and those you mentioned are certainly provocative for how much of an overlap there is between BRCA mutations and CRC risk.”

For what to do for patients with these mutations, he said, “I don’t know that our study necessarily is able to tell us when we should start colonoscopies in individuals with BRCA1 and 2 mutations, and should it be different for men versus women. Is there an actual benefit to screening them earlier with regards to reducing their colorectal cancer risk? I think those are the types of questions that naturally come out of these sorts of data.”