NEW YORK (GenomeWeb) – Broad gene panel testing can help identify the hereditary causes of early-onset colorectal cancer patients and could lead to changes in those patients' management as well as have major implications for their relatives, according to researchers from Ohio State University.
Reporting today in JAMA Oncology, the researchers found that 16 percent of 450 patients under the age of 50 with colorectal cancer carried a hereditary cancer mutation. About one-third of those patients did not meet testing criteria for the gene in which they had a mutation. For instance, some patients had mutations in genes associated with other cancer types, like breast, ovarian, and even melanoma, the results of which could have important implications not only for their management but for their relatives.
"One out of six early-onset colorectal cancer patients has a hereditary susceptibility," Heather Hampel, a genetic counselor at Ohio State's Comprehensive Cancer Center, told GenomeWeb. "The fact that it's that common says that every early-onset colorectal cancer patient should consider having a broad genetic panel test."
The study was done as a subset of a statewide colorectal cancer study that Hampel is leading called the Ohio Colorectal Cancer Prevention Initiative, which has so far recruited more than 3,000 patients between the ages of 17 and 96 and involves researchers and clinicians at 50 hospitals in Ohio. The goal of the Ohio Colorectal Cancer Prevention Initiative is to reduce colorectal cancer incidence by identifying individuals with a hereditary predisposition.
The subset published in today's JAMA Oncology study looked at 450 patients who were diagnosed earlier than the age of 50. Early-onset is a "red flag" that cancer is hereditary, Hampel said. But, the most well-known hereditary cause of colorectal cancer, Lynch syndrome, only accounts for between 4 percent and 13.5 percent of patients with early-onset disease.
So, in order to see if other known hereditary cancer genes were involved, the researchers analyzed 25 genes on next-generation sequencing panels that included both the Lynch syndrome genes and also genes known to impact other hereditary cancers.
In the study, all patients were first screened for germline defects in the mismatch repair (MMR) genes, which cause Lynch syndrome, using immunohistochemistry and microsatellite instability testing. The researchers had funding to then run an NGS assay developed by the University of Washington on the patients whose MMR genes were deficient, to see whether they had genetic mutations in those genes and also to look at other hereditary cancer mutations. Myriad Genetics donated testing for the subset of patients whose MMR genes seemed proficient, running its 25-gene MyRisk hereditary cancer panel.
Of the 450 early-onset patients, 48 had MMR deficiency. Of those, 37, or just over 8 percent of the total, had Lynch syndrome. Of the 402 patients without MMR deficiency, 32, or around 8 percent, had a mutation in a hereditary cancer susceptibility gene. Nine patients had mutations in high-penetrance genes conferring risk of colorectal cancer. Thirteen patients had a mutation in a high- or moderate-penetrance gene not traditionally associated with colorectal cancer. Those genes included BRCA1, BRCA2, PALB2, ATM, CHEK2, and CDKN2A. Ten patients had a mutation in a low-penetrance gene associated with colorectal cancer.
"We had a sense that Lynch syndrome would be common in these patients, but confirming that was important," she said. In addition, identifying other hereditary cancer risk genes, including those related not only to colon cancer but other cancers was important as well.
The findings will impact management for those patients, Hampel said. Patients with hereditary colon cancer mutations will now get colonoscopies more frequently. In addition, her team has begun to offer genetic testing to the family members of patients with germline mutations. Hampel said that, on average, about six relatives are tested per patient with a hereditary cancer mutation and three test positive.
"The implications are huge," she said. For instance, six patients had BRCA1 and/or BRCA2 mutations, Hampel said. Relatives of that patient can now get tested and results can make a huge difference, particularly for young female relatives who may have not realized they were at an increased risk for breast or ovarian cancer.
Hampel said that the researchers are also interested in studying the relationship between BRCA1 and BRCA2 and colorectal cancer. Those genes have not previously been associated with an increased risk in colorectal cancer, but because more patients than would have been expected carried germline mutations in those genes, it raises the question of whether mutations in those genes could lead to an increased risk.
Hampel said that this study does not provide enough evidence to make that conclusion. "It provides a little hint," she said, "but we'll need to look at it again."
Such a study would require a much larger cohort of patients, she said. Because breast and colorectal cancer are relatively common, by chance there will be some families who have both, so proving that the BRCA mutations are leading to an increased risk in colorectal cancer will be challenging, she said.
The researchers are also following up on the variants of unknown significance. In this study, they identified 178 VUS in 145 patients. Already, five of those variants have been upgraded to pathogenic following additional testing. For example, Hampel said that the researchers identified the same VUS in two patients from the same area in Ohio. Further investigation on Ancestry.com found that the two individuals were actually fourth cousins. Segregation analysis found that the VUS tracked with the family, and the researchers were able to reclassify that VUS as likely pathogenic.
"We've been working to get these VUS solved," Hampel said.
Aside from segregation analysis in related individuals, the researchers are also using tumor sequencing and RNA analysis to downgrade or upgrade the unknown variants. Hampel anticipated that between 90 percent and 95 percent of the VUS would ultimately be benign, but that between 5 percent and 10 percent could be reclassified as likely pathogenic or pathogenic. That would be enough to increase the percentage of early-onset patients harboring a hereditary risk mutation to as much as 20 percent from 16 percent, she said.