In a registry-based study, 3.8% of patients with pancreatic cancer had germline mutations in genes known to significantly increase cancer risk, researchers reported in the March issue of Gastroenterology.
“A small but clinically important proportion of pancreatic cancer is associated with mutations in known predisposition genes. The heterogeneity of mutations identified in this study shows the value of using a multiple-gene panel in pancreatic cancer,” said Robert Grant of the Ontario Institute for Cancer Research in Toronto and his associates.
Source: American Gastroenterological Association
Some patients had mutations in the BRCA and MMR genes, about which enough is known to help guide preventive measures and treatment decisions, the researchers noted. “Furthermore, all mutation carriers represent a high-risk subgroup of patients for pancreatic cancer researchers to consider for screening studies and for experimental targeted therapies,” the investigators wrote (Gastroenterology 2014 Dec. 2 [doi:10.1053/j.gastro.2014.11.042]).
For the study, the investigators carried out multigene next-generation sequencing of DNA from the blood and saliva of 290 patients with pancreatic cancer. The patients were randomly chosen from a population-based pancreatic cancer registry in Ontario. The researchers looked for mutations in 13 genes that have been previously linked to cancers: APC, ATM, BRCA1, BRCA2, CDKN2A, MLH1, MSH2, MSH6, PALB2, PMS2, PRSS1, STK11, and TP53. They found a total of 11 pathogenic germline mutations in seven of these genes, including 3 in the ATM gene, 1 in BRCA1,2 in BRCA2, 1 in MLH1, 2 in MSH2, 1 in MSH6, and 1 in TP53, they reported.
A total of 3.8% (95% confidence interval, 2.1%-5.6%) of the patients carried mutations, or about 1 in every 26 patients, said the researchers. Most mutations were linked to breast or colorectal cancer. Mutation carriers were significantly more likely to have been diagnosed with breast cancer or colorectal cancer themselves, or to have an affected first-degree relative (P < .01 for both), compared with other patients in the study.
Furthermore, the prevalence of mutations jumped to 10.7% (95% CI, 4.4%-11.1%) when the researchers looked only at patients with a personal or family history of breast cancer, and to 11.1% (95% CI, 3.0%-19.1%) in the subgroup of patients with a personal or family history of colorectal cancer, the investigators reported.
But mutation status was not linked with familial pancreatic cancer, nor with age at diagnosis, which “raises questions with important clinical implications,” said the researchers. “With decreasing sequencing costs, which patients should be referred for multiple-gene panel sequencing? What is the relevance of mutations in [patients] with atypical personal and family histories?”
Answering those questions would require studying many genes from large groups of patients with pancreatic cancer, they emphasized. For now, although multiple-gene panels cost about the same amount as single-gene tests, “the added value of simultaneously sequencing many genes remains uncertain, and likely depends on the clinical setting,” they said. Nonetheless, the analysis uncovered so many mutations in the cohort that the researchers would recommend multiple-gene testing of patients with pancreatic cancer, they said.
Patients were recruited into the registry based on pathology results, which would have excluded those with advanced pancreatic cancer who never had pathology testing, said the investigators. Also, study patients tended to be younger, to have been treated in academic centers, and to have resectable disease more often than other patients in Ontario with pancreatic cancer.
The work was funded by the Weston Garfield Foundation, Pancreatic Cancer Canada, National Cancer Institute, Mount Sinai Hospital Biospecimen Repository, and Teresa Bianco. The investigators declared having no conflicts of interest.