NEW YORK (GenomeWeb) – Analyzing hereditary cancer gene panel test results for the first 10,000 patients referred to its laboratory, researchers at GeneDx and their collaborators have found that almost one in 10 patients harbored either a pathogenic or a likely pathogenic variant.
The high frequency of positive test results in a variety of cancer genes "underscores both the genetic heterogeneity of hereditary cancer and the usefulness of multigene panels over genetic tests of one or two genes," the researchers, led by Wendy Chung at the Department of Pediatrics and Medicine at Columbia University Medical Center, noted in an online publication of their study in Genetics in Medicine this week.
For their analysis, the researchers looked at the first 10,030 consecutive patients referred to GeneDx, a subsidiary of Opko Health, for germline hereditary cancer panel testing by next-generation sequencing between August 2013 and October 2014. Three quarters of patients suffered from cancer, more than half of them breast cancer, with other cancer types including ovarian, colorectal, endometrial, pancreatic, and stomach cancer. More than 60 percent of women with breast cancer in the study did not previously have BRCA1/2 testing.
GeneDx performed one of eight different multigene cancer panels on patients' germline samples. The panels — a comprehensive panel, a breast/ovarian, a high/moderate risk, a colorectal, a pancreatic, an endometrical, a Lynch syndrome/colorectal high risk, and a breast cancer high risk panel — were made up of combinations of a total of 29 genes that have been associated with hereditary cancer. They include 17 high-risk genes, such as BRCA1 and BRCA2; three moderate-risk genes, ATM, CHEK2, and PALB2; and nine genes with increased but less well-defined cancer risk, such as CDK4 and RAD51C.
Following the 2007 guidelines from the American College of Medical Genetics and Genomics, variants found in those genes were classified as pathogenic, likely pathogenic, variants of unknown significance (VUS), likely benign, or benign, and reports with pathogenic or likely pathogenic variants were considered positive test results. About 200 variants underwent reclassification during the course of the study, the authors noted, 90 percent of them from "uncertain" to "likely benign."
Overall, 901 patients, or 9 percent, carried at least one pathogenic or likely pathogenic variant in a tested gene. About 90 percent of the 937 variants were pathogenic, the remainder likely pathogenic.
About half of these positive results were in well-established high-risk cancer genes, the other half in more recently described genes with moderate or unknown risk. "Positive results in genes with moderate or unknown risk can present quandaries for providers given the lack of established guidelines for medical management," the authors noted. "Furthermore, it often is not clear whether the identified variant in these genes is the sole genetic cause of the cancer in the family."
VUS rates differed depending on the number of genes included on a panel and the patient's ancestry, with the largest VUS frequency associated with the 29-gene comprehensive panel and patients with Asian or African-American ancestry, and the lowest VUS rate with the six-gene high-risk breast cancer panel and patients with Hispanic or Caucasian ancestry.
Patients with colon or stomach cancer had the most positive test results, a total of almost 15 percent, followed by about 13 percent of ovarian cancer patients and almost 10 percent of female breast cancer patients, neither of which previously had BRCA1/2 testing.
About three percent of patients with positive test results had more than one pathogenic or likely pathogenic variant.
Interestingly, a significant number of positive test results were in genes not typically associated with the cancer of the patient. For example, almost 6 percent of positive results in women with breast cancer were in highly penetrant genes other than BRCA1/2; more than a quarter of colon cancer patients tested positive for genes not considered to be classical colorectal cancer genes, such as BRCA1/2; and more than 11 percent of endometrial cancer patients with positive results had BRCA1/2 mutations.
"Together, these findings illustrate the utility of large panels in identifying pathogenic variants in high-risk genes that might not have been considered based only on a patient's personal cancer history," the authors wrote.
"Our experience demonstrates that multigene panels have the potential to identify pathogenic variants in genes that would not typically have been tested, both in high-risk genes not associated with a patient's personal cancer history and in genes with moderate or less well-defined penetrance which, before the availability of multigene panels, were not often tested," they concluded. "The high frequency of positive findings in these more recently identified cancer genes underscores their potential contribution to hereditary cancer risk and evolving impact on medical management."
The researchers noted that not all patients and their doctors are comfortable dealing with uncertain test results, and that large gene panels that include genes with unknown risks "are not appropriate in all clinical contexts or for all patients."
On the other hand, many patients "would like to have the information about these genes so that as more information becomes available they are prepared to act immediately," they wrote.