NEW YORK (GenomeWeb) – A team led by researchers at Ambry Genetics has conducted a large-scale exome sequencing study to confirm reported predisposition genes for breast and ovarian cancer, as well as identify new genetic mutations associated with these diseases.
As the researchers noted today in JAMA Oncology, multiple high- and moderate-penetrance genes have been reported as risk factors for hereditary breast cancer, ovarian cancer, or both since the discovery of BRCA1 and BRCA2, but it is still unclear whether the complete genetic map of these cancers has been drawn.
They began with a sample set from 11,416 patients with clinical features of breast cancer, ovarian cancer, or both who were referred for genetic testing from 1,200 hospitals and clinics across the US between 2014 and 2015. They also analyzed 3,988 controls who were referred for genetic testing for non-cancerous conditions such as cystic fibrosis or rare inherited cardiovascular conditions. The researchers performed whole-exome sequencing and examined the gene-phenotype associations, and conducted case-control analyses as a set of reference controls using the Genome Aggregation Database.
After cleaning and filtering the data, the team performed burden tests for 625 cancer genes among 9,639 breast cancer and 2,051 ovarian cancer cases compared to the 3,988 controls. The investigators also examined the distribution of pathogenic variants in 11 characterized breast cancer or ovarian cancer genes. They noted that the cases in the study were enriched for non-carriers of the deleterious BRCA1/2 mutations, which resulted in underestimation of risks for BRCA1/2 genes.
Overall, they identified four genes that confer significantly increased breast cancer risk — ATM, CHEK2, MSH6, and PALB2. The association with the Lynch syndrome susceptibility gene MSH6 is newly identified, the researchers said. Those findings were supported by comparisons with the gnomAD database as a reference population, in which all four genes were present and significantly associated with similar or higher estimated risks. PALB2 was associated with high breast cancer risk, whereas ATM and CHEK2 conferred two- to three-fold risks.
In contract, the researchers said they observed no significant breast cancer associations for BARD1, NF1, or PTEN. They also determined that the MMR genes MLH1, MSH2, and PMS2; the MRN complex genes MRE11A, RAD50, and NBN; the ovarian cancer susceptibility genes BRIP1, RAD51C, and RAD51D; and the melanoma predisposition gene CDKN2A did not significantly increase breast cancer risk.
In the case of ovarian cancer, the researchers found that four genes were significantly associated with increased disease risk — RAD51C, TP53, MSH6, and ATM. The associations between ovarian cancer risk and RAD51C and TP53, which had previously been studied, were confirmed. RAD51C was associated with high risk, and the researchers observed moderately increased ovarian cancer risk with MSH6 and ATM. However, they also noted no significant elevated risk associated with BRIP1, RAD51D, CDKN2A, or the MRN complex genes MRE11A, RAD50, and NBN.
"Owing to the lack of significantly powered studies, the number and selection of risk genes for hereditary cancer testing have not been standardized, and robust evidence for association with breast cancer or ovarian cancer risk is only available for a modest set of characterized genes typically included in most panels," the researchers concluded. "This is particularly the case for moderate-penetrance genes, for which there exists either conflicting evidence or insufficient data for reliable estimation."
They also noted that their study had some limitations, including incomplete information on the personal disease history of the in-laboratory controls. The team also based its analysis on the aggregation of both protein-truncating and known pathogenic variants, which could result in underestimation of risk for less-studied genes.
However, the authors added, their study showed that this approach could provide useful information for cancer genetic testing and could broaden the range of cancer risk genetic associations.
In an accompanying commentary in JAMA Oncology today, researchers from University Hospital Southampton and University Hospital Southampton NHS Foundation Trust in the UK wrote that the study's strengths include its large sample size, the analysis of accurate NGS data designed for clinical diagnostic use, and confirmation of suspected variants with Sanger sequencing.
"The PALB2-, CHEK2-, and ATM-associated breast cancer risks shown by Lu and colleagues are consistent with the published literature. Likewise, the ovarian cancer risks associated with MSH6 and RAD51C are well described," the authors noted. "By contrast, their reported MSH6-associated breast cancer risk and ATM-associated ovarian cancer risk are novel, not reflected in the current literature. Their data also challenge some previously described associations."
However, they added, despite the study's demonstration of the potential of genomics to expand on our current understanding of cancer susceptibility genes, any diagnostic genetic test aimed at informing clinical decision-making needs to have demonstrated clinical utility before it can be used for such a purpose. That includes the prediction of future cancer risks among healthy relatives in families with hereditary breast or ovarian cancer.
"Although there is accumulating evidence to support genomic testing for hereditary breast or ovarian cancer, it is important to recognize that cancer risks conferred by constitutional genotype are modifiable by polygenic factors and environmental exposures," they concluded. "As such, genotype should be viewed as an important tool for cancer risk stratification that may not be fully representative of the entire complement of risk factors that an individual may possess."