CHICAGO (GenomeWeb) – The need for more genetic diversity in large genomic datasets has long been a topic of discussion at research and clinical conferences. At the annual conference of the American Society of Clinical Oncology in Chicago this week, Boston University researcher Julie Palmer discussed results of a study she and her collaborators from institutions around the US had done to try to identify what risk cancer predisposition genes confer for breast cancer in the African-American population compared to the non-Hispanic white population.
Although clinical genetic testing has led to the identification of pathogenic mutations in breast cancer susceptibility genes, this knowledge has not helped to establish the frequency of mutations and the risks of cancer associated with breast cancer predisposition genes in the African-American population, Palmer said.
As part of the CAnceR RIsk Estimates Related to Susceptibility (CARRIERS) study, she and her colleagues tested germline DNA samples from African-American women — 5,054 breast cancer cases and 4,993 age-matched unaffected controls — from 10 US studies for mutations in 20 established breast cancer predisposition genes using a Qiagen QIAseq multiplex amplicon panel. They evaluated the frequency of mutations in each gene and analyzed the associations between the mutations and breast cancer risk, after adjusting for study design, age, and first-degree family history of breast cancer.
Palmer reported that 18.2 percent of cases and 10.8 percent of controls in the overall study reported a first-degree family history of breast cancer. Further, the researchers identified pathogenic mutations in any of the 20 breast cancer predisposition genes in 7.6 percent of breast cancer cases and 2.4 percent of controls. Mutations in BRCA1, BRCA2, and PALB2 were associated with high risk of breast cancer, while mutations in CHEK2 were associated with moderate breast cancer risk and mutations in ATM had lower clinical relevance. Mutations in BRCA1, BRCA2, PALB2, and RAD51D, were associated with increased risks of estrogen receptor-negative breast cancer.
Overall, the team concluded that cancer predisposition genes confer similar risks of breast cancer in the African-American population as in non-Hispanic whites.
In discussing Palmer's results, Memorial Sloan Kettering Cancer Center oncologist Mark Robson said the study has implications for how individuals are referred to counseling and testing for breast cancer based on self-identified race or ethnicity (SIRE) and for the possible design of differential treatment frameworks based on SIRE.
First, he said, it doesn't appear that there's a dramatic difference in mutation prevalence between SIRE groups. And although Palmer's data suggested that there was an increased prevalence of BRCA1 in self-reported African-American women, that was insufficient to explain the elevated risk for triple-negative breast cancer (TNBC) in those women.
Robson identified several ways a pathogenic variant could become enriched in a SIRE group. First is the founder effect — the loss of genetic variation that occurs when a new population is established by a very small number of individuals from a larger population. If a mutation arises in such a population, and if the population were to then go through a bottleneck event that doesn't affect reproductive fitness, that mutation might become enriched.
However, Robson said, SIRE populations, especially in the US, are genomically diverse. The transatlantic slave trade brought people from many different African populations to the Americas where they eventually mixed with each other, and with Native American, Hispanic, and Caucasian populations. Even within self-identified ethnicity groups, he added, there is a wide range of genomic variation that correlates with geographical and admixture variation.
However, Robson noted, pathogenic variants can also become enriched because of improved population fitness and an increased de novo mutation rate. He added that ascertainment bias — which is the apparent increase of pathogenic variants in a given population due to differential referral or acceptance of testing — can skew pathogenicity results in studies even though the real picture may be somewhat different. In the case of TNBC, for example, younger women with the disease are more likely to get referred for testing and to accept it than older women, which could bias any subsequent study results about pathogenicity, Robson said.
Palmer later clarified there was no ascertainment bias in her study because it included data from 10 different population studies, which included samples from women of all ages, but she did acknowledge Robson's point that lack of diversity — even age-based diversity — in databases remains a huge problem.
Robson concluded that there could be as-yet unobserved biological differences between African-American and Caucasian populations that could explain the difference in rates of TNBC between these populations while also explaining the similar penetrance of pathogenic variants. The observed increase of TNBC in African Americans is likely to be polygenic, he said. The penetrance of BRCA1 and BRCA2 is known to be modified by SNPs that influence risk in the general population, but there's enormous genomic diversity in African populations that has yet to be explored, and therefore there are risk modifiers that have yet to be found.
Robson also noted that variants of unknown significance are more common in populations other than Caucasians. African populations are generally much older than European populations in terms of population genetics, he added, and therefore they've had more time to accumulate mutations and gene modifiers. If there were better population studies done on these groups, there could be potential to gather data to build polygenic risk scores for TNBC in African American women, he said.
Importantly, however, Robson concluded that we don't yet know enough about the prevalence of breast cancer risk genes in each SIRE to recommend individualized testing and treatment regimens based on those group identifiers. Individuals should generally be referred for testing and counseling without regard to SIRE, he said. And geographic origin should only influence the threshold for mutation testing if the individual in question is from a region where founder mutations are known to be a factor for a given disease.
He also concluded that at the present time, there is insufficient evidence to recommend the differential treatment of women with pathogenic variants based solely on SIRE.