Hereditary cancer testing in an ethnically diverse U.S. population.

Abstract

10579 Background: The clinical utility of hereditary cancer multigene testing is well-established. Most testing has been performed in the White European population, with a relative shortage of data in other ethnic groups. Evaluation of hereditary cancer variants within diverse ethnic populations is important to drive accurate variant interpretation. This study aims to review the outcomes of hereditary cancer testing in an ethnically diverse U.S. population. Methods: We conducted a retrospective analysis of 8,888 test results from a 31-gene hereditary cancer test using next-generation sequencing with copy number variant analysis. Results were separated into seven categories based on the ethnicity selected on the test requisition: African American (AA), Asian (A), Ashkenazi Jewish (AJ), Hispanic (H), White (W), Other (O), and Unknown (U). O was used for any handwritten ethnicity or if multiple ethnicities were selected and U if no ethnicity was chosen. We quantified the number of negative, positive (pathogenic or likely pathogenic (P/LP) variant) and variant of uncertain significant (VUS) results for each ethnic group. We tallied the gene and variant for each result with a positive, negative or VUS. We used a Fisher's Exact test and Bonferroni-adjusted p-values to account for multiple hypothesis testing. Results: Of the 8,888 orders reviewed, 45% were non-White ethnic groups. The P/LP variant rate was lower for AA (4%, p=2e-04) compared to W (9%). The VUS rate was higher for O, AA and A (48%, 56%, and 57%; p=0.0045, 3.1e-17 and 4.8e-06, respectively), compared to 38% in W. Lower negative rates were found in A and AA (37% and 40%; p=0.00024 and 9.4e-10, respectively), and trending lower in O (46%, p=0.0097), compared to 58% for W. The VUS rate for H is 41% (p value = 0.095). Conclusions: Our results suggest that non-Whites may be at a disadvantage when it comes to hereditary cancer multigene testing due to the higher rate of VUS results. The inverse correlation between the overall rates of VUS and negative results when comparing these populations suggests there may be limited variant information for the non-White population in medical literature and available databases. This underrepresentation may make it more difficult to accurately characterize a variant. Despite the small sample size, these findings are consistent with previous publications; however, gene specific outcomes could not be evaluated. This finding suggests that providers could give these patients uninformative results more often, which has potential to impact screening protocols for these families. More research is needed to understand the impact of variant classification across ethnicities to decrease health disparities.