Skip to main content

Conference Program

Subpage Hero

Loading

Disparities in Variants of Unknown Significance Identification in Underrepresented Populations Undergoing Outpatient Genetic Testing at UMass Memorial Healthcare

Clinical Genetics and Therapeutics
  • Primary Categories:
    • Clinical Genetics
  • Secondary Categories:
    • Clinical Genetics
Introduction:
Next-generation sequencing has increased the availability of genetic testing and may have increased health-care disparity. Traditional genetic variant classification utilizes variant frequencies in a reference population to both identify novel genetic variation and classify variants as “pathogenic” “likely pathogenic,” “benign,” or “variant of uncertain significance” (VUS). However, reference databases were largely comprised of genotype and sequence data from people of Western European descent. Therefore, it is possible that patients of non-Western European descent are more likely to receive less meaningful test results due to bias in the reference genome data that lacks the diversity of the actual population. 

UMass Memorial Healthcare (UMMHC) Division of Genetics has been offering NGS testing services for over 5 years and is uniquely positioned to detect disparity in genetic testing while serving a region with broad diversity. We aimed to characterize genetic testing results, assess the utility of genetic testing in this diverse patient population, and potentially identify any racial and/or ethnic bias in our current standard of care. 

 

Methods:
We conducted a retrospective cohort analysis using demographic information and Invitae genetic testing reports from patients tested for cancer, cardiovascular, and neurodevelopmental phenotypes  seen by UMMHC Division of Genetics providers between 11/1/2017 and 6/1/2022. These patients were identified by electronic medical record review and were grouped by self-reported race/ethnicity. The primary outcome was the number of variants of unknown significance per tested gene. ANOVA and two-sample t-tests were used to compare VUS burden between groups. Secondary analyses focused on pairwise testing and phenotype specific analyses. 

 

Results:
1572 cancer, cardiovascular, and neurodevelopmental patients were identified and were sorted into self-identified White (n=1211), Black (n=44), Asian (n=35), and Other (n=102) population groups. Primary analysis identified a statistically significant difference in the number of VUS per gene in self-identified whites vs all other groups (Mean 0.0074 vs 0.012, t=5.9, p= 3.7 E-9) while there was no difference in number of pathogenic variants per gene (Mean 0.003 vs 0.0018, t= 0.80, P 0.42). Secondary one-way ANOVA for patients with cancer phenotypes revealed a significant difference in the number of VUS per gene between all 4 population groups (p=1.83 E-9). Pairwise analyses showed significant differences in Asian vs Other (p=0.004), Asian vs White (p=3 E-7), Black vs White (p=0.0083), and Other vs White (p=0.0368). T-tests also showed significant differences between whites vs all other groups in VUS per gene in the cardiovascular phenotype (Mean 0.009 vs 0.015, t= 3.3, p=0.001). The neurodevelopmental phenotype analysis did not detect intergroup difference (p=0.7). 

 

Conclusion:
Non-white patients who received Invitae genetic testing at UMMHC between 2017 and 2022 were more likely to receive results that included a VUS than white patients. This highlights the need for continued improvement in the diversity of current reference databases or migration to newer reference databases with broader diversity moving forward. Non-white individuals who have received NGS results have a higher burden of uncertainty, and will be subject to higher rates of result reclassification. It also emphasizes the need for clinicians to have a measure of skepticism when interpreting VUS to individuals. Clinicians should consider how to disseminate reclassified results and consider a mechanism for reviewing and reclassifying results for best clinical utility of NGS results.

 

Agenda

Sponsors