304P - Generalizability of 313-SNP PRS for breast cancer risk in non-European ancestries

Background

Polygenic risk scores (PRS), which summarize the combined effect of multiple common risk variants, have been associated with elevated breast cancer risk in patients who lack any identifiable monogenic risk factors. One of the most well-validated PRSs in breast cancer to date is PRS313, which was developed from a Northern European biobank (n = 33,673 cases and n = 33,381 controls). However, subsequent work has demonstrated an attenuated effect of PRS313 in African females, raising generalizability concerns. In this work, we further investigate the generalizability of the PRS313 for American women of European (EA), African (AFR), Asian (EAA), and Latinx (HL) ancestry within one institution, leveraging a singular EHR system, genotyping platform, and quality control process.

Methods

Participants were females drawn from the UCLA ATLAS Biobank (N=18,627) from 2018-2023. SNPs were genotyped on a genome-wide array and imputed to the TOPmed reference panel. Breast cancer cases and controls were identified using ICD9 and ICD10 coding. We previously identified 5 Genetically Inferred Ancestries (GIA) based on Principal Component Analysis (PCA) and k-means clustering including African Americans (AA), Hispanic Latino Americans (HLA), East Asian Americans (EAA), European Americans (EA). Breast cancer subtypes were identified based on prescription patterns.

Results

We found that the PRS313 achieved overlapping AUCs in females of Lantix (AUC, 0.68; 95% CI, 0.65-0.71) and European ancestry (AUC, 0.70; 95% CI, 0.69-0.71) with lower AUCs for the AFR and EAA populations (AFR: AUC, 0.61; 95% CI, 0.56-0.65; EAA: AUC, 0.64; 95% CI, 0.60-0.680).

Conclusions

In summary, we found that PRS313 was significantly associated with breast cancer in American females of Hispanic and East Asian ancestry but with attenuated accuracy in women of African descent. Our results are consistent with prior studies on PRS313 and further highlight the need for ethnically diverse cohorts in developing and implementing polygenic risk scores.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.