Abstract
Background: Among women with invasive breast cancer, the clinical significance of BRCA1 and BRCA2 variants of unknown significance (VUS) remain uncertain. A closer analysis of the somatic mutational profiles of women with invasive breast cancer and BRCA1 or 2 VUS could provide a better understanding of the pathogenesis of the disease and potentially allow for the development of targeted treatment regimens. We hypothesized that breast cancers with BRCA1/2 VUS will have different somatic genomic profiling patterns and tumor mutational burden (TMB) compared with BRCA1/2 wildtype (WT).
Methods: We compared the tumor molecular profiling patterns for breast cancer with somatic BRCA1/2 VUS to BRCA1/2 WT using a multiplatform profiling service (Caris Life Sciences) consisting of next generation sequencing (NGS), protein expression (immunohistochemistry [IHC]) and gene amplification (fluorescence or chromogenic in situ hybridization [FISH or CISH]). We also determined the differences in TMB between VUS and WT. We defined ER/PR/Her2 status based on IHC and/or CISH results. DNA or mutation results with >1% frequency were analyzed using Chi Square test; p<0.05 was considered statistically significant.
Results: Among 4,172 breast cancer specimens available in the Caris database acquired from more 800 participating institutions between 2013 and 2018, 125 (3%) and 341 (5.5%) had VUS in BRCA1 and BRCA2 respectively. Women with BRCA1 VUS were more likely to have tumors that are hormone receptor (HR) negative, Her2/neu positive (6.1%) compared to triple negative, triple positive or HR positive, Her2 negative (3.6%, 3.6% and 2.3% respectively) (p-value=0.0096) however BRCA2 VUS were more evenly distributed among tumor subtypes. Tumors with either BRCA1 or BRCA2 VUS were more likely to have a high TMB compared to WT (mean TMB 10.3 mut/Mb vs. 7.5 mut/Mb, p=7.2 × 10−7). Compared to WT tumors, tumors with BRCA1 or BRCA2 VUS were more likely to have pathogenic mutations in RB1 (6.4% vs 3.9%) and TP53 (62.5% vs 56%) and less likely to have GATA3 mutations (5.9% vs 8.8%) and PIK3CA amplifications (0.3% vs 35.3 %%). Furthermore, BRCA1/2 VUS cancers were 3 times more likely to have FGFR2 amplification (P=0.0065). There was no difference in AR or PD-L1 expression between VUS and WT, however breast cancer associated with BRCA1/2 VUS had a significantly less WT1 amplification compared to WT (0.3% vs 72.9%, p value <0.001).
Conclusion: Our results suggest that breast cancer associated with BRCA 1 or 2 VUS may be phenotypically and genetically different than WT. Our findings suggest the potential for different responses to immune and targeted therapies, which should be investigated in future prospective studies.