Background
The aim of this study was to investigate the correlation of BRCA1 methylation status to clinicopathological and molecular variables and outcome in patients with triple-negative breast cancer (TNBC).
Methods
145 patients enrolled 2015-2018 in the Sweden Cancerome Analysis Network - Breast (SCAN-B) study (NCT02306096) were included. 109 had chemotherapy (>90% with sequential epirubicin + cyclophosphamide and a taxane, EC–T) according to national guidelines either as neoadjuvant (32%, NAC) or adjuvant (68%, ACT) therapy. Germline screening was performed in 54 patients with pathogenic BRCA1/2 variants in 11 cases. Microsatellite instability (MSI) and BRCA1 promoter methylation (BRCA1met) using pyrosequencing were investigated in all cases, including analyses of pre-treatment biopsies and surgical resections for NAC patients with residual disease (RD). Clinicopathological and molecular variables were obtained through clinical review and complementary RNA-seq data for all patients and tumor specimens.
Results
MSI and BRCA1met was observed in 2% and 17% of patients, respectively. BRCA1met was correlated to younger age and the PAM50 Basal phenotype (92%) and was frequent (40%) in women without pathogenic BRCA1/2 germline variants. No association was observed between BRCA1met and distant recurrence-free interval in adjuvant EC-T treated patients (log-rank p=0.77), nor was there an association between BRCA1met and pathological response (pCR) in NAC patients (p=0.69). In-depth analyses of BRCA1met NAC cases revealed differences in methylation level and BRCA1 mRNA expression in pre-treatment versus surgically resected tissue in patients with RD that will be further analyzed using WGS.
Conclusions
BRCA1met is associated with young age in TNBC but hold no predictive nor prognostic value in this patient cohort. We observe a difference in BRCA1met patterns in NAC patients with RD that may potentially be related to treatment resistance.
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.