Increased risk of hypertrophic scarring in estrogen receptor-positive breast cancer: a bidirectional mendelian...

Breast cancer (BC) is a prevalent malignancy in women, often necessitating tumor resection and breast reconstruction surgeries. However, the post-operation scars can be of concern, as hypertrophic scars (HS) can profoundly impact patients' quality of life. Our study used the bidirectional Mendelian randomization (MR) method to explore the potential relationship between BC and HS. We conducted a comprehensive genetic analysis using a large meta-analysis dataset comprising the BCAC, FinnGen, MRC-IEU, and GTEx eQTL datasets. We identified independent genetic loci associated with BC and HS among individuals of European ancestry. These loci served as instrumental variables (SNPs) in MR analysis. We used the inverse-variance weighted and weighted median methods, with the odds ratio (OR) as the effect measure, to investigate the causal relationship between breast disease and HS, and all significance was corrected by FDR and Bonferroni. We assessed heterogeneity and horizontal pleiotropy through Cochran's Q, MR-PRESSO, and MR-Egger intercept tests. Moreover, a leave-one-out sensitivity analysis was performed to evaluate the impact of individual SNPs on the MR study. The MR analysis showed a significant positive association between BC and HS (OR = 1.31, 95% CI 1.10-1.56, P = 0.002). However, there was no significant association between benign breast neoplasm and HS (OR = 0.97, P = 0.73). Subgroup analysis of BC revealed that estrogen receptor-positive (ER⁺) BC increased the risk of HS (OR = 1.32, 95% confidence interval: 1.11-1.57, P = 0.0016), while ER-negative (ER^-) BC did not increase the risk of HS (OR = 0.99, P = 0.91). Analysis of the GTEx database demonstrated that increased ER level alone in healthy individuals (n = 3301) did not increase the risk of HS (P = 0.42). This result was well replicated in two independent HS datasets, and the causal effect of BC on HS was unidirectional. BC is positively associated with the incidence of HS, especially in the case of ER⁺BC. However, there is a lack of substantial evidence regarding the relationship between ER^-BC and HS.