Objective To investigate the causal association between 91 kinds of circulating inflammatory proteins and different subtypes of breast cancer (estrogen receptor-positive and -negative breast cancer) using a two-sample Mendelian randomization (MR) method.

Methods Corresponding exposure and outcome data were extracted from the genome-wide association study database. The data were analyzed by two-sample MR with inverse-variance weighting (IVW) as the primary study method, and MR-Egger, weighted median, simple mode, and weighted mode were used to complement the results. The results were complemented by sensitivity analysis to verify the reliability of the data.

Results The IVW results showed that SULT1A1 (P=0.0007) was associated with an increased risk of total BC, whereas IL-5 (P=0.0011) was associated with a decreased risk of total BC. SULT1A1 (P=0.0011) and CX3CL1 (P=0.0005) were associated with an increased risk of ER+BC, whereas beta-NGF (P=0.0001) was associated with an increased risk of ER−BC. Supplementary analysis methods validated that the findings were consistent in direction and magnitude. The results of the sensitivity analysis showed that the data were reliable and unbiased.

Conclusion Using the MR method, this study confirms that SULT1A1 is a risk factor for overall breast cancer, whereas IL-5 is a protective factor for overall breast cancer. SULT1A1 and CX3CL1 are risk factors for estrogen receptor-positive breast cancer, and beta-NGF is a risk factor for estrogen receptor-negative breast cancer.