鎮痛薬の乳がんリスクおよび生存率に対する差次的因果効果：メンデルランダム化からのエビデンス

The causal effects of widely used analgesics-paracetamol(acetaminophen), aspirin (acetylsalicylic acid), and ibuprofen-on breast cancer risk and survival remain uncertain. This Mendelian randomization (MR) study investigated their causal relationships with breast cancer incidence, mortality, and estrogen receptor (ER)-subtype heterogeneity.

Using two-sample MR, genetic instruments for analgesic use were derived from UK Biobank GWAS (N=457,547). Outcome data included breast cancer incidence (122,977 cases/105,974 controls), ER-subtypes (ER+: 69,501 cases; ER-: 21,468 cases), and survival statistics. Inverse-variance weighted (IVW) analyses were primary, supplemented by MR-Egger, weighted median/mode, and sensitivity analyses (MR-PRESSO, leave-one-out). Bidirectional MR assessed reverse causation.

Genetically predicted paracetamol use increased overall breast cancer risk (IVW OR=3.26, 95% CI:1.60-6.63, pFDR=0.005) and ER+ subtype risk (OR=3.65, 1.79-7.45, pFDR=0.003). Aspirin use showed no association with incidence but improved overall survival (HR=0.0036, 0.0001-0.1218, pFDR=0.016). Ibuprofen demonstrated no significant associations with risk or survival. Subtype-specific survival analyses were null. No reverse causation was detected (all p >0.05). Sensitivity analyses confirmed robustness, with minimal pleiotropy (MR-Egger intercept p >0.05) and consistent effects after outlier correction.

This MR study links a genetic predisposition to paracetamol use with increased breast cancer risk (especially ER+), and to aspirin use with improved survival. These divergent findings point to drug-specific mechanisms, warranting caution with long-term paracetamol use and further study of aspirin's therapeutic potential. Clinical decisions should balance analgesic benefits against these potential cancer-related outcomes.