Associations of long-term exposure to fine particulate matter and its constituents with gynecologic cancer incidence: A prospective cohort study from Beijing

There is limited evidence of the association between long-term exposure to particulate matter with aerodynamic diameter ≤2.5 μm (fine particulate matter, PM 2.5 ) and gynecologic cancer incidence, particularly in developing countries with high air pollution levels. The present study aimed to investigate the associations of long-term exposure to PM 2.5 and its constituents with the incidence of gynecologic cancer, and to identify the primary constituent. A total of 85,500 women from the National Urban Cancer Screening Program in Beijing (2013–2019) were included. 5-year average concentrations of PM 2.5 mass and its constituents, namely carbon black (BC), ammonium (NH 4 + ), nitrate (NO 3 − ), organic matter (OM), and sulfate (SO 4 2− ), were obtained from the Tracking Air Pollution in China (TAP) dataset. Cox proportional hazard models were used to analyze the association between long-term exposure to PM 2.5 and its constituents and gynecologic cancer incidence. Mixture exposure models, including quantile-based g-computation (Qgcomp) and explanatory machine learning model (SHapley Additive exPlanations, SHAP), were used to assess the effect of each constituent. During a mean follow-up of 6.22 years, 524 participants were diagnosed with gynecologic cancer. The adjusted hazard ratios (HRs) for gynecologic cancer incidence associated with 1 standard deviation (SD) increment was 1.10 (95% CI: 1.00–1.22) for PM 2.5 , 1.12 (95% CI: 1.01–1.23) for BC,1.11 (95% CI: 1.00–1.23) for NH 4 + , 1.12 (95% CI: 1.01–1.23) for OM, 1.12 (95% CI: 1.01–1.23) for SO 4 2− , and 1.08 (95% CI: 0.98–1.19) for NO 3 − . The Qgcomp model indicated the association between a mixture of PM 2.5 constituents and gynecologic cancer incidence, with positive weights assigned to BC (61.6%), SO 4 2− (35.9%), and NO 3 − (2.5%). The SHAP model showed that SO 4 2− was the most important constituent, with the mean |SHAP| value of 0.0322, followed by BC (0.0270), OM (0.0232), NO 3 − (0.0212) and NH 4 + (0.0190). Long-term exposure to PM 2.5 was associated with gynecologic cancer incidence, with SO 4 2− and BC contributing significantly to the association.