Pre-diagnosis and post-treatment exposures to PM₁, PM_2.5, and PM₁₀ and breast cancer survival: A cohort study

Ambient PM exposure is associated with poorer breast cancer survival, yet prior work rarely distinguished exposure timing or assessed multiple PM fractions. This study examined PM₁, PM_2.5, and PM₁₀ in relation to cancer-specific survival in Liaoning, China, using Mostofsky's residual models to isolate independent associations with breast cancer survival. We conducted a retrospective cohort study of 3975 breast cancer patients (2015-2020) in Liaoning, China. Pre- and post-diagnosis exposures to PM₁, PM_2.5, and PM₁₀ were estimated from high-resolution CHAP datasets. Pre-diagnosis exposure was defined as the average PM concentration during the 5 years preceding breast cancer diagnosis, whereas post-diagnosis exposure was treated as a time-varying exposure during follow-up. Time-stratified Cox proportional hazards models, adjusted for age, stage, estrogen/progesterone receptor status, NDVI, artificial light at night, and other covariates, assessed risks via Mostofsky's residual and remaining mass models. Subgroup, exploratory, and sensitivity analyses evaluated robustness. Post-treatment exposures were associated with poorer survival, with fully adjusted hazard ratios (HRs) per 1 μg/m³ increase of 1.459 (95% CI: 1.376-1.546) for PM₁, 1.215 (1.180-1.252) for PM_2.5, and 1.144 (1.120-1.169) for PM₁₀. Pre-diagnosis exposures showed limited associations, with only modest for PM₁ (1.038; 1.008-1.069). Independent associations via Mostofsky's residual models confirmed finer fractions' roles: PM_2.5 adjusted for PM₁₀ (HR: 1.132; 1.049-1.223), PM₁ adjusted for PM_2.5 (HR: 1.100; 1.018-1.187). Subgroups showed stronger associations in older patients (≥65 years) and early-stage disease (I/II). Exploratory analyses revealed escalating HRs over time and key PM_2.5 constituents: sulfate, nitrate, and ammonium in remaining mass models. Findings were robust across alternative buffers, exposure cutoffs, and stratification. In conclusion, post-treatment PM exposures, especially finer fractions, independently predict poorer breast cancer survival, emphasizing the necessity for air quality interventions to enhance outcomes in polluted regions.