Long-term exposure to PM1 and liver cancer mortality: Insights into the role of smaller particulate fractions

Ting Gan; Jie Chu; Hilary Bambrick; Xiaolei Guo; Wenbiao Hu

doi:10.1016/j.ecoenv.2025.118467

Long-term exposure to PM₁ and liver cancer mortality: Insights into the role of smaller particulate fractions

Ecotoxicol Environ Saf. 2025 Jul 15:300:118467. doi: 10.1016/j.ecoenv.2025.118467. Epub 2025 Jun 5.

Authors

Ting Gan¹, Jie Chu², Hilary Bambrick³, Xiaolei Guo⁴, Wenbiao Hu⁵

Affiliations

¹ Ecosystem Change and Population Health Research Group, School of Public Health and Social Work, Queensland University of Technology, Queensland, Australia.
² Shandong Center for Disease Control and Prevention, and Academy of Preventive Medicine, Shandong University, Jinan, Shandong, China.
³ Ecosystem Change and Population Health Research Group, School of Public Health and Social Work, Queensland University of Technology, Queensland, Australia; National Centre for Epidemiology and Population Health, Australian National University, ACT, Australia.
⁴ Shandong Center for Disease Control and Prevention, and Academy of Preventive Medicine, Shandong University, Jinan, Shandong, China. Electronic address: guoxiaolei@126.com.
⁵ Ecosystem Change and Population Health Research Group, School of Public Health and Social Work, Queensland University of Technology, Queensland, Australia. Electronic address: w2.hu@qut.edu.au.

PMID: 40480123
DOI: 10.1016/j.ecoenv.2025.118467

Abstract

Particulate matter (PM) is a recognized carcinogen, but the effects of PM₁ on liver cancer remain underexplored. This study investigates the long-term association between PM₁ and liver cancer mortality, as well as the contribution of smaller particles relative to larger PM effects. Data on 244,558 liver cancer deaths in Shandong were collected. The 10-year weighted moving average of PM was calculated to assess the long-term effects. Bayesian spatiotemporal models were applied to quantify the long-term associations between PM fractions (PM₁, PM_2.5, PM₁₀, PM_1-2.5, PM_1-10, and PM_2.5-10), ratios (PM₁/PM_2.5, PM₁/PM₁₀, PM_1-2.5/PM_1-10, and PM_2.5-10/PM_1-10) and liver cancer mortality while accounting for potential confounders and spatiotemporal effects. Nonlinear effects of PM were further explored using SHapley Additive exPlanations (SHAP) with eXtreme Gradient Boosting (XGBoost). Classification and Regression Tree (CART) models were applied to evaluate the importance of factors in regions with various PM₁ concentrations. The findings showed that PM₁ was credibly associated with an increased risk of liver cancer mortality (Relative risk (RR)= 1.813, 95 % Credible Interval (CrI): 1.647-1.997). Among larger PM, the proportion of smaller PM was associated with liver cancer mortality. (PM₁/PM₁₀: RR=1.135, 95 % CrI: 1.106-1.166; PM_1-2.5/PM_1-10: RR= 1.097, 95 % CrI: 1.079-1.115). Nonlinear relationships were identified, with an increasing risk at low PM₁ concentrations. In summary, long-term exposure to PM₁ is associated with liver cancer mortality, highlighting the role of smaller particles in PM-related risks. The integration of multiple models provides a robust approach to analyzing environmental impacts on cancer, aiding tool development in environmental oncology and supporting targeted health interventions.

Keywords: Bayesian spatiotemporal model; Liver cancer; Long term; Machine learning; PM(1).

MeSH terms

Air Pollutants* / adverse effects
Air Pollutants* / analysis
Air Pollutants* / toxicity
Air Pollution / statistics & numerical data
Bayes Theorem
China / epidemiology
Environmental Exposure* / statistics & numerical data
Humans
Liver Neoplasms* / chemically induced
Liver Neoplasms* / mortality
Particle Size
Particulate Matter* / adverse effects
Particulate Matter* / analysis

Substances

Particulate Matter
Air Pollutants