Time-varying exposure to ambient air pollution and mortality among colon cancer patients in northern Thailand: a 15-year retrospective cohort study.
Northern Thailand experiences severe annual air pollution (PM2.5 > 35 μg/m3), primarily due to agricultural burning. While ambient PM2.5 has been linked to gastrointestinal cancer deaths, its effect on colon cancer survival in high-exposure areas remains understudied.
In this 15-year retrospective cohort study, we used data on 5,018 patients with colon cancer (2003-2018) obtained from the Chiang Mai Cancer Registry. Time-varying exposure to PM2.5, PM10, NO₂, SO₂, CO, and O₃ was evaluated by exploiting annually updated Copernicus Atmosphere Monitoring Service data and linking them to residential districts. Multivariable time-dependent Cox models adjusted for demographics, tumor characteristics, and treatment were used in the analysis.
Over 18,650 person-years of follow-up (PYFU), 72% of patients died (mortality rate: 19.4 per 100 PYFU). Exposure to PM2.5 ≥ 37.5 μg/m3 (a regionally relevant threshold) was independently linked to double the risk of all-cause mortality (adjusted hazard ratio (aHR) = 1.96, 95% confidence interval: 1.49-2.58, p < 0.001), as was having a low body mass index (aHR = 1.54) and being male (aHR = 1.23). Patients who died had significantly higher mean time-updated PM2.5 exposure (with consistent annual exposure differences) compared to the survivors (37.7 vs. 35.5 μg/m3; p < 0.001). No significant associations were found between colon cancer and PM10, O₃, NO₂, SO₂, or CO exposure. The PM2.5 effect size was larger than estimates from regions with lower exposure levels.
Long-term exposure to ambient PM2.5 markedly increases the risk of death among colon cancer patients in northern Thailand. As a modifiable risk factor, PM2.5 mitigation urgently needs to be incorporated into environmental and survivorship care strategies in high-exposure areas.
In this 15-year retrospective cohort study, we used data on 5,018 patients with colon cancer (2003-2018) obtained from the Chiang Mai Cancer Registry. Time-varying exposure to PM2.5, PM10, NO₂, SO₂, CO, and O₃ was evaluated by exploiting annually updated Copernicus Atmosphere Monitoring Service data and linking them to residential districts. Multivariable time-dependent Cox models adjusted for demographics, tumor characteristics, and treatment were used in the analysis.
Over 18,650 person-years of follow-up (PYFU), 72% of patients died (mortality rate: 19.4 per 100 PYFU). Exposure to PM2.5 ≥ 37.5 μg/m3 (a regionally relevant threshold) was independently linked to double the risk of all-cause mortality (adjusted hazard ratio (aHR) = 1.96, 95% confidence interval: 1.49-2.58, p < 0.001), as was having a low body mass index (aHR = 1.54) and being male (aHR = 1.23). Patients who died had significantly higher mean time-updated PM2.5 exposure (with consistent annual exposure differences) compared to the survivors (37.7 vs. 35.5 μg/m3; p < 0.001). No significant associations were found between colon cancer and PM10, O₃, NO₂, SO₂, or CO exposure. The PM2.5 effect size was larger than estimates from regions with lower exposure levels.
Long-term exposure to ambient PM2.5 markedly increases the risk of death among colon cancer patients in northern Thailand. As a modifiable risk factor, PM2.5 mitigation urgently needs to be incorporated into environmental and survivorship care strategies in high-exposure areas.
Authors
Chitapanarux Chitapanarux, Traisathit Traisathit, Srikummoon Srikummoon, Thongsak Thongsak, Nakharutai Nakharutai, Thumronglaohapun Thumronglaohapun, Supasri Supasri, Hemwan Hemwan, Chitapanarux Chitapanarux
View on Pubmed