IARC 60th Anniversary - 19-21 May 2026
Session : Environmental and occupational cancer: an underestimated burden?
Diesel engine exhaust exposure and prostate cancer risk: early findings from 3 case-control studies
PARENT M. 1, KOGEVINAS M. 4, MENEGAUX F. 7, LAVOUÉ J. 2, MÉSIDOR M. 1, VERMEULEN R. 5, KROMHOUT H. 5, GE C. 8, YU J. 1, RICHARD H. 1, PETERS C. 6, SARAZIN P. 3
1 Institut national de la recherche scientifique , Laval, Canada; 2 Université de Montréal, Montréal, Canada; 3 Institut de recherche Robert-Sauvé en santé et en sécurité du travail, Montréal, Canada; 4 ISGlobal, Barcelona, Spain; 5 Utrecht University, Utrecht, Netherlands; 6 BC Cancer, Vancouver, Canada; 7 INSERM, Paris, France; 8 TNO, Utretcht, Netherlands
Background: Diesel engine exhaust (DEE) has a high content in chemicals with carcinogenic or hormone-disrupting properties potentially relevant to prostate cancer. Epidemiological evidence on this is very limited.
Objective: To study the association between lifetime occupational exposure to DEE assessed quantitatively and prostate cancer risk, overall and by tumour aggressiveness.
Methods: We used data from three population-based case-control studies (2005-2013) of prostate cancer from Canada (PROtEuS: 1,929 cases, 1,991 controls), Spain (MCC-Spain: 1,112 cases, 1,493 controls) and France (EPICAP: 818 cases, 875 controls) with complete work histories. DEE exposure was assigned using the quantitative DEE-JEM built around elemental carbon exposure and linked to each job by ISCO-88 codes. Unconditional logistic regression estimated odds ratios (ORs) and their 95% confidence intervals (CIs) between exposure to DEE (ever, duration, cumulative) and prostate cancer risk separately for each study, and then in a meta-analysis, for overall cancer and by cancer aggressiveness. Polytomous models were used to study cancer aggressiveness; a Gleason score ≥7 [4 + 3] defined an aggressive cancer. Potential lifestyle and occupational confounders were identified using a directed acyclic graph. Multiple imputations were applied to account for missing covariate information. We conducted a meta-analysis based on the three studies and assessed heterogeneity (I2). In sensitivity analyses, we restricted controls to participants screened in the previous 2 years, thereby limiting the potential for undiagnosed prostate cancer, and applied inverse-probability-of-censoring weighting to reduce possible bias than might ensue from excluding participants without recent screening information.
Results: There was little evidence of associations for overall prostate cancer with exposure to DEE using either of the three metrics, in any of the studies, although ORs increased slightly across duration categories. Positive associations with prolonged exposure (≥30 years) emerged for aggressive cancers in PROtEuS (OR=1.34; 95%CI 0.94-1.90) and EPICAP (OR=1.20, 95%CI 0.71-2.03), but not MCC-Spain (OR=0.82, 95%CI 0.55-1.21). The meta-OR for overall prostate cancer associated with prolonged exposure based on the three studies was 1.08 (95%CI 0.90-1.26, I2=31%), and 1.15 (95%CI 1.02-1.28, I2=0%) when restricting participants to recently screened individuals. Corresponding values for aggressive cancers were 1.12 (95%CI 0.79-1.45, I2=49%) and 1.21 (95%CI 1.00-1.42, I2=30%), respectively.
Conclusions: This is the first study to date to examine the role of lifetime occupational exposure to DEE, assessed quantitatively, in prostate cancer risk. Preliminary findings suggest that prolonged exposure to DEE is associated with a small increase in prostate cancer risk, particularly for aggressive cancers. Consideration of prostate cancer screening history influenced findings. Future analyses will evaluate risk estimates in pooled analyses, when there is limited heterogeneity across studies.