?Background: Testicular germ cell tumors (TGCT), which account for 98% of testicular cancers, are the most frequent malignant tumors in men aged 15–44 years. The incidence of these tumors has increased substantially in recent decades, particularly in Western countries, with a 2.6% annual rise in France since 1980. An environmental origin is suspected, with endocrine-disrupting chemicals (EDCs) being of particular concern. Among substances considered as potential EDCs and carcinogens, special attention has been given to various air pollutants. Previous studies have reported probable associations between occupational, domestic, and environmental pesticide exposure and TGCT risk, but residential exposure to air pollution—especially at early ages—has been poorly investigated in this cancer.
Objectives: To estimate the impact of exposure to potential carcinogenic or/and endocrine-disrupting air pollutants on TGCT risk.
Methods: The French multicenter case-control TESTIS study (2015–2018) included 451 cases and 658 matched controls, whose residential history (1968–2018) was geocoded. Lifetime average annual exposure to multiple air pollutants (benzo[a]pyrene, NO₂, O₃, cadmium, PM10, PM2.5, PCB153, dioxins, and black carbon) was estimated using several approaches. For all pollutants, the CHIMERE chemistry-transport model (1990–2010) was combined with emission data from CITEPA, the French reference organization for national air pollutant inventories, available up to 2018. A land-use regression (LUR) model (1990–2018) was used for PM2.5, PM10, and NO₂. A GIS metric (1990–2018) was applied for dioxin and cadmium exposure. Exposures before 1990 could not be assessed. The association between pollutant concentration and TGCT risk was analyzed using conditional logistic regression, adjusted for confounding variables from the study questionnaire.
Results: A significant positive association was observed between lifetime average particulate exposure (estimated by LUR) and TGCT risk ( OR_PM10=1.28, 95% CI : 1.01-1.62; OR_PM2.5=1.59, 95% CI : 1.09-2.32  per 10 µg/m³ increase). Although not significant, CHIMERE model results for particulates were consistent with these findings. Positive trends were also observed for average and cumulative PCB exposure (OR_{PCB153_avg}=1.12, 95% CI: 1.00-1.27; OR_{PCB153_cum}=1.10, 95% CI:0.97-1.25, for a one standard deviation increment among controls), but these were not significant. Positive trends were also noted for cadmium and dioxin exposures estimated by CHIMERE (OR_Cd = 1.09, 95% CI: 1.00-1.19; OR_dioxins= 1.06, 95% CI: 0.99-1.13, for a one standard deviation increment among controls), but these non-significant findings were not confirmed by the GIS metric. No associations were found for other pollutants.
Conclusion: This observational study suggests a significant association between lifetime average exposure to PM2.5 and PM10 and TGCT risk, indicating a possible role of air pollution in the etiology of this cancer. Dioxins, cadmium, and PCBs may also be involved. However, these results should be interpreted with caution and confirmed by further studies. Multi-pollutant approaches, such as Bayesian Profile Regression (BPR) or Bayesian Kernel Machine Regression (BKMR) models, are particularly needed to estimate the joint exposure to air pollutants, in order to take into account with the multicollinearity between pollutants, the potential non-linear and non-additive effects.