gms | German Medical Science

Introduction: Epidemiologic and toxicologic studies point to a potential carcinogenic effect of dinitrotoluene (DNT), particularly with respect to renal and urothelial cancer.

Methods: The cohort comprised all men born between 1920 and 1974 (N=16,441) who were ever gainfully employed between 1953 and 1990 in one of two underground copper mines in Mansfeld, Saxony-Anhalt, former German Democratic Republic and followed-up till 2006. Incident cancer cases were identified by record linkage with the Common Cancer Registry of the New Laender (GKR). In the external analysis, standardized incidence ratios (SIR) were calculated with the general population of Saxony-Anhalt as the reference. In the internal case-cohort analysis, additional cancer cases were identified by a network of pathology institutes and a subcohort (n=999) as random sample of the total cohort was defined. Vital status could be ascertained for 88% of subcohort members not suffering from renal or urothelial cancer. For all cases and subcohort members, dermal and inhalative DNT exposure was assessed by an expert-based job exposure matrix. As potential confounders, smoking status could be evaluated for 40%, alcohol abuse for 69% of the subcohort members. Hazard rate ratios (HRR) for renal and urothelial cancer were calculated with 95% confidence intervals separately for dermal and inhalative exposure.

Results: SIR for all cancers were not significantly elevated in the cohort (SIR=1.06; 95% CI 0.97-1.16). We found an increase in lung cancer (SIR=1.28; 1.13-1.45), but not in kidney cancer (74 cases, SIR=1.03; 95% CI 0.81-1.31) or invasive urothelial cancer (88 cases, SIR=0.93; 95% CI 0.72-1.20). However, an increased SIR of 1.89 (21 cases, 95% CI 1.22-2.92) was found for carcinoma in situ and papilloma of the bladder. In the internal case-cohort analysis (108 cases of kidney cancer, 139 cases of urothelial cancer), we found elevated hazard rate ratios (adjusted for smoking) for dermal (but not inhalative) DNT exposure and kidney cancer (HRR=4.08; 95% 1.32-12.54 for medium and HRR=1.69; 95% CI 0.65-4.39 for high exposure), but not for urothelial cancer.

Conclusions: The SIR analysis of workers in the copper-mining industry in comparison with the general population of Saxony-Anhalt does not reveal clearly increased risks for renal or urothelial cancer. The subsequent case-cohort analysis including expert-assessment of DNT exposure and identification of additional cancer cases by a network of pathology institutes points to a potential etiologic role of DNT exposure to renal cancer.