Regular welders had a non-significant 2× elevated testicular cancer risk; non-seminoma risk tripled, but rests on very few cases
Journal: European Journal of Public Health | Published: 2026-03-14 | Type: Case-Control Study | PMID: 41525298 Authors: Kendzia B, Behrens T, Brüning T et al. (Institute for Prevention and Occupational Medicine of the German Social Accident Insurance; University Hospital Essen; Leibniz-Institute for Prevention Research and Epidemiology) Funding/COI: Federal Ministry for Education and Research. No conflicts of interest listed.
A German case-control study of 268 testicular cancer cases and 797 controls found that regular welders had roughly double the odds of developing testicular cancer compared to non-welders, though the confidence interval crossed 1 and the result was not statistically significant. The elevated signal was strongest in non-seminoma histology among regular welders (OR 3.71), but that estimate rests on a handful of exposed cases and comes with a wide, barely-significant confidence interval. Crucially, no dose-response relationship emerged — more fume exposure did not mean more cancer — which undercuts the causal story.
This study's main methodological contribution is its use of a measurement-based Welding Exposure Matrix (WEM), linking individual job-task descriptions to quantified inhalable fume exposure values across each worker's complete occupational history. This is a meaningful step above studies that simply classify workers as "ever/never" welders. Coding of job periods was blinded to case-control status, and a reference pathologist independently reviewed all cancer diagnoses — both are genuine strengths for a study of this era.
However, the data is from 1995–1997, making it nearly 30 years old. The control response rate was 57%, raising the possibility of selection bias. The number of cases with regular welding exposure — particularly for the non-seminoma subtype — is not reported explicitly but is described as "few," rendering the headline OR of 3.71 statistically fragile. The absence of a dose-response relationship across cumulative exposure, duration, and intensity is a serious problem for any causal interpretation: if welding fumes cause testicular cancer, you would expect to see rising risk with rising exposure, and it isn't there.
The signal is interesting — regular welding showing a roughly twofold TC risk elevation is worth noting — but this paper cannot support a causal conclusion. The most striking number (OR 3.71 for non-seminoma) is almost certainly underpowered, and the failure to find any dose-response relationship with quantified fume exposure is a red flag the authors acknowledge but do not resolve. The data is also three decades old. This is a hypothesis-generating study that identifies occupational welding as a candidate exposure warranting investigation in larger, more recent cohorts — not a study that establishes welding as a testicular cancer risk factor.