Objective Study carcinogenicity of inorganic lead, classified as ‘probably carcinogenic’ to humans by the International Agency for Research on Cancer (brain, lung, kidney and stomach).

Methods We conducted internal and external analyses for cancer incidence in two cohorts of 29 874 lead-exposed workers with past blood lead data (Finland, n=20 752, Great Britain=9122), with 6790 incident cancers. Exposure was maximum measured blood lead.

Results The combined cohort had a median maximum blood lead of 29 μg/dL, a mean first blood lead test of 1977, and was 87% male. Significant (p<0.05) positive trends, using the log of maximum blood lead, were found for brain cancer (malignant), Hodgkin’s lymphoma, lung cancer and rectal cancer, while a significant negative trend was found for melanoma. Borderline significant positive trends (0.05≤p≤0.10) were found for oesophageal cancer, meningioma and combined malignant/benign brain cancer. Categorical analyses reflected these trends. Significant interactions by country were found for lung, brain and oesophageal cancer, with Finland showing strong positive trends, and Great Britain showing modest or no trends. Larynx cancer in Finland also showed a positive trend (p=0.05). External analyses for high exposure workers (maximum blood lead >40 μg/dL) showed a significant excess for lung cancer in both countries combined, and significant excesses in Finland for brain and lung cancer. The Great Britain data were limited by small numbers for some cancers, and limited variation in exposure.

Conclusions We found strong positive incidence trends with increasing blood lead level, for several outcomes in internal analysis. Two of these, lung and brain cancer, were sites of a priori interest.