Nested case-control interview studies of lung cancer (610 incident cases), stomach cancer (292 incident cases), and 959 controls were conducted to follow up leads from a proportional mortality analysis of deaths among male workers in a large integrated iron-steel complex in Anshan, China. For lung cancer, after adjusting for the significant non-occupational risk factors (smoking, other pulmonary disease, family history of lung cancer, and low consumption of fruit or tea), risks were significantly elevated for those employed for 15 or more years in smelting and rolling (OR = 1.5, CI = 1.1-2.2), in the fire-resistant brick factory (OR = 2.9, CI = 1.4-5.9), in general loading (OR = 2.5, CI = 1.0-6.1), and as coke oven workers (OR = 3.4; CI = 1.4-8.5). For stomach cancer, after adjusting for consumption of pickled vegetables, prior gastric diseases, family history of stomach cancer, low intake of fruits and vegetables, and education, risks were significantly elevated for those employed for 15 or more years in ore sintering and transportation (OR = 2.1, CI = 1.0-4.4), in the fire-resistant brick factory (OR = 2.5, CI = 1.1-5.8), in general loading (OR = 3.2, CI = 1.2-8.9), as boilerworkers and cooks (OR = 2.6, CI = 1.2-5.6), and as coke oven workers (OR = 5.4, CI = 1.8-16.0). For both lung and stomach cancers, significant dose-response gradients were observed for exposure to total dust and benzo(a)pyrene, but not for specific chemical components of dust. Overall, long-term steel workers with exposure to workplace pollutants had a 40% increased risk of both lung and stomach cancers. These case-control studies confirm many of the occupational findings reported in the proportionate mortality analysis, and suggest avenues for further work to evaluate the carcinogenicity of individual components of dust.