Cigarette smoking may increase the risk of prostate cancer by affecting circulating hormone levels or through exposure to carcinogens. Although there are plausible mechanisms that could explain an association between smoking and prostate cancer, previous studies are inconsistent. The goal of this population-based case-control study was to assess this association in middle-aged men. Cases (n = 753) were men ages 40-64 years diagnosed with prostate cancer from 1993 to 1996 identified using the Seattle-Puget Sound Cancer Registry. Age-matched controls without prostate cancer from the same region (n = 703) were identified using random digit dialing. Participants completed detailed in-person interviews. Logistic regression was used to compute adjusted odds ratios (ORs) and 95% confidence intervals (CIs) to assess the prostate cancer-cigarette smoking relationship. Current smokers had an increased risk (OR = 1.4, 95% CI 1.0-2.0) relative to nonsmokers. A dose-response relationship was noted between number of pack-years smoked and prostate cancer risk (trend P = 0.03). The OR = 1.6 (95% CI 1.1-2.2) for men with >40 pack-years of smoking, with a stronger association observed in men with more aggressive disease (OR = 2.0, 95% CI 1.3-3.1). Smoking cessation resulted in a decline in risk (trend P = 0.02). Smoking is associated with a moderately increased relative risk of prostate cancer. Furthermore, a dose-response relationship exists between number of pack-years smoked and cancer risk. Given that smoking cessation seems to reduce these risks, results from this study have public health ramifications and suggest that prostate cancer should be added to the list of tumors for which cigarette smoking is a risk factor.