The formation of DNA adducts can lead to DNA replication errors and the potential for carcinogenesis. DNA adducts have been detected in prostate cells, but the distribution of adducts with respect to prostate cancer risk factors and histology is unknown. In a study of 130 Caucasian (n ؍ 61) and African-American (n ؍ 69) men with prostate cancer who underwent radical prostatectomy, we quantified polycyclic aromatic hydrocarbon (PAH)-DNA adducts in prostate tumor and adjacent nontumor cells by immunohistochemistry. A strong correlation between paired adduct levels in the two cell types was observed (r ؍ 0.56; P < 0.0001); however, nontumor cells had a significantly higher level of adducts compared with tumor (0.30 absorbance units ؎ 0.05 versus 0.17 absorbance units ؎ 0.04; P < 0.0001). Variables significantly associated with PAH-DNA adduct levels in tumor cells included primary Gleason grade, tumor volume, and log-transformed prostate-specific antigen (PSA) at time of diagnosis. Tumors with a primary Gleason grade of 5 had significantly lower PAH-DNA adduct levels than tumor cells with a primary Gleason grade of 3 or 4 (P < 0.0001 for both). Tumors that involved 10% or less of the prostate gland had significantly higher PAH-DNA adduct levels than tumors that involved 15 to 20% of the prostate gland (P ؍ 0.004). PSA levels were inversely associated with PAH-DNA adduct levels in tumor cells (P ؍ 0.009). A similar, albeit less significant, inverse association was observed between PSA and PAH-DNA adduct levels in nontumor cells (P ؍ 0.07). Interestingly, increasing primary Gleason grade was associated with increasing PAH-DNA adduct levels in adjacent nontumor cells (P ؍ 0.008). Our results show that PAH-DNA adducts are present in the prostate but vary with regard to cellular histology. In prostate tumor cells, decreased cellular differentiation and increased tumor proliferation may reduce PAH-DNA adduct levels.