This study aimed to assess independent effects of EBV and cigarette smoking on nasopharyngeal carcinoma, which have never been assessed in long-term followup studies. A cohort of 9,622 men was enrolled from 1984 to 1986. Blood samples collected at study entry were tested for antibodies against EBV antigens (anti-EBV) viral capsid antigen immunoglobulin A and DNase. The cigarette smoking habit was inquired through questionnaire interview. Newly developed nasopharyngeal carcinoma cases were ascertained through computerized linkage with national cancer registry profile. Cox's proportional hazard regression analysis was used to estimate multivariate-adjusted hazard ratio with its 95% confidence interval (95% CI). During the follow-up of 173,706 person-years, 32 pathologically confirmed nasopharyngeal carcinoma cases were identified >1 year after recruitment. Increasing serum levels of anti -EBV viral capsid antigen immunoglobulin A and DNase were significantly associated with nasopharyngeal carcinoma risk in a dose-response relationship.The multivariate-adjusted hazard ratio (95% CI) of developing nasopharyngeal carcinoma for low and high antibody levels compared with seronegatives was 9.5 (2.2-40.1) and 21.4 (2.8-161.7), respectively, for anti -EBV viral capsid antigen immunoglobulin A (P < 0.001 for trend), and 1.6 (0.5-4.6) and 16.0 (5.4-47.1), respectively, for anti -EBV DNase (P < 0.001 for trend). The shorter the time interval between study entry and nasopharyngeal carcinoma diagnosis, the higher was the proportion of anti -EBV viral capsid antigen immunoglobulin A among nasopharyngeal carcinoma patients. The multivariate-adjusted hazard ratio (95% CI) was 3.0 (1.3-7.2) for z30 pack-years of cumulative cigarette smoking compared with <30 pack-years as the reference. The longer and heavier the cigarette smoking habit, the higher was the nasopharyngeal carcinoma risk. Anti -EBV viral capsid antigen immunoglobulin A, anti -EBV DNase, and long-term heavy cigarette smoking are independent nasopharyngeal carcinoma risk predictors. (Cancer Epidemiol Biomarkers Prev 2009;18(4):1218 -26)