Smoking and nasopharyngeal carcinoma mortality: A cohort study of 101,823 adults in Guangzhou, China

Background: Nasopharyngeal carcinoma (NPC), also known as Cantonese cancer, is rare worldwide, but has particularly high incidence in North Africa and Southeast Asia, especially in Guangdong, China, such as Guangzhou. Tobacco causes head and neck cancers, but nasopharyngeal carcinoma is not included as causally related to smoking in the 2014 United States Surgeon General's report. Prospective evidence remains limited. We used Guangzhou Occupational Cohort data to conduct the first and robust prospective study on smoking and NPC mortality in an NPC high-risk region. Methods: Information on demographic characteristics and smoking status was collected through occupational health examinations in factories and driver examination stations from March 1988 to December 1992. Vital status and causes of deaths were retrieved until the end of 1999. Cox proportional hazard model was used to assess the association of smoking with NPC mortality. Results: Of 101,823 subjects included for the present analysis, 34 NPC deaths occurred during the average 7.3 years of follow up. The mean age (standard deviation) of the subjects was 41 (5.7) years. Compared with never smokers, the hazard ratio (HR) of NPC mortality was 2.95 (95 % confidence interval 1.01-8.68; p = 0.048) for daily smokers and 4.03 (1.29-12.58; p = 0.016) for smokers with more than 10 pack-years of cumulative consumption, after adjusting for age, sex, education, drinking status, occupation and cohort status and accounting for smoking-drinking interaction. The risk of NPC mortality increased significantly with cigarettes per day (p for trend = 0.01) and number of pack-years (p for trend = 0.02). Conclusions: In this first and largest cohort in a high NPC risk region, smoking was associated with higher NPC mortality. The findings have shown statistically significant dose-response trend between smoking amount and smoking cumulative consumption and the risk of NPC mortality, but due to the small event number, further studies with larger sample size are needed to confirm the findings in the present study. Our results support that smoking is one of the risk factors likely to be causally associated with NPC mortality.