Smoking and nasopharyngeal carcinoma : prospective evidence from the Guangzhou Occupational Cohort Study and a meta-analysis

Abstract

Background: Nasopharyngeal carcinoma (NPC), also known as Cantonese cancer, has a particularly high incidence in North Africa and Southeast Asia, especially in Guangdong, China, such as Guangzhou, but it is a rare tumor worldwide. Tobacco is a well-known causal factor for head and neck cancers, except NPC. In the 2014 United States Surgeon General’s report and the 2012 China Tobacco Hazard Report, NPC is not considered as causally related to smoking. Evidence supporting the association between smoking and the risk of NPC mainly came from case-control studies, but prospective evidence remains limited. We used Guangzhou Occupational Cohort Study data to conduct the first prospective study on smoking and NPC mortality in an NPC high-risk region and updated the meta-analysis results focused on prospective cohort studies.

Methods: In the Guangzhou Occupational Cohort Study, from March 1988 to December 1992, baseline data of demographic characteristics and smoking status and alcohol drinking were collected through occupational health examinations in factories and driver examination stations. 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. For the updated meta-analysis, the most updated results of all cohort studies (including the Guangzhou Occupational Cohort Study) were included, and pooled hazard ratio (HR) was estimated based on a random-effects model.

Results: In the Guangzhou Occupational Cohort Study, 87,898 men were included for the main analysis, and 32 NPC deaths occurred during the average 7.3 years of follow up. The mean age (standard deviation) of the subjects was 41 (5.9) years at baseline. Compared with never smokers, the adjusted (adjusted by age, education, occupation and cohort status) hazard ratio (HR) was 2.70 (95% confidence interval 1.09-6.67, p=0.032) for daily smokers and 3.07 (1.18-7.96, p=0.021) for smokers with more than 10 pack-years of cumulative consumption. A significant dose-response relation was observed with daily smoking amount and number of pack-years (both p for trend=0.02). In the meta-analysis, a significant pooled HR of 1.78 (1.18-2.70, p=0.01) was observed when all five cohort studies were included. After excluding one cohort study from low-risk regions, the pooled HR was 1.57 (1.03-2.40, p=0.04) for four cohort studies in high-risk regions.

Conclusions: The Guangzhou Occupational Cohort Study is the first and largest cohort in a high NPC risk region, and smoking was found to associate with higher NPC mortality with a dose-response pattern. The significant meta-analysis results also support that smoking is likely to be causally associated with NPC.