Tobacco exposure in adolescence and the risk of nasopharyngeal carcinoma: preliminary results from a multi-center case-control study in Hong Kong, China
Jia Huang Lin 1, 2  
,   Zhi-Ming Mai 1, 2,   Tai Hing Lam 1, 2
More details
Hide details
The University of Hong Kong, School of Public Health, China
The University of Hong Kong, Center for Nasopharyngeal Carcinoma Research, China
Publication date: 2018-03-01
Tob. Induc. Dis. 2018;16(Suppl 1):A818
Download abstract book (PDF)

Incidence of many cancers increases with age, but the incidence of nasopharyngeal carcinoma (NPC) in high-risk areas (e.g. Hong Kong) peaks at age 40-55 years. The early onset suggests an important role of early childhood exposures in NPC development. In our ongoing NPC case-control study, we examined the relationship between tobacco exposure especially in adolescence and NPC.

We analysed 777 histologically confirmed NPC cases (mean age=52.8, SD=12.0), and 1,690 frequency matched non-NPC controls (mean age=50.4, SD=13.4) recruited during July 2014 to July 2017 from 5 of the largest hospitals (n=1,460) and 3 Red Cross Blood Transfusion Centres (n=230) in Hong Kong. Information on smoking and other factors was collected via a computer-assisted self-administered questionnaire. Re-interviewing of the 140 subjects showed good agreement of smoking history (Kappa or ICC: 0.65-0.99). Unconditional logistic regression was used to estimate odds ratios (ORs) of NPC for tobacco exposure, adjusting for 5-year age groups, sex, housing type at age 10, and alcohol drinking status.

Compared with never smokers, the OR (95%CIs) was 1.62 (1.32-1.98) for ever-smokers. The ORs were 2.27 (1.38-3.75) for starting smoking at 6-12 years and 1.65 (1.30-2.10) for 13-18 years (p for trend=0.001). For those who started at 6-18 years, the OR was 1.23 (0.79-1.93) for smoking 1-5 cigarettes/day, 1.71 (1.21-2.41) for 6-10 cigarettes/day and 1.78 (1.26-2.53) for 11+ cigarettes/day (p for trend< 0.001).

Our preliminary results have shown that tobacco exposure in adolescence were associated with higher risk of NPC in a high-risk area. Further analysis will be conducted after case (target=800) and control (target=1900) recruitment is completed.