Life-course trajectories of tobacco and alcohol consumption and risk of head and neck cancer; does human papillomavirus play a role? an international multi-center study.

Sreenath MADATHIL, McGill University , Canada

1 Faculty of Dentistry, McGill University, Montreal, Canada
2 Epidemiology and Biostatistics Unit, INRS-Institute Armand-Frappier, Laval, Canada
3 Division of Cancer Epidemiology, Department of Oncology, McGill University, Montreal, Canada

Purpose: To compare tobacco and alcohol life-course trajectories between two countries with different burden of head and neck cancers (H&NC), and to assess the role of the human papillomaviruses (HPV) in this relationship.

Method: We used data from a hospital‐based case‐control study conducted in Canada (460 cases, 458 controls) and India (350 cases, 371 controls). We recruited subjects newly diagnosed with primary squamous cell H&NC and non‐cancer controls frequency‐matched to cases according to age and sex. Semi-structured interviews using a life‐grid technique collected information on several life-course exposures, including a detailed history of tobacco smoking and alcohol consumption. HPV was detected and genotyped in oral mucosal samples from brush biopsies using Linear Array.
Country specific trajectories of intensity of exposures were estimated as flexible functions of age and years since habit initiation using regression splines.

Results: Cases and controls had different life-course trajectories of smoking and alcohol in both countries. Canadian cases initiated and achieved their highest intensity of tobacco and alcohol consumption 2-5 years earlier than Indian cases. However, the latter consumed higher intensities of both throughout their life. Interestingly, 41% and 0% of cases tested positive for HPV infection in Canada and India, respectively. Trajectories of smoking and alcohol clearly showed considerable difference by HPV status. Among HPV negative subjects, controls had higher cumulative exposure before 30 years of age compared to cases. Trajectories did not differ between HPV positive cases and controls. Life-course tobacco and alcohol trajectories in the Indian sample were strikingly similar to those of HPV negative Canadians.

Conclusion: Life-course trajectories of major H&NC risk factors differ in specific characteristics between Canada and India. HPV may modify the effect of these trajectories. Cross population life-course studies may lead to new prevention avenues.

Funding source: Canadian Institutes of Health Research.