In CA, Dose-Dependent Association for Smoking but Not for Drinking

LYON, FRANCE, May 16 -- Smoking and alcohol are independent risk factors for head and neck cancer, but a dose-dependent relationship was observed only for smoking, said researchers here.

A pooled analysis of data from 15 case-control studies found that among patients who never drink, smoking was associated with more than a doubling of the risk for head and neck cancers (odds ratio 2.13, 95% C.I. 1.52-2.98 P <0.001 for heterogeneity), according to findings reported in the May 16 issue of the Journal of the National Cancer Institute. The risk increased as cigarette smoking increased (P<0.001 for trend).

Among patients who never smoked, consuming three or more drinks a day was associated with an increased risk of oropharynx/hypopharynx and larynx cancers (OR 1.82 95% CI 1.10 to 2.99 P<0.001 for heterogeneity). There was no significant increase associated with light alcohol consumption -- one or two drinks a day.

"Our results suggest that, in the absence of tobacco use, the association between alcohol consumption and the risk of head and neck cancer is weak and is apparent only at high doses and only for pharyngeal and laryngeal cancers," wrote Mia Hashibe, Ph.D., of the International Agency for Research on Cancer for the INHANCE (International Head and Neck Cancer Epidemiology) consortium.

The INHANCE researchers pooled individual-level data from 15,227 case and control subjects, including 1,072 case subjects and 5,775 controls who never used tobacco products. They also identified 1,598 case subjects and 4,051 control subjects who never drink alcohol.

Approximately 24% (95% CI=16% to 31%) of the cases of head and neck cancer among never-drinkers were attributable to cigarette smoking, assuming a causal relationship, the researchers wrote. They also noted that among never-users of tobacco, 7% (95% CI=-4% to 16%) of the case of head and neck cancer were attributable to alcohol drinking, again assuming a causal relationship.

The association between smoking and head and neck cancers was stronger in studies conducted in Europe and South American, than it was in studies done in North America, India, and Sudan, the researchers said.

They acknowledged that their study was limited by use of a global value for the percentage of alcohol present in different alcoholic beverages, which could have resulted in a misclassification of exposure to ethanol. Moreover, recall bias could influence the observed associations, they said.

The researchers also noted regional differences in the classification of never drinkers or never smokers. Finally, the authors said the data were not adjusted for some potential confounders including human papillomavirus infection or diet/nutrition.