Human papillomavirus-16 modifies the exposure-response between alcohol and tobacco with head and neck squamous cell carcinoma

Abstract

1745

Introduction: Positive serology to human papillomavirus-16 (HPV16) has been found to increase the risk of squamous cell carcinoma of the head and neck (HNSCC) more than four-fold; however, it is not known how HPV16 influences the risk of HNSCC from alcohol and tobacco. Methods: Cases of incident HNSCC from the Greater Boston Metropolitan area diagnosed between December 1999 to December 2003 were identified from nine medical facilities in Boston, Massachusetts. Population-based controls were frequency-matched to cases on age, gender, and town of residence. Subjects completed a questionnaire that collected decade-specific data on smoking and alcohol consumption. Subjects also provided sera samples from which presence of HPV type 16 antibodies were detected. Unconditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) as measures of disease risk. Categories for alcohol (in average drinks per week (dpw)) and tobacco (as pack-years) were analyzed in quartiles, as determined by the combined distribution of cases and controls (alcohol quartiles: 0-3, 4-7, 8-23, ≥24 dpw; tobacco quartiles: never smoker, >0-21, 22-46, ≥47 pack-yrs). ORs were relative to the lowest consumption categories for alcohol and tobacco and controlled for age, gender, race, education and either alcohol or tobacco. Results: Alcohol, tobacco, and HPV16 serology was available for 486 cases of HNSCC and 549 controls. HPV16 status was a statistically significant modifier for the association between alcohol and HNSCC (p=0.03). Among those who were HPV16 negative, increasing alcohol consumption increased risk of HNSCC (OR_quartile2=1.1 (0.7, 1.8), OR_quartile3=1.9 (1.2, 3.1), OR_quartile4=4.0 (2.4, 6.7)). Among the HPV16 positive, HNSCC risk was no longer influenced by alcohol consumption (OR_quartile2=0.7 (0.2, 1.8), OR_quartile3=0.4 (0.2, 1.2), OR_quartile4=0.6 (0.2, 1.9)). Smoking remained a risk factor regardless of HPV16 serology, although the risk of HNSCC from smoking was stronger among the HPV16 negative (OR_quartile2=1.3 (0.8, 2.1), OR_quartile3=2.1 (1.3, 3.4), OR_quartile4=4.6 (2.8, 7.5)) than among the HPV16 positive (OR_quartile2=1.0 (0.4, 2.5), OR_quartile3=1.4 (0.5, 3.7), OR_quartile4=2.4 (0.7, 7.7)), which was a borderline statistically significant interaction (p=0.09). The joint effects between smoking and alcohol on HNSCC risk were only evident among those who were HPV16 negative. Conclusions: Our findings suggest that HPV16 abrogates the synergy of alcohol and tobacco in the induction of HNSCC.