Does human papillomavirus modify the risk of oropharyngeal cancer related to smoking and alcohol drinking? A systematic review and meta-analysis

Objective To synthesize evidence for interactions of traditional oropharyngeal squamous cell carcinoma (OPSCC) risk factors-tobacco smoking and alcohol drinking-with human papillomavirus (HPV). Data Sources MEDLINE, Embase, Cochrane Database of Systematic Reviews, ProQuest, and Global Health were searched with no restrictions on language or publication date. Methods All case-control studies assessing interactions between these factors in OPSCC were considered. Quality was assessed using the Newcastle-Ottawa Scale for case-control studies. The main outcome was the OR for developing OPSCC for the following interactions: (1) HPV and smoking, (2) HPV and alcohol drinking, and (3) HPV, alcohol drinking, and smoking. Interactions were assessed from stratified analysis (by HPV status) and/or joint effect analysis (synergy index and multiplicative index). Results The search provided 3084 relevant studies, of which 9 were included. In the stratified analysis, the OR of developing OPSCC among smokers with HPV was less than that among smokers without HPV. A similar pattern was observed for alcohol drinking. This effect persisted among smokers and heavy alcohol drinkers with HPV compared with those without HPV. Joint effect analysis on the additive scale showed sub-additive antagonistic interactions between HPV and smoking, and between HPV and alcohol. On the multiplicative scale, sub-multiplicative interactions were found between HPV and smoking, and HPV and alcohol. Conclusions This meta-analysis suggests a negative directed interaction of HPV and smoking; and HPV and heavy alcohol drinking in the development of primary OPSCC on stratified analysis and joint effect analysis. Level of Evidence 3A.