Smoking and alcohol consumption as predictors of outcome in head and neck cancer: results from the Head and Neck 5000 study.


Session type:


Rhona Beynon1,Steve Thomas2,Andrea Waylen1,Chris Penfold1,Richard Martin1,Andy Ness1
1The University of Bristol,2The University of Bristol, Bristol Royal Infirmary



Head and neck cancer (HNC) is an important cause of morbidity and mortality.  Survival is poor and has only recently started to improve.  Tobacco and alcohol consumption are established aetiological risk factors.  Small studies suggest that tobacco and alcohol use are also important determinants of survival.  We used data from Head and Neck 5000, a large prospective clinical cohort of people with HNC, to examine the association between tobacco and alcohol use and survival.


All participants with cancers of the oral cavity, larynx and oropharynx were included.  Cox regression was used to estimate the mortality hazard ratios (HR) for pre-diagnosis smoking status and alcohol consumption.  Models were adjusted for age, gender, tumour site and stage, treatment intent and co-morbidity. We repeated analyses stratified by tumour site.


Complete data were available for 2881 people, of whom 531 died during 6665 person-years of follow-up (mean follow-up 2.3 years). Being a current smoker at diagnosis was associated with higher risk of death (HR=2.34, 95%CI 1.77 to 3.11).  The risk of death was higher for people who consumed alcohol 6-7days/week (HR=1.43, 95%CI 1.15 to 1.78) compared with 0-1 days/week. The strongest associations were for oropharyngeal cancers (HR 3.38, 95%CI 2.19 to 5.24 for current smokers and HR 2.09, 95%CI 1.46 to 2.99 for daily/near-daily drinkers).  Associations were weaker for laryngeal cancers (HR 1.56, 95%CI 0.75 to 3.25 for current smokers and HR 1.35, 95% CI 0.88 to 2.09 for daily/near-daily drinkers). For oral cancers, the association was weaker for current smokers (HR 1.32, 95%CI 0.85 to 2.02); there was no association between drinking on most days (HR 0.95, 95% CI 0.65 to 1.34) and survival.  


Our results suggest that tobacco and alcohol use prior to a HNC diagnosis are predictive of survival, particularly for people with oropharyngeal cancers.