Esophageal Cancer Case-Control Study In Western Kenya Do Alcohol And Tobacco Contribute?

Diana MENYA, Moi University, Kenya

1 Department of Epidemiology and Biostatistics, Moi University, Eldoret, Kenya
2 Department of Environment and Radiation, International Agency for Research on Cancer, Lyon, France
3 Department of Environmental Biology and Health, University of Eldoret, Eldoret, Kenya
4 Environmental and Occupational Tract, University of Haifa, Haifa, Israel
5 Department of Behavioral Sciences, Moi University, Eldoret, Kenya
6 Department of Immunology, Moi University, Eldoret, Kenya

Purpose: In Kenya, esophageal squamous cell carcinoma (ESCC) is the most common cancer among men and third most common in women. In Eldoret, Western Kenya, we have commenced an ESCC case-control study to investigate the environmental, behavioral, nutritional and genetic aetiology. We report on the influence of 2 strong risk factors from other settings: alcohol and tobacco.
Methods: We recruited cases, defined as histologically confirmed ESCC patients diagnosed at Moi Teaching and Referral Hospital between August 2013-October 2014, and age and frequency-matched hospital-based controls. Participants were interviewed using a pre-tested e-questionnaire. Odds ratios (OR) minimally adjusted for age and gender and 95% confidence intervals (CI) were computed using logistic regression
Results: Mean age of the 132 cases was 58.4 and of 158 controls 56.7 years. Amongst cases, the male:female ratio was 2:1; 61.3% were of Kalenjin ethnicity, 23% Luhya and 8% Luo. Ever consumption of alcohol (prevalence 75% in cases) and tobacco (61%) were each associated with ESCC (ORs 1.8 (1.0-3.2) and 2.3 (1.4-3.9) respectively). ORs were greatest in consumers of both habits (OR 2.4 (1.3-4.6) compared to non-drinkers non-smokers, which were largely driven by tobacco smoking rather than snuff/chewing and by drinking local brews/spirits busaa and chang’aa. These effects were independent of fresh fruit and traditional vegetables consumption which were protective for ESCC (p=0.008 and p=0.014). We will also present initial findings for oral health, oral health, biomass fuels, and blood group.
Conclusion: This study supports the concept of multifactorial etiology for ESSC. Initial results point to modifiable behavioral factors concerning alcohol and tobacco. Continued recruitment is needed to confirm results and to disentangle the specific effects of these co-present habits to better enlighten the most effective primary prevention strategies.
 Funding Source: US NIH R21CA191965