COFFEE AND TEA CONSUMPTION AND GASTRIC CANCER: A POOLED ANALYSIS IN AN INTERNATIONAL CONSORTIUM OF EPIDEMIOLOGICAL STUDIES.

The main focus of my doctoral research was to examine the effect of coffee and tea drinking on gastric (stomach) cancer risk. Coffee and tea are among the most popular drinks worldwide after water, and their consumption is expected to rise by an average of 17% over the next several years. The World Cancer Research Fund and the American Institute for Cancer Research (WCRF/AICR) concluded in their most recent report (2018) that there is insufficient data to draw any conclusions about the relationship between coffee and tea consumption and gastric cancer. To estimate their relationship with gastric cancer risk, individual-participant pooled analyses in a unique global consortium of epidemiological studies on gastric cancer - the Stomach cancer Pooling (StoP) project- were carried out. The StoP project included thirty case-control (CC) studies and five nested case-control within the cohort (NCC) studies, from seventeen different countries worldwide. For each study, coffee and tea consumption were assessed using self- or interviewer-administered food frequency questionnaires (FFQ) prior to the diagnosis of gastric cancer cases or prior to the onset of disease, hospital admission for hospital-based controls, or recruitment for population-based controls. Data on coffee intake were available from eighteen studies, which included 8,198 gastric cancer cases and 21,419 controls, while on tea twenty-two studies totaled 9,438 gastric cancer cases and 20,451 controls. Coffee and tea consumption was measured by the number of cups, times, or frequency of consumption reported in each study and further expressed in the standard unit of cups per day. Coffee was considered either caffeinated coffee or decaffeinated coffee separately, as well as their combined intake. Various types of tea reported were calculated as total tea intake. In addition, the temperature and strength at which tea was consumed were assessed. Logistic generalized linear mixed-effects models including terms Dissertation Advisor: Professor Carlo La Vecchia Georgia Martimianaki for sex, age, study design, Helicobacter pylori (H. pylori) infection, and several other main recognized risk factors for gastric cancer were used to estimate the associations between gastric cancer and coffee and tea consumption, through both two- and one-stage modeling analyses. Subgroup analyses across strata of several factors and dose-response relationships were also carried out. Multinomial mixed-effects models were also applied to estimate the ORs and corresponding 95% CIs of gastric cancer by the anatomical site (cardia and non-cardia) and histological type (intestinal, diffuse, and mixed/unspecified by Lauren categorization) for regular tea drinkers and high coffee drinkers. Heterogeneity between strata was evaluated by Cohran’s Q test and by the I2 test between studies. The results showed a slightly inverse association (OR: 0.92, 95% CI: 0.82–1.05) between tea drinking and gastric cancer but no relevant association (OR: 1.03, 95% CI: 0.94-1.13) between coffee consumption and gastric cancer. Regular tea drinkers had a lower gastric cardia cancer risk (OR: 0.64, 95% CI: 0.49-0.84), than heavy coffee drinkers, who were positively associated (OR:1.61, 95% CI: 1.27–2.05) with gastric cardia cancer. Contrary to tea intake, which was more strongly linked to lower risk in Asian countries (OR: 0.67, 95% CI: 0.49-0.91, in studies from China and Japan), there was no evidence of regional differences in the effect of coffee consumption on the risk of gastric cancer. Consuming hot or very hot tea did not increase the risk of gastric cancer, drinking warm or cold tea was related to a lower risk. Last but not least, I also discuss additional research projects I have undertaken through my doctoral studies. These include other dietary analyses within the Stomach Cancer Pooling project consortium, my work with the Hellenic Health Foundation to evaluate the usual diet of the Greek adult population, and a project I carried out during my six-month visiting research period at Harvard T.H. Chan School of Public Health on the association between olive oil consumption and prostate cancer in the US Health Professionals Follow-up Study (HPFS) and the Greek part of the European Prospective Investigation into Cancer and Nutrition (EPIC-Greece) cohort study.