Background: Cancer remains a major global public health challenge, with lifestyle factors, including diet, playing an important role in its development. Tea, produced from Camellia sinensis, is one of the most commonly consumed beverages worldwide. Although numerous systematic reviews (SRs) and meta-analyses (MAs) have examined the association between tea consumption and cancer risk, findings remain inconclusive. We conducted an umbrella review within the World Cancer Research Fund International CUP Global framework to evaluate tea consumption in relation to cancer incidence and mortality.
Objectives: This umbrella review aimed to systematically synthesise and evaluate existing meta-analytical evidence and assess the strength and credibility of reported associations.
Methods: We conducted comprehensive searches of PubMed, Embase, and Central databases up to May 2025 to identify SRs with MAs of observational cohort studies and randomised controlled trials (RCTs) investigating the association between tea consumption (total or by specific type) and cancer incidence or mortality across all sites. For each association, summary relative risks (RRs) and 95% confidence intervals (CIs) were estimated using random-effects models, with heterogeneity quantified using the I² statistic. To explore credibility of the evidence, we calculated the fail-safe number (FSN), which estimates how many additional studies of average sample size with null results would be required to reduce a statistically significant meta-analytic finding to non-significant, as well as conditional power (CP), to determine the robustness of current meta-analytic evidence against future studies.
Results: These preliminary findings summarise associations between tea consumption and cancer risk. Sixteen MAs were included, comprising 73 associations of total, green or black tea consumption with cancer risk (68 incidence, 4 incidence/mortality, and 1 mortality) across 20 anatomical sites. Higher total tea consumption was inversely associated with bladder cancer (RRper 1 cup/day: 0.95; 95%CI: 0.91-0.99; I²: 4%; 1,805 cases; 5 cohorts) and glioma incidence (RRhigh vs. low: 0.81; 95%CI: 0.70-0.95; I²: 4%; 2,383 cases; 6 cohorts). Green tea consumption was inversely associated with liver cancer incidence/mortality in Asian populations (RRhigh vs. low: 0.87; 95%CI: 0.78-0.98; I²: 26%; 3,694 cases; 9 cohorts). Examining incidence separately yielded similar associations with wider CIs (RRhigh vs. low: 0.89; 95%CI: 0.72-1.09; I²: 41.9%; 1,289 cases; 7 cohorts). There was an indication of inverse associations between high green tea consumption and biliary tract cancer, and dose-response associations between tea and lower risk of liver and ovarian cancers. No other associations were observed. FSN suggested that MAs showing evidence of associations are unlikely to be substantially affected by future studies. Exceptions included the dose-response association between total tea and liver cancer and the high vs. low association of green tea association with biliary tract cancer, where only about 3 additional studies, based on the current 4 and 3 studies, respectively, would likely be sufficient to meaningfully influence the summary estimates.
Conclusions/Implications: Meta-analytical evidence suggests that tea consumption is inversely associated with glioma, liver, and bladder cancers. However, current evidence remains limited, as most MAs rely on high vs. low analyses rather than dose-response approaches and include a small number of primary studies.