Green tea consumption and breast cancer risk: three recent meta-analyses

Abstract

We read with great interest the recent meta-analysis conducted by Ogunleye et al. [1], which has examined the association between green tea consumption and risk of breast cancer incidence or recurrence, using all available epidemiologic evidence to date. They found that increased green tea consumption (more than three cups a day) was inversely associated with breast cancer recurrence [pooled relative risk (RR) = 0.73, 95% confidence interval (CI): 0.56–0.96). An analysis of case–control studies of breast cancer incidence suggested an inverse association with a pooled RR of 0.81 (95% CI: 0.75, 0.88) while no association was found among cohort studies of breast cancer incidence. However, I have several concerns. First, this meta-analysis was similar to the researches by Sun et al. [2] and Seely et al. [3] and was aimed to derive a more precise estimation since some eligible and new studies could be found. However, these meta-analyses brought a new conflicting result in the subgroup analysis by case–control studies. Ogunleye et al. [1] suggested an inverse association with a pooled RR of 0.81 (95% CI: 0.75, 0.88). Seely et al. [3] found that the pooled RR of developing breast cancer for the highest levels of green tea consumption in cohort studies was 0.89 (95% CI, 0.71–1.1), and in case control studies, the odds ratio (OR) was 0.44 (95% CI, 0.14–1.31). Second, one methodological issue need to be addressed concerning the meta-analysis by Ogunleye et al. [1]. Importantly, they found that combining all studies of breast cancer incidence resulted in significant heterogeneity, but they calculated summary effect estimates through metaanalysis using a fixed effect model. In Fig. 2c, we can easily found that there was a significant heterogeneity (P \ 0.0001) for meta-analysis of case–control studies of breast cancer risk. Why not use random effects models? Third, dose-response meta-analysis was not conducted or mentioned in the article by Ogunleye et al. [1]. It would be better if they used the method proposed by Greenland [4] and Orsini [5] to compute study-specific slopes (linear trends) from the correlated natural logarithm of the OR across categories of green tea consumption. Despite the above, the results found by Ogunleye et al. [1] confirm the main results compared with other metaanalyses [2, 3]. They have reached important conclusions that that increased green tea consumption may be inversely associated with risk of breast cancer recurrence.