In this study, we found that risk for breast cancer was weakly inversely associated with regular green tea drinking. Among pre-menopausal women, this relationship appeared to be related to years of drinking, and measures of frequency and amount. Among post-menopausal women, the relationship was stronger with recent use and lower amounts of intake. Among both pre- and post-menopausal women, there was a possible U-shaped relationship between amount of tea leaves consumed per month and risk of breast cancer. The relationship between green tea intake and breast cancer risk did not vary according to COMT rs4680 genotype.
Black tea has been evaluated in several previous studies (53
) and most, including a meta-analysis (59
), found no relationship between tea and breast cancer. Few studies have reported on use of green tea and risk for breast cancer. Three cohort studies, all conducted in Japan where green tea consumption is highly prevalent, did not find a statistically significant relationship between breast cancer and green tea (60
). However, these studies may have been limited by a small number of cases, inability to comprehensively control for confounders, and a small unexposed reference group. Nonetheless, in two meta-analyses of these studies, the summary ORs, although not statistically significant, were 0.85 and 0.89 which are consistent with our finding of a weak association (62
). Three case-control studies, a population-based study of Asian Americans in Los Angeles (64
), a population-based study of Chinese living in Singapore(65
), and a hospital-based study in Southeast China (66
), have evaluated green tea drinking and breast cancer risk. All three studies found inverse associations with at least one measure of tea drinking, although the Singapore study only observed an association among a subset of women defined by genotype (67
). Most recently, the hospital-based Chinese study found all measures of green tea drinking were associated with reduced breast cancer risk, and many were associated in a dose-dependent manner(68
). These findings are in contrast to our observation of a weak association between green tea consumption and breast cancer risk in a large population-based study. There are several notable differences between our and previous studies including the measurement, definition and prevalence of tea consumption. The Japanese studies, as mentioned previously, had virtually no abstainers from tea consumption and very high consumption patterns based on the number of cups per day(69
). In contrast to our study, consumers in two of the previous case-control studies, reported much higher prevalence of any green tea consumption, but much lower average intake of tea, with most controls consuming less than daily or weekly(71
). In our study, nearly one-third of the population consumes tea at least twice a week. These other studies reported only one measure of tea intake, cups, mL, or frequency. Only our study and the previous study in a Chinese population (73
) had multiple measurements of tea intake and only our study included strength of brew. Over half of the hospital-based controls in the previous Chinese study reported drinking green tea and most also consumed the tea at least daily, a prevalence that is much higher but a frequency that is similar to our study. However, even though the prevalence was much higher than in our study, the amount of dried tea leaves consumed per year was substantially less than the amount of tea leaves reported by our population-based controls in another part of China.
Several plausible mechanisms have been proposed for green tea’s possible chemopreventive properties. Green tea components, among other activities, affect cell cycle arrest(74
), have anti-oxidant properties(75
), down regulate telomerase(76
), inhibit vascular endothelial growth factor(77
), suppress cellular proliferation(80
), upregulate or maintain intercellular gap junction communication(82
), and increase apoptosis(84
). In addition, green tea has also been described as having anti-estrogenic properties. As a hormone-dependent cancer, estrogen plays a critical role in breast carcinogenesis. In vitro
studies found that tea polyphenols inhibit aromatase, the key enzyme converting androgens to estrone or estradiol (85
). Furthermore, consumption of green tea was associated with reduced levels of estrogens, estrone and estradiol, among pre- and post-menopausal women(86
). No previous study has evaluated the association between green tea and breast cancer by menopausal status. Our study suggests the association between green tea drinking and breast cancer risk may differ by menopausal status and the inverse association with measures of longer duration and larger dose may be more pronounced among pre-menopausal women, while only recent and mild use is associated with decreased risk among post-menopausal women. It is also possible that we did not observe consistent relationships among post-menopausal women because the sample size was smaller than for pre-menopausal women.
There are also plausible mechanisms by which COMT and green tea may interact to affect breast cancer risk. On the one hand, low COMT activity may lead to elevated levels of catechol estrogens, but, on the other hand, low activity may also lead to a slower metabolism of tea polyphenols which have anti-estrogenic effects. A population-based study of Asian Americans is the only previous study to evaluate tea consumption, including green tea, COMT
rs4680 genotype, and breast cancer risk (89
). They found the inverse relationship they initially observed between tea intake and breast cancer (90
) was primarily limited to those individuals with at least one low activity allele (91
). We, however, found no evidence COMT
rs4680 genotype affects the relationship between green tea consumption and breast cancer risk in a Chinese population. Based on the findings in our population, it appears the effect of green tea is independent of the rate of O
-methylation of both the tea polyphenols and of catechol estrogens or that the anti-carcinogenic effects of green tea, including anti-estrogenic effects, are much greater than any small genetic differences in COMT activity.
As in all case-control studies, recall bias is a potential concern. However, nearly all regular drinkers reported daily use of green tea indicating that, in this population, green tea drinking is a frequent habit which may facilitate recall for both cases and controls. Interestingly, the proportion of controls reporting regular green tea consumption remained constant in both study phases (31%) which may also indicate green tea consumption patterns are relatively stable among women in Shanghai. Although we controlled for several potential confounders, it is possible that residual confounding remained. It is also possible that some of the associations we observed were due solely to chance or were a result of multiple comparisons. This study has several strengths. This is a population-based study with a large sample size and high response rates. We were able to evaluate several measures of green tea consumption including age of initiation, duration of use, brew strength, and quantity of tea leaves. We were also able to evaluate both pre-and post-menopausal breast cancer risk related to green tea drinking.
In summary, this population-based case-control study found a weak independent role for green tea consumption in breast cancer risk. The relationship may differ by menopausal status. Future population-based or cohort studies in populations with frequent and long-term green tea consumption are needed to further investigate green tea’s potential role in breast carcinogenesis and to elucidate the part menopausal status may play in this role.