A Cochrane systematic review on green tea for cancer (protocol stage) published in 2004 does not include any studies . This review identified 48 clinical studies investigating the consumption of green tea and its association with the risk of developing cancer, including 42 epidemiological studies [17–58], one phase I trial , four randomized trials [60–63] and one meta-analysis  (Additional file 1). Twenty-two and 20 of the 42 epidemiological studies were cohort studies and case-control studies respectively, among which two studies [26, 43] reported data from four cohort studies and one cohort study [17, 18] and one case-control study [41, 42] were reported twice. All studies were published between 1984 and 2008; 44 studies were published in international journals, three in Chinese [28, 32, 56] and one in Japanese .
Participants in the included studies ranged from healthy individuals, pre-cancer patients to cancer patients. Several studies reported various cancers in their cases or cohorts, but in this review these data were presented separately in relevant categories. Average sample size for cohort studies was 31,798 (from 52 to 102,137 per study), 1,099 for case-control studies (from 213 to 3,818 per study) and 182 for randomized trials (from 60 to 400 per trial). Decaffeinated green tea, tea polyphenols (500 mg or 1000 mg) or catechins was used for 3–12 months in randomized trials, whereas regular drinking tea was used for a fixed period or lifelong time in epidemiological studies. The outcomes reported from randomized trials included 8-hydroxydeoxyguanosine (8-OHdG) in urine (an indicator for oxidative DNA damage) [60, 62], histopathological examination  and incidence of cancer . Risk of cancer development was the main outcome for epidemiological studies; a few studies reported mortality [20, 23], recurrence [47, 49], survival of cancer  and sister chromatid exchange (SCE) rates .
Four randomized trials were placebo-controlled, double-blinded (three) and of acceptable methodological quality in terms of randomization, blinding and reporting of the studies. According to the generic quality grading for all included studies, 29 studies (60%) were evaluated as good (A) and the remaining as fair (B).
Caner in general (5 studies)
A cohort study of 8,552 people with nine years of follow-up showed a negative association of green tea consumption with cancer incidence, especially among Japanese women drinking more than ten cups a day (RR 0.59; 95% CI, 0.35–0.98) [17, 18]. However, a larger cohort study with 38,540 people in Japan did not show an association between green tea consumption and sum incidence of all cancers (RR 1.0; 95% CI, 0.91–1.1 for those drinking two to four times per day; RR 0.98; 95%CI, 0.88–1.1 for those drinking five times or more per day, both compared with those drinking one time or less per day) . Another large cohort study found no significant benefit in terms of cancer mortality among a total of 1,134 cancer patients who consumed green tea and those who did not . A randomized trial comparing green tea, black tea with water in 143 heavy smokers found significant decrease in 8-OHdG levels after a 4-month intervention . A prospective cohort study in 52 male smokers demonstrated that drinking green tea inhibited cigarette-induced increase in sister chromatid exchange rates .
Oral and esophageal cancer (8 studies)
A prospective cohort study followed 20,550 men and 29,671 women for an average of 10.3 years and estimated the HRs (95% CI) in oral cancer . For women, the HRs (95% CI) were 0.51 (0.10–2.68), 0.60 (0.17–2.10) and 0.31 (0.09–1.07) for green tea consumption of one to two, three to four and five or more cups per day respectively, compared with those who drank less than one cup per day (P for trend, 0.08). For men, no trend was observed.
Inconsistent findings exist in case-control and cohort studies on green tea drinking and esophageal cancer [23–28]. Two population-based case-control studies of 3,049 subjects found a protective effect of green tea drinking against esophageal cancer among women [24, 25], while another case-control study of 1,043 subjects showed 39% decrease of risk of esophageal cancer among alcoholic drinkers and 31% decrease among smokers . The tea polyphenol epigallocatechin in urine was inversely associated with cancer risk when the data of gastric and esophageal cancer sub-sites were combined , indicating protective effect of green tea. However, in terms of death, a retrospective cohort and a pooled analysis of two cohort studies with a total of 26,723 subjects demonstrated a positive association (HR 1.67; 95% CI, 0.89–3.16) (P for trend, 0.04) of drinking green tea and mortality of esophageal cancer among men [23, 26]. A randomized controlled trial with 400 participants who were pathologically confirmed to have esophageal precancerous lesions did not show benefit of decaffeinated green tea for alleviating precancerous lesions or abnormal cell proliferation compared with calcium as placebo .
Stomach cancer (11 studies)
Seven epidemiological studies (one cohort and six case-control studies) with a total number of 77,777 subjects showed inverse association of green tea consumption (urine polyphenol epigallocatechin in one study) and the risk reduction of stomach cancer [27, 28, 30, 32, 34, 36, 37]. The cohort study with 72,943 subjects showed benefit for women who consumed five or more cups of green tea per day (RR 0.51; 95% CI, 0.30–0.86) compared with one cup per day . However, four studies including two cohort studies with 102,179 subjects did not show an inverse association of green tea consumption and risk reduction of stomach cancer or cancer-caused death [29, 30, 33, 35].
Pancreatic cancer (4 studies)
Two case-control studies (in three publications) with 522 pancreatic cancer patients and 1,694 controls showed an inverse association of drinking green tea and the risk of pancreatic cancer [38, 41, 42]. By contrast, a hospital-based case-control study of 124 patients and 124 controls demonstrated a positive association of drinking five cups or more green tea per day and pancreatic cancer . A population-based cohort study involving 102,137 participants with 11 years of follow-up did not find any association of the risk of pancreatic cancer and drinking green tea .
Liver cancer (2 studies)
A population-based case-control study involving 204 patients and 415 controls reported that drinking green tea reduced the risk of liver cancer by 78% (OR 0.25; 95% CI, 0.11–0.57) among alcoholic drinkers and by 43% among smokers . A randomized, double-blinded, placebo-controlled trial in 124 individuals with sero-positive HBsAg and aflatoxin-albumin adducts showed a significant decrease of 8-hydroxydeoxyguanosine after three months of green tea polyphenols intake .
Biliary ducts cancer (1 study)
Statistical analysis showed that green tea consumption was positively associated with the mortality of biliary duct cancer in a retrospective cohort study (P ≤ 0.01) .
Colorectal cancer (6 studies)
Two case-control studies (in three publications) involving 2,036 patients with colorectal cancer and 2,130 controls found that drinking green tea reduced the risk of colorectal cancer [41, 42, 44]. Gender difference was observed in two cohort studies [45, 46]. A prospective cohort study that followed over 60,000 subjects for an average of 8.9 years found no statistically significant difference between green tea drinkers and non-drinkers (RR 1.12; 95% CI, 0.97–1.29) . Another cohort study with six years of follow-up on 69,710 women found significant dose-response relationship (RR 0.63; 95% CI, 0.45–0.88) between regular and non-regular green tea drinkers . By contrast, a pooled analysis from two cohort studies on over 60,000 subjects showed no association between drinking green tea and a lower risk of colorectal cancer .
Breast cancer (5 studies)
In a meta-analysis of two prospective cohorts of 35,004 Japanese women , green tea intake was not associated with a lower risk of breast cancer (222 cases); and the multivariate RR for women drinking more than five cups of green tea was 0.84 (95% CI, -0.57–1.24; P = 0.69) compared with those drinking less than one cup per day. One case-control study showed significantly reduced risk of breast cancer by regular drinking a large amount of green tea . However, a cohort study did not find an association of green tea intake with lower risk of breast cancer . Another two cohort studies showed reduced recurrence of breast cancer among patients at stage I and II with high consumption of green tea (more than three cups per day) [47, 49].
Lung cancer (4 studies)
A randomized trial compared green tea and black tea with water in 143 heavy smokers for 4 months . The content of 8-OHdG in urine was reduced among the subjects in green tea group, but not in black tea or water groups. A cohort study of 52 male adults, found that green tea drinking blocked cigarette-induced increase in sister chromatid exchange rates, suggesting potential protection against lung cancer . Consumption of green tea was found to be associated with a reduced risk of lung cancer among non-smoking women in one case-control study . A phase I dose finding study showed the maximum tolerated dose of green tea extract as 3 g/m2 per day in patients with advanced lung cancer. The dose-limiting toxicities were diarrhea, nausea and hypertension .
Prostate cancer (4 studies)
A double-blind, placebo-controlled trial testing green tea catechins (600 mg per day for one year) significantly reduced the incidence of prostate cancer in a group of 60 volunteers with high-grade prostate intraepithelial neoplasia; no significant adverse effect was reported . A case-control study found prostate cancer risk declined with increasing frequency, duration and quantity of green tea consumption . A cohort study in 19,561 Japanese men showed that green tea intake was not associated with a lower risk of prostate cancer (HR 0.85; 95% CI, 0.50–1.43) between men drinking five or more cups and less than one cup per day . However, another recent cohort study with a larger sample size (n = 49,920) in Japan suggested that green tea was associated with a decreased risk of advanced prostate cancer (RR 0.52; 95% CI, 0.28–0.96) in men drinking five or more cups compared with those drinking less than one cup per day .
Urinary bladder cancer (1 study and 1 ongoing trial)
The risk of urinary bladder cancer was significantly reduced in women who consumed matcha (a powdered green tea) in a case-control study (n = 882) . A phase II randomized, double-blind, placebo-controlled, multi-center trial (n = 330) is currently carried out in the United States .
Endometrial cancer (1 study)
A population-based case-control study (n = 2082) suggested that regular green tea drinking reduced the risk of endometrial cancer (OR 0.74; 95% CI, 0.54–1.01) in pre-menopausal women .
Ovarian cancer (1 study)
A cohort study (n = 254) suggested that increasing the post-diagnosis consumption of green tea may boost the survival of patients of epithelial ovarian cancer .
Adult leukemia (1 study)
A case-control study (n = 217) demonstrated statistical association (OR 0.51; 95% CI 0.27–0.96) between higher intake of green tea and reduced risk of adult leukemia in a dose-response manner .
Safety of green tea
Green tea, as a popularly consumed beverage, is relatively non-toxic . Phase I trial in 17 patients with advanced lung cancer showed that the maximum tolerated dose of green tea extract was 3 g/m2 per day . No severe adverse effects have been reported in association with the medicinal use of green tea . Consumption of high doses of green tea or green tea extract (i.e. 5–6 litters per day) may cause nausea, vomiting, abdominal bloating/pain, dyspepsia, flatulence and diarrhea [59, 67]. Excessive consumption of caffeine from green tea may also cause central nervous system stimulation such as dizziness, insomnia, tremors, restlessness, confusion, diuresis (i.e. increasing urine output), heart rate irregularities and psychomotor agitation .
Human studies did not show severe adverse effects among volunteers who took 15 tablets of green tea per day (i.e. 2.25 g green tea extracts, 337.5 mg EGCG and 135 mg caffeine) for 6 months [68, 69]. A randomized, placebo-controlled trial (n = 40) found no adverse effect in healthy individuals who took green tea polyphenols in the amount equivalent to the EGCG content in 8–16 cups of green tea once a day or twice a day in divided doses for four weeks .