Inhibitory effects of green tea on carcinogenesis have been investigated in numerous laboratory studies using (-)-epigallocatechin gallate (EGCG) or crude green tea extract, and there is also some epidemiologic evidence. Further, EGCG has been reported to inhibit the growth of cancer cells, lung metastasis in an animal model, and urokinase activity. In this study, we first examined the association between consumption of green tea prior to clinical cancer onset and various clinical parameters assessed at surgery among 472 patients with stage I, II, and III breast cancer. We found that increased consumption of green tea was closely associated with decreased numbers of axillary lymph node metastases among premenopausal patients with stage I and II breast cancer and with increased expression of progesterone receptor (PgR) and estrogen receptor (ER) among postmenopausal ones. Since these are potential prognostic factors, we then investigated the prognosis of breast cancer with special reference to consumption of green tea, in a follow-up study of these patients. We found that increased consumption of green tea was correlated with decreased recurrence of stage I and II breast cancer (P < 0.05 for crude disease-free survival); the recurrence rate was 16.7 or 24.3% among those consuming > or = 5 cups or < or = 4 cups per day, respectively, in a seven-year follow-up of stage I and II breast cancer, and the relative risk of recurrence was 0.564 (95% confidence interval, 0.350-0.911) after adjustment for other lifestyle factors. However, no improvement in prognosis was observed in stage III breast cancer. Our results indicate that increased consumption of green tea prior to clinical cancer onset is significantly associated with improved prognosis of stage I and II breast cancer, and this association may be related to a modifying effect of green tea on the clinical characteristics of the cancer.