Resveratrol has been shown to have anticarcinogenic activity. We previously found that resveratrol inhibited growth and induced apoptosis in 2 human melanoma cell lines. In this study we determined whether resveratrol would inhibit human melanoma xenograft growth. Athymic mice received control diets or diets containing 110 micromol/L or 263 micromol/L resveratrol, 2 wk prior to subcutaneous injection of the tumor cells. Tumor growth was measured during a 3-wk period. Metabolism of resveratrol was assayed by bolus gavage of 75 mg/kg resveratrol in tumor-bearing and nontumor-bearing mice. Pellets containing 10-100 mg resveratrol were implanted into the mice, next to newly palpated tumors, and tumor growth determined. We also determined the effect of a major resveratrol metabolite, piceatannol, on experimental lung metastasis. Resveratrol, at any concentration tested, did not have a statistically significant effect on tumor growth. The higher levels of resveratrol tested (0.006% in food or 100 mg in slow-release pellets) tended to stimulate tumor growth (P = 0.08-0.09). Resveratrol and its major metabolites, resveratrol glucuronide and piceatannol, were found in serum, liver, skin, and tumor tissue. Piceatannol did not affect the in vitro growth of a murine melanoma cell line, but significantly stimulated the number of lung metastases when these melanoma cells were directly injected into the tail vein of the mouse. These results suggest that resveratrol is not likely to be useful in the treatment of melanoma and that the effects of phytochemicals on cell cultures may not translate to the whole animal system.