Resveratrol worsens survival in SCID mice with prostate cancer xenografts in a cell-line specific manner, through paradoxical effects on oncogenic pathways.

BACKGROUND: Resveratrol increases lifespan and decreases the risk of many cancers. We hypothesized resveratrol will slow the growth of human prostate cancer xenografts.
METHODS: SCID mice were fed Western diet (40% fat, 44% carbohydrate, 16% protein by kcal). One week later, human prostate cancer cells, either LAPC-4 (151 mice) or LNCaP (94 mice) were injected subcutaneously. Three weeks after injection, LAPC-4 mice were randomized to Western diet (control group), Western diet plus resveratrol 50 mg/kg/day, or Western diet plus resveratrol 100 mg/kg/day. The LNCaP mice were randomized to Western diet or Western diet plus resveratrol 50 mg/kg/day. Mice were sacrificed when tumors reached 1,000 mm(3). Survival differences among groups were assessed using Cox proportional hazards. Serum insulin and IGF axis were assessed using ELISAs. Gene expression was analyzed using Affymetrix gene arrays.
RESULTS: Compared to control in the LAPC-4 study, resveratrol was associated with decreased survival (50 mg/kg/day--HR 1.53, P = 0.04; 100 mg/kg/day--HR 1.22, P = 0.32). In the LNCaP study, resveratrol did not change survival (HR 0.77, P = 0.22). In combined analysis of both resveratrol 50 mg/kg/day groups, IGF-1 was decreased (P = 0.05) and IGFBP-2 was increased (P = 0.01). Resveratrol induced different patterns of gene expression changes in each xenograft model, with upregulation of oncogenic pathways E2F3 and beta-catenin in LAPC-4 tumors.
CONCLUSION: Resveratrol was associated with significantly worse survival with LAPC-4 tumors, but unchanged survival with LNCaP. Based on these preliminary data that resveratrol may be harmful, caution should be advised in using resveratrol for patients until further studies can be conducted.