BACKGROUND. Recent evidence suggests carbohydrate intake may influence prostate cancer biology. We tested whether a no-carbohydrate ketogenic diet (NCKD) would delay prostate cancer growth relative to Western and low-fat diets in a xenograft model. METHODS. Seventy-five male SCID mice were fed a NCKD (84% fat-0% carbohydrate-16% protein kcal), low-fat (12% fat-72% carbohydrate-16% protein kcal), or Western diet (40% fat-44% carbohydrate-16% protein kcal). Low-fat mice were fed ad libitum and the other arms fed via a modified-paired feeding protocol. After 24 days, all mice were injected with LAPC-4 cells and sacrificed when tumors approached 1,000 mm3. RESULTS. Despite consuming equal calories, NCKD-fed mice lost weight (up to 15% body weight) relative to low-fat and Western diet-fed mice and required additional kcal to equalize body weight. Fifty-one days after injection, NCKD mice tumor volumes were 33% smaller than Western mice (rank-sum, P = 0.009). There were no differences in tumor volume between low-fat and NCKD mice. Dietary treatment was significantly associated with survival (log-rank, P = 0.006), with the longest survival among the NCKD mice, followed by the low-fat mice. Serum IGFBP-3 was highest and IGF-1:IGFBP-3 ratio was lowest among NCKD mice while serum insulin and IGF-1 levels were highest in Western mice. NCKD mice had significantly decreased hepatic fatty infiltration relative to the other arms. CONCLUSIONS. In this xenograft model, despite consuming more calories, NCKD-fed mice had significantly reduced tumor growth and prolonged survival relative to Western mice and was associated with favorable changes in serum insulin and IGF axis hormones relative to low-fat or Western diet. © 2007 Wiley-Liss, Inc.