382 Background: Nearly one third of men treated with curative intent for localized prostate cancer (PC) will develop a rising PSA. The rate of PSA rise (PSA doubling time aka PSASDT) is a predictor metastases and PC death. In laboratory mice, an extreme low carbohydrate diet slows PC growth. We tested whether this diet could slow PSADT in men with recurrent PC. Methods: We are conducting a 6-month multi-center randomized phase II trial of dietary carbohydrate restriction vs. no diet intervention control. Men had to have a BMI ≥24 kg/m2, received radical prostatectomy or definitive local radiation for PC, had a PSA 0.4-20.0 ng/ml (3-20 if prior radiation therapy) within the past 3 months, and current PSADT 3-36 months. The intervention arm was instructed to eat < 20 grams/carbs/day with no other limits. The control arm was told to make no diet. In this interim analysis, we present the efficacy of the dietary intervention with regards to weight loss. Arms were compared using rank-sum. Total anticipated enrollment is 60. The primary outcome is differences in PSADT between arms. Secondary outcomes include weight loss, and dietary make-up. Results: To date, 28 patients (14 in each study arm) have completed the study. Characteristics were well-balanced at baseline. At the 6-month dietary assessment, calorie consumption was similar between the two study arms (p = 0.090) among the 16 patients (7 low-carb, 9 control) with diet information. Subjects in the low-carb arm ate fewer carbs (29 vs. 188 g, p = 0.008) and more protein (125 vs. 73 g, p = 0.044) but similar amounts of fat (75 vs. 67 g, p = 0.672) vs. subjects in the control arm. Six months on the low carb diet resulted in greater weight loss (median: 31.7 vs. 0.8 lbs, p < 0.001), lower BMI (24.4 vs. 29.6 kg/m2, p < 0.001), and smaller waist circumference (95.7 vs. 108.9 cm, p = 0.002). Conclusions: In this interim analysis of an on-going dietary study for men with a rising PSA after definitive local treatment, an extreme low carbohydrate diet results in dramatic weight loss in 6 months. Whether this weight loss slows PC growth is an on-going question. Clinical trial information: NCT01763944.