An elevated body mass (BMI) index predicts for better clinical outcomes in men with metastatic hormone refractory prostate cancer (HRPC).

4556 Background: Previous articles have reported that an elevated BMI was associated with an increased risk of biochemical failure in hormone sensitive patients. We asked the question as to whether an elevated BMI predicts for worst clinical outcomes, namely overall survival (OS) and prostate-cancer survival (PCS), among 1,216 men with HRPC. METHODS: Patients were enrolled on eight clinical trials conducted by the Cancer and Leukemia Group B (CALGB). Eligible patients had progressive prostate cancer during androgen deprivation therapy (with documented castrate levels of testosterone), an ECOG performance status of 0-2, adequate hematologic, renal and hepatic function. We used the NIH definition to classify patients as: normal (<25 kg/m(2)), overweight (25-29 kg/m(2) ), mildly obese (30-34 kg/m(2)), and moderately to severely obese (≥35 kg/m(2)). PCS was defined as the time from study entry to the time of death due to prostate cancer. The proportional hazards model was used to explore the prognostic significance of BMI in predicting OS and PCS. RESULTS: The median BMI was 27.7 kg/m(2) (inter-quartile range = 25.2-31.0 kg/m(2) ). Twenty three percent (285/1216) of the patients had normal BMI, 46% (555/1216) were overweight, 23% (280/1216) were mildly obese, and 8% (96/1216) were moderately to severely obese. In multivariate analysis, adjusting for age, race, performance status, hemoglobin, PSA, LDH, alkaline phosphatase, testosterone, years since diagnosis, presence of visceral disease and Gleason scores, BMI was a statistically significant predictor of OS and PCS. Compared to normal men, the hazard ratios (HR) of overweight patients was 0.80 (95% CI = 0.69-0.93, p-value = 0.003), for mildly obese patients was 0.86 (95% CI = 0.72-1.02, p-value = 0.087) and for moderately to severely obese men it was 0.60 (95% CI = 0.47-0.78, p-value < 0.001). In addition, the HRs for PCS for overweight patients was 0.83 (95% CI = 0.70-0.97, p-value = 0.023), was 0.88 (95% CI = 0.72-1.06, p-value = 0.179) for mildly obese and for moderately to severely obese was 0.62 (95% CI = 0.47-0.81, p-value = 0.001) compared to men with normal BMI. CONCLUSIONS: Contrary to what was reported, these findings demonstrate an inverse relationship between BMI and clinical outcomes in men with HRPC. [Table: see text].

Duke Scholars

Author Susan Halabi Biostatistics & Bioinformatics, Division of Biostatistics