1. ¹Duke Comprehensive Cancer Center, and the Duke Prostate Center, Duke University, Durham, NC, USA
2. ²Division of Medical Oncology, Department of Medicine, Duke Comprehensive Cancer Center, Duke University, Durham, NC, USA
3. ³Division of Medical Oncology, Department of Medicine, Duke Prostate Center, Duke University, Durham, NC, USA
4. ⁴Division of Urology, Department of Surgery, Duke Comprehensive Cancer Center, Duke University, Durham, NC, USA
5. ⁵Division of Urology, Department of Surgery, Duke Prostate Center, Duke University, Durham, NC, USA
6. ⁶Department of Biostatistics and Bioinformatics, Duke University, Durham, NC, USA
7. ⁷Department of Medical Oncology, Rotterdam Cancer Institute, Erasmus University Medical Center, Rotterdam, The Netherlands
8. ⁸Department of Medical Oncology, Princess Margaret Hospital, University of Toronto, Toronto, Ontario, Canada
9. ⁹Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins Medical Institutions, Baltimore, MD, USA

Correspondence: Dr AJ Armstrong, Duke Comprehensive Cancer Center and Duke Prostate Center, 2424 Erwin Road, Hock Plaza, Suite 606, Room 6127, Durham, NC 27705, USA. E-mail: andrew.armstrong@duke.edu

Received 29 April 2008; Accepted 16 May 2008; Published online 24 June 2008.

Abstract

The purpose of this study was to evaluate the relationship of baseline body mass index (BMI) and serum testosterone level with prostate cancer outcomes in men with castration-resistant metastatic prostate cancer (CRPC). BMI and testosterone levels were evaluated for their ability to predict overall survival (OS) and prostate-specific antigen (PSA) declines in the TAX327 clinical trial, an international phase III randomized trial of one of the two schedules of docetaxel and prednisone compared with mitoxantrone and prednisone. In this study of 1006 men with CRPC, the median serum testosterone level was 14.5 ng per 100 ml (range 0–270), the median BMI was 27 kg m^-2 (range 15.7–46.5), and 26% of men were obese or morbidly obese (BMI30). Obesity was associated with younger age, lower PSA and alkaline phosphatase levels, and higher performance status, primary Gleason sum, testosterone and hemoglobin compared to absence of obesity. In multivariate analysis, neither BMI, presence of obesity, nor baseline testosterone was significantly associated with OS or PSA declines. Higher testosterone levels among obese men suggest incomplete gonadal suppression with current therapies, but these differences may not be clinically relevant in men with CRPC. There was evidence of potential hemodilution of PSA and alkaline phosphatase levels in obese men.