ORIGINAL ARTICLE Metabolic syndrome, dyslipidemia and prostate cancer recurrence after primary surgery or radiation in a veterans cohort LC Macleod 1 , LJ Chery 1 , EYC Hu 2 , SB Zeliadt 2 , SK Holt 1 , DW Lin 1,2,3 , MP Porter 1,2 , JL Gore 1,3 and JL Wright 1,2,3 BACKGROUND: Metabolic syndrome (MetS) has been hypothesized to be associated with cancer, including prostate cancer (PCa), but the relationship is not well characterized. We analyze the relationship between MetS features and localized PCa recurrence after treatment. METHODS: Men having primary treatment for localized PCa were included from a multi-site regional veteran network. Recurrence was defined as nadir PSA +2 ng ml - 1 (radiation) or PSA ⩾0.2 ng ml - 1 (prostatectomy). MetS was based on consensus professional society guidelines from the American Heart Association and International Diabetes Federation (three of: hypertension 4 130/85 mm Hg, fasting blood glucose ⩾100 mg dl - 1 , waist circumference 4 102 cm, high-density lipoprotein o40 mg dl - 1 , triglycerides ⩾150 mg dl - 1 ). Closely related abnormality in low-density lipoprotein (LDL; 4 130 mg dl - 1 ) was also examined. Analysis of PCa recurrence risk included multivariable Cox proportional hazards regression with propensity adjustment. RESULTS: Of the 1706 eligible men, 279 experienced recurrence over a median follow-up period of 41 months (range 1–120 months). Adjustment variables associated with PCa recurrence included: index PSA, Gleason, and tumor stage. Independent variables of interest associated with PCa recurrence were hyperglycemia and elevated LDL. Elevated LDL was associated with PCa recurrence (multivariable hazard ratio (HR) 1.34, 95% confidence interval (CI) 1.03, 1.74; propensity adjusted HR 1.33, 95% CI 1.03, 1.72). There was also an association between impaired fasting glucose and PCa recurrence in (multivariable HR 1.54, 95% CI 1.10, 2.15; propensity adjusted HR 1.41, 95% CI 1.01, 1.95). MetS was not associated with PCa recurrence (multivariable: HR 0.96, 95% CI 0.61, 1.50; propensity adjusted HR 1.04, 95% CI 0.67, 1.62). CONCLUSIONS: PCa recurrence is not associated with MetS but is associated with elevated LDL and impaired fasting glucose. If confirmed, these data may help provide modifiable targets in preventing recurrence of PCa. Prostate Cancer and Prostatic Disease (2015) 18, 190–195; doi:10.1038/pcan.2015.12; published online 31 March 2015 INTRODUCTION Prostate cancer (PCa) is the most common cancer among men accounting for 233 000 new cases per year and 29 480 deaths in 2014. 1 Although nearly one-sixth of men will be diagnosed in their lifetime, o3% of those diagnosed with PCa will die of the disease. 1 Thus a large number of cases are over-detected and over-treated, representing a major opportunity to reduce morbid- ity and expense. Risk calculators and nomograms have achieved some success in patient counselling and selecting patients for definitive treatment as opposed to active surveillance. 2 Perhaps more important is defining low- and high-risk localized PCa. Higher-risk PCa will behave aggressively through recurrence, metastasis and hormone resistance. Adding to the controversy of who (or if) to treat PCa, data from the PIVOT randomized study of radical prostatectomy (RP) versus watchful waiting showed improved disease-free survival for men with intermediate- and high-risk disease undergoing RP, but no difference for low-risk PCa. 3 Despite this finding, death from PCa was rarely observed in the trial, resulting in a call for additional biomarkers and risk factors for PCa recurrence and progression. Observational cohorts report that among those treated 15–31% will experience bio- chemical recurrence within 15 years. 4,5 Of those with biochemical recurrence, up to 18% will have symptomatic clinical progression and up to 8% will die of PCa. 6 Some risk factors for aggressive disease have been established. For example, mortality is higher among African Americans and men with family history of fatal PCa. 7 However, obesity, 8–10 diet/vitamins/ supplements 11–13 and features of metabolic syndrome (MetS) 14–18 remain active areas of research. MetS is increasingly prevalent in countries with a ‘western diet’ making it a relevant target for investigation. 19 For this study, we based our definition on a joint consensus statement by the American Heart Association, the International Diabetes Federation and four related professional societies: hypertension 4 130/85 mm Hg, fasting blood glucose ⩾100 mg dl - 1 , waist circumference 4 102 cm, high-density lipopro- tein (HDL) o40 mg dl - 1 , and triglycerides ⩾150 mg dl - 1 . 20,21 The lengthy natural history of PCa makes it challenging to study PCa-specific survival (which is the most clinically meaningful outcome), but biochemical recurrence is indicative of mortality risk and has a higher event rate. We hypothesized that MetS, its components or closely linked measures of dyslipidemia may be associated with PCa recurrence. We therefore analyzed a cohort of veterans to determine which MetS or related clinical components were predictive of PCa recurrence. 1 Department of Urology, University of Washington School of Medicine, Seattle, WA, USA; 2 Veterans Affairs Puget Sound Health Care System, Seattle, WA, USA and 3 Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA, USA. Correspondence: Dr LC Macleod, Department of Urology, University of Washington Medical Center, 1959 NE Pacific Street, Box 656510, Seattle, WA 98195, USA. E-mail: liamcm@uw.edu Received 17 November 2014; revised 20 January 2015; accepted 18 February 2015; published online 31 March 2015 Prostate Cancer and Prostatic Disease (2015) 18, 190 – 195 © 2015 Macmillan Publishers Limited All rights reserved 1365-7852/15 www.nature.com/pcan