Objectives: Several studies have shown that obesity is associated with more aggressive prostate cancer (PCa) variants. We hypothesized that obesity, quantified as body mass index (BMI), is associated with a higher risk of lymph node invasion (LNI) in patients undergoing extended pelvic lymph node dissection (ePLND).
Methods: Clinical and pathological data were available for 994 consecutive men with PCa treated with radical prostatectomy (RP) and ePLND at a single European tertiary academic centre. Univariable and multivariable logistic regression analyses addressed the rate of LNI. Covariates consisted of pre-treatment prostate specific antigen (PSA), biopsy Gleason sum, clinical stage history of diabetes mellitus as well as BMI coded as either continuous or categorized (<25, 25.0-29.9, 30 kg/m(2) or more) variable. Predictive accuracy was assessed with area under curve estimates.
Results: Overall LNI was diagnosed in 105 patients (10.6%). Mean number of removed lymph nodes was 18.3 (range 7-60). Of all 994 patients, 372 (37.4%) were normal weight, 518 (52.1%) overweight, and 104 (10.5%) were clinically obese. Prevalence of LNI did not significantly differ across different BMI categories (<25, 25.0-29.9 and 30 kg/m(2) or more; 9.9, 10.6 and 12.5%, respectively; P = 0.75). In logistic regression models, neither continuously coded nor categorized BMI was a significant predictor of LNI at univariable or multivariable analyses (all P-values >or=0.1). Moreover, inclusion of BMI with PSA, clinical stage, biopsy Gleason sum and presence of DM did not increase the ability of these variables to predict LNI (82.2% without BMI vs 82.5% and 82.9% with BMI coded as continuous and categorized variable, respectively; all P >or= 0.4).
Conclusions: In men undergoing RP and ePLND, increased BMI was not associated with increased risk of lymph node metastases. Therefore, routinely considering patient BMI in risk stratification schemes or prognostic LNI models may not be warranted.