Sex hormone-binding globulin is a significant predictor of extracapsular extension in men undergoing radical prostatectomy

Abstract

Study Type--Therapy (case series) Level of Evidence 4. What's known on the subject? and What does the study add? Previous reports showed controversial evidence supporting the role of sex steroids, mainly testosterone, in the etiology and pathogenesis of prostate cancer (PCa). The bioavailability of sex steroids is significantly regulated by sex hormone-binding globulin (SHBG). In this context, SHBG levels have been shown to be significantly higher in PCa patients than in controls. Likewise, SHBG was reported to serve as an independent predictor for extra-prostatic extension of tumour [defined as cancer (≥pT3) with capsular penetration, seminal vesicle involvement, or lymph node invasion (LNI)] in patients with clinically localized PCa. The presence of non-organ-confined disease is significantly associated with higher biochemical recurrence rates. This study provides novel evidence that SHBG might serve as a significant multivariate predictor of extra capsular extension (ECE) in PCa patients submitted to radical prostatectomy, after accounting for preoperative clinically available variables such as patient's age, total PSA, clinical stage, biopsy Gleason sum, and BMI. Moreover, a clinical cut-off for circulating SHBG allows using this easily quantifiable molecule as a novel clinical parameter in PCa patients.

Objective: • To examine the association between sex hormone-binding globulin (SHBG) and extracapsular extension (ECE) in men treated with retropubic radical prostatectomy (RRP).

Patients and methods: • Preoperative serum SHBG levels were measured in a cohort of 629 consecutive European Caucasian men [mean (range) age of 64 (41-78) years] who underwent RRP. • No patient received any hormonal neoadjuvant treatment. SHBG levels were measured the day before RRP (08:00-10:00 hours) in all cases at the same laboratory. • Logistic regression models tested the association between predictors [including age, prostate-specific antigen (PSA) level, clinical stage, biopsy Gleason sum, body mass index (BMI), and SHBG] and ECE. • Combined accuracy of predictors was tested in regression-based models predicting ECE at RRP. SHBG was included in the model both as a continuous and categorized variable (according to the most informative threshold level of 30 nmol/L).

Results: • In all, 92 patients (14.6%) had ECE. The mean (standard deviation; median) serum SHBG levels were significantly higher in men with ECE compared with those with no ECE at 41.1 (14.7; 37.5) vs 36.4 (16.7; 34) nmol/L (P= 0.007; 95% confidence interval -8.00, -1.29). • Univariate analyses indicated that continuously coded SHBG was significantly [odds ratio (OR) 1.01; P= 0.03] associated with ECE, with a predictive accuracy of 60.1%. • At multivariate analyses, both continuous (OR 1.01; P= 0.03) and categorical SHBG (OR 3.22; P < 0.001) were significantly associated with ECE, after accounting for age, PSA level, clinical stage, biopsy Gleason sum, and BMI. • Addition of continuously coded SHBG slightly increased the predictive accuracy of the base model based on clinically established predictors from 63.3% to 65.5% (2.0% gain; P= 0.48). • In contrast, a model based on categorized-SHBG showed bootstrap-corrected predictive accuracy of 68.4% (5.1% gain; P= 0.044).

Conclusion: • This study shows that SHBG might serve as a significant multivariate predictor of ECE in men with prostate cancer that undergo RRP.