Introduction

Although there is increasing interest in focal therapy for prostate cancer as an alternative treatment option to whole-gland therapy, it remains difficult to select eligible candidates. We sought to identify preoperative factors that can predict extensive disease after radical prostatectomy (RP) in potential candidates for hemi-ablative focal therapy.

Methods

We identified 770 patients who were diagnosed with unilateral prostate cancer by biopsy, met the focal therapy consensus meeting criteria (PSA<15 ng/mL, clinical stage T1c-T2a, Gleason score 3+3 or 3+4), and had undergone RP at our institution between 2000 and 2014. Among these 770 patients, 98 had tumor maps from whole-mount slides and magnetic resonance imaging (MRI) data. Extensive disease was defined as having Gleason pattern 4 or 5 in bilateral lobes, extracapsular extension, seminal vesicle involvement, or lymph node involvement on the tumor map and RP pathology. Both lobes of the prostate were scored using the standardized 5-point Prostate Imaging Reporting and Data System version 2 on MRI. Extracapsular extension and seminal vesicle involvement were also scored using a 5-point scale. Preoperative characteristics including biopsy and MRI data were evaluated. Selected predictors of extensive disease were analyzed by multivariable logistic regression. Decision curve analysis was performed to evaluate the clinical utility of our model.

Results

Among our cohort, 39 patients (40%; 95% CI 30%, 50%) had extensive disease. The Kattan nomogram score (p=0.020) and a MRI extracapsular extension score ≥3 (p=0.002) were independent predictors of extensive disease. Our model exhibited a discrimination of 0.698 after 10-fold cross-validation. However, decision curve analysis demonstrated that our multivariable model does not add value for threshold probabilities of having extensive disease below 20%, with no increase in net benefits compared to a treat-all strategy._x000D_ _x000D_

Conclusions

Although the variables in our model are significantly associated with extensive disease, it was not able to place patients into a low enough risk of extensive disease where hemi-ablative therapy would be indicated.

Funding

none