Development of a machine learning-based predictive model for prediction of success or failure of medical management for benign prostatic hyperplasia.

OBJECTIVE: To develop a novel predictive model for identifying patients who will and will not respond to the medical management of benign prostatic hyperplasia (BPH). METHODS: Using data from the Medical Therapy of Prostatic Symptoms (MTOPS) study, several models were constructed using an initial data set of 2172 patients with BPH who were treated with doxazosin (Group 1), finasteride (Group 2), and combination therapy (Group 3). K-fold stratified cross-validation was performed on each group, Within each group, feature selection and dimensionality reduction using nonnegative matrix factorization (NMF) were performed based on the training data, before several machine learning algorithms were tested; the most accurate models, boosted support vector machines (SVMs), being selected for further refinement. The area under the receiver operating curve (AUC) was calculated and used to determine the optimal operating points. Patients were classified as treatment failures or responders, based on whether they fell below or above the AUC threshold for each group and for the whole data set. RESULTS: For the entire cohort, the AUC for the boosted SVM model was 0.698. For patients in Group 1, the AUC was 0.729, for Group 2, the AUC was 0.719, and for Group 3, the AUC was 0.698. CONCLUSION: Using MTOPS data, we were able to develop a prediction model with an acceptable rate of discrimination of medical management success for BPH.

Does prior surgical interventional therapy for BPH affect the oncological or functional outcomes after primary whole-gland prostate cryoablation for localized prostate cancer?

BACKGROUND: To assess whether prior interventional treatment for benign prostatic hyperplasia (BPH) influences oncologic or functional outcomes following primary whole-gland prostate cryoablation. METHODS: Among 3831 men with prostate cancer who underwent primary whole-gland prostate cryoablation, we identified 160 with a history of prior BPH interventional therapy including transurethral needle ablation (n = 6), transurethral microwave thermotherapy (n = 9), or transurethral resection of the prostate (n = 145). Patients with a history of medically treated or unspecified BPH therapy were excluded from the study. Oncological and functional outcomes were compared between men with and without prior BPH interventional therapy. RESULTS: In unadjusted analyses, prior interventional BPH therapy was associated with higher risks of postoperative urinary retention (17.5% vs. 9.6%, p = 0.001) and new-onset urinary incontinence (39.9% vs. 19.4%, p > 0.001) compared with no prior therapy. Interventional BPH therapy was not correlated with the risk of developing a rectourethral fistula (p = 0.84) or new-onset erectile dysfunction (ED) at 12 months (p = 0.08) following surgery. On multivariable regression, prior interventional BPH therapy was associated with increased risk of urinary retention (OR 1.9, 95%, p = 0.015) and new-onset urinary incontinence (OR 2.13, p < 0.001). The estimated 5 years Kaplan-Meier survival analysis showed no statistically significant difference (p = 0.3) in biochemical progression free survival between those who underwent interventional BPH therapy compared with those who did not. Local disease recurrence assessed by post cryoablation positive for-cause prostate biopsy showed no significant difference between the two groups (25.4% vs. 28.7%, p = 0.59). CONCLUSIONS: Prior interventional BPH therapy did not affect the oncologic outcomes nor did it increase the risk of rectourethral fistula or ED in sexually performing patients prior to cryosurgery. Prior interventional BPH therapy was associated with increased risk of urinary retention and incontinence after primary whole-gland prostate cryoablation for prostate cancer.