Preoperative neural network using combined magnetic resonance imaging variables, prostate-specific antigen, and gleason score for predicting prostate cancer biochemical recurrence after radical prostatectomy.

OBJECTIVES: To develop and test an artificial neural network (ANN) for predicting biochemical recurrence based on the combined use of pelvic coil magnetic resonance imaging (pMRI), prostate-specific antigen (PSA) measurement, and biopsy Gleason score, after radical prostatectomy and to investigate whether it is more accurate than logistic regression analysis (LRA) in men with clinically localized prostate cancer. METHODS: We evaluated 191 consecutive men who had undergone retropubic radical prostatectomy for clinically localized prostate cancer. None of the men had lymph node metastasis as determined by adequate follow-up and pathologic criteria. The preoperative predictive variables included clinical TNM stage, serum PSA level, biopsy Gleason score, and pMRI findings. The predicted result was biochemical failure (PSA level of 0.1 ng/mL or greater). The patient data were randomly split into four cross-validation sets and used to develop and validate the LRA and ANN models. The predictive ability of the ANN was compared with that of LRA, Han tables, and the Kattan nomogram using area under the receiver operating characteristic curve (AUROC) analysis. RESULTS: Of the 191 patients, 57 (30%) developed disease progression at a median follow-up of 64 months (mean 61, range 2 to 86). Using all the input variables, the AUROC of the ANN was significantly greater (P <0.05) than the AUROC of LRA, Han tables, or the Kattan nomogram for the prediction of PSA recurrence 5 years after radical prostatectomy (0.897 +/- 0.063 versus 0.785 +/- 0.060, 0.733 +/- 0.061, and 0.737 +/- 0.071, respectively). Removing the pMRI findings from the previous models, the AUROC of the ANN decreased statistically significantly (P <0.05) and was comparable to the AUROC of conventional predictive tools (P >0.05). CONCLUSIONS: Using the pMRI findings, the ANN was superior to LRA, predictive tables, and nomograms to predict biochemical recurrence accurately. Confirmatory studies are warranted.

Preoperative neural network using combined magnetic resonance imaging variables, prostate specific antigen, and Gleason score to predict prostate cancer recurrence after radical prostatectomy.

OBJECTIVE: An artificial neural network analysis (ANNA) was developed to predict the biochemical recurrence more effectively than regression models based on the combined use of pelvic coil magnetic resonance imaging (pMRI), prostate specific antigen (PSA) and biopsy Gleason score in patients with clinically organ-confined prostate cancer after radical prostatectomy (RP). METHODS: Two-hundred-and-ten patients undergoing retropubic RP with pelvic lymphadenectomy were evaluated. Predictive study variables included clinical TNM classification, preoperative serum PSA, biopsy Gleason score, transrectal ultrasound (TRUS) findings, and pMRI findings. The predicted result was a biochemical failure (PSA >or=0.1 ng/ml). Using a five-way cross-validation method, the predicted ability of ANNA for a validation set of 200 randomly selected patients was compared with those of Cox regression analysis and "Kattan nomogram" by area under the receiver operating characteristic curve (AUC) analysis. RESULTS: Seventy-three patients (35%) failed at median follow-up of 61 (mean: 60, range: 2-94) months. Using similar input variables, the AUC of ANNA (0.765, 95% Confidence Interval [CI]: 0.704-0.825) was comparable (p > 0.05) to those for Cox regression (0.738, 95%CI: 0.691-0.819) and Kattan nomogram (0.728, 95%CI: 0.644-0.819). Contrarily, adding the pMRI findings, the ANNA is significantly (p < 0.05) superior to any other predictive model (0.897, 95%CI: 0.841-0.977). The Gleason score represented the most influential predictor (relative weight: 2.4) of PSA recurrence, followed by pMRI (2.2), and PSA (2.0). CONCLUSION: ANNA is superior to regression models to predict accurately biochemical recurrence. The relative importance of pMRI and the utility of ANNA to predict the PSA failure in patients referred for RP must be confirmed in further trials.

Preoperative neural network using combined magnetic resonance imaging variables, prostate specific antigen and Gleason score to predict prostate cancer stage.

PURPOSE: We developed an artificial neural network analysis (ANNA) to predict prostate cancer pathological stage more effectively than logistic regression (LR) based on the combined use of prostate specific antigen (PSA), biopsy Gleason score and pelvic coil magnetic resonance imaging (pMRI) in patients with clinically organ confined disease before radical prostatectomy. MATERIALS AND METHODS: In 201 consecutive patients undergoing radical retropubic prostatectomy with pelvic lymphadenectomy the radiological-pathological correlation was evaluated using pMRI. Predictive variables were clinical TNM classification, preoperative serum PSA, biopsy Gleason score and pMRI findings. The predicted results were organ confined vs nonorgan confined disease and lymphatic vs no lymphatic involvement. The predicted ability of ANNA with several parameters in a set of 160 randomly selected test data was compared with that of LR and the Partin tables by area under the receiver operating characteristic curve analysis. RESULTS: The overall accuracy of ANNA and LR was 88% and 91%, and 77% and 84% for nonorgan confined and lymphatic involvement, respectively. For nonorgan confined disease and lymph node involvement the area under the curve of ANNA (0.895 and 0.899) was significantly larger than that of LR and the Partin tables (0.722 and 0.751, and 0.750 and 0.733, respectively, p <0.05). Gleason score represented the most influential predictor (relative weight 2.05) of nonorgan confined disease, followed by pMRI findings (1.96), PSA (1.73) and clinical stage (0.89). CONCLUSIONS: ANNA is superior to LR for accurately predicting pathological stage. The relative importance of pMRI findings and the usefulness of ANNA for predicting pathological stage in individuals must be confirmed in a prospective trial.

Preoperative neural network using combined magnetic resonance imaging variables, prostate-specific antigen, and Gleason score to predict positive surgical margins.

OBJECTIVES: To assess whether artificial neural network analysis (ANNA) predicts for positive surgical margins (PSMs) more effectively than logistic regression analysis (LRA) according to the combined use of the findings of pelvic coil magnetic resonance imaging (pMRI) and other preoperatively available tumor variables in patients with clinically organ-confined prostate cancer after radical prostatectomy. METHODS: A total of 205 patients with clinically localized prostate cancer, who underwent retropubic radical prostatectomy were evaluated. The predictive variables included clinical TNM stage, prostate-specific antigen (PSA) level, PSA density, biopsy Gleason score, percentage of cancer in biopsy specimens, and pMRI findings. The predicted outcome was PSMs. The patient data were randomly split into four cross-validation sets and used to develop and validate the ANNA and LRA models. For comparison, the area under the receiver operating characteristic curve was used. RESULTS: The overall PSM rate was 22% (n = 45). Using all input parameters, the accuracy of the ANNA and LRA was 84% and 75% for the prediction of PSMs, respectively. The area under the receiver operating characteristic curve of the ANNA (0.872 +/- 0.014) was significantly greater statistically (P <0.001) than that for LRA (0.791 +/- 0.006). The simplified ANNA models that used the pMRI findings in addition to PSA and Gleason score were as accurate as the model that used all the variables (P = 0.89). A high percentage of cancer in the biopsy specimens, pMRI findings, and high PSA density were equally the most influential predictors (relative weight 1.881, 1.964, and 1.493, respectively). CONCLUSIONS: All the ANNA models in this study were superior to LRA in the prediction of PSMs. The ANNA using pMRI findings, PSA level, and Gleason score as input variables performed as well as the ANNA using all the input parameters. Additional studies seem warranted.