Paul Sargos1, Nicolas Leduc1, Nicolas Giraud2, Giorgio Gandaglia3, Mathieu Roumiguié4, Guillaume Ploussard5, Francois Rozet6, Michel Soulié4, Romain Mathieu7, Pierre Mongiat Artus8, Tamim Niazi2, Vincent Vinh-Hung9, Jean-Baptiste Beauval4. 1. Department of Radiation Oncology, Institut Bergonié, Bordeaux, France. 2. Division of Radiation Oncology, Department of Oncology, McGill University, Montreal, QC, Canada. 3. Division of Oncology, Unit of Urology, Urological Research Institute, IRCCS Ospedale San Raffaele, Milan, Italy. 4. Department of Urology, CHU de Toulouse, Toulouse, France. 5. Department of Urology, Clinique La Croix du Sud, Quint-Fonsegrives, France. 6. Department of Urology, Institut Mutualiste Montsouris, Paris, France. 7. Department of Urology, CHU de Rennes, Rennes, France. 8. Department of Urology, Hôpital Saint Louis, Paris, France. 9. Department of Radiation Oncology, Hôpital Clarac, CHU de la Martinique, Fort-de-France, France.
Abstract
BACKGROUND: Use of predictive models for the prediction of biochemical recurrence (BCR) is gaining attention for prostate cancer (PCa). Specifically, BCR occurs in approximately 20-40% of patients five years after radical prostatectomy (RP) and the ability to predict BCR may help clinicians to make better treatment decisions. We aim to investigate the accuracy of CAPRA score compared to others models in predicting the 3-year BCR of PCa patients. MATERIAL AND METHODS: A total of 5043 men who underwent RP were analyzed retrospectively. The accuracy of CAPRA score, Cox regression analysis, logistic regression, K-nearest neighbor (KNN), random forest (RF) and a densely connected feed-forward neural network (DNN) classifier were compared in terms of 3-year BCR predictive value. The area under the receiver operating characteristic curve was mainly used to assess the performance of the predictive models in predicting the 3 years BCR of PCa patients. Pre-operative data such as PSA level, Gleason grade, and T stage were included in the multivariate analysis. To measure potential improvements to the model performance due to additional data, each model was trained once more with an additional set of post-operative surgical data from definitive pathology. RESULTS: Using the CAPRA score variables, DNN predictive model showed the highest AUC value of 0.7 comparing to the CAPRA score, logistic regression, KNN, RF, and cox regression with 0.63, 0.63, 0.55, 0.64, and 0.64, respectively. After including the post-operative variables to the model, the AUC values based on KNN, RF, and cox regression and DNN were improved to 0.77, 0.74, 0.75, and 0.84, respectively. CONCLUSIONS: Our results showed that the DNN has the potential to predict the 3-year BCR and outperformed the CAPRA score and other predictive models.
BACKGROUND: Use of predictive models for the prediction of biochemical recurrence (BCR) is gaining attention for prostate cancer (PCa). Specifically, BCR occurs in approximately 20-40% of patients five years after radical prostatectomy (RP) and the ability to predict BCR may help clinicians to make better treatment decisions. We aim to investigate the accuracy of CAPRA score compared to others models in predicting the 3-year BCR of PCa patients. MATERIAL AND METHODS: A total of 5043 men who underwent RP were analyzed retrospectively. The accuracy of CAPRA score, Cox regression analysis, logistic regression, K-nearest neighbor (KNN), random forest (RF) and a densely connected feed-forward neural network (DNN) classifier were compared in terms of 3-year BCR predictive value. The area under the receiver operating characteristic curve was mainly used to assess the performance of the predictive models in predicting the 3 years BCR of PCa patients. Pre-operative data such as PSA level, Gleason grade, and T stage were included in the multivariate analysis. To measure potential improvements to the model performance due to additional data, each model was trained once more with an additional set of post-operative surgical data from definitive pathology. RESULTS: Using the CAPRA score variables, DNN predictive model showed the highest AUC value of 0.7 comparing to the CAPRA score, logistic regression, KNN, RF, and cox regression with 0.63, 0.63, 0.55, 0.64, and 0.64, respectively. After including the post-operative variables to the model, the AUC values based on KNN, RF, and cox regression and DNN were improved to 0.77, 0.74, 0.75, and 0.84, respectively. CONCLUSIONS: Our results showed that the DNN has the potential to predict the 3-year BCR and outperformed the CAPRA score and other predictive models.
Authors: Timothy J Wilt; Karen M Jones; Michael J Barry; Gerald L Andriole; Daniel Culkin; Thomas Wheeler; William J Aronson; Michael K Brawer Journal: N Engl J Med Date: 2017-07-13 Impact factor: 91.245
Authors: Claire L Vale; David Fisher; Andrew Kneebone; Christopher Parker; Maria Pearse; Pierre Richaud; Paul Sargos; Matthew R Sydes; Christopher Brawley; Meryem Brihoum; Chris Brown; Sylvie Chabaud; Adrian Cook; Silvia Forcat; Carol Fraser-Browne; Igor Latorzeff; Mahesh K B Parmar; Jayne F Tierney Journal: Lancet Date: 2020-09-28 Impact factor: 79.321
Authors: Steven N Seyedin; John M Watkins; Zachary Mayo; Anthony N Snow; Michael Laszewski; J Kyle Russo; Sarah L Mott; Chad R Tracy; Mark C Smith; John M Buatti; Joseph M Caster Journal: Adv Radiat Oncol Date: 2021-08-14