MP64-07 DEVELOPMENT OF AN INDIVIDUAL POSTOPERATIVE PREDICTION MODEL FOR KIDNEY CANCER RECURRENCE USING MACHINE LEARNING (UroCCR STUDY 120)

Abstract

You have accessJournal of UrologyCME1 Apr 2023MP64-07 DEVELOPMENT OF AN INDIVIDUAL POSTOPERATIVE PREDICTION MODEL FOR KIDNEY CANCER RECURRENCE USING MACHINE LEARNING (UroCCR STUDY 120) Gaëlle Margue, Loïc Ferrer, Guillaume Etchepare, Karim Bensalah, Arnaud Mejean, Morgan Roupret, Nicolas Doumerc, Alexandre Ingels, Romain Boissier, Géraldine Pignot, Bastien Parier, Philippe Paparel, Thibaut Waeckel, Pierre Bigot, and Jean-Christophe Bernhard Gaëlle MargueGaëlle Margue More articles by this author , Loïc FerrerLoïc Ferrer More articles by this author , Guillaume EtchepareGuillaume Etchepare More articles by this author , Karim BensalahKarim Bensalah More articles by this author , Arnaud MejeanArnaud Mejean More articles by this author , Morgan RoupretMorgan Roupret More articles by this author , Nicolas DoumercNicolas Doumerc More articles by this author , Alexandre IngelsAlexandre Ingels More articles by this author , Romain BoissierRomain Boissier More articles by this author , Géraldine PignotGéraldine Pignot More articles by this author , Bastien ParierBastien Parier More articles by this author , Philippe PaparelPhilippe Paparel More articles by this author , Thibaut WaeckelThibaut Waeckel More articles by this author , Pierre BigotPierre Bigot More articles by this author , and Jean-Christophe BernhardJean-Christophe Bernhard More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000003322.07AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: Kidney cancer is of increasing incidence worldwide. It is most often diagnosed at a localized stage where surgical management is the gold standard. Current prognostic scores offer moderate predictive performance which leads to difficulties in establishing follow-up recommendations for patients after surgery and selecting patients who could benefit from adjuvant therapy. Our objective is to develop a model for individual prediction of the recurrence risk after surgery using machine learning (ML). METHODS: From the french research network on kidney cancer database UroCCR (NCT 03293563), a cohort of patients undergoing surgery between May 2000 and January 2020 for a localized or locally advanced renal cell carcinoma (pT1-T4, N0, M0) was analyzed. Patients with a genetic cancer, a concomitant malignant disease or a chronic inflammatory disease were excluded. For each patient, clinical, biological, histological, and radiological data were collected.Participating sites were randomly assigned to the training or testing cohort with a 67/33 ratio of patients. Missing data were multiply imputed using the MICE algorithm and gradient boosted trees. Several ML algorithms were trained on the training data set and parameters of each algorithm were optimized using repeated cross-validation procedure. C-index at 5 years was used as optimization metric. The predictive performance of the algorithm was then evaluated on the test dataset using C-index and AUC. RESULTS: In total, 3255 patients were split in training and test cohorts of 2172 and 1083 patients, respectively. The enclosed patients had tumors with a mean size of 4cm and 71% of ccRCC; 3.6% had locoregional recurrence, 7% had metastatic progression and 2.4% died. The median follow-up was 25 months. The best results in DFS prediction were obtained using a Cox PH model including 18 variables with a C-index of 0,75 and an AUC of 0,71 at 5 years. Comparatively, the C-index of the UISS and SSIGN scores were of 0.61 and 0.72. Shapley values graphs were generated to display the predicted DFS and the relative contribution of each feature leading to the personalized prediction. CONCLUSIONS: ML applied to data from patients undergoing surgery for localized kidney cancer appears to provide good individual DFS predictions comparing to usual scores. Source of Funding: None © 2023 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 209Issue Supplement 4April 2023Page: e883 Advertisement Copyright & Permissions© 2023 by American Urological Association Education and Research, Inc.MetricsAuthor Information Gaëlle Margue More articles by this author Loïc Ferrer More articles by this author Guillaume Etchepare More articles by this author Karim Bensalah More articles by this author Arnaud Mejean More articles by this author Morgan Roupret More articles by this author Nicolas Doumerc More articles by this author Alexandre Ingels More articles by this author Romain Boissier More articles by this author Géraldine Pignot More articles by this author Bastien Parier More articles by this author Philippe Paparel More articles by this author Thibaut Waeckel More articles by this author Pierre Bigot More articles by this author Jean-Christophe Bernhard More articles by this author Expand All Advertisement PDF downloadLoading ...