RESUMEN
PURPOSE: Machine learning (ML) models presented an excellent performance in the prognosis prediction. However, the black box characteristic of ML models limited the clinical applications. Here, we aimed to establish explainable and visualizable ML models to predict biochemical recurrence (BCR) of prostate cancer (PCa). MATERIALS AND METHODS: A total of 647 PCa patients were retrospectively evaluated. Clinical parameters were identified using LASSO regression. Then, cohort was split into training and validation datasets with a ratio of 0.75:0.25 and BCR-related features were included in Cox regression and five ML algorithm to construct BCR prediction models. The clinical utility of each model was evaluated by concordance index (C-index) values and decision curve analyses (DCA). Besides, Shapley Additive Explanation (SHAP) values were used to explain the features in the models. RESULTS: We identified 11 BCR-related features using LASSO regression, then establishing five ML-based models, including random survival forest (RSF), survival support vector machine (SSVM), survival Tree (sTree), gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), and a Cox regression model, C-index were 0.846 (95%CI 0.796-0.894), 0.774 (95%CI 0.712-0.834), 0.757 (95%CI 0.694-0.818), 0.820 (95%CI 0.765-0.869), 0.793 (95%CI 0.735-0.852), and 0.807 (95%CI 0.753-0.858), respectively. The DCA showed that RSF model had significant advantages over all models. In interpretability of ML models, the SHAP value demonstrated the tangible contribution of each feature in RSF model. CONCLUSIONS: Our score system provide reference for the identification for BCR, and the crafting of a framework for making therapeutic decisions for PCa on a personalized basis.
Asunto(s)
Aprendizaje Automático , Recurrencia Local de Neoplasia , Neoplasias de la Próstata , Humanos , Masculino , Neoplasias de la Próstata/sangre , Neoplasias de la Próstata/patología , Recurrencia Local de Neoplasia/sangre , Recurrencia Local de Neoplasia/patología , Estudios Retrospectivos , Anciano , Persona de Mediana Edad , Pronóstico , Árboles de Decisión , Modelos de Riesgos Proporcionales , Algoritmos , Máquina de Vectores de Soporte , Antígeno Prostático Específico/sangreRESUMEN
PURPOSE: It is well-established that the lack of accurate diagnostic modalities for prostate cancer (PCa) leads to overdiagnosis and overtreatments. Accordingly, this study aimed to assess the value of urine-derived exosomal prostate-specific membrane antigen (PSMA) as a biomarker for the diagnosis of PCa and clinically significant prostate cancer (csPCa). METHODS: A total of 284 urine samples were collected from patients after the digital rectal examination (DRE). Urinary exosomes were extracted using commercial kits, and urine-derived exosomal PSMA was determined via enzyme-linked immunosorbent assay (ELISA). Evaluation of diagnostic accuracy of PSMA was performed via receiver operating characteristic (ROC) analysis, decision curve analysis (DCA), and waterfall plots. RESULTS: We found that urine-derived exosomal PSMA was significantly higher in PCa and csPCa than in benign prostatic hyperplasia (BPH) and BPH + non-aggressive prostate cancer (naPCa) groups (P < 0.001). Furthermore, the urine-derived exosome PSMA yielded area under the ROC curve (AUC) values of 0.876 and 0.826 for detecting PCa and csPCa, respectively, suggesting better performance than traditional clinical biomarkers. Besides, when the cutoff value used corresponded to a sensitivity of 95%, urine-derived exosomal PSMA could avoid unnecessary biopsies in 41.2% of cases and missed only 0.7% of csPCa cases. CONCLUSIONS: Urine-derived exosomal PSMA exhibits a good diagnostic yield for detecting PCa and csPCa. Findings of the present study provide the foothold for future studies on cancer management and research in this patient population.