Machine learning approaches for predicting progression in hormone-sensitive prostate cancer patients
Machine-learning-Ansätze zur Vorhersage der Progression bei hormonsensitivem Prostatakarzinom
Frontiers in Oncology · 12.02.2026 · 16
DOI: 10.3389/fonc.2026.1704671 ↗Auf einen Blick
Bei 410 HSPC-Patienten sagte ein Ensemblemodell, besonders Random Forest, die Progression zu CRPC am besten voraus. Die Modelle zeigten gute Kalibrierung und keine deutliche Überanpassung.
Fragestellung
Kann ein Machine-Learning-Modell die Progression von HSPC zu CRPC bei Patient:innen vorhersagen?
Methode
Daten von 410 HSPC-Patienten aus einem Zentrum wurden analysiert. Getestet wurden DT, RF, XGBoost, ANN und SVM; Selektion per genetischem Algorithmus, Training 80 %, Test 20 %.
Ergebnisse
RF erzielte im Training 0,838 (95 %-KI 0,8324–0,902) und im Test 0,817 (95 %-KI 0,659–0,829); AUC 0,873 (95 %-KI 0,730–0,878). XGBoost: Training 0,814 (95 %-KI 0,790–0,878), Test 0,805 (95 %-KI 0,707–0,829); AUC 0,866 (95 %-KI 0,780–0,871).
Schlussfolgerung der Autor:innen
Die Autor:innen berichten, dass Ensemble-Methoden, besonders RF, die Progression von HSPC am besten vorhersagten. Die Studie sei ein erster Schritt zu einem prädiktiven Werkzeug auf Basis klinischer Ausgangsdaten.
Kernaussagen
- RF und XGBoost zeigten die beste Vorhersageleistung
- RF: Test 0,817; AUC 0,873
- Kalibrierung gut, keine deutliche Überanpassung
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Oncology, online erschienen am 12.02.2026. Verfasst von Bingyu Zhu und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Urology, The Affiliated Chengdu 363 Hospital of Southwest Medical University, Chengdu, Sichuan, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Künstliche Intelligenz in der Pflege, Onkologische Pflege
Abstract (Originalsprache)
Objective
Almost all hormone-sensitive prostate cancer (HSPC) cases eventually progress to castration-resistant prostate cancer (CRPC) following androgen deprivation therapy (ADT). This study aims to develop a machine learning (ML) model to predict the progression of HSPC patients. Additionally, we conducted statistical analyses on the dataset to identify significant features and clinical markers predictive of HSPC transitioning to CRPC.
Methods
Data from 410 HSPC patients treated at Yunnan Cancer Hospital between 01/01/2017, and 31/05/2022, were analyzed. Predictive analyses were performed on a series of features observed during the patients’ initial visits. The primary ML methods employed were decision tree (DT), random forest (RF), XGBoost, artificial neural network (ANN), and support vector machine (SVM). Feature selection was conducted using a genetic algorithm (GA). The ML models were trained with an 80% training set and validated with a 20% test set. Model performance was evaluated using the area under the ROC curve (AUC), calibration plots, and learning curves to assess fit and calibration. Evaluation metrics included accuracy (ACC), precision (PRE), specificity (SPE), sensitivity (SEN), and F1 score.
Results
Visualization of evaluation metrics was presented through confusion matrices and ROC curves. Ensemble learning methods, particularly RF and XGBoost, demonstrated the best model performance. RF achieved a score of 0.838 (95% CI:0.8324-0.902)on the training dataset and 0.817 (95% CI: 0.659 - 0.829) on the test dataset (AUC: 0.873, 95% CI:0.730-0.878). XGBoost achieved a score of 0.814 (95% CI:0.790-0.878) on the training dataset and 0.805 (95% CI:0.707-0.829) on the test dataset (AUC: 0.866, 95% CI:0.780-0.871). Calibration curves indicated good model calibration, and learning curves suggested no significant overfitting in both the training and test sets.
Conclusion
Our findings demonstrate that ensemble learning methods, particularly RF, exhibit superior performance in predicting HSPC progression. This study represents a preliminary step toward a predictive tool, highlighting the potential of baseline clinical data for risk stratification. Future prospective studies with larger, multi-center cohorts are warranted to validate and refine this approach for possible clinical integration.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Urology, The Affiliated Chengdu 363 Hospital of Southwest Medical University, Chengdu, Sichuan, China.
- Department of Urology I, The Third Affiliated Hospital of Kunming Medical University(Peking University Cancer Hospital Yunnan, Yunnan Cancer Hospital, Cancer Center of Yunnan Province), Kunming, Yunnan, China.
- Department of Urology II, The Third Affiliated Hospital of Kunming Medical University (Peking University Cancer Hospital Yunnan, Yunnan Cancer Hospital, Cancer Center of Yunnan Province), Kunming, Yunnan, China.
Zitieren
APA
Zhu, B., Jiang, H., Zhang, C., Wang, Q., Yang, L., Yang, H., Li, R., Li, J., Pang, X., Zheng, Y., Yan, L., Wang, Y., & Bai, Y. (2026). Machine learning approaches for predicting progression in hormone-sensitive prostate cancer patients. Frontiers in Oncology, 16. https://doi.org/10.3389/fonc.2026.1704671
Vancouver
Zhu B, Jiang H, Zhang C, Wang Q, Yang L, Yang H, et al. Machine learning approaches for predicting progression in hormone-sensitive prostate cancer patients. Frontiers in Oncology. 2026;16. doi:10.3389/fonc.2026.1704671
BibTeX
@article{Zhu2026,
author = {Zhu, Bingyu and Jiang, Haiyang and Zhang, Chongjian and Wang, Qilin and Yang, Libo and Yang, Hong and Li, Ruiqian and Li, Jun and Pang, Xusong and Zheng, Yufeng and Yan, Lingtao and Wang, Yu and Bai, Yu},
title = {Machine learning approaches for predicting progression in hormone-sensitive prostate cancer patients},
journal = {Frontiers in Oncology},
year = {2026},
volume = {16},
doi = {10.3389/fonc.2026.1704671}
}
EUVIMED