EUVIMEDEUVIMED entdecken
Systematische Übersicht

Artificial intelligence for predicting BCG response in non‐muscle‐invasive bladder cancer: a systematic review

Künstliche Intelligenz zur Vorhersage des BCG-Ansprechens bei nicht-muskelinvasivem Blasenkarzinom: systematische Übersicht

Ludovica Cella iD, Roberto Contieri iD, Marco Paciotti iD, Vittorio Fasulo iD, Alessandro Uleri iD, Pier Paolo Avolio iD, Laura S Mertens iD, Benjamin Pradere iD, Sisto Perdonà, Alberto Saita, Massimo Lazzeri, Paolo Casale und 3 weitere

BJU International · 2026-11 · 138 (5) · 786-797

DOI: 10.1111/bju.70404 ↗

Auf einen Blick

15 Studien mit >24 900 Patient:innen untersuchten KI- und ML-Modelle zur Vorhersage von BCG-Ansprechen und klinischen Ergebnissen bei NMIBC. Perzeptive Modelle, besonders digitale Pathologie, zeigten das stärkste Signal.

Fragestellung

Welche KI- und ML-Modelle sagen das Ansprechen auf intravesikales BCG und klinische Outcomes bei NMIBC voraus?

Methode

Systematische Übersicht; Suche in PubMed, EMBASE und Web of Science bis April 2026; Berücksichtigung von Studien mit quantitativer Performance; Biasbewertung mit PROBAST.

Ergebnisse

Eingeschlossen wurden 15 Studien (>24 900 Patient:innen): 7 perzeptive, 8 integrative. CHAI war über 12 internationale Zentren validiert und stratifizierte Hochgradrezidiv, Progression und BCG-unresponsives Erkrankungsgeschehen. PROGRxN-BCa und DeepSurv übertrafen Standardrechner mit kleinen Zugewinnen.

Einschränkungen

Nur 40 % der Studien hatten externe Validierung, keine prospektiv; 5 Studien waren mit hohem Verzerrungsrisiko behaftet.

Schlussfolgerung der Autor:innen

Perzeptive KI, besonders digitale Pathologie, hat die höchste externe Validierung und liefert das einzige Biomarker-Signal mit Vorhersagewert für BCG versus Alternativen. Prospektive Validierung bleibt vor klinischer Anwendung nötig.

Kernaussagen

  • 15 Studien untersuchten KI/ML-Modelle für BCG-bezogene Outcomes
  • Digitale Pathologie zeigte das stärkste Vorhersagesignal
  • Externe Validierung und prospektive Prüfung waren selten

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Systematische Übersicht in der Zeitschrift BJU International, online erschienen am 17.08.2026. Verfasst von Ludovica Cella und 14 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Sciences, Humanitas University, Milan, Italy. Hauptthemen laut MEDLINE: Adjuvants, Immunologic, Artificial Intelligence, BCG Vaccine, Non-Muscle Invasive Bladder Neoplasms. Die Arbeit nennt 39 Literaturangaben.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

To systematically identify, appraise and synthesise artificial intelligence (AI) and machine‐learning (ML) models that predict treatment response and clinical outcomes after intravesical bacillus Calmette–Guérin (BCG) in non‐muscle‐invasive bladder cancer (NMIBC), a setting in which current risk calculators underperform, and identifying non‐responders has become urgent as alternatives to BCG enter practice.

Methods

PubMed, EMBASE and Web of Science were searched through April 2026 following the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) 2020 guidelines (International Prospective Register of Systematic Reviews [PROSPERO] number CRD420261376808). Studies developing or validating AI/ML models for BCG‐associated outcomes with a quantitative performance metric were included. Risk of bias was assessed with the Prediction model Risk Of Bias ASsessment Tool (PROBAST). Evidence was synthesised along a clinically oriented framework: AI as a perceptual tool (extracting signal from histology or imaging) or integrative tool (re‐weighting clinicopathological, molecular, or urinary variables).

Results

A total of 15 studies (>24 900 patients) were included: seven perceptual, eight integrative. By input data, six used digital pathology, two radiomics, two genomics/transcriptomics, four clinicopathological markers, and one urinary biomarkers. The digital‐pathology Computational Histology Artificial Intelligence (CHAI) platform, validated across 12 international centres, stratified high‐grade recurrence (hazard ratio [HR] 2.08), progression (HR 3.87) and BCG‐unresponsive disease (HR 2.31), and was the only model providing a first signal of predictive value, demonstrating a significant BCG vs gemcitabine/docetaxel interaction ( P = 0.029). Integrative models PROGRxN‐BCa (concordance index [C‐index] 0.79) and DeepSurv (C‐index 0.881) outperformed standard calculators but with modest gains (ΔC‐index 0.05–0.10). Only 40% of studies performed external validation, none prospectively; five were at high risk of bias.

Conclusion

Perceptual AI, particularly digital pathology, has the highest external validation and provides the only biomarker with a first signal of predictive value for BCG vs alternatives, increasingly relevant in the context of the global BCG shortage. Integrative models such as PROGRxN‐BCa and DeepSurv outperform standard calculators with incremental gains and are freely accessible. Prospective validation, systematic calibration reporting and treatment‐by‐biomarker interaction analyses, ideally embedded in trials such as the BRIDGE trial (ClinicalTrials.gov identifier: NCT05538663), remain priorities before clinical adoption.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Adjuvants, ImmunologicArtificial IntelligenceBCG VaccineNon-Muscle Invasive Bladder NeoplasmsUrinary Bladder NeoplasmsHumansAdministration, IntravesicalMachine LearningPredictive Learning ModelsTreatment OutcomeClinical Trials as Topic

Stichwörter der Autor:innen

BCGartificial intelligencemachine learningnon‐muscle‐invasive bladder cancerprediction

Substanzen

Adjuvants, ImmunologicBCG Vaccine

Autor:innen und Institutionen

  • Department of Biomedical Sciences, Humanitas University, Milan, Italy.
  • Department of Urology, IRCCS Humanitas Research Hospital, Milan, Italy.
  • Urology Department, Istituto Nazionale Tumori di Napoli, IRCCS "G. Pascale", Naples, Italy.
  • Department of Urology, Josep Trueta University Hospital, Girona, Spain.
  • Department of Surgical Oncology (Urology), Netherlands Cancer Institute - Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands.
  • Department of Urology, La Croix du Sud Hôpital, Quint Fonsegrives, France.

ORCID-Profile: Ludovica Cella, Roberto Contieri, Marco Paciotti, Vittorio Fasulo, Alessandro Uleri, Pier Paolo Avolio, Laura S Mertens, Benjamin Pradere, Giovanni Lughezzani, Rodolfo Hurle

Zitieren

APA

Cella, L., Contieri, R., Paciotti, M., Fasulo, V., Uleri, A., Avolio, P. P., Mertens, L. S., Pradere, B., Perdonà, S., Saita, A., Lazzeri, M., Casale, P., Lughezzani, G., Hurle, R., & Buffi, N. M. (2026). Artificial intelligence for predicting BCG response in non‐muscle‐invasive bladder cancer: a systematic review. BJU International, 138(5), 786-797. https://doi.org/10.1111/bju.70404

Vancouver

Cella L, Contieri R, Paciotti M, Fasulo V, Uleri A, Avolio PP, et al. Artificial intelligence for predicting BCG response in non‐muscle‐invasive bladder cancer: a systematic review. BJU International. 2026;138(5):786-797. doi:10.1111/bju.70404

BibTeX
@article{Cella2026,
  author = {Cella, Ludovica and Contieri, Roberto and Paciotti, Marco and Fasulo, Vittorio and Uleri, Alessandro and Avolio, Pier Paolo and Mertens, Laura S and Pradere, Benjamin and Perdonà, Sisto and Saita, Alberto and Lazzeri, Massimo and Casale, Paolo and Lughezzani, Giovanni and Hurle, Rodolfo and Buffi, Nicolò Maria},
  title = {Artificial intelligence for predicting BCG response in non‐muscle‐invasive bladder cancer: a systematic review},
  journal = {BJU International},
  year = {2026},
  volume = {138},
  number = {5},
  pages = {786--797},
  doi = {10.1111/bju.70404}
}
Kostenlos starten →