MRI-based machine learning to differentiate true progression from pseudoprogression in glioblastoma: An updated systematic review and meta-analysis
MRT-basierte Machine Learning zur Unterscheidung von echter Progression und Pseudoprogression bei Glioblastom: aktualisierte systematische Übersichtsarbeit und Metaanalyse
Neuro-Oncology · 19.10.2026
DOI: 10.1093/neuonc/noag259 ↗Auf einen Blick
31 Studien mit 2.344 Patient:innen: MRT-basierte Machine-Learning-Modelle trennten echte Progression von Pseudoprogression mit moderater bis guter externer Validierung.
Fragestellung
Wie gut unterscheiden MRT-basierte Machine-Learning-Modelle bei Glioblastom echte Progression von Pseudoprogression?
Methode
Systematische Übersichtsarbeit und Metaanalyse; Suche in PubMed, Embase, MEDLINE und Cochrane Library bis 1. Oktober 2025; bivariate Random-Effects-Modelle.
Ergebnisse
31 Studien, davon 7 externe Testkohorten mit 219 Patient:innen-Observationen. Sensitivität 0,78 (95 %-KI 0,63–0,88), Spezifität 0,75 (95 %-KI 0,63–0,84), AUC 0,81 (Bootstrap-95 %-KI 0,69–0,91).
Einschränkungen
Häufig hohes oder unklare Verzerrungsrisiko; Transparenz der Berichterstattung heterogen; externe Testkohorten klein.
Schlussfolgerung der Autor:innen
MRT-basierte Machine Learning zeigte eine moderate bis gute externe Validierungsleistung, die Evidenz bleibt jedoch begrenzt; größere transparente multizentrische Validierungen sind vor Routineeinsatz nötig.
Kernaussagen
- Externe Validierung: AUC 0,81
- Sensitivität 0,78; Spezifität 0,75
- Evidenz begrenzt durch Bias und kleine externe Kohorten
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 Neuro-Oncology, online erschienen am 19.10.2026. Verfasst von Xin Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Biomedical Engineering & Imaging Sciences, King's College London, London, United Kingdom (X.Z., K.W., T.C.B.). Die Arbeit nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: UK Medical Research Council, King’s College London–China Scholarship Council.
Abstract (Originalsprache)
Background
Differentiating true progression (TP) from pseudoprogression (PsP) after chemoradiation for glioblastoma remains clinically difficult. We evaluated externally validated diagnostic performance of magnetic resonance imaging (MRI)-based machine learning models and appraised study quality and transparency.
Methods
PubMed, Embase, MEDLINE, and Cochrane Library were searched through October 1, 2025 (PROSPERO registration CRD420251144894). Eligible adult glioblastoma studies used MRI-based machine learning to distinguish TP from PsP against histopathology and/or clinicoradiologic follow-up. The primary meta-analysis included external test cohorts with extractable 2 × 2 data; internal and combined internal and external test analyses were secondary. Diagnostic accuracy was synthesized with bivariate random-effects models.
Results
Thirty-one studies, including 2,344 patients undergoing imaging assessment, were included. Seven external test cohorts, including 219 patient observations, entered the primary meta-analysis. Pooled sensitivity was 0.78 (95% confidence interval [CI], 0.63-0.88), specificity was 0.75 (95% CI, 0.63-0.84), and summary area under the receiver operating characteristic curve (AUC) was 0.81 (bootstrap 95% CI, 0.69-0.91). Internal test cohorts showed higher apparent performance than external cohorts (AUC, 0.91 vs 0.81), but the difference was not statistically significant (difference, −0.09; 95% CI, −0.22 to 0.02; P = .12). Risk of bias was frequently high or unclear, and reporting transparency was heterogeneous.
Conclusions
MRI-based machine learning showed moderate-to-good external validation performance for TP-versus-PsP discrimination, but evidence remains limited by small external test cohorts, bias, and incomplete reporting. Larger transparent multicenter validations are needed before routine clinical implementation.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Biomedical Engineering & Imaging Sciences, King's College London, London, United Kingdom (X.Z., K.W., T.C.B.).
- Department of Neurosurgery, Peking Union Medical College Hospital, Beijing, China (X.G., Y.W.).
- Department of Neuroradiology, King's College Hospital NHS Foundation Trust, London, United Kingdom (T.C.B.).
Förderung
- UK Medical Research Council
- King’s College London–China Scholarship Council
Zitieren
APA
Zhang, X., Guo, X., Wang, Y., Wagstyl, K., & Booth, T. C. (2026). MRI-based machine learning to differentiate true progression from pseudoprogression in glioblastoma: An updated systematic review and meta-analysis. Neuro-Oncology. https://doi.org/10.1093/neuonc/noag259
Vancouver
Zhang X, Guo X, Wang Y, Wagstyl K, Booth TC. MRI-based machine learning to differentiate true progression from pseudoprogression in glioblastoma: An updated systematic review and meta-analysis. Neuro-Oncology. 2026. doi:10.1093/neuonc/noag259
BibTeX
@article{Zhang2026,
author = {Zhang, Xin and Guo, Xiaopeng and Wang, Yu and Wagstyl, Konrad and Booth, Thomas C},
title = {MRI-based machine learning to differentiate true progression from pseudoprogression in glioblastoma: An updated systematic review and meta-analysis},
journal = {Neuro-Oncology},
year = {2026},
doi = {10.1093/neuonc/noag259}
}
EUVIMED