Prognostic Evidence for Artificial Intelligence-Derived PET Tumour-Burden Biomarkers in Cancer: A Systematic Review
Prognostische Evidenz für aus KI abgeleitete PET-Tumorlast-Biomarker bei Krebs: systematische Übersicht
Diagnostics · 29.10.2026 · 16 (21) · 3564
DOI: 10.3390/diagnostics16213564 ↗Auf einen Blick
Die Übersicht fasste 35 Berichte zu KI-abgeleiteten PET-Tumorlast-Biomarkern zusammen. Höhere MTV- oder TMTV-Werte waren in ausgewählten Krebssettings mit schlechteren Ergebnissen verbunden.
Fragestellung
Welche prognostische Evidenz gibt es für aus KI abgeleitete PET-Tumorlast-Biomarker, insbesondere im Vergleich zu Expertenmessungen?
Methode
Systematische Übersicht; PubMed und Embase am 29. August 2026, Ergänzungssuche bis 9. September; englischsprachige Originalberichte; narrative Synthese nach Krebs, Tracer, Messung, Outcome und Validierung.
Ergebnisse
Eingeschlossen wurden 35 Berichte (13 Core, 22 Contextual) aus 31 Katalogfamilien. Koppelte Hazard Ratios waren wegen Heterogenität nicht poolbar; 37 von 62 Einheiten hatten hohes Bias-Risiko für Confounding, 50 für Statistik/Reporting.
Einschränkungen
Unvollständige Suchabdeckung; keine formale GRADE-Bewertung; keine unabhängige doppelte Extraktion oder Appraisal; Heterogenität verhinderte Pooling.
Schlussfolgerung der Autor:innen
AI-abgeleitete MTV- und TMTV-Werte zeigten prognostische Assoziationen in ausgewählten Krebssettings, mit begrenztem Vertrauen in Unabhängigkeit und Transportabilität. Eine unabhängige Validierung vor klinischer Nutzung ist nötig.
Kernaussagen
- 35 Berichte zu AI-abgeleiteten PET-Tumorlast-Biomarkern eingeschlossen
- Höhere MTV/TMTV-Werte waren mit schlechteren Outcomes assoziiert
- Keine Evidenz für Austauschbarkeit mit Expertenmessungen oder bessere Outcomes
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Übersichtsarbeit in der Zeitschrift Diagnostics, online erschienen am 29.10.2026. Verfasst von Sang Jun Byun und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiation Oncology, Keimyung University Dongsan Hospital, Keimyung University School of Medicine, Daegu 42601, Republic of Korea. Die Arbeit nennt 72 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Daegu Catholic University Medical Center.
Abstract (Originalsprache)
Background/Objectives: Artificial intelligence (AI) can automate tumour-burden measurement on positron emission tomography (PET), including metabolic tumour volume (MTV) and total metabolic tumour volume (TMTV). This review evaluated prognostic evidence for AI-derived PET tumour-burden biomarkers, particularly comparisons with expert measurements.
Methods
PubMed and Embase were searched on 29 August 2026; supplementary searching closed on 9 September. Revision follow-up used an export dated 13 October 2026 while retaining the original temporal eligibility boundary. English-language original reports evaluating AI-derived PET tumour burden and clinical outcomes were eligible. Core comparative and Contextual evidence were synthesized narratively by cancer, tracer, measurement, outcome and validation setting, accounting for shared cohorts. AI-assisted extraction and appraisal underwent human verification, without fully independent duplicate extraction or appraisal. Search coverage was incomplete, and no formal certainty-of-evidence grading was performed. The review was registered in PROSPERO (CRD420261497171).
Results
We included 35 reports (13 Core and 22 Contextual) from 31 catalogue families; these families do not represent counts of independent validation cohorts. Selected studies reported associations between higher AI-derived MTV or TMTV and poorer outcomes, with confidence limited by confounding, threshold selection and validation design. Several reports found prognostic estimates similar to expert measurements without establishing equivalence. Heterogeneity in populations, outcomes, predictor scales and thresholds precluded pooled hazard ratios; incomplete paired uncertainty prevented pooling between-method differences. Of 62 prognostic factor assessment units, 37 had high risk of bias for confounding and 50 for statistical analysis and reporting. External validation of fixed prognostic thresholds remained limited.
Conclusions
AI-derived MTV and TMTV showed prognostic associations in selected cancer settings, with limited confidence in their independence and transportability. Evidence does not establish interchangeability with expert measurements or improved outcomes from biomarker-guided treatment. Independent validation is needed before clinical use.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiation Oncology, Keimyung University Dongsan Hospital, Keimyung University School of Medicine, Daegu 42601, Republic of Korea.
- Division of Hematology-Oncology, Department of Internal Medicine, Daegu Catholic University Medical Center, Daegu 42472, Republic of Korea.
- Department of Nuclear Medicine, Keimyung University Dongsan Hospital, Keimyung University School of Medicine, Daegu 42601, Republic of Korea.
- Department of Nuclear Medicine, Daegu Catholic University Medical Center, Daegu Catholic University School of Medicine, Daegu 42472, Republic of Korea.
ORCID-Profile: Yunhwi Hwang, Byungwook Choi
Förderung
- Daegu Catholic University Medical Center
Zitieren
APA
Byun, S. J., Hwang, Y., Kim, H. W., & Choi, B. (2026). Prognostic Evidence for Artificial Intelligence-Derived PET Tumour-Burden Biomarkers in Cancer: A Systematic Review. Diagnostics, 16(21), 3564. https://doi.org/10.3390/diagnostics16213564
Vancouver
Byun SJ, Hwang Y, Kim HW, Choi B. Prognostic Evidence for Artificial Intelligence-Derived PET Tumour-Burden Biomarkers in Cancer: A Systematic Review. Diagnostics. 2026;16(21):3564. doi:10.3390/diagnostics16213564
BibTeX
@article{Byun2026,
author = {Byun, Sang Jun and Hwang, Yunhwi and Kim, Hae Won and Choi, Byungwook},
title = {Prognostic Evidence for Artificial Intelligence-Derived PET Tumour-Burden Biomarkers in Cancer: A Systematic Review},
journal = {Diagnostics},
year = {2026},
volume = {16},
number = {21},
pages = {3564},
doi = {10.3390/diagnostics16213564}
}
EUVIMED