MRI radiomics prediction modelling for pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a systematic review and meta-analysis
Abdominal Radiology · 28.04.2025 · 50 (11) · 5103-5123
DOI: 10.1007/s00261-025-04953-5 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Abdominal Radiology, online erschienen am 28.04.2025. Verfasst von Jason Rai und 10 weiteren Autor:innen. Beteiligte Einrichtung: BiCyCLE Research Group, St Mark's the National Bowel Hospital, London, UK. jason.rai@nhs.net. Hauptthemen laut MEDLINE: Neoadjuvant Therapy, Rectal Neoplasms, Magnetic Resonance Imaging, Chemoradiotherapy. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 85 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Purpose
Predicting response to neoadjuvant therapy in locally advanced rectal cancer (LARC) is challenging. Organ preservation strategies can be offered to patients with complete clinical response. We aim to evaluate MRI-derived radiomics models in predicting complete pathological response (pCR).
Methods
Search included MEDLINE, Embase and Cochrane Central Register of Controlled Trials (CENTRAL) and Cochrane Database of Systematic Reviews (CDSR) for studies published before 1st February 2024. The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) and Radiomics Quality Score (RQS) tools were used to assess quality of included study. The research protocol was registered in PROSPERO (CRD42024512865). We calculated pooled area under the receiver operating characteristic curve (AUC) using a random-effects model. To compare AUC between subgroups the Hanley & McNeil test was performed.
Results
Forty-four eligible studies (12,714 patients) were identified for inclusion in the systematic review. We selected thirty-five studies including 10,543 patients for meta-analysis. The pooled AUC for MRI radiomics predicted pCR in LARC was 0.87 (95% CI 0.84–0.89). In the subgroup analysis 3 T MRI field intensity had higher pooled AUC 0.9 (95% CI 0.87–0.94) than 1.5 T pooled AUC 0.82 (95% CI 0.80–0.83) p < 0.001. Asian ethnicity had higher pooled AUC 0.9 (95% CI 0.87–0.93) than non-Asian pooled AUC 0.8 (95% CI 0.75–0.84) p < 0.001.
Conclusion
We have demonstrated that 3 T MRI field intensity provides a superior predictive performance. The role of ethnicity on radiomics features needs to be explored in future studies. Further research in the field of MRI radiomics is important as accurate prediction for pCR can lead to organ preservation strategy in LARC. Graphical abstract
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- BiCyCLE Research Group, St Mark's the National Bowel Hospital, London, UK. jason.rai@nhs.net.
- Department of Surgery and Cancer, Imperial College London, London, UK. jason.rai@nhs.net.
- BiCyCLE Research Group, St Mark's the National Bowel Hospital, London, UK.
- Department of Surgery and Cancer, Imperial College London, London, UK.
- Department of Radiology, St Mark's the National Bowel Hospital, London, UK.
- Department of Radiology, The Royal Marsden NHS Foundation Trust, London, UK.
Zitieren
APA
Rai, J., Mai, D. V. C., Drami, I., Pring, E. T., Gould, L. E., Lung, P. F. C., Glover, T., Shur, J. D., Whitcher, B., Athanasiou, T., & Jenkins, J. T. (2025). MRI radiomics prediction modelling for pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a systematic review and meta-analysis. Abdominal Radiology, 50(11), 5103-5123. https://doi.org/10.1007/s00261-025-04953-5
Vancouver
Rai J, Mai DVC, Drami I, Pring ET, Gould LE, Lung PFC, et al. MRI radiomics prediction modelling for pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a systematic review and meta-analysis. Abdominal Radiology. 2025;50(11):5103-5123. doi:10.1007/s00261-025-04953-5
BibTeX
@article{Rai2025,
author = {Rai, Jason and Mai, Dinh V C and Drami, Ioanna and Pring, Edward T and Gould, Laura E and Lung, Phillip F C and Glover, Thomas and Shur, Joshua D and Whitcher, Brandon and Athanasiou, Thanos and Jenkins, John T},
title = {MRI radiomics prediction modelling for pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a systematic review and meta-analysis},
journal = {Abdominal Radiology},
year = {2025},
volume = {50},
number = {11},
pages = {5103--5123},
doi = {10.1007/s00261-025-04953-5}
}
EUVIMED