EUVIMEDEUVIMED entdecken
Multizentrische Studie

MRI ‐Based Habitat Analysis for Predicting Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer: A Multicenter Study

Huijuan Zhu, Baiyun Zhang iD, Danni Peng, Siyi Song, Lixia Huang, Ye Zhang, Xiu Wu

Journal of Magnetic Resonance Imaging · 2026

DOI: 10.1002/jmri.70285 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Journal of Magnetic Resonance Imaging, 2026. Verfasst von Huijuan Zhu und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ultrasonography, Hunan Provincial Maternal and Child Health Care Hospital, Changsha, Hunan, China. Hauptthemen laut MEDLINE: Breast Neoplasms, Neoadjuvant Therapy, Magnetic Resonance Imaging. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 34 Literaturangaben. Gefördert von: General Project of the Hunan Provincial Health Commission, Joint Departmental Fund Project of the Natura Science Foundation of Hunan Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Habitat imaging has been widely used to assess tumor treatment response; however, the role of MRI‐based habitat analysis in identifying pathological complete response (pCR) after neoadjuvant chemotherapy (NAC) in breast cancer remains an unresolved issue.

Objectives

To evaluate the utility of dynamic contrast‐enhanced MRI (DCE‐MRI)‐based habitat imaging in identifying pCR after NAC in breast cancer patients. Study Type Retrospective. Field Strength/Sequence 1.5 T or 3.0 T, DCE‐MRI (Gradient echo). Subjects Three hundred and sixty‐three women with biopsy‐confirmed breast cancer from Center A ( n = 249, training set) and Center B and Center C ( n = 114, external validation set). Assessment DCE‐MRI peak‐enhancement images were used to generate habitat maps via supervoxel segmentation and K‐means clustering. Two intratumoral heterogeneity (ITH) metrics (Volume Entropy and Intensity Entropy) were extracted to quantify the structural and signal complexity of tumors. Three discriminative models were developed: a clinical model based on clinicopathologic variables, an ITH model incorporating Volume Entropy and Intensity Entropy, and an integrated nomogram combining both feature sets. Statistical Tests Student's t test, Wilcoxon U test, χ 2 , Fisher exact test, and receiver operating characteristic curve analysis. Significance was set at p < 0.05.

Results

Volume Entropy and Intensity Entropy were significantly lower in pCR versus non‐pCR groups. HR status, HER2 status, and both ITH features were independent indicators of pCR. The nomogram showed superior performance (AUC = 0.849 in the training set and 0.825 in the validation set), outperforming the clinical model ( DeLong test). Subgroup analysis across four molecular subtypes showed AUCs ranging from 0.762 to 0.890. An interactive online tool was developed for clinical application. Data Conclusion MRI‐based habitat analysis offers a simple, interpretable, and clinically applicable approach for noninvasive identification of pCR to NAC in breast cancer. Level of Evidence 3. Technical Efficacy Stage 3.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Breast NeoplasmsNeoadjuvant TherapyMagnetic Resonance ImagingHumansFemalePathologic Complete ResponseDynamic Contrast Enhanced Magnetic Resonance ImagingMiddle AgedRetrospective StudiesAdultContrast MediaAgedBreastTreatment OutcomeRadiomicsBiopsy

Stichwörter der Autor:innen

breast neoplasmsdynamic contrast enhanced magnetic resonance imaginghabitat imagingpathologic complete responsetreatment effect heterogeneity

Substanzen

Contrast Media

Autor:innen und Institutionen

  • Department of Ultrasonography, Hunan Provincial Maternal and Child Health Care Hospital, Changsha, Hunan, China.
  • NHC Key Laboratory of Birth Defect for Research and Prevention, Hunan Provincial Maternal and Child Health Care Hospital, Changsha, Hunan, China.

ORCID-Profile: Baiyun Zhang

Förderung

  • General Project of the Hunan Provincial Health Commission
  • Joint Departmental Fund Project of the Natura Science Foundation of Hunan Province

Zitieren

APA

Zhu, H., Zhang, B., Peng, D., Song, S., Huang, L., Zhang, Y., & Wu, X. (2026). MRI ‐Based Habitat Analysis for Predicting Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer: A Multicenter Study. Journal of Magnetic Resonance Imaging. https://doi.org/10.1002/jmri.70285

Vancouver

Zhu H, Zhang B, Peng D, Song S, Huang L, Zhang Y, et al. MRI ‐Based Habitat Analysis for Predicting Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer: A Multicenter Study. Journal of Magnetic Resonance Imaging. 2026. doi:10.1002/jmri.70285

BibTeX
@article{Zhu2026,
  author = {Zhu, Huijuan and Zhang, Baiyun and Peng, Danni and Song, Siyi and Huang, Lixia and Zhang, Ye and Wu, Xiu},
  title = {MRI ‐Based Habitat Analysis for Predicting Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer: A Multicenter Study},
  journal = {Journal of Magnetic Resonance Imaging},
  year = {2026},
  doi = {10.1002/jmri.70285}
}
Kostenlos starten →