Computational Pathology Reveals Extracellular-Matrix Imaging Biomarkers of Therapy Response in Preclinical Breast Cancer.
Simple Summary: Breast cancer treatment response is influenced not only by cancer cells but also by the surrounding tissue environment. In this preclinical study, we tested whether information contained in immunofluorescence images from a 4T1 mouse breast cancer model could help distinguish therapy...
| Published in: | Cancers Vol. 18; no. 16; pp. 2603 - 2624 |
|---|---|
| Main Authors: | , , , |
| Format: | pictorial research tables/charts Journal Article |
| Published: |
MDPI
Aug2026
|
| Online Access: | View this record in EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=196664893&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 196664893 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 20726694 B74B jtl: Cancers issn: 20726694 maglogo: N pubinfo: dt: Aug2026 vid: 18 iid: 16 pid: 97109 pub: MDPI artinfo: ui: 196664893 196664893 196664893 10.3390/cancers18162603 196664893 ppf: 2603 ppct: 21 formats: tig: atl: Computational Pathology Reveals Extracellular-Matrix Imaging Biomarkers of Therapy Response in Preclinical Breast Cancer. aug: au: Lamprou, Stelios Georgiou, Styliana Stylianopoulos, Triantafyllos Voutouri, Chrysovalantis affil: AnaBioSi-Data Ltd. Athalassas Ave. 176, MyOffice 204, Strovolos 2025, Cyprus sug: subj: Breast Neoplasms Pathology Breast Neoplasms Radiography Breast Neoplasms Therapy Extracellular Matrix Tumor Markers, Biological Image Processing, Computer Assisted Animal Studies Mice Retrospective Design Record Review Machine Learning Fluorescent Antibody Technique Hyaluronic Acid Collagen Image Interpretation, Computer Assisted Anoxia Immunohistochemistry Cell Line, Tumor Sensitivity and Specificity Descriptive Statistics Funding Source ab: Simple Summary: Breast cancer treatment response is influenced not only by cancer cells but also by the surrounding tissue environment. In this preclinical study, we tested whether information contained in immunofluorescence images from a 4T1 mouse breast cancer model could help distinguish therapy responders from non-responders. We analysed 1696 images covering nine tissue-marker panels and compared image-based deep learning with machine learning models trained on quantitative image features. The strongest signal came from extracellular-matrix markers, especially collagen and hyaluronic acid. In animals with matched hypoxia and extracellular-matrix images, collagen and hyaluronic-acid features outperformed hypoxia, and adding hypoxia did not improve prediction. These findings suggest that quantitative tissue architecture is a useful preclinical biomarker source, but the results require validation in independent animal experiments and human tumour tissue before clinical use. Background/Objectives: Histological biomarkers of therapy response remain incompletely defined in preclinical breast cancer models. We evaluated whether quantitative immunofluorescence imaging and multimodal machine learning (ML) could identify image-derived tissue biomarkers associated with response in a 4T1 murine breast cancer therapy-response setting. Methods: We analysed 1696 immunofluorescence images across nine immunofluorescence staining panels: aPDL1, alpha-smooth muscle actin-CD31, alpha-smooth muscle actin-Ki67, CD3-CD31, CD8-Ki67, collagen–hyaluronic acid (HA), granzyme B-CD8, HMGB1, and pimonidazole/hypoxia. A Python image-analysis algorithm extracted 117 RGB-channel, intensity, morphometric, and cross-channel spatial features per image. The modelling framework included image-only deep learning (DL), tabular ML on extracted histological features, and image-plus-tabular gated fusion. Models were evaluated using stratified cross-validation, train-fold-only class balancing, bootstrap confidence intervals, permutation testing, nested feature-selection sensitivity analysis, and grouped leakage controls. Results: Histological staining panels classified treatment-response status across the nine-panel benchmark. Collagen–HA was the strongest staining (AUC 0.954), followed by alpha-smooth muscle actin-Ki67 (AUC 0.851) and hypoxia (AUC 0.837). DL achieved the best performance in eight of nine stainings. In 346 collagen–HA images, combined collagen and HA features achieved AUC 0.848, collagen-only features AUC 0.833, and hyaluronic-acid-only features AUC 0.805. In 104 animals with matched hypoxia and collagen–hyaluronic-acid images, extracellular-matrix features achieved AUC 0.985, hypoxia-only features achieved AUC 0.929, and adding hypoxia did not improve extracellular-matrix-only prediction. Conclusions: Quantitative extracellular-matrix imaging provides a strong preclinical signal for therapy-response stratification in 4T1 breast cancer. The findings are hypothesis-generating and require independent preclinical and human validation before clinical translation. pubtype: Academic Journal doctype: pictorial research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|