Academic Journal
Predicting neoadjuvant breast cancer therapy response using BRIDGE from tumor transcriptomics and histopathology.
| Τίτλος: | Predicting neoadjuvant breast cancer therapy response using BRIDGE from tumor transcriptomics and histopathology. |
|---|---|
| Συγγραφείς: | Cantore T; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA. Electronic address: Thomas.Cantore@cshs.org., Hoang DT; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA., Pal LR; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA., Stemmer A; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA; Gray Faculty of Medical and Health Sciences, Tel Aviv University, Tel Aviv, Israel., Chang TG; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA., Dhruba SR; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA., Shulman ED; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA., Campagnolo E; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA., Lee JS; Department of Artificial Intelligence, Republic of Korea; Department of Precision Medicine, School of Medicine, Sungkyunkwan University, Suwon, Republic of Korea., Levy J; Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, USA., Yao K; Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, USA., Liao IC; Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, USA; Institute of Clinical Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan., Stemmer SM; Gray Faculty of Medical and Health Sciences, Tel Aviv University, Tel Aviv, Israel; Davidoff Center, Rabin Medical Center, Petah Tikva, Israel., Sammut SJ; Breast Cancer Now Toby Robins Research Centre, The Institute of Cancer Research, London, UK; The Royal Marsden Hospital NHS Foundation Trust, London, UK., Lipkowitz S; Women's Malignancies Branch, Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA., Rajagopal PS; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Women's Malignancies Branch, Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA., Filipits M; Center for Cancer Research, Medical University of Vienna, Vienna, Austria., Caldas C; Department of Clinical Biochemistry and Institute of Metabolic Science, School of Clinical Medicine, University of Cambridge, Cambridge, UK; Lautenberg Center for Immunology and Cancer Research, Institute of Medical Research Israel-Canada, Faculty of Medicine, Hebrew University of Jerusalem, Israel., Yuan Y; Department of Medicine, Cedars-Sinai Medical Center, Los Angeles, USA., Nair NU; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA. Electronic address: Nishanth.Nair@cshs.org., Ruppin E; Cancer Data Science Laboratory (CDSL), Center for Cancer Research (CCR), National Cancer Institute (NCI), National Institutes of Health (NIH), Bethesda, USA; Jim and Eleanor Randall Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, USA. Electronic address: Eytan.Ruppin@cshs.org. |
| Πηγή: | Annals of oncology : official journal of the European Society for Medical Oncology [Ann Oncol] 2026 Sep; Vol. 37 (9), pp. 1293-1307. Date of Electronic Publication: 2026 May 26. |
| Τύπος έκδοσης: | Evaluation Study; Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Elsevier Country of Publication: England NLM ID: 9007735 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1569-8041 (Electronic) Linking ISSN: 09237534 NLM ISO Abbreviation: Ann Oncol Subsets: MEDLINE |
| Imprint Name(s): | Publication: 2020- : London : Elsevier Original Publication: Dordrecht ; Boston : Kluwer Academic Publishers, c1990- |
| Ιατρικοί όροι (MeSH): | Breast Neoplasms*/genetics , Breast Neoplasms*/pathology , Breast Neoplasms*/therapy , Gene Expression Profiling*/methods , Neoadjuvant Therapy* , Prediction Algorithms*, Erb-b2 Receptor Tyrosine Kinases/metabolism ; Female ; Humans ; Datasets as Topic ; Deep Learning ; Gene Expression Regulation, Neoplastic ; Pathologic Complete Response ; ROC Curve ; Treatment Outcome |
| Περίληψη: | Background: While expression-based signatures inform adjuvant therapy in breast cancer (BC), no approved molecular biomarkers exist for the neoadjuvant setting, where early response prediction could inform treatment decisions. This challenge is compounded by intratumoral heterogeneity, as multiple malignant subtypes may coexist within a tumor and influence therapy sensitivity. Methods: We developed BRIDGE (BReast Intra-tumoral Deconvolution of Gene Expression), a computational framework that deconvolves the pretreatment bulk tumor transcriptome to estimate molecular subtype composition and predict pathological complete response to neoadjuvant therapy. BRIDGE was trained on 10 transcriptomics datasets and tested on 24 independent ones spanning different subtypes. Six additional datasets with pretreatment hematoxylin and eosin slides and response data were analyzed to evaluate histology-based predictions. Results: Analyzing measured transcriptomics, BRIDGE outperformed surrogate implementations of established commercial signatures (Oncotype DX, MammaPrint, ROR-S) in estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative tumors and exceeded other transcriptomic predictors in HER2-positive and triple-negative breast cancer (TNBC) disease. In ER-positive/HER2-negative patients, it yields an receiver operating characteristic-area under the curve (AUC) of 0.84 with a high odds ratio (OR = 8); in HER2-positive disease, an AUC of 0.77 (OR = 8.3); and in TNBC, an AUC of 0.73 (OR = 3.1). We further developed BRIDGE-Slide, which applies BRIDGE to pretreatment histopathology slides via deep learning-inferred transcriptomics. BRIDGE-Slide outperforms direct slide-to-response models, underscoring its potential as a first-of-its-kind, fast, low-cost biomarker. Exploratory leave-one-dataset-out analyses across datasets treated with alternative neoadjuvant regimens suggest generalizability to immune checkpoint blockade-treated ER-positive/HER2-negative tumors, pending validation in larger cohorts. Finally, spatial transcriptomics shows that BRIDGE-derived subtype assignments form spatially cohesive regions aligned with canonical molecular features, reinforcing its biological interpretability. Conclusions: BRIDGE is a biologically grounded framework for neoadjuvant BC response prediction, validated on a rich set of different patients' cohorts. Its histopathology-based version opens the door for fast and low-cost prediction in the neoadjuvant setting, upon further prospective testing and validation. (Copyright © 2026 The Author(s). Published by Elsevier Ltd.. All rights reserved.) |
| Competing Interests: | Disclosure ER is a cofounder of MedAware Ltd. and a cofounder (divested) and nonpaid scientific consultant of Pangea Therapeutics. ER is a member of the scientific advisory board of GSK Oncology, the WIN consortium and the ProCan program. YY reports research funding from Merck and Summit; consultancy roles with Genentech, DSI, AstraZeneca, and Stemline; and participation in the Novartis Speaker’s Bureau, all unrelated to the present study. Although unrelated to this work, SMS received research grants (to the institution) from CAN-FITE, AstraZeneca, Bioline RX, BMS, Halozyme, Clovis Oncology, CTG Pharma, Exelexis, Geicam, Halozyme, Incyte, Lilly, Moderna, Teva pharmaceuticals, and Roche, and owns stocks and options in CTG Pharma, DocBoxMD, Tyrnovo, VYPE, Cytora, CAN-FITE, Kerem. MF has received honoraria from AstraZeneca and Eli Lilly. All other authors have declared no conflicts of interest. |
| Contributed Indexing: | Keywords: AI in oncology; PRECISION MEDICINE; breast tumors; digital pathology; gene expression |
| Substance Nomenclature: | EC 2.7.10.1 (Erb-b2 Receptor Tyrosine Kinases) EC 2.7.10.1 (ERBB2 protein, human) |
| Entry Date(s): | Date Created: 20260527 Date Completed: 20260820 Latest Revision: 20260820 |
| Update Code: | 20260820 |
| DOI: | 10.1016/j.annonc.2026.05.700 |
| PMID: | 42203036 |
| Βάση Δεδομένων: | MEDLINE |
καταχωρήστε σχόλιο πρώτοι!