Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review
Abstract Background Pathological complete response (pCR) to neoadjuvant systemic therapy (NAST) is an established prognostic marker in breast cancer (BC). Multimodal deep learning (DL), integrating diverse data sources (radiology, pathology, omics, clinical), holds promise for improving pCR predicti...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-06-01
|
| Series: | Biology Direct |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13062-025-00661-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849472476332949504 |
|---|---|
| author | Eriseld Krasniqi Lorena Filomeno Teresa Arcuri Gianluigi Ferretti Simona Gasparro Alberto Fulvi Arianna Roselli Loretta D’Onofrio Laura Pizzuti Maddalena Barba Marcello Maugeri-Saccà Claudio Botti Franco Graziano Ilaria Puccica Sonia Cappelli Fabio Pelle Flavia Cavicchi Amedeo Villanucci Ida Paris Fabio Calabrò Sandra Rea Maurizio Costantini Letizia Perracchio Giuseppe Sanguineti Silvia Takanen Laura Marucci Laura Greco Rami Kayal Luca Moscetti Elisa Marchesini Nicola Calonaci Giovanni Blandino Giulio Caravagna Patrizia Vici |
| author_facet | Eriseld Krasniqi Lorena Filomeno Teresa Arcuri Gianluigi Ferretti Simona Gasparro Alberto Fulvi Arianna Roselli Loretta D’Onofrio Laura Pizzuti Maddalena Barba Marcello Maugeri-Saccà Claudio Botti Franco Graziano Ilaria Puccica Sonia Cappelli Fabio Pelle Flavia Cavicchi Amedeo Villanucci Ida Paris Fabio Calabrò Sandra Rea Maurizio Costantini Letizia Perracchio Giuseppe Sanguineti Silvia Takanen Laura Marucci Laura Greco Rami Kayal Luca Moscetti Elisa Marchesini Nicola Calonaci Giovanni Blandino Giulio Caravagna Patrizia Vici |
| author_sort | Eriseld Krasniqi |
| collection | DOAJ |
| description | Abstract Background Pathological complete response (pCR) to neoadjuvant systemic therapy (NAST) is an established prognostic marker in breast cancer (BC). Multimodal deep learning (DL), integrating diverse data sources (radiology, pathology, omics, clinical), holds promise for improving pCR prediction accuracy. This systematic review synthesizes evidence on multimodal DL for pCR prediction and compares its performance against unimodal DL. Methods Following PRISMA, we searched PubMed, Embase, and Web of Science (January 2015–April 2025) for studies applying DL to predict pCR in BC patients receiving NAST, using data from radiology, digital pathology (DP), multi-omics, and/or clinical records, and reporting AUC. Data on study design, DL architectures, and performance (AUC) were extracted. A narrative synthesis was conducted due to heterogeneity. Results Fifty-one studies, mostly retrospective (90.2%, median cohort 281), were included. Magnetic resonance imaging and DP were common primary modalities. Multimodal approaches were used in 52.9% of studies, often combining imaging with clinical data. Convolutional neural networks were the dominant architecture (88.2%). Longitudinal imaging improved prediction over baseline-only (median AUC 0.91 vs. 0.82). Overall, the median AUC across studies was 0.88, with 35.3% achieving AUC ≥ 0.90. Multimodal models showed a modest but consistent improvement over unimodal approaches (median AUC 0.88 vs. 0.83). Omics and clinical text were rarely primary DL inputs. Conclusion DL models demonstrate promising accuracy for pCR prediction, especially when integrating multiple modalities and longitudinal imaging. However, significant methodological heterogeneity, reliance on retrospective data, and limited external validation hinder clinical translation. Future research should prioritize prospective validation, integration underutilized data (multi-omics, clinical), and explainable AI to advance DL predictors to the clinical setting. |
| format | Article |
| id | doaj-art-8c69fa7995384eab8d98b8ab82c0f397 |
| institution | Kabale University |
| issn | 1745-6150 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | BMC |
| record_format | Article |
| series | Biology Direct |
| spelling | doaj-art-8c69fa7995384eab8d98b8ab82c0f3972025-08-20T03:24:32ZengBMCBiology Direct1745-61502025-06-0120111910.1186/s13062-025-00661-8Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic reviewEriseld Krasniqi0Lorena Filomeno1Teresa Arcuri2Gianluigi Ferretti3Simona Gasparro4Alberto Fulvi5Arianna Roselli6Loretta D’Onofrio7Laura Pizzuti8Maddalena Barba9Marcello Maugeri-Saccà10Claudio Botti11Franco Graziano12Ilaria Puccica13Sonia Cappelli14Fabio Pelle15Flavia Cavicchi16Amedeo Villanucci17Ida Paris18Fabio Calabrò19Sandra Rea20Maurizio Costantini21Letizia Perracchio22Giuseppe Sanguineti23Silvia Takanen24Laura Marucci25Laura Greco26Rami Kayal27Luca Moscetti28Elisa Marchesini29Nicola Calonaci30Giovanni Blandino31Giulio Caravagna32Patrizia Vici33Phase IV Clinical Studies Unit, IRCCS Regina Elena National Cancer InstitutePhase IV Clinical Studies Unit, IRCCS Regina Elena National Cancer InstitutePhase IV Clinical Studies Unit, IRCCS Regina Elena National Cancer InstituteDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteClinical Trial Center, Biostatistics and Bioinformatics Division, IRCCS Regina Elena National Cancer InstituteBreast Surgery Department, IRCCS Regina Elena National Cancer InstituteBreast Surgery Department, IRCCS Regina Elena National Cancer InstituteBreast Surgery Department, IRCCS Regina Elena National Cancer InstituteBreast Surgery Department, IRCCS Regina Elena National Cancer InstituteBreast Surgery Department, IRCCS Regina Elena National Cancer InstituteBreast Surgery Department, IRCCS Regina Elena National Cancer InstituteBreast Surgery Department, IRCCS Regina Elena National Cancer InstituteDivision of Gynecologic Oncology, Department of Woman and Child Health and Public Health, Fondazione Policlinico Universitario Agostino Gemelli IRCCSDivision of Medical Oncology 1, IRCCS Regina Elena National Cancer InstituteNuclear Medicine Unit, IRCCS Regina Elena National Cancer InstituteDepartment of Plastic and Reconstructive Surgery, IRCCS Regina Elena National Cancer InstituteDepartment of Pathology, IRCCS Regina Elena National Cancer InstituteDepartment of Radiation Oncology, IRCCS Regina Elena National Cancer InstituteDepartment of Radiation Oncology, IRCCS Regina Elena National Cancer InstituteDepartment of Radiation Oncology, IRCCS Regina Elena National Cancer InstituteRadiology Unit, IRCCS Regina Elena National Cancer InstituteRadiology Unit, IRCCS Regina Elena National Cancer InstituteOncology and Hemathology Department, Azienda Ospedaliero-Universitaria Policlinico Di ModenaHospital Pharmacy, IRCCS Regina Elena National Cancer InstituteDepartment of Mathematics, Informatics and Geosciences, University of TriesteTranslational Oncology Research Unit, IRCCS Regina Elena National Cancer InstituteDepartment of Mathematics, Informatics and Geosciences, University of TriestePhase IV Clinical Studies Unit, IRCCS Regina Elena National Cancer InstituteAbstract Background Pathological complete response (pCR) to neoadjuvant systemic therapy (NAST) is an established prognostic marker in breast cancer (BC). Multimodal deep learning (DL), integrating diverse data sources (radiology, pathology, omics, clinical), holds promise for improving pCR prediction accuracy. This systematic review synthesizes evidence on multimodal DL for pCR prediction and compares its performance against unimodal DL. Methods Following PRISMA, we searched PubMed, Embase, and Web of Science (January 2015–April 2025) for studies applying DL to predict pCR in BC patients receiving NAST, using data from radiology, digital pathology (DP), multi-omics, and/or clinical records, and reporting AUC. Data on study design, DL architectures, and performance (AUC) were extracted. A narrative synthesis was conducted due to heterogeneity. Results Fifty-one studies, mostly retrospective (90.2%, median cohort 281), were included. Magnetic resonance imaging and DP were common primary modalities. Multimodal approaches were used in 52.9% of studies, often combining imaging with clinical data. Convolutional neural networks were the dominant architecture (88.2%). Longitudinal imaging improved prediction over baseline-only (median AUC 0.91 vs. 0.82). Overall, the median AUC across studies was 0.88, with 35.3% achieving AUC ≥ 0.90. Multimodal models showed a modest but consistent improvement over unimodal approaches (median AUC 0.88 vs. 0.83). Omics and clinical text were rarely primary DL inputs. Conclusion DL models demonstrate promising accuracy for pCR prediction, especially when integrating multiple modalities and longitudinal imaging. However, significant methodological heterogeneity, reliance on retrospective data, and limited external validation hinder clinical translation. Future research should prioritize prospective validation, integration underutilized data (multi-omics, clinical), and explainable AI to advance DL predictors to the clinical setting.https://doi.org/10.1186/s13062-025-00661-8Breast cancerNeoadjuvant treatmentDeep learningMultimodal prediction |
| spellingShingle | Eriseld Krasniqi Lorena Filomeno Teresa Arcuri Gianluigi Ferretti Simona Gasparro Alberto Fulvi Arianna Roselli Loretta D’Onofrio Laura Pizzuti Maddalena Barba Marcello Maugeri-Saccà Claudio Botti Franco Graziano Ilaria Puccica Sonia Cappelli Fabio Pelle Flavia Cavicchi Amedeo Villanucci Ida Paris Fabio Calabrò Sandra Rea Maurizio Costantini Letizia Perracchio Giuseppe Sanguineti Silvia Takanen Laura Marucci Laura Greco Rami Kayal Luca Moscetti Elisa Marchesini Nicola Calonaci Giovanni Blandino Giulio Caravagna Patrizia Vici Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review Biology Direct Breast cancer Neoadjuvant treatment Deep learning Multimodal prediction |
| title | Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review |
| title_full | Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review |
| title_fullStr | Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review |
| title_full_unstemmed | Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review |
| title_short | Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review |
| title_sort | multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer a systematic review |
| topic | Breast cancer Neoadjuvant treatment Deep learning Multimodal prediction |
| url | https://doi.org/10.1186/s13062-025-00661-8 |
| work_keys_str_mv | AT eriseldkrasniqi multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT lorenafilomeno multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT teresaarcuri multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT gianluigiferretti multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT simonagasparro multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT albertofulvi multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT ariannaroselli multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT lorettadonofrio multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT laurapizzuti multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT maddalenabarba multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT marcellomaugerisacca multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT claudiobotti multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT francograziano multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT ilariapuccica multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT soniacappelli multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT fabiopelle multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT flaviacavicchi multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT amedeovillanucci multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT idaparis multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT fabiocalabro multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT sandrarea multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT mauriziocostantini multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT letiziaperracchio multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT giuseppesanguineti multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT silviatakanen multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT lauramarucci multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT lauragreco multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT ramikayal multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT lucamoscetti multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT elisamarchesini multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT nicolacalonaci multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT giovanniblandino multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT giuliocaravagna multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview AT patriziavici multimodaldeeplearningforpredictingneoadjuvanttreatmentoutcomesinbreastcancerasystematicreview |