-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

3598 Artificial Intelligence-Powered Digital Pathology to Improve Diagnosis and Personalized Prognostic Assessment in Patient with Myeloid Neoplasms

Program: Oral and Poster Abstracts
Session: 803. Emerging Tools, Techniques, and Artificial Intelligence in Hematology: Poster II
Hematology Disease Topics & Pathways:
Research, Acute Myeloid Malignancies, AML, MDS, Artificial intelligence (AI), Adult, Translational Research, MPN, Elderly, Clinical Research, Chronic Myeloid Malignancies, Diseases, Real-world evidence, Myeloid Malignancies, Biological Processes, Emerging technologies, Technology and Procedures, Multi-systemic interactions, Study Population, Human, Imaging, Machine learning, Omics technologies, Pathology
Sunday, December 8, 2024, 6:00 PM-8:00 PM

Gianluca Asti, MSc1*, Nico Curti, PhD2*, Giulia Maggioni, MD1*, Gianluca Carlini, PhD3*, Luca Lanino, MD4, Alessia Campagna, MD1*, Saverio D'Amico, MSc1,5*, Elisabetta Sauta, PhD1*, Mattia Delleani1*, Arturo Bonometti, MD1*, Cesare Lancellotti, MD6*, Daoud Rahal, MD1*, Marta Ubezio, MD1*, Gabriele Todisco, MD7,8*, Cristina Astrid Tentori, MD1*, Antonio Russo, MD1*, Alessandra Crespi, MD1*, Giulia Figini, MD1*, Alessandro Buizza, MD1*, Elena Riva, PhD4*, Matteo Zampini, PhD4*, Matteo Brindisi4,9*, Francesca Ficara, PhD1,9*, Laura Crisafulli, PhD4,9*, Denise Ventura, MSc4*, Nicole Pinocchio, MSc4*, Elena Zazzetti, MSc1*, Marilena Bicchieri, PhD1*, Maria Chiara Grondelli, BSc8*, Alessandro Forcina Barrero, BSc8*, Victor Savevski, MEng1*, Armando Santoro, MD8,10*, Valeria Santini, MD11, Francesc Sole, PhD12, Uwe Platzbecker, MD13, Pierre Fenaux, MD14, Maria Diez-Campelo, MD, PhD15*, Rami S. Komrokji, MD16, Torsten Haferlach, MD17, Shahram Kordasti, MD, PhD18,19, Luca Di Tommaso, MD1,8*, Amer M. Zeidan, MD20, Sanam Loghavi, MD21, Guillermo Garcia-Manero, MD22, Gastone Castellani, PhD23* and Matteo Giovanni Della Porta, MD1,8*

1IRCCS Humanitas Research Hospital, Rozzano, Milan, Italy
2Department of Physics and Astronomy, University of Bologna, Bologna, Italy
3Data Science and Bioinformatics Laboratory, IRCCS Institute of Neurological Sciences of Bologna, Bologna, Italy
4IRCCS Humanitas Research Hospital, Rozzano, Milano, Italy
5Train s.r.l., Rozzano, Milan, Italy
6Struttura Complessa di Anatomia Patologica, Policlinico di Modena, Modena, Italy
7IRCCS Humanitas Research Hospital, Houston, TX
8Department of Biomedical Sciences, Humanitas University, Pieve Emanuele, Milan, Italy
9Institute for Genetic and Biomedical Research, Milan Unit, CNR, Milan, Italy
10IRCCS Humanitas Research Hospital, Rozzano, Italy
11MDS Unit, Hematology, AOUC, University of Florence, Florence, Italy
12Myelodysplastic Syndromes Research Group, Institut De Recerca Josep Carreras, Badalona, Barcelona, Spain
13Medical Clinic and Policlinic 1, Hematology and Cellular Therapy, University Hospital Leipzig, Leipzig, Germany
14Département (DMU) d'hématologie et immunologie, Service d'hématologie séniors, AP-HP Hospital Saint-Louis, Paris, France
15Department of Hematology, Hospital Universitario de Salamanca-IBSAL, Salamanca, Spain
16Department of Malignant Hematology, Moffitt Cancer Center, Tampa, FL
17MLL Munich Leukemia Laboratory, Munich, Germany
18Hematology Unit, Department of Clinical and Molecular Sciences, Università Politecnica delle Marche, Ancona, Ancona, Italy
19Department of Clinical Haematology, Guy's and St Thomas' NHS Foundation Trust, London, United Kingdom
20Yale School of Medicine, Smilow Cancer Hospital at Yale New Haven, New Haven, CT
21Department of Hematopathology, The University of Texas MD Anderson Cancer Center, Houston, TX
22Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, TX
23Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy

BACKGROUND. Bone marrow (BM) cytology and histopathology images are crucial for diagnosing and prognosticating myeloid neoplasms (MNs), but their high-dimensional data are underused. Artificial Intelligence (AI) applied to tumor morphology (digital pathology, DP) has improved the use of tumor biopsies’ data for various types of malignancies, accurately detecting patterns and converting complex image information into numerical features. Here, we explored the potential of AI-based DP to improve personalized medicine in MNs which are characterized by high heterogeneity and a significant proportion of patients with unmet clinical needs.

AIMS. This project was conducted by the GenoMed4All and Synthema consortia, to build AI-based features extraction tools from BM histopathological and cytological Whole Slide Images (WSI). High-dimensional data were used to 1) assess diagnostic accuracy in MN patients; 2) elucidate the association between morphologic features, clinical variables and molecular genetics, and 3) create an innovative tool for personalized risk assessments integrating morphological features with clinical and genomic information.

METHODS. The study population included 1167 MN patients(18% AML; 44% MDS, 29% MPN, 9% MDS/MPN) from Humanitas Research Hospital, with available demographic, clinical, cytogenetic and genomic data, alongside with treatments, outcomes information and WSI collected from BM aspiration and biopsy at different timepoints during disease history. AI-based features extraction models were developed for different staining (May-Grunwald Giemsa (MGG), Perls, Hematoxylin-Eosin, Gomori, CD34, MPO, CD117, CD71, CD61, CD20, CD3, and P53). Different models and training strategies were employed to segment structures in each staining, like nuclei, cells and fibrosis; morphological and textural features were then extracted. Hematologists and hemato-pathologists feedback was included in the training process. Moreover, a classification model was developed to analyze cells on MGG smears, predicting 12 different cell types. The percentages of markers, cell types, and tissue components, combined with their spatial organization, were integrated to address the clinical aims of the project. Explainability and interpretability were implemented by SHapley Additive exPlanations approach (SHAP).

RESULTS. Multiple Gradient Boosting binary classifiers were trained using the WSI features to discriminate specific clinical entities among MN. The models predicted a correct diagnosis with an F1 Score > 90%, suggesting that extracted features capture clinically relevant information. MDS/MPN phenotype cases resulted in the most diagnostic disagreement between AI-models and pathologists.

Then we analysed the morphologic and molecular features association. Specific genomic profiles were predicted from WSI features with specialized XGBOOST models with high accuracy, in particular for SF3B1, JAK/STAT, TP53 and RUNX1 mutations (all with F1 Score > 90%).
These findings underline the capability of the morphological features to capture the biological background of MN.

On these bases, we integrated morphological features into an innovative prognostic tool for personalized prediction of overall survival (OS) and leukemia-free survival (LFS) in MN. After the feature selection process (by using a L1-penalized Cox regression) 71 morphological features were included in the model together with demographic, clinical and genomic information. Model discrimination was assessed using Harrell’s concordance index (CI).

Sequential integration of data layers into the model showed an increasing CI for OS and LFS, starting with 0.68 and 0.59, respectively with clinical and cytogenetic features alone; then raising CI up to 0.75 and 0.64 including somatic gene mutations; and finally reaching CI of 0.87 and 0.85 further integrating morphological features.

Validation is currently ongoing on an independent MN population from MD Anderson Cancer Center, US.

CONCLUSION. Our developed AI models were capable of automatically extract textural, cell features and markers from WSI. The morphological features integration significantly improved personalized patient stratification and prognostication in MN. The proposed solution provided a fully explainable interpretation of high dimensional features, thus facilitating a wide clinical implementation.

Disclosures: Santoro: Novartis: Speakers Bureau; Beigene: Speakers Bureau; Sandoz: Speakers Bureau; Lilly: Speakers Bureau; Arqule: Speakers Bureau; Astrazeneca: Speakers Bureau; Celgene: Speakers Bureau; Amgen: Speakers Bureau; Abb-vie: Speakers Bureau; Roche: Speakers Bureau; Takeda: Speakers Bureau; MSD: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Bayer: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; EISAI: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Pfizer: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Gilead: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Servier: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; BMS: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Incyte: Consultancy; Sanofi: Consultancy. Santini: Ascentage, AbbVie, Bristol Myers Squibb, CTI BioPharma, Geron, Gilead, Novartis, Servier, Syros Pharmaceuticals: Other: Advisory Board. Platzbecker: Merck: Research Funding; Amgen: Consultancy, Research Funding; MDS Foundation: Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Research Funding; BMS: Consultancy, Membership on an entity's Board of Directors or advisory committees, Other: Travel support, Research Funding; Curis: Consultancy, Honoraria, Research Funding; Geron: Consultancy; Janssen: Consultancy, Honoraria, Research Funding; Novartis: Consultancy, Research Funding. Fenaux: Agios: Research Funding; BMS: Honoraria, Research Funding; Novartis: Research Funding; Janssen: Research Funding; Jazz Pharmaceuticals: Honoraria, Research Funding; Astex: Research Funding; Servier: Research Funding; AbbVie: Honoraria, Research Funding. Diez-Campelo: BMS/Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Advisory board fees; ASTEX/OTSUKA: Membership on an entity's Board of Directors or advisory committees, Other: TRAVEL TO MEETINGS; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; KEROS: Honoraria, Membership on an entity's Board of Directors or advisory committees; BLUEPRINT MEDICINES: Consultancy, Membership on an entity's Board of Directors or advisory committees; AGIOS: Consultancy, Membership on an entity's Board of Directors or advisory committees; CURIS: Membership on an entity's Board of Directors or advisory committees; SYROS: Membership on an entity's Board of Directors or advisory committees; HEMAVAN: Membership on an entity's Board of Directors or advisory committees; Gilead: Other: Travel reimbursement. Komrokji: Janssen: Consultancy; BMS: Honoraria, Membership on an entity's Board of Directors or advisory committees; Sobi: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Keros: Membership on an entity's Board of Directors or advisory committees; DSI: Honoraria, Membership on an entity's Board of Directors or advisory committees; Servier: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; AbbVie: Consultancy, Membership on an entity's Board of Directors or advisory committees; Geron: Consultancy, Membership on an entity's Board of Directors or advisory committees; Servio: Honoraria; Servio: Membership on an entity's Board of Directors or advisory committees; CTI biopharma: Membership on an entity's Board of Directors or advisory committees; DSI: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene/BMS: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Jazz Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; PharmaEssentia: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Genentech: Consultancy; Rigel: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Taiho: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees; BMS: Research Funding; Sumitomo Pharma: Consultancy, Membership on an entity's Board of Directors or advisory committees. Kordasti: API: Consultancy; Alexion: Consultancy; Beckman Coulter: Speakers Bureau; MorphoSys: Research Funding; Pfizer: Consultancy, Speakers Bureau; Novartis: Consultancy, Honoraria, Research Funding, Speakers Bureau; Celgene: Research Funding; Boston Biomed: Consultancy. Loghavi: Pathology Education Partners; VJ HemeOnc, College of American Pathologists, OncLive, ICCS, MD Education, NCCN, MashUp Media, NCTN, Aptitude Health: Honoraria; Guidepoint; QualWorld; Gerson Lehrman Group, AlphaSight, Arima, Qiagen, Opinion Health: Consultancy; Astellas, Amgen: Research Funding; Abbvie: Current holder of stock options in a privately-held company; Syndx, Servier, BMS: Membership on an entity's Board of Directors or advisory committees; Abbvie, Daiichi Sankyo, BluePrint Medicine, Caris Diagnostics, Recordati, Servier: Consultancy. Garcia-Manero: Curis: Research Funding; Janssen: Research Funding; Onconova: Research Funding; Astex: Other: Personal fees; Genentech: Research Funding; Helsinn: Other: Personal fees; AbbVie: Research Funding; Bristol Myers Squibb: Other: Personal fees, Research Funding; Astex: Research Funding; H3 Biomedicine: Research Funding; Merck: Research Funding; Genentech: Other: Personal fees; Forty Seven: Research Funding; Helsinn: Research Funding; Novartis: Research Funding; Aprea: Research Funding; Amphivena: Research Funding. Della Porta: Bristol Myers Squibb: Consultancy.

*signifies non-member of ASH