-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.

3600 Deep Learning Predicts JAK2, Calr TET2, and ASXL1 directly from Whole Slide Marrow Images

Program: Oral and Poster Abstracts
Session: 803. Emerging Tools, Techniques, and Artificial Intelligence in Hematology: Poster II
Hematology Disease Topics & Pathways:
Research, Artificial intelligence (AI), MPN, Clinical Research, Bioinformatics, Chronic Myeloid Malignancies, Diseases, Computational biology, Myeloid Malignancies, Technology and Procedures, Imaging, Machine learning, Molecular testing, Pathology
Sunday, December 8, 2024, 6:00 PM-8:00 PM

Spencer Krichevsky, MS1,2*, Yuwei Zhang, MS2*, Madhu M Ouseph, MD, MBBS, PhD3*, Tahmeena Ahmed, MBBS4*, Ghaith Abu-Zeinah, MD5, Joseph M. Scandura, MD, PhD6 and Rajarsi Gupta, MD, PhD2*

1Richard T. Silver, M.D. Myeloproliferative Neoplasms (MPN) Center, Weill Cornell Medicine, New York, NY
2Department of Biomedical Informatics, Stony Brook University, Stony Brook, NY
3Department of Pathology and Laboratory Medicine, Weill Cornell Medicine/New York Presbyterian Hospital, New York, NY
4Department of Pathology, Stony Brook Medicine, Stony Brook, NY
5Weill Cornell Medicine, New York, NY
6Richard T. Silver, MD Myeloproliferative Neoplasms Center, Weill Cornell Medicine, New York, NY

Introduction:

Molecular subclassification and risk stratification of Philadelphia chromosome-negative myeloproliferative neoplasms (MPNs) rely on identifying myeloid driver (e.g., JAK2, MPL, CALR) and co-occurrent somatic mutations (e.g., TET2, TP53, ASXL1) [Grinfeld, 2019; Tashkandi, 2024]. This study develops deep learning (DL) models to predict these mutations using Pathomics image analysis of digital hematoxylin and eosin (H&E) whole slide images (WSIs) of MPN bone marrow core biopsies. The translational goal is to use DL Pathomics mutation for efficient, cost-effective screening, allowing rapid identification of abnormal marrows and focusing histologic review on specific regions, thereby enhancing objective diagnostic accuracy and patient care [Kather, 2019]. This result can also identify samples for confirmation with traditional laboratory molecular testing and next-generation sequencing.

Methods:

Following institutional review from Stony Brook University and Weill Cornell Medicine, we identified 271 marrow slides using the WMC MPN research data repository. These slides were from newly diagnosed patients with polycythemia vera (PV; n=83, 31%), essential thrombocytosis (ET; n=83, 31%), myelofibrosis (MF; n=76, 28%), and tumor-negative, normal controls collected for Non-Hodgkin's lymphoma staging (n=35, 13%). The dataset includes H&E WSIs with robust clinical and molecular annotations.


We trained and evaluated six DL algorithms to predict JAK2, CALR, TET2, and ASXL1 by using multiple instance learning (MIL) models [Dolezal, 2024] trained on extracted features (ResNet, HistoSSL, RetCCL, and CTransPath) and end-to-end ResNet and DenseNet models [Liechty, 2022]. A balanced selection strategy allocated 75% of WSIs for training and 25% for testing. WSIs were subdivided into 96K 256 μm square image patches at 10x magnification for computational efficiency and parallel processing. Patch-level results were aggregated into slide-level mutation predictions and used to generate attention heatmaps to examine spatial heterogeneity. Model performance was evaluated by using precision, recall, and F1-score.

Results:

The DL models showed strong performance in predicting mutations from WSIs. For JAK2 mutation prediction, the models achieved precision (P) of 90%, recall (R) of 98%, and F1 of 87%. The corresponding performance for identifying CALR mutations was P=88%, R=98%, and F1=86%. For TET2 mutation prediction, P=78%, R=100%, and F1=78% and for ASXL1 performance was P=76%, R=100%, and F1=76%.

MIL models outperformed DenseNet and ResNet models in terms of precision and F1, and the DenseNet and ResNet models were impacted by smaller training sizes. The RetCCL and CTransPath MIL models excelled, often exceeding average metrics across all prediction tasks. Attention heatmaps revealed that hematopoietic red bone marrow regions correspond to high discriminative power, underscoring the models’ ability to identify key pathologic features.

Conclusions:

Our results demonstrate that Pathomics image analysis can effectively predict myeloid driver and co-occurrent somatic mutations in MPNs. MIL feature extractors that preserve spatial relationships improve model performance. The attention heatmaps generated by our workflow assist pathologists in reviewing and interpreting mutation predictions within the bone marrow microenvironment, facilitating histologic, laboratory, and clinical correlation. This approach can expedite analysis, objectify classification, and select samples for molecular confirmation studies, ultimately improving the identification of abnormalities and focusing attention on specific regions for histologic review.

Disclosures: Scandura: Constellation: Consultancy, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Protagonist Therapeutics: Membership on an entity's Board of Directors or advisory committees; Incyte: Membership on an entity's Board of Directors or advisory committees; Morphic: Consultancy; SDP Oncology: Membership on an entity's Board of Directors or advisory committees; Medpacto: Research Funding; Calico: Consultancy. Gupta: Chilean Wool, LLC: Current equity holder in private company, Membership on an entity's Board of Directors or advisory committees.

*signifies non-member of ASH