MAGIC-DR: An interpretable machine-learning guided approach for acute myeloid leukemia measurable residual disease analysis.

Advanced Search

Kevin Shopsowitz, Jack Lofroth, Geoffrey Chan, Jubin Kim, Makhan Rana, Ryan Brinkman, Andrew Weng, Nadia Medvedev, Xuehai Wang

Author Information

Kevin Shopsowitz: Division of Hematopathology, Vancouver General Hospital, Vancouver, British Columbia, Canada.
Jack Lofroth: Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada.
Geoffrey Chan: Division of Hematopathology, Vancouver General Hospital, Vancouver, British Columbia, Canada.
Jubin Kim: Terry Fox Lab, BC Cancer, Vancouver, British Columbia, Canada.
Makhan Rana: Division of Hematopathology, Vancouver General Hospital, Vancouver, British Columbia, Canada.
Ryan Brinkman: Terry Fox Lab, BC Cancer, Vancouver, British Columbia, Canada.
Andrew Weng: Department of pathology and laboratory medicine, University of British Columbia, Vancouver, British Columbia, Canada.
Nadia Medvedev: Division of Hematopathology, Vancouver General Hospital, Vancouver, British Columbia, Canada.
Xuehai Wang: Division of Hematopathology, Vancouver General Hospital, Vancouver, British Columbia, Canada.

PMID: 38415807 DOI: 10.1002/cyto.b.22168

Multiparameter flow cytometry is widely used for acute myeloid leukemia minimal residual disease testing (AML MRD) but is time consuming and demands substantial expertise. Machine learning offers potential advancements in accuracy and efficiency, but has yet to be widely adopted for this application. To explore this, we trained single cell XGBoost classifiers from 98 diagnostic AML cell populations and 30 MRD negative samples. Performance was assessed by cross-validation. Predictions were integrated with UMAP as a heatmap parameter for an augmented/interactive AML MRD analysis framework, which was benchmarked against traditional MRD analysis for 25 test cases. The results showed that XGBoost achieved a median AUC of 0.97, effectively distinguishing diverse AML cell populations from normal cells. When integrated with UMAP, the classifiers highlighted MRD populations against the background of normal events. Our pipeline, MAGIC-DR, incorporated classifier predictions and UMAP into flow cytometry standard (FCS) files. This enabled a human-in-the-loop machine learning guided MRD workflow. Validation against conventional analysis for 25 MRD samples showed 100% concordance in myeloid blast detection, with MAGIC-DR also identifying several immature monocytic populations not readily found by conventional analysis. In conclusion, Integrating a supervised classifier with unsupervised dimension reduction offers a robust method for AML MRD analysis that can be seamlessly integrated into conventional workflows. Our approach can support and augment human analysis by highlighting abnormal populations that can be gated on for quantification and further assessment. This has the potential to speed up MRD analysis, and potentially improve detection sensitivity for certain AML immunophenotypes.

XGBoost acute myeloid leukemia dimensionality reduction flow cytometry machine learning measurable residual disease

Cava, W. L., Bauer, C., Moore, J. H., & Pendergrass, S. A. (2019). Interpretation of machine learning predictions for patient outcomes in electronic health records. AMIA Annual Symposium Proceedings. AMIA Symposium, 2019, 572–581.

Chen, T., & Guestrin, C. (2016). XGBoost: XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. Association for Computing Machinery, New York (pp. 785–794).

Chulián, S., Martínez‐Rubio, Á., Pérez‐García, V. M., Rosa, M., Goñi, C. B., Gutiérrez, J. F. R., Hermosín‐Ramos, L., Quintana, Á. M., Caballero‐Velázquez, T., Ramírez‐Orellana, M., Robleda, A. C., & Fernández‐Martínez, J. L. (2020). High‐dimensional analysis of single‐cell flow cytometry data predicts relapse in childhood acute lymphoblastic leukaemia. Cancers, 13(1), 17.

Guldberg, S. M., Okholm, T. L. H., McCarthy, E. E., & Spitzer, M. H. (2023). Computational methods for single‐cell proteomics. Annual Review of Biomedical Data Science, 6(1), 47–71.

Heuser, M., Freeman, S. D., Ossenkoppele, G. J., Buccisano, F., Hourigan, C. S., Ngai, L. L., Tettero, J. M., Bachas, C., Baer, C., Béné, M. C., Buecklein, V., Czyz, A., Denys, B., Dillon, R., Feuring‐Buske, M., Guzman, M. L., Haferlach, T., Han, L., Herzig, J. K., … Cloos, J. (2021). 2021 update measurable residual disease in acute myeloid leukemia: European LeukemiaNet working party consensus document. Blood, 138(26), 2753–2767.

Jongen‐Lavrencic, M., Grob, T., Hanekamp, D., Kavelaars, F. G., Al Hinai, A., Zeilemaker, A., Erpelinck‐Verschueren, C. A. J., Gradowska, P. L., Meijer, R., Cloos, J., Biemond, B. J., Graux, C., Kooy, M., Van, M., Manz, M. G., Pabst, T., Passweg, J. R., Havelange, V., Ossenkoppele, G. J., … Valk, P. J. M. (2018). Molecular minimal residual disease in acute myeloid leukemia. The New England Journal of Medicine, 378(13), 1189–1199.

Khoury, J. D., Solary, E., Abla, O., Akkari, Y., Alaggio, R., Apperley, J. F., Bejar, R., Berti, E., Busque, L., Chan, J. K. C., Chen, W., Chen, X., Chng, W.‐J., Choi, J. K., Colmenero, I., Coupland, S. E., Cross, N. C. P., Jong, D. D., Elghetany, M. T., … Hochhaus, A. (2022). The 5th edition of the World Health Organization classification of haematolymphoid tumours: Myeloid and histiocytic/dendritic neoplasms. Leukemia, 36(7), 1703–1719.

Lundberg, S., & Lee, S.‐I. (2017). A unified approach to interpreting model predictions. ArXiv.

Malek, M., Taghiyar, M. J., Chong, L., Finak, G., Gottardo, R., & Brinkman, R. R. (2015). FlowDensity: Reproducing manual gating of flow cytometry data by automated density‐based cell population identification. Bioinformatics, 31(4), 606–607.

McInnes, L., Healy, J., & Melville, J. (2018). UMAP: Uniform manifold approximation and projection for dimension reduction. ArXiv.

Montante, S., & Brinkman, R. R. (2019). Flow cytometry data analysis: Recent tools and algorithms. International Journal of Laboratory Hematology, 41(S1), 56–62.

Reiter, M., Diem, M., Schumich, A., Maurer‐Granofszky, M., Karawajew, L., Rossi, J. G., Ratei, R., Groeneveld‐Krentz, S., Sajaroff, E. O., Suhendra, S., Kampel, M., Dworzak, M. N., & International Berlin‐Frankfurt‐Münster (iBFM)‐FLOW‐network and the AutoFLOW consortium. (2019). Automated flow Cytometric MRD assessment in childhood acute B‐ lymphoblastic leukemia using supervised machine learning. Cytometry Part A, 95(9), 966–975.

Salama, M. E., Otteson, G. E., Camp, J. J., Seheult, J. N., Jevremovic, D., Holmes, D. R., Olteanu, H., & Shi, M. (2022). Artificial intelligence enhances diagnostic flow cytometry workflow in the detection of minimal residual disease of chronic lymphocytic leukemia. Cancers, 14(10), 2537.

Schuurhuis, G. J., Heuser, M., Freeman, S., Béné, M.‐C., Buccisano, F., Cloos, J., Grimwade, D., Haferlach, T., Hills, R. K., Hourigan, C. S., Jorgensen, J. L., Kern, W., Lacombe, F., Maurillo, L., Preudhomme, C., Thiede, C., Venditti, A., Vyas, P., … Ossenkoppele, G. J. (2018). Minimal/measurable residual disease in AML: A consensus document from the European LeukemiaNet MRD working party. Blood, 131(12), 1275–1291.

Shopsowitz, K. E., Liu, L., Setiadi, A., Al‐Bakri, M., & Vercauteren, S. (2022). Machine learning optimized multiparameter radar plots for B‐cell acute lymphoblastic leukemia minimal residual disease analysis. Cytometry Part B: Clinical Cytometry, 102(5), 342–352.

Vial, J. P., Lechevalier, N., Lacombe, F., Dumas, P.‐Y., Bidet, A., Leguay, T., Vergez, F., Pigneux, A., & Béné, M. C. (2021). Unsupervised flow cytometry analysis allows for an accurate identification of minimal residual disease assessment in acute myeloid leukemia. Cancers, 13(4), 629.

Weijler, L., Kowarsch, F., Wödlinger, M., Reiter, M., Maurer‐Granofszky, M., Schumich, A., & Dworzak, M. N. (2022). UMAP based anomaly detection for minimal residual disease quantification within acute myeloid leukemia. Cancers, 14(4), 898.

Wood, B. L. (2020). Acute myeloid leukemia minimal residual disease detection: The difference from normal approach. Current Protocols in Cytometry, 93(1), e73.

Xu, J., Jorgensen, J. L., & Wang, S. A. (2017). How do we use multicolor flow cytometry to detect minimal residual disease in acute myeloid leukemia? Clinics in Laboratory Medicine, 37(4), 787–802.

/BD Bioscience Investigator Sponsored Study
/Faculty of Medicine, University of British Columbia

Humans

Leukemia, Myeloid, Acute

Neoplasm, Residual

Flow Cytometry

Machine Learning

Immunophenotyping

Journal Article

OpenLB
Open Library of Bioscience