Academic Journal
Leveraging Kappa-Lambda Signatures in a Multistage Machine Learning Pipeline for B-Cell Lymphoma Detection by Flow Cytometry.
| Title: | Leveraging Kappa-Lambda Signatures in a Multistage Machine Learning Pipeline for B-Cell Lymphoma Detection by Flow Cytometry. |
|---|---|
| Authors: | Zhang I; Department of Biostatistics, School of Global Public Health, New York University, New York, New York., Chalise S; Memorial Sloan Kettering Cancer Center, New York, New York., Roshal M; Memorial Sloan Kettering Cancer Center, New York, New York., Gao Q; Memorial Sloan Kettering Cancer Center, New York, New York., Zhu M; Memorial Sloan Kettering Cancer Center, New York, New York. Electronic address: zhum1@mskcc.org., Feng Y; Department of Biostatistics, School of Global Public Health, New York University, New York, New York; Memorial Sloan Kettering Cancer Center, New York, New York. Electronic address: yf31@nyu.edu. |
| Source: | The American journal of pathology [Am J Pathol] 2026 May; Vol. 196 (5), pp. 1158-1168. Date of Electronic Publication: 2026 Mar 05. |
| Publication Type: | Journal Article |
| Language: | English |
| Journal Info: | Publisher: Elsevier Country of Publication: United States NLM ID: 0370502 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1525-2191 (Electronic) Linking ISSN: 00029440 NLM ISO Abbreviation: Am J Pathol Subsets: MEDLINE |
| Imprint Name(s): | Publication: 2011-: New York : Elsevier Original Publication: Philadelphia [etc.] American Assn. of Pathologists [etc.] |
| MeSH Terms: | Flow Cytometry*/methods , Lymphoma, B-Cell*/diagnosis , Lymphoma, B-Cell*/immunology , Immunoglobulin kappa-Chains*/metabolism , Immunoglobulin lambda-Chains*/metabolism , Machine Learning*, Immunophenotyping/methods ; Humans ; Classification Algorithms ; Boosting Machine Learning Algorithms ; Biomarkers, Tumor |
| Abstract: | Flow cytometry immunophenotyping is essential for diagnosing B-cell lymphomas, but manual interpretation of high-dimensional data remains subjective, time-consuming, and prone to interoperator variability. Previous computational approaches often overlook clinically relevant principles, such as Ig light chain restriction. To address this gap, a biologically informed, three-stage machine learning pipeline that integrates Ig κ (IGK) and Ig λ (IGL) signatures to improve B-cell lymphoma detection was developed. A total of 200 peripheral blood samples (100 normal, 100 abnormal) were analyzed, comprising >15 million single-cell events characterized by 21 immunophenotypic markers. Three XGBoost models were trained sequentially: the first classified light chain expression (IGK, IGL, or nuisance), the second identified cell phenotypes using marker intensities and IGK/IGL-based neighborhood enrichment, and the third produced sample-level predictions based on aggregated cell features. The IGK/IGL classifier achieved 88.0% test accuracy [area under the receiver operating characteristic curve (AUC), 0.957], whereas the cell-level classification reached 92.9% accuracy (AUC, 0.983), with IGK/IGL enrichment as the most informative feature. Similarly, sample-level classification achieved 94.7% accuracy (AUC, 0.976), with improved performance when IGK/IGL enrichment was included. These findings demonstrate that incorporating biologically grounded features enhances both the accuracy and interpretability of automated flow cytometry analysis. This approach offers a scalable, reproducible, and clinically aligned alternative to the manual review of flow cytometry data for B-cell lymphomas. (Copyright © 2026 The Author(s). Published by Elsevier Inc. All rights reserved.) |
| Competing Interests: | Disclosure Statement None declared. |
| Substance Nomenclature: | 0 (Immunoglobulin kappa-Chains) 0 (Immunoglobulin lambda-Chains) 0 (Biomarkers, Tumor) |
| Entry Date(s): | Date Created: 20260307 Date Completed: 20260710 Latest Revision: 20260710 |
| Update Code: | 20260711 |
| DOI: | 10.1016/j.ajpath.2026.02.006 |
| PMID: | 41794128 |
| Database: | MEDLINE |
| ISSN: | 1525-2191 |
|---|---|
| DOI: | 10.1016/j.ajpath.2026.02.006 |