Post-treatment infection prediction in CLL using domain adaptation of lymphoma electronic health records
DOI:
https://doi.org/10.2340/1651-226X.2026.44569Keywords:
CLL, adverse events, Infection, Treatment, prediction, machine learningAbstract
Background and purpose: Infections are the leading cause of morbidity and mortality in patients with chronic lymphocytic leukemia (CLL) and occur during and after treatment. When deciding on the type of CLL treatment, the risk of infections is typically assessed based only on age and comorbidities; therefore, there is a need to develop a predictive model that incorporates information from multiple data sources. However, training an effective machine learning model requires a large sample size.
Patient/material and methods: In this study, we developed a machine learning approach using domain adaptation (DA) to predict the risk of severe infection during treatment in patients with CLL. We implemented a DA strategy using lymphoma patient data and compared it with a domain-specific (DS) strategy across multiple models.
Results: The DA strategy outperformed the DS strategy across all models, with an odds ratio of 4.43 for infection risk between high-risk and low-risk groups, compared with an odds ratio of 3.69 for the best DS model and 2.27 for the CLL-IPI alone. Explainability analysis identified predictive features for both the DA and DS models, including medication data and biochemistry tests. Specifically, C-reactive protein levels and non-therapeutic drugs were common features identified by both DA and DS models, while the DA models relied more heavily on alimentary tract drugs, solvents and diluting agents, and antibacterial medications.
Interpretation: These findings highlight the value of integrating data from different diseases (lymphoma) to improve predictions in a target disease (CLL), and represent a step toward data-driven identification of CLL patients at high risk of infection during treatment.
Downloads
References
Matsukane R, Watanabe H, Minami H, Hata K, Suetsugu K, Tsuji T, et al. Continuous monitoring of neutrophils to lymphocytes ratio for estimating the onset, severity, and subsequent prognosis of immune related adverse events. Sci Rep. 2021;11(1):1–11.
https://doi.org/10.1038/s41598-020-79397-6 DOI: https://doi.org/10.1038/s41598-020-79397-6
Morelli T, Fujita K, Redelman-Sidi G, Elkington PT. Infections due to dysregulated immunity: an emerging complication of cancer immunotherapy. Thorax. 2021;77(3): 304–311.
https://doi.org/10.1136/thoraxjnl-2021-217260 DOI: https://doi.org/10.1136/thoraxjnl-2021-217260
Svanberg R, Janum S, Patten PEM, Ramsay AG, Niemann CU. Targeting the tumor microenvironment in chronic lymphocytic leukemia. Haematologica. 2021;106(9):2312–24.
https://doi.org/10.3324/HAEMATOL.2020.268037 DOI: https://doi.org/10.3324/haematol.2020.268037
Teglgaard RS, Marquart HV, Hartling HJ, Bay JT, Da Cunha-Bang C, Brieghel C, et al. Improved innate immune function in patients with chronic lymphocytic leukemia treated with targeted therapy in clinical trials. Clin Cancer Res. 2024;30(9):1959–71.
https://doi.org/10.1158/1078-0432.CCR-23-2522 DOI: https://doi.org/10.1158/1078-0432.CCR-23-2522
Lipsky A, Lamanna N. Managing toxicities of Bruton tyrosine kinase inhibitors. Hematol Am Soc Hematol Educ Progr. 2020;2020(1):336.
https://doi.org/10.1182/HEMATOLOGY.2020000118 DOI: https://doi.org/10.1182/hematology.2020000118
Vainer N, Aarup K, Andersen MA, Wind-Hansen L, Nielsen T, Frederiksen H, et al. Real-world outcomes upon second-line treatment in patients with chronic lymphocytic leukaemia. Br J Haematol. 2023;201(5):874–886.
https://doi.org/10.1111/bjh.18715 DOI: https://doi.org/10.1111/bjh.18715
Ruppert AS, Booth AM, Ding W, Bartlett NL, Brander DM, Coutre S, et al. Adverse event burden in older patients with CLL receiving bendamustine plus rituximab or ibrutinib regimens: Alliance A041202. Leukemia. 2021;35(10):2854.
https://doi.org/10.1038/S41375-021-01342-X DOI: https://doi.org/10.1038/s41375-021-01342-x
Goyal RK, Nagar SP, Kabadi SM, Le H, Davis KL, Kaye JA. Overall survival, adverse events, and economic burden in patients with chronic lymphocytic leukemia receiving systemic therapy: real‐world evidence from the medicare population. Cancer Med. 2021;10(8):2690.
https://doi.org/10.1002/CAM4.3855 DOI: https://doi.org/10.1002/cam4.3855
Launay CP, Lubov J, Galery K, Vilcocq C, Maubert É, Afilalo M, et al. Prognosis tools for short-term adverse events in older emergency department users: result of a Québec observational prospective cohort. BMC Geriatr. 2021;21(1):1–10.
https://doi.org/10.1186/s12877-020-01999-6 DOI: https://doi.org/10.1186/s12877-020-01999-6
Gu Y, Li Q, Lin R, Jiang W, Wang X, Zhou G, et al. Prognostic model to predict postoperative adverse events in pediatric patients with aortic coarctation. Front Cardiovasc Med. 2021;8:1–12.
https://doi.org/10.3389/fcvm.2021.672627 DOI: https://doi.org/10.3389/fcvm.2021.672627
Buckley SA, Othus M, Vainstein V, Abkowitz JL, Estey EH, Walter RB. Prediction of adverse events during intensive induction chemotherapy for acute myeloid leukemia or high-grade myelodysplastic syndromes. Am J Hematol. 2014;89(4):423–8.
https://doi.org/10.1002/ajh.23661 DOI: https://doi.org/10.1002/ajh.23661
Parviz M, Agius R, Rotbain EC, Vainer N, Aarup K, Niemann CU. Identifying CLL patients at high risk of atrial fibrillation on treatment using machine learning. Leuk Lymphoma. 2024;65(4):449–59.
https://doi.org/10.1080/10428194.2023.2299737 DOI: https://doi.org/10.1080/10428194.2023.2299737
Kim JW, Lee YG, Hwang IG, Song HS, Koh SJ, Ko YH, et al. Predicting cumulative incidence of adverse events in older patients with cancer undergoing first-line palliative chemotherapy: Korean Cancer Study Group (KCSG) multicentre prospective study. Br J Cancer. 2018;118(9):1169–75.
https://doi.org/10.1038/s41416-018-0037-6 DOI: https://doi.org/10.1038/s41416-018-0037-6
Sakakibara T, Shindo Y, Kobayashi D, Sano M, Okumura J, Murakami Y, et al. A prediction rule for severe adverse events in all inpatients with community-acquired pneumonia: a multicenter observational study. BMC Pulm Med. 2022;22(1):1–13.
https://doi.org/10.1186/s12890-022-01819-0 DOI: https://doi.org/10.1186/s12890-022-01819-0
Parviz M, Brieghel C, Agius R, Niemann CU. Prediction of clinical outcome in CLL based on recurrent gene mutations, CLL-IPI variables, and (para)clinical data. Blood Adv. 2022;6(12):3716–28.
https://doi.org/10.1182/bloodadvances.2021006351 DOI: https://doi.org/10.1182/bloodadvances.2021006351
Agius R, Brieghel C, Andersen MA, Pearson AT, Ledergerber B, Cozzi-Lepri A, et al. Machine learning can identify newly diagnosed patients with CLL at high risk of infection. Nat Commun. 2020;11(363):1–16.
https://doi.org/10.1038/s41467-019-14225-8 DOI: https://doi.org/10.1038/s41467-019-14225-8
Agius R, Riis-Jensen AC, Wimmer B, da Cunha-Bang C, Murray DD, Poulsen CB, et al. Deployment and validation of the CLL treatment infection model adjoined to an EHR system. NPJ Digit Med. 2024;7(1):1–12.
https://doi.org/10.1038/s41746-024-01132-6 DOI: https://doi.org/10.1038/s41746-024-01132-6
Leusder M, Porte P, Ahaus K, van Elten H. Original research: cost measurement in value-based healthcare: a systematic review. BMJ Open. 2022;12(12):e066568.
https://doi.org/10.1136/BMJOPEN-2022-066568 DOI: https://doi.org/10.1136/bmjopen-2022-066568
Schwarze K, Buchanan J, Fermont JM, Dreau H, Tilley MW, Taylor JM, et al. The complete costs of genome sequencing: a microcosting study in cancer and rare diseases from a single center in the United Kingdom. Genet Med. 2020;22(1):85–94.
https://doi.org/10.1038/S41436-019-0618-7 DOI: https://doi.org/10.1038/s41436-019-0618-7
Daumé H, Marcu D. Domain adaptation for statistical classifiers. J Artif Intell Res. 2006;26:101–26.
https://doi.org/10.1613/jair.1872 DOI: https://doi.org/10.1613/jair.1872
Agius R, Parviz M, Niemann CU. Artificial intelligence models in chronic lymphocytic leukemia – recommendations toward state-of-the-art. Leuk Lymphoma. 2022;63(2):265–78.
https://doi.org/10.1080/10428194.2021.1973672 DOI: https://doi.org/10.1080/10428194.2021.1973672
Bungărdean RM, Şerbănescu MS, Streba CT, Crişan M. Deep learning with transfer learning in pathology. Case study: classification of basal cell carcinoma. Rom J Morphol Embryol. 2021;62(4):1017–28.
https://doi.org/10.47162/RJME.62.4.14 DOI: https://doi.org/10.47162/RJME.62.4.14
Bjerregaard-Michelsen S, Poulsen LØ, Bjerrum A, Bøgsted M, Vesteghem C. Machine learning for prediction of 30-day mortality in patients with advanced cancer: comparing pan-cancer and single-cancer models. ESMO Real World Data Digit Oncol. 2025;8:100146.
https://doi.org/10.1016/J.ESMORW.2025.100146 DOI: https://doi.org/10.1016/j.esmorw.2025.100146
Wu P, Dietterich TG. Improving SVM accuracy by training on auxiliary data sources. Proceedings, Twenty-First Int. Conf. Mach. Learn. ICML. 2004;871–8.
https://doi.org/10.1145/1015330.1015436 DOI: https://doi.org/10.1145/1015330.1015436
Lundberg SM, Lee SI. A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems, 2017, pp. 4766–75. [Cited 8th of July 2024]. Available from: https://github.com/slundberg/shap
Da Cunha-Bang C, Geisler CH, Enggaard L, Poulsen CB, De Nully Brown P, Frederiksen H, et al. The Danish National Chronic Lymphocytic Leukemia Registry. Clin Epidemiol. 2016;8:561–5.
https://doi.org/10.2147/CLEP.S99486 DOI: https://doi.org/10.2147/CLEP.S99486
Arboe B, Josefsson P, Jørgensen J, Haaber J, Jensen P, Poulsen C, et al. Danish National Lymphoma Registry. Clin Epidemiol. 2016;8:577–81.
https://doi.org/10.2147/CLEP.S99470 DOI: https://doi.org/10.2147/CLEP.S99470
Brieghel C, Werling M, Frederiksen CM, Parviz M, Lacoppidan T, Faitova T, et al. The Danish Lymphoid Cancer Research (DALY-CARE) Data Resource: The Basis for Developing Data-Driven Hematology. Clin Epidemiol. 2025;17:131–145. DOI: https://doi.org/10.2147/CLEP.S479672
https://doi.org/10Bernstorff M, Enevoldsen K, Damgaard J, Danielsen A, Hansen L. timeseriesflattener: a Python package for summarizing features from (medical) time series. J Open Source Softw. 2023;8(83):5197.
https://doi.org/10.21105/JOSS.05197 DOI: https://doi.org/10.21105/joss.05197
Andersen MA, Moser CE, Lundgren J, Niemann CU. Epidemiology of bloodstream infections in patients with chronic lymphocytic leukemia: a longitudinal nation-wide cohort study. Leukemia. 2018;33(3):662–70.
https://doi.org/10.1038/s41375-018-0316-5 DOI: https://doi.org/10.1038/s41375-018-0316-5
Sørrig R, Klausen TW, Salomo M, Vangsted A, Gimsing P. Risk factors for blood stream infections in multiple myeloma: a population-based study of 1154 patients in Denmark. Eur J Haematol. 2018;101(1):21–7.
https://doi.org/10.1111/EJH.13066 DOI: https://doi.org/10.1111/ejh.13066
International T. An international prognostic index for patients with chronic lymphocytic leukaemia (CLL-IPI): a meta-analysis of individual patient data. Lancet Oncol. 2016;17(6):779–90.
https://doi.org/10.1016/S1470-2045(16)30029-8 DOI: https://doi.org/10.1016/S1470-2045(16)30029-8
de Mathelin A, Atiq M, Richard G, de la Concha A, Yachouti M, Deheeger F, et al. ADAPT : Awesome Domain Adaptation Python Toolbox. 2021 [cited 2024 Jun 05]. Available from: https://arxiv.org/abs/2107.03049v2
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12(85):2825–30. [Cited 8th of July 2024] Available from: http://jmlr.org/papers/v12/pedregosa11a.html
Pölsterl S. Scikit-survival: a library for time-to-event analysis built on top of scikit-learn. J Mach Learn Res. 2020;21:1–6.
https://doi.org/10.1007/978-1-4842-5373-1_1 DOI: https://doi.org/10.1007/978-1-4842-5373-1_1
Davidson-Pilon C. Lifelines: survival analysis in Python. J Open Source Softw. 2019;4(40):1317.
https://doi.org/10.21105/joss.01317 DOI: https://doi.org/10.21105/joss.01317
Chicco D, Jurman G. The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genomics. 2020;21(1):1–13. DOI: https://doi.org/10.1186/s12864-019-6413-7
https://doi.org/10.1186/S12864-019-6413-7/TABLES/5
Therneau TM, Grambsch PM. A Package for Survival Analysis in S, version 2.38. Model Surviv. Data Extending Cox Model; 2000 [cited 2024 Jul 08]. Available from: http://cran.r-project.org/package=survival
Eriksson F, Li J, Scheike T, Zhang MJ. The proportional odds cumulative incidence model for competing risks. Biometrics. 2015;71(3):687.
https://doi.org/10.1111/BIOM.12330 DOI: https://doi.org/10.1111/biom.12330
Guan H, Liu M. Domain adaptation for medical image analysis: a survey. IEEE Trans Biomed Eng. 2022;69(3):1173–85.
https://doi.org/10.1109/TBME.2021.3117407 DOI: https://doi.org/10.1109/TBME.2021.3117407
Laparra E, Mascio A, Velupillai S, Miller T. A review of recent work in transfer learning and domain adaptation for natural language processing of electronic health records. Yearb Med Inform. 2021;30(1):239.
https://doi.org/10.1055/S-0041-1726522 DOI: https://doi.org/10.1055/s-0041-1726522
Crassini KR, Best OG, Mulligan SP. Immune failure, infection and survival in chronic lymphocytic leukemia. Haematologica. 2018;103(7):e329.
https://doi.org/10.3324/HAEMATOL.2018.196543 DOI: https://doi.org/10.3324/haematol.2018.196543
Andersen MA, Niemann CU. Immune failure, infection and survival in chronic lymphocytic leukemia in Denmark. Haematologica. 2018;103(7):e330.
https://doi.org/10.3324/HAEMATOL.2018.197889 DOI: https://doi.org/10.3324/haematol.2018.197889
Syed-Ahmed M, Narayanan M. Immune dysfunction and risk of infection in chronic kidney disease. Adv Chronic Kidney Dis. 2019;26(1):8–15.
https://doi.org/10.1053/J.ACKD.2019.01.004 DOI: https://doi.org/10.1053/j.ackd.2019.01.004
Eichhorst B, Ghia P, Niemann CU, Kater AP, Gregor M, Hallek M, et al. ESMO Clinical Practice Guideline interim update on new targeted therapies in the first-line and at relapse of chronic lymphocytic leukaemia. Ann Oncol Off J Eur Soc Med Oncol. 2024:35(9):762–768.
https://doi.org/10.1016/j.annonc.2024.06.016 DOI: https://doi.org/10.1016/j.annonc.2024.06.016
Rotbain EC, Niemann CU, Rostgaard K, da Cunha-Bang C, Hjalgrim H, Frederiksen H. Mapping comorbidity in chronic lymphocytic leukemia: impact of individual comorbidities on treatment, mortality, and causes of death. Leukemia. 2021;35(9):2570–80.
https://doi.org/10.1038/S41375-021-01156-X DOI: https://doi.org/10.1038/s41375-021-01156-x
Gordon MJ, Huang J, Chan RJ, Bhargava P, Danilov AV, Medical comorbidities in patients with chronic lymphocytic leukaemia treated with idelalisib: analysis of two large randomised clinical trials. Br J Haematol. 2021;192(4):720–8.
https://doi.org/10.1111/BJH.16879 DOI: https://doi.org/10.1111/bjh.16879
Levinson T, Wasserman A. C-Reactive Protein Velocity (CRPv) as a new biomarker for the early detection of acute infection/inflammation. Int J Mol Sci. 2022;23(15):1–10. DOI: https://doi.org/10.3390/ijms23158100
Wiedermann CJ. Hypoalbuminemia as surrogate and culprit of infections. Int J Mol Sci. 2021;22(9):1–25.
https://doi.org/10.3390/IJMS22094496 DOI: https://doi.org/10.3390/ijms22094496
Additional Files
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Mehdi Parviz, Christian Brieghel, Mikkel Werling, Thomas Lacoppidan, Emelie Rotbain, Carsten U. Niemann, Rudi Agius
This work is licensed under a Creative Commons Attribution 4.0 International License.
