Comparative analysis of machine learning models for predicting anthracycline-induced cardiotoxicity in patients with hematologic malignancies

Objective: a comparative study of various machine learning algorithms (AdaBoost, k-nearest neighbors, linear discriminant analysis, logistic regression, neural networks, random forest, stochastic gradient descent, support vector machines, XGBoost) for predicting anthracycline-induced cardiotoxicity (AIC) in patients with haematological cancer using clinical and instrumental predictors.

Material and Methods. A prospective study included 155 haematological cancer patients receiving anthracycline-containing therapy. The age of the patients ranged from 18 to 74 years. Clinical data, biomarker levels (NT-proBNP, troponin I), and echocardiographic parameters of diastolic function (E’, E/E’, LAVI) were analyzed. Data underwent preprocessing (standardization, one-hot encoding), and class imbalance was addressed using SMOTETomek. Models were trained and evaluated via 5-fold stratified cross-validation using F1-score, AUC-ROC, precision, and recall metrics.

Results. Statistically significant predictors of AIC included NT-proBNP (p<0.001), troponin I (p=0.004), and echocardiographic parameters E’ (p<0.001) and LAVI (p<0.001). Incorporating age and E/E’ ratio further enhanced the model predictive value. Logistic regression demonstrated optimal performance (F1 0.943 ± 0.070, AUC-ROC 0.963 ± 0.051) with perfect precision (1.00 ± 0.00) and high recall (0.90 ± 0.12). Linear discriminant analysis yielded comparable results (F1 0.921 ± 0.066, AUC-ROC 0.963 ± 0.046). Linear models outperformed more complex algorithms (neural networks, ensemble methods).

Conclusion. Linear models, particularly logistic regression, exhibit high accuracy and reliability in predicting AIC using combined biomarkers and echocardiographic diastolic function parameters. These models show potential for clinical implementation in risk stratification and timely initiation of cardioprotective therapy. Further validation across multi-center patient cohorts is warranted.

1. Vasyuk Y.A., Shkolnik E.L., Nesvetov V.V., Shkolnik L.D., Varlan G.V. Cardiooncology: Current Aspects of Prevention of Anthracycline Toxicity. Cardiology. 2016; 56(12): 72–79. (in Russian). doi: 10.18565/cardio.2016.12.72-79. EDN: XIMODJ.

2. Morgillo A., Mazzarella M., Randazzo M.F., Mancino N., Talessi M., Marovino E. Anthracycline Cardiotoxicity: From Mechanisms to Prevention Strategies. J Med – Clin Res Rev. 2023; 7(1): 1–5. doi: 10.33425/2639944X.1309.

3. Valcovici M., Andrica F., Serban C., Dragan S. Cardiotoxicity of anthracycline therapy: current perspectives. Arch Med Sci. 2016; 12(2): 428–35. doi: 10.5114/aoms.2016.59270.

4. Narezkina A., Narayan H.K., Zemljic-Harpf A.E. Molecular mechanisms of anthracycline cardiovascular toxicity. Clin Sci (Lond). 2021; 135(10): 1311–32. doi: 10.1042/CS20200301.

5. Agunbiade T.A., Zaghlol R.Y., Barac A. Heart Failure in Relation to Anthracyclines and Other Chemotherapies. Methodist DeBakey Cardiovasc J. 2019; 15(4): 243–49. doi: 10.14797/mdcj-15-4-243.

6. Sobczuk P., Czerwińska M., Kleibert M., Cudnoch-Jędrzejewska A. Anthracycline-induced cardiotoxicity and renin-angiotensin-aldosterone system-from molecular mechanisms to therapeutic applications. Heart Fail Rev. 2022; 27(1): 295–319. doi: 10.1007/s10741-020-09977-1.

7. Idrees I., Ahmad R., Nadeem M., Khan A., Gilani A., Khattak R. Antracycline Induced Early Onset Chronic Cardiotoxicity in Cancer Patients. Pak Armed Forces Med J. 2024; 74(6): 1492–95. doi: 10.51253/pafmj.v74i2.6795.

8. Tariq M., Ibrahim M., Khan S.R., Baloch S.S., Shahzadi M., Rashid Y.A., Zahid M.J., Khan M.D. Frequency of Cardiac-Toxicity in Patient Treated with Anthracycline, An Institutional Perspective. Pak J Med Health Sci. 2022; 16(4): 488. doi: 10.53350/pjmhs22164488.

9. Lotrionte M., Biondi-Zoccai G., Abbate A., Lanzetta G., D’Ascenzo F., Malavasi V., Peruzzi M., Frati G., Palazzoni G. Review and meta-analysis of incidence and clinical predictors of anthracycline cardiotoxicity. Am J Cardiol. 2013; 112(12): 1980–84. doi: 10.1016/j.amjcard.2013.08.026.

10. Sawaya H., Sebag I.A., Plana J.C., Januzzi J.L., Ky B., Tan T.C., Cohen V., Banchs J., Carver J.R., Wiegers S.E., Martin R.P., Picard M.H., Gerszten R.E., Halpern E.F., Passeri J., Kuter I., Scherrer-Crosbie M. Assessment of echocardiography and biomarkers for the extended prediction of cardiotoxicity in patients treated with anthracyclines, taxanes, and trastuzumab. Circ Cardiovasc Imaging. 2012; 5(5): 596–603. doi: 10.1161/CIRCIMAGING.112.973321.

11. Bhagat A.A., Kalogeropoulos A.P., Baer L., Lacey M., Kort S., Skopicki H.A., Butler J., Bloom M.W. Biomarkers and Strain Echocardiography for the Detection of Subclinical Cardiotoxicity in Breast Cancer Patients Receiving Anthracyclines. J Pers Med. 2023; 13(12): 1710. doi: 10.3390/jpm13121710.

12. Gulati G., Zhang K.W., Scherrer-Crosbie M., Ky B. Cancer and cardiovascular disease: the use of novel echocardiography measures to predict subsequent cardiotoxicity in breast cancer treated with anthracyclines and trastuzumab. Curr Heart Fail Rep. 2014; 11(4): 366–73. doi: 10.1007/s11897-014-0214-8.

13. Zhang L., Zhang R., Shuai P., Chen J., Yin L. A global case metaanalysis of three-dimensional speckle tracking for evaluating the cardiotoxicity of anthracycline chemotherapy in breast cancer. Front Cardiovasc Med. 2022; 9: 942620. doi: 10.3389/fcvm.2022.942620.

14. Wei X., Lin L., Zhang G., Zhou X. Cardiovascular Magnetic Resonance Imaging in the Early Detection of Cardiotoxicity Induced by Cancer Therapies. Diagnostics (Basel). 2022; 12(8): 1846. doi: 10.3390/diagnostics12081846.

15. Musella F., Librera M., Sibilio G., Boccalatte M., Tagliamonte G., Cavaglià E., Ferrara I., Puglia M., Dell’Aversana S., Ducci C.B., Dellegrottaglie S., Savarese G., Scatteia A. Cardiovascular magnetic resonance parametric techniques to characterize myocardial effects of anthracycline therapy in adults with normal left ventricular ejection fraction: a systematic review and meta-analysis. Curr Probl Cardiol. 2024; 49(7): 102609. doi: 10.1016/j.cpcardiol.2024.102609.

16. Stevens P.L., Lenihan D.J. Cardiotoxicity due to Chemotherapy: the Role of Biomarkers. Curr Cardiol Rep. 2015; 17(7): 603. doi: 10.1007/s11886-015-0603-y.

17. Díaz-Antón B., Madurga R., Zorita B., Wasniewski S., MorenoArciniegas A., López-Melgar B., Ramírez Merino N., Martín-Asenjo R., Barrio P., Amado Escañuela M.G., Solís J., Parra Jiménez F.J., Ciruelos E., Castellano J.M., Fernández-Friera L. Early detection of anthracyclineand trastuzumab-induced cardiotoxicity: value and optimal timing of serum biomarkers and echocardiographic parameters. ESC Heart Fail. 2022; 9(2): 1127–37. doi: 10.1002/ehf2.13782.

18. Stolojanu C., Steflea R., Micsescu-Olah A.M., Alexandra I., Popoiu A., Doros G. Combined Utility of Speckle Tracking Echocardiography and Cardiac Biomarkers for Early Detection of Anthracycline-Induced Cardiotoxicity in Pediatric Oncology Patients. Biomedicines. 2024; 12(12): 2849. doi: 10.3390/biomedicines12122849.

19. Mahjoob M.P., Sheikholeslami S.A., Dadras M., Mansouri H., Haghi M., Naderian M., Sadeghi L., Tabary M., Khaheshi I. Prognostic Value of Cardiac Biomarkers Assessment in Combination with Myocardial 2D Strain Echocardiography for Early Detection of Anthracycline-Related Cardiac Toxicity. Cardiovasc Hematol Disord Drug Targets. 2020; 20(1): 74–83. doi: 10.2174/1871529X19666190912150942.

20. Horacek J.M., Vasatova M., Pudil R., Tichy M., Zak P., Jakl M., Jebavy L., Maly J. Biomarkers for the early detection of anthracycline-induced cardiotoxicity: current status. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub. 2014; 158(4): 511–17. doi: 10.5507/bp.2014.004.

21. Feng W., Wang Q., Tan Y., Qiao J., Liu Q., Yang B., Yang S., Cui L. Early detection of anthracycline-induced cardiotoxicity. Clin Chim Acta. 2025; 565: 120000. doi: 10.1016/j.cca.2024.120000.

22. Murtagh G., Januzzi J.L., Scherrer-Crosbie M., Neilan T.G., Dent S., Ho J.E., Appadurai V., McDermott R., Akhter N. Circulating Cardiovascular Biomarkers in Cancer Therapeutics-Related Cardiotoxicity: Review of Critical Challenges, Solutions, and Future Directions. J Am Heart Assoc. 2023; 12(21): e029574. doi: 10.1161/JAHA.123.029574.

23. Liu Z., Wang Y., Huang X., Qi X., Qian C., Zhao S. Development and Validation of a Diagnostic Nomogram to Predict the AnthracyclineInduced Early Cardiotoxicity in Children with Hematological Tumors. Cardiovasc Toxicol. 2022; 22(9): 802–12. doi: 10.1007/s12012-02209755-5.

24. Gómez-Vecino A., Corchado-Cobos R., Blanco-Gómez A., GarcíaSancha N., Castillo-Lluva S., Martín-García A., Mendiburu-Eliçabe M., Prieto C., Ruiz-Pinto S., Pita G., Velasco-Ruiz A., Patino-Alonso C., Galindo-Villardón P., Vera-Pedrosa M.L., Jalife J., Mao J.-H., Macías de Plasencia G., Castellanos-Martín A., Sáez-Freire M.D.M., FraileMartín S., Rodrigues-Teixeira T., García-Macías C., Galvis-Jiménez J.M., García-Sánchez A., Isidoro-García M., Fuentes M., García-Cenador M.B., García-Criado F.J., García-Hernández J.L., Hernández-García M.Á., Cruz-Hernández J.J., Rodríguez-Sánchez C.A., García-Sancho A.M., Pérez-López E., Pérez-Martínez A., Gutiérrez-Larraya F., Cartón A.J., García-Sáenz J.Á., Patiño-García A., Martín M., Alonso-Gordoa T., Vulsteke C., Croes L., Hatse S., Van Brussel T., Lambrechts D., Wildiers H., Hang C., Holgado-Madruga M., González-Neira A., Sánchez P.L., Pérez Losada J. Intermediate Molecular Phenotypes to Identify Genetic Markers of Anthracycline-Induced Cardiotoxicity Risk. Cells. 2023; 12(15): 1956. doi: 10.3390/cells12151956.

25. Sharafeldin N., Zhou L., Singh P., Crossman D.K., Wang X., Hageman L., Landier W., Blanco J.G., Burridge P.W., Sapkota Y., Yasui Y., Armstrong G.T., Robison L.L., Hudson M.M., Oeffinger K., Chow E.J., Armenian S.H., Weisdorf D.J., Bhatia S. Gene-Level Analysis of Anthracycline-Induced Cardiomyopathy in Cancer Survivors: A Report From COG-ALTE03N1, BMTSS, and CCSS. JACC CardioOncol. 2023; 5(6): 807–18. doi: 10.1016/j.jaccao.2023.06.007.

26. Jacquemyn X., Chinni B.K., Barnes B.T., Rao S., Kutty S., Manlhiot C. Unsupervised machine learning identifies distinct phenotypes in cardiac complications of pediatric patients treated with anthracyclines. Cardiooncology. 2024; 10(1): 74. doi: 10.1186/s40959-024-00276-4.

27. Yagi R., Goto S., Himeno Y., Katsumata Y., Hashimoto M., MacRae C.A., Deo R.C. Artificial intelligence-enabled prediction of chemotherapy-induced cardiotoxicity from baseline electrocardiograms. Nat Commun. 2024; 15(1): 2536. doi: 10.1038/s41467-024-45733-x.

28. Zhou Y., Hou Y., Hussain M., Brown S., Budd T., Tang W.H.W., Abraham J., Xu B., Shah C., Moudgil R., Popovic Z., Cho L., Kanj M., Watson C., Griffin B., Chung M.K., Kapadia S., Svensson L., Collier P., Cheng F. Machine Learning-Based Risk Assessment for Cancer TherapyRelated Cardiac Dysfunction in 4300 Longitudinal Oncology Patients. J Am Heart Assoc. 2020; 9(23): e019628. doi: 10.1161/JAHA.120.019628.

29. Artico J., Abiodun A., Shiwani H., Kurdi H., Chen D., Tyebally S., Moon J.C., Westwood M., Manisty C.H. Multimodality Imaging for Cardiotoxicity: State of the Art and Future Perspectives. J Cardiovasc Pharmacol. 2022; 80(4): 547–61. doi: 10.1097/FJC.0000000000001281.

30. Taruna R.N., Wihandono A., Budiarto R.M. Cardiac Function in Breast Cancer Patients Undergoing Anthracycline Chemotherapy: A Comprehensive Review of Mechanisms, Monitoring, and Management Strategies. Int J Sci Adv. 2024; 5(6): 1565–72. doi: 10.51542/ijscia.v5i6.85.

31. Chang W.T., Liu C.F., Feng Y.H., Liao C.T., Wang J.J., Chen Z.C., Lee H.C., Shih J.Y. An artificial intelligence approach for predicting cardiotoxicity in breast cancer patients receiving anthracycline. Arch Toxicol. 2022; 96: 2731–37. doi: 10.1007/s00204-022-03341-y.

32. Zhou X., Weng Y., Jiang T., Ou W., Zhang N., Dong Q., Tang X. Influencing factors of anthracycline-induced subclinical cardiotoxicity in acute leukemia patients. BMC Cancer. 2023; 23(1): 976. doi: 10.1186/s12885-023-11060-5.