{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T09:25:28Z","timestamp":1783416328672,"version":"3.54.6"},"reference-count":41,"publisher":"MDPI AG","issue":"13","license":[{"start":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T00:00:00Z","timestamp":1783382400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background: Thymic epithelial tumors (TETs) are the most common primary neoplasms of the anterior mediastinum and present a dual classification challenge, namely anatomical staging according to the Masaoka\u2013Koga system and histological risk stratification according to the World Health Organization (WHO) classification. Both tasks rely on expert pathological assessment and may be affected by interobserver variability. This study applied supervised machine learning (ML) to quantitative immunohistochemical (IHC) H-score profiles to predict Masaoka\u2013Koga stage and WHO risk group in TETs. Methods: Logistic regression (LR) and XGBoost were applied to 19 biomarkers, including cellular localization, across two parallel analyses. Masaoka\u2013Koga stage prediction was performed in 81 patients, including 59 early-stage and 22 advanced-stage cases, using the Synthetic Minority Oversampling Technique (SMOTE) across 100 train\/test splits. WHO risk group prediction was performed in 89 patients, including 45 low-risk and 44 high-risk tumors, without oversampling. A cross-endpoint analysis applied the optimal Masaoka\u2013Koga model to the WHO endpoint. Results: LR consistently outperformed XGBoost. The optimal Masaoka\u2013Koga model combined Eph receptor A6 (EphA6) membranous, Yes-associated protein (YAP) nuclear, and histone deacetylase 4 (HDAC4) cytoplasmic H-scores, achieving an area under the curve (AUC) of 0.756. The optimal WHO model combined transcriptional coactivator with PDZ-binding motif (TAZ) cytoplasmic, EphA6 membranous, and YAP nuclear H-scores, achieving an AUC of 0.936. The Masaoka\u2013Koga triad predicted WHO risk group with an AUC of 0.901. No tetrad improved trivariate performance. Conclusions: IHC H-score profiling combined with supervised ML identifies biologically interpretable candidate signatures for TET classification, although prospective external validation is required before clinical application.<\/jats:p>","DOI":"10.3390\/diagnostics16132118","type":"journal-article","created":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T08:04:49Z","timestamp":1783411489000},"page":"2118","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Machine Learning-Based Prediction of Masaoka\u2013Koga Stage and WHO Histological Risk Group in Thymic Epithelial Tumors Using Biomarker Combinations"],"prefix":"10.3390","volume":"16","author":[{"given":"Konstantinos","family":"Kitrou","sequence":"first","affiliation":[{"name":"Department of Business Administration, University of Patras, 26504 Rio, Achaia, Greece"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2974-4084","authenticated-orcid":false,"given":"Georgios","family":"Mandrakis","sequence":"additional","affiliation":[{"name":"First Department of Pathology, School of Medicine, National and Kapodistrian University of Athens, 75 Mikras Asias Street, 11527 Athens, Greece"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Georgios","family":"Tsirogiannis","sequence":"additional","affiliation":[{"name":"Department of Food Science and Technology, University of Patras, 26504 Rio, Achaia, Greece"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stamatios","family":"Theocharis","sequence":"additional","affiliation":[{"name":"First Department of Pathology, School of Medicine, National and Kapodistrian University of Athens, 75 Mikras Asias Street, 11527 Athens, Greece"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7924-5075","authenticated-orcid":false,"given":"Constantinos","family":"Halkiopoulos","sequence":"additional","affiliation":[{"name":"Department of Management Science and Technology, University of Patras, 26504 Patras, Greece"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8925-9427","authenticated-orcid":false,"given":"Yannis","family":"Stamatiou","sequence":"additional","affiliation":[{"name":"Department of Business Administration, University of Patras, 26504 Rio, Achaia, Greece"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"7531","DOI":"10.21037\/jtd-2019-thym-02","article-title":"Epidemiology of thymoma","volume":"12","author":"Rich","year":"2020","journal-title":"J. Thorac. Dis."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"200","DOI":"10.1016\/j.jtho.2021.10.010","article-title":"The 2021 WHO Classification of Tumors of the Thymus and Mediastinum: What Is New in Thymic Epithelial, Germ Cell, and Mesenchymal Tumors?","volume":"17","author":"Marx","year":"2022","journal-title":"J. Thorac. Oncol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"3242","DOI":"10.1111\/1759-7714.14725","article-title":"Prognostic factors and genetic markers in thymic epithelial tumors: A narrative review","volume":"13","author":"Agrafiotis","year":"2022","journal-title":"Thorac. Cancer"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"2485","DOI":"10.1002\/1097-0142(19811201)48:11<2485::AID-CNCR2820481123>3.0.CO;2-R","article-title":"Follow-up study of thymomas with special reference to their clinical stages","volume":"48","author":"Masaoka","year":"1981","journal-title":"Cancer"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"420","DOI":"10.1002\/cncr.10665","article-title":"New WHO histologic classification predicts prognosis of thymic epithelial tumors: A clinicopathologic study of 200 thymoma cases from China","volume":"95","author":"Chen","year":"2002","journal-title":"Cancer"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1501","DOI":"10.1200\/JCO.2004.10.113","article-title":"Tumor recurrence and survival in patients treated for thymomas and thymic squamous cell carcinomas: A retrospective analysis","volume":"22","author":"Strobel","year":"2004","journal-title":"J. Clin. Oncol."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"734","DOI":"10.1111\/his.14167","article-title":"Interobserver variation in the classification of thymic lesions including biopsies and resection specimens in an international digital microscopy panel","volume":"77","author":"Wolf","year":"2020","journal-title":"Histopathology"},{"key":"ref_8","first-page":"716","article-title":"Estrogen receptor analyses. Correlation of biochemical and immunohistochemical methods using monoclonal antireceptor antibodies","volume":"109","author":"McCarty","year":"1985","journal-title":"Arch. Pathol. Lab. Med."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Palamaris, K., Levidou, G., Kordali, K., Masaoutis, C., Rontogianni, D., and Theocharis, S. (2023). Searching for Novel Biomarkers in Thymic Epithelial Tumors: Immunohistochemical Evaluation of Hippo Pathway Components in a Cohort of Thymic Epithelial Tumors. Biomedicines, 11.","DOI":"10.3390\/biomedicines11071876"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Palamaris, K., Tzimou, L.M., Levidou, G., Masaoutis, C., Theochari, I., Rontogianni, D., and Theocharis, S. (2023). Histone Deacetylases (HDACs): Promising Biomarkers and Potential Therapeutic Targets in Thymic Epithelial Tumors. Int. J. Mol. Sci., 24.","DOI":"10.3390\/ijms24054263"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Masaoutis, C., Georgantzoglou, N., Sarantis, P., Theochari, I., Tsoukalas, N., Bobos, M., Alexandrou, P., Pergaris, A., Rontogianni, D., and Theocharis, S. (2021). Ephrin Receptors (Ephs) Expression in Thymic Epithelial Tumors: Prognostic Implications and Future Therapeutic Approaches. Diagnostics, 11.","DOI":"10.3390\/diagnostics11122265"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Stergiou, I.E., Palamaris, K., Levidou, G., Tzimou, M., Papadakos, S.P., Mandrakis, G., Masaoutis, C., Rontogianni, D., and Theocharis, S. (2024). PD-L1 Expression in Neoplastic and Immune Cells of Thymic Epithelial Tumors: Correlations with Disease Characteristics and HDAC Expression. Biomedicines, 12.","DOI":"10.3390\/biomedicines12040772"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1186\/s12957-021-02259-6","article-title":"CT imaging-based machine learning model: A potential modality for predicting low-risk and high-risk groups of thymoma: \u201cImpact of surgical modality choice\u201d","volume":"19","author":"Orhan","year":"2021","journal-title":"World J. Surg. Oncol."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"253","DOI":"10.1186\/s13244-024-01798-2","article-title":"Risk stratification of thymic epithelial tumors based on peritumor CT radiomics and semantic features","volume":"15","author":"Zhang","year":"2024","journal-title":"Insights Imaging"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"287","DOI":"10.21037\/atm.2020.02.183","article-title":"Development of a deep learning model for classifying thymoma as Masaoka-Koga stage I or II via preoperative CT images","volume":"8","author":"Yang","year":"2020","journal-title":"Ann. Transl. Med."},{"key":"ref_16","first-page":"2825","article-title":"Scikit-learn: Machine Learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016, January 13\u201317). \u201cXGBoost\u201d. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1613\/jair.953","article-title":"SMOTE: Synthetic Minority Over-sampling Technique","volume":"16","author":"Chawla","year":"2002","journal-title":"J. Artif. Intell. Res."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Yang, Y., and Mirzaei, G. (2024). Performance analysis of data resampling on class imbalance and classification techniques on multi-omics data for cancer classification. PLoS ONE, 19.","DOI":"10.1371\/journal.pone.0293607"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Steyerberg, E.W., Moons, K.G., van der Windt, D.A., Hayden, J.A., Perel, P., Schroter, S., Riley, R.D., Hemingway, H., Altman, D.G., and Group, P. (2013). Prognosis Research Strategy (PROGRESS) 3: Prognostic model research. PLoS Med., 10.","DOI":"10.1371\/journal.pmed.1001381"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"1373","DOI":"10.1016\/S0895-4356(96)00236-3","article-title":"A simulation study of the number of events per variable in logistic regression analysis","volume":"49","author":"Peduzzi","year":"1996","journal-title":"J. Clin. Epidemiol."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1016\/j.cell.2008.03.011","article-title":"Eph-ephrin bidirectional signaling in physiology and disease","volume":"133","author":"Pasquale","year":"2008","journal-title":"Cell"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"165","DOI":"10.1038\/nrc2806","article-title":"Eph receptors and ephrins in cancer: Bidirectional signalling and beyond","volume":"10","author":"Pasquale","year":"2010","journal-title":"Nat. Rev. Cancer"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"22587","DOI":"10.18632\/oncotarget.4088","article-title":"EphA6 promotes angiogenesis and prostate cancer metastasis and is associated with human prostate cancer progression","volume":"6","author":"Li","year":"2015","journal-title":"Oncotarget"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Real, S.A.S., Parveen, F., Rehman, A.U., Khan, M.A., Deo, S.V.S., Shukla, N.K., and Husain, S.A. (2018). Aberrant Promoter Methylation of YAP Gene and its Subsequent Downregulation in Indian Breast Cancer Patients. BMC Cancer, 18.","DOI":"10.1186\/s12885-018-4627-8"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1186\/1476-4598-13-110","article-title":"Fbxw7 is an independent prognostic marker and induces apoptosis and growth arrest by regulating YAP abundance in hepatocellular carcinoma","volume":"13","author":"Tu","year":"2014","journal-title":"Mol. Cancer"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"3395","DOI":"10.2147\/CMAR.S194512","article-title":"FRMPD1 activates the Hippo pathway via interaction with WWC3 to suppress the proliferation and invasiveness of lung cancer cells","volume":"11","author":"Rong","year":"2019","journal-title":"Cancer Manag. Res."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1101\/gad.1843810","article-title":"A coordinated phosphorylation by Lats and CK1 regulates YAP stability through SCF(beta-TRCP)","volume":"24","author":"Zhao","year":"2010","journal-title":"Genes Dev."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"7835","DOI":"10.1073\/pnas.140199597","article-title":"Regulation of histone deacetylase 4 and 5 and transcriptional activity by 14-3-3-dependent cellular localization","volume":"97","author":"Grozinger","year":"2000","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"jcs230425","DOI":"10.1242\/jcs.230425","article-title":"YAP\/TAZ functions and their regulation at a glance","volume":"133","author":"Pocaterra","year":"2020","journal-title":"J. Cell Sci."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"783","DOI":"10.1016\/j.ccell.2016.05.005","article-title":"YAP\/TAZ at the Roots of Cancer","volume":"29","author":"Zanconato","year":"2016","journal-title":"Cancer Cell"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"2747","DOI":"10.1101\/gad.1602907","article-title":"Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth control","volume":"21","author":"Zhao","year":"2007","journal-title":"Genes Dev."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"van der Ploeg, T., Austin, P.C., and Steyerberg, E.W. (2014). Modern modelling techniques are data hungry: A simulation study for predicting dichotomous endpoints. BMC Med. Res. Methodol., 14.","DOI":"10.1186\/1471-2288-14-137"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"e2200039","DOI":"10.1200\/CCI.22.00039","article-title":"A Comparison of Logistic Regression Against Machine Learning Algorithms for Gastric Cancer Risk Prediction Within Real-World Clinical Data Streams","volume":"6","author":"Huang","year":"2022","journal-title":"JCO Clin. Cancer Inform."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1186\/s41512-020-00075-2","article-title":"Logistic regression has similar performance to optimised machine learning algorithms in a clinical setting: Application to the discrimination between type 1 and type 2 diabetes in young adults","volume":"4","author":"Lynam","year":"2020","journal-title":"Diagn. Progn. Res."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"2580","DOI":"10.21037\/jtd.2016.08.82","article-title":"Prognostic value of immunohistochemical markers in malignant thymic epithelial tumors","volume":"8","author":"Leisibach","year":"2016","journal-title":"J. Thorac. Dis."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"304","DOI":"10.1074\/mcp.M200008-MCP200","article-title":"Discordant protein and mRNA expression in lung adenocarcinomas","volume":"1","author":"Chen","year":"2002","journal-title":"Mol. Cell Proteom."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"227","DOI":"10.1038\/nrg3185","article-title":"Insights into the regulation of protein abundance from proteomic and transcriptomic analyses","volume":"13","author":"Vogel","year":"2012","journal-title":"Nat. Rev. Genet."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"965","DOI":"10.1002\/(SICI)1097-0258(19970515)16:9<965::AID-SIM509>3.0.CO;2-O","article-title":"A comparison of goodness-of-fit tests for the logistic regression model","volume":"16","author":"Hosmer","year":"1997","journal-title":"Stat. Med."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"475","DOI":"10.1016\/j.jclinepi.2004.06.017","article-title":"Substantial effective sample sizes were required for external validation studies of predictive logistic regression models","volume":"58","author":"Vergouwe","year":"2005","journal-title":"J. Clin. Epidemiol."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"128","DOI":"10.1097\/EDE.0b013e3181c30fb2","article-title":"Assessing the performance of prediction models: A framework for traditional and novel measures","volume":"21","author":"Steyerberg","year":"2010","journal-title":"Epidemiology"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/13\/2118\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T08:55:05Z","timestamp":1783414505000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/13\/2118"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,7]]},"references-count":41,"journal-issue":{"issue":"13","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["diagnostics16132118"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16132118","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,7]]}}}