<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE root>
<article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" article-type="research-article" dtd-version="1.2" xml:lang="en"><front><journal-meta><journal-id journal-id-type="publisher-id">Science and Innovations in Medicine</journal-id><journal-title-group><journal-title xml:lang="en">Science and Innovations in Medicine</journal-title><trans-title-group xml:lang="ru"><trans-title>Наука и инновации в медицине</trans-title></trans-title-group></journal-title-group><issn publication-format="print">2500-1388</issn><issn publication-format="electronic">2618-754X</issn><publisher><publisher-name xml:lang="en">FSBEI of Higher Education SamSMU of Ministry of Health of the Russian Federation</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">686422</article-id><article-id pub-id-type="doi">10.35693/SIM686422</article-id><article-categories><subj-group subj-group-type="toc-heading" xml:lang="en"><subject>Oncology and radiotherapy</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="ru"><subject>Онкология, лучевая терапия</subject></subj-group><subj-group subj-group-type="article-type"><subject>Research Article</subject></subj-group></article-categories><title-group><article-title xml:lang="en">Prediction of recurrence-free survival in patients with renal cell carcinoma and tumor thrombosis of the renal and inferior vena cava of levels I–II using an extended Cox model and machine learning methods</article-title><trans-title-group xml:lang="ru"><trans-title>Прогнозирование безрецидивной выживаемости больных с почечно-клеточным раком и опухолевым тромбозом почечной и нижней полой вены I–II уровней с использованием расширенной модели Кокса и методов машинного обучения</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0003-8365-7672</contrib-id><name-alternatives><name xml:lang="en"><surname>Mirzabekov</surname><given-names>Musabek K.</given-names></name><name xml:lang="ru"><surname>Мирзабеков</surname><given-names>М. К.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>postgraduate student</p></bio><bio xml:lang="ru"><p>аспирант</p></bio><email>musabek.mirzabekoff@yandex.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0001-3077-1776</contrib-id><name-alternatives><name xml:lang="en"><surname>Tikhonskii</surname><given-names>Nikolai D.</given-names></name><name xml:lang="ru"><surname>Тихонский</surname><given-names>Н. Д.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>lecturer at the Department of Physics and Informatics</p></bio><bio xml:lang="ru"><p>преподаватель кафедры физики и информатики</p></bio><email>wirelessm8@mail.ru</email><xref ref-type="aff" rid="aff2"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-0589-7999</contrib-id><name-alternatives><name xml:lang="en"><surname>Shkolnik</surname><given-names>Mikhail I.</given-names></name><name xml:lang="ru"><surname>Школьник</surname><given-names>М. И.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Dr. Sci. (Medicine), chief researcher, Professor</p></bio><bio xml:lang="ru"><p>д-р мед. наук, главный научный сотрудник, профессор</p></bio><email>shkolnik_phd@mail.ru</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-5860-9076</contrib-id><name-alternatives><name xml:lang="en"><surname>Bogomolov</surname><given-names>Oleg A.</given-names></name><name xml:lang="ru"><surname>Богомолов</surname><given-names>О. А.</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>MD, Cand. Sci. (Medicine), senior researcher, Associate professor</p></bio><bio xml:lang="ru"><p>канд. мед. наук, старший научный сотрудник, доцент</p></bio><email>urologbogomolov@gmail.com</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-7894-4779</contrib-id><name-alternatives><name xml:lang="en"><surname>Trukhacheva</surname><given-names>Nina V.</given-names></name><name xml:lang="ru"><surname>Трухачева</surname><given-names>Нина Васильевна</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>Cand. Sci. (Pedagogics), Associate professor of the Department of Physics and Informatics</p></bio><bio xml:lang="ru"><p>канд. пед. наук. доцент кафедры физики и информатики</p></bio><email>tn10@mail.ru</email><xref ref-type="aff" rid="aff2"/></contrib></contrib-group><aff-alternatives id="aff1"><aff><institution xml:lang="en">Russian Scientific Center for Radiology and Surgical Technologies named after Academician A.M. Granov</institution></aff><aff><institution xml:lang="ru">ФГБУ «Российский научный центр радиологии и хирургических технологий имени академика А.М. Гранова» Минздрава России</institution></aff></aff-alternatives><aff-alternatives id="aff2"><aff><institution xml:lang="en">Altai State Medical University</institution></aff><aff><institution xml:lang="ru">ФГБОУ ВО «Алтайский государственный медицинский университет» Минздрава России</institution></aff></aff-alternatives><pub-date date-type="preprint" iso-8601-date="2025-08-12" publication-format="electronic"><day>12</day><month>08</month><year>2025</year></pub-date><pub-date date-type="pub" iso-8601-date="2025-08-24" publication-format="electronic"><day>24</day><month>08</month><year>2025</year></pub-date><volume>10</volume><issue>3</issue><issue-title xml:lang="en"/><issue-title xml:lang="ru"/><fpage>237</fpage><lpage>242</lpage><history><date date-type="received" iso-8601-date="2025-06-30"><day>30</day><month>06</month><year>2025</year></date><date date-type="accepted" iso-8601-date="2025-07-14"><day>14</day><month>07</month><year>2025</year></date></history><permissions><copyright-statement xml:lang="en">Copyright ©; 2025, Mirzabekov M.K., Tikhonskii N.D., Shkolnik M.I., Bogomolov O.A., Trukhacheva N.V.</copyright-statement><copyright-statement xml:lang="ru">Copyright ©; 2025, Мирзабеков М.К., Тихонский Н.Д., Школьник М.И., Богомолов О.А., Трухачева Н.В.</copyright-statement><copyright-year>2025</copyright-year><copyright-holder xml:lang="en">Mirzabekov M.K., Tikhonskii N.D., Shkolnik M.I., Bogomolov O.A., Trukhacheva N.V.</copyright-holder><copyright-holder xml:lang="ru">Мирзабеков М.К., Тихонский Н.Д., Школьник М.И., Богомолов О.А., Трухачева Н.В.</copyright-holder><ali:free_to_read xmlns:ali="http://www.niso.org/schemas/ali/1.0/"/><license><ali:license_ref xmlns:ali="http://www.niso.org/schemas/ali/1.0/">https://creativecommons.org/licenses/by/4.0</ali:license_ref></license></permissions><self-uri xlink:href="https://innoscience.ru/2500-1388/article/view/686422">https://innoscience.ru/2500-1388/article/view/686422</self-uri><abstract xml:lang="en"><p><bold>Aim</bold> – to compare the predictive accuracy of Cox regression and machine learning (ML) methods regarding recurrence-free survival in patients with locally advanced renal cell carcinoma after radical treatment. Additionally, to investigate an extended Cox model in which the risk function is formed using a neural network approximator (DeepSurv).</p> <p><bold>Material and methods.</bold> This study conducted a retrospective analysis of data from patients diagnosed with renal cell carcinoma who underwent radical nephrectomy with thrombectomy from the renal and inferior vena cava between 2007 and 2024 at the Federal State Budgetary Institution “RSC for Radiology and Surgical Technologies named after Academician A.M. Granov”. The study included 100 patients (54 men and 46 women). The median age was 61.5 years (IQR: 59.7–63). Of the total observations, disease progression was recorded in 41 cases, while in the remaining 59 cases, the data were censored. The models were evaluated based on the concordance index (C-index) and interpreted using SHAP analysis.</p> <p><bold>Results.</bold> The DeepSurv neural network model demonstrated higher predictive accuracy on the test dataset compared to the classical Cox model (C-index: 0.8056 vs. 0.7917, respectively). This indicates a superior ability of DeepSurv to rank patients by individual risk of disease progression. Using SHAP analysis, the key predictors contributing most significantly to the prognosis were identified: tumor size, ISUP grade, level of tumor thrombosis, and histological tumor type. The DeepSurv model enabled the capture of complex nonlinear interactions between features, thereby improving both the interpretability and clinical applicability of the results.</p> <p><bold>Conclusion.</bold> The obtained data confirm the feasibility of using machine learning methods for personalized prognosis and optimization of monitoring strategies in patients with RCC.</p></abstract><trans-abstract xml:lang="ru"><p><bold>Цель</bold> – сравнить прогностическую точность регрессии Кокса и методов машинного обучения (ML) в отношении безрецидивной выживаемости пациентов с местно-распространенным почечно-клеточным раком после радикального лечения, а также исследовать расширенную модель Кокса, в которой функция риска формируется с использованием нейросетевого аппроксиматора (DeepSurv).</p> <p><bold>Материал и методы.</bold> В данном исследовании был проведен ретроспективный анализ данных пациентов с диагнозом «почечно-клеточный рак», перенесших радикальную нефрэктомию с тромбэктомией из почечной и нижней полой вены в период с 2007 по 2024 годы в ФГБУ «РНЦРХТ им. акад. А.М. Гранова». В исследование включены 100 пациентов (54 мужчины и 46 женщин). Медианный возраст составил 61,5 года (IQR: 59,7–63). Из общего числа наблюдений в 41 случае было зафиксировано прогрессирование заболевания, в остальных 59 случаях данные были цензурированные. Оценка моделей проводилась на основе индекса конкордации (C-index) и интерпретировалась с использованием SHAP-анализа.</p> <p><bold>Результаты.</bold> Нейросетевая модель DeepSurv продемонстрировала более высокую прогностическую точность на тестовой выборке по сравнению с классической моделью Кокса (C-index: 0,8056 против 0,7917 соответственно). Это свидетельствует о лучшей способности модели DeepSurv ранжировать пациентов по индивидуальному риску прогрессирования. С помощью SHAP-анализа установлены ключевые предикторы, вносящие наибольший вклад в прогноз: размер опухоли, степень злокачественности (ISUP-грейд), уровень опухолевого тромбоза и морфологический тип опухоли. Модель DeepSurv позволила учесть сложные нелинейные взаимодействия между признаками, что повысило интерпретируемость и клиническую применимость результатов.</p> <p><bold>Заключение.</bold> Полученные данные подтверждают целесообразность применения методов машинного обучения для персонализированного прогноза и оптимизации тактики наблюдения у больных с почечно-клеточным раком.</p></trans-abstract><kwd-group xml:lang="en"><kwd>recurrence-free survival</kwd><kwd>renal cell carcinoma</kwd><kwd>tumor thrombosis</kwd><kwd>Cox model</kwd><kwd>DeepSurv</kwd><kwd>machine learning</kwd><kwd>SHAP</kwd><kwd>prognosis</kwd><kwd>oncourology</kwd></kwd-group><kwd-group xml:lang="ru"><kwd>безрецидивная выживаемость</kwd><kwd>почечно-клеточный рак</kwd><kwd>опухолевый тромбоз</kwd><kwd>модель Кокса</kwd><kwd>DeepSurv</kwd><kwd>машинное обучение</kwd><kwd>SHAP</kwd><kwd>прогноз</kwd><kwd>онкоурология</kwd></kwd-group><funding-group/></article-meta></front><body></body><back><ref-list><ref id="B1"><label>1.</label><mixed-citation>Cox DR. Regression models and life tables. Journal of the Royal Statistical Society: Series B. 1972;34(2):187-220. DOI: 10.1111/j.2517-6161.1972.tb00899.x</mixed-citation></ref><ref id="B2"><label>2.</label><mixed-citation>Reva SA, Shaderkin IA, Zyatchin IV, Petrov SB. Artificial intelligence in cancer urology. Experimental and Clinical Urology. 2021;14(2):46-51. [Рева С.А., Шадеркин И.А., Зятчин И.В., и др. Искусственный интеллект в онкоурологии. Экспериментальная и клиническая урология. 2021;14(2):46-51]. DOI: 10.29188/2222-8543-2021-14-2-46-51</mixed-citation></ref><ref id="B3"><label>3.</label><mixed-citation>Du M, Haag DG, Lynch JW, et al. Comparison of the tree-based machine-learning algorithms to Cox regression in predicting the survival of oral and pharyngeal cancers: analyses based on SEER database. Cancers. 2020;12(10):2802. DOI: 10.3390/cancers12102802</mixed-citation></ref><ref id="B4"><label>4.</label><mixed-citation>Qiu X, Gao J, Yang J, et al. A comparison study of machine learning (random survival forest) and classic statistic (Cox proportional hazards) for predicting progression in high-grade glioma after proton and carbon ion radiotherapy. Frontiers in Oncology. 2020;10:551420. DOI: 10.3389/fonc.2020.551420</mixed-citation></ref><ref id="B5"><label>5.</label><mixed-citation>Lundberg SM, Lee S-I. A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems. 2017;30:4765-4774.</mixed-citation></ref><ref id="B6"><label>6.</label><mixed-citation>Gonen M, Heller G. Concordance probability and discriminatory power in proportional hazards regression. Biometrika. 2005;92(4):965-970. DOI: 10.1093/biomet/92.4.965</mixed-citation></ref><ref id="B7"><label>7.</label><mixed-citation>Liu Y, Zhou S, Wei H, An S. A comparative study of forest methods for time-to-event data: variable selection and predictive performance. BMC Medical Research Methodology. 2021;21(1):193. DOI: 10.1186/s12874-021-01386-8</mixed-citation></ref><ref id="B8"><label>8.</label><mixed-citation>Katzman JL, Shaham U, Cloninger A, et al. DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC Medical Research Methodology. 2018;18(1):24. DOI: https://doi.org/10.1186/s12874-018-0482-1</mixed-citation></ref><ref id="B9"><label>9.</label><mixed-citation>Alabi RO, Elmusrati M, Leivo I, et al. Machine-learning explainability in nasopharyngeal cancer survival using LIME and SHAP. Scientific Reports. 2023;13(1):8984. DOI: 10.1038/s41598-023-35795-0</mixed-citation></ref><ref id="B10"><label>10.</label><mixed-citation>Kantidakis G, Putter H, Lancia C, et al. Survival prediction models since liver transplantation – comparisons between Cox models and machine-learning techniques. BMC Medical Research Methodology. 2020;20(1):277. DOI: 10.1186/s12874-020-01153-1</mixed-citation></ref><ref id="B11"><label>11.</label><mixed-citation>Byun S-S, Heo TS, Choi JM, et al. Deep-learning-based prediction of prognosis in non-metastatic clear cell renal cell carcinoma. Scientific Reports. 2021;11(1):1242. DOI: 10.1038/s41598-020-80262-9</mixed-citation></ref><ref id="B12"><label>12.</label><mixed-citation>GraphPad Software. Multicollinearity in Cox proportional hazards regression. URL: https://www.graphpad.com/guides/prism/latest/statistics/stat_cox_reg_results_multicollinearity.htm</mixed-citation></ref><ref id="B13"><label>13.</label><mixed-citation>Moncada-Torres A, van Maaren MC, Hendriks MP, et al. Explainable machine learning can outperform Cox regression predictions and provide insights in breast-cancer survival. Scientific Reports. 2021;11(1):6968. DOI: 10.1038/s41598-021-86327-7</mixed-citation></ref><ref id="B14"><label>14.</label><mixed-citation>Park H, Jeong CW, Yuk H, et al. Influence of tumor thrombus on occurrence of distant venous thromboembolism and survival in patients with renal cell carcinoma after surgery. Clinical and Applied Thrombosis/Hemostasis. 2019;25:1076029618823288. DOI: 10.1177/1076029618823288</mixed-citation></ref><ref id="B15"><label>15.</label><mixed-citation>Shin D, Lim B, Song C, et al. Comparative analysis of oncologic outcomes in surgically treated patients with renal cell carcinoma and renal-vein thrombosis by pathologic subtypes. Scientific Reports. 2025;15(1):15946. DOI: 10.1038/s41598-025-00452-1</mixed-citation></ref><ref id="B16"><label>16.</label><mixed-citation>Yang B, Liu C-X, Wu R, et al. Development and validation of a DeepSurv nomogram to predict survival outcomes and guide personalized adjuvant chemotherapy in non-small-cell lung cancer. Frontiers in Oncology. 2022;12:895014. DOI: 10.3389/fonc.2022.895014</mixed-citation></ref><ref id="B17"><label>17.</label><mixed-citation>Schulz S, Woerl A-C, Jungmann F, et al. Multimodal deep learning for prognosis prediction in renal cancer. Frontiers in Oncology. 2021;11:788740. DOI: 10.3389/fonc.2021.788740</mixed-citation></ref><ref id="B18"><label>18.</label><mixed-citation>Mahootiha M, Qadir HA, Aghayan D, et al. Deep-learning-assisted survival prognosis in renal cancer: a CT-scan-based personalized approach. Heliyon. 2024;10(2):e24374. DOI: 10.1016/j.heliyon.2024.e24374</mixed-citation></ref></ref-list></back></article>
