Abstract
Kidney cancer is one of the most heterogeneous malignant tumors of the urinary system. Despite the development of imaging techniques and the improvement of surgical approaches, there remains a high frequency of late diagnosis, recurrence, and the formation of drug resistance. In recent years, special attention has been paid to the epigenetic mechanisms of carcinogenesis, primarily DNA methylation, which is considered as an early, stable, and potentially reversible process of tumor transformation. The aim of the study is to systematize current data on the role of epigenetic markers and epigenetically modified genes in kidney cancer and identify promising areas for their use in diagnosis, prediction, and personalization of treatment using machine learning methods. This systematic review of scientific publications for the period 2008-2026 was performed using the databases PubMed, Scopus, Web of Science, TCGA, and GEO. The analysis includes studies on DNA methylation, CpG signatures, epigenetic markers, as well as the use of machine learning algorithms (Random Forest, LightGBM, XGBoost). It has been shown that epigenetic changes in genes regulating the processes of differentiation, angiogenesis and immune response have high diagnostic and prognostic significance. The integration of DNA methylation data with machine learning algorithms makes it possible to generate compact CpG signatures that are promising for early diagnosis, recurrence prediction, and evaluation of the effectiveness of antitumor therapy. The use of epigenetic markers in combination with machine learning methods opens up new opportunities for personalized oncourology, providing more accurate stratification of patients and optimization of treatment tactics.