Abstract Clear cell renal cell carcinoma (ccRCC) is the most prevalent form of kidney cancer, distinguished by intricate interactions between metabolic reprogramming, immune microenvironment dynamics, and genetic mutations. In this detailed investigation, we analyzed the ccRCC cohort from The Cancer Genome Atlas (TCGA) alongside 81 metabolic signaling pathways from the KEGG database. By utilizing Gene Set Variation Analysis (GSVA), we performed hierarchical clustering of patients based on their metabolic pathway activity profiles, identifying three distinct clusters with notable differences in pathway activity and survival outcomes. Cluster 1 displayed high metabolic activity and more favorable survival outcomes, while Cluster 3 was characterized by low metabolic activity and poorer prognosis. Clinical comparisons revealed significant disparities in gender, histological stage, and survival status, with Cluster 3 exhibiting a higher proportion of patients at advanced stages and those who had passed away. Genetically, Cluster 1 showed the highest mutation burden, with prominent mutations in genes such as VHL and PBRM1. Biological process analysis indicated that pathways like organic carboxylic acid metabolism and ATP synthesis were upregulated in Cluster 1 but suppressed in Cluster 3. Machine learning models (GBM, CoxBoost, and LASSO regression) enabled the identification of four pivotal genes—BCAT1, IL4I1, ACADM, and ACADSB—which were subsequently used to construct a multifactorial Cox regression model. This model successfully stratified patients into high- and low-risk groups, correlating with marked differences in immune activities. The high-risk group showed elevated expression of chemokines, TNF, and HLA molecules. Drug sensitivity analysis suggested that AKT inhibitor III was more effective in the low-risk cohort, while Bortezomib might be more beneficial for high-risk patients. Additionally, a clinical prediction model integrating risk scores and clinical factors demonstrated strong predictive power for patient survival. Methylation profiling of the core genes via the UALCAN platform revealed distinct epigenetic signatures in ccRCC, providing deeper insight into the disease’s molecular mechanisms. This study contributes to a more comprehensive understanding of ccRCC and proposes valuable directions for personalized treatment strategies and enhanced patient management. Keywords: Clear cell renal cell carcinoma, Metabolic pathways, Immune microenvironment, Prognostic biomarkers, Personalized therapy Background Renal cell carcinoma (RCC) represents a significant challenge in oncology, accounting for approximately 2–3% of all adult cancer cases [[30]1, [31]2]. Among its subtypes, clear cell renal cell carcinoma (ccRCC) stands out as the most prevalent and aggressive form [[32]3]. Known for its resistance to both radiation therapy and conventional chemotherapy, ccRCC proves particularly difficult to treat [[33]4, [34]5]. Despite considerable advances in understanding its molecular and genetic underpinnings, the prognosis for patients with advanced-stage ccRCC remains poor [[35]6–[36]8]. This persistent challenge underscores the urgent need for novel therapeutic strategies and more reliable prognostic tools, marking a critical gap in current cancer treatment efforts. The heterogeneity of ccRCC is well-established, with distinct histological and genetic profiles emerging across its subtypes [[37]9, [38]10]. This diversity complicates clinical management but also offers an opportunity to explore targeted molecular pathways for therapeutic intervention [[39]11, [40]12]. Recent research has increasingly recognized the pivotal role of altered metabolic pathways in ccRCC, including disruptions in lipid, amino acid, and glucose metabolism [[41]13, [42]14]. These metabolic shifts are not merely a consequence of cancer progression but are now understood to actively drive tumorigenesis, enabling cancer cells to thrive in hostile tumor microenvironments and promoting disease advancement [[43]15, [44]16]. In particular, the metabolic reprogramming in ccRCC cells supports rapid cell growth, survival, and resistance to therapies, providing a promising avenue for therapeutic targeting. Another critical area of interest is the tumor microenvironment (TME), especially the involvement of the immune system [[45]17, [46]18]. The TME of ccRCC is typically infiltrated by a variety of immune cells that can either suppress or facilitate tumor growth [[47]19]. The dynamic interaction between tumor cells and immune cells within the TME significantly impacts prognosis and treatment response, particularly to emerging immune checkpoint inhibitors [[48]20, [49]21]. Given the rapid progress in understanding the molecular biology of ccRCC, there is an increasing need to further explore its complexities [[50]22–[51]24]. This knowledge is vital for developing more effective treatment protocols and prognostic tools [[52]25]. The advent of high-throughput sequencing technologies and advanced bioinformatics approaches has revolutionized our capacity to examine the genetic and epigenetic features of ccRCC [[53]26, [54]27]. However, translating these molecular discoveries into clinical applications remains challenging. One major hurdle is the identification of reliable biomarkers that can accurately predict disease progression and response to treatment [[55]28, [56]29]. The inherent heterogeneity of ccRCC adds complexity to this task, making it difficult to adopt a one-size-fits-all approach [[57]30]. Therefore, the future of ccRCC treatment lies in personalized medicine, which tailors therapies to each patient’s unique molecular profile, offering more precise and effective treatment options [[58]31]. Furthermore, the integration of molecular data with clinical metrics to construct comprehensive risk models is still underexplored. Such models could not only predict patient outcomes but also guide clinical decision-making, enabling more targeted and adaptive treatment strategies. The motivation for our study stems from these pressing challenges in ccRCC research and treatment. By focusing on the molecular landscape of ccRCC, particularly metabolic pathways and the immune microenvironment, our study seeks to fill critical gaps in the current understanding of this disease. Utilizing data from public resources such as The Cancer Genome Atlas (TCGA) and leveraging advanced bioinformatics tools, we aim to identify key molecular alterations in ccRCC. Our goal is to develop an integrative risk stratification model that combines both molecular and clinical data, ultimately improving patient care through personalized treatment approaches and advancing the field of personalized medicine in ccRCC. Methods Data collection Our data collection incorporated two primary sources: the TCGA-KIRC dataset for renal clear cell carcinoma and metabolic signaling pathway data from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. The TCGA-KIRC dataset includes 72 adjacent non-tumor tissue samples and 541 ccRCC samples, which were utilized for differential expression analysis. Of these, 528 patients with available survival and clinical data were selected for subgroup classification and risk model development. Additionally, we integrated 81 metabolic-related signaling pathways from the KEGG database. Metabolic pathway analysis in renal patients To assess the activity of 81 metabolic pathways, we employed a multi-faceted approach. Gene Set Variation Analysis (GSVA) was performed to score pathway activities using data from the KEGG database ([59]https://www.kegg.jp/kegg/pathway.html). The GSVA and GSEABase R packages facilitated the GSVA enrichment analysis. To categorize patients based on metabolic pathway activity, hierarchical clustering was performed using Euclidean distances, and the results were visualized with heatmaps generated by the pheatmap package. Principal Component Analysis (PCA) was then conducted using the ggplot2 package to further stratify patients according to their enrichment scores. Survival curves, generated using the survival and survminer packages, were analyzed to assess overall survival differences between groups. To explore the relationship between metabolic pathways and the immune system, we calculated enrichment scores for 29 immune cell types and processes using the ssGSEA algorithm. These immune enrichment scores were then correlated with metabolic pathway activity, and the findings were visualized as a correlation heatmap using ggplot2 and dplyr. Clinical and genetic differences among renal patient subgroups To assess clinical and genetic differences across renal patient subgroups, we first evaluated clinical differences, focusing on gender, histological stage, and survival status. These differences were visualized through heatmaps generated using the ComplexHeatmap package. To explore genetic variation, we analyzed mutation patterns of the top 20 most frequently mutated genes in each subgroup. Single nucleotide variant (SNV) data were obtained using the TCGAbiolinks package, and mutation patterns were visualized in waterfall plots created with the maftools package. Biological variances in renal patient subgroups via enrichment analyses To investigate biological differences, Gene Ontology (GO) enrichment analysis was performed using the clusterProfiler package, identifying key biological processes, with results expressed as z-scores for clarity. Gene Set Enrichment Analysis (GSEA) was then conducted on KEGG gene sets from the MSigDB database, also using clusterProfiler. The results of the GO and GSEA analyses were visualized using the GOplot and GseaVis packages for enhanced interpretation of biological processes and pathway enrichments. Survival analysis and prognostic modeling in ccRCC To evaluate the survival impact of 81 metabolic pathways, we performed a univariate Cox regression analysis using the survival package. Differential analysis, conducted via the limma package, identified genes with an absolute fold change (FC) > 1.5 and an adjusted P-value < 0.05. The TCGA-KIRC dataset was divided into a training set (372 samples) and a validation set (156 samples) in a 7:3 ratio using the caret package. To identify prognostic genes, we applied three machine learning methods: Gradient Boosting Machine (GBM) with the gbm package, Coxboost through the CoxBoost package, and LASSO regression using the glmnet package. A Venn diagram, created via an online tool ([60]http://bioinformatics.psb.ugent.be/webtools/Venn/), was used to identify common core genes across the three methods. A stepwise Cox regression analysis was then conducted to develop the prognostic model, and survival analysis was performed using the survival and survminer packages. To assess the model’s predictive accuracy for 1-, 3-, and 5-year survival, time-dependent Receiver Operating Characteristic (ROC) curves were generated with the timeROC package. Additionally, t-SNE analysis using the Rtsne package was conducted to visualize the patient risk groups. Immune landscape assessment in high and low-risk groups To assess differences in the immune landscape between high- and low-risk groups, we analyzed the expression levels of key immune activity markers (chemokines, TNF family molecules, HLA family molecules) to identify immune profile variations. Expression differences were visualized using boxplots generated with the ggpubr package. To evaluate immune components further, we analyzed the ESTIMATE scores, which encompass stromal, immune, and tumor microenvironment scores, using cloud plots created with the gghalves package. Additionally, the Tumor Immune Dysfunction and Exclusion (TIDE) score, which reflects immune therapy sensitivity, was calculated and visualized using ggbeeswarm and ggplot2. Microsatellite instability (MSI) was assessed through violin plots from ggpubr. Finally, the progression of tumor-related processes, including angiogenesis [[61]32], cell cycle regulation [[62]33], and epithelial-mesenchymal transition (EMT) [[63]34], was evaluated using relevant gene sets, with the results visualized through color dot plots created with ggplot2 and ggpubr. Drug sensitivity analysis To explore the correlation between the IC50 values of various drugs and patient risk scores, we utilized the pRRophetic package for calculating IC50 values. The drugs selected for this analysis were chosen based on their relevance to ccRCC treatment and their potential efficacy in high- and low-risk patient subgroups. These included targeted therapies such as tyrosine kinase inhibitors and immune checkpoint inhibitors, which are commonly used in clinical practice for ccRCC. The results were visualized using box-and-whisker and scatter plots, which were generated with the ggpubr package. Development and validation of clinical predictive model for ccRCC To develop a clinical predictive model, risk scores and clinical characteristics were integrated into a univariate Cox regression analysis to evaluate their relationship with survival outcomes. Forest plots, generated with the ggplot2 package, visualized these relationships. Nomograms, created with the StepReg and regplot packages, were used to predict 1-, 3-, and 5-year survival rates. The model’s calibration was assessed using calibration curves from the rms package, and its discriminatory ability was evaluated with time-dependent ROC curves from the timeROC package. Survival curves were plotted using the survival package. Finally, the methylation status of core genes in the prognostic model was analyzed through the UALCAN website. Statistical analysis All data were analyzed and visualized using R software (version 4.2.1), unless otherwise specified. Statistical significance was determined using a two-sided P-value < 0.05. Results Metabolic pathway activity and patient clustering We initially conducted GSVA on 81 metabolic pathways across ccRCC patients, followed by hierarchical clustering based on Euclidean distance. This analysis revealed three distinct patient clusters. Cluster 1 displayed high metabolic pathway activity, while Cluster 3 showed significantly lower activity. PCA confirmed this clustering, emphasizing that metabolic pathway activity reflects key features of ccRCC patient characteristics (Fig. [64]1A, B). Fig. 1. [65]Fig. 1 [66]Open in a new tab Metabolic pathway analysis and survival correlation in renal cancer patients. A Heatmap depicting hierarchical clustering of renal cancer patients based on the GSVA scores of 81 metabolic-related pathways. B PCA plot illustrating the separation of renal cancer patients into three distinct clusters according to metabolic pathway enrichment scores. C Kaplan–Meier survival curves comparing the survival outcomes across the three identified patient clusters. D Correlation plot displaying the relationships between GSVA-derived enrichment scores for 29 immune cells/processes and metabolic pathway activity Survival and molecular profiles of the subgroups Our survival analysis revealed significant differences in overall survival (OS) across the three clusters. Cluster 1, with the highest metabolic activity, showed the best survival outcomes, while Cluster 3, with low pathway activity, had the poorest prognosis (Fig. [67]1C). Further analysis using ssGSEA assessed 29 immune cells and processes, showing positive correlations between metabolic pathway activity and immune functions (Fig. [68]1D). Clinical characteristics, such as gender, histological stage, and survival status, further distinguished the clusters. Cluster 3 had a higher proportion of patients in advanced stages (G4/GX) and a greater number of deaths (Fig. [69]2A). Mutation analysis of the top 20 frequently mutated genes revealed distinct mutation profiles for each cluster. Cluster 1 exhibited the highest mutation rate (87.1%) with mutations in VHL, PBRM1, TTN, BAP1, and MTOR (Fig. [70]2B–D). Fig. 2. [71]Fig. 2 [72]Open in a new tab Clinical and genetic features of renal cancer subgroups. A Heatmap showing the clinical characteristics of the three clusters, highlighting significant differences in gender, histological stage, and survival status. B–D Mutation profile waterfall plots for the 20 most frequently mutated genes in renal cancer across the three clusters. Cluster 1 (B) exhibits the highest mutation rate with frequent mutations in genes like VHL and PBRM1, Cluster 2 (C) has moderate mutation rates, and Cluster 3 (D) shows the lowest mutation frequency, each cluster exhibiting distinct genetic alterations Biological differences between clusters To explore the biological variations between clusters, Gene Ontology (GO) enrichment analysis of differentially expressed genes (DEGs) revealed significant upregulation of organic carboxylic acid metabolism and ATP production in Cluster 1, while these processes were downregulated in Cluster 3 (Fig. [73]3A, B). Gene Set Enrichment Analysis (GSEA) of KEGG pathways highlighted upregulation of energy-related pathways like oxidative phosphorylation, glycolysis, and fatty acid metabolism in Cluster 1, whereas Cluster 3 showed downregulation of these pathways and upregulation of cancer-related pathways such as ECM receptor interactions and TGF-beta signaling (Fig. [74]3C, D). These findings suggest that elevated metabolic and energy pathway activity correlates with better survival, while alterations in the extracellular matrix may contribute to poorer outcomes. Fig. 3. [75]Fig. 3 [76]Open in a new tab Gene ontology and pathway enrichment analysis of ccRCC subgroups. A GO enrichment analysis for Cluster 1 showing upregulated biological processes. B GO enrichment analysis for Cluster 3 indicating downregulated biological processes, suggesting a reduction in metabolic functions. C GSEA for Cluster 1 indicating the activation of certain pathways. D GSEA for Cluster 3, showing downregulation of the PPARG pathway and activation of cancer-associated pathways Prognostic model development and validation Using univariate Cox regression analysis, we identified 53 metabolic pathways with survival significance (P < 0.05, Fig. [77]4A). The top five pathways—Nitrogen metabolism, Fatty acid degradation, and others—were analyzed to extract 118 unique genes, of which 33 were differentially expressed (Fig. [78]4B). After further univariate Cox regression, we pinpointed 21 survival-related genes (Fig. [79]4C). Machine learning approaches, including Gradient Boosting Machine (GBM) (Fig. [80]4D), CoxBoost (Fig. [81]4E), and LASSO regression (Fig. [82]4F–G), identified four core genes: BCAT1, IL4I1, ACADM, and ACADSB (Fig. [83]4H). The prognostic model was constructed using stepwise Cox regression, resulting in a risk score formula: Risk score = 0.293BCAT1 + 0.239IL4I1—0.383ACADM—0.339ACADSB. This model effectively stratified patients into high- and low-risk groups with significant survival differences in both training and validation datasets (Fig. [84]5A–C). The ROC curves confirmed the model’s predictive accuracy for 1-, 3-, and 5-year survival rates (Fig. [85]5D–F), and t-SNE analysis further validated the distinct patient characteristics represented by these risk groups (Fig. [86]5G–I). Fig. 4. [87]Fig. 4 [88]Open in a new tab Core gene identification in ccRCC using metabolic pathway analysis and machine learning. A Univariate Cox regression analysis identified 53 metabolic pathways with significant survival associations (P < 0.05). The top five pathways, including Nitrogen metabolism and Fatty acid degradation, are highlighted. B Differential expression analysis revealed 118 unique genes across the significant pathways, with 33 genes showing differential expression. C Univariate Cox regression analysis narrowed down 21 survival-related genes. D–H Machine learning models, including Gradient Boosting Machine (GBM) (D), CoxBoost (E), and LASSO regression (F–G), were used to identify four core genes (H)—BCAT1, IL4I1, ACADM, and ACADSB Fig. 5. [89]Fig. 5 [90]Open in a new tab Evaluation and risk stratification of prognostic model. A–C Kaplan–Meier survival curves for high- and low-risk groups based on the prognostic model. D–F ROC curves for predicting survival at 1, 3, and 5 years. G–I t-SNE plots visualizing the separation of patients into distinct risk groups Immune landscape differences between high- and low-risk groups Immune activity analysis showed that the high-risk group exhibited elevated expression of chemokines, TNF family molecules, and HLA family molecules (Fig. [91]6A–C), indicating heightened immune activity. Higher stromal and immune scores, as well as tumor microenvironment (TME) scores, were also observed in the high-risk group (Fig. [92]6D–F), suggesting a more intense immune-inflammatory microenvironment. Fig. 6. [93]Fig. 6 [94]Open in a new tab Divergence in immune landscape between high- and low-risk ccRCC groups. A Heatmap comparing chemokine expression levels between high- and low-risk groups, revealing significantly elevated chemokine expression in the high-risk group. B Higher expression of TNF family molecules in the high-risk group, suggesting increased inflammatory activity. C Elevated HLA family molecule expression in the high-risk group, indicating enhanced immune presentation activity. D ESTIMATE stromal scores showing higher stromal content in the high-risk group’s tumor microenvironment. E Immune scores indicating more robust immune cell infiltration in the high-risk group, reflecting an intensified immune response. F Overall tumor microenvironment scores are elevated in the high-risk group, supporting the conclusion of a more severe immune-inflammatory environment in these patients Immune therapy sensitivity and tumor progression The low-risk group demonstrated a lower TIDE score, indicating reduced resistance to immune therapies (Fig. [95]7A). A significant positive correlation between risk score and TIDE score (Fig. [96]7B) suggested that low-risk patients are more likely to respond to immunotherapy. Additionally, the low-risk group had higher microsatellite instability (MSI) scores, indicating better suitability for immune checkpoint inhibitors (Fig. [97]7C). Tumor progression analyses revealed that the low-risk group had higher angiogenesis scores (Fig. [98]7D), while the high-risk group exhibited elevated scores in cell cycle regulation (Fig. [99]7E) and epithelial-mesenchymal transition (EMT) (Fig. [100]7F), suggesting different therapeutic targets for these two groups. Fig. 7. [101]Fig. 7 [102]Open in a new tab Immune therapy sensitivity and tumor progression in ccRCC. A Scatter plot comparing TIDE scores between high- and low-risk groups. B Correlation plot showing a positive relationship between risk score and TIDE score. C Violin plot illustrating the distribution of Microsatellite Instability (MSI) scores. D Scatter plot of angiogenesis scores versus risk scores. E Scatter plot showing the correlation between cell cycle scores and risk scores. F Scatter plot of EMT scores correlating with risk scores Drug sensitivity and correlation with risk scores Drug sensitivity analysis revealed a correlation between risk scores and the IC50 values for several drugs. The low-risk group showed greater sensitivity to AKT inhibitors, while the high-risk group was more sensitive to Bortezomib (Fig. [103]8A–L), highlighting potential therapeutic strategies based on individual risk profiles. Fig. 8. [104]Fig. 8 [105]Open in a new tab Drug sensitivity and risk score correlation. A–D Scatter and box plots demonstrating a positive correlation between risk scores and IC50 values for AKT inhibitor III and other drugs. E–L Scatter and box plots showing a significant negative correlation between risk scores and IC50 values for Bortezomib and seven other drugs Prognostic model efficacy and clinical application We integrated the risk score with clinical variables such as age, pathological staging, and TNM staging to create a comprehensive clinical prediction model. This model demonstrated strong predictive power for 1-, 3-, and 5-year survival (Fig. [106]9A–E), with a nomogram and calibration curve confirming its accuracy (Fig. [107]9B–C). Methylation analysis revealed distinct methylation patterns for the four core genes in ccRCC, with ACADM and ACADSB exhibiting higher methylation in cancer tissues, suggesting their potential as epigenetic biomarkers (Fig. [108]9F–I). Fig. 9. [109]Fig. 9 [110]Open in a new tab Clinical prediction model development and validation. A Forest plot illustrating the association of clinical factors (age, pathological staging, TNM staging) and risk score with patient survival. B Survival nomogram incorporating risk score and clinical characteristics. C Calibration curve assessing the accuracy of the prognostic model. D ROC curve analysis demonstrating the discriminative ability of the model to predict patient outcomes. E Kaplan–Meier survival curves stratified by the prognostic model. F–I Box plots showing methylation levels of ACADM and ACADSB in renal cancer compared to higher methylation of the other two genes in control groups Discussion ccRCC, known for its aggressive behavior and poor prognosis in advanced stages, presents a unique challenge due to its complex metabolic reprogramming, altered immune microenvironment, and diverse genetic landscape [[111]35, [112]36]. This study utilized data from the TCGA-KIRC cohort and 81 metabolic pathways from the KEGG database to explore the molecular landscape of ccRCC, providing new insights into its pathophysiology and potential targets for therapy. Our analysis highlighted the metabolic heterogeneity within ccRCC. By employing GSVA and hierarchical clustering, we identified distinct patient subgroups based on metabolic pathway activity. This approach builds on the work of the Cancer Genome Atlas (2013), which also highlighted metabolic alterations in ccRCC, but our study further investigates the correlation between metabolic activity and survival outcomes, an area not fully explored in prior studies. The patient clustering based on metabolic activity is a novel aspect of our research. We observed that Cluster 1, characterized by high metabolic activity, had better survival outcomes compared to Cluster 3, which exhibited low metabolic activity. This aligns with recent studies suggesting that metabolic reprogramming in cancer cells significantly impacts patient prognosis [[113]37]. Moreover, the distinct clinical characteristics of these clusters—such as gender distribution, histological stages, and survival status—highlight the potential of metabolic profiling for patient stratification in ccRCC. In our study, we examined the genetic alterations in ccRCC and observed that Cluster 1 exhibited the highest mutation rate, with prominent mutations in genes such as VHL, PBRM1, and BAP1 [[114]38]. While VHL mutations are generally associated with better overall survival (OS), PBRM1 and BAP1 mutations have been linked to worse OS outcomes. This aligns with existing research that highlights the complex prognostic implications of these mutations in ccRCC. However, our study did not specifically analyze the direct influence of these individual mutations on OS but rather focused on the broader integration of genetic alterations with metabolic and immune pathway changes. Although the association between genetic alterations and clinical outcomes in ccRCC is well established, the relationship between these mutations and specific metabolic reprogramming or immune microenvironment alterations remains less explored. Our study sheds light on how such genetic alterations might interact with metabolic pathways, offering valuable insights into the molecular mechanisms of ccRCC, which could help inform the development of more targeted therapeutic strategies. The evaluation of the immune microenvironment across different risk groups revealed a more intense immune-inflammatory response in the high-risk group. This observation supports recent studies emphasizing the role of the immune microenvironment in cancer progression and treatment response [[115]39, [116]40]. Our study builds on this by linking immune activity with metabolic pathway alterations, suggesting that metabolic reprogramming in ccRCC could influence the tumor immune microenvironment. Additionally, the correlation between risk score and TIDE score in our study indicates a lower likelihood of immune therapy rejection in the low-risk group. This finding has important implications for selecting patients for immunotherapy, a promising treatment in ccRCC that requires better stratification for optimal efficacy [[117]41]. Our drug sensitivity analysis showed a correlation between risk score and IC50 values for drugs like AKT inhibitor III, providing a foundation for more personalized treatment strategies. The significant negative correlation for drugs such as Bortezomib in the high-risk group suggests potential therapeutic options for these patients. These insights could guide the development of tailored treatment regimens based on patient risk profiles, a strategy not fully explored in current ccRCC treatments. The clinical prediction model we developed, incorporating both molecular and clinical data, demonstrated excellent predictive power for patient survival. This model outperforms traditional staging systems by offering a more nuanced risk assessment, potentially enhancing treatment decision-making in clinical practice. Our analysis also revealed notable gender differences in the clinical outcomes and metabolic characteristics of ccRCC patients. Specifically, male patients were more likely to be assigned to the high-risk group, exhibiting higher metabolic activity and poorer prognosis. This observation is consistent with previous studies that have indicated that male patients with ccRCC tend to have more aggressive disease and may present with higher mutation rates in key genes such as VHL and PBRM1, which are critical in ccRCC tumorigenesis. In contrast, female patients in the low-risk group exhibited better survival outcomes, potentially due to a more favorable immune response and lower levels of inflammation in the tumor microenvironment. These findings are supported by reports suggesting that female patients may have stronger immune surveillance, which could enhance their ability to mount an effective anti-tumor response, thereby contributing to improved prognosis. Clinical validation could involve correlating the model’s risk scores with patient outcomes, such as overall survival and response to therapy, to determine its prognostic value. Additionally, circulating biomarkers in blood samples, such as RNA or proteins corresponding to these four genes, could potentially reflect these genetic changes and serve as non-invasive indicators for disease progression and prognosis. Liquid biopsy technologies, including circulating tumor DNA (ctDNA) or exosome-based assays, could be particularly valuable in tracking gene expression changes over time and assessing how they relate to tumor dynamics and therapeutic responses. If validated, this model could be integrated into clinical decision-making to guide personalized treatment strategies for ccRCC patients, particularly in choosing between targeted therapies or immunotherapies based on individual molecular profiles. Finally, our analysis of gene methylation patterns using the UALCAN website revealed distinct methylation profiles in ccRCC, adding an epigenetic dimension to our understanding of the disease [[118]42]. This aspect of our study contributes to the growing body of research exploring the role of epigenetics in cancer. While our study provides valuable insights into ccRCC, it has limitations. It relies on retrospective data from the TCGA database, which may not fully capture the diversity of ccRCC. The findings are dependent on the quality of the available data and require external validation in larger cohorts for broader applicability. Additionally, our analysis focuses primarily on genomic and transcriptomic data, potentially overlooking important proteomic and metabolomic aspects of ccRCC. Environmental and lifestyle factors, crucial in cancer etiology and management, were also not considered. Translating these molecular insights into clinical practice presents significant challenges and necessitates further research and validation. In conclusion, this study provides a comprehensive analysis of the molecular landscape of clear cell renal cell carcinoma (ccRCC), revealing critical insights into the interplay between metabolic reprogramming, immune microenvironment dynamics, and genetic alterations. By integrating multi-omics data, we identified distinct patient subgroups based on metabolic pathway activity, highlighting how metabolic shifts actively influence prognosis, with high metabolic activity correlating with better survival outcomes and low metabolic activity linked to poorer prognosis. The immune microenvironment also plays a pivotal role in ccRCC progression, as the high-risk group exhibited elevated immune-inflammatory responses, while the low-risk group demonstrated increased sensitivity to immune therapies. Our developed prognostic model, combining molecular and clinical data, effectively stratified patients into high- and low-risk categories and showed strong predictive power for survival. Additionally, drug sensitivity analysis revealed potential therapeutic strategies tailored to risk profiles, with specific drugs showing differential efficacy. Finally, methylation analysis of key genes provided valuable insights into epigenetic regulation, offering potential biomarkers for early detection and therapeutic targeting. This study advances our understanding of ccRCC, proposing novel biomarkers and an integrative model to guide personalized treatment strategies, with further research needed to validate these findings and explore therapeutic targeting of metabolic and immune pathways in larger cohorts. Acknowledgements