Abstract 2028: Development of a miRNA-based prognostic signature for clear cell renal cell carcinoma

Abstract Background: Clear cell renal cell carcinoma (ccRCC) represents about 80% of renal cell carcinoma cases. MicroRNAs (miRNAs) play a critical role in RNA silencing and post-transcriptional regulation of gene expression, impacting tumorigenesis and cancer progression. This study aimed to develop a miRNA-based prognostic model to predict overall survival in ccRCC patients by leveraging genome-wide miRNA expression profiling. Methods: Sequencing data from tumor kidney samples were obtained from the Dartmouth Renal Tumor Biobank. Data was processed using the Nextflow smrnaseq pipeline and analyzed in R. Log-transformed Transcripts Per Million (logTPM) values were used for downstream analysis. After quality control, survival-associated miRNAs were identified using Cox proportional hazards (CoxPH) models, controlling for batch effects and tumor stage. miRNAs with a p-value < 0.01 were further refined through the LASSO method, yielding an optimal set of miRNAs for prognostic modeling. The prognostic risk score was calculated as: Risk score = ∑(logTPM×coefficient). Patients were stratified into high and low risk groups based on the median risk score. Kaplan-Meier survival analyses and CoxPH models were used to assessed survival differences between risk groups. Covariates, including batch, cell proportion, grade, stage, age, and sex, were included in adjusted analyses. Predictive accuracy was assessed by calculating the area under the curve (AUC) for 1, 3, and 5 years using the survivalROC package with the nearest neighbor estimation. Results: 147 tumor samples with complete covariates passed quality control. For the initial analysis, 37 miRNAs were associated with survival, and 19 miRNAs were retained using the LASSO method. Of these, 15 miRNAs were previously reported to have tumor-related functions. Notably, miR-466, -4437, and -524-3p were associated with prognosis in other cancers. The risk score, derived from the expression levels and coefficients of the 19 miRNAs, followed a normal distribution (Shapiro-Wilk normality test p-value = 0.15) with a median value of -4.29. Patients in the high-risk group had a significantly higher hazard for death compared to those in the low-risk group, with a hazard ratio (HR) of 6.21 (95% CI: 3.94-9.78, p-value = 3.28 E-15). After adjusting for relevant covariates, the high-risk group remained an independent predictor of poor survival (HR: 9.06, 95% CI: 4.99-16.46 , p-value = 4.66 E-13). The predictive performance was robust, with AUCs of 0.85, 0.86, and 0.81 for 1, 3, and 5 years, respectively. Conclusion: The miRNA-based risk score effectively predicts survival in ccRCC patients, offering a potential prognostic biomarker for ccRCC. It may provide valuable insights for predicting survival outcomes and guiding clinical decision-making. Future studies should validate these findings in external cohorts and explore the underlying biological mechanisms of the identified miRNAs. Citation Format: Minghui Zhang, Brock C. Christensen, Lucas A. Salas. Development of a miRNA-based prognostic signature for clear cell renal cell carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2028.