A Combinatorial Model of Serum tsRNAs as a Diagnostic Biomarker for Non-Small Cell Lung Cancer

As a promising candidate in liquid biopsy, tRNA-derived small RNAs (tsRNAs) have been implicated as potential diagnostic biomarkers for cancer; however, the diagnostic potential of serum cell-free tsRNAs in non-small cell lung cancer (NSCLC) remains unexplored. This study aimed to develop a combinatorial model based on serum cell-free tsRNA signatures and evaluated its diagnostic performance for NSCLC detection. Through systematic screening from the tsRFun database, we identified 32 NSCLC-related differentially expressed tsRNAs, whose predicted target genes were significantly enriched in the MAPK pathway. Four MAPK pathway-associated tsRNAs were selected for further validation due to their robust discriminatory power in distinguishing NSCLC patients from healthy control. Subsequent validation via quantitative RT-PCR confirmed significantly elevated serum levels of tsRNA-Thr-5-0039, tsRNA-Thr-5-0044, and tsRNA-Cys-5-0011 in NSCLC patients compared to healthy controls. The final combinatorial diagnostic model incorporating these three tsRNAs achieved an AUC of 0.83 (95% CI: 0.70-0.95), with 71.8% accuracy, 52.6% sensitivity, and 90.0% specificity. The three-tsRNA model outperformed individual tsRNAs in diagnostic efficiency. In conclusion, these findings highlight the diagnostic potential of circulating tsRNA signatures as a minimally invasive liquid biopsy approach, offering new perspectives for early diagnosis of NSCLC.