Identification and validation of diagnostic alternative splicing events in tumor-educated platelets for non-small cell lung cancer in patients with ground-glass opacity: a multicenter study

Background: The diagnostic potential of tumor-educated platelets (TEPs) across various cancer types has gained increasing recognition; however, the relationship between alternative splicing (AS) events in TEPs and tumor development remains understudied. Early detection of non-small cell lung cancer (NSCLC) in ground-glass opacities (GGOs) is critical for improving patient outcomes, yet current methods lack sufficient accuracy. Our research identified diagnostic-related alternative splicing events (DASEs) in TEPs, revealing several promising biomarkers for NSCLC, specifically in patients presenting with GGOs. Methods: Patients with GGOs from two hospitals were prospectively enrolled [Hospital 1-Platelet (H1- P) and Hospital 2-Tissue (H2-T) in the validation cohort; Hospital 2-Platelet (H2-P) in the test cohort]. Benign/malignant diagnosis of GGOs was confirmed by pathological examination according to the World Health Organization (WHO) classification. TEPs from the H1-P cohort were collected for transcriptome sequencing and AS analysis. Chi-square tests, least absolute shrinkage and selection operator (LASSO) regression analysis, and protein-protein interaction (PPI) network were used for the preliminary screening of DASEs. Pathological tissue from the H2-T cohort was collected to validate the diagnostic efficacy of hub DASEs in NSCLC against the pathological gold standard. Moreover, TEPs from the H2-P cohort were used to assess the predictive performance of hub DASEs for GGOs using receiver operating characteristic (ROC) curves. Decision curve analysis (DCA) was used to determine whether diagnosing NSCLC in the GGOs population via hub DASEs could benefit patients. Results: A total of 285 patients with GGOs were enrolled, including 151 NSCLC and 128 inflammatory nodules confirmed by pathological examination. Thirteen DASEs were screened with the chi-square test and LASSO regression analysis to identify diagnostic TEP AS markers for GGOs NSCLC. The PPI network identified four hub diagnostic-related alternative splice genes (DASGs) (TMEM219, MPV17, FIBP, and VPS28). Pathological tissues and platelets were collected to validate the four hub DASEs of these four hub DASGs. MXE-32112-TMEM219 yielded an area under the curve (AUC) of 0.82 [95% confidence interval (CI): 0.729–0.902], with a sensitivity of 83.33% and a specificity of 80.00%; RI-3259-VPS28 yielded an AUC of 0.77 (95% CI: 0.677–0.870) with a sensitivity of 93.33% and a specificity of 78.33%; and RI-3641-MPV17 yielded an AUC of 0.82 (95% CI: 0.728–0.901) with a sensitivity of 90.00% and a specificity of 80.00%. The DCA results suggested that using hub DASEs in diagnosing NSCLC in individuals with GGOs could provide benefits. Conclusions: The specific diagnostic AS events (MXE-32112-TMEM219, RI-3259-VPS28, and RI-3641- MPV17) identified in TEP samples demonstrated high sensitivity and specificity for diagnosing NSCLC in patients with GGOs. These findings suggest that TEP-related AS events may serve as non-invasive biomarkers to guide biopsy decisions for NSCLC in GGOs, reducing unnecessary procedures.