RESUMEN
PURPOSE: Lung cancer accounts for one in five cancer deaths. Broad screening strategies for high-risk populations are unavailable, and the validation of biomarkers for early cancer detection remains a prime interest. Therefore, we investigated the value of circulating U2 small nuclear RNA fragments (RNU2-1f) as a biomarker for diagnosis, prognosis estimation and treatment monitoring in a large lung cancer cohort. METHODS: We determined RNU2-1f abundance in sera of patients with treatment-naive lung cancer (n = 211, 25.6 % early stage), chronic lung disease (n = 56) and healthy controls (n = 58) by reverse transcription quantitative PCR. Initial levels and changes after one chemotherapy cycle were correlated with treatment outcomes in patient subsets. RESULTS: Relative serum RNU2-1f expression levels (REL) were elevated in lung cancer patients compared with patients with chronic lung disease and healthy controls (p < 0.0001). The area under the receiver operating characteristic curve for the complete data set (lung cancer vs. healthy) was 0.91 (95 % CI 0.87-0.95). By applying a REL of -4.505 as diagnostic cutoff (Youden's criterion), sensitivity and specificity reached 0.86 and 0.81, respectively. To determine the generalization error, in a subsampling study, sensitivity and specificity were estimated as 0.82 and 0.77 for the application to future, independent samples. High initial RNU2-1f REL were associated with shorter median survival in stage IIIB/IV disease (RNU2-1fhigh = 228 days/RNU2-1flow = 484 days; p = 0.009, log-rank test, HR1.43 95 % CI 1.23-1.66). Multivariate analysis confirmed RNU2-1f as an independent prognostic factor. Patients with subsequent RNU2-1f reduction had a trend toward better treatment outcome. CONCLUSIONS: Serum RNU2-1f may serve as a biomarker for lung cancer detection, prognosis prediction and treatment monitoring.