Circulating micro-RNA profiling in patients with advanced non-squamous non small-cell lung cancer receiving bevacizumab/erlotinib first-line treatment followed by platinum-based chemotherapy at disease progression (SAKK 19/05)
Completed · 2012 until 2013
Jörger Markus, Zappa Francesco, Gautschi Oliver, Pless Miklos, Ochsenbein Adrian, Bubendorf Lukas, von Moos Roger, Betticher Daniel, Stahel Rolph, Früh Martin, Baty Florent, Brutsche Martin
Background: This study was initiated by the Swiss Group for Clinical Cancer Research (SAKK) 19/05 study to assess the predictive and prognostic value of circulating miRNA profiles from peripheral blood of patients with non-squamous non small-cell lung cancer (NSCLC) receiving 1st-line bevacizumab/erlotinib followed by platinum-based chemotherapy at the time of progression . Patients and Methods: We included 50 patients with available baseline and 24 hours blood samples. The primary objective of the study was to identify prognostic miRNAs for overall survival (OS). Following confirmation of RNA content and quality, patient samples were analyzed with Agilent human miRNA 8x60K microarrays (release 16.0). One glass slide formatted with eight high-definition 60K arrays. Based on miRBase release 16.0. Each array containing 40 probes targetting each of the 1’347 miRNAs. Data preprocessing included quantile normalization and range migration algorithm filtering. Predictive and prognostic miRNA expression profiles were identified by using the unpaired two-tailed parametric T-test (for quantiative endpoints such as tumor shrinkage) or log-rank testing (for survival endpoints). Results: Data preprocessing kept 49 patients and 424 miRNA’s for further analysis. Ten miRNA’s significantly predicted OS, including miR-29a (HR=6.44, 95%-CI 2.39-17.33, p-value 0.0002). Six miRNA candidates (hsa-miR-29a, hsa-miR-542-5p, hsa-miR-502-3p, hsa-miR-376a, hsa-miR-500a, hsa-miR-424) were found to be insensitive to perturbations according to internal crossvalidation (jackknife) on their HR for OS. The respective principal component analysis defined a multi-miRNA signature including these 6 miRNA candidates that significantly predicted OS in the study population. MicroRNA-665 significantly correlated with percentage tumor shrinkage following 1st-line bevacizumab/erlotinib (rho=0.62, p-value=0.0002). Conclusions: Profiling of easily accessible circulating miRNAs successfully identified a prognostic 6-microRNA signature in patients with advanced non-squamous NSCLC, as well as predictive miRNAs for molecularly-targeted treatment and conventional chemotherapy. This technique should be further evaluated for potential treatment monitoring in patients with advanced NSCLC.