OS1.1 Role of RNU6-1 isolated from circulating exosomes as a differential biomarker for GBM versus non-neoplasic brain lesions and PCNSL

Abstract

Background

Glioblastoma (GBM) is the most common malignant primary brain tumour in adults, and is still associated with a dismal prognosis despite intensive research and therapeutic efforts. The identification of circulating biomarkers for tumor detection and response assessment would be of great clinical help. Recent studies have shown that GBM cells release microvesicles containing a select subset of cellular proteins and RNA. Previously, we demonstrated that exosomes isolated from the serum of GBM patients had an increased expression of the small non-coding RNA RNU6-1 compared to control samples, and hence could serve as a non-invasive diagnostic biomarker for GBM.In this study, we set to investigate the role of RNU6-1 as a differential biomarker of GBM versus other brain diseases with similar radiological features.

Material and Methods

We analyzed the expression of RNU6-1 in circulating exosomes of GBM patients (n=18), healthy controls (n=28), and patients with different brain lesions that can mimic GBM on Magnetic Resonance Imaging: subacute stroke (n=30), acute-subacute haemorrhage (n=29), acute demyelinating lesions (n=19), brain metastases (n=21) and Primary CNS Lymphomas (PCNSL) (n=12). First, we isolated the exosomes from the serum of healthy subjects and patients with these pathological conditions using an adsorption method (Exoquick). Then, RNU6-1 levels were analyzed by digital droplet PCR (ddPCR).

Results

Corroborating our previous results, we found that the expression of RNU6-1 was significantly higher in GBM patients (412 ± 550.48 copies/20µL) than in healthy controls (150 ± 224.35 copies/20µL; p=0.039). In addition, RNU6-1 levels were higher in exosomes from GBM patients than in exosomes from patients with non-neoplasic lesions (stroke [223 ± 709.8 copies/20µL; p=0,067], haemorrhage [127 ± 198.7 copies/20µL; p=0.010], demyelinating lesions [111.5 ± 250.35 copies/20µL; p=0.019]) and PCNSL [18.15 ± 245.7 copies/20µL; p=0.004]). In contrast, RNU6-1 levels were similar between brain metastases and GBM patients [325 ± 632 copies/20µL; p=0.573]. In addition, analyzing whether this small non-coding RNA could be a predictive marker of GBM by ROC curves analysis, we demonstrated that RNU6-1 was a robust diagnostic biomarker of GBM compared to subacute stroke [AUC=0.659; p=0.004], acute/subacute haemorrhage [AUC=0.724; p=0.006], acute demyelinating lesions [AUC=0.728; p=0.011] and PCNSL [AUC=0.814; p<0.001]; on the contrary, it did not allow differentiating GBM from brain metastases [AUC=0.552; p=0.575].

Conclusion

Our data suggest that RNU6-1 isolated in circulating exosomes could serve as a differential biomarker for GBM versus non-neoplastic brain lesions and PCNSL.(Grant: 42/2015 Dpto. de Salud Gobierno de Navarra)