Article
Circulating biomarker for glioblastoma and primary central nervous system lymphoma – next-generation sequencing of small noncoding RNA
Zirkulierender Biomarker für Glioblastome und primäre Lymphome des zentralen Nervensystems – next generation sequencing von kleinen nicht-kodierenden RNA
Search Medline for
Authors
Published: | June 26, 2020 |
---|
Outline
Text
Objective: Glioblastoma (GBM) and Primary Central Nervous System Lymphoma (PCNSL) are common intracranial malignant tumors. The treatment of these tumors is quite different, accurate preoperative differentiation is the essential of clinical relevance.However, they sometimes present similar radiological findings and diagnoses could be difficult without surgical biopsy.For improving the current management, development of non-invasive biomarkers are desired. In this study, we explored the differently expressed circulating small noncoding RNA (sncRNA) in serum for specific diagnostic tool of GBM and PCNSL.
Methods: Serum samples were obtained from three groups: 1) GBM patients (N=26), 2) PCNSL patients (N=14), 3) healthy control (N=114). Written informed consent to participate in the study were obtained. The total small RNAs were extracted from serum by using the QIAGEN® kit. The whole expression profiles of serum sncRNAs were measured using Next-Generation Sequencing System (Thermo Fisher, Ion S5). We analyzed serum levels of sncRNAs (15-55 nt) in each serum samples. The difference of sncRNAs expression profile among three groups were compared. Data analysis was performed by logistic regression analysis followed by leave-one-out cross-validation (LOOCV). The accuracy of diagnostic models of sncRNAs combination were evaluated by receiver operating characteristic (ROC) analysis.
Results: We created the combination models using three sncRNA in each models based on the logistic regression analysis. The model 1 (based on sncRNA-X1, X2 and X3) enabled to differentiate GBM patients form healthy control with a sensitivity of 92.3% and a specificity of 99.2% (AUC : 0.993). The model 2 (based on sncRNA-Y1, Y2 and Y3) enabled to differentiate PCNSL patients form healthy control with a sensitivity of 100% and a specificity of 93.9% (AUC: 0.984). The model 3 (based on sncRNA-Z1, Z2 and Z3) enabled to differentiate GBM patients form PCNSL patients with a sensitivity of 92.3% and a specificity of 78.6% (AUC: 0.920).
Conclusion: We found three diagnostic models of serum sncRNAs as non-invasive biomarkers potentially useful for detection of GBM and PCNSL from healthy control, and for differentiation GBM from PCNSL.