COT-11 Relationship between preoperative liquid biopsy and prognosis of glioblastoma -Next Generation Sequencing of small noncoding RNA-

BACKGROUND: Non-invasive biomarkers are required in clinical practice of glioblastoma (GBM). We have previously reported the liquid biopsy for differentiating glioblastoma, central nervous system primary lymphoma and healthy control. In this study, we analyzed the relationship between the preoperative serum expression of circulating small non-coding RNAs and the prognosis of GBM patients. METHODS: Preoperative blood samples of GBM, IDH-wildtype patients (N=26) were centrifuged and collected all small RNAs in serum. The expression of small non-coding RNAs were analyzed using a next-generation sequencing system. The small non-coding RNAs that could predict short-term survivals in GBM patients were selected by the stepwise analysis. A diagnostic model was created using the combination of these RNAs and evaluated with ROC curve. RESULTS: GBM patients treated with adjuvant therapy of temozolomide and radiotherapy were divided into two groups: (1) a short-term survival group (N=11) with a survival time less than 15 months and (2) a long-term survival group (N=15) with a survival time more than 15 months. In the short-term survival group, the preoperative serum expression levels of small RNA-X and small RNA-Y were high, and the expression levels of small RNA-Z and small RNA-W were low. Using these four small non-coding RNAs, a prognostic model was created. The model was able to predict the short-term survival group of GBM patients with a sensitivity of 90.9% and specificity of 93.3% (AUC: 0.969). CONCLUSION: The prognostic model developed with preoperative small non-coding RNA in GBM patients may be useful for estimating the survival of GBM patients treated with adjuvant therapy of temozolomide and radiotherapy.