Objective To investigate the relationships of serum microRNA-9-3p (miR-9-3p), microRNA-27b-3p (miR-27b-3p), with brain glioma (BG) grading and their impacts on prognosis.

Methods A total of 172 BG patients admitted in Suqian Hospital of Jiangsu Provincial People′s Hospital from May 2020 to May 2023 were enrolled and divided into high-grade group (grade Ⅲ to Ⅳ, n=101) and low-grade group (grade Ⅰ to Ⅱ, n=71) according to the World Health Organization (WHO) classification of central nervous system tumors. Additionally, 85 healthy individuals of the same age range during the same period were selected as control group. General information, serum levels of miR-9-3p, and miR-27b-3p were compared among the three groups. The correlations of pretreatment serum levels of miR-9-3p and miR-27b-3p with BG grading were analyzed. The prognosis of BG patients one year after treatment was recorded, and clinical data were compared between the poor prognosis group and the good prognosis group. Factors influencing poor prognosis in BG patients as well as the value of pretreatment serum miR-9-3p and miR-27b-3p in predicting poor prognosis were analyzed. Conventional factors influencing poor prognosis were used to establish a conventional prediction model, and model combined with pretreatment serum miR-9-3p and miR-27b-3p levels was used as a new prediction model. The value of these two models in predicting poor prognosis in BG patients was compared.

Results Before treatment, serum levels of miR-9-3p and miR-27b-3p were lower in the high-grade group than those in the low-grade and control groups, and lower in the low-grade group than those in the control group (P < 0.05). Correlation analysis showed that pretreatment serum levels of miR-9-3p and miR-27b-3p were negatively correlated with BG grading (r=-0.573, P < 0.001; r=-0.498, P < 0.001), and serum miR-9-3p level was positively correlated with miR-27b-3p level in BG patients (r=0.509, P < 0.05). Before treatment, the poor prognosis group had a higher proportion of patients with WHO grade Ⅲ to Ⅳ, low differentiation, Karnofsky Performance Scale (KPS) < 70, and higher serum interleukin-17 (IL-17) levels compared to the good prognosis group, while serum levels of miR-9-3p and miR-27b-3p were lower (P < 0.05). Logistic regression analysis showed that pretreatment WHO grading, differentiation, KPS, serum IL-17, miR-9-3p, and miR-27b-3p levels were all influencing factors of poor prognosis in BG patients (P < 0.05). Receiver operating characteristic (ROC) curves showed that the areas under the curve (AUCs) for pretreatment serum miR-9-3p and miR-27b-3p alone in predicting poor prognosis in BG patients were 0.714 and 0.720, respectively, with no statistically significant difference (Z=0.086, P=0.413). The AUC of the conventional prediction model for predicting poor prognosis in BG patients was 0.829, and the AUC of the new prediction model was 0.926 (P < 0.05).

Conclusion Serum levels of miR-9-3p and miR-27b-3p are negatively correlated with BG grading and are factors influencing poor prognosis, demonstrating certain value in predicting poor prognosis.