Glioma, the most common primary malignant brain tumor arising from the brain orspinal cord tissue, has a tendency to invade the surrounding brain tissue. Gliomas represent approximately one-third of all intrinsic neoplasms of the central nervous system in both adults and children. The World Health Organization (WHO) classification divides gliomas into grades I through IV, and these values correspond to increasing levels of malignancy, including well-differentiated low grade, astrocytomas and glioblastoma multiforme [1]. Glioma has a poor prognosis due to the characteristic progressive overgrowth and diffuse invasion. Although the WHO classification can serve as a criterion to predict the patient clinical outcomes, several recent studies have indicated that this criteria alone may not be sufficient to estimate patient prognosis [2, 3]. Therefore, investigating molecular mechanisms of gliomas may produce better prognostic markers to anticipate patient survival.

MicroRNAs (miRNAs) are a class of small, evolutionarily conserved, short non-coding endogenous RNA molecules that regulate gene expression at the post-transcriptional level [4]. Primary transcripts of miRNAs (pri-miRNA) are generated by RNA polymerase II and then they are sequentially processed by RNase III enzymes, Drosha and Dicer, to first produce pre-miRNAs and finally mature miRNAs. MiRNAs act by base-pairing with their target mRNAs according to the degree of complementarities with their target 3′ untranslated regions (UTRs) leading to their translational regulation and/or direct cleavage [5, 6]. MiRNAs have been demonstrated to play normal physiologic roles in cell proliferation and differentiation, epithelial-mesenchymal transition, apoptosis and metabolism [7, 8]. In addition, studies suggest that miRNAs also play important roles in tumorigenesis and tumor progression, acting as oncogenes or tumor suppressors depending on their target genes. Finally, some miRNAs may be markers for cancer diagnosis and prognosis [9, 10].

In the present study, we focus on miR-21, which has been demonstrated to act as either an oncogenic miRNA or an anti-oncomiR (miRNA that negatively regulates oncogenes) in various human malignancies including glioma [11–15]. However, to our knowledge, no correlation between miR-21 and patient prognosis has not been addressed in glioma at this time. To investigate this problem, miR-21 expression in human gliomas and nonneoplastic brain tissues was measured using real-time quantitative RT-PCR assay. The association of miR-21 with clinicopathological factors or glioma patient prognosis was also statistically analyzed.

Glioma is one of the most common malignant tumors with a low survival rate, especially in the face of high WHO tumor grades (III, IV). Accurately predicting postoperative outcomes in glioma patients and choosing the correct adjuvant therapy is a present and persistent problem. At this time, traditional pathologic variables are used to make such predictions. Currently, WHO glioma grades are the gold standard for determining glioma patient prognosis, but this information is less useful for individual patients and often fails to discriminate the biological nature of a large number of gliomas. Therefore patients at the same disease stage often have significant discrepancies in survival times. Until now, there are several studies showed that single molecule could be the relapse and prognosis indicator [17, 18]. Thus, discovering more molecular markers associated with the relapse and prognosis of glioma is essential.

To date, studies have shown that miRNA expression is correlated with clinical and biological features of tumors, including tissue type, differentiation and aggressiveness, which can be a potential biomarker for therapy and prognosis [10, 12–14]. MiR-21 is one of the first discovered miRNAs that is known to be widespread in human tissues. Up-regulation of miR-21 has been observed in numerous human cancers [13, 15]. Several publications confirm that miR-21 can promote cell proliferation and invasion and that it inhibits cell apoptosis [19–21]. Inhibition of miR-21 could upregulate the expression of a miRNA-target tumor suppressor gene such as protein tyrosine phosphatase (PTEN) and B cell translocation gene 2 (BTG2), and thereby decreases tumor progression [22–24]. However, to date there has been no research on the prognostic role of miR-21 in glioma. New molecular prognostic factors for glioma may contribute to a better assessment of survival probability and the tailoring of treatments for individual patients.

We investigated miR-21 expression in glioma and determined if this marker could predict disease relapse and patient outcomes. We observed that increased miR-21 expression was evident in human glioma tissues compared with non-neoplastic brain tissues. High miR-21 expression in glioma tissues was significantly correlated with aggressive clinicopathological features as well. Kaplan-Meier analysis revealed that glioma patients with high miR-21 expression had poorer overall survival. Moreover, multivariate analysis revealed that high miR-21 expression was a marker of worse patient outcomes which was independent of known clinical prognostic indicators such as KPS and WHO grades. Gliomas with grade I have different genetic abnormality compared with other grade gliomas. In this study, we divided all the gliomas into two groups: low grade glioma (WHO grade I and II) and high grade glioma (WHO III and IV) and performed multivariate analysis. The result also indicated that miR-21 could be the marker of poor outcomes in these two groups (Relative Risk is 4.947, P value < 0.001, 95% confidence interval is 3.12-7.843). All these data indicate that increased expression of miR-21 is correlated with poor glioma patient outcomes. Thus, miR-21 expression may serve as a potential biomarker for overall survival prediction in this patient population.

Our results revealed that miRNA-21 was overexpressed in glioma tumors and that high expression level of miRNA-21 was strongly associated with poor prognosis, but further prospective studies are needed to determine the actual clinical relevance of this observation. Our data suggest that miR-21 may be an appropriate prognostic marker for evaluating advanced tumors with high pathological grades and that this marker may assist with the appropriate choice of adjuvant chemotherapy as well as serve as a potential therapeutic target in molecular therapy for glioma.