Prognostic Factors for Survival in Glioblastoma: A Retrospective Analysis Focused on the Role of Hemoglobin.

Background: Although several prognostic factors for survival have been identified in glioblastoma, there are numerous other potential markers (such as hemoglobin) whose role has not yet been confirmed. The aim of this study was to evaluate a wide range of potential prognostic factors, including HIF-1α and hemoglobin levels, for survival in glioblastoma. A secondary aim was to determine whether hemoglobin levels were associated with HIF-1α expression. Methods: A retrospective study of 136 patients treated for glioblastoma at our institution between 2012 and 2021 was performed. Cox univariate and multivariate analyses were carried out. Kaplan-Meier survival curves were generated. In addition, bivariate non-parametric correlation analyses were performed for key variables. Results: Median survival was 11.9 months (range: 0-119.4). According to the univariate analysis, 13 variables were significantly associated with survival: age, performance status, extent of surgery, tumor depth, tumor size, epilepsy, postoperative chemoradiotherapy, IDH mutations, CD44, HIF-1α, HIF-1ß, vimentin, and PDFGR. According to the multivariate regression analysis, only four variables remained significantly associated with survival: age, extent of surgery, epilepsy, and HIF-1α expression. No significant association was observed between hemoglobin levels (low <120 g/L in females or <140 g/L in males vs. high ≥120 or ≥140 g/L) and survival or HIF-1α/HIF-1ß expression. Conclusions: In this retrospective study of patients with glioblastoma, four variables-age, extent of surgery, HIF-1α expression, and epilepsy-were significant prognostic factors for survival. Hemoglobin levels were not significantly associated with survival or HIF-1α expression. Although hypoxia is a well-recognized component of the glioblastoma microenvironment, more research is needed to understand the pathogenesis of onset tumor hypoxia and treatment implication.

Expression of STAT3 and hypoxia markers in long-term surviving malignant glioma patients.

BACKGROUND: Glioblastoma is a malignant and aggressive type of central nevous system malignancy characterized by many distinct biological features including extensive hypoxia. Hypoxia in glioblatoma associates with complex signaling patterns including activation of several pathways such as MAPK, PI3K-AKT/mTOR and IL-6/JAK/STAT3 with the master regulator HIF-1, which in turn drive particular tumor behaviors determining, in the end, treatment outcomes and patients fate. Thus, the present study was designed to investigate the expression of selected hypoxia related factors including STAT3 in a small set of long-term surviving glioma patients. METHODS: The expression of selected hypoxia related factors including STAT3 was evaluated in a time series of formalin fixed paraffin embedded and cryopreserved glioma samples from repeatedly resected patients. In addition, comparative studies were also conducted on primary glioma cells derived from original patient samples, stabilized glioma cell lines and tumor-xenograft mice model. Obtained data were correlated with clinical findings too. RESULTS: Glioblastoma samples of the analyzed patients displayed heterogeneity in the expression of hypoxia- related and EMT markers with most interesting trend being observed in pSTAT3. This heterogeneity was subsequently confirmed in other employed models (primocultures derived from glioblastoma tissue resections, cryopreserved tumor specimens, stabilized glioblastoma cell line in vitro and in vivo) and concerned, in particular, STAT3 expression which remained stable. In addition, subsequent studies on the role of STAT3 in the context of glioblastoma hypoxia demonstrated opposing effects of its deletion on cell viability as well as the expression of hypoxia and EMT markers. CONCLUSIONS: Our results suport the importance of STAT3 expression and activity in the context of hypoxia in malignant glioblastoma long-term surviving glioma patients while emphasizing heterogeneity of biological outcomes in varying employed tumor models.

Diagnostic Utility of Immunohistochemical Detection of MEOX2, SOX11, INSM1 and EGFR in Gliomas.

Histological identification of dispersed glioma cells in small biopsies can be challenging, especially in tumours lacking the IDH1 R132H mutation or alterations in TP53. We postulated that immunohistochemical detection of proteins expressed preferentially in gliomas (EGFR, MEOX2, CD34) or during embryonal development (SOX11, INSM1) can be used to distinguish reactive gliosis from glioma. Tissue microarrays of 46 reactive glioses, 81 glioblastomas, 34 IDH1-mutant diffuse gliomas, and 23 gliomas of other types were analysed. Glial neoplasms were significantly more often (p < 0.001, χ2) positive for EGFR (34.1% vs. 0%), MEOX2 (49.3% vs. 2.3%), SOX11 (70.5% vs. 20.4%), and INSM1 (65.4% vs. 2.3%). In 94.3% (66/70) of the glioblastomas, the expression of at least two markers was observed, while no reactive gliosis showed coexpression of any of the proteins. Compared to IDH1-mutant tumours, glioblastomas showed significantly higher expression of EGFR, MEOX2, and CD34 and significantly lower positivity for SOX11. Non-diffuse gliomas were only rarely positive for any of the five markers tested. Our results indicate that immunohistochemical detection of EGFR, MEOX2, SOX11, and INSM1 can be useful for detection of glioblastoma cells in limited histological samples, especially when used in combination.

Small RNA Sequencing Identifies PIWI-Interacting RNAs Deregulated in Glioblastoma-piR-9491 and piR-12488 Reduce Tumor Cell Colonies In Vitro.

Glioblastoma (GBM) is the most frequently occurring primary malignant brain tumor of astrocytic origin. To change poor prognosis, it is necessary to deeply understand the molecular mechanisms of gliomagenesis and identify new potential biomarkers and therapeutic targets. PIWI-interacting RNAs (piRNAs) help in maintaining genome stability, and their deregulation has already been observed in many tumors. Recent studies suggest that these molecules could also play an important role in the glioma biology. To determine GBM-associated piRNAs, we performed small RNA sequencing analysis in the discovery set of 19 GBM and 11 non-tumor brain samples followed by TaqMan qRT-PCR analyses in the independent set of 77 GBM and 23 non-tumor patients. Obtained data were subsequently bioinformatically analyzed. Small RNA sequencing revealed 58 significantly deregulated piRNA molecules in GBM samples in comparison with non-tumor brain tissues. Deregulation of piR-1849, piR-9491, piR-12487, and piR-12488 was successfully confirmed in the independent groups of patients and controls (all p < 0.0001), and piR-9491 and piR-12488 reduced GBM cells' ability to form colonies in vitro. In addition, piR-23231 was significantly associated with the overall survival of the GBM patients treated with Stupp regimen (p = 0.007). Our results suggest that piRNAs could be a novel promising diagnostic and prognostic biomarker in GBM potentially playing important roles in gliomagenesis.