This study determined the expression of five novel biomarker candidates in IDH wild-type glioblastoma (GBM) tissues compared to non-malign brain parenchyma, as well as their prognostic relevance for the GBM patients\' outcomes. The markers were analysed by immunohistochemistry in tumour tissues (n = 186) and healthy brain tissues (n = 54). The association with the patients\' overall survival (OS) and progression-free survival (PFS) was assessed by Kaplan-Meier and log-rank test. The prognostic value of the markers was determined using multivariate Cox proportional hazard models. AGTRAP, DIVERSIN, cytoplasmic NEDD8 (NEDD8c) and RRM1 were significantly overexpressed in tumour tissues compared to the healthy brain, while the opposite was observed for ALKBH3. AGTRAP, ALKBH3, NEDD8c and RRM1 were significantly associated with OS in univariate analysis. AGTRAP and RRM1 were also independent prognostic factors for OS in multivariate analysis. For PFS, only AGTRAP and NEDD8c reached significance in univariate analysis. Additionally, AGTRAP was an independent prognostic factor for PFS in multivariate models. Finally, combined analysis of the markers enhanced their prognostic accuracy. The combination AGTRAP/ALKBH3 had the strongest prognostic value for the OS of GBM patients. These findings contribute to a better understanding of the GBM pathophysiology and may help identify novel therapeutic targets in this type of cancer.

Malignant mesothelioma (MM) is a highly aggressive tumor associated with asbestos exposure. The identification of a marker specific for MM may be of considerable value for the early detection of this tumor and may be used in particular to screen groups with a history of asbestos exposure. The aim of this study was to evaluate serum soluble mesothelin-related peptide (SMRP) levels as a diagnostic marker for MM and investigate whether its diagnostic value is enhanced by combination with other biomarkers. Serum SMRP levels were measured using a quantitative enzyme-linked immunosorbent assay in 96 patients with MM, 55 patients with lung cancer and 39 individuals with a history of asbestos exposure. Receiver operating characteristic curves were constructed for performance evaluation. Stepwise logistic regression analysis was used to select marker combinations (MCs). Serum SMRP levels in patients with MM were significantly higher compared to those in the other groups (P<0.001). The sensitivity of SMRP levels in diagnosing MM was 56% and its specificity for MM vs. lung cancer and individuals with asbestos exposure was 87 and 92%, respectively. The area under the curve (AUC) was 0.76 [95% confidence interval (CI): 0.68-0.83] for the differentiation between MM and lung cancer and 0.78 (95% CI: 0.71-0.86) for the differentiation between MM and individuals with asbestos exposure. For the MC of presence of effusion, SMRP and carcinoembryonic antigen (CEA) levels, the AUC for the differentiation between MM and lung cancer (0.92; 95% CI: 0.88-0.97) and the differentiation between MM and individuals with asbestos exposure (0.93; 95% CI: 0.87-1.0) was significantly higher compared to that for SMRP alone (P=0.0001 and 0.0058, respectively). While the specificity of this MC was comparable to SMRP alone, its sensitivity was ∼20% higher compared to that of SMRP alone. Therefore, combining SMRP and CEA improves the diagnostic performance of SMRP alone. A combination of serum biomarkers, including SMRP, may facilitate the non-invasive diagnosis of MM.