The prognostic impact of TERT mutations has been controversial in IDH-wild tumors, particularly in glioblastomas (GBM). The controversy may be attributable to presence of potential confounding factors such as MGMT methylation status or patients’ treatment. This study aimed to evaluate the impact of TERT status on patient outcome in association with various factors in a large series of adult diffuse gliomas. We analyzed a total of 951 adult diffuse gliomas from two cohorts (Cohort 1, n = 758; Cohort 2, n = 193) for IDH1/2, 1p/19q, and TERT promoter status. The combined IDH/TERT classification divided Cohort 1 into four molecular groups with distinct outcomes. The overall survival (OS) was the shortest in IDH wild-type/TERT mutated groups, which mostly consisted of GBMs (P < 0.0001). To investigate the association between TERT mutations and MGMT methylation on survival of patients with GBM, samples from a combined cohort of 453 IDH-wild-type GBM cases treated with radiation and temozolomide were analyzed. A multivariate Cox regression model revealed that the interaction between TERT and MGMT was significant for OS (P = 0.0064). Compared with TERT mutant-MGMT unmethylated GBMs, the hazard ratio (HR) for OS incorporating the interaction was the lowest in the TERT mutant-MGMT methylated GBM (HR, 0.266), followed by the TERT wild-type-MGMT methylated (HR, 0.317) and the TERT wild-type-MGMT unmethylated GBMs (HR, 0.542). Thus, patients with TERT mutant-MGMT unmethylated GBM have the poorest prognosis. Our findings suggest that a combination of IDH, TERT, and MGMT refines the classification of grade II-IV diffuse gliomas.Electronic supplementary materialThe online version of this article (doi:10.1186/s40478-016-0351-2) contains supplementary material, which is available to authorized users.

Object The precise natural history of incidentally discovered meningiomas (IDMs) remains unknown. It has been reported that for symptomatic meningiomas, tumor location can be used to predict growth. As to whether the same is true for IDMs has not been reported. This study aims to answer this question and provide biological evidence for this assumption by extending the study to involve symptomatic cases. Methods A total of 113 IDMs were analyzed by fine volumetry. A comparison of growth rates and patterns between skull base and non–skull base IDMs was made. Subsequently, materials obtained from 210 patients with symptomatic meningiomas who were treated in the authors' hospital during the same period were included for a biological comparison between skull base and non–skull base tumors using the MIB-1 index. Results The 110 patients with IDMs included 93 females and 17 males, with a mean follow-up period of 46.9 months. There were 38 skull base (34%) and 75 non–skull base (66%) meningiomas. Forty-two (37%) did not exhibit growth of more than 15% of the volume, whereas 71 (63%) showed growth. Only 15 (39.5%) of 38 skull base meningiomas showed growth, whereas 56 (74.7%) of 75 non–skull base meningiomas showed growth (p = 0.0004). In the 71 IDMs (15 skull base and 56 non–skull base), there was no statistical difference between the 2 groups in terms of mean age, sex, follow-up period, or initial tumor volume. However, the percentage of growth (p = 0.002) was significantly lower and the doubling time (p = 0.008) was significantly higher in the skull base than in the non–skull base tumor group. In subsequently analyzed materials from 94 skull base and 116 non–skull base symptomatic meningiomas, the mean MIB-1 index for skull base tumors was markedly low (2.09%), compared with that for non–skull base tumors (2.74%; p = 0.013). Conclusions Skull base IDMs tend not to grow, which is different from non–skull base tumors. Even when IDMs grow, the rate of growth is significantly lower than that of non–skull base tumors. The same conclusion with regard to biological behavior was confirmed in symptomatic cases based on MIB-1 index analyses. The authors' findings may impact the understanding of the natural history of IDMs, as well as strategies for management and treatment of IDMs and symptomatic meningiomas.

Molecular biological characterization of tumors has become a pivotal procedure for glioma patient care. The aim of this study is to build conventional MRI-based radiomics model to predict genetic alterations within grade II/III gliomas attempting to implement lesion location information in the model to improve diagnostic accuracy. One-hundred and ninety-nine grade II/III gliomas patients were enrolled. Three molecular subtypes were identified: IDH1/2-mutant, IDH1/2-mutant with TERT promoter mutation, and IDH-wild type. A total of 109 radiomics features from 169 MRI datasets and location information from 199 datasets were extracted. Prediction modeling for genetic alteration was trained via LASSO regression for 111 datasets and validated by the remaining 58 datasets. IDH mutation was detected with an accuracy of 0.82 for the training set and 0.83 for the validation set without lesion location information. Diagnostic accuracy improved to 0.85 for the training set and 0.87 for the validation set when lesion location information was implemented. Diagnostic accuracy for predicting 3 molecular subtypes of grade II/III gliomas was 0.74 for the training set and 0.56 for the validation set with lesion location information implemented. Conventional MRI-based radiomics is one of the most promising strategies that may lead to a non-invasive diagnostic technique for molecular characterization of grade II/III gliomas.

No abstract

Identification of genotypes is crucial for treatment of glioma. Here, we developed a method to predict tumor genotypes using a pretrained convolutional neural network (CNN) from magnetic resonance (MR) images and compared the accuracy to that of a diagnosis based on conventional radiomic features and patient age. Multisite preoperative MR images of 164 patients with grade II/III glioma were grouped by IDH and TERT promoter (pTERT) mutations as follows: (1) IDH wild type, (2) IDH and pTERT co-mutations, (3) IDH mutant and pTERT wild type. We applied a CNN (AlexNet) to four types of MR sequence and obtained the CNN texture features to classify the groups with a linear support vector machine. The classification was also performed using conventional radiomic features and/or patient age. Using all features, we succeeded in classifying patients with an accuracy of 63.1%, which was significantly higher than the accuracy obtained from using either the radiomic features or patient age alone. In particular, prediction of the pTERT mutation was significantly improved by the CNN texture features. In conclusion, the pretrained CNN texture features capture the information of IDH and TERT genotypes in grade II/III gliomas better than the conventional radiomic features.