BACKGROUND:Anaplastic oligodendrogliomas IDH-mutant and 1p/19q codeleted (AO) occasionally have a poor outcome. Herein we aimed at analyzing their characteristics. METHODS:We retrospectively analyzed the characteristics of 44 AO patients with a cancer-specific survival <5 years (short-term survivors, STS) and compared them with those of 146 AO patients with a survival ≥5 years (classical survivors, CS) included in the POLA network. RESULTS:Compared to CS, STS were older (P = .0001), less frequently presented with isolated seizures (P < .0001), more frequently presented with cognitive dysfunction (P < .0001), had larger tumors (P = .= .003), a higher proliferative index (P = .= .0003), and a higher number of chromosomal arm abnormalities (P = .= .02). Regarding treatment, STS less frequently underwent a surgical resection than CS (P = .= .0001) and were more frequently treated with chemotherapy alone (P = .= .009) or with radiotherapy plus temozolomide (P = .= .05). Characteristics independently associated with STS in multivariate analysis were cognitive dysfunction, a number of mitosis > 8, and the absence of tumor resection. Based on cognitive dysfunction, type of surgery, and number of mitosis, patients could be classified into groups of standard (18%) and high (62%) risk of <5 year survival. CONCLUSION:The present study suggests that although STS poor outcome appears to largely result from a more advanced disease at diagnosis, surgical resection may be particularly important in this population.

IDH (isocitrate dehydrogenase) mutation, hypoxia, and neo-angiogenesis, three hallmarks of diffuse gliomas, modulate the expression of small non-coding RNAs (miRNA). In this paper, we tested whether pro-angiogenic and/or pro-hypoxic miRNAs could be used to monitor patients with glioma. The miRNAs were extracted from tumoral surgical specimens embedded in the paraffin of 97 patients with diffuse gliomas and, for 7 patients, from a blood sample too. The expression of 10 pro-angiogenic and/or pro-hypoxic miRNAs was assayed by qRT-PCR and normalized to the miRNA expression of non-tumoral brain tissues. We confirmed in vitro that IDH in hypoxia (1% O2, 24 h) alters pro-angiogenic and/or pro-hypoxic miRNA expression in HBT-14 (U-87 MG) cells. Then, we reported that the expression of these miRNAs is (i) strongly affected in patients with glioma compared to that in a non-tumoral brain; (ii) correlated with the histology/grade of glioma according to the 2016 WHO classification; and (iii) predicts the overall and/or progression-free survival of patients with glioma in univariate but not in a multivariate analysis after adjusting for sex, age at diagnosis, and WHO classification. Finally, the expression of miRNAs was found to be the same between the plasma and glial tumor of the same patient. This study highlights a panel of seven pro-angiogenic and/or pro-hypoxic miRNAs as a potential tool for monitoring patients with glioma.

We aimed to describe the characteristics, patterns of care and predictive geriatric factors of elderly patients with IDH-mutant (IDHm) high-grade gliomas (HGG) included in the French POLA network, dedicated to HGG (including 68% of IDHm HGG). For IDHm HGG patients over the age of 70 years, geriatric features were collected: G8 score items (appetite, weight loss, mobility, neuropsychological disorders, body mass index, medications, self-rated health, age), Activities and Instrumental Activities of Daily Living (ADL, IADL) scores, Charlson’s comorbidity Index (CCI) and biological markers. Out of the 1433 HGG patients included in the POLA Network, 119 (8.3%) occurred in patients ≥ 70 years. Among them, 39 presented with IDHm HGG. Of these 39 patients, estimated G8 score was ≤ 14/17 for 16 patients (64%), ADL score was < 6 for 33.3%, IADL score was < 4 for 47% and CCI was ≥ 5 for 72%. Regarding treatment feasibility, 6 of the 19 patients treated by temozolomide prematurely discontinued chemotherapy including 2 for toxicity and 4 for progression. Five of the 10 patients treated by PCV prematurely discontinued chemotherapy, all for toxicity. In multivariate analysis, loss of mobility (p=0.018; p=0.008), severe neuropsychological disorders (p=0.005; p=0.047), body mass index < 21 kg/m2 (p=0.002; p=0.006) and ADL score < 6 (p=0.002; p=0.01) were significantly predictive of poor PFS and OS. Then we generated a specific brain geriatric score including these four items with a sensibility, specificity and AUC for long term survivor (≥ 48 months) of 100%, 83% and 0.948 respectively. Using a cutoff of < 10/13, this score was significantly correlated to PFS and OS (p< 0.001 both). In conclusion, geriatric predictive factors may contribute to the elderly management improvement: the brain geriatric score must now be validated in a prospective independent cohort including IDHm and IDHwt elderly patients.

Ras association domain family (RASSF)/Hippo pathway alterations are poorly characterized in diffuse gliomas. We assayed promoter methylation of LATS1/2, MST1(STK4)/MST2(STK3), RASSF1, RASSF2, Nore1A/RASSF5, RASSF6, and RASSF10 genes in 133 diffuse gliomas. The RASSF/Hippo pathway was highly silenced in gliomas, particularly RASSF1A (79.4%) and LATS2 (35.9%). The most frequent combination of promoter hypermethylation of one RASSF gene and one Hippo pathway member's gene was RASSF1/LATS2-coupled hypermethylation [n = 44 (33.08 /0)]. Hypermethylated profiles were related to IDH mutation, yet not randomly in IDH-mutated gliomas, because LATS2 promoter hypermethylation was more frequent in oligodendroglioma than in astrocytoma. RASSFI and LATS2 promoter hypermethylation predicted a longer overall survival (OS). Considering hypermethylation of these two promoters, Cox proportional hazard regression analysis categorized the patients into three prognostic groups: i) high risk of death (n = 24; both RASSF1 and LATS2 unmethylated promoters; median OS, 13 months); ii) intermediate risk of death (n = 65; RASSFI or LATS2 hypermethylated promoter; median OS, 50.5 months; HR = 3.3; 95% CI, 1.6-6.4; P = 0.001); and iii) tow risk of death (n = 44; both RASSFI and LATS2 hypermethylated promoters; median OS, 119 months; HR = 75.1; 95% CI, 3.3-15.1; P = 0.001). We have thus highlighted a simple two-gene (RASSF1/LATS2) methylation signature as a tool to stratify different prognostic groups of patients with diffuse glioma, adding further prognostic information within the IDH-mutated group.

BACKGROUND1p/19q-codeleted anaplastic gliomas have variable clinical behavior. We have recently shown that the common 9p21.3 allelic loss is an independent prognostic factor in this tumor type. The aim of this study is to identify less frequent genomic copy number variations (CNVs) with clinical importance that may shed light on molecular oncogenesis of this tumor type.MATERIALS AND METHODSA cohort of 197 patients with anaplastic oligodendroglioma was collected as part of the French POLA network. Clinical, pathological, and molecular information was recorded. CNV analysis was performed using single-nucleotide polymorphism arrays. Computational biology and feature selection based on the random forests method were used to identify CNV events associated with overall survival and other clinical-pathological variables.RESULTSRecurrent chromosomal events were identified in chromosomes 4, 9, and 11. Forty-six focal amplification events and 22 focal deletion events were identified. Twenty-four focal CNV areas were associated with survival, and five of them were significantly associated with survival after multivariable analysis. Nine out of 24 CNV events were validated using an external cohort of The Cancer Genome Atlas. Five of the validated events contain a cancer-related gene or microRNA: CDKN2A deletion, SS18L1 amplification, RHOA/MIR191 copy-neutral loss of heterozygosity, FGFR3 amplification, and ARNT amplification. The CNV profile contributes to better survival prediction compared with clinical-based risk assessment.CONCLUSIONSeveral recurrent CNV events, detected in anaplastic oligodendroglioma, enable better survival prediction. More importantly, they help in identifying potential genes for understanding oncogenesis and for personalized therapy.IMPLICATIONS FOR PRACTICEGenomic analysis of 197 anaplastic oligodendroglioma tumors reveals recurrent somatic copy number variation areas that may help in understanding oncogenesis and target identification for precision medicine. A machine learning multivariable model built using this genomic information enables better survival prediction.

Hypoxia in gliomas is associated with tumor resistance to radio- and chemotherapy. However, positron emission tomography (PET) imaging of hypoxia remains challenging, and the validation of biological markers is, therefore, of great importance. We investigated the relationship between uptake of the PET hypoxia tracer [18F]-FMISO and other markers of hypoxia and angiogenesis and with patient survival.

Confronted with brain metastases (BM), pathologists aim to rule out a primary central nervous system (CNS) tumor and to identify or verify the primary tumor site to guide the clinician to specific therapies. Apart from morphological features, ancillary immunohistochemical analysis is the most effective tool for characterizing a metastatic neoplasm of unknown origin. A limited array of antibodies is used, taking into account relevant clinical information and the known brain tropism of lung cancer, breast cancer and melanoma. Recently, targeted therapies have enriched the therapeutic arsenal, in particular for patients with non-small cell lung cancer or melanoma and for patients carrying molecular anomalies. These therapies can lead to a substantial tumor response, brain metastases included, which justifies rapid determination of a molecular profile. To combine different tools and provide timely results, good tumor sample management and careful attention at the pre-analytical phase are critical. Appropriate strategies for molecular and immunohistochemical analysis are needed to identify theranostic markers. This article aims to review the anatomopathological diagnostic approach for BM in the age of targeted therapies.

BACKGROUND: High grade gliomas in children and young adults are clinically and genetically different from their adult counterparts. Recurrent mutations of the H3F3A gene coding for the H3.3 isoform of the H3 histone have recently been described. These mutations have been shown to be exclusive to high grade gliomas, and the majority occurs in the pediatric setting. OBSERVATION: We report on 2 cases of confirmed mutations of the H3F3A gene at the K27M site. The first patient was a young adult with a left cerebellar mass. Imaging and pathological analysis were typical of glioblastoma. The second patient was a 14-year old boy with a left thalamic lesion, rather circumscribed on MRI but with features of malignity on spectroscopy and perfusion. Pathologic analysis showed a heterogeneous tumor with regions in favor of ganglioglioma, and regions with marked anisokaryosis and rare mitoses, a slightly elevated proliferation index and high p53 positivity. Due to the presence of a histone mutation, the tumor was classified as a high grade glioma. DISCUSSION: Pathological analysis of pediatric tumors can be a challenge. Nowadays, molecular biology is an important tool to help us classify these tumors. In our first patient, the presence of a histone mutation confirmed the initial diagnosis. In the second patient however, the discovery of a histone mutation finally permitted us to classify this pathologically challenging tumor as a high grade glioma. CONCLUSION: When confronted with childhood or young adult brain tumors, it is important to search for histone mutations, since their presence can influence the final diagnosis.