Glioblastomas (GBM) are devastating tumors in which there has been little clinical improvement in the last decades. New molecularly directed therapies are under development. EGFR is one of the most promising targets, as this receptor is mutated and/or overexpressed in nearly half of the GBMs. However, the results obtained with first-generation tyrosine-kinase inhibitors have been disappointing with no clear predictive markers of tumor response. Here, we have tested the antitumoral efficacy of a second-generation inhibitor, dacomitinib (PF299804, Pfizer), that binds in an irreversible way to the receptor. Our results confirm that dacomitinib has an effect on cell viability, self-renewal, and proliferation in EGFR-amplified AE EGFRvIII GBM cells. Moreover, systemic administration of dacomitinib strongly impaired the in vivo tumor growth rate of these EGFR-amplified cell lines, with a decrease in the expression of stem cell-related markers. However, continuous administration of the compound was required to maintain the antitumor effect. The data presented here confirm that dacomitinib clearly affects receptor signaling in vivo and that its strong antitumoral effect is independent of the presence of mutant receptor isoforms although it could be affected by the PTEN status (as it is less effective in a PTEN-deleted GBM line). Dacomitinib is being tested in second line for EGFR-amplified GBMs. We hope that our results could help to select retrospectively molecular determinants of this response and to implement future trials with dacomitinib (alone or in combination with other inhibitors) in newly diagnosed GBMs. Mol Cancer Ther; 14(7); 1548-58. Ó2015 AACR.

Background Giant cell glioblastoma (gcGBM) is a rare morphological variant of IDH-wildtype (IDHwt) GBM that occurs in young adults and have a slightly better prognosis than “classic” IDHwt GBM. Methods We studied 36 GBMs, 14 with a histopathological diagnosis of gcGBM and 22 with a giant cell component. We analyzed the genetic profile of the most frequently mutated genes in gliomas and assessed the tumor mutation load (TML) by gene-targeted next-generation sequencing. We validated our findings using The Cancer Genome Atlas (TCGA) data. Results p53 was altered by gene mutation or protein overexpression in all cases, while driver IDH1, IDH2, BRAF, or H3F3A mutations were infrequent or absent. Compared to IDHwt GBMs, gcGBMs had a significant higher frequency of TP53, ATRX, RB1, and NF1 mutations, while lower frequency of EGFR amplification, CDKN2A deletion, and TERT promoter mutation. Almost all tumors had low TML values. The high TML observed in only 2 tumors was consistent with POLE and MSH2 mutations. In the histopathological review of TCGA IDHwt, TP53-mutant tumors identified giant cells in 37% of the cases. Considering our series and that of the TCGA, patients with TP53-mutant gcGBMs had better overall survival than those with TP53wt GBMs (log-rank test, P < .002). Conclusions gcGBMs have molecular features that contrast to “classic” IDHwt GBMs: unusually frequent ATRX mutations and few EGFR amplifications and CDKN2A deletions, especially in tumors with a high number of giant cells. TML is frequently low, although exceptional high TML suggests a potential for immune checkpoint therapy in some cases, which may be relevant for personalized medicine.

Glioblastoma (GBM) is the most common malignant adult primary brain tumor. Despite its high lethality, a small proportion of patients have a relatively long overall survival (OS). Here we report a study of a series of 74 GBM samples from 29 long-term survivors ([LTS] OS ≥36 months) and 45 non-LTS. Using next-generation sequencing, we analyzed genetic alterations in the genes most frequently altered in gliomas. Approximately 20% of LTS had a mutation in the IDH1 or IDH2 (IDH) genes, denoting the relevance of this molecular prognostic factor. A new molecular group of GBMs harbored alterations in ATRX or DAXX genes in the absence of driver IDH or H3F3A mutations. These patients tended to have a slightly better prognosis, to be younger at diagnosis, and to present frontal or temporal tumors, and, morphologically, to present giant tumor cells. A significant fraction of LTS GBM patients had tumors with 1 or more alterations in the relevant GBM signaling pathways (RTK/PI3K, TP53 and RB1). In these patients, the PDGFRA alteration is suggested to be a favorable molecular factor. Our findings here are relevant for developing future targeted therapies and for identifying molecular prognostic factors in GBM patients.

Background: Gliomas remain refractory to all attempted treatments, including those using immune checkpoint inhibitors. The characterization of the tumor (immune) microenvironment has been recognized as an important challenge to explain this lack of response and to improve the therapy of glial tumors. Methods: We designed a prospective analysis of the immune cells of gliomas by flow cytometry. Tumors with or without isocitrate dehydrogenase 1/2 (IDH1/2) mutations were included in the study. The genetic profile and the presence of different molecular and cellular features of the gliomas were analyzed in parallel. The findings were validated in syngeneic mouse models. Results: We observed that few immune cells infiltrate mutant IDH1/2 gliomas whereas the immune content of IDH1/2 wild-type tumors was more heterogeneous. Some of them contained an important immune infiltrate, particularly enriched in myeloid cells with immunosuppressive features, but others were more similar to mutant IDH1/2 gliomas, with few immune cells and a less immunosuppressive profile. Notably, we observed a direct correlation between the percentage of leukocytes and the presence of vascular alterations, which were associated with a reduced expression of Tau, a microtubule-binding protein that controls the formation of tumor vessels in gliomas. Furthermore, overexpression of Tau was able to reduce the immune content in orthotopic allografts of GL261 cells, delaying tumor growth. Conclusions: We have confirmed the reduced infiltration of immune cells in IDH1/2 mutant gliomas. By contrast, in IDH1/2 wild-type gliomas, we have found a direct correlation between the presence of vascular alterations and the entrance of leukocytes into the tumors. Interestingly, high levels of Tau inversely correlated with the vascular and the immune content of gliomas. Altogether, our results could be exploited for the design of more successful clinical trials with immunomodulatory molecules.

No abstract