Tami-51. Identifying New Tumor Microenvironment (Tme) Contexts of Vulnerability in Glioblastoma

Abstract

Glioblastoma (GBM) is the most frequent and aggressive adult brain tumor with 85% of patients dying within two years. New effective precision medicine therapies are urgently required, especially for isocitrate dehydrogenase wild-type (IDHwt) disease. Despite efforts to subtype patients based on molecular profiles, this approach has yet failed to direct treatment strategies. Further interrogation of the tumor microenvironment (TME) across molecular subtypes and identification of new TME specific subtypes may guide new directions for future therapies. Here, we analysed transcriptomic data from selected GLIOTRAIN(www.gliotrain.eu)(n=120) and TCGA(n=69) IDHwt patients. Firstly, the microenvironment cell population (MCP)-counter method (a gene-expression-based TME deconvolution tool) was validated for use in the brain tumor setting using quantitative multiplex immunohistochemistry. In this context, immune markers (CD20/CD3/CD68/CD8) were significantly correlated with MCP-counter scores. We are currently optimizing and validating a vessel-density and microglial RNA-signature to provide a more robust representation of brain TME. Next, using MCP-counter, the TME composition of IDHwt tumors was assessed within proneural (24%), classical (38%) and mesenchymal (38%) subtypes. We initially classified the GLIOTRAIN cohort into 3 novel clusters characterised by differences in TME composition and validated our findings in the TCGA cohort. A TME-high group (37%) is characterized by elevated presence of lymphocytes and myeloid cells, and presents a high level of immune checkpoint genes: PDCD1(PD1) and CTLA4. In addition, the presence of tertiary lymphoid structures (TLS) is a feature of TMEhigh/mesenchymal+ patients. This finding has been validated by IHC and RNA-signature. TME-med (38%) displayed heterogenous immune populations and the TME-low (25%) represented an ‘immune-desert’ group. There was no significant difference in OS based on these TME subtypes(p=0.50). We hypothesise that PD1/CTLA4 blockade might be an effective treatment strategy in TME-high patients. These hypotheses will be tested (in the adjuvant setting) using appropriate syngeneic disease models which incorporate surgical resection.

Systemic and local immunosuppression in glioblastoma and its prognostic significance

“…In 2019, using a 500-gene set classifier (only 108 genes matched Verhaak et al's 840-gene set) on six different datasets (three TCGA-cohorts and three Asian-cohorts), Teo et al also confirmed the classical, mesenchymal, and proneural subtypes, with similar survival outcomes between subtypes (293). Finally, in 2023, White et al also found no significant difference in overall survival between subtypes in the GLIOTRAIN (n=123), TCGA (n=164), and CGGA (n=693) cohorts after stratifying the mesenchymal, proneural, and classical subtypes according to Wang et al' report (294). Interestingly, the authors introduced novel glioblastoma tumor microenvironment subtypes for IDH wild type glioblastoma (TME High , TME Med , and TME Low ), characterized by significantly different expression of genes specific to all immune and endothelial cell markers.…”

“…Interestingly, the authors introduced novel glioblastoma tumor microenvironment subtypes for IDH wild type glioblastoma (TME High , TME Med , and TME Low ), characterized by significantly different expression of genes specific to all immune and endothelial cell markers. However, stratification into these new subtypes showed no association with overall survival in the GLIOTRAIN, TCGA, CGGA, and DUKE cohorts (294).…”

“…Considering the enormous inter-and intratumoral cellular heterogeneity and plasticity, as well as the evolution of the cancer genome and phenotype, clinically meaningful subtyping of glioblastoma based on the transcriptome alone is challenging. However, there is some retrospective clinical evidence that mesenchymal or TME High glioblastoma might respond better to immunotherapy (vaccine, checkpoint inhibitor or oncolytic virus) (294,297).…”

Advancements, Challenges, and Future Directions in Tackling Glioblastoma Resistance to Small Kinase Inhibitors

“…Another source of complexity influencing heterogeneity is given by the tumor microenvironment (TME) [23]. TME is not constituted only by cancerous cells, but also by different populations of immune cells, stromal cells, endothelial cells, and pericytes, creating different types of niches within the tumor [24].…”

Spatial transcriptomics in glioblastoma: is knowing the right zip code the key to the next therapeutic breakthrough?

“…One of the hypothesized reasons for the aggressiveness and treatment resistance displayed by GBM is its complex spatial organization, seemingly seeking to mimic the spatial architecture of its s u r r o u n d i n g m i c r o e n v i r o n m e n t . T h e G B M t u m o r microenvironment (TME) is comprised of endothelial cells, neurons, astrocytes, oligodendrocytes, microglia, tumor-associated macrophages, tumor-infiltrating lymphocytes, and noncellular components such as apocrine and paracrine signaling molecules, exosomes, extracellular matrix (ECM) components, and secreted ECM remodeling enzymes (32,33). Each of these components play not only individual roles but also orchestrate an incredibly complex and spatially distinct TME in which GBM can evade immune system detection and uncontrollably proliferate.…”