Integrated approach to delineating molecular traits of recurrent glioblastoma using glioma stem cell models and tumor tissue profiling

The inevitable recurrence after standard therapy poses a major challenge for improving clinical outcomes of patients with glioblastoma (GB). Clinical experience accumulated over the past decades calls for a rethink of the current “one-treatment-fits-all” approach to GB treatment and necessitates new strategies to refine diagnostic criteria and increase the precision of patient stratification. There has been a growing realization that recurrent GB (recGB) comprise a molecularly distinct subgroup of tumours that differ from newly diagnosed GB in the mutation spectrum, frequency of genomic rearrangements and gene expression patterns [1]. It appears that emergence of new genomic alterations/gene expression patterns in recGBs is pro-actively driven by cytotoxic treatments used for GB treatment. The implication following from these findings is that a systematic molecular profiling could lead to the identification of targetable lesions in recGB. A conceptual challenge in the interpretation of profiling data is that distinct molecular signatures can co-exist within the same tumor [2]. Recent research indicates that molecular classifications based on the traditional approach, which relies on assessments of a single tumour biopsy, is not adequate for GBs [3]. The emerging consensus is that a switch from a single- to multisampling approach is required to improve the accuracy of molecular diagnostics in GB. Apart from the methodological difficulties that hinder the implementation of a multisampling approach to recGB operated in only about 30% of cases, the key question to be addressed is whether an association exists between distinct molecular signatures and ability to promote tumor growth after (or under) initial therapy. Glioma stem-like cells (GSCs), which have been implicated as the major underlying reason for intratumour heterogeneity and a causative factor of post-therapy recurrence in GB [4] may provide an answer. In order to delineate molecular and functional consequences of adjuvant treatments used for GB (fractionated low-dose radiation and temozolomide), we developed an integrated approach, which takes into consideration impacts of intratumor heterogeneity and functional hierarchy among different types of GSCs. Our results are fully concordant with the paradigm of intratumoral molecular heterogeneity of GBs and reveal that exposure to clinically relevant regiments of radiation elicits global re-programming distinct in the GSCs transcriptome manifest in quantitative and qualitative changes in the strength of activation of core glioma pathways and stemness-related pathways. Our findings further support a treatment-driven mechanism of molecular drifting in GB and indicate that the ultimate outcome of radiation in GSCs is determined by the composite impact of GSCs hierarchical position and integrity of pathways that govern self-renewal, differentiation and proliferation.

References:

[1] TCGA Consortium 2008 Nature 455:1061 – 1068

[2] Morrissy et a. 2017 Nat Genet 49:780 – 788

[3] Sttoriva et al 2013 Proc Natl Acad Sci USA 110:4009 – 14

[4] Lathia et al. Genes Dev 2015 29: 1203 – 1217