Identification and validation of DKK1 as a novel candidate therapeutic target for glioblastoma

Abstract

Glioblastoma (GBM) is a very aggressive and invasive tumour that relapses within nine months of diagnosis and remains incurable despite advances in multimodal therapy including surgical resection, chemotherapy and radiation. Poor patient outcome has been correlated to specific markers of brain tumour initiating cells (BTIC) and intratumoural heterogeneity (ITH), which have also been associated with treatment resistance and tumour recurrence. ITH can be explained at the cellular level by the existence of multiple populations of cancer cells, including some which have acquired stemness properties like self-renewal, proliferation, and multilineage differentiation, also known as cancer stem cells (CSCs). In brain tumours, CSCs or BTICs, have been shown to be resistant to both chemotherapy and radiation treatment, allowing them to escape therapy and consequently generate for tumour recurrence. As a result, therapies that focus on targeting the BTIC compartment within the bulk GBM tumour would provide better treatment and prognosis for patients. To profile GBM BTICs we conducted two transcriptomic screens. The first compared GBM BTICs to neural stem cells (NSCs), their healthy counterparts, and for the second we developed a pipeline utilizing a dynamic BTIC patient-derived xenograft (PDX) model of human GBM recurrence allowing for the profiling of GBM BTICs at engraftment, after chemoradiotherapy delivery in a phase we have termed "minimal residual disease" (MRD), and at tumour recurrence. In this study, Dickkopf-1 (DKK1) was identified as a potential therapeutic target for GBM from each transcriptomic screen and was studied using short hairpin knockdowns, blockade with monoclonal antibodies, and subsequent functional stem cell assays.