A Multistep In Silico Approach Identifies Potential Glioblastoma Drug Candidates via Inclusive Molecular Targeting of Glioblastoma Stem Cells

Abstract

Aims Glioblastoma (GBM) is the highest-grade of glioma for which no effective therapy is currently available. Despite extensive research in diagnosis and therapy there has been no signi cant improvement in GBM outcomes, with a median overall survival continuing at a dismal 15-18 months. In recent times, glioblastoma stem cells (GSCs) have been identi ed as crucial drivers of treatment resistance and tumor recurrence; and GBM therapies targeting GSCs are expected to improve patient outcomes. We used a multistep screening strategy to identify repurposed candidate drugs against therapeutic molecular targets in GBM with potential to concomitantly target GSCs. Main methods Common differentially expressed genes (DEGs) were identi ed through analysis of multiple GBM and GSC datasets from the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA). For identi cation of target genes, we further selected the genes with most signi cant effect on overall patient survival. The relative mRNA and protein expression of the selected genes in TCGA control versus GBM samples was also validated and their cancer dependency scores were assessed. Drugs targeting these genes and their corresponding proteins were identi ed from LINCS database using Connectivity Map (CMap) portal, and by in silico molecular docking against each individual target using FDA approved drugs library from DrugBank database, respectively. The molecules thus obtained were further evaluated for their ability to cross blood brain barrier (BBB) and their likelihood of resulting in drug resistance by acting as p-glycoprotein (p-Gp) substrates. The effect of these nal shortlisted compounds on a panel of GBM cell lines was examined and compared with temozolomide through the drug sensitivity EC50 values and AUC from the PRISM Repurposing Secondary Screen, and IC50 values obtained from GDSC portal. Key ndings We identi ed RPA3, PSMA2, PSMC2, BLVRA and HUS1 as molecular targets in GBM including GSCs with signi cant impact on patient survival. Our results show GSK-2126458, linifanib, drospirenone, eltrombopag, nilotinib and PD198306 as candidate drugs which can be further evaluated for their anti-tumor potential against GBM. Signi cance Through this work we identi ed repurposed candidate therapeutics against GBM utilizing a GSC inclusive targeting approach, which demonstrated high in vitro e cacy. These drugs have the potential to be developed as individual or combination therapy to improve GBM outcomes.

Overcoming Barriers in Glioblastoma—Advances in Drug Delivery Strategies

“…Yet, upcoming in silico approaches including AI can help to prescreen interesting candidate drugs to select the GBM drug candidates with the highest likelihood of being successful in clinical trials. If successful, this can substantially improve the preselection of interesting repurposed drug candidates [ 139 ].…”