Glioblastoma multiforme (GBM) is the most lethal type of brain tumour in the adult humans. The cancer-initiating cell (CIC) hypothesis supports the notion that failures in current approaches to GBM treatment might be attributed to the survival of the CIC subpopulation. Recent evidence shows the idea that using CIC-enriched cell lines derived from human GBM as new targets for drug discovery programs, may improve the chance of successfully translating the basic research findings into clinical trials. Although this approach appears promising, many important biological and technical issues (characterization of functional CIC markers, inter- and intra-tumoral CIC heterogeneity, and isolation and maintenance inconsistency) need to be resolved.