Overcoming glioblastoma intractability : pre-clinical characterisation of TRAIL sensitisation by marizomib and novel treatment perspectives

Abstract

Glioblastoma (GBM) is the most aggressive cancer of the central nervous system (CNS). Surgical resection, adjuvant temozolomide-based chemotherapy and radiation are the primary treatments, yet the outcome of GBM patients remains poor with a median life expectancy of 15 to 17 months. Therefore, novel and effective treatment options are required, as are reliable pre-clinical experimental models that are suitable for exploratory studies on novel drugs and drug combinations. In this work, patient-derived cell line models (PDCL), generated from fresh primary or recurrent glioblastoma tumours, have been examined to assess prevalence of responsiveness to a highly stable hexavalent format of TRAIL receptor agonist (IZI1551) and to the blood brain barrier (BBB)-permeant proteasome inhibitor marizomib (MRZ). Serum-free medium and limited cultivation times of both 2D and 3D cancer cell cultures were adopted to maintain the characteristics of primary tumour cells. The degree of BBB permeability of marizomib was evaluated in the human hCMEC/D3 cell line model, which was also employed to test the efficacy of the IZI1551+MRZ combination in pre-clinical settings. It was found that IZI1551 and marizomib acted synergistically to induce apoptosis in the majority of low-passage PDCLs, both under 2D and 3D cultivation conditions. Altering the relative timing of drug exposure, specifically marizomib pre-treatment, led to even enhanced responses and allowed to lower drug concentrations without losing treatment efficacy. Importantly, the amount of marizomib that can cross the simple BBB model was sufficient to confer sensitisation to IZI1551. In cases of treatment resistance against IZI1551 and marizomib, lowering the mitochondrial apoptosis threshold with BH3 mimetics appeared sufficient to restore apoptosis sensitivity. Taken together, these results demonstrated that marizomib is a potent sensitiser of apoptosis induced by a 2nd generation TRAIL receptor agonist in glioblastoma. The optimized synergism between marizomib and IZI1551 in time-shifted treatment schedules, together with the ability of marizomib to cross the BBB, suggests this combination as a promising strategy to be tested in clinical settings. In the second part of this work, an alternative cell death pathway, namely ferroptosis, has been investigated as a strategy to bypass the obstacle of the apoptosis refractory state of highly resistant cancers such as glioblastoma. Ferroptosis is a recently identified form of iron-dependent regulated cell death that presents distinct features compared to apoptosis and that is characterised by the accumulation of toxic lipid peroxides. Here it was shown that the U-87 MG, a bona-fide glioblastoma cell line that was reported to be TRAIL resistant, displays a dose-dependent cell death response to the ferroptosis inducer RSL3. Surprisingly, it was found that BH3 mimetics antagonised this cytotoxicity. The unexpected consequences of combining these agents highlight the need to better understand the interactions between these drugs in order to advance their use as cancer therapeutics. Overall, this thesis presents diverse treatment options against glioblastoma that exploit either drugs classically inducing apoptosis or the alternative cell death modality of ferroptosis. Considering the limited availability of approved treatments, studies aiming at expanding the choice of glioblastoma therapeutics, such as those conducted in this work, are of particular importance and pave the way for their implementation at a clinical and pre-clinical level.