Background: IDH mutations cause aberrant accumulation of (R)-2-hydroxyglutarate (2-HG), an “oncometabolite” that has pleotropic effects on tumorigenesis. Mutant-IDH (mIDH) inhibition significantly improves progression free survival in patients with IDH-mutated WHO grade 2 glioma, however most patients will progress, with a lack of understanding on mechanistic reasons for failure.
Methods: In the first pre- and post-treatment perioperative trial in patients with mIDH1 low-grade glioma, we tested safusidenib (AB-218 / DS-1001b), an oral, blood brain barrier penetrant IDH1 R132X inhibitor, in a single-arm, open-label study. Patients with IDH1 mutated glioma (no prior radiation or chemotherapy), received safusidenib for four weeks following surgical biopsy and prior to definitive resection (Part A) and then continued Safusidenib until toxicity or progression (Part B). The primary endpoint demonstrated the feasibility and acceptability of conducting a two-stage perioperative trial. Translational endpoints include paired whole genome transcriptome sequencing (WGTS), single nuclei RNA sequencing (snRNAseq), spatial transcriptomics and spatial metabolomics as well as patch-clamp electrophysiology. Here we report outcomes for Part A.
Expected Results: As of May/02/2024 10 patients with IDH mut glioma have been enrolled. The mean age was 37, with 40% male; six had prior surgical resection. Treatment has been well tolerated with ongoing follow-up for safety, and patients reported a positive response to recommend joining a research study, despite the demanding nature of the trial. The mean safusidenib tumour concentration was 2654 ng/g (range 957.0 – 4810) with tumour:plasma ratio 0.33. On-target activity was confirmed with reduction in 2-HG 88% from biopsy with homogenous reduction in 2-HG throughout the infiltrative tumour demonstrated with spatial metabolomics. As expected, increased metabolites associated with the reductive citric acid cycle seen prior to treatment were significantly reduced post-treatment, with concomitant increases in downstream metabolites in the oxidative cycle. Paired snRNAseq demonstrates transcriptional shift post treatment with alterations in differentiation, and immune infiltration supported by spatial transcriptomics. Notably, we identify reduction in neural excitability, functionally validated by patch clamp electrophysiology.
Conclusions: Taken together, these results provide detailed understanding on mechanism of mIDH inhibition for glioma, and a proof of concept of the perioperative approach to advance drug development in glioma.