Enhancing demethylation-induced differentiation in IDH-mutant glioma

Elevated levels of D-2-hydroxyglutarate (2HG) produced by mutant IDH1 or IDH2 enzymes have been shown to have a role in establishing the glioma CpG island methylator phenotype (G-CIMP), characterized by pathogenic global hypermethylation.^1–3 As an oncometabolite, 2HG inhibits a wide array of α-ketoglutarate-dependent dioxygenases, a subset of which normally function as DNA and histone demethylases. A 2HG-induced change in the epigenetic landscape contributes to gliomagenesis, with the disruption of typical modes of gene regulation causing an impairment to differentiation and aberrant maintenance of a glioma stem-cell program. Despite these substantial advances in understanding of IDH-mutant glioma development, the progress in developing effective, IDH-mutant-specific therapies remains limited, with the current standard of care continuing to be comprised of radiation therapy and chemotherapeutics.

The demethylating agents 5-azacytidine (5-aza) and decitabine (DAC) are DNA methyltransferase 1 (DNMT1) inhibitors that are under investigation for the treatment of IDH-mutant glioma. Both agents have demonstrated efficacy in slowing tumor growth and prolonging survival in IDH-mutant glioma xenograft models.^4–6 In this issue of Neuro-Oncology, da Costa Rosa et al now show enhanced effectiveness of this approach with the addition of pro-differentiation agent all-trans retinoic acid (atRA), an agent that is used clinically to promote terminal differentiation in the treatment of acute promyelocytic leukemia (APL).⁷

The investigators observed that many retinoic acid-related genes are hypermethylated in IDH-mutant tumors, a pattern that was reversed with 5-aza treatment to patient-derived IDH-mutant glioma cells. Based on these findings, they postulate that a hypomethylating agent may enhance sensitivity to atRA by reactivating the methylated/silenced genes. In line with this concept, combination treatment led to activation of retinoic acid response elements specifically in the context of the IDH mutation, though several retinoic acid-related genes were upregulated upon both 5-aza and atRA treatment in both IDH-wildtype and IDH-mutant cell lines. Excitingly, when administered together, the combination of 5-aza and atRA significantly slowed tumor growth in a subcutaneous IDH1-mutant xenograft model. While the influence on expression of retinoic acid-related genes was variable, there was a clear increase in glial fibrillary acidic protein (GFAP), a marker of astrocytic differentiation, evidence that the combination can overcome the 2HG-induced differentiation block. These data advance a promising drug combination with translational potential that warrants clinical exploration and could be rapidly trialed in systemic IDH-mutant cancers, like cholangiocarcinoma and chondrosarcoma.

Notably, however, further development will be needed in gliomas, as the investigators demonstrate that the augmented anti-tumor effect was not apparent when testing the combination of 5-aza and atRA in an orthotopic IDH-mutant glioma model.⁷ This difference in therapeutic efficacy between subcutaneous and intracranial models underscores the unique aspects of the brain tumor microenvironment which need to be accommodated in the development of such regimens for IDH-mutant gliomas, and more practically the potential issues with drug bioavailability in the CNS compartment. Indeed, these experiments identify the existing barriers in glioma-differentiating therapy, and highlight the need for in vivo studies to guide the optimal development of such therapies. To address this question, a currently active trial is enrolling recurrent IDH-mutant glioma patients to a “window of opportunity” study treating with ASTX727, an oral formulation of DAC in combination with cedazuridine, a cytosine deaminase inhibitor that prevents DAC degradation in the gut and liver (ClinicalTrials.gov Identifier: NCT03922555). The results of this trial will examine CNS penetration of this related demethylating agent and provide important intratumoral pharmacodynamic information that could further support the clinical deployment of demethylating agents in IDH-mutant glioma. Confirmation of a differentiation phenotype in glioma samples resected after ASTX727 treatment could set the stage for further investigation of combination treatments with the pro-differentiation agent atRA.

All told, such investigations hold significant promise for the addition of such novel therapeutic demethylating and differentiating agents to the armamentarium of available effective treatments for IDH-mutant gliomas.

Acknowledgments

The text is the sole product of the authors and no third party had input or gave support to its writing.

References