Targeting the retinoid acid pathway: a new therapeutic strategy for IDH1/2-mutant chondrosarcomas

Chondrosarcomas are the most common primary sarcoma of the bone in patients older than age 50. Although the majority of chondrosarcomas is of low or intermediate grade, high-grade chondrosarcomas are very aggressive tumors with high metastatic potential and a poor prognosis. Because chondrosarcomas are generally resistant to systemic chemotherapy as well as radiation, new therapeutic strategies for unresectable, metastatic or refractory disease are urgently needed. 71% of conventional chondrosarcomas and 57% of dedifferentiated chondrosarcomas carry a point mutation in the isocitrate dehydrogenase (IDH) 1 or 2 gene, which code for enzymes in the TCA (tricarboxylic acid) cycle. Mutant IDH leads to accumulation of the oncometabolite D-2­Hydroxyglutarate (2HG), which contributes to malignant progression through DNA methylation and inhibition of differentiation. IDH1/2 mutations are also present in over 80% of adult diffuse gliomas. Studies from our group (Zhang et al., Neuro-Oncology 2016) have identified genes involved in the vitamin A (retinol) pathway, particularly retinol-binding protein 1 (RBP1), as consistently hypermethylated and downregulated in glioma. RBP1 is crucial for the metabolism of retinol to retinoic acid (RA) and thus transcription of multiple genes that are important for tumor suppression, cell differentiation and immune activation. Notably, our group has also shown that the poor immunogenicity and resistance of IDH1-mutant gliomas to natural killer (NK) cell-mediated cytotoxicity is due to epigenetic silencing of NK group 2D (NKG2D) ligands. This process can be evoked by prolonged treatment of IDH1-wt astrocytes with the oncometabolite 2HG and seems therefore intrinsic to the metabolic changes induced by mutant IDH1/2. Importantly, treatment with RA can rescue both, impaired RA signaling and NK cell cytotoxicity in glioma models.

Our preliminary data strongly indicate that dysregulation of the RA pathway also mediates the malignant phenotype and a reduced immunogenicity of IDH1/2-mutant chondrosarcomas. Targeting this pathway therapeutically in chondrosarcoma will be the overarching goal of this proposal.

Specifically, the objective of Aim 1 will be to characterize the RA pathway and NK-cell mediated toxicity in IDH1/2-mutant chondrosarcomas. In Aim 2, we will focus on preclinical studies to target the RA pathway (alone and in combination with IDH inhibitors) as a new therapeutic strategy for IDH1/2-mutant chondrosarcomas. Molecular and biochemical characterization of chondrosarcomas including patient-derived cell lines and patient samples will be key to this proposal.

Our work will determine whether RA is an effective treatment for IDH1/2-mutant chondrosarcoma, a cancer for which systemic therapies have remained ineffective. Notably, RA is already in clinical trials for the treatment of IDH-mutant gliomas at our institution and could thus be rapidly translated to trials for chondrosarcoma patients.