Identification of De Novo Pyrimidine Synthesis as a Targetable Vulnerability in a Novel IDH1 Mutant Engineered Astrocytoma Model

• Published in: International journal of radiation oncology, biology, physics Vol. 111; no. 3; p. S86
• Main Authors: Shi, D.D., Wang, A.C., Gao, W., Khanal, J., Levitt, M.M., Jennings, R.B., Signoretti, S., Nguyen, Q.D., Endress, J.E., Xu, M., Gradl, S., Sutter, A., Jeffers, M., Janzer, A., Cahill, D.P., Abdullah, K.G., Ligon, K.L., Harris, I.S., Kaelin, W.G., McBrayer, S.K.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.11.2021
• ISSN: 0360-3016; 1879-355X
• DOI: 10.1016/j.ijrobp.2021.07.205

Despite high prevalence of IDH1-R132H mutations in grade II-III gliomas, effective therapies remain limited. We sought to identify tumor-specific vulnerabilities induced by the IDH1-R132H oncogene and test the translational relevance of targeting them using a new genetically engineered mouse model (GEMM) of IDH1 mutant anaplastic astrocytoma. We conducted a synthetic lethality screen using isogenic IDH1 mutant and IDH1 wild-type (WT) glioma cells and a novel drug screening platform developed by our group, called MAPS. To create a GEMM of anaplastic astrocytoma, we developed a strategy to engineer Tp53, Atrx, Pik3ca, and Idh1 mutations in the brains of adult mice. We intracranially injected adeno-associated virus (AAV) expressing Cre recombinase and CRISPR sgRNAs targeting murine Atrx and Tp53 genes into 4 mouse strains: 1) LSL-Cas9; 2) LSL-Cas9; LSL-Pik3caH1047R, 3) LSL-Cas9; LSL-Idh1R132H, and 4) LSL-Cas9; LSL-Idh1R132H; LSL-Pik3caH1047R. Our screen revealed that IDH1 mutant cells are hypersensitive to drugs targeting enzymes in the de novo pyrimidine nucleotide synthesis pathway, including dihydroorotate dehydrogenase (DHODH). We demonstrated that these cytotoxic effects are on-target and are associated with increased DNA damage in IDH1 mutant cells. Next, we showed that IDH1 mutant patient-derived glioma stem-like cell lines (GSCs) are also hyperdependent on de novo pyrimidine synthesis compared to IDH1 WT lines. In addition, we found that the novel brain penetrant DHODH inhibitor, BAY2402234 (currently undergoing testing in leukemia patients), decreased tumor growth in an orthotopic xenograft model of IDH1 mutant, but not IDH1 WT, glioblastoma. We then sought to create and use a GEMM of IDH1 mutant anaplastic astrocytoma to test whether dependence on de novo pyrimidine synthesis manifests across tumor grade. Following intracranial AAV injection (see methods), astrocytomas preferentially formed after 9-14 months in mice carrying both Idh1 and Pik3ca conditional alleles. These tumors histologically resembled grade III astrocytomas, expressed astrocytoma lineage markers, and displayed elevated (R)-2-hydroxyglutarate. To create an additional model with shorter tumor latency, we performed secondary transplants of GSCs derived from our GEMM into recipient mice. We found that BAY2402234 blocked tumor growth in these orthotopic astrocytoma allografts. Our findings establish IDH1 mutations as predictive biomarkers of DHODH inhibitor efficacy in gliomas across tumor grade, highlight BAY2402234 as a candidate glioma therapeutic, and unveil new genetically faithful mouse models of IDH1 mutant glioma. In addition, we show that BAY2402234 induces preferential DNA damage in IDH1 mutant cells, thereby supporting evaluation of BAY2402234 as a potential tumor-selective radiosensitizer.