Thank you for the opportunity to present my work entitled Exploiting Inherent DNA Damage Repair Defects in IDH1 and 2 Mutated Gliomas with the CNS Penetrant PARP Inhibitor BGB290. I have no disclosures. Since these two seminal papers, it has become well known in glioma that patients with IDH1 and 2 mutant tumors exhibit significantly longer survival compared to their wild-type counterparts, conferring a favorable response to chemotherapy and radiation. Mechanistically, the heterozygous missense mutation confers a neomorphic enzymatic activity in the encoded protein, in which the normal product of the wild-type enzyme, alpha-ketoglutarate, is converted into the oncometabolite R2-hydroxyglutarate. R2-hydroxyglutarate subsequently has been shown to harbor many pleiotropic effects in the cell, including methylation and cellular differentiation. Recently, our lab discovered that IDH1 and 2 mutant tumors harbor a homologous recombination defect, which renders them exquisitely sensitive to poly-ADP ribose polymerase, or PARP, inhibitors. We validated these findings in patient-derived glioma cell lines and also in genetically matched tumor xenografts in vivo. We termed this phenotype as BRCA-ness, echoing modern synthetic lethality paradigms, that mutations in key homologous recombination genes, such as BRCA1 and BRCA2, are sufficient to induce marked sensitivity to specific drugs, such as PARP inhibitors. More recently, we demonstrated that two structurally related oncometabolites, fumarate and succinate, which accumulate in patients with hereditary renal cancer syndromes harboring germline loss-of-function mutations and key citric cycle enzymes, also suppress homologous recombination. We observed similar levels of PARP inhibition in tumor cells producing these oncometabolites. Taken together, this schematic illustrates the concept of oncometabolite-induced suppression of homologous recombination of DNA double-strand breaks via inhibition of two key alpha-ketoglutarate-dependent dioxygenases, Kdm4a and Kdm4b. To understand the efficacy of PARP inhibitors within the BRCA-ness phenotype, a simplified background of DNA damage repair may be helpful. After a single-strand break in DNA, PARP is recruited to the site and subsequently undergoes poly-ADB ribosylation, or pyrolation, for enzyme activation. Subsequently, the pyrolated enzyme recruits various DNA repair complexes to the site of the break for base excision repair. When PARP is inhibited pharmacologically, normal base excision repair cannot occur, and single-strand breaks may subsequently be converted to double-strand breaks. In the presence of normal homologous recombination, the double-strand break can be repaired with high fidelity. However, in BRCA-mutated cancers and what we have shown also in oncometabolite-induced BRCA-ness in IDH-mutated tumors, normal homologous recombination cannot proceed, leading to alternative modalities of double-strand DNA break repair that are highly error-prone and can lead to genomic instability and cell death. Currently, the following PARP inhibitors have been approved by the FDA for treatment of various systemic cancers, including germline BRCA-mutated cancers. While all of these drugs inhibit PARP via preventing pyrolation, those that exhibit PARP trapping may be more cytotoxic as single agents and reflects a characteristic of the drug that traps the PARP enzyme bound to the damaged DNA. To date, PARP inhibitors have not been approved in CNS tumors, in part due to their relatively poor penetration across the blood-brain barrier. The most CNS penetrant of these drugs is valiparib, but exhibits the poorest PARP trapping potency and unfortunately has not demonstrated significant efficacy in clinical trials. BGP290 is a relatively new PARP inhibitor that has demonstrated potent PARP trapping capability and high CNS penetrance relative to the other FDA-approved PARP inhibitors. As such, we hypothesized that IDH1 and 2 mutant glioma can be targeted by therapeutic combinations of PARP inhibition with standard-of-care DNA-damaging therapies, temozolomide and radiation. We first aimed to test therapeutic synergy utilizing BGP290 with temozolomide and radiation for IDH1-mutated glioma in vitro. Using isogenic pairs of IDH1-mutated and wild-type U87 glioma cells, we performed short-term viability assays after 5 days of drug exposure. As expected, we saw preferential cell killing in IDH1 mutant cells versus wild-type cells after treatment with temozolomide, with a 60-fold more potent IC50. Likewise, treatment with BGP290 selectively targeted IDH1 mutant cells with an IC50 within the nanomolar range, compared to a micromolar range for wild-type cells. Lastly, we also demonstrated the FDA-approved PARP inhibitor dilaparib preferentially targeted IDH1 mutant cells with a nearly 4-fold more potent IC50. We subsequently sought to corroborate the results of the short-term viability assays with clonogenic survival assays, in which cells are exposed to treatment and subsequently allowed to grow for approximately 2 weeks, and colonies are counted as a proxy of cell proliferation and survival. We found that single-agent treatment with temozolomide led to robust cell killing in both IDH1 mutant and wild-type and mutant cell lines, although with greater potency, particularly at higher doses of the drug. Notably, treatment with BGP290 selectively targeted IDH1 mutant cells at a medium to high doses of the drug, consistent with short-term viability assay data, and likewise, we found that IDH1 mutated cells were preferentially targeted with increasing doses of radiation, most notably at the highest doses of 5G exposure. We then proceeded to treat these cells with dual therapies to explore potential synergistic relationships between BGP290 and temozolomide or radiation. IDH1 wild-type and mutant cells were exposed to incremental doses of temozolomide and BGP290 for 5 days, and cell viability was measured. Synergy between the two treatments was assessed with synergy scores calculated via standard open-access software, in which red shows antagonism, while deep blue illustrates treatment synergy. In these representative examples of experiments performed in triplicate, combination therapy with BGP290 and temozolomide in IDH mutant cells, IDH wild-type cells on the left, did not reveal any significant synergism. In contrast, in the same treatment in IDH1 mutant cells, there were significant synergistic relationships, particularly at higher doses of BGP290 and relatively lower doses of TMZ. This effect was even more pronounced when synergy was analyzed between BGP290 treatment and radiation exposure, as seen in IDH1 wild-type cells, where there was antagonism between the two treatments across most of the tested concentrations for BGP290 and radiation doses. In contrast, for IDH1 mutant cells, we saw particularly potent synergy at lower doses of radiation across a range of concentrations of BGP290. In our second aim, we sought to perform preliminary experience of pharmacokinetic markers of on-target drug activity in the brain, utilizing a murine intracranial glioma model treated with oral BGP290 administration. Utilizing established luciferase lentivirus constructs for bioluminescence imaging of U87 wild-type and IDH1 mutant glioma cells, these cells were orthotopically injected into the cortex of nude mice to establish an intracranial in vivo model. There were nine mice each for the wild-type and IDH1 mutant cohorts, respectively, within each group. Three mice were treated with low-dose BGP290, three with high-dose BGP290, and three acted as controls. After intracranial injection of the cells, tumors were allowed to grow for three weeks, and tumor formation was subsequently assessed with serial BLI measurements, as demonstrated in this image. Subsequently, the animals were treated daily with BGP290 for eight days, and then sacrificed for harvesting of tumor tissue, normal brain from the contralateral cortex, and blood plasma. Samples from six mice were subsequently sent for liquid chromatography and mass spectrometry, which revealed increasing concentrations of BGP290 in tumor tissue compared to normal brain. There were processing errors with the controlled tumor samples, which we are currently refining. This pilot in vivo experiment did not reveal a dose-dependent response in tumor levels of the drug, but was limited by small sample size. So in conclusion, single-agent treatment with temozolomide and radiation preferentially targets IDH1 mutant cells in an isogenic U87 cell line. BGP290 treatment also preferentially targets IDH1 mutant cells, consistent with potent PARP trapping ability. BGP290 demonstrates therapeutic synergism with temozolomide and radiation in these cells in vitro. And BGP290 crosses the blood-brain barrier and selectively penetrates tumor tissue in an in vivo intracranial gliama model. And with that, I'd like to thank all the members of my lab, particularly our lab manager, Dr. Ranjini Sundaram, and my PI, Dr. Ranjit Bindram. Thank you for your attention.

DNA Damage Repair

IDH1 Mutated Gliomas

PARP Inhibitor

Therapeutic Synergy

Blood-Brain Barrier