Function of IDH Mutations Can Be Exploited by PARP Inhibitors Instead of IDH Inhibitors

A phase II basket trial has been announced to test that mutations in isocitrate dehydrogenase 1 and 2 are not driver mutations that should be targeted with direct IDH inhibitors, but instead they create vulnerabilities that can be exploited through treatment with PARP inhibitors.

Ranjit S. Bindra, MD, PhD

Mutations in isocitrate dehydrogenase 1 and 2 (IDH1/2) are not the mutations commonly found in cancer cells, according to a presentation from the 2017 AACR Annual Meeting.1 These mutations, which are typically found in patients with brain tumors and acute myelogenous leukemia (AML), are not driver mutations that should be targeted with direct IDH inhibitors, but instead they create vulnerabilities that can be exploited through treatment with poly (ADP-ribose) polymerase (PARP) inhibitors. A phase II basket trial has been announced to test this theory in patients with IDH1/2 mutations.

“Our data strongly suggest that exploiting the DNA repair deficiency in IDH-mutant tumors, rather than inhibiting the function of the mutant IDH proteins, likely will be a better strategy for treating brain tumors with these specific mutations, a devastating disease with an urgent need for better therapies,” Ranjit S. Bindra, MD, PhD, said in a statement before his presentation.

IDH1/2 mutations are heterozygous, missense mutations that induce both a gain- and a loss-of-function phenotype. “The mechanism of action of these mutations is quite fascinating. These are neomorphic mutations that take the normal product of the wild-type enzyme and convert it to this oncometabolite that is foreign to the cell called 2-hydroxyglutarate [2HG],” explained Bindra, assistant professor of therapeutic radiology and experimental pathology, Yale School of Medicine, during the presentation.

Previous research has shown that IDH mutations may drive the initial formation of tumor cells but quickly become “passenger” mutations once the cells have been transformed.2 As these mutations are not typical driver mutations, they cannot be targeted as such.

Bindra noted that IDH mutations could serve as either a tumor suppressor gene, like the BRCA1/2 mutations in patients with breast and ovarian cancer—creating a “BRCAness”, or an oncogene, similar to EGFR mutations. The oncometabolite 2HG, when it comes from mutant IDH1/2 (R-2HG) rather than as a response to hypoxia (S-2HG), suppresses KDM4A/B and leads to decreased homologous recombination repair and sensitivity to PARP inhibitors, as is often seen in patients with BRCA mutations. Bindra commented that this could be part of a much larger underlying pathway to genetic instability that could involve other mutations as well.

These findings were previously published by Bindra and his colleagues in Science Translational Medicine demonstrating the potential sensitivity of IDH mutations to PARP inhibitors due to the function of the mutation.3

“Our findings raise the question of whether we should target this Achilles heel, meaning should we exploit the DNA repair defect, or should we repair that Achilles heel, that would be in line with the oncometabolite hypothesis. This is a critical question because it really raises questions about how should we be approaching the therapeutic management for these patients,” Bindra said.

The oncometabolite hypothesis suggests that if both growth signals and metabolic rewiring could be blocked, rather than just one or the other, it could have a broad, profound impact on these tumor cells.

Bindra commented that exploitation of the DNA repair defects seen in IDH mutations is a powerful cytotoxic approach to treating cancer that is potentially a game changer as significant improvements in progression-free survival rates have already been seen with PARP inhibitors, such as olaparib (Lynparza) and talazoparib (BMN-673).

In in vitro studies of IDH1/2 mutations treated with olaparib and talazoparib, a significant difference in survival was found with IDH mutations versus wild-type tumors favoring the mutant tumor cells. This effect was carried over to in vivo studies with tumor xenograft models where a large difference in survival was noted in treatment with olaparib when an IDH mutation was lacking.

In addition, Bindra noted that there was a significant synergistic interaction for combination approaches with PARP inhibitors in IDH-mutant cells. Synergy was marked for a combination of talazoparib and cisplatin as well as for talazoparib and VE-822, an ATR inhibitor.

Alternatively, if IDH1 inhibitors, which are currently under development, were to be used and 2HG would be suppressed, then it would reverse the sensitivity to PARP inhibitors and revert the DNA repair defect.

In preclinical studies of IDH wild-type and IDH-mutant cancer cell line models treated with IDH inhibitors or PARP inhibitors, Bindra and his colleagues found that IDH1 inhibitors did not have a significant effect on IDH1-mutant tumor cell growth whereas PARP inhibitors did.

“These data suggest that…[not] all mutations are drivers that should be inhibited, but rather there are mutations that are actually creating vulnerabilities that can be exploited,” explained Louis M. Weiner, MD, who moderated the session.

To put this theory to the test, a basket trial has been designed to look at treating patients with IDH1/2-mutant tumors with a PARP inhibitor. Bindra announced the new trial during his presentation: “For the first time, today we are pleased to announce conditional approval by the National Cancer Institute [NCI] of this phase II trial where we are seeking to test this concept in the clinic.”

The phase II basket trial will enroll patients with IDH1/2-mutant tumors of any histology, although extracranial solid tumors are certainly a focus of the trial, and treat them with 300 mg of olaparib twice daily in a 28-day cycle. The trial will be led by principal investigator Patricia M. LoRusso, DO, and will begin enrollment later this year.

“If, in fact, [the sensitivity of IDH mutations to PARP inhibitors] can be confirmed in this exciting basket trial, it would be a very interesting new twist on how we begin interpreting molecular data in the future,” Weiner said.


  1. Sulkowski P, Corso C, Robinson N, et al. Oncometabolites induce a BRCAness state that can be exploited by PARP inhibitors. Presented at: 2017 AACR Annual Meeting; April 1-5, 2017; Washington, DC. Abstract LB-290.
  2. Johannessen TA, Mukherjee J, Viswanath P, et al. Rapid conversion of mutant IDH1 from driver to passenger in a model of human gliomagenesis. Mol Cancer Res. 2016;14(10):976-983.
  3. Sulkowski PL, Corso CD, Robinson ND, et al. 2-Hydroxyglutarate produced by neomorphic IDH mutations suppresses homologous recombination and induces PARP inhibitor sensitivity. Sci Transl Med. 2017;9(375):eaal2463. doi: 10.1126/scitranslmed.aal2463.


View more from the 2017 AACR Annual Meeting