Agios Pharmaceuticals Inc. has reported the first mutant-selective inhibitors of the metabolic enzymes IDH1 and IDH2 and has shown that the molecules have a therapeutic effect in preclinical cancer models.1,2 The company also announced a partnership with Foundation Medicine Inc. to develop companion diagnostics to identify patients with cancer who have the mutations.

The metabolic enzymes IDH1 (isocitrate dehydrogenase 1) and IDH2 are somatically mutated in multiple cancers, including about 50%-80% of low-grade gliomas and 9% of acute myelogenous leukemias (AMLs). All of the identified mutations alter one of three arginine residues found in the enzymes' active site.

Isocitrate dehydrogenases catalyze the conversion of isocitrate to a-ketoglutarate (a-KG). In 2009, Agios and academic collaborators reported in Nature that the glioma-specific mutation in IDH1 leads to a gain of function in which the mutant enzyme can further convert a-KG to an oncogenic metabolite called (R)-2-hydroxyglutarate (R-2HG)3 that appears to be the key driver of the cancer-promoting effects of the IDH1 mutation.

In 2010, oncometabolite production was also found to occur with cancer-associated IDH2 mutations.4

IDH1 and IDH2 mutations are believed to promote cancer because R-2HG can inhibit a-KG-dependent dioxygenases such as histone and DNA methylases, leading to cancer cells that are less differentiated and thus more proliferative than noncancerous cells.

An open question was whether inhibiting the mutant enzymes would have a therapeutic effect.

The Agios-led team has now identified the first mutant-selective inhibitors of each IDH enzyme. An in vitro screen of a small molecule library identified initial hits that inhibited the enzyme activity of either mutant target.

Optimization of hits against mutant IDH1 led to the discovery of AGI-5198, which inhibited the R132H mutant of IDH1 with an IC50 of 70 nM. It did not inhibit the wild-type enzyme or any of the IDH2 isoforms tested.

Medicinal chemistry optimization of a chemically distinct series led to the identification of AGI-6780, which inhibited the R140Q mutant of IDH2 with nanomolar potency and was selective for mutant enzyme over wild-type IDH2 and other dehydrogenases. Crystallographic studies revealed that the inhibitor bound an allosteric site at the enzyme dimer interface that is not present on IDH1, providing an explanation for the molecule's selectivity.

Next, the researchers tested the inhibitors in preclinical cancer models.

In patient-derived AML cells expressing the R140Q mutant of IDH2, AGI-6780 decreased intracellular and extracellular levels of the oncometabolite R-2HG compared with vehicle. The inhibitor induced cancer cell differentiation in the mutant AML cells but not in primary AML cells with wild-type IDH2.

In a xenograft model for R132H-IDH1 glioma, oral AGI-5198 lowered R-2HG levels, increased tumor cell differentiation and decreased tumor growth by about 50% compared with vehicle. The inhibitor did not affect the growth of wild-type IDH1 xenografts.

The results of both studies were reported in Science.

Clinical setup

Agios is continuing preclinical development of mutant-selective IDH1 and IDH2 inhibitors and hopes to start Phase I testing for both programs over the next year or two. The biotech also wants to explore the effects of mutant IDH inhibitors in additional cancers.

"We are continuing to look at preclinical animal models, focusing a lot on solid tumor models in which IDH1 is mutated, particularly chondrosarcoma and cholangiocarcinoma," said Katharine Yen, director of biology at Agios.

CSO Scott Biller added, "There is no standard of care for either indication, and so we see great opportunity there."

Biller said the reported molecules are tool compounds. He declined to disclose whether the company has identified clinical candidates but noted that "we've made a lot of progress in both programs and expect to be in the clinic in 12-24 months. Our intention is to only test the compounds in patients with IDH mutations."

Biller added that the initial indications are not disclosed.

Agios also will continue preclinical "exploratory studies to better understand the downstream biology" induced by the oncometabolite and the mutant-selective inhibitors, said Biller.

The company intends to monitor the downstream pathways during Phase I testing. "We can look, for instance, at the actual cytology of the cells in AML and at histone modifications and changes in gene expression in glioma and figure out which of these are most meaningful for response," said Yen.

In the Phase I trials, the company plans to use R-2HG, the direct product of mutant IDH enzymatic activity, as the primary marker of target engagement. Biller noted that the oncometabolite could provide an early marker of drug efficacy.

"Studies in collaboration with MGH [Massachusetts General Hospital] looking at 2HG as a biomarker in AML showed that when you treat patients with standard of care, 2HG levels go down as the AML burden decreases and come up with relapse. Thus, 2HG is potentially also a marker for response and relapse," he said.

The partnership with Foundation Medicine may also help refine a patient selection strategy. In Phase I studies, Foundation Medicine's genomic profiling will be used to determine if there are oncogenic alterations beyond IDH mutations that correlate with response, according to Biller and Yen.

The company did not disclose the IP status of the programs. Agios' IDH1 and IDH2 programs are partnered with Celgene Corp. Under a 2010 deal, Celgene received an option to license selected Agios compounds after Phase I testing in exchange for $130 million up front and up to $120 million in milestones per program, plus royalties. Agios will retain certain co-development and marketing rights.

Kotz, J. SciBX 6(14); doi:10.1038/scibx.2013.328 Published online April 11, 2013


1.   Wang, F. et al. Science; published online April 4, 2013; doi:10.1126/science.1234769 Contact: Katharine E. Yen, Agios Pharmaceuticals Inc., Cambridge, Mass. e-mail:

2.   Rohle, D. et al. Science; published online April 4, 2013; doi:10.1126/science.1236062 Contact: Katharine E. Yen, Agios Pharmaceuticals Inc., Cambridge, Mass. e-mail: Contact: Ingo K. Mellinghoff, Memorial Sloan-Kettering Cancer Center, New York, N.Y. e-mail:

3.   Dang, L. et al. Nature 462, 739-744 (2009)

4.   Ward, P.S. et al. Cancer Cell 17, 225-234 (2010)


Agios Pharmaceuticals Inc., Cambridge, Mass.

Celgene Corp. (NASDAQ:CELG), Summit, N.J.

Foundation Medicine Inc., Cambridge, Mass.

Massachusetts General Hospital, Boston, Mass.