IDH Inhibitors Target Common Glioma Mutation

Mutant-selective IDH inhibitors engage their target and shrink gliomas with an acceptable safety profile, according to phase I trial data on three drug candidates. The findings, reported at the 2019 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago, IL, in June validate a therapeutic strategy that previously had been shown to work only against hematologic malignancies.

“It's very exciting,” says Ashley Love Sumrall, MD, of the Levine Cancer Institute in Charlotte, NC, who was not involved in the research. “We're so used to our patients having a terminal illness, [so] to see any signal is encouraging.” However, some neuro-oncologists caution that targeting this early driver of disease may be futile—or could even make tumors more aggressive.

Two drugs are approved for IDH-mutant acute myeloid leukemia: the IDH1 inhibitor ivosidenib (Tibsovo; Agios) and the IDH2 inhibitor enasidenib (Idhifa; Celgene/Agios). Both target oncogenic forms of isocitrate dehydrogenase, a metabolic enzyme that normally catalyzes a key step in the Krebs cycle, but when mutated produces the cancer-causing metabolite 2-hydroxyglutarate (2-HG).

IDH mutations are particularly common in oligodendrogliomas, astrocytomas, and secondary glioblastomas. They are thought to be early activators of tumorigenesis, so researchers began studying IDH inhibitors in gliomas in parallel with their development for leukemia.

At ASCO, Atsushi Natsume, MD, PhD, of Nagoya University in Japan, reported results of a phase I study of DS-1001b (Daiichi Sankyo), an IDH1 inhibitor with high blood–brain barrier permeability (J Clin Oncol 37, 2019 [suppl; abstr2004]). Two of nine evaluable patients with nonenhancing gliomas experienced minor responses, as measured by RANO imaging criteria; the rest achieved stable disease. In another cohort of 29 patients with more aggressive, contrast-enhancing gliomas, one had a complete response, three had partial responses, and 10 experienced stable disease.

Ingo Mellinghoff, MD, of Memorial Sloan Kettering Cancer Center in New York, NY, detailed early data from a separate U.S.-based study involving patients with nonenhancing gliomas who, for a month before surgery, received either ivosidenib or an experimental inhibitor of both IDH1 and IDH2 called vorasidenib (Agios) that suppresses 2-HG to a greater extent and more readily crosses the blood–brain barrier (J Clin Oncol 37, 2019 [suppl; abstr 2003]).

Mellinghoff reported that only one of 27 patients given either ivosidenib or vorasidenib had experienced progressive disease and discontinued treatment at the time of analysis. The rest had been taking their drugs for 2 to 10 months and counting.

Mellinghoff also showed that 2-HG levels in resected tumor tissue were about 92% lower on average among those who got an IDH inhibitor compared with those who didn't. That result—the primary outcome measure of the trial—was based on an interim analysis of just 22 patient samples, yet it points to dramatic target modulation in the brain. “The drugs are effective in patients in inhibiting the enzyme,” says Ovidiu Andronesi, MD, PhD, of Massachusetts General Hospital in Boston, MA, a site investigator for the study.

The 2-HG changes also hint at the potential for therapeutic benefit. Andronesi previously showed that decreased 2-HG correlates with improved well-being in patients undergoing chemotherapy and radiation therapy (Clin Cancer Res 2016;22:1632–41). More recently, a group from Princess Margaret Cancer Centre in Toronto, Canada, demonstrated that tissue levels of different forms of 2-HG can help predict survival for patients with IDH-mutant brain tumors (Clin Cancer Res 2019;25:3366–73).

Ivosidenib, vorasidenib, and DS-1001b caused similar side effects—the most common being diarrhea, electrolyte disturbances, nausea, and headaches—although severe toxicities were more frequent with the Daiichi drug. Agios now plans to advance vorasidenib into phase III testing, with a trial slated to launch by year's end. Novartis and Forma Therapeutics also have mutant-selective IDH1 inhibitors in earlier-stage glioma trials.

Ranjit Bindra, MD, PhD, co-director of Yale's Brain Tumor Center in New Haven, CT, describes IDH inhibition as a “very attractive approach,” but adds that “there's really a fork in the road about how to address these IDH-mutant tumors.” Bindra and others have shown in preclinical studies that, although IDH mutations trigger glioma development, other oncogenic drivers take over as the disease progresses, at which point blocking IDH has little effect (Sci Transl Med 2017;9:eaal2463; Cancer Cell 2015;28:773–84; Mol Cancer Res 2016;14:976–83). “You have three labs showing IDH-mutant gliomas do not respond to IDH inhibitors,” Bindra says. “It's the elephant in the room.”

Further, because gliomas that harbor IDH defects generally have better prognoses than gliomas that don't, blocking the enzyme could actually make the cancer “more angry biologically,” notes Yale's Nicholas Blondin, MD. “My concern is you inhibit IDH and you further change around the metabolism in those cells,” he says.

Bindra thus proposes using a different class of drugs to treat patients with IDH-mutant gliomas: PARP inhibitors. As his team has shown, the 2-HG oncometabolite produced by mutant IDH enzymes alters DNA-repair pathways and induces a “BRCAness” that makes glioma cells sensitive to PARP blockade (Sci Transl Med 2017;9:eaal2463).

With PARP inhibition, explains Bindra, the goal is to allow 2-HG production and exploit the consequences for therapeutic gain, not to shut off oncometabolite production as IDH blockade does. The two strategies could not be more different, and as Bindra notes, “rarely do you have trials going on where they're testing the exact opposite thing.” –Elie Dolgin

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.