Acute myeloid leukemia (AML) is the most common acute leukemia in adults, with nearly 20,000 new cases expected in the U.S. for 2018. Generally speaking, AML is a malignancy of older age, with an average age at diagnosis of approximately 68 years.
Between 6 and 10 percent of AML patients have disease that is driven by somatic mutations to the gene which encodes isocitrate dehydrogenase 1 (IDH1). Recently, results were published from a first-in-human phase I clinical study (NCT02074839) which evaluated the use of the mutant IDH1 inhibitor ivosidenib in relapsed or refractory (R/R) AML patients (N Engl J Med 2018;378:2386-2398).
One investigator that participated in this study was Courtney DiNardo, MD, Associate Professor in the Division of Cancer Medicine at University of Texas MD Anderson Cancer Center, Houston. “As a result of the positive findings obtained in this clinical study, on July 18, 2018, this compound was approved by the FDA for the treatment of R/R AML patients having confirmation of an IDH1 mutation using the FDA-approved test,” DiNardo noted.
Isocitrate Dehydrogenase & Cancer
Isocitrate dehydrogenase is an enzyme that catalyzes the in vivo conversion of isocitrate to α-ketoglutarate (αKG). There are two different isoforms of this enzyme: 1) IDH1, which is found in the cytosol, and 2) IDH2, which is the mitochondrial variant of this protein. Mutated forms of IDH1 have been observed in a number of different malignances, including AML, glioma, cholangiocarcinoma, and chondrosarcoma. Additionally, mutated forms of the mitochondrial homolog, IDH2, are also associated with a number of cancers, including AML (Cancer Cell 2010;17:225-234).
Interestingly, these mutated forms of IDH have alterations at key active site arginine residues (R132 for IDH1 and R172 for IDH2). As a result of these mutations, the enzymes display a loss-of-function for the conversion of isocitrate to αKG and a gain-of-function for the reduction of αKG to (R)-2-hydroxyglutarate (R-2HG)(Nature 2009;462:739-744). R-2HG, which has been called an oncometabolite, has been shown to competitively inhibit a number of αKG-dependent enzymes, including histone lysine demethylases, which results in epigenetic alteration (EMBO Rep 2011;12(5):463-469). This oncometabolite has also been implicated in impaired hematopoietic differentiation (Cancer Cell 2010;18:553-567). In August 2017, the FDA gave approval for the use of enasidenib in adults with IDH2-mutant R/R AML, as confirmed using an FDA-approved mutation test.
Formerly referred to as AG-120, ivosidenib is a selective inhibitor of several different forms of R132-mutant IDH1, showing no activity against IDH2 isoforms (wild-type or mutant), even at concentrations up to 100 mm (ACS Med Chem Lett 2018;9:300-305).
Ivosidenib showed rapid equilibrium inhibition against the R132-mutant IDH1 homodimer in preclinical in vitro studies. Attempts to further ascertain the mechanism of action using kinetic binding studies were inconclusive as a result of persistently pre-bound NADP(H) in R132-mutant IDH1 enzyme preparations. In subsequent preclinical studies, ivosidenib displayed good cellular potency across several R132-mutant IDH1 endogenous and overexpressing cell lines, which implied its potential utility across all such driven cancers. Additionally, robust reduction of tumor 2HG levels was noted in HT1080 cell-inoculated female nude BALB/c mice with ivosidenib administration.
Phase I Clinical Study
“This was a phase I, multicenter, open-label, dose-escalation, and dose-expansion study with ivosidenib being orally administered daily in 28-day cycles,” DiNardo noted. Patient eligibility criteria included age >18 years; ECOG performance status score of 0, 1, or 2; and presence of confirmed IDH1-mutant hematologic malignancy.
“The primary objectives were to assess the safety, maximum tolerated dose (MTD), and recommended phase II dose (RP2D) of ivosidenib, and to assess clinical activity in R/R IDH1-mutant AML,” she stated, “while secondary objectives included the characterization of pharmacokinetics (PK) and pharmacodynamics (PD) profiling.
“For the purposes of this study, dose-limiting toxicities were defined as nonhematologic toxic effects of grade 3 or higher or prolonged myelosuppression with persistent grade 4 or higher neutropenia or thrombocytopenia in the absence of leukemia (i.e., blast count of less than 5%) at least 42 days after treatment initiation.”
The overall population included those patients receiving at least one dose of ivosidenib, while the primary efficacy population consisted of the first 125 R/R AML patients dosed at 500 mg daily who were eligible for group 1 and whose first dose of ivosidenib was at least 6 months prior to the analysis cutoff date of May 12, 2017.
Between March 12, 2014, and May 8, 2017, 268 patients were enrolled in this study, and of these, 258 received at least one dose of the investigational medication. In this patient population, 78 were included in the dose-escalation phase while 180 took part in the dose-expansion phase.
A total of 179 participants had R/R AML with a starting dose of 500 mg ivosidenib daily. From these patients, the first 125 comprised the primary efficacy population, including 92 from expansion group 1 (dose-expansion phase) and 33 from the dose-escalation phase also eligible for group 1.
The rate of complete remission (CR) or complete remission with partial hematologic recovery (CRi) was 30.4 percent (95% CI: 22.5-39.3%) in the primary efficacy population, while the rate of CR was 21.6 percent (95% CI: 14.7-29.8%) and the overall response rate (ORR) was 41.6 percent (95% CI: 32.9-50.8%).
“The median duration of CR or CRi was 8.2 months (95% CI: 5.5-12.0 months), the median duration of CR was 9.3 months (95% CI: 5.6-18.3 months), and the median duration of response was 6.5 months (95% CI: 4.6-9.3 months),” DiNardo noted.
“The median time to CR or CRi was 2.7 months (range: 0.9-5.6 months), the median time to CR was 2.8 months (range: 0.9-8.3 months), and the median time to response was 1.9 months (range: 0.8-4.7 months).”
The median overall survival (OS) for the primary efficacy population was 8.8 months (95% CI: 6.7-10.2 months) with a median follow-up time of 14.8 months (range: 0.2-30.3 months). The 18-month survival rate was 50.1 percent among patients who had a CR or CRi, while the median was not reached as of the data cutoff date. Estimates for the median OS were 9.3 months among those patients showing a response other than CR or CRi and 3.9 months among patients who had no response.”
Two episodes of dose-limiting toxicities were noted in the dose-escalation phase: one grade 3 QT interval prolongation in a patient who received an ivosidenib dose of 800 mg daily and one grade 3 rash in a patient receiving 1,200 mg daily. “Although the MTD was not determined, the ivosidenib dose of 500 mg daily was selected for the dose-expansion phase,” DiNardo stated. “This decision was based on the available safety, pharmacokinetic, and clinical data and data on 2HG levels.”
Nearly all of the 179 relapsed or refractory AML patients who received 500 mg ivosidenib daily as their starting dose experienced a treatment-related adverse event (AE) (98.9%). “The most common AEs observed, irrespective of a relationship to ivosidenib, were diarrhea (30.7%), leukocytosis (29.6%), febrile neutropenia (28.5%), nausea (27.9%), fatigue (25.7%), dyspnea (24.6%), QT interval prolongation (24.6%), peripheral edema (21.8%), anemia (21.8%), pyrexia (21.2%), and cough (20.7%).”
Ivosidenib displayed rapid absorption, with a mean serum half-life of 72-138 hours after a single dose administered 3 days before the scheduled first dose of the escalation phase in a subset of patients. “This finding supported the use of a daily dose regimen,” DiNardo explained. A steady state concentration was attained within 14 days of dosing; however, when the dosage was increased, the resulting increase in plasma exposure was not proportional.
“The greatest inhibition of 2HG production in plasma and bone marrow was observed by day 14 in those receiving 500 mg ivosidenib daily, with no additional inhibition being observed at higher doses (e.g., 800 or 1,200 mg daily),” she said. Mean plasma 2HG levels approached those of healthy individuals after multiple daily doses of 500 mg ivosidenib.
Regarding the clinical trial's endpoints, DiNardo stated, “This phase I first-in-human study met its primary and secondary endpoints. Although the MTD was never identified, the dose of 500 mg was chosen as the RP2D based on favorable efficacy, safety, and PK/PD data demonstrating effective 2HG suppression.”
When asked how the responses obtained in this study compared to those commonly noted for R/R AML patients, DiNardo replied, “This was an older and refractory population, with a median of two prior therapies, 29 percent of which were post stem cell transfer, where historical data would suggest a CR rate of less than 15 percent and a median OS of under 6 months.
“The outcomes seen with ivosidenib included a CR rate of ~20 percent, an ORR of ~40 percent, and a median OS of 9 months,” she stated, then adding, “the rates of transfusion independence (i.e., lasting more than 2 cycles) on this study (40% overall) were impressive and speak to the clinical benefit in the responding (and even some ‘non-responding’) patients.”
When asked if she was surprised by the robustness of the study results, DiNardo commented, “It is amazing to me that a well-tolerated, oral, and non-toxic targeted therapy can so effectively treat acute leukemia.”
Offering her opinion on the FDA approval, she stated, “Thankfully ivosidenib is not experimental anymore—ivosidenib was approved in July of this year for the treatment of R/R AML. Notably, this study also highlights the importance of molecular testing and genomic medicine in cancer to identify patients for whom there are personalized and targeted therapeutic options available.
“For R/R AML patients, a statistically significant association was noted between a lower co-mutational burden and CR or CRi as compared with other responses (P=0.02). No specific preexisting single gene mutation, including in NRAS, was shown to be significantly predictive of clinical response or resistance to ivosidenib. Although, more frequent mutations in receptor tyrosine kinase pathway genes were noted in R/R AML patients who did not have a response as compared with those who had a CR or CRi.”
Looking to the future, DiNardo commented, “The next frontier will be putting ivosidenib in rational combinations both in the relapsed setting, but even more importantly in the upfront, newly diagnosed patient setting to further improve clinical outcomes.”
Richard Simoneaux is a contributing writer.