Summary Basis of Decision for Idhifa

 

Clinical Pharmacology

Enasidenib (the medicinal ingredient in Idhifa) is a small molecule inhibitor of the isocitrate dehydrogenase 2 (IDH2) mutant enzyme. Isocitrate dehydrogenase 2 mutations confer a gain of function, whereby the aberrant enzyme catalyzes the production of the oncogenic metabolite 2 hydroxyglutarate (2-HG). The metabolite 2-HG induces a block of cell differentiation by inhibiting the activity of chromatic-modifying histone and deoxyribonucleic demethylases. Enasidenib targets the mutant IDH2 variants R140Q, R172S, and R172K at approximately 40-fold lower concentrations than the wild-type enzyme in vitro. Inhibition of the mutant IDH2 enzyme by enasidenib led to decreased 2-HG levels and induced myeloid differentiation in vitro and in vivo in human xenograft models of IDH2 mutated acute myeloid leukemia (AML).

The pharmacokinetic disposition of enasidenib was characterized in 3 studies conducted in healthy volunteers (AG-221-CP-002, AG-221-CP-001, AG-221-C-002) and in the Phase I/II study (AG221-C-001; i.e., the pivotal and the support studies) conducted in patients with advanced hematologic malignancies including relapsed or refractory acute myeloid leukemia (R/R AML). The clinical pharmacology package also included a population pharmacokinetic analysis (AG-221-MPK-001), a concentration and QTc analysis (C-001-QTCPK) and 5 pharmacokinetic/pharmacodynamic analysis reports.

The absolute bioavailability of enasidenib under fasting condition following a single dose of 100 mg tablet in healthy subjects was 57.2%. Maximum plasma concentration (C_max) was achieved at approximately 3 hours and terminal half-life (t_1/2) was 29 hours. The mean clearance is 1.37 L/h and the mean volume of distribution is 55.8 L.

After a single dose of 100 mg enasidenib in patients with advanced hematologic malignancies under fasting condition, C_max was 1.343 mcg/mL and the area under the plasma drug concentration time curve from time 0 to 24 hours (AUC_0-24) was 23.227 mcg*h/mL. Cross-study comparison indicates that the exposure (C_max and AUC_0-72) of enasidenib was higher (1.3~1.9 fold) in patients compared to healthy subjects as a result of slower clearance in patients.

Upon multiple doses of enasidenib under fasting conditions, plasma concentration reached steady-state starting around Cycle 2, Day 1 (after approximately 29 days of treatment) for the 100 mg once daily (QD) dose. On Cycle 2 Day 1, enasidenib C_max and AUC_0-24 were 11.6 mcg/mL and 258.506 mcg*h/mL, respectively, for 100 mg QD. Enasidenib had a long plasma half-life [190 hours based on the population pharmacokinetic analysis] with 9- to 11-fold accumulation in patients.

Enasidenib exposure increased in a dose proportional manner from 50 mg to 450 mg single daily dose in patients, but increased in a less than dose proportional manner after having received multiple doses at steady state.

The metabolic pathways included N-dealkylation, oxidation, direct glucuronidation, and combinations of these pathways. The parent compound was the most predominant component in circulation. AGI-16903 was the most prominent metabolite which showed inhibitory activity against IDH2 R140 in vitro. The major route of elimination is fecal excretion, while urinary excretion is a minor elimination pathway.

A population pharmacokinetic analysis indicated that there was no obvious relationship between drug clearance or volume of distribution and any of the covariates tested (gender, age, race, mild hepatic impairment and renal impairment) with the exception of estimated apparent clearance (CL/F) which was lower in the patients compared to healthy subjects (CL/F in patients is 70% lower than in healthy subjects). The effect of hepatic impairment on enasidenib plasma exposure could not be considered conclusive due to the small sample size.

There were no clinical drug-drug studies conducted with enasidenib (the medicinal ingredient in Idhifa). The population pharmacokinetic analysis suggests that concurrent cytochrome (CYP) 450 enzyme inhibitors administration was not a significant covariate affecting exposure. In vitro results indicated that enasidenib was metabolized by multiple CYP and uridine diphosphate-glucuronyltransferase (UGT) isoforms and was an inhibitor of certain CYP isoforms and UGT1A1. Enasidenib is also an inhibitor of P-glycoprotein (P-gp), Breast Cancer Resistance Protein (BCRP), organic anion transporter 1 (OAT1), OATP1B1, OATP1B3, and organic cation transporter 2 (OCT2), but was not a substrate for P-gp or BCRP transporters. Upon initiation or discontinuation of Idhifa in patients being treated with other drugs that are substrates of these enzyme or transporters and have a narrow therapeutic index, monitoring of the expected effect or drug concentration of the other.

In the exposure-response analyses, no apparent relationship between systemic enasidenib exposure levels and the clinical best response was observed in the overall patient population enrolled in the pivotal trial, but the logistic regression analyses show a statistically significant correlation between enasidenib exposure and overall response rate (ORR) for R/R AML patients with the R172 mutation and for all R/R AML patients pooled regardless of mutation status. Exposure-safety analyses with logistic regression revealed a statistically significant relationship between enasidenib exposure and blood bilirubin elevation regardless of mutation status. Other safety endpoints of febrile neutropenia, leukocytosis, tumor lysis syndrome and isocitrate dehydrogenase (IDH) differentiation syndrome were not correlated with enasidenib exposure.

The potential for QTc prolongation with Idhifa was evaluated in Study AG-221-C-001 in patients with advanced haematological malignancies with an IDH2 mutation. Data visualization of change in baseline of corrected QT interval using Fridericia formula (ΔQTcF) and plasma concentration of enasidenib suggested a decrease in ΔQTcF with increasing concentration of enasidenib. Based on the QTc data for a single dose of 30 mg to 650 mg and multiple doses of 100 mg daily in the fasted state, no large mean changes in the QTc interval (>20 ms) were observed following treatment with enasidenib under fasting condition.

Based on exploratory analysis of pharmacokinetic/pharmacodynamic correlations, daily doses of 100 mg enasidenib consistently suppressed 2-HG at Cycle 2 Day 1 in both patients with the R140 IDH2 mutation (median 92.8% inhibition) and patients with the R172 IDH2 mutation (median 27.6% inhibition). Maximum suppression in 2-HG appeared to have been achieved in the patients with the R140 mutation as there were minimal additional decreases in 2-HG levels with increasing enasidenib doses. While no correlation was observed between 2-HG reduction and clinical response in those patients with the R140 mutation, a statistically significant association in reduction of 2-HG was observed with clinical response in those with the R172 mutation.

Study AG221-C-002 was a Phase I, two-way crossover study to assess the pharmacokinetics and safety of a single dose of Idhifa in healthy male subjects when administered under fed and fasted conditions. The study results indicate that administration of enasidenib with food produces an approximately 50% increase in enasidenib extent of absorption (AUC), and a 63% increase in the rate of absorption (C_max) of enasidenib, relative to administration of enasidenib mesylate under fasted conditions. Study AG221-C-001, the pivotal trial, was conducted under fasting conditions yet Celgene Inc. proposed that enasidenib be given without regard to food. During the course of the review, sufficient pharmacologic and clinical justification was provided to support this recommendation. The increased exposure of enasidenib when administered with food is not expected to be relevant clinically. This is based on the available data including the maximum administered dose, the exposure-safety analyses, and the variability in enasidenib exposure observed between patients, all of which support the safe administration of enasidenib with or without food. Additionally, compliance with treatment is expected to improve if enasidenib can be administered without regard to food.

For further details, please refer to the Idhifa Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

Efficacy of Idhifa in the treatment of IDH2 mutated patients with R/R AML has been demonstrated in the Phase II portion of the AG221-C-001 Phase I/II study which was an open-label, single-arm, multicentre clinical trial.

The AG221-C-001 Phase I/II study consisted of three parts: a Phase I dose escalation, a Phase I expansion, and a Phase II expansion (referred to as the Phase II portion). The Phase II portion was considered as the pivotal study in support of efficacy. The Phase I portion was considered as providing additional support.

The Phase I dose escalation portion was designed to determine the maximum tolerated dose (MTD) and recommended Phase II dose (RP2D), and to evaluate clinical activity and safety of Idhifa in patients with advanced hematologic malignancies. Total daily doses from 50 mg to 650 mg were evaluated in the dose escalation portion of the study. No MTD was reached. However, an increasing proportion of subjects treated at daily doses over 300 mg required dose modification or reduction for adverse events (that did not qualify as dose limiting toxicities) in order to remain on treatment. The starting dose of enasidenib recommended for evaluation was 100 mg daily. This was based on the safety, pharmacokinetics, pharmacodynamics, and efficacy of enasidenib observed in the dose escalation portion of the study. In the 34 subjects with a dose increase from 100 mg to 200 mg per day in the Phase I expansion portion of the study, the best response was achieved at the 100 mg daily dose, and in general the increase to the higher 200 mg daily dose was not associated with better objective responses. The pre-specified lower bound (25%) of the 95% CI for investigator-assessed ORR was met, with an ORR of 40.4% (median duration of response [DoR] of 5.6 months) in 109 R/R AML subjects who were assigned the 100 mg daily dose. A CR rate of 19.3% was seen (median DoR of 16.3 months).

Based on the clinical activity observed in R/R AML patients in Phase I, the Phase II portion of Study AG221-C-001 was designed to assess efficacy of Idhifa at the RP2D and further evaluate safety in patients with R/R AML and with IDH2 mutation.

In the pivotal Phase II portion of Study AG221-C-001, 105 patients with IDH2 mutated R/R AML were treated with Idhifa. Patients enrolled were selected to meet the unmet medical need in AML, and consisted of the following:

• Patients who relapse after allogenic-hematopoietic stem cell transplantation (allo-HSCT);
• Patients in second or later relapse;
• Patients who are refractory to initial induction or re-induction treatment;
• Patients who relapse within one year of initial treatment, excluding patients with favorable-risk status according to National Comprehensive Cancer Network (NCCN) Guidelines (NCCN, 2015).
  • Favorable-risk cytogenetics: inv(16), +(16;16), t(8;21), t(15;17).
• Patients who failed two or more cycles of first line therapy (consisting of intermediate intensity chemotherapy, hypomethylating agent or low dose cytarabine) [failed is defined as refractory or subjects who relapsed after receiving regimens involving low-intensity therapy]
• Phase II patients were required to have IDH2 mutation tested centrally in samples of bone marrow aspirate and peripheral blood, and confirmed positive in bone marrow aspirate and/or peripheral blood during screening prior to study treatment. The Abbott RealTime IDH2 assay was used in the majority of patients.

A starting dose of 100 mg Idhifa given once daily was administered to all enrolled patients on Days 1 to 28 in 28-day cycles. Starting at the first cycle Day 1, dosing was continuous with no inter-cycle rest periods.

The median age of patients in the study was 68 years with 24% of patients ≥ 75 years; 63% of patients were male; 61% of patients were Eastern Cooperative Oncology Group Performance Status (ECOG PS) 1 and 15% were ECOG PS 2; 66% of patients had relapsed AML; 16% of patients had prior stem cell transplantation for AML; and, the median number of prior therapies was two.

The primary endpoint was overall response rate (ORR) based on investigator assessment and was defined as the rate of responses, including complete response (CR), CR with incomplete neutrophil recovery (CRi), CR with incomplete platelet recovery (CRp), partial response, and morphologic leukemia-free state. Responses were based on revised International Working Group 2003 criteria for AML. Response was assessed by the investigator and retrospectively by an Independent Response Adjudication Committee (IRAC).

An ORR of 37.1% (95% confidence interval [27.9, 47.1], median DoR of 5.6 months) was observed. This observation met a threshold predefined in the study protocol as indicative of clinical benefit as the lower bound of the 95% confidence interval was greater than 25%; thus, the study had a positive outcome. CR was observed in 20.0% of patients and CR plus CRi/CRp in 31.4% of patients. The median durations of response were 5.6 months, 6.7 months and 6.5 months respectively for ORR, CR, and CR plus CRi/CRp. Similar CR+CRi/CRp rates were observed in patients with either R140 or R172 mutation.

Key secondary endpoints in the AG221-C-001 study, including the number of required red blood cell (RBC) and/or platelet transfusions and overall survival, also supported the overall benefit of Idhifa treatment. Of the 76 patients who were either RBC and/or platelet transfusion dependent at baseline, 27 patients (35.5%) became both RBC and platelet transfusion independent during any 56-day post-baseline period. Additionally, 19 of the 29 patients (65.5%) who were both RBC and platelet transfusion independent at baseline remained both RBC and platelet transfusion independent during any 56-day post-baseline period. The ability for patients to be independent of transfusions in this treatment setting is considered to be of considerable clinical relevance for the patient. Also observed was a decrease in AML complications during periods of response, including febrile neutropenia, infections and bleeding events, likely signifying improvement in their underlying disease. Additionally, a median overall survival of 7.0 months was observed which appears promising compared with historical controls; however, the prognostic significance of the IDH2 mutation is controversial.

Additionally, there was minimal evidence of myelosuppression with Idhifa which is commonly observed with chemotherapy in this treatment setting and nine patients (8.6%) discontinued treatment and proceeded to an allo-HSCT. These were patients who were not considered candidates for therapy with curative intent at the time of study enrolment.

A durable CR is a clinical endpoint in AML studies that has been associated with improved survival based on studies with cytotoxic chemotherapy. The CR rate in Study AG221-C-001 was 20.0% with a median DoR of 6.7 months. In comparison, the CR rate with available therapies can be expected to be approximately 10.5% in a general R/R AML population.

Overall response rate, the primary endpoint in the study, was defined by inclusion of CR and CRi/CRp, a typical definition in AML studies, but also included morphologic leukemia free-state responses and partial responses. This broader definition, while appropriate in this early phase study where an attempt was being made to observe any efficacy signal attributable to the mechanism of action of Idhifa (i.e., differentiation of leukemic cells), has not been validated as providing clinical benefit.

The overall survival (OS) results observed appear promising although OS from a single-arm study must be interpreted with caution. The prognostic significance of the IDH2 mutation is controversial so it is not clear how this may impact survival as compared to an AML population not harbouring this mutation.

Indication

The New Drug Submission for Idhifa was filed by the sponsor with the following indication, which Health Canada subsequently approved:

Idhifa is indicated for the treatment of adult patients with relapsed or refractory acute myeloid leukemia (AML) with an isocitrate dehydrogenase-2 (IDH2) mutation.

Treatment with Idhifa should be initiated following confirmation of IDH2 mutation through a validated test.

Overall Analysis of Efficacy

Treatment of patients with R/R AML with Idhifa in Study AG221-C-001 demonstrated an ORR of 37.1% of patients, CRs in 20.0% of patients, the ability of patients to achieve or maintain RBC or platelet transfusion independence and a median OS of 7.0 months. However, as the primary endpoint (ORR) has not been validated as a surrogate of clinical benefit and as data are conflicting regarding the prognostic significance of the IDH2 mutation, these efficacy data are considered to be promising.

For more information, refer to the Idhifa Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Safety

The safety evaluation of single-agent Idhifa is based on the pooled Phase I/II data from the clinical study AG221-C-001 previously described in the Clinical Efficacy section. Two-hundred and fourteen patients (214) patients were treated with Idhifa 100 mg daily.

Overall, Idhifa was well-tolerated with relatively few dose modifications or discontinuations required to manage adverse events. The most common (≥15%) adverse events as assessed as related to treatment by the investigator were nausea, vomiting, diarrhea, elevated bilirubin and decreased appetite. The most frequent serious adverse events as assessed as related to treatment by the investigator were differentiation syndrome (7.5%), febrile neutropenia (4.2%), leukocytosis (3.7%), nausea (3.3%), dyspnea (2.8%), decreased appetite (1.9%), pyrexia (1.9%), and vomiting (1.9%).The most important of these was IDH differentiation syndrome which has the potential for fatal outcomes if not identified early in the course of the syndrome. In the AG221-C-001 study, 13.1% patients treated with Idhifa experienced an adverse event of differentiation syndrome. Additionally, Idhifa related differentiation syndrome with fatal outcome has been reported outside of clinical trials related to a delay in recognition or to a delay in treatment initiation.

Differentiation syndrome is associated with rapid proliferation and differentiation of myeloid cells. While there is no diagnostic test for differentiation syndrome, symptoms and signs reported by more than 50% of patients included acute respiratory distress represented by dyspnea and/or hypoxia with need for supplemental oxygen, unexplained fever, pulmonary infiltrates and renal impairment. Less common symptoms and signs may include pleural effusion, lymphadenopathy, bone pain, peripheral edema with rapid weight gain, and pericardial effusion. An alanine aminotransferase (ALT) or aspartate aminotransferase (AST) elevation has been observed. Differentiation syndrome has been observed with and without concomitant hyperleukocytosis, and as early as 10 days and up to 5 months after Idhifa initiation. The condition can be managed with the administration of corticosteroids.

Other important adverse events associated with Idhifa include non-infectious leukocytosis, tumor lysis syndrome, elevation of blood bilirubin, and gastrointestinal disturbances.

Non-infectious leukocytosis

Idhifa can induce myeloid proliferation resulting in a rapid increase in white blood cell count without evidence of infection or clinical signs of differentiation syndrome. In the pooled AG221-C-001 study, 13.6% of patients were reported with an adverse event of non-infectious leukocytosis with 6.5% of cases reported as serious. The majority of cases occurred within the first 3 months of treatment. Non-infectious leukocytosis led to dose interruption in 1.9% of patients and treatment discontinuation in 0.9% of patients.

Tumor lysis syndrome

An increase in uric acid associated with imbalance in electrolytes, consistent with signs and symptoms of tumor lysis syndrome, has been observed. The adverse event of tumor lysis syndrome was reported in 6.1% of patients of the AG221-C-001 with 4.7% of cases reported as serious. Tumor lysis syndrome usually occurred within the first 3 months of treatment. Tumor lysis syndrome led to treatment discontinuation in 0.9% of patients.

Elevated bilirubin

Idhifa may interfere with bilirubin metabolism through inhibition of UGT1A1. Idhifa caused dose dependent bilirubin elevation beginning by the first week of treatment and stabilizing by the end of first month of treatment. Elevated bilirubin led to dose reductions in 0.5% (1/214), and treatment interruption in 4.2% (9/214) of patients.

Patients with congenital UGT1A1 deficiency (Gilbert Syndrome) who received Idhifa experienced a more rapid increase in bilirubin values, as compared to patients without this mutation and more frequently experienced bilirubin increase >3 times the upper limit of normal (ULN).

Gastrointestinal disturbance

Adverse events such as nausea, diarrhea, vomiting, and other reactions such as dysgeusia and decreased appetite were usually mild to moderate in severity, did not lead to treatment discontinuation and only infrequently required dose reduction or interruption. These events were not dose related and generally occurred during the first month of treatment and often resolved with continued treatment.

Overall Analysis of Safety

Overall, the safety profile of Idhifa is considered to be acceptable in patients with IDH2 mutated R/R AML who were being treated palliatively. The most serious toxicity associated with Idhifa was IDH differentiation syndrome which has the potential for fatal outcomes if not identified early in the course of the syndrome. The adverse events associated with Idhifa are considered to be manageable when employing the recommended dose modifications as outlined in the Idhifa Product Monograph. Reports of patients, some with fatal outcome, in whom differentiation syndrome was not recognized or where patients did not receive the necessary treatment have been identified from post-market surveillance of Idhifa. To mitigate this risk, differentiation syndrome has been labelled within the Serious Warnings and Precautions Box of the Idhifa Product Monograph to alert the prescriber to the potential for this event. Additionally, patient/caregiver information wallet cards will be included in all cartons of Idhifa. The wallet card lists differentiation syndrome symptoms and provides a section for physician and/or hospital/centre contact information. Prescribers are recommended to encourage patients to keep the Patient Wallet Card with them and to share the associated Companion Card with a caregiver. The card is to be used in the event any of the symptoms of differentiation syndrome are observed. The patient or caregiver should show this card to any new treating healthcare professionals to alert them to the potential of differentiation syndrome.

For more information, refer to the Idhifa Product Monograph, approved by Health Canada and available through the Drug Product Database.