FDA approval summary: decitabine and cedazuridine tablets for myelodysplastic syndromes

Abstract

On July 7, 2020, the Food and Drug Administration approved Inqovi™ (Otsuka Pharmaceutical Co.), an oral fixed dose combination tablet comprising 35 mg decitabine, a hypomethylating agent, and 100 mg cedazuridine, a cytidine deaminase inhibitor (abbreviated DEC-C) for treatment of adult patients with myelodysplastic syndromes (MDS). Evidence of effectiveness of DEC-C was established in Phase 3 ASTX727-02 (N=133) in adults with MDS. The study involved a 2-sequence crossover comparing DEC-C and intravenous (IV) decitabine 20 mg/m² once daily for the first 5 days of each 28-day cycle in the first 2 cycles. From Cycle 3 onwards, patients received DEC-C. Five-day cumulative area under the curve (5-d AUC) of decitabine for DEC-C was similar to that of IV decitabine with geometric mean ratio 0.99 (90% confidence interval: 0.93, 1.06). Clinical benefit was supported by Study ASTX727-02 and the similarly designed Phase 2 Study ASTX727-01-B (n=80), with complete remission (CR) of 21% and 18% and median duration of CR 7.5 and 8.7 months, respectively. Adverse reactions were consistent with IV decitabine. Post-marketing assessments were issued to address the effect of cedazuridine on QT prolongation, food effect, moderate and severe hepatic impairment, and severe renal impairment on the pharmacokinetics and safety of DEC-C.

Introduction

Myelodysplastic syndromes (MDS) are a heterogeneous group of hematopoietic stem cell disorders resulting in cytopenias and risk of transformation to acute myelogenous leukemia (AML) (1). Current standard of care for patients with newly-diagnosed higher-risk MDS includes treatment with a hypomethylating agent (HMA), i.e. azacitidine or decitabine (1).

Decitabine was approved as an intravenous (IV) formulation in 2006 for treatment of patients with MDS (2). The recommended dose of IV decitabine is 20 mg/m² daily for 5 days of each 28-day cycle. In clinical studies leading to its approval, the complete remission (CR) + partial remission (PR) rate was 17% and CR rate was 9% (3), based on International Working Group (IWG) 2000 criteria (2, 4).

The development of an orally bioavailable HMA for patients with MDS was previously limited by rapid metabolism of nucleosides by cytidine deaminase (CDA) in the gastrointestinal (GI) mucosa and liver. This led to development of cedazuridine, a CDA inhibitor that prevents the rapid metabolism of decitabine in the GI tract (5).

Herein, we provide a summary of FDA’s review of the marketing application that led to approval of 35 mg decitabine and 100 mg cedazuridine fixed dose (DEC-C) tablets for the treatment of adults with MDS, including previously treated and untreated, de novo and secondary MDS with the following French-American-British subtypes (refractory anemia, refractory anemia with ringed sideroblasts, refractory anemia with excess blasts, and chronic myelomonocytic leukemia [CMML]) and intermediate-1 (INT-1), intermediate-2 (INT-2), and high-risk International Prognostic Scoring System (IPSS) groups.

Clinical Pharmacology

Overview of Clinical Trials

The clinical program involved two clinical trials: ASTX727-01, which included a dose-escalation portion studying oral decitabine and cedazuridine (ASTX727-01-A) and a dose-confirmation portion (ASTX727-01-B) (NCT02103478) (6) studying oral decitabine and cedazuridine as a combination; and ASTX727-02 (NCT03306264), a Phase 3 study to serve as a PK bridge between DEC-C tablets and IV decitabine.

ASTX727-01-B and ASTX727-02 were multicenter, open-label, randomized, 2-sequence crossover studies in adults ≥ 18 years with MDS or CMML who received up to 1 cycle of prior decitabine or azacitidine. In ASTX727-01-B, patients with MDS were eligible if they had INT-1, INT-2, or high-risk IPSS scores. In ASTX727-02, patients with all FAB classification criteria were included, to match the labeled indication for IV decitabine.

In Studies ASTX727-01-B and ASTX727-02, patients were randomized 1:1 to receive 35 mg decitabine and 100 mg cedazuridine orally in Cycle 1 and decitabine 20 mg/m² IV in Cycle 2 or the reverse sequence, once daily Days 1–5 of each 28-day cycle. Starting with Cycle 3, all patients received 35 mg decitabine and 100 mg cedazuridine orally once daily Days 1–5 of each 28-day cycle until disease progression or unacceptable toxicity. Patients were stratified by IPSS risk of INT-1/INT-2/high-risk in ASTX727-01-B. In both studies, the oral decitabine and cedazurudine formulations were administered under modified fasted conditions (i.e., 2 hours before and 2 hours after dosing).

ASTX727-01-B consisted of two stages: a dose confirmation stage and a fixed dose combination stage. For the dose confirmation stage, 42 patients were used to compare the 5-day cumulative area under the curve (5-d AUC) of decitabine of 35 mg decitabine and 100 mg cedazuridine as individual capsules against IV decitabine. For the fixed-dose combination stage, 18–24 patients were used to compare the 5-d AUC of decitabine between DEC-C fixed-dose tablets and IV decitabine.

For Study ASTX727-02, a sample size of 118 subjects was estimated for the 2 one-sided equivalence tests for the geometric mean ratio (GMR) of 5-day AUC of decitabine between DEC-C tablets and IV decitabine to provide 90% power at the statistical significance level of 0.05, assuming a geometric mean ratio of 1.0, the CV of 0.55, and the 90% confidence interval (CI) limits of 80 to 125%.

Dose Selection

The dose escalation study (ASTX727-01-A) in patients with MDS (n=41) indicated that cedazuridine increased exposure of oral decitabine (Supplemetary Figure S1).

Based on results of ASTX727-01-A, an oral dose combination of 35 mg decitabine and 100 mg cedazuridine was selected for ASTX727-01-B in patients with MDS/CMML (n=64). Study ASTX727-01-B confirmed the oral dose combination of 35 mg decitabine and 100 mg cedazuridine, as the 5-d AUC of decitabine for the oral dose combination was 94–98% of the 5-d AUC of IV decitabine.

Pharmacokinetic Bridge - Primary Support of Efficacy

The primary endpoint of ASTX727-02 was the 5-d AUC of decitabine with response rate as the secondary endpoint. Of 123 evaluable patients, the GMR and 90% CI of the 5-d AUC of decitabine for DEC-C tablets compared to 20 mg/m² IV decitabine was 0.99 (0.93, 1.06) (Table 1), which was within the bioequivalence limits of 0.80 to 1.25.

Table 1:

Comparison of Decitabine Exposures in Study ASTX727-02^a,b

PK Parameter	Comparison (Paired Analysis)		N	GMR	90% CI
Daily AUC	IV	Day 1 vs. Day 5	111	1.03	0.98, 1.07
	DEC-C	Day 1 vs. Day 5	122	0.58	0.55, 0.62
	DEC-C	Day 2 vs. Day 5	124	1.07	1.02, 1.13
	DEC-C vs. IV	Day 1	115	0.60	0.55, 0.65
	DEC-C vs. IV	Day 5	112	1.06	0.98, 1.14
5-day AUC	DEC-C vs. IV	--	123	0.99	0.93, 1.06

Abbreviations: AUC, area under the curve; CI, confidence interval; DEC-C, 35 mg decidabine and 100 mg cedazuridine tablet; IV, intravenous 20 mg/m² decitabine; GMR, geometric mean ratio

^aAdapted from FDA review (11)

^bDEC-C and IV decitabine were administered once daily for 5 consecutive days in Cycles 1 or 2.

Pharmacokinetics

The AUC of decitabine administered as DEC-C tablets was 60% of that observed with IV decitabine after the first dose (on Day 1) and was comparable to that of IV decitabine after the subsequent doses (Figure 1, Table 1). Therefore, DEC-C tablets must not be substituted for IV decitabine within a cycle.

Figure 1:

Daily Decitabine AUC of DEC-C and Intravenous Decitabine in Individual Patients in Study ASTX727-02^a

Abbreviations: AUC, area under the curve

^aAdapted from FDA review (11)

PK parameters and other clinical pharmacology characteristics of decitabine and cedazuridine are presented in Supplementary Table S1 and Supplementary Figure S2. Age (32–90 years), sex, and mild hepatic impairment (HI) (total bilirubin > 1–1.5 × ULN or AST > ULN) did not have an effect on the PK of decitabine or cedazuridine after dosing with DEC-C tablets. Increasing body size was associated with decreased decitabine AUC and higher creatinine clearance (CLcr) correlated with lower cedazuridine AUC; however, neither had any clinically meaningful effect on the PK of decitabine or cedazuridine. Effects of moderate (total bilirubin > 1.5–3 × ULN and any AST) and severe HI (total bilirubin > 3 × ULN and any AST) or severe renal impairment (RI) (CLcr 15 to <30 mL/min) and end stage renal disease (CLcr <15 mL/min) on the PK of decitabine and cedazuridine are unknown.

Assessment of Efficacy

Efficacy Analysis

Efficacy in ASTX727-01-B and ASTX727-02 was assessed according to IWG 2006 MDS response criteria (7). However, neither study was powered to establish efficacy (rather, bioequivalency). In ASTX727-01-B, response was assessed with bone marrow and peripheral blood assessments at the start of Cycle 3, and every other cycle. In ASTX727-02, response was assessed with bone marrow and peripheral blood assessments at the start of Cycles 3, 5, and 7, then every 3 months in the first year, and every 6 months thereafter unless indicated earlier.

Eighty-six patients were randomized in study ASTX727-01-B and 80 patients received any amount of study treatment: 41 received Sequence A (DEC-C Cycle 1, IV decitabine Cycle 2) and 39 received Sequence B (IV decitabine Cycle 1, DEC-C Cycle 2). In study ASTX727-02, 138 patients were randomized and 133 patients were treated: 66 on Sequence A and 67 on Sequence B.

Demographics and Disposition

Demographics of the efficacy population for studies ASTX727-01-B and ASTX727-02 are shown in Table 2. In Study ASTX727-01-B, the median time since diagnosis was 52.5 days. Only 9% received up to 1 cycle of prior HMA therapy prior to enrolling on the trial, and 19% received any prior anticancer therapy (e.g. HMA, lenalidomide, erythropoietin stimulating agents [ESAs]). In Study ASTX727-02, median time since diagnosis was 83 days. Only 8% received up to 1 cycle of prior HMA therapy, and 23% received any prior anticancer therapy.

Table 2:

Characteristics of Patients in Study ASTX727-01-B and Study ASTX727-02^a

	ASTX7272-01-B (n = 80)	ASTX727-02 (n = 133)
Median age (range)	71 (32, 90) years	71 (44, 88) years
Sex
Male	61 (76%)	87 (65%)
Female	19 (24%)	46 (35%)
Race
White	74 (93%)	121 (91%)
Black or African American	2 (3%)	4 (3%)
Asian	1 (1%)	3 (2%)
Other or Not Reported	3 (4%)	5 (4%)
ECOG PS
0–1	73 (92%)	133 (100%)
2	7 (9%)	0
Disease category / IPSS
MDS INT-1	35 (44%)	59 (44%)
MDS INT-2	19 (24%)	26 (20%)
MDS High Risk	9 (11%)	21 (16%)
MDS Low Risk	0	11 (8%)
CMML	17 (21%)	16 (12%)
Prior HMA therapy b
Prior Azacitidine	3 (4%)	6 (5%)
Prior Decitabine	3 (4%)	4 (3%)
HMA (not specified)	1 (1%)	0
Transfusion dependence
RBC Transfusion Dependence	38 (48%)	52 (39%)
Platelet Transfusion Dependence	12 (15%)	10 (8%)

^aAdapted from FDA review (11)

^bOne cycle only, per the Exclusion Criteria.

At the time of the efficacy analyses for Studies ASTX727-01-B and ASTX727-02, median follow-up was 24.0 and 12.6 months and 1% and 45% remained on therapy with DEC-C tablets, respectively. Reasons for treatment discontinuation (ASTX727-01-B %, ASTX727-02 %) included primary disease (35%, 13%), adverse event (21%, 9%), hematopoietic stem cell transplantation (15%, 20%), withdrawal by patient (6%, 2%), alternative therapy (0%, 2%), or other (21%, 9%). Median treatment duration was 6.6 months (range, < 0.1–27.9) on ASTX727-01-B and 8.2 months (range, 0.2–19.7) on ASTX727-02.

Efficacy Results

Efficacy results for Study ASTX727-01-B and Study ASTX727-02 are presented in Table 3. CR rate was 18% (95% CI 10–28%) on Study ASTX727-01-B and 21% (95% CI 15–29%) on Study ASTX727-02. No patients achieved a PR. CR rates were comparable across most demographic and disease subgroups (see Supplementary Figure S3 for Phase 3 results). The median duration of CR was 8.7 and 7.5 months on ASTX727-01-B and ASTX727-02, respectively.

Table 3:

Efficacy Results in Patients in Study ASTX727-01-B and Study ASTX727-02^a

	ASTX7272-01-B (n = 80)	ASTX727-02 (n = 133)
Complete remission
Number (%)	14 (18%)	28 (21%)
95% CI	(10–28%)	(15–29%)
Median time to CR (range)	4.8 (1.7, 10.0) months	4.3 (2.1, 15.2) months
Median duration of CR (range)	8.7 (1.1, 18.2) months	7.5 (1.6, 17.5) months
Conversion from TD to TI b
Number (%)	20/41 (49%)	30/57 (53%)
Maintenance of TI c
Number (%)	25/39 (64%)	48/76 (63%)

^aAdapted from FDA review (11)

^bRate of conversion from RBC and/or platelet transfusion dependence (TD) at baseline to RBC and platelet transfusion independence (TI) during any 56-day post-baseline period

^cNumber (percentage) of patients who were both RBC and platelet transfusion independent (TI) at baseline and maintained transfusion independence during any 56-day post-baseline period.

Assessment of Safety

Safety was assessed in 208 patients (78 from ASTX727-01-B, 130 from ASTX727-02) with MDS or CMML treated with at least 1 dose of DEC-C tablets. Safety data was pooled across both studies given the similar randomized crossover study designs and patient populations. FDA performed a comparative safety analysis of DEC-C tablets (n=107) and IV decitabine (n=106) during Cycle 1, in addition to an assessment of safety during the entire treatment period in all 208 patients.

Twelve on-treatment deaths (6%) were concluded to have potentially resulted from a direct toxicity of treatment. Proximate causes of death included sepsis (n=3), septic shock (n=3), pneumonia (n=2), respiratory failure (n=2), cerebral hemorrhage (n=1), and sudden death (n=1). One death (1%) was noted on each of the DEC-C tablets and IV decitabine arms in the first cycle.

Adverse reactions (ARs) led to dose interruption in 86 (41%), dose reduction in 39 (19%), and withdrawal of therapy in 11 (5%) patients. ARs leading to dose interruption in >5% of patients were neutropenia, febrile neutropenia, thrombocytopenia, and anemia. ARs leading to dose reduction in >2% of patients were neutropenia, anemia, and thrombocytopenia. The only ARs leading to discontinuation in more than 1 patient were febrile neutropenia and pneumonia. There were no meaningful differences in ARs leading to dose interruption, reduction, or withdrawal between the DEC-C tablets and IV decitabine arms during Cycle 1. The first DEC-C tablet dose reductions involved switching from a 5-day to a 4-day regimen, followed by reduction to a 3-day regimen, then an alternating 3-day regimen if necessary (i.e. Days 1, 3, and 5). Although the decitabine exposure with an alternating 3-day regimen is only about 40% of that with a 5-day regimen, it may be beneficial to continue decitabine treatment with this regimen for patients who experience adverse reactions with the 3-day regimen (Days 1–3).

Common ARs are listed in Table 4. Nonlaboratory adverse reactions ≥ 5% more common with DEC-C tablets than IV decitabine in Cycle 1 included nausea, dyspnea, diarrhea, dizziness, headache, transaminase increased, and hemorrhage. Adverse reactions ≥ 5% more common with IV decitabine than DEC-C tablets in Cycle 1 included mucositis and myalgia.

Table 4:

Non-laboratory Adverse Reactions (≥ 10%) in the Pooled Safety Population^a

Preferred termb	DEC-C Cycle 1 (n = 107)		IV Decitabine Cycle 1 (n = 106)		DEC-C All Cycles (n = 208)
	Any Grade	Grade 3–4	Any Grade	Grade 3–4	Any Grade	Grade 3–4
Fatigue	29%	2%	25%	0	55%	5%
Constipation	20%	0	23%	0	44%	0
Hemorrhage	24%	2%	17%	0%	43%	3%
Myalgia	9%	2%	16%	1%	42%	3%
Mucositis	18%	1%	24%	2%	41%	4%
Arthralgia	9%	1%	13%	1%	40%	3%
Nausea	25%	0	16%	0	40%	0.5%
Dyspnea	17%	3%	9%	3%	38%	6%
Diarrhea	16%	0	11%	0	37%	1%
Rash	12%	1%	11%	1%	33%	0.5%
Febrile neutropenia	10%	10%	13%	13%	33%	32%
Dizziness	16%	1%	11%	0	33%	2%
Edema	10%	0	11%	0	30%	0.5%
Cough	7%	0	8%	0	28%	0
Headache	22%	0	13%	0	30%	0
Decreased appetite	10%	1%	6%	0	24%	2%
URI	6%	0	3%	0	23%	1%
Pneumonia	7%	7%	7%	5%	21%	15%
Transaminase increased	12%	1%	3%	0	21%	3%
Abdominal pain	9%	0	7%	0	19%	1%
Pyrexia	7%	0	7%	0	19%	1%
Renal impairment	9%	0	8%	1%	18%	0
Vomiting	5%	0	5%	0	15%	0
Sepsis	6%	6%	2%	1%	14%	11%
Neuropathy	4%	0	8%	0	13%	0
Fall	4%	0	1%	0	12%	1%
Insomnia	6%	0	2%	0	12%	0.5%
Cellulitis	4%	1%	3%	2%	12%	5%
Hypotension	4%	0	6%	1%	11%	2%
Arrhythmia	3%	0	2%	0	11%	1%
Weight decreased	5%	0	3%	0	10%	1%

Abbreviations: DEC-C, oral decitabine and cedazuridine; URI, upper respiratory tract infection

^aAdapted from Otsuka INQOVI (decitabine and cedazuridine tablets) [drug label] (12).

^bIncludes grouped terms. See Supplementary Table S2 for further information.

The most common laboratory abnormalities were decreased blood counts (Supplementary Table S3). The overall incidence of decreased blood counts worsening from baseline by one or more grades was comparable between DEC-C tablets and IV decitabine during Cycle 1. However, Grade 3–4 decreased blood counts worsening from baseline by one or more grades were slightly more common with DEC-C tablets than IV decitabine. For example, 65% of subjects had Grade 3–4 leukocytes and neutrophils decreased with DEC-C tablets (vs. 59% with IV decitabine) and 41% had Grade 3–4 hemoglobin decreased with DEC-C tablets (vs. 36% with IV decitabine). The only exception was Grade 3–4 platelet count decreased, which was slightly less common with DEC-C tablets compared to IV decitabine (65% vs. 67%). Continued treatment with DEC-C was associated with improved hematologic parameters over time.

Common (non-hematologic) laboratory abnormalities that were more frequent with DEC-C tablets during Cycle 1 were glucose increased, albumin decreased, alkaline phosphatase increased, ALT increased, calcium decreased, and AST increased (Supplementary Table S3). Grade ≥ 3 nonhematologic laboratory abnormalities were rare, both in Cycle 1 and in all cycles.

Regulatory Insights

The clinical studies of DEC-C tablets were primarily designed to demonstrate that oral administration over 5 days results in comparable exposure to IV decitabine daily for 5 days. However, response rates and achievement of transfusion independence were also assessed. Ultimately, the demonstration of comparable exposure, with supportive efficacy and safety data in Studies ASTX727-01-B and ASTX727-02 supported the regular approval of DEC-C tablets for adult patients with MDS. The once daily dose of DEC-C tablets for the first 5 days of each 28-day cycle was recommended. This provides an oral option for decitabine administration which may reduce time in clinic compared to the IV formulation, a particular benefit during the current COVID-19 pandemic (8). Based on the fasted conditions used in Studies ASTX727-01-B and ASTX727-02, patients should avoid food 2 hours before and after each DEC-C tablet dose. Although patients in the studies received one cycle of IV decitabine in either Cycle 1 or 2, the observed efficacy was mostly attributed to DEC-C tablets since 1 cycle of IV decitabine is not considered sufficient to achieve a CR.

The traditional response endpoint appropriate for regulatory decision-making for higher-risk MDS is overall response rate, defined as CR + PR responses. There were no PR responses in Studies ASTX727-01-B or ASTX727-02; therefore, only CR responses were considered for demonstration of clinical benefit. Hematologic improvement and marrow CR responses were not considered, as their association with clinical benefit has not been established (9). Of note, the pivotal trials of IV decitabine used IWG 2000 criteria for response evaluation, whereas Studies ASTX7272-01-B and ASTX727-02 used IWG 2006 response criteria. The major differences are that the IWG 2000 criteria require an ANC of ≥ 1.5 Gi/L with no dysplasia for CR responses, whereas the 2006 criteria only require an ANC of ≥ 1 Gi/L with allowance of persistent dysplasia (4, 7). Accounting for this change in response criteria, FDA assessed the CR rates with DEC-C tablets (18% in ASTX727-01-B and 21% in ASTX727-02) to be comparable to the CR rate of 9–15% and CR+PR rate of 16–17% previously reported for IV decitabine.

The CR responses in the ASTX7272-01-B and ASTX727-02 trials were sufficiently durable to demonstrate clinical benefit, with median duration of response (DOR) 8.7 months observed in ASTX7272-01-B and 7.5 months in ASTX7272-02. Although the durability of response appeared to be lower than with IV decitabine (median 9.6–14.7 months) (10), many of the responses were ongoing at the time of data cutoff. Therefore, the median DOR is expected to increase with longer follow-up. Furthermore, the CR benefit was supported by a transfusion benefit on both trials.

Safety data was pooled across Studies ASTX727-01-B and ASTX7272-02 given similar study designs and patient populations. For comparative purposes, FDA evaluated safety between the DEC-C tablets and IV decitabine arms during Cycle 1 of therapy. In the comparative analysis of safety in Cycle 1, Grade 3–4 decreases in blood counts (leukocytes, neutrophils, hemoglobin) were slightly more common in subjects treated with DEC-C tablets. Thus, it will be important for clinicians to closely monitor patients on DEC-C tablets, despite the convenience of oral administration at home. Overall, however, the safety profile of DEC-C tablets across all cycles was similar to that of IV decitabine, with no new adverse reactions identified. The risk of embryo-fetal toxicity with DEC-C tablets was unchanged from IV decitabine. Both myelosuppression and embryo-fetal toxicity are listed under Warnings and Precautions in the USPI.

Post-marketing studies in patients with moderate to severe hepatic impairment and severe renal impairment were required at the time of approval, due to absence of data in these patients receiving DEC-C tablets. Post-marketing assessments to evaluate the effect of food, and the effect of cedazuridine on QT interval prolongation were also issued, to inform future labeling.

Conclusions

The benefit-risk analysis of the results of Study ASTX727-01-B and ASTX727-02 supported regular approval of DEC-C tablets for treatment of adult patients with MDS and CMML. Approval was based on demonstration of comparable bioavailability to IV decitabine, CR rate comparable to IV decitabine, and conversion from transfusion dependence to independence. Both common and serious toxicities were similar to IV decitabine.