The European Medicines Agency Review of Decitabine (Dacogen) for the Treatment of Adult Patients With Acute Myeloid Leukemia: Summary of the Scientific Assessment of the Committee for Medicinal Products for Human Use

This review focuses on the European Medicines Agency approval for decitabine, a hypomethylating agent for the treatment of adult patients age 65 years and older with newly diagnosed de novo or secondary AML who are not candidates for standard induction chemotherapy.

Abstract

On September 20, 2012, a marketing authorization valid throughout the European Union (EU) was issued for decitabine for the treatment of adult patients aged 65 years and older with newly diagnosed de novo or secondary acute myeloid leukemia (AML) who are not candidates for standard induction chemotherapy. Decitabine is a pyrimidine analog incorporated into DNA, where it irreversibly inhibits DNA methyltransferases through covalent adduct formation with the enzyme. The use of decitabine was studied in an open-label, randomized, multicenter phase III study (DACO-016) in patients with newly diagnosed de novo or secondary AML. Decitabine (n = 242) was compared with patient’s choice with physician’s advice (n = 243) of low-dose cytarabine or supportive care alone. The primary endpoint of the study was overall survival. The median overall survival in the intent-to-treat (ITT) population was 7.7 months among patients treated with decitabine compared with 5.0 months for those in the control arm (hazard ratio [HR], 0.85; 95% confidence interval [CI], 0.69–1.04; p = .1079). Mature survival data after an additional year of follow-up were consistent with these results, with a median overall survival of 7.7 months in patients treated with decitabine and 5.0 months in the control arm (HR, 0.82; 95% CI, 0.68–0.99; p = .0373). Secondary endpoints, including response rates, progression-free survival, and event-free survival, were increased in favor of decitabine when compared with control treatment. The most common adverse drug reactions reported during treatment with decitabine are pyrexia, anemia, thrombocytopenia, febrile neutropenia, neutropenia, nausea, and diarrhea. This paper summarizes the scientific review of the application leading to approval of decitabine in the EU. The detailed scientific assessment report and product information (including the summary of product characteristics) for this product are available on the EMA website (http://www.ema.europa.eu).

Implications for Practice:

Acute myeloid leukemia (AML) remains an area of significant unmet need, especially in older patients. Older patients and those with comorbidities are often considered ineligible for standard induction therapy, and outcome for these patients is poor. Decitabine has favorable effects in terms of overall survival, which were considered clinically meaningful in the context of a manageable toxicity profile and after consideration of the lack of therapeutic alternatives for these patients. Decitabine is widely used in the treatment of AML in patients aged >60 years, as per current guidelines, including the European LeukemiaNet and the U.S. National Cancer Comprehensive Network.

Introduction

Acute myeloid leukemia (AML) is a clonal disorder caused by malignant transformation of a bone marrow-derived cell, myeloid stem cell, or progenitor cell. These myeloid leukemic blasts accumulate in the bone marrow, peripheral blood, and/or other tissues. The consequent reduction of normal bone marrow cells results in anemia, thrombocytopenia, and neutropenia. The clinical presentation of AML is directly related to these cytopenias; patients typically present with signs and symptoms of fatigue, hemorrhage, or infection [1]. Untreated, AML is a rapidly progressive and fatal disease that often requires prompt intervention [2].

Acute myeloid leukemia is the most common form of acute leukemia in adults [3] but remains a rare (orphan) disease. In the European Union (EU), the median age at diagnosis is 64 years, and approximately 18,000 new cases are diagnosed each year (roughly 0.6% of all cancers) [4]. Incidence rates are 3.62 cases of AML per 100,000 people for the period between 2000 and 2002 [5].

For 2001–2007, the 5-year survival rate for the AML population was 39.6% for patients <65 years of age and 5.2% for patients age 65 years or older [3]. In the older segment, median survival generally is short (2.4 months overall); it ranges from 3.9 months among those age 65–74 years to 1.4 months among those age 85 years or older [6–8]. Cytogenetic abnormalities represent an important prognostic factor for predicting remission rate, relapse risk, and survival outcome. The 5-year survival rates for patients with favorable, intermediate, and unfavorable risk cytogenetics were 55%–65%, 24%–41%, and 5%–14%, respectively [9–11]. For older patients, the survival rates were lower at 34%, 13%, and 2%, respectively [12].

Standard therapy for patients with AML is intensive combination induction chemotherapy, usually an anthracycline plus cytarabine, aiming to attain a complete remission and prolong survival. Older patients and those with comorbidities are often considered ineligible for standard induction therapy, and outcome for these patients is poor. According to recent guidelines from the European LeukemiaNet [13], the British Committee for Standards in Haematology [14], and the U.S. National Cancer Comprehensive Network [15], treatment options in older patients are guided by disease biology (cytogenetic abnormalities), performance status, and the presence of comorbidities (e.g., cardiovascular, cerebrovascular, pulmonary, hepatic, or renal dysfunction), which limit the efficacy of the therapy or the ability of the patient to tolerate cytotoxic induction chemotherapy.

Currently, no single standard of care exists for older patients with AML. Treatment options can include the intensive standard anthracycline plus cytarabine regimen, low-dose cytarabine, the hypomethylating agent azacitidine, or best supportive care. Azacitidine is approved in patients with marrow blast percentages between 20% and 30% who fulfill the criteria for AML according to the 2008 World Health Organization (WHO) classification. Despite improvements in outcomes for younger patients in recent decades, there has been little progress in improving prognosis for patients age ≥60 years [7, 8]. AML, especially in older patients, remains an area of significant unmet need.

This review focuses on the EMA approval for decitabine (Janssen-Cilag International N.V., B-2340 Beerse, Belgium, http://www.janssen.com/), a hypomethylating agent for the treatment of adult patients age 65 years and older with newly diagnosed de novo or secondary AML who are not candidates for standard induction chemotherapy. In a treatment cycle, decitabine is administered at a dose of 20 mg/m² body surface area by i.v. infusion over 1 hour daily for 5 consecutive days. The cycle should be repeated every 4 weeks depending on the patient’s clinical response and observed toxicity

Nonclinical Aspects and Clinical Pharmacology

Decitabine (5-aza-2′-deoxycytidine) is a cytidine deoxynucleoside analog (Figure 1). The antitumoral activity of decitabine was evaluated in vitro in several leukemic cell lines and solid tumors, where the 50% inhibitory concentration ranged from 10 nM to 10 μM. Concentrations of decitabine optimal to induce DNA hypomethylation reached a plateau above concentrations of 0.5 μM to 1 μM, whereas higher concentrations increased cytotoxicity [16–19].

Figure 1.

Structure of decitabine.

Decitabine specifically inhibits DNA methyltransferases (DNMTs), leading to hypomethylation of DNA including gene promoter regions, which might lead to re-expression of silenced tumor suppressor and other genes involved in cellular proliferation and differentiation. In many leukemia cell lines, decitabine induced morphological and functional differentiation at concentrations that inhibited cellular DNA methylation [16]. Decitabine also induces cytotoxicity by incorporation into DNA and the generation of DNA-DNMT adducts that interfere with DNA synthesis during cell replication, which results in apoptosis. In addition, it affects angiogenesis, decreasing vessel formation in different tumor models. The relative importance of the different mechanisms of action remains under investigation.

In vivo, decitabine increased the survival of treated animals in several mouse leukemia models (AKR, P388, L1210), as well as in a rat myeloid leukemia (BNML) model, when administered at intermittent dose schedules that were well-tolerated [20–27].

Multicycle repeated-dose toxicity studies in rats and rabbits indicated that the primary toxicity was myelosuppression, which was reversible on cessation of treatment [28–32]. Gastrointestinal toxicity was also observed. In males, the testes was a target organ in all animal species tested (mouse, rat, rabbit, and dog). In one mouse study in which decitabine was administered via i.v. route at 0.25 mg/kg per day for 5 days per week for 4 weeks, testicular degeneration was not reversible after a 4-week recovery period. In a rabbit study in which decitabine was administered via i.v. route as a 3-hour infusion 3 times a day for 4 cycles of 3 consecutive days at up to 4.8 mg/kg per day, testicular atrophy did not reverse after a recovery period of up to 6 weeks.

Decitabine administration to neonatal/juvenile rats showed a general toxicity profile similar to that seen in adult rats [33–35]. Neurobehavioral development and reproductive capacity were unaffected when neonatal/juvenile rats were treated at dose levels inducing myelosuppression. Decitabine has genotoxic potential, as expected from its interaction with DNA. No formal carcinogenicity studies were conducted as part of the decitabine development program on the basis of its intended clinical use as an anticancer drug for treatment of older patients with AML who have limited life expectancy; however, evidence from the literature indicates that decitabine has carcinogenic potential. Available data also indicate that decitabine has adverse effects on all aspects of the reproductive cycle, including fertility, embryo-fetal development, and postnatal development [36–39]. Most toxicities observed from treatment with decitabine are likely to be related to the cytoxicity of the molecule.

The population pharmacokinetics (PK) parameters of decitabine after 20 mg/m² daily as a 1-hour i.v. infusion for 5 consecutive days every 4 weeks were evaluated in two phase II studies and one phase III study in 45 patients with AML or myelodysplastic syndrome (MDS). Decitabine exhibited linear PK, and steady-state concentrations were reached within 0.5 hour. The steady-state volume of distribution was 116 L for a 70-kg cancer patient with body surface area of 1.73 m², indicating distribution of the drug into peripheral tissues. Median maximum concentration, area under the curve, and plasma half-life were 107 ng/mL (0.47 µM), 580 ng⋅hour/mL, and 68.2 minutes, respectively. The in vitro plasma-binding protein was negligible (<1%). Additionally, decitabine is a poor P-glycoprotein (P-gp) substrate, as well as a poor P-gp inhibitor. There were no dependencies on age, creatinine clearance, total bilirubin, or disease.

Intracellularly, decitabine is activated through sequential phosphorylation via phosphokinase activities to the corresponding triphosphate, which is then incorporated into DNA by DNA polymerases. In vitro, decitabine metabolism is not mediated by the P450 enzyme system but by oxidative deamination in the liver, kidneys, intestinal epithelium, and blood. As a consequence, CYP-mediated drug-drug interactions are unlikely. The mean plasma clearance after i.v. administration in cancer patients was 298 L/hour. The predominant route of excretion is renal, with approximately 90% excreted in the urine. Excretion of unchanged drug in the urine is low (4.2%), and the major circulating metabolites are not believed to be pharmacologically active.

Clinical Efficacy

The pivotal study DACO-016 was a randomized, open-label, multicenter phase III study of decitabine versus supportive care or low-dose cytarabine [40]. Eligible patients were aged 65 years or older; had newly diagnosed histologically confirmed de novo or secondary AML; and had a poor- or intermediate-risk cytogenetic profile, Eastern Cooperative Oncology Group (ECOG) performance status of 0–2, and adequate organ function determined by laboratory evaluation. Patients were excluded if they had acute promyelocytic leukemia or any other active systemic malignancy; had received previous chemotherapy (except hydroxyurea), including azacitidine, cytarabine, or decitabine, for any myeloid disorder; or were potential candidates for bone marrow or stem cell transplant within 12 weeks after randomization.

Decitabine was administered at a dose of 20 mg/m² by 1-hour i.v. infusion once daily for 5 consecutive days every 4 weeks. Cytarabine was administered at a dose of 20 mg/m² by subcutaneous injection once daily for 10 consecutive days, repeated every 4 weeks. Patients in both arms received supportive care and could continue on randomized therapy until they were no longer deriving clinical benefit. Patients were stratified by age (65–69 vs. >70 years), ECOG performance status (0–1 vs. 2), and cytogenetic risk group (poor versus intermediate).

DACO-016 enrolled 485 patients from 65 international sites, of whom 242 received decitabine and 243 received control treatment (TC). Demographic and baseline characteristics are presented in Tables 1 and 2.

Table 1.

Demographic and baseline characteristics (DACO-016 study) (intent-to-treat analysis population)

Table 2.

Baseline disease characteristics (DACO-016 study) (intent-to-treat analysis population)

The primary endpoint was overall survival (OS), measured from the date of randomization to the date of death from any cause or the last date known to be alive. At the time of the primary analysis, 396 patients had died: 197 in the decitabine arm and 199 in the control arm (clinical cutoff date [CCO], October 2009). The median OS in the ITT population was 5.0 months (95% confidence interval [CI], 4.3–6.3 months) in the TC arm and 7.7 months (95% CI, 6.2–9.2 months) in the decitabine arm. The estimated hazard ratio (HR; decitabine arm/TC arm) was 0.85 (95% CI, 0.69–1.04; p = .1079). In an analysis with an additional year of mature survival data (CCO, October 2010), the effect of decitabine was consistent with these results, with a median OS of 5.0 months in the TC arm and 7.7 months in the decitabine arm (HR, 0.82; 95% CI, 0.68–0.99; p = .0373).

Secondary endpoints included clinical response based on morphologic complete remission (CR) or CR with incomplete platelet recovery (CRp). In the decitabine arm, 17.8% of patients achieved CR or CRp compared with 7.8% in the TC arm (odds ratio, 2.5; 95% CI, 1.40–4.78; p = .0011). For patients who achieved CR or CRp in the decitabine group, the responses were achieved in a median of 4.3 months and were durable for a median of 8.3 months. Results on event-free survival (HR, 0.75; 95% CI, 0.62–0.91; p = .0029) and progression-free survival (PFS) (HR, 0.75; 95% CI, 0.62–0.91; p = .0036) supported the primary analysis. A summary of efficacy results is shown in Table 3 and Figure 2.

Table 3.

DACO-016 summary of efficacy results (intent-to-treat analysis)

Figure 2.

Overall survival data (DACO-016 study: intent-to-treat population; clinical cutoff date, 2009).

Abbreviations: CI, confidence interval; HR, hazard ratio; TC, patient's choice of treatment with physician's advice.

Subgroup analyses by age, ECOG performance status, cytogenetic risk, and geographic region are shown in Figure 3. Median OS varied considerably across geographic regions, particularly in western Europe, where median survival in the TC arm was longer.

Figure 3.

Overall survival subgroup analysis (clinical cutoff date, 2009) (DACO-016 study: intent-to-treat analysis set).

Note: p value is based on two-sided log rank test stratified by age, cytogenetic risk, and ECOG performance status.

Abbreviations: AML, acute myeloid leukemia; Aus, Australia; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; Med, median months; TC, patient's choice of treatment with physician's advice.

Results from the supportive study DACO-017 were consistent with the pivotal study. DACO-017 was a multicenter, single-arm, phase II study of decitabine as front-line therapy in older patients with AML who were at least 60 years of age with newly diagnosed histologically confirmed de novo or secondary AML and who had poor- or intermediate-risk cytogenetics at baseline. Eligible patients had an ECOG performance status of 0–2 and were considered to have adequate renal and hepatic function by laboratory evaluation. Treatment consisted of decitabine 20 mg/m², administered as a 1-hour i.v. infusion daily for 5 consecutive days every 4 weeks. The primary efficacy endpoint, morphologic complete remission, was achieved in 13 of 55 patients (23.6%) with a median time to CR of 4.1 months. The secondary endpoint overall survival was 7.6 months (95% CI, 5.7–11.5 months).

Clinical Safety

The safety dataset for decitabine comprised 293 patients with AML (DACO-016 and DACO-017 studies) and 192 patients with MDS (DACO-020, ID03-0180, D-0007, EORTC-06011 studies). Although MDS and AML can be considered a continuum of the same myeloid malignancy, the two diseases differ in terms of prognosis and patient characteristics. Therefore, the integrated analysis in AML studies was considered more relevant to the claimed indication.

The most frequently observed adverse drug reactions (ADRs) in AML patients in the decitabine group versus control treatment were directly related to myelosuppression (thrombocytopenia, 41% vs. 37%; anemia, 38% vs. 31%; and neutropenia, 32% vs. 21%) or its consequences (pyrexia, 48% vs. 37%; febrile neutropenia, 34% vs. 23%; diarrhea, 31% vs. 23%; and pneumonia, 24% vs. 21%). Nausea occurred in 33% patients (Table 4). ADRs were defined as treatment-emergent serious adverse events occurring in more than 5% of patients in any treatment group and reported more frequently in the decitabine group compared with the cytarabine group in DACO-016. In addition, anemia was classified as treatment-related by investigators, although the frequency was equivalent to that in the cytarabine group. The most common grade 3/4 ADRs (≥20%) included pneumonia, thrombocytopenia, neutropenia, febrile neutropenia, and anemia. ADRs leading to dose delays were primarily due to myelosuppression or infection (16% and 18%, respectively). The incidence of ADRs leading to dose reductions and treatment discontinuation was 4% and 7%, respectively. The decitabine treatment group had a higher incidence of treatment discontinuation because of adverse events in women compared with men (43% vs. 32%).

Table 4.

Adverse drug reactions identified with decitabine

Thirty percent of AML patients treated with decitabine and 25% of patients treated with the comparator had adverse events with an outcome of death during treatment or within 30 days after last dose. The most common adverse events leading to death in AML patients were septic shock (4% in the decitabine arm vs. 3% in the TC arm), sepsis (3% vs. 1%), and pneumonia (2% vs. 4%). Serious bleeding-related adverse reactions were reported in the context of severe thrombocytopenia, some of which led to a fatal outcome, such as central nervous system hemorrhage (2% in the decitabine arm vs. 1% in the TC arm) and gastrointestinal hemorrhage (2% vs. 5%).

Discussion and Benefit-Risk Assessment

An increase in median survival of 2.7 months was considered clinically meaningful in the context of a manageable toxicity profile and after consideration of the lack of therapeutic alternatives for these patients. The primary efficacy results were supported by increased response rates (CR + CRc), progression-free survival, and event-free survival in favor of decitabine when compared with control treatment, which included supportive care or low-dose cytarabine. In addition, mature survival data collected for an additional year of follow up was highly consistent with those of the primary analysis.

The pivotal study has demonstrated favorable effects in terms of overall survival, considered the preferred primary outcome measure with unbiased results in this group of patients with very poor prognosis. Moreover, secondary and tertiary endpoints, including clinical response, showed a highly significant benefit in favor of decitabine and were durable. Although statistical significance was not reached for overall survival at the time of the primary analysis (HR, 0.85; 95% CI, 0.69–1.04; p = .1079, the submission of a single pivotal study has been considered sufficient because it has been supported adequately with respect to internal and external validity, clinical relevance, data quality, and internal consistency. The design of the pivotal study followed previous Committee for Medicinal Products for Human Use scientific advice given to the applicant on key design and analysis aspects, including dosing regimen, choice of comparator, study endpoints, and patient selection criteria.

Potential factors that could have contributed to the lack of statistical significance in the primary endpoint in the pivotal study included (a) imbalance in subgroup characteristics, (b) use of subsequent therapy, and (c) data censoring in the period following the median of the curve. In this regard, sensitivity analyses censoring data at the time patients received subsequent therapies (induction chemotherapy, azacitidine, and decitabine) showed a 20% reduction in the risk for death in patients in the decitabine arm, with an increased median survival of 3.2 months (HR, 0.80; 95% CI, 0.64–0.99; p = .0437). Interestingly, the change in results might have been driven by the western European region (17.5% of patients), where a strong qualitative interaction that was initially present disappeared after the CCO 2010 reanalysis. It is considered that the large crossover to the other hypomethylating agent azacitidine in the control arm in the western European (32.4%) compared with the non-western European regions (4.3%) may have led to the difference in the results between regions. In addition, results for the endpoints PFS and response rate, which are not influenced by subsequent therapy, are in line with other regions’ results and the overall study results. Furthermore, clinical experience with the hypomethylating agent azacitidine has recently shown an increase in median OS in older AML patients treated with azacitidine compared with conventional care regimens [41, 42].

Currently, an ongoing phase III trial is further investigating decitabine in older patients with AML (≥60 years; NCT01633099), with the aim to study the therapeutic effect and safety of a 10-day regimen on the basis of the higher response rates in the single-center phase II study with the 10-day regimen (NCT01786343). In addition, ongoing pediatric development is evaluating decitabine for treatment of pediatric patients with acute myeloid leukemia and high-risk cytogenetics or as a second-line treatment.

Of note, decitabine was not recommended for approval in AML in the United States after FDA review in 2012. After an Oncologic Drugs Advisory Committee meeting, it was concluded that the pivotal study failed to demonstrate benefit based on rigorous statistical interpretation (HR, 0.85; 95% CI, 0.69–1.04; p = .1079). Although panel members generally agreed that statistical significance is not imperative for demonstration of clinical benefit, it was viewed as a major component in assessing the big picture of supportive evidence, especially for single trials [43]. On the contrary, EMA acknowledged the lack of statistical significance; however, it considered the totality of evidence presented, including secondary endpoints, mature survival data, and the unmet medical need in this patient population and deemed the results clinically meaningful. Further debate on the different regulatory positions has taken place in the AML scientific community since the publication of the regulatory decisions [43, 44].

During the assessment, discussions also focused on the patient population included in the pivotal studies because for some patients included in the study, standard induction therapy might have been beneficial. According to the applicant, only 6% of the enrolled population could have been candidates for induction therapy on the basis of being age ≥75 years or ≥60 years with one of the following: ECOG performance status >2, poor-risk cytogenetics, presence of comorbidities, or unwillingness to undergo intensive chemotherapy. Although it was acknowledged that decitabine had not been intended to replace standard induction treatment, further analyses showed borderline significance in terms of OS for this patient population. In addition, the pivotal trial included only patients ≥65 years of age, and therefore benefit with decitabine has been shown only for this specific patient population. In line with the study population, it was agreed that the use of decitabine should be limited to adult patients age 65 years and older with newly diagnosed de novo or secondary AML according to the WHO classification who are not considered candidates for standard induction chemotherapy.

In terms of safety, adverse reactions with decitabine are predominantly myelosuppressive. The safety profile of decitabine was similar to that of low-dose cytarabine, and although some undesirable effects were seen more often with decitabine (such as infections and febrile neutropenia), the safety profile may be due to the longer exposure to decitabine treatment (median, 4.4 cycles of decitabine versus 2.4 cycles of cytarabine). The use of supportive anti-infective agents and cytokines was also similar in both treatment arms. Overall, the safety profile of decitabine is globally in accordance with what could be expected from a cytotoxic compound in such a setting. In addition, there is an extensive understanding of its risk profile because decitabine has been approved in the United States for use in patients with MDS since 2006. Therefore, on the basis of the totality of evidence available, the benefit-risk balance for decitabine was considered positive in view of the clinically relevant effect in terms of overall survival and an acceptable safety profile.

Author Contributions

Data analysis and interpretation: Pierre Demolis, Eliane Béhanzin, Alexandre Moreau, Ian Hudson, Beatriz Flores, Henry Stemplewski, Tomas Salmonson

Manuscript writing: Maria Nieto, Christian Gisselbrecht, David Bowen, Francesco Pignatti

Final approval of manuscript: Maria Nieto, Pierre Demolis, Eliane Béhanzin, Alexandre Moreau, Ian Hudson, Beatriz Flores, Henry Stemplewski, Tomas Salmonson, Christian Gisselbrecht, David Bowen, Francesco Pignatti

Disclosures

The authors indicated no financial relationships.