Summary Basis of Decision for Exjade

3.3.1 Pharmacodynamics

A pharmacodynamic study evaluated the dose response of overall iron balance, i.e., the difference between total iron intake (iron in consumed diet and medication) and iron excreted in urine and feces, following multiple oral doses of deferasirox. Negative iron balance was achieved at all three doses of active drug, and averaged approximately 0.119 mg/kg/day at the 10 mg/kg dose, 0.329 mg/kg/day at the 20 mg/kg dose, and 0.445 mg/kg/day at the 40 mg/kg dose. For each active dose, the fraction of iron excreted in urine was on average approximately 2-5% of the total dose.

The studies have confirmed that deferasirox is an active and effective iron-chelating compound in humans. There appears to be a dose-dependent effect of this compound, at least in the range of 20-30 mg/kg/day. Higher doses appear more effective than lower ones, and more so in patients with more severe iron overload at baseline. It is clear that the dosing of any iron-chelating agent needs to be adjusted based on the patient's response, since excessive chelation may lead to morbid events, while underchelation will result in progression of hemosiderosis.

3.3.2 Pharmacokinetics

Absorption

Following the administration of a single oral dose of 2.5 to 80 mg/kg of deferasirox, the maximum plasma concentration and exposure of deferasirox increased approximately proportionally with the dose, but with substantial variability. The mean half-life of the drug was between 11 and 19 hours.

The mean percentage of deferasirox chelated to iron ranged from 15 to 31%. The pharmacokinetic profile of deferasirox and its iron complex appears compatible with a once a day oral administration.

Drug accumulation occurred in all age groups after multiple dosing of deferasirox. Total drug exposure (AUC) was 60% higher after 2 weeks of treatment compared to Day 1 (90% CI for the ratio: 1.38, 1.85). After 4 weeks of deferasirox treatment, the AUC0-24 had doubled compared to Day 1 (90% CI for the ratio: 1.72, 2.31), showing a 25% further increase in the last 2 weeks (90% CI for the ratio: 1.08, 1.45).

A food effect on bioavailability was reported. Food increased the fraction available to the systemic circulation.

Distribution

Deferasirox was extensively bound (>98%) to plasma proteins. In blood, only 5% of the radiolabelled deferasirox was associated with red blood cells.

Metabolism

The main metabolic pathway of deferasirox was glucuronidation and the primary enzyme responsible was UGT1A1. Based on molecular mass and comparison with reference material, M3 was identified as the acyl glucuronide metabolite of deferasirox. Studies with radiolabelled deferasirox showed that M3 exposure accounted for 3.3% of the total radioactivity in plasma.

Oxidative metabolism by CYP enzymes was a minor elimination process, therefore any inhibition or induction of those enzymes by co-medications is not expected to significantly affect the pharmacokinetics of deferasirox.

Excretion

Deferasirox was primarily eliminated in the faeces (84% of the dose) with enterohepatic recirculation likely but not proven. The predominant radioactive compound excreted in the faeces was unchanged deferasirox. Renal excretion amounted to only 8% of the dose. In the collection period of 7 days, the mean total recovery amounted to 91%, with low inter-individual variability.

Drug Interactions

Potential for drug-drug interactions are considered to be minimal although this conclusion is based more on theoretical reasoning based on the extremely high protein binding of deferasirox and of its iron complex rather than on definitive studies.

Special Populations

The pharmacokinetics (PK) of deferasirox in children were significantly different from that in adults. In children under 6 years of age, blood levels of deferasirox were half that obtained in adults at equivalent doses, for reasons that remain to be elucidated. There were no studies in patients with significant renal or hepatic impairment although transaminase levels of approximately five times the upper limit of normal did not seem to affect the PK of deferasirox. The sponsor is currently conducting a single dose (20 mg/kg per day) PK study of deferasirox in patients with hepatic impairment and the results will be submitted when available.

3.3.3 Clinical Efficacy

The efficacy of Exjade was investigated primarily in two studies, Study 107 and Study 108.

Study 107 was a one-year, multi-centre, open-label, randomized Phase III active comparator control study that compared Exjade (deferasirox) to deferoxamine in patients ≥ 2 years of age with β thalassemia and transfusional hemosiderosis. The starting doses were based on LIC (liver iron concentration) at baseline [2-3, >3-7, >7-14 and >14 mg Fe/g dry weight (dw)]. Patients treated with deferasirox (n=296) received starting doses of 5, 10, 20 or 30 mg/kg orally once daily and patients treated with deferoxamine (n=290) received subcutaneous starting doses of 20 to 60 mg/kg for at least 5 days per week. The ratio of deferasirox to deferoxamine doses for the two lower LIC categories was disproportionately low (1:4) compared to the two upper LIC categories (1:2). The primary efficacy endpoint, was defined as a reduction in LIC of ≥ 3 mg Fe/g dw for baseline values ≥ 10 mg Fe/g dw, reduction of baseline values between 7 and <10 to <7 mg Fe/g dw, or maintenance or reduction for baseline values <7 mg Fe/g dw.

Non-inferiority of Exjade (deferasirox) to deferoxamine in terms of success rates based on changes in LIC was not shown in the overall study population due to the disproportionately low dosing of deferasirox relative to deferoxamine in patients with baseline LIC <7 mg Fe/g dw. Non-inferiority was demonstrated in patients with LIC 7 mg Fe/g dw in which doses of deferasirox and deferoxamine were used as specified in the protocol. On average, 20 mg/kg deferasirox in regularly transfused adult and pediatric β-thalassemia patients was able to maintain iron balance whereas a negative iron balance was induced at 30 mg/kg. Doses of 5 and 10 mg/kg deferasirox on average did not maintain iron balance in this regularly transfused population. Dose-dependent responses in all age groups were consistently observed for the secondary efficacy parameters including absolute changes in LIC, serum ferritin and iron balance, and also for liver pathology. Serum ferritin trends over time closely reflected changes in LIC supporting the use of repeated measurements of this parameter to monitor iron accumulation and removal during chelation therapy.

Study 108 was a one year, multi-centre, open-label, non-comparative, Phase II study of Exjade (deferasirox) in patients ≥ 2 years of age with chronic anemias and transfusional hemosiderosis unable to be treated with deferoxamine. Patients (n=184) received 5, 10, 20, or 30 mg/kg per day of deferasirox based on baseline LIC. The primary efficacy endpoint was to demonstrate a success rate significantly greater than 50% with deferasirox.

The overall success criteria based on changes in LIC were not reached in the combined disease groups due to a relatively high drop-out rate due to the underlying disease and inappropriately low dosing of deferasirox in patients with baseline LIC <7 mg Fe/g dw. These success criteria were met in the combined disease groups in patients with LIC ≥ 7 mg Fe/g dw in which doses of 20 and 30 mg/kg deferasirox were employed. On average, 20 mg/kg deferasirox in regularly transfused adult and pediatric patients was able to maintain iron balance in both disease groups whereas a negative iron balance was induced at 30 mg/kg. Doses of 5 and 10 mg/kg, on average, did not maintain iron balance in this regularly transfused population. However, in patients with myelodysplastic syndrome (MDS), who had a lower transfusional iron intake, this dose response relationship was shifted such that 10 mg/kg deferasirox was able to maintain iron balance and marked negative iron balance was induced at 20 and 30 mg/kg. The analyses of iron balance indicate that the frequency and amount of blood transfusions, and hence the quantity of transfusional iron administered, has a direct impact on the dose of deferasirox needed to maintain or induce negative iron balance.

Both efficacy studies, Study 107 and Study 108, did not meet the criteria for their primary efficacy endpoints, but post-hoc subgroup analyses showed that patients with LIC ≥ 7 mg Fe/g  dw who were treated with 20 to 30 mg/kg per day Exjade had a significant reduction in LIC from baseline that was no different from deferoxamine. Dose dependent effects in serum ferritin and in the ratio of iron excretion to iron intake were also observed following doses of 5 to 30 mg/kg per day Exjade. The studies did not establish the efficacy of Exjade with moderate iron overload, i.e. LIC <7 Fe/g dw. Additional studies of Exjade at 20-30 mg/kg per day in these populations have been scheduled.

3.3.4 Clinical Safety

The safety of Exjade was investigated in several Phase II clinical studies (105, 106, 108, 109) and one Phase III study (107). Evidence regarding therapeutic benefit of deferasirox was based mainly on Study 107 which comprised of β-thalassemia patients aged ≥ 2 years, Study 108 which comprised of β-thalassemia patients and patients with other rare anemias aged ≥ 2 years [including elderly patients with myelodysplastic syndrome (MDS)], and Study 109 which comprised of patients with sickle cell disease aged ≥ 2years. Studies 107 and 108 (described in section 3.3.3 Efficacy) contributed 480 patients exposed to Exjade and Studies 105, 106, and 109 contributed another 220 patients treated with Exjade.

Deferasirox had a favourable safety profile with a low rate of discontinuation due to adverse events (AEs). The most frequently reported AEs most likely related to Exjade included transient, manageable, mild to moderate gastrointestinal (GI) complaints (abdominal pain, nausea, vomiting and diarrhea) and skin rash. Major adverse drug reactions that often required dose reductions involved renal, hepatobiliary, skin rash and ocular effects.

In Study 107, Exjade appeared to be associated with a higher incidence of GI adverse effects compared to deferoxamine. Effects on vision and hearing appeared to be comparable. In a number of cases, serious adverse reactions, such as cardiac rhythm disturbances and retinal degeneration, were not reported or discussed because they were deemed to be not drug related. The causality assessments for these serious adverse reactions were not explicitly explained in the study report. In approximately one third of the patients treated with Exjade , mild non-progressive increases in serum creatinine (generally within the normal range) were reported. Some of the increases were transient and in less than 10% of the patients precautionary dose reductions were performed. Four patients discontinued due to transaminase elevations. Probable drug-induced hepatotoxicity with a marked increase in transaminase levels occurred in at least one or possibly two patients. There were no episodes of drug-induced neutropenia, agranulocytosis or thrombocytopenia in either treatment group. There was no evidence of consistent relevant effects on any of the cardiac safety parameters measured. The results of vital signs and ECG monitoring were unremarkable. The higher incidence of renal effects, including increased serum creatinine and proteinuria, and hepatic transaminases, should be further characterized in adequate post-marketing studies.

Study 108 was a difficult study due to the number of various diseases with a wide spectrum of transfusion requirements and morbidity. In particular, the MDS group had significant withdrawal due to the complications of the underlying disease. This was true for all patients in the rare disease group. Additional long-term studies with MDS patients are required to better define the safety profile. The β-thalassemia population did not contribute any new information regarding Exjade Similar to Study 107, the major adverse finding (laboratory finding) was a non-progressive increase in serum creatinine in patients treated with Exjade which on average measured approximately 10 μmol/L.  Dose reductions were performed in 32 of the 73 patients (43.8%) who had >33% creatinine increases from baseline. Of the 32 patients who underwent dose reductions, only 8 patients returned to normal baseline serum creatinine levels. 

Study 109 assessed the safety of Exjade in patients with sickle cell disease. As in Study 107, patients received 5, 10, 20, or 30 mg/kg per day of Exjade or subcutaneous deferoxamine at doses of 20 to 60 mg/kg for five days per week based on baseline LIC. With a median exposure of six months, Exjade had an acceptable safety profile in patients ≥ 2 years old with sickle cell disease. The side effects of Exjade in patients with sickle cell disease were similar to that observed in patients with β-thalassemia. Mild, non-progressive increases in serum creatinine were observed in 25% of patients receiving Exjade. The most significant finding was that the renal toxicity was not more prevalent or serious in this group of patients with a high probability of renal pathology associated with their sickle cell disorders.

The safety profile in children less than 6 years of age was generally comparable to that seen in adolescents and adults with the exception of an increased incidence of upper respiratory tract symptoms, cough, rhinitis, fever, and diarrhea. This may be indicative of increased incidence of infection due to depression of the immune system, a finding supported by non-clinical data. The clinical program enrolled more than 1000 patients worldwide, but there are limited data on paediatric patients aged 2 to 5 (52 patients). A multi-national pediatric registry is being set up as a five-year observational study specifically to generate additional safety information in the 2- to 6-year age group of patients with transfusional hemosiderosis.

Additional studies are required to evaluate cardiac iron concentration and cardiac function, as well as clinical outcomes with patients treated with Exjade. The relationship between LIC and iron levels in the heart remains unclear but needs to be investigated as most inadequately chelated transfusion-dependent patients succumb to fatal cardiotoxicity.

Although no adverse effects were observed in a variety of anthropomorphic measurements or in indices of sexual development, the one-year studies are not sufficiently long in duration nor sufficiently powered to definitively conclude that Exjade is devoid of adverse effects in this respect. Overall, the relatively small numbers of patients investigated for each dose level and age group precluded the detection of adverse effects that occur with an incidence below 5%. Thus, more infrequently occurring potential adverse events, such as agranulocytosis and neutropenia, (toxicities associated with deferiprone, an oral iron chelating agent marketed in Europe) would not have been identified. Furthermore, studies longer than one year in duration are required to determine whether reduction in morbidity and mortality is similar to that achieved with deferoxamine patients. In recognizing the need for additional long-term studies, the sponsor has implemented four-year extensions to Studies 105, 106, 107, 108 and 109.

3.3.5 Issues Outstanding

Under the Notice of Compliance with Conditions (NOC/c) policy, the sponsor is required to submit the results of the following confirmatory studies:

• Extension studies 105E2, 106E1, 107E1, 108E1 and 109E1 for a total of four years after the core trial.
  
  
• A registry for children aged 2 to <6 years to enroll approximately 200 patients with a 5-year follow-up.
  
  
• A single dose (20 mg/kg per day) pharmacokinetics study of Exjade in subjects with hepatic impairment.
  
  
• A single-arm study in patients (adults and pediatrics) with congenital or acquired anemias and chronic iron overload to obtain additional data in patients with LIC <7 mg Fe/g dw treated with Exjade doses of 20 or 30 mg/kg per day.
  
  
• A study of long-term follow-up (3 years) in 150 patients with myelodysplastic syndromes (MDS) receiving Exjade including patients with baseline serum creatinine values up to two times the upper limit of normal to evaluate safety and haematologic and clinical benefit of Exjade in these patients.
  
  
• Studies of cardiac iron concentration, cardiac function, as well as clinical outcomes to explore the relationship amongst iron body burden as measured by LIC, cardiac iron, serum ferritin, and transfusion rates in patients treated with Exjade, using a validated methodology.

The sponsor has agreed to conduct additional post-market safety and mechanistic studies to further characterize the safety profile of Exjade and final reports will be provided upon completion. The sponsor will include monitoring and follow-up of adverse events from these studies as well as events noted in post-marketing data and in the major safety concerns identified. The sponsor has agreed to submit Periodic Safety Reports (PSURs) until the conditions have been fulfilled and removed from the NOC/c by Health Canada.