| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Artemether-lumefantrine (coartemether; Coartem) (Novartis Pharma AG, Basel, Switzerland) is currently the only fixed-combination ACT widely available. Each tablet consists of 20 mg artemether and 120 mg lumefantrine. The rationale for this combination is that artemether rapidly reduces parasite biomass, and the long-acting lumefantrine eliminates residual parasites. Accumulating data suggest that, for optimum efficacy, coartemether should be given as six doses and with food containing fat.8 Although the four-dose regimen seemed sufficient to provide good cure rates in regions where patients were semi-immune and in areas without multidrug resistance,9 WHO expressed the desire to have a single global dosing regimen for coartemether to avoid confusion and to assure its long-term effectiveness. However, the safety and tolerability of the higher dose regimen is an obvious concern in using the six-dose regimen as the global standard. Clinical studies investigating the six-dose regimen of coartemether showed high day 28 parasitological cure rates and good tolerability in adults and children from multi-drug–resistant areas of Southeast Asia and Africa.10–14 Moreover, a high adherence to the six-dose regimen was observed, with day 28 cure rates close to 100%, irrespective of whether given under supervision or under unsupervised conditions of routine clinic practice in Africa.13,15 In a head-to-head study performed in Thailand, the six-dose administration schedule of coartemether was more efficacious and equally well tolerated than the four-dose regimen.10 A recently performed pooled analysis of data from randomized clinical trials confirmed the superior efficacy of the six-dose over the four-dose regimen in adolescents and adults, without altering tolerability and safety (E.A. Mueller and others, unpublished data).
Similar reassuring comparative data from children are lacking. In a multicenter non-comparative study in Kenya, Nigeria, and Tanzania, 310 children were treated with the six-dose regimen of coartemether. The drug was well tolerated, and the cure rate corrected by polymerase chain reaction (PCR) at day 28 was approximately 94%.14 The study had no comparative four-dose regimen arm, for ethical reasons. It was considered that the four-dose regimen might not be sufficiently efficacious in nonimmune patients likely to be enrolled in that study. Against this background, we performed additional analyses on data assessed in children. Individual patient data from clinical trials performed under the auspices of coartemether’s manufacturer in pediatric patients weighing between 5 and 25 kg were pooled to facilitate comparisons between the four-dose and six-dose regimens.
The aim of this pooled analysis was to confirm that the six-dose regimen of coartemether is more efficacious than the four-dose regimen in pediatric patients, while remaining equally safe and well tolerated. The results of the analysis were used for supporting the registration of the six-dose regimen as a single global dosing regimen, irrespective of age, body weight, immunologic status, and anti-malarial multi-drug resistance.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
).
|
12 years; 2) body weight 5–25 kg; 3) baseline parasitemia 1,000–100,000/µL; 4) same individual dose of coartemether, according to body weight group allocation. These criteria were selected in a way to match the characteristics of the patients included in Study 2403, which was the pivotal study to gain regulatory approval for using the six-dose regimen in kids. It was designed in partnership with the WHO. All studies screened were conducted according to the Declaration of Helsinki with all its amendments. Informed consent was obtained from parents or legal guardians, with an appropriate institutional review board having approved the studies.
Study site, study design, and population. In the four-dose regimen group, there was a comparable distribution between African and Asian patients selected for pooling. Study 2403 was the primary source of data for the six-dose regimen and provided the vast majority of patients on this regimen. Studies 008, 025, and 026 included both adults and children. Although most studies were of open-label nature, the six-dose regimen group included a lower proportion of patients from double-blind studies than the four-dose group (6% versus 29%). Comparators included sulfadoxine/pyrimethamine (SP, Study 010), chloroquine (Study 011), quinine (Study 003), and the combination of mefloquine and artesunate (MAS, Studies 008 and 026). In one case (Study 025), both coartemether regimens were compared within a study, in which two different six-dose regimens were used (same total dose), with doses given over 3 or 5 days; both six-dose regimens were considered for pooling. Studies 009 and 011 used a pediatric tablet formulation of coartemether (one half the dose of a standard tablet) that was produced during the original development program but is no longer available. To maintain comparability between dose regimens, only patients who received individual coartemether doses equivalent to the marketed formulation (i.e., two or four pediatric tablets per dose depending on body weight) were considered for pooling.
Before enrollment into the studies, male and female patients presenting with symptoms suggestive of malaria were assessed to be suffering from acute, uncomplicated P. falciparum malaria, and the diagnosis was confirmed by a positive blood smear. Patients with signs and symptoms of severe or complicated malaria were excluded.
In most of the trials, study endpoints included parasitological cure rate on days 7, 14, and 28. Cure rates were not evaluated at day 28 in Studies 010 and 011. In Studies 003, 009, 025, 026, and 2403, PCR genotyping of paired blood samples was used to account for reinfections. Secondary outcomes included parasite clearance time (PCT), fever clearance time (FCT), and gametocyte carriage. Safety assessments consisted of monitoring and recording of all adverse events (AEs) and serious adverse events (SAEs), physical examinations, regular monitoring of hematology and blood chemistry, and regular recordings of electrocardiograms (ECGs) and vital signs. Two formulae (Fridericia and Bazett) were used to calculate the heart rate–adjusted QT interval (QTc).
Drug regimens.
Table 2 summarizes the drug regimen of coartemether by body weight used in the trials that provided patients for pooling. Each standard tablet of coartemether contained 20 mg of artemether and 120 mg of lumefantrine, whereas the pediatric tablet formulation contained only 10 mg of artemether and 60 mg of lumefantrine. For reasons outlined above, patients taking one or three pediatric tablets were excluded from pooling. Patients on the four-dose regimen received their doses at 0, 6–8, 24, and 48 hours, whereas patients on the six-dose regimen received coartemether at 0 and 8 hours, and then twice daily on each of the following 2 days. In one treatment arm of Study 025, the six doses were administered over 5 days. Patients were advised to take coartemether with fat-containing food, if tolerated.
|
Day 7 cure rate. Proportion of patients with clearance of asexual parasitemia within 7 days of initiation of trial treatment, without subsequent recrudescence within these 7 days.
Day 14 cure rate. Proportion of patients with clearance of asexual parasitemia within 7 days of initiation of trial medication, without subsequent recrudescence within 14 days after study start.
Day 28 cure rate. The 28-day cure rate was considered to be of greatest importance, defined as the proportion of patients with clearance of asexual parasitemia within 7 days of initiation of trial drug, without subsequent recrudescence within 28 days after study start. Recrudescence was defined as the existence of positive blood smears after initial clearance of parasites from the peripheral blood. When available, results from PCR analysis were used to distinguish between recrudescence and reinfection.
Parasite clearance time. Time from first dose until first total and continued disappearance of asexual parasite forms that remained for at least a further 48 hours.
Fever clearance time. Time from first dose until the first time the body temperature decreased below and remained less than 37.5°C for at least a further 48 hours.
Gametocyte carriage. Proportions of patients with gametocytes at a given time point.
Subgroup analyses for efficacy were performed for patients 5 to < 10, 10 to < 15, and 15 to 25 kg. Safety was analyzed in terms of AEs, SAEs, clinical laboratory evaluations, and ECGs. AEs were defined as events that occurred after baseline but before recurrence of malaria.
Statistical analysis.
All efficacy parameters were analyzed for the intent-to-treat (ITT) population, which was defined as all treated patients who met the pooling criteria and had confirmed malaria (i.e., baseline parasite count 
0). Of note, all patients who discontinued the study before evaluation of cure rate (day 7, 14, or 28) for any reason (including losses to follow-up) were considered a treatment failure in the ITT analysis for cure rates (i.e., conservative approach). Cure rates were also analyzed for the evaluable population (i.e., all patients in the ITT population who had parasite counts recorded up to the day of evaluation for cure rate: day 7, 14, or 28; if completed 7, 14, or 28 days) and took no other anti-malarial drugs effective on P. falciparum or discontinued before day 7, 14, or 28, as relevant, because of "unsatisfactory therapeutic effect" because of recurrence of P. falciparum. Safety was analyzed for the safety population, defined as all patients who received at least one dose of study medication.
For parasitological cure rates and gametocyte counts, a treatment comparison between the four- and six-dose regimen of coartemether was performed using the Fisher exact test, two-sided (P 0.05). Both, PCT and FCT were analyzed by survival analysis methods (Kaplan-Meier estimators with appropriate censoring and corresponding 95% confidence intervals [CIs] for median event times). Otherwise, all data analyses were done using summary descriptive statistics.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
shows the numbers of patients in each defined analysis population. The 5- to < 10-kg body weight group treated with the four-dose regimen was small because of the exclusion from this population of patients who received doses of one pediatric tablet.
|
); the difference mainly was in the proportion of patients discontinuing because of unsatisfactory therapeutic effect (0.6% versus 13.4%) and who were lost to follow-up (2.0% versus 11.9%). The latter may have been related to better study monitoring and conduct in more recent studies (with the six-dose regimen) than in earlier trials (with the four-dose regimen). There was no difference in the rates of discontinuation caused by AEs, which were low for both regimens (six-dose: one patient [0.3%]; four-dose: two patients [1.0%]). One patient on the six-dose regimen died, which was unrelated to treatment (severe gastroenteritis contracted at home 5 days after the end of study treatment).
|
). Patients treated with the six-dose regimen tended to be younger and of lower body weight than those treated with the four-dose regimen. This reflected two factors: first, the exclusion of most patients in the 5- to < 10-kg body weight group for the four-dose regimen because they received doses of one pediatric tablet, and second, there was an emphasis in Study 2403 on recruitment of patients in the lowest body weight group. Moreover, mean and median parasite density tended to be somewhat lower in patients who received the six-dose regimen than in those on the four-dose regimen.
Parasitological cure rates.
Cure rates (Table 5) with the six-dose regimen were consistently higher than the four-dose regimen in all analysis populations, with the difference being greatest at 28 days. PCR-corrected day 28 cure rates were as high as 96% for the six-dose regimen in the evaluable population. In both the ITT and evaluable populations, the overall difference between the two coartemether treatments in terms of day 28 cure rates (with and without PCR correction) was clinically and statistically significant (P < 0.0001).
|
Times to parasite and fever clearance. Median PCT was shorter with the six-dose (24.2 hours; 95% CI: 24.0, 35.8) than the four-dose regimen (45.0 hours; 95% CI: 40.0, 48.0). The proportion of patients with rapid parasite clearance (i.e., within 48 hours) was more than 80% for the two regimens (six-dose: 94.2%; four-dose: 83.1%). Subanalyses according to body weight yielded similar results (data not shown).
Analysis of FCT data suggested a more rapid resolution of fever with the six-dose coartemether treatment regimen, although results could have been confounded by the extensive use of antipyretic medications. For the six-dose regimen, median FCT was 7.9 hours (95% CI: 7.8, 7.9) and for the four-dose regimen was 24.0 hours (95% CI: 23.8, 26.5). Again, subanalyses according to body weight yielded similar results (data not shown).
Gametocyte counts.
The proportion of patients with gametocytes at early time-points (up to day 7) seemed to be lower for the six-dose regimen than the four-dose regimen, suggesting a rapid clearance of gametocytes, particularly with the six-dose regimen, or clearance of merozoites before gametocyte formation can occur (Table 6). Significantly fewer patients had circulating gametocytes at day 28 with the six-dose than the four-dose regimen. Subanalyses according to body weight were in line with the results obtained with the total population (data not shown).
|
The majority of patients reported at least one AE after baseline (Table 7). In both groups, most AEs were reported during the first 3 days of treatment, which were generally mild or moderate in severity. Severe AEs were infrequent (six-dose: 5.2%; four-dose: 7.0%). The most frequent AEs observed with both regimens included cough, anemia, anorexia, vomiting, hepato-/splenomegaly, headache, and diarrhea, most of which could have been disease related.20 The proportion of patients with AEs was similar between the 5 to < 10, 10 to < 15, and 15–25 kg body weight groups, with incidence rates of 75.5%, 72.8%, and 77.0%, respectively, for the six-dose regimen, compared with 87.5%, 91.1%, and 91.3%, respectively, for the four-dose regimen. There were some differences between body weight groups in particular AEs, such as headache or dizziness, which appeared less common in very small infants. This may have been related to the limited ability of these patients to communicate subjective symptoms without outward sign. Moreover, because the AEs accounting for the differences between body weight groups were generally of mild to moderate severity, those differences were not considered to be clinically relevant.
|
). With regard to AEs of special interest, fewer patients in the six-dose than the four-dose group showed AEs related to the central nervous system. Of these, headache (six-dose: 9.9%; four-dose: 40.3%) and dizziness (six-dose: 3.8%; four-dose: 16.4%) were by far the most frequent ones. Other events included clonus in 3.8% of patients taking the six-dose regimen and 1.0% of those treated with the four-dose regimen, hyperreflexia (six-dose: 1.7%; four-dose: 0.5%), convulsions (six-dose: 0.3%; four-dose: 1.0%), and nystagmus, ataxia, and coordination abnormal, all of which occurred in 1.0–1.5% of patients treated with the four-dose regimen, but were not reported with the six-dose regimen. Decreased hearing (hypoacusis) was reported in 3/201 (1.5%) of patients taking the four-dose regimen, but was not reported in patients taking the six-dose regimen. All cases of hypoacusis were of mild severity, and none were considered to be drug-related.
AEs suspected to be drug-related are shown in Table 8. Only a rather small proportion of patients who experienced AEs had events that were suspected to be drug-related. Such AEs were overall higher with the coartemether six-dose regimen than the four-dose regimen.
|
Electrocardiography.
For both coartemether treatment regimens, the incidence in QTc changes within specific ranges were overall comparable for the six-dose and four-dose regimens (Table 9). Specifically, the proportion of patients who had clinically non-significant QTc changes of less than 30 ms from baseline was greater for the six-dose than the four-dose regimen. On the other hand, the four-dose regimen had greater proportions of patients with QTc increases of 30–60 or more than 60 ms than the six-dose regimen. This was generally consistent when QTc was calculated using either Bazett or Fridericia formulae. Given the low number of patients with ECG assessments in the four-dose group, differences between treatment groups should be interpreted with caution.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The main outcomes of this pooled analysis of individual patient data from eight studies in children with acute, uncomplicated P. falciparum malaria, matched for demographic and baseline disease characteristics and individual coartemether doses, can be summarized as follows: 1) the parasitological cure rate at 28 days is clinically and statistically significantly higher with the six-dose than the four-dose regimen of coartemether; 2) the six-dose regimen is associated with a more rapid clearance of parasites and fever and a faster and more sustained reduction in gametocyte carriage; 3) patients in the six-dose group show a markedly lower discontinuation rate because of insufficient efficacy than patients on the four-dose regimen; 4) the six-dose regimen is equally well tolerated as the four-dose regimen in terms of AEs, SAEs, and laboratory parameters; 5) ECG changes (in terms of QTc prolongations) are infrequent and comparable between the two coartemether dose regimens.
Efficacy. This analysis showed that the six-dose regimen of coartemether is highly efficacious in the treatment of children with a body weight as low as 5 kg, including those who are likely to be non-immune. Analyses based on the ITT and evaluable populations yielded PCR-corrected day 28 cure rates of 93% and 96%, respectively. If compared with the four-dose regimen, cure rates were significantly higher with the six-dose regimen, irrespective of analysis population, whereas the results in the ITT population might have been slightly biased to the disadvantage of the four-dose regimen because of a higher proportion of patients being lost to follow-up, which were counted as treatment failures. Although PCT and FCT were shorter with the six-dose regimen, the proportion of patients with parasite clearance within 48 hours was high for both coartemether regimens. Given the fact that the two regimens are identical for the first 36 hours, this was not surprising.
Our findings confirmed the results from Falade and others14 who showed that the coartemether six-dose treatment of children with falciparum malaria is highly efficacious in endemic countries of Africa, using a non-comparative study design. Moreover, results from our analysis corroborate findings of a study in Thai adults that compared the six-dose and four-dose regimens directly10 and of a recently performed pooled analysis of data from adolescents and adults participating in randomized clinical trials (E. A. Mueller and others, unpublished data).
With regard to reduction of gametocyte carriage, in this analysis, the six-dose regimen acted faster and significantly fewer patients showed circulating gametocytes at day 28 than the four-dose regimen. Thus, the six-dose regimen may have a favorable impact on malaria transmission. Decreasing the transmission of any drug-resistant parasites that might have been selected is of major public health interest.5,21
Tolerability and safety. The safety profile of the six-dose regimen was not worse than that of the four-dose regimen. In fact, the six-dose regimen was well tolerated in the treatment of children down to 5-kg body weight. The observation of lower AE rates with the six-dose regimen is most likely a reflection of the fact that most of the observed AEs were related to malaria symptoms, and the better efficacy associated with the six-dose regimen allowed a more rapid resolution of malaria and thus prevented worsening of symptoms more effectively than the four-dose regimen. This is supported by the fact that most AEs were reported during the first 3 days of treatment. Moreover, the six-dose regimen group was composed of more very small children with a limited ability to communicate subjective symptoms.
Suspected drug-related AEs were overall more frequent with the six-dose regimen of coartemether, which may look surprising as a lower proportion of AEs of any cause was observed with that regimen. However, this finding has to be interpreted with caution because drug relationship, according to the investigator’s judgment, might have been assessed differently across studies. In addition, Study 2403, which provided most of the data on the six-dose regimen, followed an open, non-comparative design, in which the investigators knew they were prescribing a higher dose of coartemether than would previously have been used in the patient population represented in the study. This may have increased the index of suspicion. In contrast, almost one third of the patients in the pooled population for the four-dose regimen participated in double-blind comparative studies.
There has been some concern about the possible risk of neurotoxicity with artemisinin derivatives in animals and hearing decreases with coartemether in humans.22,23 In our pooled analysis in children, the six-dose regimen was not associated with a worse neurologic safety profile than the four-dose regimen. Headache and dizziness were by far the most frequent neurologic AEs and were almost certainly symptoms of malaria. Decreased hearing was not reported in patients taking the six-dose regimen, thereby confirming results of a recent study that indicated that coartemether has no influence on auditory function.24
SAEs were infrequent, and only one (severe urticaria) reported with the six-dose regimen was suspected to be drug-related. One patient died of causes unrelated to study treatment.
In terms of clinical laboratory parameters, the information was limited, but findings were consistent with acute malaria and its resolution after treatment, without major differences between coartemether treatment regimens. In the six-dose group, a lower decrease to day 3 in hemoglobin levels and hematocrit and a greater increase to day 28 was observed than in the four-dose group, most likely because of the better therapeutic effect with the six-dose regimen.
Lumefantrine has some chemical similarities with halofantrine, a drug known to cause QTc interval prolongation. Consequently, QTc values were evaluated in this pooled analysis. Although the number of patients with ECG assessments in the four-dose group was small, the six-dose regimen of coartemether did not seem to be associated with a greater frequency of QTc prolongation than the low rate observed with the four-dose regimen. Significant prolongation of the QTc interval has been observed in young African children with malaria (at day 4 versus baseline) in patients being treated with coartemether, chloroquine, or sulfadoxine/pyrimethamine (SP). The QTc prolongation seemed to be independent of the treatment given but showed a weak correlation to both baseline parasitemia and temperature.25 In view of this, and the electrocardiographic safety of coartemether shown before in healthy volunteers and malaria patients,26,27 it seems unlikely that the administration of the six-dose regimen of coartemether represents an increased cardiac risk in the patient population studied.
Limitations. There was an imbalance between coartemether treatment groups with regard to the proportion of patients from double-blind studies compared with open-label studies. This might have influenced the assessment of drug-relationship of AEs. There were too few patients treated with the four-dose regimen in the 5- to < 10-kg body weight group to allow meaningful comparisons with the six-dose regimen for that particular body weight range. Moreover, Study 2403 provided the vast majority of patients on the six-dose regimen. Hence, the patient sample providing data for that regimen might have been not fully representative of patients with falciparum malaria from other geographical regions.
Conclusion. Based on this pooled analysis, the six-dose administration schedule of coartemether is more efficacious than the four-dose regimen in children with uncomplicated P. falciparum malaria while remaining equally well tolerated. The six-dose regimen is associated with high parasitological cure rates even in non-immune infants of body weight as low as 5 kg. This pooled analysis supports the harmonized use of the six-dose regimen of coartemether in children as single global dosing regimen.
Received December 21, 2005. Accepted for publication February 6, 2006.
Acknowledgments: Biometry and statistical analysis was carried out by DATAMAP GmbH, Freiburg, Germany. We are grateful to Mrs. Mailis Virtanen for her help and support in conducting this analysis.
Financial support: This pooled analysis was funded by Novartis Pharma AG, Basel, Switzerland.
Disclosure: Some of the authors are Novartis employees and have no conflicts of interest concerning the work reported in this paper.
* Address correspondence to Edgar A. Mueller, Institute for Clinical Pharmacology, Public Health Research Association Saxony, Medical Faculty, Technical University, Fiedlerstrasse 27, D-01307 Dresden, Germany. E-mail: Edgar.Mueller{at}mailbox.tu-dresden.de 
Authors’ addresses: Michael Makanga, European and Developing Countries Clinical Trials Partnership (EDCTP), PO Box 19070, Francie van Zijl Drive, Parow Valley 7500, Cape Town, Tygerberg 7505, South Africa. Zul Premji, Department of Parasitology and Entomology, Muhimbili Medical Centre, PO Box 65001, Dar es Salaam, Tanzania. Catherine Falade, Department of Pharmacology and Therapeutics, College of Medicine, University of Ibadan, Ibadan, Nigeria. Juntra Karbwang, WHO, Avenue Appia 20, CH-1211 Geneva 27, Switzerland. Edgar A. Mueller, Institute for Clinical Pharmacology, Public Health Research Association Saxony, Medical Faculty, Technical University, Fiedlerstrasse 27, D-01307 Dresden, Germany, E-mail: Edgar.Mueller{at}mailbox.tu-dresden.de. Kim Andriano, Novartis Pharmaceuticals Corporation, One Health Plaza, East Hanover, NJ 07936-1080. Philip Hunt, Novartis Horsham Research Centre, Wimblehurst Road, Horsham, RH12 5AB, UK. Patricia Ibarra de Palacios, Novartis Pharma AG, Clinical Research, Lichtstrasse 35, CH-4056 Basel, Switzerland.
Reprint requests: Edgar A. Mueller, MD, Institute for Clinical Pharmacology, Public Health Research Association Saxony, Medical Faculty, Technical University, Fiedlerstrasse 27, D-01307 Dresden, Germany. E-mail: Edgar.Mueller{at}mailbox.tu-dresden.de.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
