In vivo efficacy of artemether-lumefantrine and artesunate-amodiaquine for the treatment of uncomplicated falciparum malaria in children: a multisite, open-label, two-cohort, clinical trial in Mozambique

The study was conducted using the standard WHO in vivo efficacy protocol[14] in five hospitals or health centres across Mozambique, namely: 1) Hospital Rural de Montepuez, in Cabo Delgado province (northern region); 2) Centro de Saúde de Dondo, in Sofala province (central region); 3) Hospital provincial de Tete, in Tete province (central region); 4) Hospital rural de Chokwe, in Gaza province (southern region); and 5) Hospital distrital de Manhiça, in Maputo province (southern region). Malaria transmission in Mozambique is perennial, with a peak transmission period normally coinciding with the rainy season, from November to April. Study cohort 1 testing AL began in June 2011 and involved the five sites, while study cohort 2 assessing the combination ASAQ started in August 2012, once the follow-up for the first cohort had been concluded, and was only conducted in three of the five sites (Montepuez, Dondo and Chokwe).

The study population comprised children aged six to 59 months with microscopically confirmed, acute uncomplicated malaria. Other inclusion criteria included body weight ≥5 kg, the presence of fever (≥37.5°C axillary) or a history of fever in the preceding 24 hours, P. falciparum malaria mono-infection with an asexual blood density ≥2,000/μL and <200,000/μL, and the absence of severe signs of complicated malaria as defined by WHO[15]. Key exclusion criteria included mixed malarial infections, haemoglobin <5 g/dL, severe malnutrition, intake of anti-malarials within the preceding seven days, ongoing prophylaxis in HIV-positive patients with cotrimoxazole or the intake of any other drug with anti-malarial activity, and any serious underlying disease. Patients satisfying the inclusion criteria were enrolled if the parent/guardian signed a detailed written informed consent.

Eligible patients were consecutively assigned to the cohort and treated with AL (cohort 1) or ASAQ (cohort 2). AL (Coartem™, Novartis, each tablet contains 20 mg artemether and 120 mg lumefantrine) was administered twice daily for three days (six doses in total) with dosage determined according to body weight: one tablet for children 5 to <15 kg, two tablets for children 15 to <25 kg, and three tablets children 25 to <35 kg. ASAQ (Winthrop™, Sanofi Aventis, each tablet contains 25 mg artesunate and 67.5 mg amodiaquine) was administered once daily according to body weight: one tablet tablet for children <9 kg, two tablets for children 9–17.9 kg; and four tablets for children >18-35 kg. All treatments were directly observed for a minimum of 30 min. Vomiting occurring within the first 30 min implied the repetition of the full dose of treatment. For those patients living far away from the health facilities, and for which direct observation of the evening doses of AL was challenging, admission was offered for the first three days of the study.

Antipyretics, such as paracetamol, were used to control fever > = 38°C. In the event of severe malaria or danger signs, the patient was hospitalized and received intravenous quinine, according to the national malaria treatment policy. Rescue therapy according to national malaria treatment guidelines was also administered in cases of early or late treatment failure with parenteral quinine[16].

Follow-up visits took place on days 1, 2, 3, 7, 14, 21 and 28 after enrolment or at any time point whenever the child was sick. Patients who prematurely discontinued either study drug or the study were excluded from the study. Vital signs and body temperature were assessed during each follow-up visit. Adverse events were recorded and assessed for severity and association with study medication.

Thick and thin Giemsa-stained blood slides were prepared before each dose was administered and at every follow-up visit of days 2, 3, 7, 14, 21, and 28. Slides were examined by two independent microscopists and considered negative if no parasites were seen after examination of 200 oil-immersion fields in a thick blood film. Parasite density was estimated by counting the number of asexual parasites in 200 white blood cells (WBC), assuming a standard WBC count of 8,000 /μl. Species determination (and thus conformation of mono-infection) was made based on assessment of thin films. Blood spots for PCR analysis were collected from every patient using 3 M Whatman™ filter papers at baseline and at days 7, 14, 21 and 28, day of treatment failure or at any other unscheduled visit, and subsequently stored in plastic zip bags containing silica gel dessicant. PCR was performed centrally for all cases of recurrent parasitaemia from day 7 onwards, including DNA extraction using a QIAamp DNA Mini Kit (Qiagen), and investigation of the three polymorphic genetic markers MSP1, MSP2, and GluRP, which were used to distinguish recrudescence from new infections, according to WHO recommended procedures[17]. Recrudescence was defined as at least one identical allele for each of the three markers in the pre-treatment and post-treatment samples. New infections were diagnosed when all alleles for at least one of the markers differed between the two samples. Cases with new infection were excluded from the analysis.

The primary efficacy outcomes were the PCR-corrected early treatment failure (ETF), late clinical failure (LCF), late parasitological failure (LPF) and adequate clinical and parasitological response (ACPR) at day 28. Secondary outcomes included 28 day-uncorrected ACPR (crude efficacy), safety and tolerability profiles, time to parasite, fever and gametocyte clearance, and haemoglobin changes from baseline to day 28.

Data were recorded using specifically designed, standardized case report forms based on those proposed by WHO[14]. All study questionnaires were doubled-entered into a study-specific database created using open clinica software (OpenClinica Enterprise - Electronic Data Capture Software for Clinical Trials version 3.1.2, OpenClinica LLC, Waltham, MA, USA). Two populations were defined for the analysis: the intent-to-treat (ITT) population (safety population) comprised all patients who received ≥ one dose of study medication and underwent at least one post-baseline safety assessment. Efficacy was calculated in the according-to-protocol population (ATP), which included all patients fulfilling the protocol eligibility criteria, having completed the three-day course of study medication, accomplishing the day-28 assessment and having an evaluable PCR in case of recurrent parasitaemia. Cure rates were calculated as the number of patients with clinical and parasitological cure by day 28 divided by the total number of patients who could be evaluated. Additionally, Kaplan-Meier estimates of the cumulative risk of failure were computed for day 28. For such an analysis, losses to follow-up and study withdrawals were censored on the last day of follow-up. Cases with re-infections were also censored from the analysis. Statistical analyses were done with Stata 13 (Stata Corp, College Station, TX, USA), and the statistical significance level was set at 5%. No formal comparisons were made between the two study cohorts.

The only available published data regarding the efficacy of the two studied ACT in Mozambique come from previous studies conducted in Manhiça, where AL was shown to have an efficacy ranging from 96.9%[13] to over 98%[12] and ASAQ of 97.2%[10]. Sample size calculations were based on WHO-proposed methodologies[14] using a slightly more conservative expected efficacy estimate (95%) and a level of precision around the estimate of 5%. To achieve this, and with an expected loss to follow-up rate of 20% on day 28, a minimum 87 patients would need to be recruited at each of the sites and for each of the two study arms.

The protocol was approved by the National Mozambican Ethics Review Committee (Ref 134/CNBS/11) and the Hospital Clínic of Barcelona Ethics Review Committee. The trial was conducted according to Good Clinical Practice guidelines. Written consent was obtained from the parents/guardians of the study children. The clinical-trial identifier is NCT02168569 (http://​www.​clinicaltrials.​gov).

Between June 2011 and January 2012, both cohort 1 (AL) and cohort 2 (ASAQ) were recruited and followed up to 28 days. Some 2,587 febrile children were screened for cohort 1, of which 439 (16.9%) ended up being recruited and 335 (76.3%) completed the study, with or without a recurring parasitaemia. For cohort 2, 2,154 febrile children were screened, 261 (12.1%) recruited and 232 (89.0%) successfully completed the study, with or without a recurring parasitaemia (Figure 1). Main reasons for exclusion in both cohorts was the absence of malaria parasites at screening and/or the presence of concomitant illnesses. The ITT and ATP populations for cohort 1 included 439 and 335 individuals, respectively, whereas for cohort 2 these numbers were 261 and 232, respectively. Table 1 summarizes the baseline characteristics for both cohorts, which were comparable in terms of gender, mean age, weight, and malarial infection.

During the 28-day follow-up period, 66 patients (cohort 1; 15%) and 28 patients (cohort 2; 10.7%) did not successfully complete the study on account of loss to follow-up, withdrawal of consent or other protocol violations (Figure 1). Table 2 and Figures 2 and3 summarize both cohort 1 and cohort 2 treatment outcomes. The day-28 PCR-uncorrected cure rate (i.e., the proportion of patients with ACPR, ATP population) was 89.6 (335/374; 95% CI 86.0-92.5) for AL, and 99.6 (232/233; 95% CI 97.6-99.9) for ASAQ. Of the 38 cases of recurring parasitaemia in cohort 1, 25 (65.8%) proved to be new infections according to PCR, yielding a day-28 PCR-corrected cure rate of 96.0 (335/349; 95% CI 93.4-97.8). In the ASAQ, only one recurring parasitaemia was detected, which proved to be a recrudescence of the original infection according to PCR. Thus, day-28 PCR-corrected cure rate was identical to the uncorrected one (99.6 [232/233; 95% CI 97.6-99.9]). All patients in both cohorts cleared their parasitaemia by day 3 following treatment, 100% patients treated with ASAQ by the end of the first 24 hours of follow-up, while 5.0% (22/439) of the patients treated with AL took longer than 24 hours to clear parasitaemia (of these, 77.2% [17/22] cleared parasitaemia on day 2, while 22.7% [5/22] on day 3).

Fever, present at recruitment in about three-quarters of all study patients, receded rapidly during the first 72 hours of follow-up, with all patients being afebrile by day 7 of follow-up (Figures 4 and5). Tolerability of the two drugs, as judged by solicited adverse events during drug intake, was good (Table 3). Some patients reported vomiting (20/439, 4.5% in the AL cohort; 25/261, 9.6% in the ASAQ group) during the three days of treatment, and the occurrence of other solicited adverse events was rare. All the vomiting episodes (with the exception of two episodes in different children, both occurring in the ASAQ group during the first day of treatment and after >1 hour post treatment, deemed unrelated to the study drug) occurred within the first 30 minutes of treatment, and thus required re-dosing. No patient had repeated vomiting episodes, and no rescue treatment was required on account of vomiting. Six urticarial-like episodes (2/6 [33.3%] in the AL group, 4/6 [66.6%] in the ASAQ group) occurred in both groups, being all of them self-limited, transient and of mild nature. In the majority of these cases (5/6; 83.3%) these events were judged unrelated to the study drug, as alternative explanations (“atopy”; “eczema”; “scabies-related rash” (twice); “viral rash”) were provided, although in one case (16.7%; post AQAS), investigators deemed it “possibly” drug-related, and the skin manifestations disappeared rapidly without requiring any treatment.

Haemoglobin recovery from day 0 to day 28 occurred slowly in both groups (mean increase in the AL cohort being 1.0 (SD 4.2) and in the ASAQ cohort being 1.6 (SD 1.7) (Table 4). Seven serious adverse events (SAE) were documented in the study: four in the AL group (three severe malaria cases and a Kwashiorkor with pellagra diagnosis) and three in the ASAQ group (two severe malaria cases one of which with concomitant severe pneumonia), none deemed by the investigators related to the study drugs. Six of the seven SAEs recovered completely, but an 11-month-old female child in the Montepuez site died after a severe malaria episode, which resolved adequately at hospital. The death occurred at home, on the third day of follow-up, and according to the verbal autopsy subsequently performed, appeared related to the uncontrolled administration of traditional medicine.