Study area
This trial was conducted at Yombo primary health facility, Bagamoyo district, Tanzania, between July and November, 2014. The health facility is located southwest, about 20 km, from Bagamoyo town. It serves approximately 7000 people and has capability to carry out routine malaria microscopy and rapid diagnostic test.
Malaria transmission is high and occurs throughout the year with peaks related to the long rain season from May to July and short rain season from November to December. In the study area,
P. falciparum is the major malaria species and
Anopheles gambiae sensu stricto the main vector [
7‐
9]. AL has been used as the first-line treatment for uncomplicated malaria since 2006. Sulfadoxine-pyrimethamine is used for intermittent preventive treatment in pregnant women. Long-lasting, insecticide-treated mosquito nets is the major vector control method [
10]. G6PD deficiency prevalence in the study area has previously been estimated to 13.6 % in hemizygous males and 4.5 % in homozygous females [
11].
Study design
This was a randomized, single-blinded, clinical trial comparing treatment outcome and safety of AL plus a single low-dose PQ (AL + PQ) versus standard AL regimen. Safety outcomes have been presented in a separate publication [
11]. Patients with uncomplicated microscopically confirmed
P. falciparum mono-infection were enrolled, randomly assigned to either AL + PQ or AL treatment, admitted during the first 3 days of treatment and thereafter followed up until day 28 after treatment initiation.
Treatment outcome was based on polymerase chain reaction (PCR)-adjusted parasitological cure. Therapeutic failures were classified as; early treatment failure (ETF), late clinical failure (LCF), or late parasitological failure (LPF) [
12].
Study population
Patients presenting at the study site with suspected acute uncomplicated malaria were screened for eligibility. Inclusion criteria were age ≥1 year, weight ≥10 kg, body temperature ≥37.5 °C or history of fever in the last 24 h, microscopy confirmed P.
falciparum mono-infection, any parasitaemia level, ability to swallow oral medication, ability and willingness to abide by the study protocol and the stipulated follow-up visits, and a written informed consent (in case of children a proxy consent from a parent/guardian). Exclusion criteria were evidence of severe malaria or danger signs, known allergy to trial medicines, reported anti-malarial intake ≤2 weeks, haemoglobin (Hb) <8 g/dL, blood transfusion within last 90 days, febrile condition other than malaria, known underlying chronic or severe disease (including severe malnutrition), pregnancy and breastfeeding.
Randomization and blinding
Treatment allocation was done using sex-stratified, block randomization with four blocks, two per treatment arm using RESEARCH RANDOMIZER (version 4) (computer software) (Wesleyan University, Connecticut, USA) [
13].
Opaque envelopes containing the pre-determined treatment codes were kept serially in a male and female strata. The envelopes were opened by a study nurse just before the first treatment dose was administered. Patients were blinded to the assigned treatment.
Treatment
A standard, weight-based, three-day course of AL (Coartem
®, Novartis) was administered to all patients according to Tanzania national treatment guideline for uncomplicated
P. falciparum malaria [
14]. AL dispersible tablets suspended in water were administered to children who were not able to swallow tablets. A single 0.25 mg/kg PQ dose (Primaquine phosphate, Sanofi) was administered together with AL first dose to patients assigned AL + PQ treatment. The accuracy of PQ dose among patients weighing less than 60 kg body weight was ensured by administering the drug in aqueous solution, whereas for adults weighing 60 kg and above the drug was administered as tablet. To make aqueous solution, a 15 mg PQ tablet was suspended in 15 mL of water, and the dose was measured using a sterile syringe based on body weight. To achieve the single blinding, the PQ dose was prepared in the absence of patients, and a glucose-based syrup was added to mask the PQ taste, whereas for patients allocated AL alone, the same glucose-based syrup was administered with AL first dose. Adult patients allocated to the AL + PQ arm, during consenting, never knew in which treatment arm they were allocated, and during drug administration, PQ tablet was mixed with AL tablets, however, in case patients asked about the additional tablet, they were told that the tablet was given to prevent treatment side effects. In order to optimize AL absorption and minimize PQ gastro-intestinal side effects, biscuits were administered prior to all drug doses [
15]. A study nurse administered/supervised intake of all drug doses. Patients were monitored for 30 min after each drug dose. Treatment was re-administered in case of vomiting within this period.
Procedures
Clinical assessment was performed at 0, 8, 24, 36, 48, 60, and 72 h and on days 7, 10, 14, 21, 28 or on any day of recurrent illness. The assessment included history of clinical symptoms, possible adverse events, concomitant drug consumption and clinical examination including measurement of axillary temperature. Fever was defined as body temperature ≥37.5 °C. A case record form was used to record all clinical and laboratory data. Laboratory assessment involved collection of finger-prick blood samples for Hb concentration, thick smears for microscopy-determined asexual and sexual parasitaemia, thin smears for species determination (at enrolment) and filter-paper blood samples for parasite detection and genotyping by PCR. Hb assessment was done on days 0, 1, 2, 3, 7, 10, 14, 21, 28 or on any day of recurrent illness, whereas, filter paper and blood slide samples were collected at 0, 8, 24, 36, 48, 60, and 72 h and on days 7, 10, 14, 21, 28 or on any day of recurrent illness. Hb concentration was measured using a portable spectrophotometer, HemoCue Hb 201+ (HemoCue AB, Ängelholm Sweden), with a precision of ±0.3 g/dL. A control cuvette at 16.0 ± 0.3 g/dL was used for daily calibration according to manufacturer’s instruction.
A 10 % Giemsa solution was used to stain thick and thin blood smears. Thin smears were prepared once, i.e., at enrolment, whereas thick smears were prepared at all sampling time points. Asexual parasites were counted against 200 white blood cells (WBC). The obtained number was multiplied by 40, assuming 8000 leukocytes per μL of blood, to gain an approximate parasite count. In case no parasites were observed after examining 100 fields, the blood slide was considered negative. Each slide was read by two independent microscopists. In case they disagreed on presence of parasitaemia or if density differed by more than 25 %, a third independent reading was performed. In case of positive versus negative results, a third independent reading was used to confirm the reading of the first two readers. The filter paper (3MM Whatman) blood samples were labelled, air-dried at room temperature for 3–4 h and then packed in individual plastic bags and stored. After study completion, all filter papers were transported to Karolinska Institutet, Sweden, for molecular analysis.
Molecular analysis
A 10 % Chelex-100
® method was used to extract genomic DNA from dried blood spots [
16]. Paired blood samples (pre-treatment and day of recurrent parasitaemia) from patients classified as LCF or LPF were genotyped to differentiate recrudescence from re-infection by stepwise genotyping of
P. falciparum block 3 of merozoite surface protein (msp) 2, block 2 of msp 1 and region II (RII) of glutamate-rich protein (glurp) [
17]. The respective initial amplifications were followed by individual nested PCR reactions using family specific primers for msp1 (K1, MAD20 and RO33) and msp2 (FC27 and IC) and semi-nested for RII of glurp [
17]. The amplicons were loaded on a GelRed™ (Biotium, Hayward, CA, USA) stained agarose gel, separated by electrophoresis and then visualized under ultraviolet transillumination (Gel Doc™, Bio-Rad, Hercules, CA, USA), and sized by Image Lab™ software (Bio-Rad, Hercules, CA, USA). Alleles in each family were considered the same if fragments size were within 20 base pair interval. Subjects with recurrent parasitaemia by microscopy, which could not be confirmed by PCR were considered to have uncertain-PCR-adjusted outcome.
Recrudescence was defined as presence of at least one matching allelic band, and re-infection was defined as absence of any matching allelic band at baseline and on the day of parasite recurrence [
17]. In addition
, P. falciparum multi-drug resistant gene 1 (
Pfmdr1) asparagine (N)-86-tyrosine (Y) and chloroquine resistance transporter gene (
Pfcrt) lysine (K)-76-threonine (T) were genotyped using nested PCR followed by restriction fragment length polymorphism using
ApoI restriction enzyme as previously described [
18].
Study end-points
The primary outcome was the proportion of patients with PCR-adjusted adequate clinical and parasitological response (ACPR) by day 28. Secondary outcomes included: fever and asexual parasite clearance, proportion of patients with PCR-determined parasitaemia on day 3, PCR- determined re-infection rate, and proportion of patients with Pfmdr1 N86Y and Pfcrt K76T on days 0, 3 and day of recurrent infection.
Statistical analysis
Sample size calculation was based on equivalence, defined as margin of 10 % PCR-adjusted ACPR, between the treatments. Allowing for 10 % attrition, 80 % power at 0.05 significance, a sample size of 110 per treatment arm was required. Data were double-entered in an electronic database and analysed using SPSS software version 16 (SPSS Inc, Chicago, USA) as per protocol. Independent sample t test was used to compare means. Non-parametric data were compared using Chi square, Fisher’s or McNemar tests as appropriate. Cure rate end points were analysed by survival analysis, and the survival curves of the two treatment arms were plotted and compared using log-rank test. Data were censored at the time of withdrawal for patients lost to follow up, withdrew consent and PCR determined re-infection or uncertain PCR outcome. A p ≤ 0.05 was considered significant.