Safety of single low-dose primaquine in glucose-6-phosphate dehydrogenase deficient falciparum-infected African males: Two open-label, randomized, safety trials

Background Primaquine (PQ) actively clears mature Plasmodium falciparum gametocytes but in glucose-6-phosphate dehydrogenase deficient (G6PDd) individuals can cause hemolysis. We assessed the safety of low-dose PQ in combination with artemether-lumefantrine (AL) or dihydroartemisinin-piperaquine (DP) in G6PDd African males with asymptomatic P. falciparum malaria. Methods and findings In Burkina Faso, G6PDd adult males were randomized to treatment with AL alone (n = 10) or with PQ at 0.25 (n = 20) or 0.40 mg/kg (n = 20) dosage; G6PD-normal males received AL plus 0.25 (n = 10) or 0.40 mg/kg (n = 10) PQ. In The Gambia, G6PDd adult males and boys received DP alone (n = 10) or with 0.25 mg/kg PQ (n = 20); G6PD-normal males received DP plus 0.25 (n = 10) or 0.40 mg/kg (n = 10) PQ. The primary study endpoint was change in hemoglobin concentration during the 28-day follow-up. Cytochrome P-450 isoenzyme 2D6 (CYP2D6) metabolizer status, gametocyte carriage, haptoglobin, lactate dehydrogenase levels and reticulocyte counts were also determined. In Burkina Faso, the mean maximum absolute change in hemoglobin was -2.13 g/dL (95% confidence interval [CI], -2.78, -1.49) in G6PDd individuals randomized to 0.25 PQ mg/kg and -2.29 g/dL (95% CI, -2.79, -1.79) in those receiving 0.40 PQ mg/kg. In The Gambia, the mean maximum absolute change in hemoglobin concentration was -1.83 g/dL (95% CI, -2.19, -1.47) in G6PDd individuals receiving 0.25 PQ mg/kg. After adjustment for baseline concentrations, hemoglobin reductions in G6PDd individuals in Burkina Faso were more pronounced compared to those in G6PD-normal individuals receiving the same PQ doses (P = 0.062 and P = 0.022, respectively). Hemoglobin levels normalized during follow-up. Abnormal haptoglobin and lactate dehydrogenase levels provided additional evidence of mild transient hemolysis post-PQ. Conclusions Single low-dose PQ in combination with AL and DP was associated with mild and transient reductions in hemoglobin. None of the study participants developed moderate or severe anemia; there were no severe adverse events. This indicates that single low-dose PQ is safe in G6PDd African males when used with artemisinin-based combination therapy. Trial registration Clinicaltrials.gov NCT02174900 Clinicaltrials.gov NCT02654730

In the current study we aim to assess safety of PQ, in particular the risk of acute haemolysis, when given together with artemether-lumefantrine (AL) in G6PD deficient individuals. Forty male participants with asymptomatic malaria, normal haemoglobin levels (≥11g/dL), reduced G6PD enzyme function, and aged between 18 and 45 years will be sequentially enrolled into two dosing cohorts consisting of 20 individuals and doses of 0.25 and 0.4 mg/kg PQ in combination with a full three-day course of AL. Participants will first be assigned to the lowest open cohort; enrolment in the next cohort is initiated after tolerability and short-term safety is demonstrated at the preceding lower dose. Furthermore, we will include 3 control groups into the first cohort: i) 10 male participants aged 18-45 years with asymptomatic malaria, normal haemoglobin levels (≥11g/dL) and a reduced G6PD enzyme function will receive AL only, ii) 10 male participants with asymptomatic malaria, normal haemoglobin levels (≥11g/dL) and normal G6PD enzyme function will receive 0.25 mg/kg PQ in combination with a full three-day course of AL, and iii) 10 male participants with asymptomatic malaria, normal haemoglobin levels (≥11g/dL) and normal G6PD enzyme function will receive 0.4 mg/kg PQ with AL. Clinical follow-up of participants and sampling will be done twice daily for the first 4 days (days 0, 1, 2 and 3) and once daily on days 4, 5, 7, 10, 14 and 28. At each timepoint laboratory safety parameters are measured, including haematology, biochemistry, urine dipstick for haemoglobinuria/urobilinogen and parasite density. Five individuals from each intervention group (total of 10) will also be asked to provide seven venous blood samples on days 0, 1 and 2 to study pharmokinetics of PQ in G6PD deficient individuals.

INTRODUCTION/BACKGROUND
A new global effort is underway to step up malaria control and push towards the elimination of malaria as a public health problem [1]. Some substantial successes have been achieved in shrinking the global distribution of malaria. In Africa, effective elimination programmes have been initiated in Zanzibar and South Africa. There are now several African countries with a commitment to malaria elimination. This call for elimination has created a drive for the deployment of tools to reduce malaria transmission. One such tool is primaquine (PQ). It is a drug which can efficiently block the transmission of Plasmodium falciparum malaria from humans to mosquitoes by clearing the parasite's transmission stages, gametocytes that persist after artemisinin-combination therapy (ACT) [2,3]. A study conducted in Tanzania in symptomatic parasitized children demonstrated a dramatic reduction of gametocyte circulation time from 28.6 days with ACT alone to 6.3 days with ACT-PQ [4].
A drawback of using PQ is, however, that is has been shown to cause haemolysis in glucose-6phosphate dehydrogenase (G6PD) deficient individuals [5][6][7]. G6PD is an essential enzyme linked to the pentose phosphate pathway; in erythrocytes, defence against oxidative stress is dependent on G6PD since they do not contain mitochondria. Its X-linked deficiency is one of world's most common genetic polymorphisms affecting 400 million people worldwide, mostly in P. falciparum malaria endemic areas [8].
A recent study in G6PD-sufficient individuals in Uganda found that lower doses of PQ than the then WHO standard dose of 0.75 mg/kg, retain efficacy for gametocyte clearance [Eziefula et al, under review]. In 2012, the WHO recommended that "A single 0.25 mg base/kg primaquine dose should be given to all patients with parasitologically-confirmed P. falciparum malaria on the first day of treatment in addition to an ACT, except for pregnant women and infants <1 year of age" (from Malaria Policy Advisory Committee Meeting 2012) [9]. However, detailed data about the lowest effective PQ dose and robust data regarding the safety of various PQ dosages in glucose-6phosphate dehydrogenase (G6PD) deficient individuals are lacking [10]. This is remarkable, since all parasitologically-confirmed P. falciparum malaria patients in malaria elimination areas will now receive a single dose of PQ on the first day of treatment with ACT. This includes G6PD deficient individuals, who can account up to 30% of the population in some malaria-endemic areas in Africa with a mean prevalence of approximately 10% [8,11]. Determining the safety profile of PQ in G6PD deficient individuals has become crucial and high quality clinical studies in G6PD deficient individuals are urgently needed.

History of primaquine use and safety in G6PD deficiency
Over the last fifty years many people have received PQ. Particularly in the 1960s when malaria eradication seemed feasible, entire regions and millions of people were exposed to repeated mass drug administration (MDA) with regimens containing PQ without individual G6PD deficiency screening. In 1973, the WHO recommended a single dose of PQ (0.75 mg/kg) for malaria transmission-blocking and considered prior screening for G6PD deficiency unnecessary [12].
Although little detailed information is available on the safety of these practices [13,14], there are no reports of major toxic effects when given weekly or in single doses (usually 0.75 mg base/kg PQ) [14][15][16][17][18].
Safety data for PQ use in Africa or African Americans are limited. Burgess and Bray comment that PQ was "well-tolerated" [19]. In a large scale MDA conducted by Clyde in 1962, PQ was paired with amodiaquine (AQ) in a hyperendemic area of Tanzania. More than 15,000 subjects were studied in three clusters: weekly administration, fortnightly administration, and monthly administration. Clyde reported no safety data and it is unclear who was excluded from treatment [16]. In a series of studies in G6PD deficient African-American volunteers, Alving et al. showed that, in three individuals, haemolysis occurred with daily administration of 30mg (approximately 0.5 mg/kg) of PQ.
But, after three weeks, the haematocrit levels recovered and lower doses of PQ resulted in less haemolysis. Eight weekly doses of 60 mg and 45 mg were not associated with haemolysis [20,21].
Daily administration of 30mg of PQ to African Americans resulted in significant haemolysis in 1%, compared to no severe haemolysis when 15 mg was administered [22]. Tolerance in a pregnant woman (28 weeks gestation) has only been reported by Burgess and Bray, but there was no documentation of birth outcomes [19]. In a more recent study, Kenyan school children were randomized to receive 15mg PQ daily or three times a week as a malaria prophylactic. It is not clear whether G6PD deficient individuals were included and haemoglobin (Hb) levels are not reported but again the authors note simply that "PQ was remarkably well tolerated in our studies" [23]. In nearly 1500 patients studied prospectively in trials of single dose PQ given as a gametocytocide in Africa, Asia and South-America, no severe haemolysis requiring blood transfusion was reported [3,9,[24][25][26][27][28][29].
Based on these previously mentioned observations, the risk of haemolysis with single-dose PQ was regarded as acceptable except in areas where severe G6PD deficiency is prevalent. Shekalaghe

G6PD deficiency worldwide and in Burkina Faso
Haemolysis after PQ is strongly associated with G6PD deficiency, which is expressed in males carrying a variant gene that may result in sufficient enzyme deficiency to lead to symptoms. The mean red blood cell enzyme activity in heterozygous females may be normal, moderately reduced, or grossly deficient depending upon the degree of lyonization and the degree to which the abnormal G6PD variant is expressed [32]. A heterozygous female with 50 percent normal G6PD activity, due to The WHO has classified different G6PD variants according to the magnitude of the enzyme deficiency and the severity of haemolysis [11]. Classes IV and V are of no clinical significance.
 Class I variants have severe enzyme deficiency (less than 10 percent of normal enzyme activity) and have chronic (nonspherocytic) hemolytic anemia.
are usually only intermittent episodes of acute haemolysis associated with infection, drugs, or chemicals.
 Class III variants, such as G6PD A-, have mild to moderate enzyme deficiency (10 to 60 percent of normal enzyme activity) with intermittent episodes of acute haemolysis usually associated with infection, drugs, or chemicals.
 Class IV variants have no enzyme deficiency or haemolysis.
 Class V variants have increased enzyme activity.
The most common G6PD variant in Burkina Faso, and Africa as a whole, is the Class III A-variant [33,34]. In 352 samples from clinical malaria patients from around Ouagadougou, all G6PD A-mutations involved the A376G/G202A mutation [34]. The G6PD A-variant encoded by these mutations has up to 80% enzyme function compared to wild type. This variant is associated with mild to moderate haemolysis in the presence of stimuli such as PQ. By contrast, some Southeast Asian and Mediterranean variants are Class I and II variants, being associated with severe deficiency, and PQ may provoke a severe haemolysis in these G6PD deficient individuals.

Side effects of primaquine in G6PD deficient individuals
Given its widespread use over the last fifty years, side effects of PQ are well known. The main side effects are as follows: i) Gastro-intestinal symptoms if not given with food ii) Methaemoglobinaemia (although there are no reports in the literature of individuals who required treatment for methaemoglobinemia due to PQ [35]) iii) Transient haemolysis in individuals with a predisposition, such as G6PD deficiency Haemolysis occurs mostly in senescent erythrocytes as these cells become increasingly G6PD deficient, and therefore, more depleted in reduced glutathione, which is their principal antioxidant defence [8]. Therefore, reticulocytosis (proliferation of young red blood cells) in acute symptomatic malaria affords some protection, as the population of red cells is relatively younger.
The haemolytic side effect of PQ is dose-related and in proportion to the degree of G6PD deficiency (WHO Classification, paragraph 6.2) [8,11,36]. In individuals with mild or moderate deficiency, such as the A-G6PD variant (Class III), maximum rates of red-cell destruction are usually reached within 1-2 days and clinically detectable haemolysis and jaundice can typically arise within 24-72 hours of drug dosing [8,9]. In these less-severe cases, haemolysis stops in about 7 days, because haematocrit values start to rise due to replacement of old G6PD-deficient erythrocytes with younger, more PQresistant red cells [11]. Furthermore, PQ is eliminated rapidly from the body (PQ's plasma half life is between 4 and 9 hours [35]) and so haemolysis is a self-limiting process provided no further drug is taken. In severe cases, such as those associated with G6PD Mediterranean variant (Class I and II), individuals may experience nausea, vomiting, abdominal or loin pain, and passage of dark urine due to haemoglobinuria [8,9]. Anaemia develops rapidly and patients become jaundiced. This is associated with an abrupt fall in haemoglobin concentration of 3-4 g/dL, during which time the peripheral blood smear reveals red cell fragments, microspherocytes, and eccentrocytes or "bite" cells. Special stains document the production of Heinz bodies, which are collections of denatured globin chains often attached to the red cell membrane. Haemolysis is both extravascular and intravascular [8]. Rarely, haemoglobinuric renal failure occurs. In patients with this severe G6PD deficiency, continuation of PQ results in further haemolysis and life-threatening anaemia.
Haemolysis is more commonly observed after prolonged PQ treatment and 13 deaths from unsupervised prolonged PQ use have been reported: four in Sri Lanka, five in Turkey, two in Brazil, and one each in the UK and the USA [5]. No deaths are known to have followed single-dose administration [9], but haemolysis has also been observed in African populations following a single dose of PQ [2,26]. This haemolysis was self-limiting, largely restricted to G6PD deficient individuals and did not lead to clinical symptoms.

Assessing a safe and effective dose of PQ -data from Uganda
Since haemolysis is dose-dependent, side effects are expected to be less or absent when PQ is administered as a single low dose. If a lower dose of PQ would be efficacious in clearing gametocytes, this would make it a safe and powerful tool for community-wide malaria elimination strategies. Two studies in Thailand have suggested that lower doses of PQ than previously recommended by WHO (0.75 mg/kg) can be equally efficacious in clearing gametocytes. Bunnag compared the effect of 15 mg daily for 5 days, 30 mg single dose and 45 mg single dose in Thai adults and found no significant difference in gametocyte clearance between doses [37].
Pukrittayakamee compared 0.25 mg/kg and 0.5 mg/kg PQ in adults and found both to have shorter gametocyte clearance times (GCT) than non-PQ-containing regimens, with no significant difference in outcomes between the two doses of PQ [38]. A mass drug administration programme in Cambodia used a 9 mg stat dose of PQ (approximately 0.15 mg/kg) every ten days, with a significant reduction in microscopic gametocyte carriage from 13.1% to 0.8% after 3 years [39]. These studies have shortcomings in their safety and efficacy assessments. None of the studies included complete safety assessments and microscopy was used as gametocyte detection tool, which is notoriously insensitive for detecting low density gametocytes [40]. Furthermore, none of these studies determined the transmissibility of gametocytes persisting after PQ. These studies support the hypothesis that a lower, safer but efficacious dose of PQ may be found.
A more detailed study to provide evidence on efficacy of low dose PQ was recently completed in In the current study, we propose to assess the risk of haemolysis associated with low doses of PQ (0.25 mg/kg and 0.40 mg/kg) in individuals with G6PD deficiency. given to parasitaemic individuals with normal G6PD activity as well as G6PD deficient individuals, who have an increased risk of PQ-induced haemolysis. Therefore, using a dose-escalation design, the current study will assess safety of PQ when given together with an ACT (in this case AL) in asymptomatic parasitaemic G6PD deficient individuals. The current WHO-recommended dose (0.25 mg/kg) and the lowest efficacious dose (0.4 mg/kg), as defined by the recent trial, will be tested.

MAIN OBJECTIVE
To evaluate the tolerability and safety of increasing doses of PQ in combination with AL in G6PD deficient males with an asymptomatic P. falciparum malaria infection.

General study design
The study is an open-label, dose-escalation Phase I trial. A minimum of 70 males between 18 and 45 years will be sequentially enrolled into five treatment groups. Two out of five treatment groups are intervention groups and 3 are control groups. The two intervention groups consist of 20 individuals each; doses of 0.25 and 0.4 mg/kg PQ + AL will be given, respectively. The 3 control groups consist of 30 individuals: 10 male participants with G6PD deficiency will receive AL only, 10 males with normal G6PD enzyme function will receive 0.25 mg/kg PQ + AL, and 10 males with normal G6PD enzyme function will receive 0.4 mg/kg PQ + AL.
There are two dosing cohorts. The first cohort will consist of: i) the intervention group with G6PD deficiency receiving 0.25 mg/kg PQ + AL (n=20) ii) the control group with G6PD deficiency receiving AL only (n=10) iii) the control group with normal G6PD enzyme function receiving 0.25 mg/kg PQ + AL (n=10) iv) the control group with normal G6PD enzyme function receiving 0.4 mg/kg PQ + AL (n=10) The second cohort will consist of: i) the intervention group with G6PD deficiency receiving 0.4 mg/kg PQ + AL (n=20) The trial will be stopped according to criteria stated in section 10.5. Male participants will be recruited from Banfora health district, in the southwest of Burkina Faso, using village census lists and house-to-house visits of study teams. Screening lists are prepared based on census lists. Individuals aged 18-45 years who indicate a willingness to participate in the trial and sign an informed consent form for the screening procedure, will be tested for the presence of malaria parasites by microscopy and for haemoglobin levels by HemoCue; if parasites are present and Hb >11g/dL, G6PD deficiency will be tested by Beutler Fluorescent Spot test [41] (paragraph 10.11). Clinical screening will take place if malaria parasites are detected at any density in the absence of a history of fever and haemoglobin > 11g/dL. Study clinicians will assess eligibility for each individual to enter the study by i) taking medical history (including inclusion and exclusion criteria) and laboratory testing and ii) physical examination.
If individuals are eligible for enrolment, they will be enrolled into the lowest open cohort of the trial.
Treatment allocation in cohort 1 will be randomised, using sealed envelopes. If G6PD deficiency test indicates reduced enzyme function (<30% of normal function [41]), individuals will be enrolled in the intervention group (receiving 0.25 mg/kg PQ + AL) or control group (receiving AL only), or -if the first cohort has been completed -they will be enrolled without randomization in the intervention group receiving 0.4 mg/kg PQ + AL (details in table 2). If G6PD activity is normal, individuals will be enrolled in one of the other two control groups in cohort 1 (0.25 mg/kg PQ or 0.40 mg/kg PQ; details in table   2).
All enrolled individuals will receive a full three-day course of AL and -depending on treatment arm -PQ with the first dose of AL (day 0). All doses of AL and PQ will be directly observed. Participants will be reimbursed for travel to and from the clinic for all scheduled and non-scheduled visits during the time they are enrolled in the study. During follow up and beyond the last scheduled visit at day 28, all participants will be encouraged to attend the clinic for any medical concerns and the cost of travel to the clinic will be reimbursed.
Sampling will be as follows: all individuals will be sampled on T0, T12 (D0); T24, T36 (D1); T48, T60 (D2); T72, T84 (D3); D4, D5, D7, D10, D14 and D28. At these time-points, various measurements including full haematology, biochemistry and urine dipstick will be performed and samples will be tested for the presence of malaria parasites (asexual and sexual) by microscopy and molecular methods (for details: table 3 Sampling schedule). Five individuals in each intervention group with G6PD deficiency will be invited to participate in pharmacokinetic analysis, which will involve taking a total of 7 venous blood samples of less than 1 ml each on days 0, 1 and 2.

Study site and population
The study will be carried out in the clinical trial field site of Banfora, Comoé province, 90 kilometres southwest of Bobo-Dioulasso. This site is being used for malaria vaccine trials and has an excellent infrastructure for clinical trials, including a well equipped clinical laboratory and staffed by skilled medical personnel.

Participant screening and recruitment
Study subjects will be recruited from villages surrounding the Banfora clinical trial centre. Census lists will be prepared to screen individuals in the proposed age-range who will have a blood slide and hemoglobin levels measured by HemoCue. Screening blood smears will be read and counted by the outpatient laboratory technicians; parasitemic individuals with normal Hb levels (Hb>11g/dL) will be tested for G6PD enzyme function by Beutler test [41].
Upon referral to the study clinic, a standardized screening interview and physical examination will be conducted by study clinicians. This interview will go through the initial screening selection criteria that are given above. All individuals with fever or other symptoms suggestive of infection are excluded from study enrolment since infection (symptomatic malaria, pneumonia, hepatitis, typhoid fever, among others) itself might precipitate G6PD-related haemolysis; these individuals will be referred back to the standard outpatient clinic for diagnosis and treatment of their acute clinical condition.

Enrolment
Individuals who agree to participate for screening and meet all inclusion criteria will be invited for enrolment. During enrolment procedures, they will again be informed about the objectives and practical consequences of participation and asked to sign an informed consent form. The possibility of withdrawal from the study, at any time and without any declaration of the reason will again be pointed out. Participants will then be invited to sign the written consent form adjoining the written study information and approved by the IRBs to participate in this clinical trial; participants will also be invited to sign consent for the future use of biological specimens obtained during the course of the study. If the participant is unable to write, their fingerprint will be used in substitute for a signature, and a signature from an impartial witness to the informed consent discussion will be obtained. Two copies of the consent form must be signed. The participant / impartial witness will sign / fingerprint one copy for the study staff and one copy to keep for themselves.
Following the informed consent discussion, participants will be given their copy of the form to keep which includes the study information, the signed consent form and contact names and telephone numbers to use if they have further questions regarding the study or follow up procedures.

Dose escalation protocol
After enrolment, participants will be assigned to the lowest-dose open cohort, with enrolment in the second cohort initiated after tolerability and short-term safety is demonstrated at the preceding lower dose (this enrolment to the second cohort accounts for G6PD deficient participants only).
Within each cohort, the first 2 participants of the intervention group are treated and monitored for 6 days for immediate side-effects and haematological abnormalities before the rest of the participants of that particular intervention group are enrolled and treated. Once safety of these first 2 participants is confirmed, the next 4 subjects are enrolled and treatment for the next 4 subjects is initiated on day 2 of the last treated participant of the preceding 4 subjects. The last two groups for each intervention group comprise 5 individuals, making a total of 20. After inclusion of the intervention group of the first cohort (n=20) is completed (follow-up day 14 of last participant in that group), a 10-day safety observation period is installed before enrolment of the intervention group of the second cohort is initiated.

Total participants 20
The following scenarios will prompt urgent DSMB review (< 24hours) to consider whether it is appropriate to stop the trial: 1. There is any drop in Hb below 5 g/dL 2. There is any need for blood transfusion 3. 3 or more participants have a drop in Hb >4 g/dL 4. There is any requirement for haemodialysis 5. Any severe adverse event defined as any untoward medical occurence or effect that at any dose: results in death, is life threatening, requires hospitalization (attributable to the study drug), results in persistent or significant disability or incapacity.
If an enrolled participant discontinues the study for reasons unrelated to treatment before completing the final visit on day 28, an additional participant will be enrolled in the same group. This process will be repeated until either both cohorts have their follow-up completed or enrolment to a subsequent cohort is suspended because stopping criteria are met.

Treatment and administration procedures
All participants will be treated with artemether-lumefantrine (AL; Coartem®; Novartis Pharma) For participants who develop anaemia (Hb <11 g/dL) during the study, we will follow international and Burkina Faso national treatment guidelines.

Follow-up and samples taken
After initiation of treatment, participants are asked to return to the clinic on the evening of day 0, twice daily on days 1, 2, and 3, and once daily on days 4, 5, 7, 10, 14 and 28. On days 0, 1 and 2 study drugs are administered twice daily. On all visits, participants will be examined clinically (except for day 28) and a structured questionnaire is used to determine the occurrence of side effects. During screening, twice daily on days 0, 1, 2, 3, and once daily on days 4, 5, 7, 10, 14 and 28 after initiation of treatment, participants will have a Hb sample taken by finger prick. Hb concentration is determined by HemoCue photometer [Angelholm, Sweden].
A blood slide will be taken once daily on all days -except day 4 and 5 -to determine whether asexual parasites and gametocytes have been cleared. A longer follow-up is not needed since currently, there are no indications that PQ influences the clinical efficacy of AL and the study is purely designed to determine safety in the first four weeks after initiation of treatment in G6PD deficient individuals.
An RNA sample for gametocyte detection by quantitative nucleic acid sequence based amplification (QT-NASBA) will be collected on days 0, 2, 3, 7, 10 and 14 [2]. Biochemistry (ASAT, ALAT, total Bilurubin, Creatinin) and full blood count (including reticulocytes) are done for all individuals twice daily on days 0 (enrolment) and 1, and once daily on days 3, 7, 14 and 28. Filter paper blood samples will be collected at screening and on day 14 for quantification of G6PD activity. Dried blood spots will also be used to determine G6PD genotype and mutations in the CYP450 genes retrospectively, which may influence PQ metabolism and therefore haemolysis.
Urine dipstick will be done twice daily on days 0, 1, 2 and 3 and once daily on days 4, 5, 7 and 28. Table 3. Sampling schedule

Day of follow-up
Pfs25 QT-NASBA technique is gametocyte specific and has an approximate detection limit of 0.1 gametocytes/µL in the blood volume taken. NucliSens Basic kits are used for amplification in accordance with the manufacturer's instructions. A standard dilution series of mature, in vitro cultured NF54 gametocytes [44] is included in each run. Gametocyte detection by Pfs25 QT-NASBA is done for all samples on days 0, 2, 3, 7, 10 and 14 after initiation of treatment. Day 28 is excluded from this analysis because gametocyte prevalence on this day may be due to re-infections [45].

Haematology and biochemistry
On all time points Hb will be assessed using a HemoCue® photometer (Ängelholm, Sweden). This is done in order to avoid discrepancies between HemoCue® and full blood count Hb results. The HemoCue® produces a point-of-care result. A venipuncture sample will be taken twice daily (1.5 mL per sample for haematology; 2.5-3 mL per sample for biochemistry) on days 0 (enrolment) and 1, and once daily on days 3, 7, 14 and 28 for a full blood count (e.g. haemoglobin, white blood cell count, reticulocytes, platelets and haematocrit) and biochemistry (renal and liver function tests). On any day or time when the clinician feels this is required for the investigation of adverse events, additional samples will be taken. These samples will be used for a full blood count and biochemistry including, for example, renal and liver function testing and urinalysis for haemolysis.

G6PD deficiency testing
An EDTA tube or heparinized haematocrit capillary tube will be used to acquire blood for glucose-6phosphate-dehydrogenase semi-quantitative fluorescent spot test for initial screening in the clinic site. This will require approximately 0.5 ml of blood. A reagent solution containing Glucose-6-P + NADP+ is mixed with whole blood. Samples obtained from normal or slightly reduced G6PD activity will show strong fluoresce. Failure to fluoresce after 10-minutes of incubation suggests a total or marked deficiency of G6PD. This test may fluoresce falsely if the study participant has had a blood transfusion within the last 90 days hence, these persons will be excluded from the study. In addition, dried blood spots obtained on filter paper on screening and day 14 will be stored for quantitative G6PD testing. This will be performed using the G-6-PD OSMMR 2000 kit [R&D Diagnostics]. The dried blood spots will also be utilized for G6PD &CYP450 genotyping.

Pharmacological analysis
Pharmacokinetic evaluations will be obtained on approximately one quarter of the participants in intervention groups with G6PD deficiency, a maximum of 10 G6PD deficient participants.
Information on pharmacokinetic analysis is included in the study information sheet and procedures will be explained to participants on day 0; those who are selected and agree to participate will be asked to come for sampling on days 0 to 2. The sampling on day 0 will happen whilst they are at the clinic for their first day of AL and the study dose of PQ.
The pharmacokinetic sampling will involve taking a total of 7 venous blood samples of less than 1mls. The total amount sampled, being approximately 7 mL in 3 days. The first sample is just prior to the PQ dose (a baseline sample) and the subsequent six doses are at intervals up to 72 hours after the dose of PQ. The blood samples will be taken at fixed times between 8am to 5pm. The first 5 samples are taken on day 0 and they will be taken through a venflon, placed when the baseline pharmacokinetic sample is taken. If a venflon is not placed successfully, a butterfly needle may be used. The last two samples (one on day 1 and one on day 2) will be taken by individual blood draws (venipuncture). The participant will be asked to stay in the clinic between sampling times on day 0.
In order to minimize the total number of blood draws per participant, the sampling timeframe has been randomized so that over the total population of participants, a population pharmacokinetic model can be constructed for analysis. Six randomized sample times will be allocated to sequential consenting participants in opaque envelopes. Each sample time is within a window so that there are 5 samples on day 0 and one each on days 1 and 2. Pharmacokinetic samples will be analysed in Professor Niklas Lindegardh's laboratory in Mahidol University, Bangkok, Thailand, where the randomized sampling framework was generated.

Selection for pharmacokinetic analysis
On day 0, pharmacokinetic procedures will be explained and consent obtained (in Clinical study informed consent form); every 4 th G6PD deficient adult who was enrolled in an intervention group will be invited to participate in pharmacokinetic sampling. If this study participant declines inclusion in the pharmacokinetic study, the next consecutive participant will be invited. If a participant is included and vomits within 30 minutes of primaquine dose, the next participant will be invited. The Clinical Study consent interview will be conducted by study clinicians. . All participants undergoing pharmacokinetic sampling must satisfy the following criteria in addition to the study's inclusion criteria: Pharmacokinetic analysis inclusion criteria:

1.
Siting of secure blood sampling access feasible (venflon/ butterfly needle) on day 0

2.
Willing and able to attend study clinic by 7.30am on days 0-2

3.
Willing to stay on study clinic premises between 8am to 5pm on day 0 Study subjects included in the pharmacokinetic analysis will be seen by a study clinician who will select the next available pharmacokinetic opaque envelope. This will contain a sheet with the sampling times for the participant. The clinician will label this sheet with the participant's study number. The clinician will be responsible for adhering to the sampling times.
The clinician will fix venous access (using a venflon or butterfly) and take the baseline sample. All blood will be taken into heparinised tubes. The subsequent 4 samples will be taken at the allocated sampling times (Table 4). Then the fixed venous access will be removed. The participant will be informed as to what time they need to present by on days 1 and 2 for the last two samples. On these days, the sample will be taken by phlebotomy with no fixed access.

SAMPLE SIZE CONSIDERATIONS
For each of the 2 intervention groups (PQ in G6PD deficients) 20 individuals will be enrolled. This number of individuals per group will give us a 88% probability of detecting at least one severe event and ~61% probability of detecting at least 2 severe adverse events if we assume that the probability of a serious adverse event is 10% for each individual. The control groups are included to support the interpretation of haemoglobin concentrations following treatment (i.e. a reference group where a reduction in Hb is not due to primaquine in combination with low G6PD enzyme activity); the size of these control groups was based on expert opinion (WRAIR, Tafenoquine group) and not based on sample size calculations.

DATA ANALYSIS PLAN
The primary safety outcome, risk of haemolysis, defined as the proportion of individuals in each group that has a drop of >2.5 g/dL in haemoglobin levels, will be calculated. Mean (± SD) maximal fall (±) in Hb (haemoglobin, g/dl) from enrolment to day 28 of follow-up, follow-up day of Hb nadir, maximal percentage fall in Hb level compared to enrolment value, and evidence of black urine (urine dipstick positive for Hb) will be determined. The incidence of serious adverse events by sign, symptom, laboratory parameter and relationship to taking study drug will also be determined.

ETHICAL CONSIDERATIONS
This study includes G6PD deficient individuals who are at an increased risk of haemolytic side-effects of PQ as well as individuals with normal G6PD function. Within the G6PD deficient population, safety of the AL-PQ combination will be studied thoroughly. This information is essential since all parasitologically-confirmed P. falciparum malaria patients in elimination programmes will now receive a single dose of PQ on the first day of treatment with ACT according to new WHO policy guidelines. This includes G6PD deficient individuals who account for up to 30% of the population in some malaria-endemic areas [8,11].
The individual participant will benefit from treatment for their malaria infection. The treatment given, AL-PQ, is recommended by the WHO for elimination settings although the PQ component provides no immediate benefit to the individual. At a population level, treatment with PQ can reduce malaria transmission [9] and lead to a lower risk of infection experienced by individual community members [46]. This study is an essential requirement before transmission-reducing interventions with community benefit can be implemented.
The current study will restrict enrolment to asymptomatically infected males with normal Hb values.
Careful monitoring during 28 days will minimize the risk of participation for individuals. Clinical staff will receive training on standardised protocols for the detection and management of haemolysis.
Blood sampling will be restricted to the absolute minimum to answer the study questions, but may be increased in some individuals to monitor safety parameters. The total blood volume that is taken per participant during their 28-day period of participation will be ~25mL. Participants will be compensated for their participation.

Confidentiality
All participant records are to be used only for the purpose of this research project. Names will not appear on labels on laboratory specimens or in any report resulting from the study.
Identifying information will be kept in a metal cabinet that is locked and only accessible to the study clinician or his clinical representative in case he is absent. Paper forms are only accessible to senior research staff (i.e. clinician, nurse and scientific personnel but not field workers or laboratory assistants); names and addresses will be removed from digital files which will be password protected. All materials collected in this study will be labelled with a study identification number that cannot be directly linked to identifying information.

Participant reimbursement
Participants will receive a small parcel with items such as sugar, rice and flour on a daily basis to compensate them for the time lost during the study. This is common practice for all clinical trials at CNRFP.

Good clinical practice
The study will be conducted in accordance with the latest South Africa revision of the Declaration of Helsinki and local regulatory requirements, and study staff will receive a local workshop in Good Clinical Practice before the study starts. AL is the first line antimalarial drug in Burkina Faso; PQ is a registered drug that has been extensively used over the last 50 years and is recommended by the WHO. A data safety and monitoring board will be installed to ensure the trial is completed according to GCP standards.

Withdrawal of individual subjects
Subjects can leave the study at any time for any reason if they wish to do so without any consequences. The principal investigators and trial clinicians can decide to withdraw a subject from the study for urgent medical reasons.
Subjects can be withdrawn from the study procedures for the following reasons:  Any serious adverse event (SAE)  Any AE that, according to clinical judgment of the trial clinicians, is considered as a definite contra-indication to proceeding with the study procedures  Withdrawal of informed consent by subject  Failure to comply with the prescribed study protocol or complete loss to follow-up  Repeated vomiting within 30 minutes after administration of study drug

Premature termination of the study
The study may be discontinued for the following reasons:

Management of haemolysis
In previous studies PQ has never been exclusively administered to G6PD deficient individuals.
However, we know that in nearly 1500 patients studied prospectively in trials of single dose PQ given as a gametocytocide in Africa, Asia and South-America, no severe haemolysis requiring blood transfusion was reported [3,9,[24][25][26][27][28][29]. Furthermore, in Tanzania [2], none of the participants (who all had a normal G6PD enzyme function) experienced symptoms of anaemia and no child required a blood transfusion. In the second study in Tanzania [26], one G6PD deficient child who received PQ 0.75mg/kg had severe anaemia, but did not require a blood transfusion and recovered with haematinic drug treatment. In Sudan [25], there were no severe or serious or adverse events and severe anaemia was not reported. In Uganda [Eziefula, under review], a subanalysis by G6PD genotype revealed that individuals with A-genotype had a reduced fall in haemoglobin with 0.4mg/kg and 0.1mg/kg, compared to the fall with 0.75 mg/kg and there were no episodes of severe haemolysis or requirement for blood transfusion.
Asymptomatic malaria infection is expected to precipitate a small degree of haemolysis, due to erythrocive parasitisation. However, given that we test lower doses of PQ than those previously associated with haemolysis and we administer PQ at a single time-point, we do not expect significant haemolysis to occur at any point during the study. Furthermore, the dose-escalation schedule and halting criteria allow us to conduct this trial safely and will minimise the risk of haemolysis.
For the purposes of systematic and responsible safety monitoring, the following detailed protocols have been developed for the management of participants in whom haemolysis is suspected.

Measures of haemolysis
Haemolysis will be diagnosed according to criteria in a study SOP, detailing the size of haemoglobin fall (measured by Hemocue®) after PQ treatment and the absolute haemoglobin value. In addition, any participant presenting with or complaining of tea-coloured or black urine will be assessed for haemolysis. If haemolysis is suspected, a venipuncture sample will be taken for a full blood count, biochemistry and Hb by Hemocue, a blood film will be prepared and analysed for abnormal red blood cells, urine dipsticks and urine microscopy will be taken and a clinical examination will be performed.

Anaemia
Participants with haemoglobin below 11 g/dL at screening will not be included and treated with haematinic drugs and de-worming according national guidelines.

Haemolysis
Participants with evidence of mild haemolysis will be observed and monitored and haematinic drugs will be considered according to national guidelines. Haemoglobin testing will be repeated according to the participant's clinical progress. Any participant reporting black urine will be assessed by a study clinician. Black urine is defined as any dark-coloured urine with brown or black pigments (not orange). A full clinical exam and history will be obtained, a urine sample sent for analysis (evidence of haematuria, proteinuria) and a venous blood sample drawn for assessment of full blood count, biochemistry and renal function (including bicarbonate). A blood film will be assessed for abnormal red blood cells (schistocytes). Requirement for blood transfusion will be assessed by trial clinicians, considering the size and rate of Hb drop and signs of clinical compromise, according to national guidelines.

Adverse events
Adverse events are defined as any undesirable experience occurring to a subject during the study, whether or not considered related to trial. An AE can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding), symptom, or disease (new or exacerbated) temporally associated with a study intervention. AEs may include events that occur as a result of protocol-mandated procedures (i.e. invasive procedures). All AEs reported spontaneously by the subject or observed by the trial clinicians or their staff will be recorded.
Abnormal laboratory findings or other abnormal assessments that are judged by the trial clinicians to be clinically significant will be recorded as AEs or serious adverse events (SAEs) if they meet the definition. The principal investigators and trial clinicians will exercise their medical and scientific judgement in deciding whether an abnormal laboratory finding or other abnormal assessment is clinically significant.

Serious adverse events
A serious adverse event is any untoward medical occurrence or effect that at any dose:  Results in death  Is life threatening (at the time of the event)  Results in persistent or significant disability or incapacity  Is a new event of the trial likely to affect the safety of the subjects, such as an unexpected outcome of an adverse reaction or a major safety finding from a newly completed animal study etc.
All SAEs will be reported immediately to the principal investigators. The latter will subsequently

Adverse Event Data Collection
Safety assessments will be performed, and recorded by the trial clinicians. All AEs/reactions, observed by the trial clinicians or by the subject, will be accurately documented in the case report form. For each event/reaction the following details will be recorded:

Follow-up of subjects that voluntarily withdraw participation
If a subject fails to appear for follow-up, effort will be undertaken to locate or recall him/her or at least to determine his/her health status. These efforts will be documented in the subject's CRF and source documents.
In the event that a subject discontinues the study for any reason, the trial clinicians will conclude appropriate safety assessments for the subject and, in the case of an (S)AE, which has not yet resolved, they will schedule follow-up visits until the AE resolves or stabilizes.

Follow-up of adverse events
All AEs will be followed until they have abated, or until a stable situation has been reached.
Depending on the event, follow-up may require additional tests or medical procedures as indicated, and/or referral to other specialists as the trial clinicians see fit.
The trial clinicians will follow-up subjects: -with SAEs or those withdrawn from the study as a result of an AE, until the event has resolved, subsided, stabilized, disappeared, the event is otherwise explained, or the subject is lost to follow-up.
-with other non-serious AEs, until the subject has completed the study or is lost to follow-up.
Clinically significant laboratory abnormalities will be followed up until they have normalized, or until an alternative explanation, that is not related to the study has been provided. Additional information (including but not limited to laboratory results) relative to the subsequent course of such an abnormality will be made available to the LSM.
AEs reported during the trial will be assessed and classified as:

Follow-up of serious adverse events
In the case of a suspected SAE, the nurses or trial clinicians will immediately report this occurrence to the principal investigators (PIs). The latter will subsequently report to the Sponsor (London School of Hygiene and Tropical Medicine) within 24hrs of awareness and to the chairpersons of the DSMB and Ethics Committee. In their report, the PIs will provide as much details as possible about the suspected SAE. At the same time the PIs, in consultation with the trial clinicians will stop any administration of study drugs, if the SAE is thought to be as a result of the study drug. The trial clinicians and consultant clinicians will immediately institute appropriate medical treatment to the subject who develops an SAE until the event has resolved, subsided, stabilized, disappeared, or the event is otherwise explained.
All findings of investigations and treatments administered will be recorded on the subject's CRF and also on the SAE Report Form. The trial clinicians may also refer the subject to other specialists as they see fit. The PIs will forward the completed SAE form to the LSM to assess whether the SAE could have been a result of the study procedures, as per standard criteria. Findings of this assessment will be forwarded to the sponsor (LSHTM) and to the chairpersons of the DSMB and Ethics Committee.

LSM and Data Safety Monitoring Board (DSMB)
For this study, a LSM is appointed, who is based in Banfora and will be involved in the review of SAEs and subject safety. He is an experienced clinician and independent of the investigators team. His main responsibility will be the assessment of the events and recommendation regarding halting further study procedures.
Furthermore, an independent DSMB has been appointed, which consists of three skilled professionals constituted according to sponsor procedures that respect international standards.
These experts will monitor the progress of the study with particular interest in the safety of trial subjects. The LSM and DSMB will receive progress reports by email after the first 10 individuals completed day 14, then after the second 10 individuals completed day 14, etc. This will result in a total 7 safety reports at the end of the trial. These reports will be prepared by the PIs and trial clinicians and include a list of all reported AEs and any safety laboratory values (e.g. Hb, ALAT/ASAT) outside the normal range.
All SAEs will be reported by the PIs to the Sponsor (LSHTM), the LSM, the chairpersons of the DSMB and Ethics Committee within 24 hours of awareness. The DSMB is empowered to suspend the enrolment to the trial pending review of potential safety issues.

Data monitoring
A local independent data monitor from CNRFP will undertake data monitoring at site (e.g. source data verification). In addition, there will be central monitoring by a monitor that will be appointed in consultation with the sponsor. This central monitoring will undertake statistical monitoring of inliers and outliers, data management review etc.

TIME FRAME AND DURATION OF THE PROJECT
Data collection will take place in 2014. The enrolment is envisaged to take 8 weeks; the clinical follow-up 4 weeks per individual. Community sensitization is expected to start in February 2014, the data collection will take place between March 2014 -May 2014. The project is followed by a community meeting to present some of the preliminary findings. Data analysis will take place in the second half of 2014.