Evaluation of pharmacokinetics of single-dose primaquine in undernourished versus normally nourished children diagnosed with Plasmodium vivax malaria in Mumbai.

BACKGROUND: The pharmacokinetics of primaquine [PQ] have been the subject of studies in both adults and healthy participants. However, there is no study on its pharmacokinetics in a setting of undernourishment. In India, there is evidence to show considerable malnourishment in children that in turn can affect drug pharmacokinetics. Given that the country is moving towards malaria elimination, the present study was planned with the objective of comparing pharmacokinetics of the drug in undernourished children relative to normally nourished children.
MATERIALS AND METHODS: After Institutional Ethics Committee approval, children of either gender between the ages of 5 and 12 years and smear-positive for Plasmodium vivax malaria were included. Nourishment status was determined using the Indian Academy of Pediatrics classification of protein energy malnutrition based on Khadilkar's growth charts. Twelve children each were enrolled in the two groups. PQ was given in the dose of 0.3 mg/kg/d and blood collections were made at 0, 1, 2, 3, 4, 6, 8 and 24 hours post-dosing. Levels were estimated by high-performance liquid chromatography. Chloroquine in the dose of 25 mg/kg was given over three days along with supportive care.
RESULTS: Of the 24 children, there were 17 boys and 7 girls. There was a statistically significant difference in the body weight between the undernourished and the normally nourished children [21.5 ± 5.52 vs. 28.8 ± 8.84, P < 0.05]. PQ levels showed wide inter-individual variation in both groups. No significant difference was seen in any pharmacokinetic parameter between the two groups. DISCUSSION: This study adds to the limited body of evidence on the pharmacokinetics of PQ in children with malaria and indicates that the dosing of primaquine could potentially be independent of the nourishment status.

Introduction

The Global Technical Strategy for Malaria (2016-2030) has set an ambitious target of eliminating the disease from at least 35 countries [one of which is India] by the year 2030.[1] A key to its success will be addressing the problem of Plasmodium vivax malaria. The World Malaria Report of 2018 estimates that more than 80% of the world's P. vivax malaria burden is due to three countries with India being one of them.[2] Chloroquine along with primaquine [PQ], has been the treatment of choice for the radical cure of P. vivax malaria. Since its approval by the US FDA in 1952, PQ remains the sole available drug to prevent relapses and needs to be given for 14 days. Additionally, it is also a useful transmission-blocking agent for Plasmodium falciparum malaria.[34] Hence, its optimal use is a key component of all malaria control and elimination efforts.

The pharmacokinetics of PQ have been the subject of several studies that have largely involved adult patients or participants.[5] There are however limited studies on its pharmacokinetics in children[678] and no study on its pharmacokinetics in a setting of undernourishment. In India, there is evidence to show that 46% of children under the age of 5 years are moderately to severely underweight, 38% are moderately to severely stunted and yet another 19% show moderate to severe wasting.[9] Given that the country is moving towards malaria elimination, pharmacokinetics of PQ in undernourished children will add to the body of evidence in this field and help take policy decisions on its use in this particular population. The present study was thus carried out with the primary objective of evaluating the pharmacokinetics of a single oral dose of PQ (0.3 mg/kg) in undernourished children relative to children who were normally nourished and diagnosed with P. vivax malaria. Safety formed the secondary objective.

Methods

Ethics – The study protocol was approved by the Institutional Ethics Committee [IEC approval EC/88/2010] and written informed consent was taken from the parents/guardians. Assent was in addition taken from children over the age of 7 years. The study protocol was retrospectively registered with the Clinical Trials Registry of India (CTRI/2012/12/003188).

Setting – A tertiary referral center in the city of Mumbai, India

Study design- An open-label study in consecutive patients.

Selection criteria- Inclusions were children of either gender between the ages of 5 and 12 years, smear positive for P. vivax malaria and classified either as undernourished or not. Nourishment status was determined using the Indian Academy of Pediatrics classification of protein energy malnutrition [PEM] based on Khadilkar's growth charts[10] Those with a past history of hypersensitivity to primaquine, any clinically significant renal, hepatic or cardiac disease, history of having received antimalarials in the past one month, G6PD deficiency and those with severe or complicated malaria or history of HIV infection were excluded.

Sample size- Since there was no similar study in the country and the objective was only single-dose pharmacokinetics, n = 12 children were enrolled in each of the two groups without any formal sample size calculations being made.

Treatment for vivax malaria- All participants received chloroquine [CQ] in the dose of 10 mg/kg on the first day followed by 10 mg/kg on Day 2 and 5 mg/kg on Day 3.[11] They also received supportive management in the form of intravenous fluids, anti-pyretics and anti-emetics as and when required.

PQ dosing, time points of blood collection and subsequent follow up- The drug was given in the daily dose of 0.3 mg/kg/d after food for a total period of 14 days.[12] Therapy with PQ was initiated on Day 4 after completion of treatment with CQ. Blood collection for pharmacokinetics was done only after the first dose and patients were discharged on Day 5. The drug was given under supervision on Days 4 and 5 after which parents were counseled to complete the entire 14-day course. They were also advised to follow up weekly up to day 28 for clinical evaluation and for peripheral smear examination for malaria parasites. Blood for primaquine levels was collected at baseline [0h], 1, 2, 3, 4, 6, 8, and 24 hours post-dosing.

Analysis of PQ levels - Plasma PQ concentration was analyzed by high-performance liquid chromatography (HPLC) using the method of Dua et al.[13] Briefly, the HPLC conditions were a flow rate of 1.2 ml/min, wavelength 254 nm, drug retention time of 8.5 minutes using a 3 x 30 mm Bondapak column. The assay sensitivity was 20 ng/ml and it was linear in the range 28.5 – 3365 ng/ml.

Pharmacokinetic analysis- All analysis was done assuming first-order kinetics and a one-compartment model using the Win non-Lin software version 6.1. Primary pharmacokinetic parameters Cmax, T max were obtained directly from the data, while Area Under the Curve [AUC 0-t and AUC0-inf], Clearance (CL), Volume of distribution (Vd) and Elimination rate constant (Kel) were calculated using the linear trapezoidal rule.

Statistical analysis – Both descriptive and inferential statistics were used. Age, body weight and the dose of PQ are described as mean [SD]. Quantitative pharmacokinetic data are presented as median and range. They were assessed for normality using the Kolmogorov-Smirnov test and between-group comparison was done using either the two-sample student's t-test or Mann–Whitney U test as appropriate. All analyses were done using Microsoft Excel at 5% significance.

Results

Demographics

Of the 24 children, there were 17 boys and 7 girls. Their mean [SD] age in years was 9.52 [2.0], body weight [kg] was 27.2 [7.85], and the daily dose of primaquine [mg] was 7.72 [2.35]. The only significant difference was seen in the body weight between the undernourished and the normally nourished children [21.5 ± 5.52 vs. 28.8 ± 8.84, P < 0.05]. Demographics are depicted in Table 1.

Table 1

Demographics

Variable	Normally nourished children n=12	Malnourished children n=12
Age in months [range]	89-164	63-143
Gender	8 males, 4 females	9 males, 3 females
Body weight in [range] kg	15-46	15-32
Dosage (mg)*range and mean (SD)	4.5-13.8 (8.64±2.65)	4.5-9.6 (6.8±1.65)

P<0.05 for all variables except body weight which was significantly different between the two groups

Among the n = 12 children enrolled in the undernourished group, 6 had Grade I PEM, 4 Grade II and 2 had Grade III PEM. There were 9 boys and 3 girls who were undernourished.

All 24 patients responded to treatment with CQ and were clinically disease free as also negative on the peripheral smear for the malarial parasite at Day 28 follow up.

Single-dose PQ [0.3 mg/kg] pharmacokinetics

The single-dose was rapidly absorbed and completely eliminated by 24 hours in all children regardless of the nourishment status. Wide inter-individual variation was seen in the levels in both groups of patients. The dose-adjusted pharmacokinetic data (for body weight) for all 24 children is given in Table 2 for all the pharmacokinetic parameters. No significant difference was seen in any parameter between the two groups.

Table 2

Comparison of pharmacokinetics of single-dose [0.3 mg/kg] primaquine in undernourished versus normally nourished children [n=12/group, dose adjusted for body weight]

Pharmacokinetic parameters [units]*	Normally nourished children n=12	Undernourishment n=12
Cmax[ng/ml]	9.63 (2.15-16.40)	8.08 (2.05-44.23)
AUC (0-inf) [ng.hr/ml]	96.10 (12.94-241.10)	63.4 (20.54-782.45)
t1/2 [hrs]	6.67 (1.32-39.88)	3.69 (1.38-15.68)
Tmax[hrs]	2.5 (1-6)	3 (1-8)
Cl [L/hr]	1.80 (0.72-19.06)	2.55 (0.29-10.14)
Vd[L]	16.19 (2.50-121.79)	12.13 (2.03-92.71)

Safety

The drug was well tolerated and no adverse events were noted in any of the 24 children.

Discussion

This study is perhaps among the few studies worldwide that have evaluated the pharmacokinetics of PQ in children. It's strength lies in evaluation of kinetics of the drug in P. vivax malaria against a backdrop of undernourishment. Wide inter-individual variability was seen in the drug levels in the two groups of patients. However, the difference in pharmacokinetic parameters between the normally nourished and undernourished children was not statistically significant. Given that there were no safety concerns with the 14-day dosing, the drug's short half-life and evidence in adults to show a lack of difference between single and multiple-dose pharmacokinetics,[14] it is likely that the drug's dosing does not require alteration in undernourished children.

Optimizing drug dosing in malaria requires an understanding of four determinants- antimalarial pharmacokinetics, parasite susceptibility, host defense, and parasite burden[15] Of these, anti-malarial pharmacokinetics is known to be affected by age and nourishment status as is host defense.[16] We compared the pharmacokinetic parameters seen by us with the other two pharmacokinetic studies with primaquine done in African children. One study by Moore et al.[7] was conducted in healthy children in Papua New Guinea and included population pharmacokinetics while the second was conducted in asymptomatic children aged 2-14 years with P. falciparum malaria and evaluated single-dose kinetics similar to the present study.[7] The median Cmax seen by us was far lower than that seen by Moore et al. (8.94(2.05-44.23) vs. 115 (56.5-226)).

The difference between our study and similar studies in literature could be attributed to several reasons. Moore et al. [Papua New Guinea] studied pharmacokinetic properties of much higher single-dose abbreviated regimens [0.5 mg/kg and 1 mg/kg respectively in two studies] in healthy children; whereas our children had a diagnosis of vivax malaria. In addition, the method of estimation used by them was liquid chromatography mass spectrometry [LCMS, greater sensitivity] whereas we used HPLC. When we compared our values of Cmax, Tmax and AUC 0-inf [noncompartmental comparison] to the values seen in the study by Goncalves et al.,[6] while on an average, our levels were lower, our values were closer to Goncalves et al. relative to the study by Moore et al. where we saw a much larger difference. Goncalves et al. have again used LCMS as the method of estimation and the children studied had falciparum rather than vivax malaria. A more recent study(2020) from Brazil[8] included 85 children with vivax malaria, similar to our study who were dosed using an age-based dosing regimen. The median [range] Cmax values are again much higher than those seen by us {94 (51-144) vs 9.6 [2.2, 16,4]}. These estimations were however done on day 7 of dosing while we estimated levels immediately after a single-dose. Between study comparisons are depicted in Table 3.

Table 3

Comparison of Cmax, Tmax and AUC 0-inf of present study with similar studies in literature

Pharmacokinetic parameter/units	Present study, India n=12 normally nourished patients with vivax malaria, aged 5-12 years Dose=0.3 mg/kg, single-doseHPLC estimation	Goncalves PB et al. n=40 afebrile children [2-14 years], asymptomatic with P falciparum malaria, Burkina FasoDose=0.25 mg/kg single-doseLCMS estimation	Goncalves PB et al. n=40 afebrile children [2-14 years], asymptomatic with P falciparum malaria, Burkina FasoDose=0.4 mg/kg single-doseLCMS estimation	Moore RB et al. [2014], n=28 healthy children, Papua New Guinea, Population Pharmacokinetic data based on simulationsDose=0.5 mg/kg daily x 14 daysLCMS estimation	Moore RB et al. [2014], n=28 healthy children, Papua New Guinea, Population Pharmacokinetic data based on simulationsDose=1 mg/kg daily for 7 daysLCMS estimation	Moore RB et al. [2014], n=28 healthy children, Papua New Guinea, Population Pharmacokinetic data based on simulationsDose=1 mg/kg twice daily for 3.5 daysLCMS estimation	Vieira MVDF et al.[2020]n=85 children with vivax malaria in Brazil that used age based dosingDose=3.22 mg/kg TOTAL median dose[Recommended dosing 0.25 mg/kg/d x 14 days or 0.3 mg/kg/d x 7 days]HPLC estimation. Levels estimated on Day 7
Cmax [ng/ml] Median [range]	9.63 (2.15-16.40)	50.2 (20.2-138.7)	88.9 [24.4-391.2]	115 (56.5-226)	230 (112-442)	275 (142-508)	94 (51-144)
Tmax [h]	2.5 (1-6)	1.6 [1.2-1.9]	1.6 [1.0-2.1]	Not given	Not given	Not given	Not given
AUC (0-∞) [ng.hr/ml]	96.10 (12.94-241.10)	450.4 (259.9-875.9)	730.7 (334.2-3315.4)	Not given	Not given	Not given	Not given

Our study has several limitations. We evaluated only single-dose kinetics and also did not estimate the major metabolite of primaquine, namely, carboxyprimaquine.[17] No children under the age of five years were studied and there were just two children with grade III PEM. All children had uncomplicated vivax malaria and it is possible that kinetics of the drug in severe P. vivax malaria may be different. Also, we did not evaluate the CYP2D6 status of the children enrolled. There is mounting evidence to show that poor metabolizers of CYP2D6 have higher concentrations of the parent drug and genotype may be an important determinant of dosing.[18] CYP2D6 as a determinant of primaquine levels was also shown by Goncalves et al.[7] in their study. In addition, our study is a traditional/classical pharmacokinetic study [non- population pharmacokinetics]. In Table 1, we have compared our results with studies where population pharmacokinetic modeling has been done. The latter studies are inherently limited by unknown concomitant variables, interaction or collinearity between these variables and the final model fit and the extent of uncertainty that it carries with it.[19]

The utility of primaquine today lies in its single-dose [0.25 mg/kg] use in P. falciparum malaria along with artemisinin-based combination regimes [ACT], where its use produces a greater reduction in gametocytemia relative to when ACTs are used alone.[20] In vivax malaria, research is now focused upon giving larger doses of the drug for shorter durations so as to improve compliance[21] as radical cure is now known to be a function of the total dose rather than duration of treatment. PQ regimens in children are usually age based. The present study adds to the limited body of evidence on the pharmacokinetics of PQ in children with malaria and indicates that the dosing could potentially be used independent of the nourishment status. Finally, appropriate dosing however must take into account at least the three key determinants identified thus far- age, body weight, and CYP2D6 status for individualizing and optimizing therapy in malaria [vivax or falciparum] in children regardless of nourishment status.

Financial support and sponsorship

Research Society of the Seth GS Medical College and KEM Hospital and the Indian Council of Medical Research

Conflicts of interest

Dr. Sunil Karande is the Editor of the Journal of Postgraduate Medicine.