Does gender, food or grapefruit juice alter the pharmacokinetics of primaquine in healthy subjects?

Abstract

Aims

To evaluate the effects of gender, food and grapefruit juice on the pharmacokinetics of primaquine in healthy subjects.

Methods

In a randomized, two-phase cross-over study, 10 male and 10 female healthy Vietnamese subjects were administered 30 mg primaquine in the fasting state or with food, followed by administration of primaquine with grapefruit juice.

Results

The pharmacokinetics of primaquine were comparable between male and female subjects, with geometric mean ratios of C_max = 0.89 [95% confidence interval (CI) 0.65, 1.22] and AUC = 0.80 (95% CI 0.56, 1.15). The mean CL/F of primaquine was slightly higher in males than in females [0.52 l h⁻¹ kg⁻¹vs. 0.43 l h⁻¹ kg⁻¹, mean difference of 0.09 (95% CI –0.10, 0.28), P= 0.32]. When compared with fasting state values, food increased the geometric mean C_max of primaquine by 26% (95% CI 12, 40) and the AUC by 14% (95% CI 3, 27). Similarly, grapefruit juice increased the geometric mean C_max by 23% (95% CI 4, 45) and the AUC by 19% (95% CI 4, 37).

Conclusions

The disposition of primaquine was comparable between genders, suggesting no need to modify the dose of primaquine for malaria treatment or prophylaxis. Food increased the oral bioavailability of primaquine, which may lead to higher antimalarial efficacy. Grapefruit juice increased the bioavailability of primaquine, with marked interindividual differences suggesting that people should not take primaquine with grapefruit juice.

Introduction

Plasmodium vivax malaria is responsible for significant morbidity among people in Asia, the Middle East and Latin America. Annually there are 70–80 million cases of vivax malaria reported throughout the world, causing tremendous human suffering [1]. Primaquine, an 8-aminoquinoline, is the only drug currently available for the radical cure or terminal prophylaxis of both P. vivax and P. ovale malaria. Because of the spread of primaquine-tolerant P. vivax strains in South-East Asia and Western Pacific countries, the World Health Organization International Travel and Health 2005 recommended that the adult dose of primaquine for these regions be increased from 15 mg daily to 30 mg daily for 14 days [2]. Primaquine is also a highly effective prophylactic agent against both falciparum and vivax malaria at a daily adult dose of 30 mg and because of its causal prophylactic activity it needs to be given only for 7 days after leaving a malarious area [3]. The US Control of Disease and Prevention recommends that, in special situations when other antimalarial drugs cannot be taken and in consultation with malaria experts, primaquine may be used for malaria prophylaxis at an adult dose of 30 mg daily [4]. Clinically important adverse events of primaquine use include gastrointestinal disturbances, methaemoglobinaemia and acute intravascular haemolysis in individuals deficient in glucose-6-phosphate dehydrogenase (G6PD) [5].

In spite of considerable clinical experience with primaquine, there is limited information on the pharmacology and metabolism of the drug. The pharmacokinetics of primaquine was studied by Mihaly and colleagues in the 1980s in healthy caucasian male subjects [6, 7]. They reported that primaquine is rapidly absorbed with an elimination half-life of about 7 h. It is a low to intermediate clearance drug and the low plasma primaquine concentrations obtained after ingestion are due to rapid and extensive tissue distribution. Primaquine exhibits linear pharmacokinetics over the dose range of 15–45 mg [7]. The pharmacokinetics of primaquine has been well characterized in males and, although virtually no data are available for females, a limited study in Thailand suggested that plasma clearance of primaquine was higher in females than in males [8].

Primaquine is usually taken with food to minimize gastrointestinal disturbances but there are no data on whether food affects the bioavailability of the drug. Food affects the absorption of many drugs [9], including the antimalarials chloroquine [10], atovaquone [11] and mefloquine [12]. Alteration of the rate and extent of absorption of drugs with narrow therapeutic indices, such as primaquine, could be clinically important in optimizing treatment and prophylactic efficacy.

Primaquine’s metabolism has been difficult to study because the expected aminophenol metabolites of primaquine and their amphoteric nature are unstable [13]. The major plasma metabolite of primaquine is carboxyprimaquine, which is not detected in urine, suggesting that it is subjected to further metabolism prior to excretion. Human liver microsomal studies have shown that the formation of carboxyprimaquine is cytochrome P450 (CYP450) dependent [14], but it is unclear which isozyme(s) is responsible for primaquine’s metabolism.

Grapefruit juice increases the oral bioavailability of a variety of CYP3A4 substrates [15, 16], including the antimalarial drugs halofantrine [17] and artemether [18]. Increase in plasma concentrations of these substrates is due to inhibition of mucosal CYP3A4 activity by grapefruit juice [16]. For drugs such as primaquine that are difficult to study metabolically, coadministration with grapefruit juice may provide an insight into whether gut CYP3A4 activity is involved in the drug’s first-pass metabolism.

The present study was conducted to investigate the effects of gender and food on the pharmacokinetics of primaquine and carboxyprimaquine. Since primaquine’s metabolism has not been well defined, we evaluated the possible interaction of grapefruit juice with primaquine.

Methods

Subjects and study site

Twenty healthy Vietnamese subjects [10 males, mean (± SD) age 34.4 (6.5) years; weight 57.7 (6.7) kg and 10 females, mean age 35.1 (10.3) years; weight 55.2 (7.6) kg] participated in the study, which was conducted at the Department of Infectious Diseases, Central Military Hospital 108, Hanoi, Vietnam from August 2002 to January 2003. The subjects were judged healthy based on medical history, clinical examination and routine laboratory testing (haematology and biochemistry). Subjects were not allowed to smoke or drink alcohol, caffeine-containing beverages, grapefruit juice or to eat grapefruit in the 24 h before and on the study day. The subjects were not on other medication and the female subjects were not pregnant or lactating. All subjects were classified as G6PD normal using the Sigma Diagnostic G6PD Kit (Immunodiagnostics Pty Ltd, St Peters, Australia), which is based on fluorescence of the subject’s red cell–substrate mixture containing G6PD on filter paper. The Review and Scientific Board of Central Military Hospital 108 gave ethical approval for the study and the subjects gave written informed consent before entering the trial.

Study design and drug administration

Each participant was studied on three occasions with a washout period ranging from 2 to 8 weeks (average 3.3 weeks). A randomized, open-label, cross-over study design was used for the first two occasions in which each subject was administered a single oral dose of 30 mg primaquine (four tablets of 7.5 mg primaquine base per tablet; Boucher & Muir Pty Ltd, Crows Nest, Australia) either in the fasting state following an overnight fast or with food (two bread rolls with a spread of 30 g of butter ‘President, Beurre, France; 82% fat, total fat content ∼ 28 g). The diet of bread and butter was selected to achieve a moderate amount of fat to evaluate whether a fatty meal would impact on the disposition of primaquine. The drug was also administered with 300 ml of water in the fasting and fed phases. Primaquine was administered within 5 min of ingesting the meal. After completion of the first two phases, the subjects were given the same dose of primaquine with 300 ml of grapefruit juice (50% concentration, grapefruit necta made from concentrated grapefruit juice; Don Simon Calcio, J. Garcia Carrion, Murcia, Spain). This phase of the trial was not randomized as no order effect was expected and to prevent a possible long-lasting carry-over effect of grapefruit juice. Subjects were allowed to have a standard hospital meal 4 h after each drug administration. They were also asked the nonleading question ‘How do you feel since you took the primaquine tablets?’ at 24 h after taking the drug. If a subject responded affirmatively with symptoms, the timing and intensity of the complaint were recorded.

Blood sampling

Venous blood samples were collected through an in-dwelling cannula inserted into a forearm vein and kept patent with heparinized saline. Blood samples (7 ml) were collected in lithium heparin tubes (Vacuette; Interpath, West Heidelberg, Australia) within 0.5 h (baseline) prior to primaquine administration and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, 12 and 14 h after drug administration. Subsequent blood samples were collected by venepuncture at 21 and 24 h after dosing. All blood samples were centrifuged at 1400 g for 15 min and the separated plasma samples were transferred to cryotubes (Nunc tubes; Medos, Noble Park North, Australia) stored at − 25 °C prior to transport to the Australian Army Malaria Institute on dry ice, where they were kept at − 80 °C until analysis.

Drug analysis

The reversed-phase high-performance liquid chromatographic (HPLC) methods described by Mihaly et al.[6] were used to measure plasma primaquine and carboxyprimaquine concentrations, with minor modifications. Briefly, the following modifications were made: (i) a Waters Symmetry C8 HPLC cartridge (150 × 3.9 mm)/guard holder (Waters, Woolloongabba, Australia) was used to measure both compounds; (ii) all glassware was silanized with AquaSil™ (Pierce, USA, Quantum Scientific, Milton, Australia); (iii) plasma extraction volumes were 0.5 ml for primaquine and 0.25 ml for carboxyprimaquine; (iv) the mobile phase for primaquine analysis consisted of acetonitrile : water (30 : 70) with 5 mm PIC B-5 low UV Reagent (Waters) and the flow rate was 0.6 ml min⁻¹; (v) the retention times for the internal standard [6-methoxy-8-(3-amino-1-methylpropyl-aminoquinoline)] and primaquine were 5.6 and 7.3 min, respectively; (vi) the mobile phase for carboxyprimaquine analysis consisted of acetonitrile:water (50 : 50) with 5 mm PIC B-5 low UV Reagent (Waters) and the flow rate was 0.6 ml min⁻¹; (vii) the retention times for carboxyprimaquine and the internal standard (indomethacin) were approximately 6.2 and 12.1 min, respectively; and (viii) the wavelength for UV detection was 267 nm for both primaquine and carboxyprimaquine. The limit of quantification of primaquine and carboxyprimaquine were 5 ng ml⁻¹ and 25 ng ml⁻¹, respectively. The interday assay coefficients of variation (CV%) for the measurement of primaquine at 5, 50 and 200 ng ml⁻¹ were 18.7%, 4.4% and 1.3% (n = 11), respectively. The interday assay CV% for carboxyprimaquine at 25, 250 and 1000 ng ml⁻¹ were 15.6%, 6.6% and 2.3% (n = 27), respectively. The inaccuracy of the method at 50 ng ml⁻¹ was 2% for both primaquine and carboxyprimaquine. The primaquine tablets used in this study were assayed for content and found to be within 4% (range 2–5%, n = 5 tablets) of the target value of 7.5 mg base per tablet.

Pharmacokinetic analysis

Maximum plasma drug concentration (C_max) and time to maximum concentration (t_max) were obtained from the plasma drug concentration–time curve. The elimination rate constant (k_el) was estimated by least-squares regression analysis of the post-absorption and distribution log plasma drug concentration–time data using at least 4 points. The elimination half-life (t_1/2) was calculated from the ratio ln2/k_el. The area under the drug concentration–time curve from zero to infinity (AUC_0–∞) and the area under the first moment curve (AUMC_0,∞) were calculated by the linear trapezoidal rule from the beginning of primaquine administration to the last data point and with extrapolation to infinity. The mean residence time (MRT) was calculated as MRT = AUMC_0,∞/AUC_0–∞. The oral clearance (CL/F) was expressed as a function of bioavailability (F) and calculated as the dose divided by AUC_0–∞, with complete systemic bioavailability assumed (F= 1). The apparent volume of distribution (V/F) was calculated as (CL/F)/k_el. Both CL/F and V/F were normalized to body weight. The pharmacokinetic parameters were calculated for each subject.

Statistical analysis

Normally distributed data are expressed as mean values (± SD), with 95% confidence intervals (CI) of mean differences for the pharmacokinetic parameters between the phases. Statistical comparisons among the three treatments for each pharmacokinetic parameter were assessed by repeated measures analysis of variance (RM-anova) with post hoc adjustment for multiple comparisons using the Bonferroni correction or RM-anova on ranks and Dunnett’s test post hoc (SigmaStat version 3.0; Jandel Scientific, San Rafael, CA, USA). When comparing gender and the three treatments (fasting, fed and grapefruit juice), geometric mean ratios (GMR) were calculated for C_max and AUC. The C_max and AUC values were log transformed before comparison by statistical analysis and the data are presented after reverse transformation. Data were accepted as significant using the 5% significance level. Based on an earlier Thai study [8], we expected a 45% difference in the AUC of primaquine between males and females. Assuming a SD of the difference of 30%, a power of 80% and a significance level of 0.05, we would need nine males and nine females to detect this major difference. Since primaquine is not a highly lipophilic drug (Lop P = 1.89 using the Crippen Fragmentation Method [19], personal communication R. K. Haynes), we expected a smaller increase in the AUC of primaquine of about 20% when the drug was given with food or grapefruit juice. To detect this difference, we needed 20 subjects in a paired study assuming a SD of the difference of 30%, a power of 80% and a significance level of 0.05.

Results

The single oral dose of 30 mg primaquine administered on three occasions (without food, with food or with grapefruit juice) was well tolerated in the 20 healthy Vietnamese subjects (10 males; 10 females), with no adverse events reported.

Effect of gender on the pharmacokinetics of primaquine and carboxyprimaquine

Mean plasma concentration–time curves of primaquine and carboxyprimaquine were similar between male and female subjects in the fasting state (Figure 1). Primaquine was rapidly absorbed in the fasting state, with 90% (18 of 20) of subjects having measurable primaquine concentrations at 0.5 h after drug administration. Table 1 shows the pharmacokinetic parameters of primaquine and carboxyprimaquine in males and females after fasting. The t_max, t_1/2 and MRT values were about 2 h, 8 h and 11 h, respectively, for both males and females. Although not statistically significant, the mean CL/F of primaquine was higher in males than females (0.52 l h⁻¹ kg⁻¹vs. 0.43 l h⁻¹ kg⁻¹). Similar to the fasting state, no statistical differences were detected in the pharmacokinetics of primaquine between male and female subjects during the fed and grapefruit juice phases. In the fed state, the geometric mean AUC_0–∞ of primaquine in males and females were 1233 ng ml⁻¹ h and 1582 ng ml⁻¹ h (GMR of 0.78, 95% CI 0.59, 1.04, P = 0.08), respectively. Corresponding values after grapefruit juice administration were 1349 ng ml⁻¹ h and 1564 ng ml⁻¹ h (GMR of 0.86, 95% CI 0.68, 1.09, P = 0.20). No significant differences were detected in the C_max, t_max and AUC_{0−24 h} values of carboxyprimaquine between males and females after the three treatment phases (data for fed and grapefruit phases not shown).

Figure 1.

Mean (± SD) plasma concentrations of primaquine (PQ) and carboxyprimaquine (CPQ) vs. time profiles in healthy Vietnamese subjects (10 males and 10 females) following a single oral dose of 30 mg primaquine. PQ-males (•), PQ-females (○), CPQ-males (▴), CPQ-females (▵)

Table 1.

Comparison of the pharmacokinetic parameters of primaquine and carboxyprimaquine after a single oral dose of 30 mg primaquine in male and female healthy Vietnamese subjects in the fasting state

Pharmacokinetic parameters	Males (n = 10)	Females (n = 10)	Difference between males and females (P)
Primaquine
Cmax (ng ml−1)b	120 (81–151)	135 (73–245)	0.89 (0.65, 1.22)c (P = 0.43)
tmax (h)e	2.0 (1.0–2.0)	1.8 (1.0–3.0)	(P = 0.97)
AUC0–∞ (ng ml−1 h)b	1 094 (454–1723)	1 364 (922–2266)	0.80 (0.56, 1.15)c (P = 0.21)
t1/2 (h)a	7.9 ± 2.5	8.0 ± 2.0	0.1 (− 2.2, 2.1, P = 0.97)d
MRT (h)a	11.1 ± 2.8	11.8 ± 2.6	0.7 (− 3.3, 1.8, P = 0.57)d
CL/F (l h−1 kg−1)a	0.52 ± 0.23	0.43 ± 0.16	0.09 (− 0.10, 0.28, P = 0.32)d
V/F (l kg−1)a	5.3 ± 1.3	4.6 ± 1.0	0.7 (− 0.3, 1.8, P = 0.16)d
Carboxyprimaquine
Cmax (ng ml−1)b	1 110 (873–1540)	1 181 (858–1893)	0.93 (0.76, 1.14)c (P = 0.47)
tmax (h)e	4.0 (3.0–10.0)	5.0 (2.0–12.0)	(P = 0.60)
AUC0−24 h (ng ml−1 h)b	21 391 (16 302–31 771)	22 054 (14 600–32 443)	0.97 (0.78, 1.20)c (P= 0.77)

^aMean values (± SD).

^bGeometric mean (range).

^cGeometric mean ratios (male:female) (95% CI).

^dMean difference between males and females (95% CI).

^et_max values are given as medians with ranges.

Effect of food and grapefruit juice on the pharmacokinetics of primaquine and carboxyprimaquine

Mean plasma concentration–time curves of primaquine and carboxyprimaquine are shown in Figure 2 for the three treatment phases. A comparison of the pharmacokinetics of primaquine and its metabolite in the fasting and fed states and after coadministration with grapefruit juice is presented in Table 2. With the exception of the t_max for primaquine, there were significant differences in the pharmacokinetics of primaquine between the three treatments. When compared with the fasting state, food significantly increased the geometric mean C_max and AUC_0–∞ of primaquine by 26% (127 ng ml⁻¹ to 160 ng ml⁻¹, 95% CI 12, 40, P< 0.001) and by 14% (1222 ng ml⁻¹ h to 1396 ng ml⁻¹ h, 95% CI 3, 27, P = 0.013), respectively. Similarly, grapefruit juice significantly increased the C_max of primaquine by 23% (127 ng ml⁻¹ to 156 ng ml⁻¹, 95% CI 4, 45, P = 0.02) and the AUC_0–∞ by 19% (1222 ng ml⁻¹ h to 1453 ng ml⁻¹ h, 95% CI 4, 37, P = 0.02) compared with fasting state values. In contrast to primaquine, neither food nor grapefruit juice significantly altered the C_max and AUC₀₋₂₄ of carboxyprimaquine.

Figure 2.

Mean (± SD) plasma concentrations of primaquine (PQ) and carboxyprimaquine (CPQ) vs. time profiles in 20 healthy Vietnamese subjects following a single oral dose of 30 mg primaquine administered either in the fasting and fed states or with grapefruit juice. PQ-fasting (▾), PQ-food (□), PQ-grapefruit juice (○), CPQ-fasting (▴), CPQ-food (◊) CPQ-grapefruit juice ()

Table 2.

Comparison of the pharmacokinetic parameters of primaquine and carboxyprimaquine after a single oral dose of 30 mg primaquine in healthy Vietnamese subjects in the fasting state, fed state and with the coadministration of grapefruit juice

Pharmacokinetic parameters	Fasting (control) (n = 20)	Fed (n = 20)	Grapefruit juice (n = 20)	Differences in mean and median values among 3 treatment groups (P)
Primaquine
Cmax (ng ml−1)b	127 (73–245)	160 (113–287)	156 (99–230)	0.019**
tmax (h)c	2.0 (1.0–3.0)	1.5 (0.5–2.0)	1.5 (1.0–3.0)	0.137**
AUC0–∞ (ng ml−1 h)b	1 222 (454–2266)	1 396 (806–2601)	1 453 (828–1969)	0.024**
t1/2 (h)a	8.0 ± 2.2	7.2 ± 1.5	7.7 ± 1.7	0.041*
MRT (h)a	11.5 ± 2.7	10.3 ± 2.2	11.2 ± 2.1	0.003*
CL/F (l h−1 kg−1)c	0.41 (0.25–1.02)	0.37 (0.22–0.82)	0.36 (0.26–0.79)	0.024**
V/F (l kg−1)c	5.0 (3.2–7.6)	4.0 (2.3–6.7)	3.8 (2.7–7.2)	<0.001**
Carboxyprimaquine
Cmax (ng ml−1)b	1 140 (858–1893)	1 061 (753–1471)	1 081 (762–1446)	0.157**
tmax (h)c	4.0 (2.0–12.0)	8.0 (4.0–21.0)	8.0 (1.5–12.0)	0.043**
AUC0−24 h (ng ml−1 h)b	21 720 (14 600–32 443)	19 820 (15 047–26 276)	21 156 (13 774–31 118)	0.212**

^aMean values (± SD).

^bGeometric mean (range).

^ct_max, CL/F and V/F values are given as medians with ranges.

^*RM-anova and Bonferroni t-test.

^**RM-anova on ranks and Dunnett’s test post hoc.

Discussion

This is the first study describing the effects of gender, food and grapefruit juice on the pharmacokinetics of primaquine and its major plasma metabolite, carboxyprimaquine. Although gender disparity in the pharmacokinetics of numerous drugs has been identified, the differences are generally only subtle [20]. Physiological and molecular differences such as drug transporters and drug-metabolizing enzymes between genders can cause sex-related differences in the pharmacokinetics of drugs. For example, females tend to have lower body weight, a greater percentage of body fat and higher CYP3A4 activity than males. Although the present study provided no significant evidence for sex-related differences in the pharmacokinetics of primaquine and carboxyprimaquine, the number of subjects was small, which may lead to a Type II error. Our finding, however, differed from that of Singhasivanon et al.[8], who reported healthy Thai females to have significantly higher C_max and AUC_0–∞ values of primaquine than males (252 ng ml⁻¹vs. 139 ng ml⁻¹ and 1900 ng ml⁻¹ h vs. 1307 ng ml⁻¹ h, respectively) following the first dose of a daily 15-mg regimen of primaquine. This discordance in results between the two studies is difficult to explain, as the C_max and AUC_0–∞ of primaquine were about twofold less in the Vietnamese subjects given twice the dose than that administered to the Thai subjects. The C_max and AUC_0–∞ values of primaquine in the Vietnamese subjects, particularly in the males, were also comparable to those of Mihaly et al.[7] in healthy caucasian males administered 30 mg primaquine.

It is generally recommended that primaquine should be taken with a meal to minimize any gastrointestinal disturbances associated with the use of this drug for the treatment or prophylaxis of malaria. Since food is known to stimulate splanchnic and hepatic blood flow, delay gastric emptying, induce the release of bile acids and can cause chemical or physical drug interactions [9], the bioavailability of primaquine may be altered by the intake of food. Our study showed a meal containing ∼ 28 g of fat significantly increased the bioavailability of primaquine by 26% and 14%, as measured by C_max and AUC, respectively, and that the presence of food did not delay the rate of primaquine absorption. These findings suggest that the bioavailability of primaquine can be maximized by administering the drug with a meal. However, the increase in primaquine concentrations with food was relatively modest and is unlikely to cause an increase in the incidence and severity of adverse events associated with the drug. Improved bioavailability of primaquine may be an additional reason for stressing that this drug should be taken with a meal, since higher primaquine concentrations may be beneficial in the treatment of increasingly drug-tolerant infections of vivax malaria [3].

The increased bioavailability of primaquine when coadministered with grapefruit juice indicates that gut CYP3A4 activity is involved in the drug’s first-pass metabolism. Grapefruit juice–primaquine interaction was quite variable, with some individuals having no increase in primaquine concentrations and others showing up to a twofold increase (e.g. the grapefruit juice to fasting AUC_0–∞ ratio ranged from 0.71 to 2.18). This marked variability is presumably associated with the large interindividual differences in the content of mucosal CYP enzyme activity in the small intestine among individuals. Although the extent of absorption of primaquine was increased by grapefruit juice as determined by elevated maximum primaquine concentrations, the rate of absorption was not altered, with similar t_max values between the fasting and grapefruit juice phases. Thus, grapefruit juice does not appear to accelerate gastric emptying or enhance dissolution of the primaquine tablet. The increase in oral bioavailability of primaquine by grapefruit juice would have led to more primaquine being available to be metabolized by the liver to carboxyprimaquine. This increase in parent drug would have compensated for the reduced carboxyprimaquine formation in the stomach due to gut metabolism inhibition by grapefruit juice. Such a sequence of events would explain the lack of change in carboxyprimaquine concentrations between the fasting and grapefruit juice phases, since grapefruit juice does not inhibit liver CYP3A4 metabolism [16].

The mechanism of grapefruit juice–drug interaction is complex and as yet not fully understood [21]. In vitro studies have identified the flavonoid, naringin and the furanocoumarins, bergamottin and 6’,7′-dihydroxybergamottin as probable active constituents in grapefruit juice [22, 23]. These constituents can vary considerably among types and brands of grapefruit juice [24]. However, when pure naringin was administered to humans a significant grapefruit juice interaction with the prototypical CYP3A4 substrate, felodipine, failed to materialize [25]. Recently, in human studies, bergamottin and 6′,7′-dihydroxybergamottin have been reported to be important inhibitors of CYP3A4 in both Seville orange and grapefruit juices [26]. Other possible mechanisms for grapefruit juice–drug interactions could include inhibition of intestinal P-glycoprotein-mediated efflux transport of drugs and organic anion-transporting polypeptides [21].

In conclusion, no apparent sex-related differences were found in the pharmacokinetics of primaquine and carboxyprimaquine after a single oral dose of primaquine to warrant a different dose regimen for males and females. Food increased primaquine concentrations, suggesting that primaquine should be taken with or immediately after a meal to maximize bioavailability and clinical efficacy. Although diluted grapefruit juice increased primaquine concentrations, the marked interindividual variability demonstrated that the quantitative effects of grapefruit juice on the pharmacokinetics of primaquine are not predictable at an individual level. Because of this unpredictability and the lack of information on the effects of full-strength grapefruit juice on the pharmacokinetics of primaquine, it would be prudent to advise people not to take primaquine with grapefruit juice or any grapefruit products.