Abstract
Aims
To investigate whether coadministration of the antimalarials artesunate and artemisinin alters the clearance of either drug.
Methods
Ten healthy Vietnamese males (Group AS) were randomized to receive a single dose of 100 mg oral artesunate (pro-drug of dihydroartemisinin) on day −5 and then once daily for 5 consecutive days (days 1–5). Oral artemisinin (500 mg) was coadministered on days 1 and 5. Another 10 subjects (Group AM) were given 500 mg oral artemisinin on day −5 and then further doses on days 1–5. Artesunate 100 mg was given on days 1 and 5. Artemisinin and dihydroartemisinin plasma concentrations on days −5, 1 and 5 were quantified by h.p.l.c. with on-line postcolumn derivatization and u.v. detection.
Results
In Group AS, dihydroartemisinin oral clearance values (mean (95% CI)) were similar on day 1 (32 (22, 47)) l h−1 and day 5 (38 (28, 51)) l h−1 of daily artesunate administration but these mean values were approximately three fold higher compared with day −5 after a single dose (95 (56, 159)). In this group, artemisinin oral clearance increased from 196 (165, 232) l h−1 on day 1–315 (241, 410) l h−1 on day 5. In Group AM, dihydroartemisinin oral clearance on day 1 was 39 (34, 46) l h−1 and increased 1.6 fold to 64 (48, 85) l h−1 on day 5. In this group, artemisinin oral clearance increased sequentially (1.5 and 4.7 fold, respectively) from 207 (151, 285) l h−1 on day −5–308 (257, 368) l h−1 on day 1 and to 981 (678, 1420) l h−1 on day 5. The increase in artemisinin oral clearance between days −5 and 1 (in the absence of artesunate) was similar to that between days 1 and 5 in Group AS subjects who took daily artesunate. Dihydroartemisinin was not a significant metabolite of artemisinin.
Conclusions
Artesunate (dihydroartemisinin) did not alter the elimination of artemisinin. However, dihydroartemisinin elimination was inhibited by artemisinin. Artemisinin induced its own elimination even 5 days after a single oral dose. There was no evidence for the formation of dihydroartemisinin from artemisinin.
Keywords: artemisinin, artesunate, dihydroartemisinin, induction, inhibition, pharmacokinetics
Introduction
Artemisinin and derivatives such as artesunate are rapidly acting, well-tolerated endoperoxide compounds widely used in uncomplicated and severe P. falciparum malaria in south-east Asian countries and other parts of the world [1–4]. A therapeutic drawback is the high recrudescence rates associated with monotherapy of less than 1 week's duration [5]. To improve cure rates, and possibly to prevent development of resistance, combination chemotherapy based on one of the artemisinin derivatives has been advocated [6, 7]. The parent compound of this class of antimalarials, artemisinin (Figure 1), exhibits remarkable time-dependent pharmacokinetics. Artemisinin oral clearance (CL/F) increased 5–7 fold in healthy subjects and patients during a 5–7 day course of 500 mg daily [8–12]. This change has been interpreted as an example of auto-induction of drug metabolism [8], a hypothesis supported by the finding that artemisinin is capable of inducing the in vivo metabolism of omeprazole to 5-OH-omeprazole, putatively by a CYP2C19-mediated pathway [13]. In contrast, the in vitro metabolism of artemisinin appears to be primarily mediated by CYP2B6 [14]. It is not known which enzymes are induced by artemisinin. The most widely used artemisinin derivative is artesunate (Figure 1), a hemisuccinate ester pro-drug, which is rapidly cleaved in vivo to the active metabolite dihydroartemisinin (Figure 1). After i.v. administration, artesunate is rapidly hydrolysed (t½ = 2–3 min) to dihydroartemisinin which has an elimination t½ of 30–40 min [15, 16]. After oral administration, artesunate concentrations are very low or undetectable but the relative bioavailability of dihydroartemisinin is 82–85% [15, 16].
Figure 1.
Chemical structure of artemisinin (a), artesunate (b) and its active metabolite dihydroartemisinin (c).
Auto-induction of metabolism may compromise the clinical efficacy of drugs like artemisinin, and contribute to recrudescence. Since artemisinin and its derivatives are being used increasingly in countries such as Vietnam, this would also have implications for the increasingly proposed multidrug treatment regimens under current investigation. Therefore, we aimed to determine whether coadministration of artesunate (dihydroartemisinin) increases the elimination of artemisinin thereby demonstrating a class effect with respect to enzyme induction. Secondary objectives were to determine whether dihydroartemisinin kinetics would be influenced by administration of artemisinin, and whether dihydroartemisinin was a clinically significant metabolite of artemisinin.
Methods
Subjects
Twenty healthy male Vietnamese subjects, aged 21–45 years (median 29) and with body weights ranging between 44 and 73 kg (median 53) were studied. Four were nonsmokers and 16 were smokers of less than 10 cigarettes per day. Subjects who had taken artemisinin or artesunate within 1 month before recruitment, those with a history of alcohol or drug abuse, and those taking any other regular medication were excluded. Written informed consent was obtained from all subjects for study procedures that were approved by the Ministry of Health, Hanoi, the Ethics Committee of the Medical Faculty of Uppsala University and the Medical Products Agency, Uppsala.
Study design
The study was conducted in the Clinical Unit at the Institute of Malariology, Parasitology and Entomology (IMPE), Hanoi, Vietnam, from June to July 1998. Subjects were randomized to receive artemisinin and artesunate according to one of two different dosing schedules (Table 1). In each case, randomization was achieved by blinded selection of a prenumbered envelope. The effect of repeated artesunate administration on artemisinin kinetics was studied in the test arm (Group AS). In the control arm (Group AM), the effect of repeated artemisinin administration on artesunate kinetics was assessed. This group served as a positive control for artemisinin induction well as providing a means to determine the effects of artemisinin on dihydroartemisinin kinetics.
Table 1.
Study design with repeated administration of artesunate (Group AS, n = 10) or with repeated administration of artemisinin (Group AM, n = 10).
| Day −14 | Day −5 | Day1 | Day2 | Day3 | Day4 | Day5 | |
|---|---|---|---|---|---|---|---|
| Physical examination | X | X | |||||
| Sampling for pharmacokinetics | X | X | X | ||||
| Group AS | |||||||
| AS (100 mg orally daily) | X | X | X | X | X | X | |
| AM (500 mg orally daily) | X | X | |||||
| Group AM | |||||||
| AS (100 mg orally daily) | X | X | |||||
| AM (500 mg orally daily) | X | X | X | X | X | X | |
Artesunate was administered orally in the morning as tablets (2 × 50 mg once daily, batch 061297, the Pharmaceutical Factory Nr 2, Hanoi, Vietnam). Artemisinin was given in the same manner as hard gelatine capsules (2 × 250 mg once daily, batch 060996, the Pharmaceutical Factory Nr 2, Hanoi, Vietnam). Assay of artesunate tablets and artemisinin capsules from the Pharmaceutical Factory Nr 2 yielded 95% confidence intervals for actual content that included the stated dose in each case. On pharmacokinetic study days, all subjects were fasted for 12 h before and 2 h after drug administration. During this period, only fluids (except tea, coffee, and cola soft drinks) permitted. Drugs were administered under supervision and with 150 ml of water.
Physical examination, and routine biochemical and haematological tests, were performed in each subject on days −14 and 5. Subjects were asked to report side-effects on days 1 and 5 and a checklist of prespecified symptoms and signs was also completed at these times.
Sampling
On days −5, 1 and 5, blood samples were collected in 5 ml fluoride-oxalate Vacutainers™ via an indwelling Venflon™ catheter at 5 min prior to, then at 20 and 40 min, and 1, 1.5, 2, 3, 4, 5, 6, 7, 8, 9, 10 h after drug administration, except when artesunate was given alone, in which case the last sample was at 6 h. Blood samples were centrifuged at ambient temperature immediately after collection (1500 g for 5 min). Plasma was transferred into two separately labelled Nunc™ cryo-tubes, and stored at −70 ° C until transported on dry ice to Uppsala (artemisinin measurement) or Perth (dihydroartemisinin measurement). Samples were stored at −80 ° C and assayed within 3 months. Dihydroartemisinin, the more labile of the analytes, is stable for at least 8 months [15].
Drug assay
Artemisinin concentrations in plasma were determined at the Division of Biopharmaceutics and Pharmacokinetics, Uppsala University, by h.p.l.c. with postcolumn derivatization and u.v. detection (289 nm) based on a modified method of Edlund et al. [17], as previously described [13]. The only further modification was adjustment of pH of the 0.1 mol l−1 acetate buffer used in mobile phase to 3.6. All samples from one subject were analysed on the same occasion together with eight standard curve samples and with the interspersion of six quality control samples (two at each of three concentrations) (Table 2). The limit of quantification was set at 35 nmol l−1 (intra-assay CV = 7%, n = 10). Dihydroartemisinin plasma concentrations were determined at the Department of Pharmacology, University of Western Australia, by h.p.l.c. based on the method of Edlund et al. [17] as modified by Batty et al. [18] and using artemisinin as internal standard. The limit of quantification for dihydroartemisinin was 316 nmol l−1 and the limit of detection 176 nmol l−1. Intra-assay CV was 2–9% (n = 3–5) and interassay CV 10–11% (n = 22–33). For plasma samples containing both artemisinin and artesunate (dihydroartemisinin), the assay was carried out without the use of an internal standard (intra-assay CV 9–13%, n = 5). The assay used enables both artesunate and dihydroartemisinin concentrations to be quantified specifically in the same sample [18]. Consistent with previous data [15], plasma concentrations of artesunate per se after oral administration were below the limit of quantification in most of the present samples and therefore only dihydroartemisinin concentrations are reported.
Table 2.
Assay interoccasion precision and accuracy for artemisinin and dihydroartemisinin based on the analyses of quality control (QC) samples interspersed with authentic samples on each experimental occasion.
| Precisiona (%) | Accuracyb (%) | |
|---|---|---|
| Artemisinin QCs (n > 38) | ||
| 293.6 nmol l−1 | 7.8 | 103 |
| (100, 106) | ||
| 2936 nmol l−1 | 5.2 | 99 |
| (97, 100) | ||
| 5872 nmol l−1 | 8.0 | 100 |
| (98, 103) | ||
| Dihydroartemisinin QCs | ||
| 911 nmol l−1 (n = 7) | 11.9 | 111 |
| (99–123) | ||
| 4234 nmol l−1 (n = 11) | 7.5 | 105 |
| (100–110) | ||
Intermediate precision expressed as the relative standard deviation (% coefficient of variation) of found concentrations.
Closeness of agreement between found and theoretical concentrations calculated as 100 ○ measured concentration/theoretical concentration as mean (95%).
Data analysis
Artemisinin and dihydroartemisinin pharmacokinetic parameters were calculated by noncompartmental methods from plasma concentration-time data as previously described [11]. Oral clearance (CL/F) was calculated by dose/AUC(0, ∞). Dihydroartemisinin oral clearance was estimated assuming quantitative conversion of artesunate to dihydroartemisinin.
Statistics
A sample size of 10 subjects in the test arm had 90% power to detect an intraindividual average two-fold increase in artemisinin oral clearance after artesunate administration assuming a relative standard deviation of 70% and at a significance level of 0.05. Another 10 subjects were included in the positive control arm.
Pharmacokinetic parameters for both artemisinin and dihydroartemisinin conformed to a logarithmic distribution. Log-transformed dihydroartemisinin and artemisinin parameters obtained on days −5, 1 and 5 were compared by anova with within subject repeated measurement with planned comparison (contrast) using the Statistica™ software (Statistica, StatSoft, Tulsa, USA). The Mann–Whitney U-test was used for comparisons between study arms. Correlation was investigated using linear regression and Spearman's rank correlation coefficient. The significance level was set at < 0.05 in all statistical analyses. Variables and their intraindividual changes with time are presented as geometric means (95% CI from log-transformed data).
Results
Repeated artesunate administration (Group AS)
During repeated administration of artesunate, dihydroartemisinin CL/F values were similar on days 1 and 5 when coadministrated with artemisinin, but both were significantly lower compared with when artesunate was given alone on day −5 (P < 0.001; Table 3, Figure 2a). Dihydroartemisinin t½ and AUC were significantly higher on days 1 and 5 compared with day −5 (P < 0.001; Table 3). Due to insufficient concentration-time data in the elimination phase (three subjects), dihydroartemisinin pharmacokinetic parameters on day −5 (except Cmax) were based on the results from seven individuals. All other data sets were complete. Artemisinin oral clearance increased on average by 1.6-fold (1.3, 2.1) from day 1 to day 5 with an increase of > 20% seen in all but one subject (P < 0.01; Table 4, Figure 2b).
Table 3.
Dihydroartemisinin pharmacokinetic parameters determined on days −5, 1 and 5 in Group AS (n = 10, artesunate was given on day −5 and from day 1 through day 5, artemisinin was given on days 1 and 5) and in Group AM (n = 10, artemisinin was given on day −5 and from day 1 through day 5, artesunate was given on days 1 and 5). Values of parameters are shown as geometric mean (95% CI).
| Dihydroartemisinin parameters | Day −5 | Day 1 | Day 5 |
|---|---|---|---|
| Group AS | |||
| AUC(0, t) (nmol l−1 h) | 2289***† | 6776 | 6065 |
| (1567, 3343) | (4664, 9844) | (4412, 8339) | |
| AUC(0, ∞) (nmol l−1 h) | 2765***† | 8121 | 6910 |
| (1637, 4670) | (5534, 11917) | (5098, 9365) | |
| CL/F (l h−1) | 94***† | 32 | 38 |
| (56, 159) | (22, 47) | (28, 51) | |
| t½ (h) | 0.55***† | 1.63 | 1.33 |
| (0.44, 0.70) | (1.34, 1.99) | (1.12, 1.57) | |
| Cmax (nmol l−1) | 1664** | 2821 | 2620 |
| (999, 2772) | (1968, 4043) | (1793, 3827) | |
| Group AM | |||
| AUC(0, t) (nmol l−1 h) | 5061 | 3120** | |
| (3465, 7391) | (2001, 4864) | ||
| AUC(0, ∞) (nmol l−1 h) | 5907 | 3579** | |
| (4226, 8255) | (2370, 5404) | ||
| CL/F (l h−1) | 39 | 64** | |
| (34, 46) | (48, 85) | ||
| t½ (h) | 1.42 | 1.02*** | |
| (1.26, 1.60) | (0.85, 1.24) | ||
| Cmax (nmol l−1) | 2505 | 1970* | |
| (1973, 3179) | (1526, 2543) | ||
n = 7
P < 0.05
P < 0.01
P < 0.001; (compared with the values on day 1).
Figure 2.
Individual oral clearance (CL/F) values for dihydroartemisinin determined on days −5, 1 and 5 (a, n = 10 except on day −5 when n = 7) and for artemisinin on days 1 and 5 (b, n = 10) when 100 mg artesunate was given orally once daily on day −5 and from day 1 through day 5 and with oral administration of 500 mg artemisinin on days 1 and 5 in healthy, male Vietnamese subjects (Group AS). Values above horizontal lines represent the geometric average and its 95% confidence interval of the intraindividual ratios of oral clearance values between study days (CL/Fday 1/CL/Fday −5 and CL/Fday 5/CL/Fday 1), and vertical arrows depict occasions of drug administration.
Table 4.
Artemisinin pharmacokinetic parameters determined on days −5, 1 and 5 in Group AS (n = 10, artesunate was given on day −5 and from day 1 through day 5, artemisinin was given on days 1 and 5) and in Group AM (n = 10, artemisinin was given on day −5 and from day 1 through day 5, artesunate was given on days 1 and 5). Values of parameters are shown as geometric mean (95% CI).
| Artemisinin parameters | Day −5 | Day 1 | Day 5 |
|---|---|---|---|
| Group AS | |||
| AUC(0, t) (nmol l−1 h) | 8339 | 5219** | |
| (7153, 9720) | (4028, 6761) | ||
| AUC(0, ∞) (nmol l−1 h) | 9056 | 5638** | |
| (7652, 10719) | (4320, 7358) | ||
| CL/F (l h−1) | 196 | 315** | |
| (165, 232) | (241, 410) | ||
| t½ (h) | 2.48 | 2.50 | |
| (2.05, 3.01) | (1.99, 3.13) | ||
| Cmax (nmol l−1) | 2794 | 2047* | |
| (2353, 3317) | (1546, 2710) | ||
| Group AM | |||
| AUC(0, t) (nmol l−1 h) | 7733*** | 5081 | 1572*** |
| (5674, 10539) | (4320, 5975) | (1051, 2350) | |
| AUC(0, ∞) (nmol l−1 h) | 8555** | 5763 | 1807*** |
| (6212, 11781) | (4813, 6901) | (1249, 2614) | |
| CL/F (l h−1) | 207** | 308 | 981*** |
| (151, 285) | (257, 368) | (678, 1420) | |
| t½ (h) | 2.21 | 2.97 | 2.53 |
| (1.72, 2.84) | (2.27, 3.88) | (1.99, 3.23) | |
| Cmax (nmol l−1) | 2408** | 1803 | 629*** |
| (1824, 3179) | (1413, 2299) | (414, 956) | |
P < 0.05
P < 0.01
P < 0.001; (compared with the values on day 1).
Repeated artemisinin administration (Group AM)
Dihydroartemisinin CL/F increased from 39 (34, 46) l h−1 on day 1–64 (48, 85) l h−1 on day 5 (P < 0.01; Table 3, Figure 3a). Dihydroartemisinin t½ on day 5 was significantly lower by a factor of 0.72 (0.64, 0.82) compared with that on day 1 (P < 0.001). When artemisinin was given alone on day −5, a chromatographic peak with a retention time corresponding to α-dihydroartemisinin and exceeding the limit of detection (176 nmol l−1) was observed in only one out of 80 plasma samples assayed from the 10 individuals.
Figure 3.
Individual oral clearance (CL/F) values for dihydroartemisinin on days 1 and 5 (a, n = 10) and for artemisinin on days −5, 1 and 5 (b, n = 10) when 500 mg artemisinin was given orally once daily on day −5 and from day 1 through day 5 with single oral administrations of 100 mg artesunate on days 1 and 5 in healthy male Vietnamese subjects (Group AM). Values above horizontal lines represent the geometric average and its 95% confidence interval of the interindividual ratios of oral clearance values between study days (CL/Fday 1/CL/Fday −5 and CL/Fday 5/CL/Fday 1), and vertical arrows depict occasions of drug administration.
Artemisinin oral clearance (CL/F) increased 1.5-fold (1.2, 1.8) from day −5 to day 1 (P < 0.01; Table 4), with an increase exceeding 20% in 8 out of 10 subjects. Artemisinin oral clearance exhibited a further 3.2-fold (2.2, 4.6) increase from day 1 to day 5 (Figure 3b). All artemisinin pharmacokinetic parameters on days −5 and 1 were similar to those on days 1 and 5 in the study arm with repeated artesunate administration (P > 0.19 vs Group AS; Table 4).
Artemisinin t½ did not change significantly between different study days or between the two treatment arms. Pharmacokinetic parameters and their time-dependencies for artemisinin and dihydroartemisinin were similar for smokers and nonsmokers (data not shown). This is in agreement with a previous study of artemisinin kinetics (M. Ashton, unpublished data). The extrapolated relative to the total AUC averaged (s.d.) 10% (6%) for artemisinin and 13% (7%) for dihydroartemisinin.
Correlations
There was a weak but significant correlation between the individual dihydroartemisinin and artemisinin AUC(0, t) values on days 1 and 5 (rs = 0.59, P < 0.001, n = 40). A similar correlation was also found between t½ of dihydroartemisinin and AUC(0, t) of artemisinin determined on the same days (rs = 0.60, P < 0.001, n = 40). Figure 4 shows the mean (± 95% CI) dihydroartemisinin AUC(0, t) values plotted against artemisinin AUCs for each group and for each occasion on which artesunate was given with or without artemisinin. The AUCs for each compound appeared linearly and positively related over a wide range of AUCs in each case.
Figure 4.
Relationship between dihydroartemisinin AUC(0, t) and artemisinin AUC(0, t) (geometric means and 95% confidence intervals) in 10 healthy Vietnamese subjects (Group AS) who received the combination on days 1 (a) and 5 (b)) of a 5 day course of daily 100 mg artesunate preceded by a single artesunate administration on day −5 and in another 10 subjects (Group AM) who received the combination on days 1 (c) and 5 (d)) of a 5 day course of daily 500 mg artemisinin preceded by a single artemisinin dose on day −5. The fitted regression line for the four mean values is shown.
Clinical assessments
No adverse events were reported, and all laboratory values were within normal reference ranges on days −14 and 5.
Discussion
With increasing interest in combination chemotherapy for malaria, the potential of metabolic interactions between artemisinin, its derivatives and other antimalarial drugs requires investigation. The known ability of artemisinin to induce drug metabolism, if restricted to artemisinin, could make it a less suitable drug for combination treatment. We aimed to determine whether the (auto-)induction seen with artemisinin was also a feature of the commonly used, semisynthetic derivative artesunate. Since it is not known which enzyme(s) are induced by artemisinin, we coadministered artemisinin with artesunate to assess time-related changes in drug clearance which might reflect altered activity of the enzymes induced by artemisinin. If induction was a class effect, it is proposed that artesunate (or its active metabolite dihydroartemisinin) would induce the same enzyme(s) as artemisinin and its coadministration would thereby affect the pharmacokinetics of the latter.
There was a significant, 1.6-fold increase in artemisinin oral clearance after daily administration of artesunate for five days in Group AS. However, this increase cannot be attributed to induction by artesunate (dihydroartemisinin). It was quantitatively similar to the increase in artemisinin oral clearance in Group AM (1.5-fold) after single artemisinin doses given five days apart with 100 mg artesunate administered only on the second occasion. These changes in artemisinin clearance in Group AM were unexpected. Even though an earlier study had suggested that changes in artemisinin kinetics after single oral dose were persistent [19], more recent results had demonstrated normalization of drug metabolizing capacity within 1 week of a 7 day daily regimen [13]. We therefore expected the 5 day washout period between the single dose administration (day −5) and the first combined administration (day 1) to be long enough for the effects of auto-induction to have abated. However, despite the relatively short t½ of artemisinin, its auto-induction persisted for at least 5 days in Group AM subjects.
The 3.2-fold increase in artemisinin oral clearance during 5 days of repeated administration in Group AM, was lower than in previous studies [9, 11, 13]. This apparently diminished capacity for auto-induction could be explained by the fact that a 1.5-fold increase had been induced by first dose given on day −5. The total increase in artemisinin clearance during the 10 days of the study in Group AM was 4.7-fold. Although artemisinin t½ did not change in parallel with oral clearance, this can be explained by its high hepatic extraction ratio [8].
The changes in artemisinin pharmacokinetic parameters between the first single dose of 500 mg artemisinin and the next dose given 5 days later were similar in the two study arms. This suggested that coadministration of artesunate, when given either as a single dose or daily for a duration of five days, did not to appreciably alter the elimination of artemisinin. Therefore, our results do not support the hypothesis that the ability of the artemisinin derivatives to induce their own metabolism is a class effect. By contrast, three reports published since the completion of our study all suggest that (auto-)induction might be common to the endoperoxide antimalarials. In a population pharmacokinetic study in patients with uncomplicated malaria, Ezzet et al. reported that ratio of artemether: dihydroartemisinin decreases sequentially after each of four oral artemether doses [20]. In a conventional multiple-dose pharmacokinetic study, an identical observation was recently reported for the same drug [21]. Khanh et al. found relatively low plasma dihydroartemisinin concentrations on the 5th day of an oral artesunate regimen in six patients [22]. In a recent study in Vietnamese patients with uncomplicated falciparum malaria, a time-dependent increase in dihydroartemisinin clearance was also noted (Ilett K.F., unpublished data). However, since malaria infection in rats significantly decreases hepatic clearance of dihydroartemisinin [23], it is possible that the increased oral clearance of this drug in patients during recovery from malaria may be due to restoration of hepatic enzyme activity compromised in the acute phase of the infection.
Some early reports suggested that dihydroartemisinin was an active metabolite not only of derivatives such as artesunate and artemether, but also of artemisinin itself [4, 7, 24–27]. The present study shows that, if any dihydroartemisinin is formed from artemisinin, it has to be a minor metabolite of little clinical consequence. We estimate the fraction of artemisinin metabolized to dihydroartemisinin to be a maximum of 7%, assuming (i) complete conversion of artesunate to dihydroartemisinin (ii) that maximum concentrations of the latter were equal to the limit of detection (176 nmol l−1) and (iii) according to the linear relationship (AUC(0, ∞) = 0.79 × Cmax+1130, r2 = 0.92, n = 7) between AUC(0, ∞) and Cmax of dihydroartemisinin when artesunate was given alone. Our present findings should lay to rest some of the confusion which has surrounded this issue.
Artemisinin clearly inhibited the elimination of dihydroartemisinin with two to three-fold increases in both dihydroartemisinin AUC and t½ values when the drugs were coadministered. The dependency of dihydroartemisinin elimination on plasma artemisinin concentrations was evident from the strong linear relationship between dihydroartemisinin and artemisinin AUCs seen in Figure 4. This suggests a competitive drug interaction. The mechanism for such inhibition of dihydroartemisinin metabolism is unknown. Since the major metabolite of dihydroartemisinin in humans is its 12-glucuronide [28], we suggest that artemisinin is an inhibitor of glycuronosyl transferase.
In summary, we found that, in contrast to recent data from malaria patients [22], there was no increased clearance of dihydroartemisinin with repeated doses of artesunate. However, our results are complicated by the simultaneous presence of an inhibition of dihydroartemisinin metabolism by artemisinin. In addition, data from malaria patients could reflect infection-specific changes in drug disposition. We also found no evidence that dihydroartemisinin alters the metabolism of artemisinin. Therefore, artesunate, or indeed dihydroartemisinin, may be the drug of choice when an artemisinin derivative is required for combination antimalarial regimens. However, recent literature suggests the inducing capacity of the different derivatives requires further elucidation. Since coadministration of artemisinin inhibited the elimination of dihydroartemisinin, we also suggest that a fixed dose combination of artemisinin and artesunate could be used to prolong the clinical efficacy of artesunate. Artemisinin was capable of inducing its own elimination even 5 days after a single oral dose but dihydroartemisinin is not a metabolite of any clinical importance.
Acknowledgments
The skilful help of Nguyen Thi Nieu is gratefully acknowledged as is the support from Le Dinh Công. We also acknowledge the excellent technical assistance of Shane Powell with the artesunate and dihydroartemisinin assays. This study was supported by the Swedish Agency for International Development and by a grant (TMED and KFI) from the National Health and Medical Research Coucil of Australia. Shu-Qiu Zhang was the recipient of a scholarship from Shanxi Province Government, People's Republic of China.
References
- 1.Arnold K, Hien TT, Chinh NT, Phu NH, Mai PP. A randomised comparative study of artemisinin (qinghaosu) suppositories and oral quinine in acute falciparum malaria. Trans Roy Soc Trop Med Hyg. 1990;84:499–502. doi: 10.1016/0035-9203(90)90012-4. [DOI] [PubMed] [Google Scholar]
- 2.Cao XT, Bethell DB, Pham TP, et al. Comparison of artemisinin suppositories, intramuscular artesunate and intravenous quinine for the treatment of severe childhood malaria. Trans Roy Soc Trop Med Hyg. 1997;91:335–342. doi: 10.1016/s0035-9203(97)90099-7. [DOI] [PubMed] [Google Scholar]
- 3.Vinh H, Huong NN, Ha TTB, et al. Severe and complicated malaria treated with artemisinin, artesunate or artemether in Viet Nam. Trans Roy Soc Trop Med Hyg. 1997;91:465–467. doi: 10.1016/s0035-9203(97)90287-x. [DOI] [PubMed] [Google Scholar]
- 4.Hien TT, White NJ. Qinghaosu. Lancet. 1993;341:603–608. doi: 10.1016/0140-6736(93)90362-k. [DOI] [PubMed] [Google Scholar]
- 5.Sy DN, Hoan DB, Dung NP, et al. Treatment of malaria in Vietnam with oral artemisinin. Am J Trop Med Hyg. 1993;48:398–402. [PubMed] [Google Scholar]
- 6.White NJ. Preventing antimalarial drug resistance through combinations. Drug Resistance Updates. 1998;1:3–9. doi: 10.1016/s1368-7646(98)80208-2. [DOI] [PubMed] [Google Scholar]
- 7.Peters W. The prevention of antimalarial drug resistance. Pharmac Ther. 1990;47:499–508. doi: 10.1016/0163-7258(90)90067-c. [DOI] [PubMed] [Google Scholar]
- 8.Ashton M, Sy ND, Gordi T, et al. Evidence for time-dependent artemisinin kinetics in adults with uncomplicated malaria. Pharm Pharmacol Lett. 1996;6:127–130. [Google Scholar]
- 9.Hassan Alin M, Ashton M, Kihamia CM, Mtey GJB, Björkman A. Multiple dose pharmacokinetics of oral artemisinin and comparison of its efficacy with that of oral artesunate in falciparum malaria patients. Trans Roy Soc Trop Med Hyg. 1996;90:61–65. doi: 10.1016/s0035-9203(96)90480-0. [DOI] [PubMed] [Google Scholar]
- 10.Ashton M, Hai TN, Sy ND, Huong NV, Nieu NT, Cong LD. Artemisinin pharmacokinetics is time-dependent during repeated oral administration in healthy male adults. Drug Metab Dispos. 1998;26:25–27. [PubMed] [Google Scholar]
- 11.Ashton M, Sy ND, Huong NV, et al. Artemisinin kinetics and dynamics during oral and rectal treatment of uncomplicated malaria. Clin Pharmacol Ther. 1998;63:482–493. doi: 10.1016/S0009-9236(98)90044-3. [DOI] [PubMed] [Google Scholar]
- 12.Sidhu JS, Ashton M, Huong NV, et al. Artemisinin population pharmacokinetics in children and adults with uncomplicated falciparum malaria. Br J Clin Pharmacol. 1998;45:347–354. doi: 10.1046/j.1365-2125.1998.t01-1-00686.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Svensson USH, Ashton M, Hai TN, et al. Artemisinin induces omeprazole metabolism in human beings. Clin Pharmacol Ther. 1998;64:160–167. doi: 10.1016/S0009-9236(98)90149-7. [DOI] [PubMed] [Google Scholar]
- 14.Svensson USH, Ashton M. Identification of the human cytochrome P450 enzymes involved in the in vitro metabolism of artemisinin. Br J Clin Pharmacol. submitted. [DOI] [PMC free article] [PubMed]
- 15.Batty KT, Thu LTA, Davis TME, et al. A pharmacokinetic and pharmacodynamic study of intravenous vs oral artesunate in uncomplicated falciparum malaria. Br J Clin Pharmacol. 1998;45:123–129. doi: 10.1046/j.1365-2125.1998.00655.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Batty KT, Thu LTA, Ilett KF, et al. A pharmacokinetic and pharmacodynamic study of artesunate for vivax malaria. Am J Trop Med Hyg. 1998;59:823–827. doi: 10.4269/ajtmh.1998.59.823. [DOI] [PubMed] [Google Scholar]
- 17.Edlund PO, Westerlund D, Carlqvist J, Bo-Liang W, Yun-Hua J. Determination of artesunate and dihydroartemisinin in plasma by liquid chromatography with post-column derivatization and UV-detection. Acta Pharm Suec. 1984;21:223–234. [PubMed] [Google Scholar]
- 18.Batty KT, Davis TME, Thu TA, Binh TQ, Anh TK, Ilett KF. Selective high-performance liquid chromatography determination of artesunate and α-, β-dihydroartemisinin in patients with falciparum malaria. J Chromatogr B. 1996;677:345–350. doi: 10.1016/0378-4347(95)00428-9. [DOI] [PubMed] [Google Scholar]
- 19.Ashton M, Gordi T, Hai TN, et al. Artemisinin pharmacokinetics in healthy adults after 250, 500 and 1000 mg single oral doses. Biopharm Drug Dispos. 1998;19:245–250. doi: 10.1002/(sici)1099-081x(199805)19:4<245::aid-bdd99>3.0.co;2-z. 10.1002/(sici)1099-081x(199805)19:4<245::aid-bdd99>3.0.co;2-z. [DOI] [PubMed] [Google Scholar]
- 20.Ezzet F, Mull R, Karbwang J. Population pharmacokinetics and therapeutic response of CGP 56697 (artemether+benflumetol) in malaria patients. Br J Clin Pharmacol. 1998;46:553–561. doi: 10.1046/j.1365-2125.1998.00830.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.van Agtmael MA, Cheng-Qi S, Qing JX, et al. Multiple dose pharmacokinetics of artemether in Chinese patients with uncomplicated falciparum malaria. Int J Antimicrob Agents. 1999;12:151–158. doi: 10.1016/s0924-8579(99)00063-1. [DOI] [PubMed] [Google Scholar]
- 22.Khanh NX, de Vries PJ, Ha LD, et al. Declining concentrations of dihydroartemisinin in plasma during 5-day oral treatment with artesunate for falciparum malaria. Antimicrob Agents Chemother. 1999;43:690–692. doi: 10.1128/aac.43.3.690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Batty KT, Ilett KF, Edwards G, et al. Assessment of the effect of malaria infection on hepatic clearance of dihydroartemisinin using rat liver perfusion and microsomes. Br J Pharmacol. 1998;125:159–167. doi: 10.1038/sj.bjp.0702023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.White NJ. Antimalarial pharmacokinetics and treatment regimens. Br J Clin Pharmacol. 1992;34:1–10. doi: 10.1111/j.1365-2125.1992.tb04100.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Van Hensbroek MB, Onyiorah E, Jaffar S, et al. A trial of artemether or quinine in children with cerebral malaria. N Engl J Med. 1996;335:69–75. doi: 10.1056/NEJM199607113350201. [DOI] [PubMed] [Google Scholar]
- 26.White NJ. Artemisinin: current status. Trans Roy Soc Trop Med Hyg. 1994;88(Suppl 1):3–4. doi: 10.1016/0035-9203(94)90459-6. [DOI] [PubMed] [Google Scholar]
- 27.Nosten F, Luxemburger C, ter Kuile FO, et al. Treatment of multidrug-resistant Plasmodium falciparum malaria with 3-day artesunate-mefloquine combination. J Infect Dis. 1994;170:971–977. doi: 10.1093/infdis/170.4.971. [DOI] [PubMed] [Google Scholar]
- 28.Ilett KF, Davis TME, Batty KT, et al. Vol. 13. Australia: Proceedings 5th International ISSX Meeting Cairns; October 25–29 1998. Glucuronidation of dihydroartemisinin following administration of artesunate to humans and by human liver microsomes; p. 105. [Google Scholar]