| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
To establish the efficacy of the single dose artesunate-mefloquine regimen for uncomplicated P. falciparum malaria in Vietnam, and to study the tolerance, pharmacokinetics, and pharmacodynamics of different timing of the mefloquine dose, a controlled, randomized, double-blind study was performed in which a single dose of 4 mg/kg of artesunate was followed by 15 mg/kg of mefloquine, given at various intervals.
|
METHODS |
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
6 years old with a symptomatic P. falciparum (asexual blood stages with parasite counts between 1,000 and 200,000/µL) infection were enrolled at five primary health care posts and one district hospital in Binh Thuan Province in southern Vietnam. Giemsa-stained thick and thin blood smears were obtained for identification of asexual parasites by light microscopy prior to patient inclusion. Exclusion criteria included a P. vivax infection, pregnancy, lactation, any sign of severe or complicated malaria, concurrent chronic disease, inability to take oral medication, known allergy to the study drugs, and intake of antimalarial agents in the preceding period as follows: artemisinin or derivatives (e.g., artesunate) in the previous 24 hours; mefloquine, tetracycline, or doxycycline in the previous seven days, or quinine in the previous 12 hours. Patients could be entered into the study for a second time if there were at least three months between the first and second date of entry.
The sample size was sufficient to detect a difference between approximately 30% and 15% response rates with statistical significance at 
= 0.05 and ß = 0.2. Artesunate (Plasmotrim® Lactabs), 50 mg and mefloquine (Mephaquine®) tablets, 250 mg, and mefloquine placebo tablets were obtained from Mepha AG (Aesch, Switzerland).
All patients were admitted into the study. After informed consent was obtained, the treatment regimen was allocated by drawing an envelope that included a computer-generated randomized dosage regimen, with the mefloquine and placebo tablets and a timetable for drawing blood samples, just before the first drug was administered (t0). All patients received an initial dose of 4 mg/kg of artesunate. Mefloquine was administered simultaneously, (regimen A), 8 hours later (regimen B), or 24 hours later (regimen C). By breaking tablets, the dose nearest to 15 mg/kg of body weight was chosen. Mefloquine placebo was given to complete three dosages for all patients. Medication was taken under supervision. Vomiting within one hour after intake of trial medication was a reason for withdrawal and installation of another treatment. These patients were evaluated for tolerance and for efficacy on an intention to treat basis.
Written informed consent was obtained from all patients or their parents. The study protocol was reviewed and approved by the research committee of Cho Ray Hospital, Ho Chi Minh City, and the Medical Ethics Committee of the Academic Medical Center, Amsterdam.
Follow-up of patients. The parasite count was determined every eight hours until three negative smears were obtained. Blood smears were repeated on 7, 14, 21, and 28 days after t0. Parasites were counted against 200 white blood cells in the thick smear, and, by recalculating to the patient’s total leukocyte count, expressed as the number of parasites per microliter of blood. A full blood count was determined at t0 and t = 48 hours. All blood smears were retained and reviewed by an experienced technician at the Department of Parasitology of Cho Ray Hospital. His results were taken as gold standard in case of conflicting results.
Blood was collected four times for measuring plasma concentrations of mefloquine. To ensure variable sampling points for a population pharmacokinetic analysis, the first 48 hours were divided in intervals of four hours from which the first two sampling points were chosen at random, starting at t = 4 for regimen A, t = 12 for regimen B, and t = 28 for regimen C. Additional samples were collected at t = 168 hours (7 days, Cd7) and day 28 (Cd28). The selection procedure of sampling times was unknown to researchers and clinicians to avoid revealing the treatment regimen. Blood was drawn into heparinized tubes, centrifuged immediately, and plasma was stored at –30°C until analysis. Mefloquine and its inactive carboxy metabolite were measured using high-performance liquid chromatography as described previously.9 Mefloquine recovery in this assay was > 95% with a detection level of 50 µg/L or lower. Patients who had a recurrence of parasites received additional medication. They returned on day 28 for determination of the plasma mefloquine concentration Clinical and parasitologic outcomes were assessed separately.10,11 Cure was defined as a disappearance of symptoms and of parasites; deterioration of the patient’s condition within the first 24 hours after initiation of therapy was classified as early clinical failure, and later as treatment failure, irrespective of the parasitologic outcome. Early clinical failure was not attributed to failure of study drugs. Parasitologic response was defined as follows: parasite recurrence: parasite clearance within 7 days but reappearance within the first 14 days after t0 (early recurrence) or from day 14 until 28 (late recurrence); R-II: initial reduction of parasite count (> 75%) without clearance within the first 7 days; and R-III: no or less response. Molecular techniques to distinguish re-infection from recrudescence were not available. Fever and parasite clearance times (FCT and PCT) were defined as the time from t0 to the first of three consecutive normal temperature readings (< 37.0°C axillary) or negative blood smears, respectively.
Early clinical failure was treated with intravenous artesunate, parasite recurrence with proguanil plus atovaquone (Malarone®; GlaxoWellcome, Stevenage, Hertfordshire, United Kingdom). To avoid interference with the pharmacokinetics, no mefloquine was prescribed to treat recrusdescence.
Data analysis. The time course of the parasite count was fitted to a mono-exponential elimination model. Log-linear mixed effects model were applied using maximum likelihood techniques in S-plus version 4.5 (Math Soft, Inc., Seattle, WA) as previously reported.5 The model estimates of the initial parasite count, P(0)est and the parasite elimination rate constant, k, were generated for every patient using restricted maximum likelihood method, and from this the time to clear 50% (PC50, parasite elimination half-life) and 90% (PC90) of the parasites was calculated. The interaction with the variables regimen, age, sex, artesunate, mefloquine dose, and pharmacokinetic parameters were entered as covariates. The mefloquine pharmacokinetics were analyzed in a similar way with non-linear mixed effects models to estimate the pharmacokinetic parameters for every individual.
Further analysis of the patients’ data was carried out using SPSS version 11.0 (SPSS, Inc., Chicago, IL). Intention to treat analysis included survival analysis and calculation of extreme case scenarios in which patients who were lost to follow were reallocated to either complete cure or to parasite recurrence.12 Contingency tables and chi-square tests with continuity correction were applied to categorical variables. Numerical variables were tested for normality before it was decided whether to use analysis of variance (ANOVA), or the Kruskal-Wallis test to compare groups. Parasite clearance and recrudescence were analyzed with survival analysis (Kaplan Meier plots and the Cox proportional hazard model). Correlation was expressed by Pearson’s correlation coefficient r. Statistical significance was accepted if P < 0.05.
A blinded interim analysis of outcome was performed midway during the study by a statistician not involved in the study. Differences between the three treatment groups were not found at that time.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The patient flow sheet (Figure 1
) shows which patients were used for analysis of proportions and which patients were only eligible for the extreme case scenarios and survival analysis. Twenty-three patients were excluded from the analysis because upon review of the blood slides the parasite count at baseline was less than 1,000/µL. The characteristics of the patients, included in the analysis, are shown in Table 1. One patient in group C deteriorated within eight hours after the start of therapy and was classified as an early clinical failure. Parasites were still detectable on day 7 in three patients. These patients had recovered completely and were classified as clinical cure with a parasitologic R-II response. Three patients left before any endpoint was reached and two patients left after parasite clearance but did not return on day 7 so that an R-II response could not be excluded. There was no significant difference among the three groups with respect to initial treatment outcome. Of the remaining 328 patients who were initially cured and completed parasite clearance before day 7, 11 cases were lost during follow-up at some stage. Their results were entered into extreme case scenarios. The radical cure rates were in a best case scenario 75% in group A, 75% in group B, and 69% in group C (P = 0.49, by chi-square test). In a worst case scenario, the ratios were 72%, 73%, and 63% (P = 0.22, by chi-square test), respectively. The rate of recrudescence was not associated with the patients’ date of enrollment. The proportional cumulative parasite clearance and recrudescence curves are shown in Figures 2A and B. The mean (95% confidence interval [CI]) PCT and FCT were 40.6 (38.5–42.8) hours and 23.7 (22.4–25.0) hours, respectively.
|
|
|
. Patients with an early recrudescence had a significantly slower parasite clearance and higher P(0)est than patients with radical cure or late recrudescence, and this was confirmed in a Cox survival model of parasite recurrence (P < 0.001 for P(0) and P < 0.001). In terms of relative risks, this means that for every hour prolongation of the PC50, the risk of parasite recurrence increases 1.36 times (95% CI = 1.06–1.21), and for every 10-fold increase of P(0), the risk of recurrence increases 1.7 fold (95% CI = 1.3–2.1).
|
 |
After exclusion of plasma samples with illegible identification codes or leakage, plasma samples of 250 patients were available for measurement of mefloquine concentrations, of which 179 had a complete series of four measurements available for population pharmacokinetics.
A two-compartment elimination model with first-order absorption yielded the best fit in the formula in which C(t) is the plasma concentration (µg/L), t is the time after intake of the mefloquine dose (h), k1 and k2 are the distribution and elimination rate constants (h–1) with their respective intercepts C01 and C02 (µg/L), and ka is the absorption rate constant (h–1). With four data points per patient, a maximum of four kinetic parameters can be estimated. Therefore, the terminal elimination k2 and its intercept C02 were fitted first and their mean value was entered as constants in the population model. For k2, the value was 0.00217 h–1 corresponding to an elimination half life of 319 hours (13 days). C02 was 1,506 µg/L. The maximum plasma concentration, Cmax, was derived numerically. The area under the concentration time curve, from t0 until infinity (AUC0–
) was calculated as
 |
The pharmacokinetic data are specified per treatment regimen in Table 1
. Figure 3 shows that the fitted individual concentration time curves are close to the observed values, but tend to underestimate Cmax in the lower ranges and Cd28. The plasma concentrations on day 7, observed as well as estimated by the pharmacokinetic model, were lowest in group A and highest in group C due to the difference in mefloquine dosing time, but this did not reach statistical significance. The observed mefloquine plasma concentrations on day 28 showed a similar trend with the highest values in group C and the lowest in group A (Table 1).
|
 |
Table 2 shows that the mefloquine concentration was significantly different among the sub-groups with cure and early and late parasite recurrence, especially Cd28. Cmax and AUC0– were lower in patients with an early recrudescence, but this did not reach statistical significance (ANOVA and Cox regression model of parasite survival). This also applies to the observed and estimated mefloquine plasma concentration on day 7. The terminal elimination half-life was shorter and Cd28 was lower in patients with early parasite recurrence.
The minimum in vivo parasitocidal plasma concentration of a fully sensitive parasite strain, 500 µg/L, is also shown in Figure 3. It shows that for all patients, the plasma concentration was above this value for a certain period. On day 28, the mefloquine plasma was still higher than 500 µg/L in 28% of the patients with complete cure, 45% in late recrudescence, and 10% in early recrudescence (P = 0.037). Based on the population model, the estimated mefloquine plasma concentrations were above 500 µg/L during a mean interval of 507 (range = 504–560) hours. This was similar for the three regimens and treatment outcomes. However, this interval is a conservative estimate since the model underestimates the observed terminal plasma concentrations and, consequently, the interval of plasma concentrations above 500 µg/L.
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Early recurrence during the first two weeks is usually a true recrudescence, while late parasite recurrences may include re-infections. Due to the long plasma residence time of mefloquine and the low incidence of malaria in our study area at present, there is little over-estimation of the true recrudescence rate in this study. Conversely, other studies indicated that the follow-up after mefloquine treatment should be extended until 63 days.13 Since the follow-up in this study was only 28 days, the real parasite recurrence rate may be somewhat higher.
The pharmacodynamics of this antimalarial combination can be divided into two components: initial parasite elimination and complete eradication. The fast initial parasite killing and clinical recovery are the effects of artesunate and they were similar to values found in other studies with artemisinin.2,3,5,6 No synergy with mefloquine was observed. The explanation of the unacceptably high rate of incomplete parasite eradication is more complex. Recrudescence, a frequent complication of monotherapy with artemisinin drugs, is associated with high initial parasite counts and with slower parasite clearance rates, but not with resistance to artemisinin.5,6,14 Exposure to effective artemisinin blood concentrations is often not long enough due to their short plasma residence time and time-dependent decrease of blood concentrations.15–17 This is the pharmacodynamic basis for combining artemisinin drugs with longer acting agents, to ensure long suppressive drug concentrations.18
Complete eradication of all parasites therefore depended on exposure to mefloquine, but apparently this was insufficient. This may be caused by two factors: sub-therapeutic plasma concentrations and resistant parasites. The plasma concentrations depend on dose, bioavailability, and disposition. The dose was strongly correlated with Cmax and AUC0–, but these were not associated with parasite recurrence. Variable disposition caused variation in plasma concentrations on day 28. Low plasma concentrations on day 28 were associated with parasite recurrence, indicating that duration of exposure to parasitocidal plasma concentrations is important complementary to the height of peak concentrations.
Synergy between artesunate and mefloquine was not detected, mainly because by the time that mefloquine starts to exert its activity, artesunate is no longer present in the blood.17 Mefloquine pharmacokinetics were not different between the three treatment groups, except for a difference in the observed Cd28. This difference is explained by the different timing of mefloquine administration, although the magnitude of the difference is not completely explained by the 8-and 24-hour differences in dosing time and the estimated t1/2el. Pharmacokinetic interactions between artesunate and mefloquine, as suggested by other investigators, were not found and thus do not explain the subtherapeutic mefloquine plasma concentrations in patients with parasite recurrence.7 Similarly, the improvement of bioavailability of mefloquine or artesunate during recovery from malaria, reported by investigators in Thailand, was not confirmed in this study.8,19,20 Other potential sources of variation in drug disposition not studied here are stereoselective pharmacokinetics and pharmaceutical bioavailability.21,22 The pharmaceutical bioavailability of the mefloquine formulation used in this study is possibly lower than of other formulations, but this does not explain inter-individual variation as a cause of parasite recurrence.23
Incomplete eradication of parasites may also have been caused by resistance of parasites. The official definition of resistance is based on the administration and absorption of a therapeutic, non-toxic, dose of an anti-malarial drug, and not on plasma concentrations. The mefloquine dose used in this study (15 mg/kg) was lower than the therapeutic dose that is recommended for southeast Asia (25 mg/kg), and that would preclude any statement on parasite resistance. However, the population pharmacokinetic model yielded a rather accurate description of the individual concentration time curves, even at the extremes of the population, and based on this we can draw some tentative conclusions. The observed mefloquine plasma concentrations and the concentrations fitted by the model, even though this tends to underestimate the observed plasma concentrations on day 28, were all above the in vivo suppressive (chemoprophylactic) plasma concentration for a fully sensitive strain (500–625 µg/L) for a certain period of time, even in the cases with early recrudescence.24,25 This points to a reduced sensitivity of P. falciparum to mefloquine in southern Vietnam.
Mefloquine resistance has not been previously documented in Vietnam. It may have been overlooked, but the general view is that resistance is absent because mefloquine was used mainly in combination with artemisinin drugs. The Vietnamese drug policy has been restrictive towards use of antimalarial agents in the private health sector. Mefloquine was introduced in the public health sector in 1992 and always used in combination with artemisinin drugs. This is different from Thailand, where mefloquine resistance developed within several years of applying mefloquine monotherapy in highly endemic areas.26–29
Reduced mefloquine sensitivity is a reason for concern. First, it would show that a policy to use mefloquine only in combination with artemisinin drugs may not prevent development of mefloquine resistance.18 Second, a 15 mg/kg dose of mefloquine is not enough in Vietnam, despite the induction of rapid parasite clearance by artesunate. This has implications for the use of mefloquine and artesunate in other regions.30
This study raises questions if a dose of 15 mg/kg mefloquine should be applied at all, even in areas without documented mefloquine resistance. Notably, in a non-scientific setting, subtherapeutic plasma concentrations will be common because the dose will not always be adapted exactly to body weight and pharmaceutical bioavailability is variable. Even in combination with artemisinin drugs, such a low dose of mefloquine could lead to recrudescence and selection of resistant parasite strains. A related question is if only increasing the mefloquine dose is sufficient to prevent recrudescence. The experience in Thailand showed that a regimen with 25 mg/kg of mefloquine, divided over a two-day period, plus three days of treatment with artesunate, is superior to single-dose regimens with respect to tolerance and efficacy.31,32
In conclusion, there is no effect of different dosing times of a 15 mg/kg dose of mefloquine within the first 24 hours after an initial single dose of artesunate. The pharmacodynamics of this low dose artesunate-mefloquine combination reflect the characteristics of artesunate monotherapy, i.e., parasite recurrence being associated with high initial parasite counts and slower initial parasite clearance, and confirm the association between parasite recurrence and lower mefloquine plasma concentrations. However, the data also suggest reduced mefloquine sensitivity of P. falciparum in Vietnam. Therefore, the mefloquine dose should be increased to 25 mg/kg, in combination with artemisinin drugs, preferably given for three days.
Received September 18, 2003. Accepted for publication February 27, 2004.
Authors’ addresses: Le Q. Hung and Phan T. Giao, Division of Infectious Diseases, Tropical Medicine & AIDS, Academic Medical Center F4-217, PO Box 22700, 1100 DE Amsterdam, The Netherlands and Department of Tropical Diseases, Cho Ray Hospital, 201 B Nguyen Chi Thanh, District 5, Ho Chi Minh City, Vietnam, E-mail: lequochung{at}saigonnet.vn. Peter J. De Vries and Piet A. Kager, Division of Infectious Diseases, Tropical Medicine & AIDS, Academic Medical Center F4-217, PO Box 22700, 1100 DE Amsterdam, The Netherlands, E-mails: p.j.devries{at}amc.uva.nl and p.a.kager{at}amc.uva.nl. Tran Q. Binh, Department of Tropical Diseases, Cho Ray Hospital, 201 B Nguyen Chi Thanh, District 5, Ho Chi Minh City, Vietnam, E-mail: binhtq{at}hcm.vnn.vn. Nguyen V. Nam, Binh Thuan Provincial Malaria Center, Phan Thiet, Binh Thuan Province, Vietnam. R. Holman, Department of Clinical Epidemiology and Biostatistics, Academic Medical Center J2-212, PO Box 22700, 1100 DE Amsterdam, the Netherlands, E-mail: r.holman{at}amc.uva.nl.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. A. Ashley, K. Stepniewska, N. Lindegardh, R. McGready, R. Hutagalung, R. Hae, P. Singhasivanon, N. J. White, and F. Nosten Population pharmacokinetic assessment of a new regimen of mefloquine used in combination treatment of uncomplicated falciparum malaria. Antimicrob. Agents Chemother., July 1, 2006; 50(7): 2281 - 2285. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), median (
), and lowest (
) mefloquine plasma concentrations, coincidentally, all treated with regimen C. The mean inhibitory concentration (500 µg/L) of a fully sensitive strain is also indicated by the dashed line. h = hours.