This pharmacokinetic study was part of an open-label, randomized uncontrolled, dose optimization clinical trial in 114 patients with acute uncomplicated falciparum malaria. The study was conducted during 2006 at the two main study sites – Bangkok (Hospital for Tropical Diseases; n = 84), and Tak Province (Mae Sot General Hospital and Umphang Hospital; n = 30). The latter is located in an area along the Thai-Myanmar border, where multidrug resistance exists. Approval of the study protocol was obtained from the Ethics Committees of the Ministry of Public Health of Thailand and the Faculty of Tropical Medicine, Mahidol University, Bangkok. A total of 80 patients, 77 males and 3 females, of varying ethic groups (30 Karen, 20 Burmese and 20 Thai) were included in the pharmacokinetic study.

The age, body weight and admission parasitaemia ranges were 15 – 61 years, 45.0 – 68.0 kg and 1,152 – 254,800 asexual parasites/μl respectively. Pre-treatment investigations consisted of clinical assessments (general medical history, demographic data, significant medical history, previous drug administration, physical examinations, monitoring of vital signs and malaria signs and symptoms), laboratory assessments (thick and thin blood smears for parasite identification/quantification), routine haematology and biochemistry, urinalysis and stool examination for parasites and ova). Most patients showed baseline laboratory parameters outside the normal ranges, which are common in patients with acute falciparum malaria and all were mild or moderate in severity.

All patients were admitted to the hospitals for 7 days and were requested to attend for follow-up visits on days 14, 21 and 28 or until suspected signs and/or symptoms of malaria appeared.

This study was planned as a proof-of-concept study to investigate the combination regimen of fosmidomycin-clindamycin that would produce approaching 100% cure rate with short duration of 3 days and less frequent dosing interval. The design of dosage regimens used in the present study was based on background information from previous studies 111213. The total number of patients recruited for Group I and II was at the ratio of approximately 1: 3.5 (25 vs 89 cases, respectively). Patients from each study site were sequentially entered into the study and treated with one of two different combination regimens of fosmidomycin and clindamycin as follows: Group I (n = 25): fosmidomycin sodium capsules (Alphamed PHARBIL Arzneimittel GmbH, Germany) 450 mg × 2 tablets, six-hourly for 3 days plus clindamycin hydrochloride capsules (Alphamed PHARBIL Arzneimittel GmbH, Germany) 150 mg × 2 tablets, six-hourly for 3 days; and Group II (n = 89): fosmidomycin sodium capsules 450 mg × 4 tablets, twelve-hourly for 3 days plus clindamycin hydrochloride capsules 150 mg × 4 tablets, twelve-hourly for 3 days.

The patients fasted overnight (8–10 hr) and emptied their bladders immediately before drug administration. The drugs were administered orally with 250 ml of water and a standard hospital meal (20–25% fat content) under supervision of the assigned study staff. After ingestion of the medication, patients were observed for 1 hr and in the event of vomiting, the medication was re-administered. Meals were provided at 4 hr post-dosing. Fruit juice and water were freely available throughout the period of hospitalization. The administration of prescription drugs, deemed necessary for the well-being of the patients, were permitted during the study period and documented.

Pharmacokinetic investigation was performed in all (n = 25) patients in Group I, while only 54 cases in Group II from both study sites were randomly recruited for pharmacokinetic investigation. Blood samples (4 ml) were obtained at the following time points:- Group I: 0, 1, 2, 3, 4, 6, 8, 12, 18, 24, 26, 30, 36, 42, 48, 50, 54, 60, 66, 72 and 74 hours from the commencement of dosing; Group II: 0, 1, 2, 3, 4, 6,12, 14,24, 26, 36, 38, 48, 50, 60, 62, 72 and 74 hours from the commencement of dosing. Blood samples were drawn via an indwelling venous cannula and collected in heparinized tubes. Where appropriate, blood samples were taken immediately prior to the next dose. Immediately after collection, the tubes were centrifuged (2,000 g) for 10 minutes. The plasma samples were stored at -80°C pending analysis.

Twelve hourly urine collections were made during the first four days as follows: 0–12, 13–24, 25–36, 37–48, 49–60, 61–72, 73–84 and 85–96 hours. The total volume of each collection was recorded. A 20 ml aliquot was taken from each collection and stored at -80°C until analysis.

Efficacy assessments included clinical and parasitological evaluations. Clinical signs/symptoms of malaria including body temperature were monitored on days 0, 1, 2, 3, 4, 5, 6, 7, 14, 21 and 28. Finger-prick blood smears were examined at 6 hourly intervals until three consecutive negative readings were obtained, then once daily during hospitalization, and then on days 14, 21 and 28. The films were stained with Giemsa and parasite counts were determined by counting the number of asexual parasites per 1,000 red blood cells on a thin film or per 200 white blood cells on a thick film. The protocol used to determine the response to treatment was adapted from the World Health Organization 14. Efficacy was assessed using the following criteria: (i) polymerase chain reaction (PCR)-adjusted 28 day cure rate:- the proportion of patients with clearance of asexual parasitaemia within 7 days of the commencement of treatment and without subsequent recrudescence during the 28 day follow-up period: (ii) parasite clearance time (PCT):- as calculated from the commencement of therapy to the time of the first negative film for asexual parasites subject to two consecutive films remaining negative: (iii) fever clearance time (FCT):- as calculated from the commencement of therapy to the time that the temperature falls below 37.5°C and remains below 37.5°C for 48 hours.

Safety and tolerability were assessed on the clinical findings and abnormal laboratory tests (routine haematology and biochemistry) that first occur or increase in severity during the study period in accordance with the Common Toxicity Criteria CTC grade 15.

Fosmidomycin concentrations in plasma and urine were determined by a bioassay system based on agar disk diffusion using Enterobacter cloacae ATCC 23355 as the test organism 16. The precision of the assay method was examined and the within-day (repeatability) and day-to-day (reproducibility) variations were below 5% (% coefficient of variation: % C.V.). Good accuracy was also observed for both the intra-day and inter-day assays. The limit of quantification was 1 ng using 40 μl of plasma or 7.5 μl of urine.

Plasma concentrations of clindamycin were measured by high performance liquid chromatography as described previously 17. The precision of the method within-day and day-to-day variations were below 15% (% C.V.) Good accuracy was observed for both the intra-day and inter-day assays, as indicated by the minimal deviation of mean values found with measured samples compared to theoretical values (below ± 15%). The limitation of quantification was considered acceptable at 0.07 μg, using plasma sample quantities of 1 ml.

The non-compartmental pharmacokinetic analysis was used to determine the pharmacokinetic parameters of fosmidomycin and clindamycin after a single initial dose of the two drugs 18. C_max was defined as the maximum plasma concentrations and t_max was the time required to reach maximum plasma concentrations following a single initial dose or the final dose. C_min-obs was the observed trough concentration. The area under the concentration-time curve from the zero time point to the last sampling point yielding a quantifiable concentration (AUC_{0, t}) was calculated with the log-linear trapezoidal method. A minimum of three final sampling points were used to determine the first order constants generated by the terminal portion of the curve (λ_z). Terminal phase elimination half-life (t_1/2z) was defined as t_1/2z = ln(2)/λ_z. The AUC from time zero to infinity (AUC) was calculated as AUC_0-t + Ct/λ_z, where C_t stands for the last measurable concentration. Total clearances (CL/F) were calculated from Dose/AUC_0,8 and apparent volumes of distribution (V_z/F) were calculated as CL/F/λ_z, where F is bioavailability. The maximum plasma concentrations at steady-state (C_max-ss), minimum plasma concentrations at steady-state (C_min-ss) and average plasma concentrations at steady-state (C_ave-ss) following a fast release oral formulation could all estimated using:

C max ⁡ − ss = Dose V z F ( 1 − e − λ ⋅ τ ) MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGaciGaaiaabeqaaeqabiWaaaGcbaGaem4qam0aaSbaaSqaaiGbc2gaTjabcggaHjabcIha4jabgkHiTiabbohaZjabbohaZbqabaGccqGH9aqpjuaGdaWcaaqaaiabbseaejabb+gaVjabbohaZjabbwgaLbqaaiabdAfawnaaBaaabaGaeeOEaOhabeaacqGGgbGrcqGGOaakcqaIXaqmcqGHsislcqqGLbqzdaahaaqabeaacqGHsislcqaH7oaBcqGHflY1cqaHepaDaaGaeiykaKcaaaaa@4BC2@

C_min-ss = C_max-ss ·e^-λ.τ

C ave-ss = Dose CL / F ⋅ τ MathType@MTEF@5@5@+=feaagaart1ev2aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xI8qiVKYPFjYdHaVhbbf9v8qqaqFr0xc9vqFj0dXdbba91qpepeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGaciGaaiaabeqaaeqabiWaaaGcbaGaee4qam0aaSbaaSqaaiabbggaHjabbAha2jabbwgaLjabb2caTiabbohaZjabbohaZbqabaGccqGH9aqpjuaGdaWcaaqaaiabbseaejabb+gaVjabbohaZjabbwgaLbqaaiabboeadjabbYeamjabc+caViabbAeagjabgwSixlabes8a0baaaaa@4438@

Where τ is the dosing interval (hr).

Fluctuations between C_max-ss and C_min-ss were calculated from the ratio between the two parameters.

For the purpose of future stimulation and prediction as well as pharmacokinetic/pharmacodynamic modelling, the compartment open model (one- or two-) with first-order absorption and elimination with absorption lag time was fitted to the data by iterative, weighted non-linear regression using ADAPT II (release 4.0) 19. The observed concentrations of fosmidomycin and clindamycin were weighted as the reciprocal of the analytical variance. Data reported in the previous study 11 were used for the initial estimation of the pharmacokinetic parameters. The adequacy of the pharmacokinetic models chosen was based on statistical methods for assessing the validity of the models for describing the concentration-time data, i.e., F-ratio test, Akaike information criterion (AIC), Schwartz and Imbimbo criteria.

Total urinary excretion of fosmidomycin was calculated from the sum of amounts of drug excreted during each 12-hour interval multiplied by total urine volume at each interval. The dose excreted (%) was calculated from total amounts of drug excreted in urine during 72 hours divided by total dose given.

Descriptive statistics were calculated for the pharmacokinetic variables. Geometric means were determined for baseline parasite counts. All pharmacokinetic parameters were presented as median values (95% C.I.). Comparisons of the pharmacokinetic parameters between the two groups were performed using the Mann-Whitney U test. The level of statistical significance was set at α = 0.05 for all tests.

The concentration-time data of both fosmidomycin and clindamycin were in all cases best fitted with one-compartment open model with first-order absorption and elimination. Observed and model-predicted (one-compartment model with first-order absorption and elimination with absorption lag time) median plasma concentrations of fosmidomycin and clindamycin in patients with sensitive responses following implementation of the two combination regimens are shown in Figures 2 and 3, respectively. The pharmacokinetic parameters of both drugs analysed by model-independent and model-dependent methods are summarized in Table 1 and 2 respectively. Variability as expressed as % C.V. for all parameters varied between 7.7 and 40.5% for fosmidomycin and 12.2 and 38.9% for clindamycin. Results obtained from model-dependent analysis for both fosmidomycin and clindamycin were generally in agreement with those obtained from model-independent analysis. Fluctuation was significantly higher (260% for fosmidomycin and 120% for clindamycin) following the administration of the high dose combination regimen.