TRANSACTIONS OFTHE ROYAL SOCIETY OFTROPICAL MEDICINE AND HYGIENE (1997) 91,322-327

ChlorproguaniVdapsone for uncomplicated Pfasmodium falciparum malaria in young children: pharmacokinetics and therapeutic range

Peter Winstanley1p2, William Watkins Marsh’,3

ly2, David Muhial, Simon Szwandt2, Evans Amukoye’ and Kevin ‘Kenya Medical Research Institute (KEMRI), CRC Research Unit, Kilji, Kenya;2Department of Pharmacology

and Therapeutics, University of Liverpool, Liverpool, UK;3Nufield Department of Clinical Medicine, University of Oxford, Oxford, UK

Abstract The disposition of chlorproguanil/dapsone (one daily dose for 3 d of 1.2 and 2.4 mg/kg respectively) has been studied in young children with Plasmodium falciparum malaria, to provide data complementary to a clinical trial of this drug combination. Unbound concentrations of chlorcycloguanil (the active metabolite of chlorproguanil) and dapsone in clinical samples have been related to the unbound drug concentrations which produced defined outcomes in tests in vitro of drug efficacy and toxicity. Twelve children with un- complicated malaria were treated: all cleared parasitaemia within 72 h and made uneventful recoveries, After the first dose of chlorproguanil/dapsone the maximum unbound chlorcycloguanil concentration in clinical samples (19 ng/mL [about 60 KIM]) was 2 orders of magnitude above the 50% inhibitory concen- tration (X50) value for this drug against the K39 strain of I? fulciparum, while falling 2 orders of magnitude below its IC50 against human bone marrow cells; the maximum unbound dapsone concentra- tion in clinical samples (160 ng/mL [about 645 no]) was lo-fold higher than its IC50 against the K39 strain. However, because of the rapid elimination of chlorproguanil from the body (half-life 12.ti6.3 h), the minimum fractional inhibitory concentrations of unbound chlorcycloguanil/dapsone against the K39 strain were probably exceeded for no more than 6 d. These data, together with the clinical trial, will be helpful in deciding whether current chlorproguanilidapsone doses are optimal for the treatment of falci- parum malaria.

Keywords: malaria, Plasmodium fulciparum, chemotherapy, chlorproguanil/dapsone, chlorcycloguanil, pharmacoki- nerics, children

Introduction Although Plasmodium falciparum malaria is a problem

throughout the tropics, the greatest burden of disease is found in sub-saharan Africa: at a conservative estimate the disease kills 2 million people per year in Africa, main- ly young children (WHO, 1995). For logistic reasons, chemotherapy of uncomplicated malaria (mainly in out- patients) is currently the principal means of disease ‘con- trol’ in Africa. Chloroquine has been the mainstay of such chemotherapy, but the prevalence of clinically sig- nificant chloroquine resistance in many areas is now such that alternative drugs are needed @LOLAND et al., 1993). Antifolate drugs ire considerably cheaper than other alternatives (FOSTER. 199 1; SUDRE et al.. 1992). including halofant&e, mefioquine, pyronaridine (WIG: STANLEY, 1996), atovaquone/proguanil (RADLOFF et al., 1996) and artemisinin derivatives, and pyrimeth- amine/sulfadoxine is becoming widely used as the ‘suc- cessor’ to chloroquine-especially in high-risk patients like young children. However, pyrimethamine and sulf- adoxine are both eliminated slowly (WINSTANLEY et al., 1992), favouring the selection bf resistant parasites (WATKINS & MOSOBO, 1993), and their widespread use is likelv to result in raoid develonment of clinical drug. re- sistande, as happene; in south-kast Asia (WHO, 1955). This would be disastrous in the absence of affordable al- ternative drugs.

In contrast, the antifolate combination chlorpro- guaniudapsone (which has been shown to clear asymp- tomatic I? falcipancm parasitaemia; WATKINS et al., 1988) is eliminated rapidly from the body (WATKINS et al., 1987), and so may impose less selection pressure for resistance than pryimethamine-sulfadoxine. Further- more, in vitro data suggested that the combination of chlorcycloguanil (the active merabolite of chlorpro- guanil) plus dapsone had considerably greater activity against a Kenyan pyrimethamine-resistant strain of l? falciparum than did pyrimethamine/sulfadoxine, but was

Address for correspondence: Dr Peter Winstanley, Depart- ment of Pharmacology and Therapeutics, University of Liver- pool, Liverpool, L69 3GE, UK; phone +44 (0) 151 794 5539, fax +44 (0) 151 794 5540, e-mail peterwin@liv.ac.uk

no more toxic to human bone marrow cells in culture (WINSTANLEY et al., 1995). However, as a guide to clin- ical use, such in vitro data are of limited value, in partic- ular because (i) the ratio of free to bound drug in vivo may differ from that in vitro because of variation in plas- ma protein concentration, (ii) it is impossible to deter- mine from in vitro data for how long synergistic drug concentrations remain within the therapeutic range in viva, and (iii) whole-animal toxicity cannot be deduced from cell culture experiments. A double-‘blind’ clinical trial has just been completed in Kenya (results in prep- aration), comparing 2 dosing regimens of chlorpro- guaniudapsone with pyrimethamine/sulfadoxine in the treatment, as outpatients, of young children with un- complicated falciparum malaria. The chlorproguanill dapsone doses used in the trial (1.2 mg/kg and 2.4 mg/ kg respectively, either as a single dose of the combina- tion or daily for 3d) were derived empirically. Dose op- timization is now a pressing issue: it is important to define therapeutic range, drug disposition, and the max- imum tolerable dose. The present work addresses the first 2 of these issues; assessment of drug tolerability is in progress. We have estimated the therapeutic range of chlorproguanil/dapsone by measuring the unbound fractions of each compound in the culture media em- ployed for previous work in vitro and in plasma. We have also studied the disposition of the drugs in children with uncomplicated malaria, and estimated how long the current treatment regimen achieves drug concentrations above the therapeutic target.

Methods Clinical study

Patients. The study was approved by the National Ethics Committee of Kenya; written consent was ob- tained after explanation to parents in their chosen lan- guage. The work was done in the paediatric high- dependency ward of the Kenya Medical Research Insti- tute unit (KEMRI) in the grounds of Kilifi District Hos- pital in Kenya, which has a nurse to patient ratio of at least 1:6. Children admitted to hospital with uncompli- cated falciparum malaria were studied if they (i) were aged 6 to 71 months (inclusive), (ii) had pure l? fulci-

PHARMACOKINETICS OF CHLOPROGUANIUDAPSONE

parum parasitaemia (2000-250000 per & or <lo% parasitaemia if assessed from a thin blood film), and (iii) had a haemoglobin level >4*9 g/dL. Children were ex- cluded if they (i) had severe malaria (WHO, 1990), (ii) had an intercurrent infection, (iii) were allergic to sul- phonamides, or (iv) had been treated with Fansida& or Metakelfin@ within the last month. Prior treatment with chloroquine was not an exclusion criterion. All patients were given oral paracetamol syrup (10 mg/kg every 6 h) for fever, and were tepid-sponged if their rectal temper- ature exceeded 40°C. Children were examined daily by a doctor, and parasitaemia was estimated every 6 h dur- ing the first 48 h. Patients were detained in hospital until they were well (minimum 72 h).

Antimalarial chemotherapy. Patients were treated with oral chlorproguanil (1.2 mglkg) and dapsone (2.4 mgl kg) at 24 h intervals for 3 d; all doses were given by 2 of the authors (P.W. and W.W.) and children were careful- ly observed for vomiting. Chlorproguanil/dapsone was formulated extempore as a suspension. Chlorproguanil (LapudrineQ Zeneca) and dapsone (BP) tablets were powdered and extracted with ethanol/propylene glycol; to this suspension was added syrup (BP), sodium ben- zoate (0.1% w/v, as preservative), and flavouring. The final concentrations of chlorproguanil and dapsone were 2.4 mg/mL and 4.8 mg/mL, respectively. This suspen- sion was identical to that used in the randomized trial.

Blood sampling. Sampling was done initially, and for as long as possible, through an in-dwelling Teflon@ can- nula inserted in a forearm vein, and kept patent with heparinized saline, and thereafter by venepuncture. Blood (2 mL) was drawn before the first dose and at the following times: 1 h, 2 h, 4 h, 6 h, 12 h, 24 h (before the second dose), 28 h, 36 h, 48 h (before the final dose), 52 h, 60 h, 72 h, 96 h and 120 h. The maximum volume of blood drawn for drug measurement was 30 mL over 5 d. Blood was centrifuged (at about 2000g for about 10 min) within 12 h of sampling, and plasma was stored at -20°C until assayed.

Plasma protein binding Protein binding in human plasma. Because the children

recruited for pharmacokinetic work were young, and blood volumes were therefore strictly limited, it was necessary to recruit other Kenyan children with uncom- plicated falciparum malaria (fulfilling the same entry criteria) for the plasma protein binding study: a single blood sample (6 mL) was drawn from each child before antimalarial treatment was given. Blood was centrifuged and plasma was stored as already described. With all the samples, plasma was ‘spiked’ with chlorproguanil (final concentration 100 ng/mL), chlorcycloguanil (100 ngi mL), or dapsone (1 mg/mL), adjusted to pH 7.4 (using 0.1 M HCl) and incubated (37°C; 0.5 h). Centrifugation dialysis (25 min; 1000 g at ambient temperature) was then done with commercially-available kits (Amicon@ micropartition systems). Concentrations of al -acid glycoprotein in plasma samples were measured by radial immunodiffusion (Beringwerke, Germany).

Protein binding in culture media. Plasma protein bind- ing experiments employed the media used in previous work in vitro (WINSTANLEY et al., 1995) for culture of l? falciparum (RPMI-1640 medium containing HEPES buffer [25 mM], bicarbonate [25 mu], and normal pooled human serum [lo%]) and human bone marrow cells (Iscove’s modified Dulbecco’s medium containing fetal calf serum [20%], streptomycin [lo0 mg/mL] and penicillin [lo0 iu/mL]). Parasite medium and bone marrow cell medium were ‘spiked’ with chlorcyc- loguanil(lO0 ng/mL and 3.0 pg/mL, respectively; 4 rep- licates of each medium) or dapsone (12.4 and 5.0 ug/ mL, respectively; 4 replicates of each medium). Ali- quots were incubated at 37°C for 0.5 h; pH (7.5 for both media) was not adjusted before dialysis. Centrifugation dialysis was performed as already described.

Binding to single proteins. To determine the association

323

constants (KA) of chlorproguanil, chlorcycloguanil and dapsone with the major drug binding proteins albumin and at-acid glycoprotein, each drug (at final concentra- tions of 150 ng/mL, 100 ng/mL and 1 l.tg/rnL, respec- tively; 6 replicates) was incubated (37”-C for O-5 h) with human serum albumin (33.3 giL) or at-acid glycopro- tein (0.5 g/L) in phosphate buffer (pH 7.4; 100 rnM). Centrifugation dialysis was done under the conditions described above. KA was determined from the expres- sion.

Yy c... p

where Ct,d=the concentra:on of drug bound, C,=the concentration of drug unbound and P=the protein con- centration.

Drug assays Plasma concentrations of chlorproguanil, chlorcyc-

loguanil, dapsone and monoacetyl dapsone were meas- ured in Nairobi by reversed-phase high performance liquid chromatographic methods (HELSB~ et al., 1990; MAY et al., 1990). For chlorproguaniVchlorcycloguani1, 1 mL of plasma was assayed; for dapsone/monoacetyl dapsone, 100 ~.IL were assayed. Drug concentrations in dialysate were measured by the same methods, but sam- ple extraction was not required and dialysate was inject- ed directly on to the column.

Pharmacokinetic analysis Data were analysed in a non-compartmental manner

using commercially-available software (Topfit@, Scher- ing, Germany). The elimination rate constant Cp) for chlorproguanil was determined from the elimination phase of the first dose; this was not possible for dapsone or monoacetyl dapsone (because of slower elimination rates, and the need for a second drug dose), and /I was therefore determined from the elimination phase of the third dose. Areas under the concentration versus time curves (AUC) from dosing to the end of the first dose interval (AUCe-24) were derived by summation of trap- ezoids; AUC from the last measured concentration (C,) to infinity was derived from the expression C,/p, and the total AUC was obtained by summation. Oral clearance (CIfF) was derived from the expression dose/AUC and the apparent volume of distribution (I’d/F) was derived from the expression p/CL.

Results Response to treatment

Twelve children with uncomplicated falciparum ma- laria were recruited for pharmacokinetic study: their ad- mission characteristics are described in Table 1. All patients recovered without complications within 72 h and were discharged from hospital. Parasitaemia re- mained unchanged or rose for the first 18 h after the

Table 1. Admission characteristics of children with Plasmodium falciparum malaria

Age Weight Haemoglobin Parasitaemia No. (monW (kg) 6W-J (per NJ

11 18 8.6 13 20 8.2 14 44 12.6

30 22 8.7 35 33 10.6

46

2

;z 8.2

11.0

;: 10.1 9.6

46 14.6 18 7.8 12 9.2

5.7 5.5 9.4 8.7

10.4 9.5

10.8

;:; 7.5 6.6 9.9

233520 6724

230000 5952 8652

167200 22010

5340 250000 250000 137200 173760

324

Fig. 1. Mean Plasmodium fdciparum parasitaernia in 12 chil- dren after treatment with chlorproguanil plus dapsone; vertical lines indicate +l SD, asterisks (*) indicate times when treat-

0 20 40 60 80 loo 120

Time (Hours)

Fig. 2. Concentration versus time curves of dapsone (0), monoacetyl dapsone (W), chlorproguanil (0) and chlorcyc- loguanil (0) following 3 oral doses of chlorproguanil (1.25 mg/ kg) plus dapsone (2.4 mg/kg) at 24 h intervals.The data points are means of determinations in 12 children with I! falciparum infection; asterisks (*) indicate times when treatment was giv- en.

start of treatment. Thereafter, parasite counts fell in a log-linear manner (Figure 1); all patients cleared parasi- taemia within 72 h of the first chlorproguanil/dapsone dose.

Pharmacokinetics Plasma concentration versus time curves for chlor-

PETERWINSTANLEY ETAL.

proguanil, chlorcycloguanil, dapsone and monoacetyl dapsone are shown in Fig. 2; Tables 2 and 3 give the de- rived pharmacokinetic parameters. Blood sampling ter- minated prematurely in 3 cases (last sample at 36 h in one child and at 48 h in 2 others) because of failure of venous access (one child) or withdrawal of consent (2 children).

After the first dose of chlorproguanilidapsone, dap- sone and monoacetyl dapsone were detectable in the plasma at 1 h in 11 of the 12 children, and at 4 h in all children. Dapsone concentrations (expressed as mearbSD) rose slowly to reach a peak concentra- tion(C,,& of 5345 164 ng/mL. C,,, values for dapsone after the second and third doses were 747?356 and 824,424 ng/mL, respectively. The elimination half- time (tl, ) of dapsone could not be estimated after the first and second doses, but was 24.M6 h (mea&SD) af- ter the final dose. The ratios of AUC values of parent sulphone to metabolite ranged between 1.2 and 2.8.

After the first dose of the combination, chlorpro- guanil was detectable at 1 h in 8 children, and at 2 h in 11; chlorproguanil was undetectable in the plasma of one child throughout. Concentrations (meanfSD) rose quickly to reach qm,=95f 67 ng/mL; C,,, values after the second and third doses were 12017 1 and 84+26 ngt mL, respectively. After the first dose, the mean tf, of chlorproguanil was 12.6 h (s~f6.3); individual values varied 3.8-fold.

Chlorcycloguanil was detected in the plasma of 8 of the 12 patients; in 3 of these children the active metab- olite was detected in only one plasma sample. After the first chlorproguanil/dapsone dose, C,,, (mea&SD) for chlorcycloguanil was 274 11 ng/mL. The elimination phase of chlorcycloguanil could not be estimated after any of the chlorproguanil doses. The ratios of AU%-24 values of parent drug to those of the active metabolite ranged between I.1 and 7.3.

Plasma protein binding Blood for the plasma protein binding study was

drawn from 7 children with uncomplicated falciparum malaria aged between 29 and 61 months. Admission parasite count ranged from 5341 to 125 OOOimL and haemoglobin levels ranged from 6.4 to 10.4 g/dL. The mean concentration (*SD) of at-acid glycoprotein in plasma samples was 2.8kO.6 g/L. Table 4 gives the un- bound fractions (f,) of the 3 compounds, in plasma and culture media. KA values of each drug for each protein are given in Table 5.

Assay specifications The lowest detectable plasma concentrations of chlor-

proguanil, chlorcycloguanil, dapsone and monoacetyl dapsone were 5, 10, 15 and 10 ng/mL, respectively. The

Table 2. First dose pharmacokinetics of dapsone in children with Plasmodium falciparum malaria

Dapsone Monoacetyl dapsone C max knax Au%-24 AUC tJl, a VdlF CL’F AucO-24 AUC

No. (ng/mL) 04 @g/mL.h) (Clg/mL.h) (h) (L/kg) (mUmin/kg) @g/mL.h) (pg/mL.h)

:: 530 630 24 12 1’2.40 23.6 - 21.1 - 3.10 - 1.70 - 7.7 2.2 5.8 -

;“o 460 520 24 6 7.8 - 27 490 6 ;:;

- - - - - - 10.7 4.8 - - 29.0 34.0 4;5 1.38 5.6 10.5

30 460 12 35 660 12

17.58 12.8 24.3 6.54 3.12 1.3 - 25.2 17.3 1.29 1.59 6.0 10.1

39 640 ii 11.6 24.1 21.7 3.13 1.66 5.2 10.9 43 290 4.1 -

2i.l - 2.3

46 860 12 15.2 40.4 2.50 oG9 6.9 25.1 58 260 24 4.4 - - - - - 2.9 60 610 24 7.8 - - - - - 3.3

Mean 534 25.8 24.5 3.43 1.74 4.9 12.4 SD 164

;:; 8.9 6.0 1.77 0.72 2.7 7.3

aThe elimination rate constant was derived from final dose data.

PHARMACOKINETICS OF CHLOPROGUANIUDAPSONE 325

Table 3. First dose pharmacokinetics of chlorproguanil in children withPlasmodium fakiparum malaria

Chlorproguanil C Gnax

(rig/mm) (h) AucO-24 AUC t1, a

No. (6) 2;:) (mI%E/kg)

ChlorcycloguCanil AUCO-24 ma*

(ng/mL.h) (ng/mL.h) (ng/mL.h) (ng/mL)

11 13 14 20

i: 35 39 43 46 58 60 Mean SD

126 6 1957 69 6 1362 49 4 784

279 12 1835 47 6 659 83 4 1138 59 4 604 54 4 929 89 4 1012 72 6 1205

123 95 67

- 6

-

- 1672 1196 464

3262

922 -

756 1656 682

1926 1081

- - -

1469 916

16.2 8.6 -

16.0 3;1 - 7.5 Gl

13.0 13.4 6.7 17.1

23.0 20.7 6.0 9.6 - -

- - 12.6 16.7 6.3 7.3

6.1 -

21.7 -

26.4 12.1 29.3 10.4 18.5

- -

17.8 8.6

275 40a 25 608 38 7a

- 14a

536 - 27

138 - 11

662 35 - - - -

444 - 226 -

aOne datum point only.

Table 4. Fractions of chlorproguanil, chorcycloguanil and dapsone which were unbound in human plasma from patients with Plasmodium falczparum malaria and culture mediaa

Chlorproguanil Chlorcycloguanil Dapsone

Patient’s plasma 0.366(0.036) 0,721(0.075) 0.321(0.023) Culture medium l? falciparum NT 0+70(0.079) 0,770(0.038) Bone marrow NT l,OOO(-) 0~904(0.047)

aMeans (standard deviations in parentheses); NT=not tested.

Table 5. Association constants (KA) of cblor- proguanil, chlorcycloguanil and dapsone for albumin and al-acid glycoprotein

Albumin KA c”-‘ja al-Acid glycoprotein

Chlorproguanil 7.2(1.1)x103 Chlorcycloguanil 0.3 (0.3)x103

148(13)x103 Nil

Dapsone 2.6(0.2)x103 Nil

aIn 100 mM phosphate buffer at pH 7.4; values are means (standard deviations in parentheses). Nil=no binding detected.

within-day coefficients of variation were below 10% for all 4 compounds, and calibration curves were linear (00.990).

Discussion We were interested in chlorproguanil//dapsone be-

cause (i) it probably exerts less selection pressure for resistance than pyrimethamineisulfadoxine (WATKINS & MOSOBO, 1993), (ii) it is comparable to pyrimeth- amine/sulfadoxine in terms of cost and clinical efficacy (E. Amukoye et al., paper in preparation), and (iii) it will probably retain efficacy as pyrimethamine/sulfadoxine failure becomes more prevalent-for which data are beginning to emerge (TRIGG et al., 1996). In this paper we sought to provide some of the data needed for dose optimization.

Dapsone was absorbed slowly, reaching C,,, as late as 24 h after dosing. Cm,, was probably lower than pre- viously reported in healthy adults: EDSTEIN et al. (1990) reported a mean C,,, of 1.13 WmL at the end of 8 weekly doses of 100 mg (about 1.4 mg/kg), compared with 534 ng/mL after the single dose of 2.4 mg/kg which we used. As a result of these low levels, our estimates of Vd/F and CUF were higher than previously calculated in adults-l.2 L/kg and 0.63 mL’min/kg respectively in one study (EDSTEIN et al., 1990). These differences may

be functions of age, malaria, or race. However, it is also possible that the low dapsone levels resulted from redu- ced bioavailability (MIROCHNICK et d., 1993). The tli, of dapsone in the present study was 24.5 h, similar to previous estimates (CARR et al., 1978; ZUIDEMA et al., 1986; EDSTEIN et al., 1990; MIROCHNICK et al., 1993); this was slow enough to allow dapsone to accumulate with successive doses. No ‘slow acetylator’ was identi- fied in this study (CARR et al., 1978).

Chlorproguanil was absorbed more rapidly than dapsone: C max, tmax and tl,, were similar to those reported elsewhere (EDSTEIN & VEENENDAAL, 1987; VEENENDAAL et al., 1988; PETERSEN et al., 1990). The bioavailability of chlorproguanil is unknown but, if it is close to 1.0, the drug has a large volume of distribution. Chlorproguanil was rapidly metabolized to chlor- cycloguanil. As noted previously (EDSTEIN & VEENENDAAL, 1987; VEENENDAAL et al., 1988; PE- TERSEN et al.,, 1990), chlorcycloguanil is difficult to detect-especially with the small blood sample volumes which are inescapable when studying young children. Chlorcycloguanil was undetectable at any time in 4 of the 12 patients (including the one child with undetecta- ble chlorproguanil). The ratio of AUCo-24 values for chlorcycloguanil to those for chlorproguanil was 0.37: 1; consequently all those in whom chlorcycloguanil could be measured were ‘extensive metabolizers’ (WARD et al., 1989).

The present parasite concentration versus time curves were indistinguishable from those for pyrimethamine/ sulfadoxine (WINSTANLEY et al., 1992), and all 12 chil- dren recovered, but this does not prove that the chlor- proguaniVdapsone regimen was optimal. Optimization would be facilitated if we knew the therapeutic range and the length of time for which therapeutic concentra- tions must be maintained. Drug levels derived from es- timates of potency obtained in vitro (WINSTANLEY et aZ., 1995) cannot be used directly as a therapeutic range, one of the main limitations being the difference in un- bound drug fractions between human plasma and cul- ture media. Measurement of unbound concentrations in vitro and in T&O probably allows a reasonable estimate of therapeutic range to be made (WINSTANLEY et al., 1993), since the unbound concentration of most drugs at steady state is in equilibrium between the plasma and tissues-including erythrocytes (ROWLAND & TOZER, 1995). The present study showed that chlorcycloguanil and dapsone do not bind avidly to plasma proteins, but that chlorproguanil binds to both al-acid glycoprotein and albumin. The maximum unbound concentrations of chlorproguanil, chlorcycloguanil and dapsone after the first dose of the drug combination can be calculated

326 PETERWINSTANLEY ETAL.

to have been 28, 19 and 160 ng/mL, respectively (about 90, 60 and 645 no, respectively).

In vitro, against a pyrimethamine-resistant strain ofZ? falciparum, the 50% inhibitory concentration values (IC& of chlorcycloguanil and dapsone (alone) were 4.8 nM and 5 1 p, respectively (WINSTANLEY et al., 1995); in the present study we have shown that the fractions of each drug unbound in the culture medium were 0.87 and 0.77, respectively. The unbound concentrations of chlorcycloguanil and dapsone, at their ICsos, were therefore about 4 no and 40 C(M, respectively. However, when used in combination, chlorcycloguanil and dap- sone exhibit marked synergy and, in our previous work, the minimum fractional inhibitory concentrations (MFIC) were found to be 0.535 and 80.6 no, respec- tively (WINSTANLEY et al., 1995). The unbound con- centrations of chlorcycloguanil and dapsone at their MFIC were therefore about 0.5 and 62 no. resoective- ly, and the maximum unbound concentrahons*of each achieved in viva were therefore about 120-fold and lo- fold higher. The MFIC of dapsone was exceeded in these 12 children for about 240 h; we cannot calculate this duration for chlorcycloguanil because it could not be measured for long enough. If the tliz of chlorcyc- loguanil is similar to that of chlorproguanil, as seems likely (WATKINS et al., 1987), then MFIC was exceeded for no longer than 144 h. It is elimination of chlorcyc- loguanil which determines the duration of effective syn- ergy, and in some children this might be unacceptably short with the current regimen.

Dapsone causes dose-related haemolysis and methaemoglobinaemia, but its most worrying adverse effects, on skin and bone marrow, are idiosyncratic rather than clearly dose-related (DOLLBRY, 1991). In contrast, chlorproguanil rarely causes idiosyncratic toxicity but, like other dihydrofolate reductase inhibitors, its metabolite chlorcycloguanil causes concentration-related inhibition of human bone marrow cell replication: in a model in vitro, the IC50 of chlorcycloguanil against granulocyte colony forming units from normal human marrow was 9.6 m (WINSTANLEY et al., 1995). In the present study we found that chlorcycloguanil was 100% unbound in the medium previously used for bone marrow culture. The maximum unbound chlorcycloguanil concentration achieved in the clinical study was therefore 160-fold lower than the IC50 in vitro against bone marrow cells. This is reassuring, and suggests that there is room to increase the chlorproguanil dose if required. Of course, whole-body toxicity cannot be deduced from cell culture experiments, and a rising dose tolerance study is currently in progress in which daily chlorproguanil doses of up to 3.6 mg/kg will be examined (in combination with 2.4 mg/kg of dapsone).

We did not expect the current study to give definitive guidance on the need to change the chlorproguanill dapsone dose schedule-this will be decided mainly by recrudescence rates in the clinical trial (E. Amukoye et al., paper in preparation). This study has, however, shown that, after current doses, (i) the maximum un- bound chlorcycloguanil concentration is 2 orders of magnitude above MFIC, while falling 2 orders of mag- nitude below its IC50 against bone marrow cells and (ii) the maximum unbound dapsone concentration is lo- fold higher than MFIC, but (iii) the MFIC of unbound chlorcycloguanil/dapsone may be exceeded for no long- er than 6 d.

If results from the concomitant randomized trial sug- gest that the clinical response is unsatisfactory, chlorcy- cloguanil/dapsone concentrations could be maintained above MFIC longer in 3 ways: (i) giving further doses of the present combination (which may reduce compli- ance), (ii) formulating the present combination for ex- tremely sustained release (which may be technically impossible and expensive), or (iii) increasing drug dos- es. The present data suggest that chlorproguanil doses

could probably be increased; in contrast, dapsone is dangerous in overdose, causing dose-related haemolysis and methaemoglobinaemia, and achieves acceptable unbound concentrations at the current dose. It seems wise to fix the dapsone component of this combination for the time being, while investigating the tolerability of higher chlorproguanil doses.

Acknowledgements We thank the Director of KEMRI for permission to publish

these results. The study would have been impossible without the enthusiastic help of nurses, technicians and support staff at Kilifi. The work was supported by the UNDPNCTorld Bank/ WHO Snecial Proaramme for Research and Training in Tron- ical Diseases (TD@, the Wellcome Trust, and KEMRI. I’.*. and S.S. are grateful to SmithKline Beecham for financial sup- port; K.M. is a Wellcome Trust Senior Research Fellow in clinical science; W.W. and D.M. are grateful to the Wellcome Trust for personal and project support.

References BiGrkman, A. & Phillins-Howard. P. A. (199 1). Adverse reac-

‘tions to-sulfa drugs:-implicatio& for malaria chemotherapy. Bulletin of the World Health Organization, 69,297-304.

Bloland, l? B., Lackritz, E.. Kazembe, P.. Were 1. B. O., Steket- ee, R..& Campbell, 6. 6. (1993). Beiond chloroq&ne: im- plications of drug resistance for evaluating malaria therapy efficacy and treatment policy. Journal of Infectious Diseases, 167,932-937.

Carr, K., Oates, J. A., Nies, A. S. & Woosley, R. L. (1978). Si- multaneous analvsis of dapsone and monoacetvl dapsone employing high p&forma&e liquid chromatography: a-rapid method for determination of acetylator phenotype. British Journal of Clinical Pharmacology, 6,42 1-427.

Dollerv, C. (editor) (1991). Dapsone. Therabeutic Drugs. Edin- bur& etc..: Chu&ll Lkingstone, pp. Dll-D17. -

Edstein, M. D. &Veenendaal, J. R. (1987). Chlorproguanil and chlorcycloguanil concentrarions in human plasma and urine after Lapudrine@ administration. Transactions of the Royal Society of Tropical Medicine and Hygiene, 81, 136-139.

Edstein! M. D., Rieckmann, K. H. &Veenendaal, J. R. (1990). Multiple-dose pharmacokinetics and in vitro antimalarial ac- tivity of dapsone plus pyrimethamine (Maloprim@) in man. BritishJournal of Clinical Pharmacology, 30, 259-265.

Foster, S. D. (1991). Pricing, distribution and use of antima- larial drugs. Bulletin of the World Health Organization, 69, 349-363.

Helsby, N. A., Ward, S. A., Howells, R. E. & Breckenridge, A. M. (1990). In vitro metabolism of the biguanide antimalar- ials in human liver microsomes: evidence for a role of the mephenytoin hydroxylase (P450MP) enzyme. British Jour- nal of Clinical Pharmacology, 30,287-291.

Jones, C. & Ovenell, S. M. (1979). Determination of plasmsa concentrations of dapsone, mono-acetyl dapsone and py- rimethamine in human subjects dosed with MaloprimB. journal of Chromatoeraahv. 163. 179-185.

M&, D. G:, Porter, JYA:, cetrecht, J. I?, Wilkinson, G. R. & Branch, R. A. (19901. The contribution of N-hvdroxvlation and ac&vlatio& to d&one nharmacokinetics innormal sub- jects. &&al I’harmdcology’and Therapeutics, 48, 619-627.

Mirochnick, M., Michaels, M., Clarke, D., Brena, A., Regan, A. & Pelton. S. (1993). Pharmacokinetics of dansone in chil- dren. Jo&al of‘Pedia&ics, 122,806-809. A

Pasvol, G., Newton, C. R. J. C., Winstanley, I? A., Watkins, W. M., Peshu, N. M., Were, J. B. O., Marsh, K. & Warrell, D. A. (1992). Quinine treatment of severe falciparum malaria in African children; a randomized trial of three regimens. American Journal of Tropical Medicine and Hygiene, 45, 702-713.

Petersen, E., Flachs, H., Hogh, B., Hanson, A. P., BjSrkman, A. & Hvidberg, E. F. (1990). Plasma, erythrocyte and urine concentrations of chlorproguanil and two metabolites in man after different doses. Journal of Tropical Medicine and Hygiene, 94, 199-205.

Phillips-Howard, P. A. & Bjiirkman, A. (1990). Ascertainment of risk of serious adverse reactions associated with chemo- Drophvlactic antimalarial drugs. Bulletin of the World Health br~anization, 68,493-504. -

Radloff, P. D., Phillips, J., Nkeyi, M., Hutchinson, D. & Krem- sner, P. G. (1996). Atovaquone and proguanil for Plasmodi- urn falciparum malaria. Lance& 347, 1511-1514.

Rowland, M. & Tozer, T. N. (1995). Clinical Pharmacokinetics: Concepts andApplications. Baltimore etc.: Williams &Wilkins.

Sudre, P., Breman, J. G., McFarland, D. & Koplan, J. P.

PHARMACOKINETICS OF CHLOPROGUANIL’DAPSONE

(1992). Treatment of chloroquine-resistant malaria in Afri- can children: a cost-effectiveness analysis. ZnternationalJour- nal of Epidemiology, 21, 146-154.

Trigg, J. K., Mbwana, H., Chambo, O., Hills, E., Curtis, J., Is- sifou, S., Durasingh, M. & Curtis, C. F. (1996). Resistance in Tanzanian Plasmodium falciperum to pyrimethamine-sulf- adoxine and a test of chlorproguanil-dapsone. Annals of Tropical Medicine and Parasitology, 90,408 (abstract).

Veenendaal, J. R., Edstein, M. D. & Rieckmann, K. H. (1988). Pharmacokinetics of chlorproguanil in man after a single oral dose of Iapudrine@. Chemotherapy, 34,275-283.

Ward, S. A., Watkins, W. M. & Mberu, E. (1989). Intersubject variability in the metabolism of proguanil to the active me- tabolite cycloguanil in man. British Journal of Clinical Phar- macology, 27,781-787.

Watkins, W. M. & Mosobo, M. (1993). Treatment of Plasmodi- urn falciparum malaria with pyrimethamine-sulphadoxine: selective pressure for resistance is a function of long elimina- tion half-life. Transactions of the Royal Society of Tropical Med- icine and Hygiene, 87,75-78.

Watkins, W. M., Chulay, J. D., Sixsmith, D. G., Spencer, H. C. & Howells, R. E. (1987). A preliminary pharmacokinetic study of the antimalarial drugs proguanil and chlorpro- guanil.Journal of Pharmacy and Pharmacology, 39,261-269.

Watkins, W. M., Brandling-Bennett, A. D., Nevill, C. G., Cart- er, J. Y., Boriga, D. A., Howells, R. E. & Koech, D. K. (1988). Chlorproguanilidapsone for the treatment of non- severe Plasmodium falciparum malaria in Kenya: a pilot study. Transactions of the Royal Society of Tropical Medicine and Hy- giene, 82, 398-403.

WHO (1990). Severe and complicated malaria, 2nd edition. Transactions of the Royal Society of Tropical Medicine and Hy- giene, 84, supplement 2.

321

WHO (1995). Twelfth Programme Report of the UNDP-World Bank-WHO Special Programme for Research and Training in Tropical Diseases (TDR). Geneva: World Health Organiza- tion.

Winstanley, l? A. (1996). Pyronaridine-a promising drug for Africa? Lancet, 341,2-3.

Winstanley, P. A., Watkins, W. M., Newton, C. R. J. C., Nevill, C., Mberu, E., Warn, I? A., Waruiru, C. M., Mwangi, I. N., Warrell, D. A. & Marsh, K. (1992). The disposition of oral and intramuscular pyrimethamine/sulfadoxine in Kenyan children with high parasitaemia but clinical non-severe falci- parum malaria. British Journal of Clinical Pharmacology, 33, 143-148.

Winstanley, P., Newton, C., Watkins, W., Mberu, E., Ward, S., Warn, P., Mwangi, I., Waruiru, C., Pasvol, G., Warrell, D. & Marsh, K. (1993). Towards optimal regimens of parenteral quinine for young African children with cerebral malaria: the importance of unbound quinine concentration. Transactions of the Royal Society of Tropical Medicine and Hygiene, 87, 201-206.

Winstanley, P. A., Mberu, E. K., Szwandt, I. S. F., Brecken- ridge, A. M. &Watkins, W. M. (1995). The in vitro activity of novel antifolate drug combinations against Plasmodium falci- parum and human granulocyte colony-forming-units. Anti- microbialAgents and Chemotherapy, 39, 948-952.

Zuidema, J., Hilbers-Modderman, E. & Merkus, F. (1986). Clinical pharmacokinetics of dapsone. Clinical Pharmacoki- netics, 11, 299-315.

Received 5 August 1996; revised 10 October 1996; accepted for publication 10 October 1996

( Announcement (

The Tropical Health and Education Trust

Fellows of the Society have always been actively involved in many tropical countries in establishing and devel- oping medical schools and other training institutions. But some of these schools, particularly in poorer African countries, face severe hardships. Students have no books, there is no foreign exchange for journals, equipment lacks spares, research cannot be supported and external aid is directed towards primary health care.

The Tropical Health Education Trust has started to relieve, with support from many individuals, Trusts and organizations, some of these disadvantages.

Basic books have been sent to all the rural hospitals in two African countries, sets of books have been given for students in a number of others. Links between medical schools overseas and home departments have been started with fellowships for students in training and research methods also.

The Tropical Health and Education Trust aims to extend support like this to more countries, hospitals, med- ical schools and students and needs funds to do it: Fellows of the Society who would like to take this opportunity to help our colleagues overcome some of their obstacles can do so through a single gift, a four-year or a deposited covenant, or even through a legacy.

Trustees include: R. M. Anderson, K. l? W. J. McAdam, E. H. 0. Parry (Chairman), D. A. Warrell. For more information about THET please write or telephone: 2 1 Edenhurst Avenue, Fulham, London, SW6

3PD, UK; phone +44 (0) 171 927 2411, fax +44 (0) 171 637 4314.