Am. J. Trop. Med. Hvg.. 50(5), 1994,pp. 64I—645 Copyright C 1994 by The American Society of Tropical Medicine and Hygiene

PROGUANIL PLUS SULFAMETHOXAZOLE IS NOT CAUSALLY PROPHYLACTIC IN THE MACACA MUL47TA—PLASMODIUM CYNOMOLGI MODEL

G. DENNIS SHANKS, MICHAEL D. EDSTEIN. A. LEE CHEDESTER, C. DAHLEM SMITH, KEVIN D. CORCORAN, MONTIP NGAMPOCHJANA, PRANEE HANSUKJARIYA. JETSUMON SATrABONGKOT, ANDH. KYLE WEBSTER Departments of Medicine, immunology, Veterinary Medicine, and Entomology. Armed Forces Research institute of Medical Sciences, Bangkok. Thailand

Abstract. New drugs for causal prophylaxis of malaria are needed. A proguan il/sulfamethoxazole combination was investigated using a rhesus monkey model (Macaca mulaiia infected with Plasmodium cynomolgi) to determine whether causal prophylaxis could be achieved. When a five-day regimen of proguanil (40 mg/kg/day) combined with sulfamethoxazole (100 mg/kg/day) was used, infection of all animals (6 of 6) was observed, with an extended prepatent period (median 40 days). Two control animals became infected on days 9 and 23 following sporozoite inoculation. Plasma concentrations indicated that proguanil and sulfamethoxazole were adequately absorbed and metabolized to cycloguanil and N4-acetylsulfamethoxazole, respectively. Analysis of liver biopsy specimens demon strated that the drugs were present two days following sporozoite inoculation but were not detectable one week later. Proguanil plus sulfamethoxazole does not eliminate exoerythro cytic-stage parasites in the rhesus monkey—P. cynomolgi model.

Antifolates have been popular chemoprophylac ia-naive rhesus monkeys of either sex weighing tic agents against malaria because of the infre between 2 and 5 kg were infected intravenously quency of side effects, low cost, and good com with 0.5—1 x 10―P. cynomolgi sporozoites on pliance. Although resistance to single agents such day 0. Proguanil (40 mg/kg/day) plus sulfame as pyrimethamine and proguanil has made antifol thoxazole (100 mg/kg/day) was given daily on ates ineffective in many areas, proguanil combi days —2,—1, 0, + 1, and +2. Daily blood smears nations are often recommended for causal prophy were used to detect parasitemia until 100 days. laxis in areas of malaria drug resistance.'2 at which time negative blood smears were as Proguanil plus sulfamethoxazole has recently been sessed to indicate successful prophylaxis. Un shown to provide effective prophylactic protection treated control animals were used to insure the in human field trials, with marked synergy be infectivity of the sporozoites and estimate the tween the components.34 The problem of relapsing prepatent period. in Plasmodium vivax infections following the end Animals were housed as a closed colony in of doxcycline chemoprophylaxis (up to 30% in double-screened, open-air ventilated animal fection within one month after ending doxycy dine) was not seen with proguanil plus sulfame rooms. They were fed a standard monkey thoxazole.5 It seemed possible that a proguanil chow, supplemented three times a week with combination might have increased causal prophy approximately 100 g of fresh fruit. Water was lactic activity against relapsing forms of malaria. freely available via automatic watering devices. The only established primate model of causal Animals were sedated intramuscularly with ke prophylaxis, Macaca mulatta infected with P. cy tamine hydrochloride (5—10mg/kg) during na nomolgi, was used to test this hypothesis. This sogastric administration of test drugs, liver bi simian parasite is biologically related to P. vivax opsies. intravenous sporozoite injection, and in its relapse patterns and its response to standard blood sample collections. Daily blood smears antimalarial drugs.6 were made without sedation from the margin of the ear using sterile, disposable, pediatric skin MATERIALS AND METHODS lancets. After the conclusion of each set of ex Animal experiments periments, animals were cleared of any remain The methods of Schmidt and others were used ing parasites by the administration of chloro for the M. mulaiia—P. cynomolgi model.―Malar quine and primaquine. Curative treatment was 641 642 SHANKS AND OTHERS

TABLE I stored at —70°Cuntil analysis. When testing Summary of parasitemia data for Macaca mulatta proguanil plus sulfamethoxazole, open liver bi monkeys infected with Plasmodium cynomolgi fol opsies were done on one control and three mon lowing multiple dosing with proguanil (PRO) plus sulfamethoxazole (SMX) keys that had received the drug combination on day 2 postinfection and the remaining control blood and three other monkeys that had received the smear was first drug combination on day 9 postinfection. Liver positiveDA312Monkey Drug'Drug daystDay specimens were dissected to a wet weight of ap 4OPRO/IOOSMX proximately 250 mg and stored at —70°C. DA 3 17 40 PRO/l00 SMX —2—+2 28 DA 3 19 40 PRO/I 00 SMX —2—+2 40 DA 327 40 PRO/l00 SMX —2-+2 40 Drug analysis DA330 4OPRO/100SMX -2-+2 44 DA355 4OPRO/IOOSMX —2-+240 38 Plasma and liver concentrations of proguanil DA 320 None 9 and sulfamethoxazole and their principal metab DA23K347 None—2-+2 olites, cycloguanil and N4-acetylsulfamethoxa values are in mg/kg. t Days are numbered from day 0 (sporozoIc inoculation). zole, were measured by high-performance liquid chromatography. The drugs were analyzed on a Waters 840 chromatographic system (Millipore verified by consistently negative blood smears Waters, Milford, MA) consisting of a Model 510 for 100 days after the last treatment. Routine pump, Model U6K injector, and a Lambda-Max preventive medicine and health assessment pro Model 41 8 spectrophotometer. Chromatographic cedures for all nonhuman primates in the col separation of proguanil, cycloguanil, and the in ony included semiannual physical examinations ternal standard chlorcycloguanil were performed and tuberculin skin testing. This research was on a Spherisorb-5-Phenyl column (particle size conducted according to the principles enunci 5 pm, 150 x 4.6 mm internal diameter; HPLC ated in the Guidefor the Care and Use of Lab Technology, Cheshire, UK). The mobile phase oraiory Animals, Institute of Laboratory Ani consisted of methanol and water (29:7 1 v/v) mal Research, National Research Council, NIH containing 5 mM low ultraviolet pentane-sulfon Pub. No. 85-23, and applicable Federal laws ic acid (Millipore-Waters). Absorbance was and regulations. measured at 238 nm. The run time was less than Drugs used included proguanil hydrochloride 17.5 mm. Plasma samples (0.5 ml) containing (Paludrine@; Imperial Chemical Industries, 200 ng of chlorcycloguanil were prepared for London, UK) and sulfamethoxazole (Ganitol@; solid-phase extraction based on the method of Roche Laboratories, Nutley, NJ). Tablets of Taylor and others.7 The interassay and intraassay proguanil and sulfamethoxazole were crushed coefficients of variation for both proguanil and to a fine powder using a mortar and pestle. cycloguanil at 10 ng/ml were less than 10% (n Drugs were individually weighed on an analyt = 6). Liver samples were homogenized with an ical scale and maintained in a dry sealed vial equal volume of water and then ultrasonicated. until just prior to administration. Drugs were Liver suspensions (100 p1) were precipitated suspended in approximately 25 ml of distilled with methanol (I ml), centrifuged, and the su water not more than 30 mm prior to adminis pernatant (1 ml) was subjected to solid-phase tration. Placement of a size 8 french disposable extraction. The intraassay coefficients of va.ria infant nasogastric feeding tube for drug admin tion for proguanil and cycloguanil at 200 ng/100 istration was verified by aspiration of gastric p1 of liver suspension were 4.6% and 4.5%, re contents through the tube or administration of spectively(n = 4). 1—2ml of distilled water through the catheter Sulfamethoxazole, N4-acetylsulfamethoxazo without subsequent coughing by the animal. le, and the internal standard sulfalene were sep Following administration of the drugs, the feed arated on a Nova-Pak C I 8 column (particle size ing tube was flushed with 15 ml of distilled 4 p.m. 150 x 3.9 mm internal diameter; Milli water. pore-Waters). The mobile phase consisted of Blood samples for plasma drug determina acetonitrile, water, and glacial acetic acid (18.5: tions were obtained by venipuncture. After sep 8 1:0.5, v/v/v). Absorbance was measured at 268 aration from red blood cells, the plasma was nm. The total run time was less than I2 mm. PROGUANIL PLUS SULFAMETHOXAZOLE CHEMOPROPHYLAXIS 643

400

A @PRO 300

@2OO

U

-2 ‘-I 0 I 2 3 4 5 6 DAY

L 4 4, + DRUG DOSING

FIGURE 1 . Minimum plasma concentration-time profiles (mean ± SEM) of proguanil (PRO) and cycloguanil (CYC) in rhesus monkeys (n = 6) infected with Plasmodium cvnomolgi. Arrows indicate the days of drug administration.

Plasma samples (200 p1) containing sulfalene (5 parasitemia from 28 to 44 days (median 40) fol pg) were extracted using liquid-liquid extraction lowing sporozoite inoculation (Table I). The at pH 5.6 based on the method of Edstein and control monkeys who received no drug were others for sulfadoxine analysis.― The interassay first parasitemic on days 9 and 23. There was no and intraassay coefficients of variation for sul obvious explanation for the unusually long pre famethoxazole (1 pg/200 p1) and N4-acetylsul patent period in one of the control monkeys. The famethoxazole (0.25 pg/200 p.1) were less than usual prepatent period is eight days.9 10% (n = 5). The detection limit was 20 ng/200 The mean minimum plasma concentrations p.1 for both compounds. Liver samples were ho (drawn just prior to the next dose) of proguanil mogenized in an equal volume of water, ultra and cycloguanil are shown in Figure 1, which sonicated, and this suspension (200 p.1) was cx demonstrates little or no drug present after day tracted as described for plasma analysis. The 5. A similar result is seen in Figure 2, which intraassay coefficients of variation for sulfame shows the mean minimum plasma concentra thoxazole (0.5 p.g) and N4 acetylsulfamethoxa tions of sulfamethoxazole and N4-acetylsulfa zole (0. 125 jig) were 2.8% and 4.8%, respec methoxazole. The liver concentrations (mean ± lively (n = 4). SD, n = 3) were 49.0 ± 53.3 pg/gm for pro guanil, 15.3 ±4.7 pg/gm for cycloguanil, 15.9

RESULTS ±8.6 pg/gm for sulfamethoxazole, and 24.1 ± 4.5 pg/gm for N4-acetylsulfamethoxazole. No Monkeys receiving proguanil plus sulfame drug was detected in the livers of three drug thoxazole for five days had their first detectable treated monkeys on day 9. 644 SHANKS AND OTHERS

200 •SMX o N4SMX

I50@

‘@ 100

S's OK

50@

0 I I I —3 —2 —I 0 I 2 3 4 56 DAY

DRUG DOSING

FIGURE 2. Minimum plasma concentration-time profiles (mean ± SEM) of sulfamethoxazole (SMX) and N4- acetylsulfamethoxazole (NJSMX) in rhesus monkeys (n —6) infected with Plasmodium cynomolgi. Arrows indicate the days of drug administration.

DISCUSSION The pharmacologic data presented can be ap proximately compared with previous studies in With the M. mulaiia—P. cvnomolgi model humans. Proguanil has been reported to be ab used, we found no evidence of causal prophy sorbed in humans and extensively metabolized lactic activity using the combination of proguan to cycloguanil, the triazine metabolite that is re ii plus sulfamethoxazole. Previous work with 40 sponsible for the antimalarial activity.'2 Rhesus mg/kg of proguanil given for three days showed monkeys appear to convert more proguanil to a delay in the onset of patency (mean 36 days) cycloguanil than that found in humans with a in three of three monkeys.9 The delayed prepa cycloguanil:proguanil concentration ratio greater tent period without protection may be explained than I .0, compared with less than 1.0 in the ma on the basis of two different populations of jority of humans.'3 The plasma elimination half schizonts in the exoerythrocytic cycle as previ life of cycloguanil in rhesus monkeys (9.3 hr) is ously proposed by Bray.'° The synergistic com shorter than that reported for humans (I 5. 1 hr).'4 bination of proguanil plus sulfamethoxazole was The elimination half-life of sulfamethoxazole, inadequate to eliminate these tissue schizonts, however, is similar in the rhesus monkey and which do appear to have some antifolate sensi humans (9.2 hr versus I I .5 hr).'5 Both rhesus tivity.― One possible explanation might be that monkeys and man metabolize sulfamethoxazole hypnozoites are not metabolically active enough to its principal N4-acetylated metabolite. The to be killed by antifolates. Thus, the primary tis five-day drug regimen used did not produce sue schizonts that multiply rapidly after infec measurable concentrations of drug at a time tion are killed, but the hypnozoites that undergo when merozoites would be expected to be re multiplication later escape the antifolate action. leased from the liver. PROGUANIL PLUS SULFAMETHOXAZOLE CHEMOPROPHYLAXIS 645

The data presented here are in conflict with 2. Bradley Di, Phillips-Howard PA, 1989. Prophy some suggestions of causal prophylaxis from hu laxis against malaria for travellers from the United Kingdom. Br Med J 299: 1087—1089. man field trials.34 This could be explained by 3. Pang LW, Limsomwong N, Singharaj P. Canfield the intrinsic differences in the human and mon CJ. 1989. Malaria prophylaxis with proguanil key tests. Plasmodium vivax may have a differ and sulfisoxazole in children living in a malaria ent drug sensitivity pattern than P. cynomolgi. endemic area. Bull World Health Organ 67: In field trials with humans, subjects are gener 5 1—58. 4. Karwacki ii, Shanks GD, Limsomwong N, ally exposed to small intermittent sporozoite in Singharaj P. 1990. Proguanil-sulphonamide for oculations by mosquitoes during long periods of malaria prophylaxis. Trans R Soc Trop Med continuous drug administration. The M. mulai Hyg 84: 55—57. ia—P.cynomolgi model uses a large artificial 5. Pang LW, Limsomwong N, Boudreau E, Singharaj P, 1987. Doxycycline prophylaxis for sporozoite inoculation during a brief drug falciparum malaria. Lancet i: 1161—1164. course. Although the primate model system used 6. Schmidt LH, Fradkin R, Genther CS, Hughes HB. here was successful in screening 8-aminoquin 1982. III. Delineation of the potentials of pri oline drugs similar to primaquine, it has limits maquine as a radical curative and prophylactic drug. Am J Trop Med Hvg 31: 666—680. tions for the causal prophylactic assessment of 7. Taylor RB. Moody RR. Ochekpe NA, 1987. Dc antifolates.6 Care must be taken when interpret termination of proguanil and its metabolites cy ing data from animal models of malaria infection cloguanil and 4-chlorphenylbiguanide in plas to classes of drug for which they were not orig ma, whole blood and urine by high-performance inally developed. liquid chromatography. J Chromatogr 416: 394—399. 8. Edstein MD, Lika ID, Chongsuphajaisiddhi T, Acknowledgments: We thank Drs. David Davidson, Sabchareon A, Webster HK, 1991 . Quantitation Brian Schuster, and Wilbur Milhous for discussions and of Fansimef (mefloquine + sulfadoxine + pyr encouragement, and Dr. David Jacobus for supplying the proguanil and sulfamethoxazole. We also thank the imethamine) components in human plasma by two high-performance liquid chromatographic staff of the Department of Veterinary Medicine of the methods. Ther Drug Monit 13: 146—151. Armed Forces Research Institute of Medical Sciences for care of the animals in this study. Cycloguanil and 9. Schmidt LH, Fradkin R, Genther CS, Rossan RN. Squires W, 1982. II. Responses of sporozoite N4-acetylsulfamethoxazole were donated by Imperial induced and trophozoite-induced infections to Chemical Industries (London, UK) and Hoffrnann-La Roche (Basel, Switzerland), respectively. standard antimalarial drugs. Am J Trop Med Hvg 31: 646—665. Financial support: This work was supported by the Unit 10. Bray RS, 1984. The response of Plasmodium ed States Army Medical Research and Development vivax to antifols. Trans R Soc Trop Med Hvg Command. 78: 420—421. I I . Jiang JB. Bray RS, Krotowski WA, Canning EU, Disclaimer: The opinions or assertions contained in this Liang DS, Huang JC, Liao JY, Li DS, Lun ZR, paper are private views of the authors and are not to be Landau I, 1988. Observations on early and late construed as official or as reflecting the views of the U.S. Army. post-sporozoite tissue stages in primate malaria. Trans R Soc Trop Med Hyg 82: 56—58. Authors' addresses: G. Dennis Shanks and Michael D. 12. Carrington HC, Crowther AF, Davey DG, Levi Edstein, Army Malaria Research Unit, MILPO Ingle AA, Rose FL, 1951. A metabolite of paludrine burn 2174, New South Wales, Australia. A. Lee with high antimicrobial activity. Nature 168: Chedester, C. Dahlem Smith, Kevin D. Corcoran, Mon l080. tip Ngampochjana, and Prance Sattabongkot, Depart 13. Ward SA, Watkins WM, Mberu E, Saunders JE, ment of Veterinary Medicine, Armed Forces Research Koeck DK, Gilles HM, Howells RE, Brecken Institute of Medical Sciences, Bangkok, Thailand. Jet ridge AM, 1989. Inter-subject variability in the sumon Sattabongkot, Department of Entomology, metabolism of proguanil to the active metabolite Armed Forces Research Institute of Medical Sciences, cycloguanil in man. Br J C/in Pharmacol 27: Bangkok, Thailand. H. Kyle Webster, Becton Dickin 781—787. son, 2350 Qume Drive, San Jose, CA 95731. 14. Edstein MD, Veenendaal JR. Rieckmann KH, 1990. Multiple-dose kinetics in healthy volun teers and in vitro antimalarial activity of pro REFERENCES guanil plus dapsone. Chemotherapy 36: 169— I76. 1. Limsomwong N, Pang LW, Singharaj P. 1988. 15. Grose WE, Bodey GP, Loo TL, 1979. Clinical Malaria prophylaxis with proguanil in children pharmacology of intravenously administered living in a malaria-endemic area. Am J Trop tn methoprim-sulfamethoxazole. Antimicrob Med Hyg 38: 23 1—236. Agents Chemother 15: 447—451.