Efficacy of amodiaquine for uncomplicated Plasmodium falciparum malaria in Harper, Liberia

TRANSACTIONSOFTHEROYALSOCIETYOFTROPICALMEDICINEANDHYGIENE(2002)96,670-673

Efficacy of amodiaquine Harper, Liberia

for uncomplicated

Plasmodium

falciparum

malaria

in

F . Checchi”‘, S. Balkan’, B. T. Vonhm*, M. Massaquoi’, P. Bibersonl, I’. Eldin de Pecoulas3, I’. ‘Medecins Suns Frontieres, 8 rue Saint-Sabin, 75011 Paris, France; 2Malaria Brasseur4 and J.-P. Guthmann’ Control Program, Ministry of Health and Social Welfare, Monrovia, Liberia; 3Laboratoire de Parasitologic, Faculte de Pharmacie, Boulevard Becquerel, 14032 Caen, France; 4Laboratoire de Parasitologic, Faculte de Medecine-Pharmacie, 22 Boulevard Gambetta, 76183 Rouen, France; jEpicentre, 8 rue Saint-Sabin, 75011 Pari>, France Abstract In the face of spreading chloroquine and sulfadoxine-pyrimethamine (SP) resistance, amodiaquine remains a cheap and efficacious alternative for treating uncomplicated Plasmodium falcipamm malaria in many settings. In Harper, south-eastern Liberia, a previous study we conducted showed very high levels of resistance to both chloroquine and SF’. In 2001, in an effort to look for possible alternatives, we measured in the same setting the efficacy of amodiaquine in a 28-d study in vivo, with results corrected by polymerase chain reaction genotyping to distinguish recrudescences from reinfections. In total, 107 children were included in the study and received a 3-d supervised course of 25 mg/kg amodiaquine. Of these, 81 were analysable at day 28. The overall failure rate was 19.8% (95% CI 11.7-30.1%) considering both parasitological and clinical outcomes. These results provide hitherto missing data on amodiaquine in Liberia, and confirm that the drug may still be efficacious in settings where chloroquine and SF are failing. We recommend the introduction of amodiaquine in association with artesunate as a first-line antimalarial in Harper. Keywords: malaria, Plasmodium falciparum, chemotherapy, resistance, amodiaquine, Liberia Introduction In Africa, Plasmodium.falciparum resistance to chloroquine is widespread, and resistance to sulfadoxine(SP). which in some East African coun-uvrimethamine -. tries has already redlaced chloroquine as first-line therapy, is spreading rapidly (R@NN it al., 1996; BLOLAND, 2001: MUTABINGWA et al., 2001). We thus face a situation in which the drug t-hat was once the mainstay of antimalarial treatment throughout the continent is no longer efficacious, and its natural successor may have a very short therapeutic lifespan (WHITE et al., 1999). Among the few possible alternatives, amodiaquine is by far the most affordable: a full adult dose currently costs about US$O.lS (WHO, 2001a). Past use of this drug was limited due to reports of severe adverse reactions among travellers taking it as prophylaxis. Recently, however, safety and tolerability of the drug have been reviewed, showing a safety profile similar to that of chloroquine and SP (NEVILL et al., 1994; WHITE, 1996; STAEDI(E et al., 2001). In addition, evidence from various African settings shows that amodiaquine remains effective against l? falciparum even in areas where chloroquine fails (OLLIARO et al., 1996; BRASSEUR et al., 1999; GORISSEN et al., 2000; STAEDKE et al.. 2001). The WHO now recommends testing and use of amodiaquine as an alternative to chloroquine in areas of high antimalarial drug resistance (WHO, 2001a). In Harrier. Liberia. where the humanitarian oraanization Midecins Saris Front&es (MSF) has beenpresent since 1996, a previous study we conducted yielded parasitological failure rates in vivo for both chloroquine (74% at day 14) and SP (52% at day 28 with polymerase chain reaction [PCR] adjustment), respectively the first- and second-line therapies in Liberia. A related genotype analysis showed that 100% of pretreatment I? falciparum isolates carried mutations at gene pfcrt associated with chloroquine resistance, whereas 84% and 79% carried mutations associated with SP resistance at genes dhfr and dhps respectively (CHECCHI et al., 2002). The need for a change in antimalarial therapy was therefore obvious and in an attempt to explore

Address for correspondence: Jean-Paul Guthmann, Epicentre, 8 rue Saint-Sabin, 75011 Paris, France; phone +33 1 40 21 28 48, fax +33 140 il28 03, e-mail jguthmann@epicentre. msf.org

possible alternatives, we decided to evaluate the efficacy in vivo of amodiaquine. Materials and Methods Study site Harper is a city of 10000 located in south-eastern Liberia, near the border with the Ivory Coast, where most inhabitants fled during the 1989-96 civil war. JJ Dossen Memorial Hospital, located in Harper, is the only referral health care facility for a catchment area of some 90000 people. Malaria is the main health problem in the hospital, accounting for about one-third of consultations (400-450 cases per month), admissions and deaths. Among children under 5 years old, almost half of cases and deaths are attributed to malaria. Transmission appears stable throughout the year and all infections are caused by P. falciparum. Patients from both JJ Dossen Hospital and Sacred Heart Clinic participated in this study, with the approval and support of the Liberian Ministry of Health and Social Welfare. Study design Our methodology essentially followed the WHO guidelines for assessment of therapeutic efficacy of antimalarial drugs for uncomplicated P. falciparum malaria in areas with intense transmission (WHO, 1996). Accordingly, children aged 6-59 months referred to the studv were, if eligible for inclusion, treated on site with amodiaquine and followed-up for 28 d. Outcomes were both clinical and parasitological, with PCR genotyping methods used to distinguish reinfections from recrudescences due to therapeutic failure. Sample size We expected amodiaquine resistance to be 15%, to be detected with 7.5% precision. At a significance level of 5%, and with a power of 80%, we required 88 patients. Allowing for a 10% loss to follow-up, the final sample size was thus set at 97. Enrolment of study patients Clinicians at JJ Dossen Hospital and Sacred Heart Clinic were asked to refer to the study all children who had clinically suspected malaria. Referrals received a thorough clinical examination, and, in the laboratory, thick and thin blood films were obtained, haemoglobin was measured, and a capillary blood sample was collected on filter-paper for PCR genotyping. Children

AMODIAQUINEFORUNCOMPLICATEDFALCIl'ARUMMALARIA

were eligible for the study if they (i) were aged 6-59 months; (ii) had fever (axillary temperature 2 37.5 “C) or history of fever in the past 24 h; (iii) had a l? falciparum monoinfection with an asexual count 2000100 OOOipL; and (iv) were likely to complete 28 d of follow-up (living no more than 1 h away from the hosoitalj. Children were not eligible in case of (i) severe disease signs or signs of seiere malaria (WH& 1996); (ii) concomitant febrile illnesses; (iii) severe underlying condition; and (iv) history of allergy to amodiaquine. A history of previous antimalarial intake was not a reason for exclusion, though this information was recorded carefully. Written informed consent was obtained from parents or guardians of each study subject. Treatment

andfollow-up

Study children received amodiaquine (153 mg base tablets; Laboratoires Creat, Vernouillet, France) at a dosage of 25 mg/kg, divided equally over 3 d. All doses were supervised, and intake was observed for 30 min. Children who vomited the first dose were retreated. After the treatment period (days 0, 1, and 2) children were reassessed on days 3, 7, 14, and 28, and parents/ guardians were also instructed to bring their children on any other day irrespective of schedule if their health was of concern. At each follow-up visit, a complete clinical examination was made, and thick and thin blood films were obtained. In case of any recurrent parasitaemia after day 3, a second capillary blood sample was collected for PCR genotyping. Defaulters were visited at home, and considered as lost to followup after 2 unsuccessful visits. Patients were withdrawn from the study for any of the following reasons: (i) failure to take any dose of amodiaquine; (ii) severe allergic reaction to the study drug; (iii) onset of a nonmalaria febrile illness or other serious condition during follow-up; (iv) self-medication with antimalarials during follow-up; and (v) decision of parent or guardian. Therapeutic

outcomes

Patients were classified as early treatment failure (ETF) if they met any of the following criteria: (i) progression to severe malaria on days 1, 2, or 3; (ii) parasitaemia on day 2 higher than on day 0 in the presence of fever; (iii) any parasitaemia in the presence of fever on day 3; or (iv) day 3 parasitaemia 2 25% of the day 0 count, irrespective of fever. After day 3 patients with recurrent parasitaemia were classified as late clinical failure (LCF) if febrile, or late parasitological failure (LPF) if they remained afebrile until day 28. All others were classified as adequate treatment response (ATR), provided that they completed perprotocol follow-up. Rescue treatment (quinine hvdrochloride 10 mg/Gg/8 h for 5-7 d) was immediately administered to children who fulfilled the criteria of ETF or LCF. Children who were parasitaemic but asymptomatic after day 3 were visited at home daily by a health worker, in addition to scheduled visits, to monitor closely any onset of malaria symptoms. If they became symptomatic, these children were immediately treated with rescue medication. All children who were parasitaemic on day 28 received quinine irrespective of symptoms.

671

mann No. 3 filter-paper, air-dried, and stored in the dark in sealed bags at room temperature. PCR genotyping

technique

Capillary blood samples were analysed at the Laboratoire de Parasitologie, Facultt de Mkdecine-Pharmatie, Rouen according to a method previously described (BASCO & RINGWALD, 2000). A polymorphic region of the l? falciparum msp.2 gene was amplified bv nested PCR, and amplification products were run through an electroohoresis chamber. The band orofiles of oaired pretreitment and failure-day samples were thenA compared. Paired samples with different band profiles were classified as reinfections, and if the band profiles were similar they were classified as recrudescences. Similar profiles with additional or missing bands in either sample were also classified as a recrudescence: the difference in the number of bands was explained by selection of previously undetected clones, presence of a novel infection superimposed on the recrudescent one, or disappearance of drug-sensitive clones present in the pretreatment infection. Allele frequencies in our population were not studied because we considered that the nzsp2 locus used to differentiate recrudescence from reinfection has a large degree of polymorphism (FELGER et al., 1999). LTFs shown to be reinfections through PCR genotyping were reclassified as ATR. Data entry and analysis Data were double-entered on Epi-Info 6.04 software (CDC, Atlanta, GA, USA). The failure rate of amodiaquine was expressed as the total number of ETF, LCF, and LPF outcomes over the total number of patients followed-up according to protocol, with associated 95% confidence intervals. ‘Mean age’ and ‘proportion of children with a history of chloroquine intake’ among failures and adequate responders were compared using a Student’s t test and a x2 test respectively. Results Enrolment in the study lasted from March 7 to May 11 2001, and details are shown in the Figure. The main reasons for exclusion were parasitaemia below

234 excluded

14 withdrawn 7 acute resi%rarotv infecrion 1 meningitis 1 suspected AIDS 1 abscess 1 urinary infection 1 received antimalarial at home 1 mistakenly underdosed 1 consent withdrawn 5 lost to follow-up

Laborato y procedures

Thick and thin blood films were Giemsa-stained at 10% dilution for 13 min: asexual counts were obtained on the thick film by counting the number of trophozoites against 200 leucoc$es (white blood cells NCTBCl). assuming a normal blood level of 8000 tic/&. A sampg of 217 thick films (22.3% of total) were re-read by independent MSF technicians. Haemoglobin was measured using the Lovibond undiluted technique (Assistant Co., Sondheim Rhon, Germany). Samples for genomic analysis were collected on What-

+ 81 completed follow-up

28-d

Figure. Study enrolment details for 341 children with clinically suspected malaria, Harper, Liberia, 2001.

F. CHECCHI ETAL.

672

2OOO/uL (147 children), negative smear (61 children), and firesence of acute respiratory infection or otitis (40 children). Ten mixed or monoinfections with l? malariae were detected on day 0 (4% of all smear-positive referrals). Baseline characteristics of inclusions are shown in Table 1. Haemoglobin levels were relatively low in our sample, and a considerable proportion of patients had a history of antimalarial intake before inclusion. Parasitological and clinical outcomes for the 81 children who completed 28 d follow-up are shown in Table 2. PCR genotyping results were available for 28 out of the 33 late failures (2 samples were not collected and 3 results were not interpretable) and they revealed that 15 (54%) were actually reinfections, of which 3 occurred before day 14, 2 between day 14 and day 28, and 10 on day 28. Assuming that the 5 failures not genotyped were recrudescences, the day 28 PCR-adjusted failure rate thus becomes 19/81 (Table 2) considering both clinical and parasitological criteria, and 6181, (7.4%; 95% CI 2.8-15.4%) considering only clinical criteria. If the proportion of reinfections to recrudescences among the 28 late failures genotyped is applied to the 5 late failures not genotyped, we can assume that only 2 of the 5 were true recrudescences and the overall failure rate becomes 16/81, (19.8%; 95% CI 11*7-30.1%). When taken as a continuous variable, age appeared lower among treatment failures (mean = 22.5 months) (mean = 26.6 than among adequate responders months), but this difference was not significant (P = 0.36). No significant difference in failure rate was found between children with a history of chloroquine intake in the week prior to inclusion and those without a history of antimalarial administration (33% vs. 19%, P = 0.18). By day 2, all but 2 children (102/l 04) were afebrile (98.1%: 95% CI 93.2-99.8%). and bv dav 3 75190 (83.3%; 95% CI 73.7790.1%) had cleared their initial parasitaemia. All but one of 13 children with considerable anaemia on day 0 (haemoglobin < 8.0 a/dL) who did not receive iron or folic acid sunnleGents had improved their haemoglobin level by day 14 (median on day 0 = 7.3 g/dL and median on day 14 = 8.7 g/dL). Two of these children were still parasitaemic on day 14 because of reinfection.

No serious drug-related adverse events were noticed in children treated with amodiaquine in this study during the 28-d follow-up period. There were, however, 2 deaths in the study population. One 6-monthold girl child was referred to the in-patients department on day 3 with rising fever no longer attributable to malaria (all films after day 3 were negative): she was diagnosed for meningitis and expired on day 9. One 6month-old boy was given a home course of quinine on day 17, because, although he was still asymptomatic, his parasitaemia had increased from the day 14 level, and his parents wished to travel away from Harper. He was rushed back to the hospital on day 28 with severe malaria (the mother reported that the quinine course had not been completed), and died on day 31 of renal failure.

I

Table 1. Baseline beria, 2001

characteristics

Discussion The results of this study show that, in an area with alarming rates of resistance to both chloroquine and SP, amodiaquine remains a relatively efficacious alternative for antimalarial therapy for outpatients. To our knowledge, this is the first investigation of amodiaquine resistance carried out in Liberia in the past 15 years. In this study, the drug provided fast defervescence and parasitological clearance, and contributed to haematological recovery. The observed clinical and parasitological failure rate of 23.5% may be an overestimation, as it includes 5 failures that were not confirmed by PCR genotyping; 19.8% (derived by extrapolating PCR results to these 5 failures) seems like a more likely approximation. Amodiaquine recrudescences in this study occurred late (10 out of the 13 PCR-confirmed failures took place on day 28), and most remained asymptomatichuring the follow-up period. Amodiaauine has not been used in the official Liberian health *system at least for a decade according to local clinicians. Limited supplies are available locally in 3 private pharmacies, but according to these vendors sales of amodiaquine are very infrequent. In Harper, P. faZciparum strains with mutations at gene pfcrt conferring resistance to chloroquine were ubiquitous, and three-quarters of children treated with chloroquine had a treatment failure (CHECCHI et al., 2002). Though not confirmed on a genetic level, the relatively low extent of of 107 study children,

Harper,

Li-

Characteristic Mean (SD) age (months) Gender ratio (M/F) Mean (SD) weight/height as % of median Mean (SD) initial axillary temperature (“C) Geometric mean (range) initial parasitaemia (@L) Median (range) initial haemoglobin (g/dL) Reported antimalarial intake in past 7 d (%)

Table 2. Therapeutic response after polymerase chain reaction

of 81 study children adjustment, Harper, Failures before PCR

Outcome Early treatment failure Late clinical failure Late parasitological failure Failure (ETF + LCF + LPF) Adequate treatment response

n (%) l(1.2) 9 (11.1) 24 (29.6) 34 (42.0) 47 (58.0)

95% CI 0.0-6.7 5.2-20.0 20.0-40.8 31.1-53.5 46.5-68.9

22.8 0.98 95.6 37.8 10251 84;

(14.1) (53154) (8.6) (1.1)

(2 000-82 196) i369;;; 2.7) cl

at day 28, before Liberia, 2001

and

Failures after PCR n (%) 1 (1.2) 5 (6.2) 13” (16.0) 19 (23.5)

95% CI 0.0-6.7 2.0-13.8 8.8-25.9 14.8-34.2

62 (76.5) 65.8-85.2

PCR, polymerase chain reaction; 95% CI, 95% confidence interval. “Includes 5 late failures not genotyped and assumed to be recrudescences (see Results).

AMODIAQUINEFORUNCOMPLICATEDFALCIPARUMMALARIA resistance to amodiaquine in this study contradicts previous reports of cross-resistance between the 2 drugs (BASCO, 1991), and suggests that widespread chloroquine resistance in any setting should not discourage testing of amodiaquine. At the same time, given the intense Z? falciparmm transmission and resulting heavy drug pressure, it is likely that the efficacy of amodiaquine would gradually decline if introduced as monotherapy in Harper. A more long-term approach might be to combine it with another equally efficacious drug: it is agreed that combination therapy substantially decreases the risk that strains resistant to either drug will emerge (WHITE & OLLIARO, 1996; D'ALESSANDRO & BUTTIENS, 2001). In Harper, the high failure rate of SP precludes it as a possible combination drug. A combination of amodiaquine and artesunate, however, has been tested in various settings, including West Africa, with favourable results (WHO, 2001 b). Advantages of such combination therapy would include high cure rates, rapid relief of symptoms and reduction in parasite levels, and a possible decrease in community transmission of malaria given the gametocydal properties of artesunate. Indeed, the deployment of artemisinin-based combinations is recommended as one of the key malaria control strategies in the face of rapidly spreading resistance to monotherapies (WHITE etaZ.,1999). In summary, this study confirms that amodiaquine may be considered as an alternative first-line drug in an malaria is a major health area where l? falcipamm problem, and where both chloroquine and SP are failing. In Harper, we recommend its introduction as firstline antimalarial, in combination with artesunate. In the light of our findings, testing of amodiaquine should be intensified and integrated in efforts to document antimalarial efficacy in Liberia and neighbouring regions. However, given its known potential for causing serious side effects at the hepatic level (ORRELL et al., 2001), the safety of amodiaquine both as a single and combination drug should be continuously monitored. Acknowledgements

We are very grateful to all the children and families who participated in this study, and to the authorities of JJ Dossen Memorial Hospital and Sacred Heart Clinic for making it possible. We thank the study team for their excellent work (counsellor Wede Clarke, data collector Andrew Colley, nurses Jomah Kollie and Moses Gayflor, home visitors Robert Kun and John Weah, laboratory technician Alfred Nyuma, and nurse aide Agatha Satia), as well as local clinicians and hospital staff for their cooperation. On MSF’s side, Dr Martine Savourk provided daily supervision of the study-related clinical work, and Katherine Johnson was instrumental in setting up the laboratory facilities and performed some of the quality control of slides, along with Laurence Bonte and Laurence Flevaud. Thanks also to Dounia Bitar, Marc Gastellu and Brigitte Vasset (MSF) for support and advice during the study, and to Dominique Legros (Epicentre) for commentary and review of the manuscript. References

Basco, L. K. (199 1). Inefficacy of amodiaquine against chloroquine-resistant malaria. Lancet, 33, 1460. Basco, L. K. & Ringwald, P. (2.000). Molecular epidemiology of malaria in Yaounde,, Cameroon. VII. Analysis of recrudescence and reinfecnon in patients with uncomplicated falciparum malaria. American Journal of Tropical Medicine andHygiene, 63, 215-221. Bloland, P. B. (2001). Drug resistance in malaria. Geneva: World Health Organization, WHOICDSICSIUDRSI2001.4. Brasseur, P., Guiguemde, R., Diallo, S., Guiyedi, V., Kombi-

673 la, M., Ringwald, P. & Olliaro, P. (1999). Amodiaquine remains effective for treating uncomplicated malaria in West and Central Africa. Transactions of the Royal Society of Tropical Medicine and Hygiene, 93, 645-650.

Checchi, F., Durand, R., Balkan, S., Vonhm, B. T., Kollie, J. Z., Biberson, I’., Baron, E., Le Bras, J. & Guthmann, J.-P. (2002). High Plasmodium fulciparum resistance to chloroquine and sulfadoxine-pyrimethamine in Harper, Liberia: results in viuo and analysis of point mutations. Transactions of the Royal Society of Tropical Medicine and Hygiene, 96,

664-669. D’Alessandro, U. & Butt&s, H. (2001). History and importance of antimalarial drug resistance. Tropical Medicine and International Health. 6. 845-848.

Felger, I., Irion, A., Steiger, S. & Beck, H. P. (1999). Epidemiology of multiple Plasmodium falciparium infections. Genotypes of merozoite surface protein 2 of Plasmodium falciparum in Tanzania. Transactions of the Roval Society of ‘Tro$icalMedicine andHygiene, 93, supplement i, S113-Si19: Gorissen, E., Ashruf, G., Lamboo, M., Bennebroek, J., Gikunda, S., Mbaruku, G. & Kager, P. A. (2000). In viva efficacy study of amodiaquine and sulfadoxineipyrimethamine in Kibwezi, Kenya and Kiaoma, Tanzania. Tro&zl Medicine and Inteknatiokzl Health.>. 459-463. 1 Mutabingwa, T., Nzila, A., Mbei, b., Nduati, E., Winstanley, P., Hills, E. & Watkins, W. (2001). Chlorproguanildapsone for treatment of drug-resistant ialciparu-m malaria in Tanzania. Lancet. 3.58.1218-1223. Nevill, C. G., Verhoeff, F.‘H., Munafu, C. G., ten Howe, W. R., van der Kaay, H. J. & Were, J. B. (1994). A comparison of amodiaquine and chloroquine in the treatment therapy of falciuarum malaria in Kenva. East African Medical I7ournal._ 71. i67-170. Olliaio, P., Nevill, C., Le Bras, J., Ringwald, P., Mussano, P., Garner, P. & Brasseur, P. (1996). Systematic review of amodiaquine treatment in uncomplicated malaria. Lancet, 348, 1196-1201. Orrell, C., Taylor, W. R. J. & Olliaro, P. (2001). Acute asymptomatic hepatitis in a healthy normal volunteer exposed to 2 oral doses of amodiaquine and artesunate. Transactions of the Royal Society of Tropical Medicine and

Hygiene, 95, 517-518. Rann, A. M., Msangeni, H. A., Mhina, J., Wernsdorfer, W. H. & Bygbjerg, I. C. (1996). High level of resistance of Plasmodium falciparum to sulphadoxine-pyrimethamine in children in Tanzania. Transactions of the Royal Society of Tropical Medicine and Hygiene, 90, 179- 18 1. Staedke, S. G., Kamya, M. R., Dorsey, G., Gasasira, A., Ndeezi, G., Charlebois, E. D. & Rosenthal, P. J. (2001). Amodiaquine, sulfadoxineipyrimethamine, and combination therapy for treatment of uncomplicated falciparum malaria in Kampala, Uganda: a randomised trial. Lancet, 358,368-374. White, N. J. (1996). Can amodiaquine be resurrected? Lancet, 748. 1184. White, N. J. & Olliaro, P. (1996). Strategies for the prevention of antimalarial drug resistance: rationale for combination chemotherapy for n&aria. Parasitology Today, 12, 399-401. White, N. J., Nosten, F., Looareesuwan, S., Watkins, W. M., Marsh, K., Snow, R. W., Kokwaro, G., Ouma, J., Hien, T. T., Molyneux, M. E., Taylor, T. E., Newbold, C. I., Ruebush, T. K., Danis, M., Greenwood, B. M., Anderson, R. M. & Olliaro, P. (1999). Averting a malaria disaster. Lancet, 353, 1965-1967. WHO (1996). Assessment of therapeutic efficacy of antimalarial drugs for uncomplicatedfalciparum malaria in areas with intense transmission. Geneva: World Health Organization, WHO/

MAL’96.1077. WHO (2001a). The use of antimalarial drugs. Geneva: World Health Organization, WHO/CDSIRBiW2001.33. WHO (2001b). Antimalarial drug combination therapy. Geneva: World Health Organization, WHO/CDS/RBM/2001.35. Received 27 March acceptedforpublication

2002; revised 15 August 21 August 2002

2002;