the methyl ether of dihydroartemisinin, has been combined with benflumetol, a novel aryl amino alcohol

Transcription

1 Am. J. Trop. Med. Hyg., 58(5), 1998, pp Copyright 1998 by The American Society of Tropical Medicine and Hygiene A RANDOMIZED CONTROLLED TRIAL OF ARTEMETHER/BENFLUMETOL, A NEW ANTIMALARIAL AND PYRIMETHAMINE/SULFADOXINE IN THE TREATMENT OF UNCOMPLICATED FALCIPARUM MALARIA IN AFRICAN CHILDREN LORENZ VON SEIDLEIN, KALIFA BOJANG, PETER JONES, SHABBAR JAFFAR, MARGARET PINDER, STEVEN OBARO, TOM DOHERTY, MICHELLE HAYWOOD, GEORGES SNOUNOU, BEATRICE GEMPERLI, INSA GATHMANN, CATHERINE ROYCE, KEITH McADAM, AND BRIAN GREENWOOD Medical Research Council Laboratories, Fajara, Banjul, The Gambia; Department of Epidemiology and Population Sciences, and Department of Medical Parasitology, London School of Hygiene and Tropical Medicine, London, United Kingdom; Department of Infection and Tropical Medicine (Lister Unit), Imperial College School of Medicine Northwick Park Hospital, Harrow, Middlesex, United Kingdom; International Clinical Research, Novartis Pharma AG, Basel, Switzerland Abstract. We report here the results of a randomized double blind trial comparing coartemether (), a combination of artemether and benflumetol, with pyrimethamine/sulfadoxine (). Two hundred eighty-seven children 1 5 years of age with uncomplicated falciparum malaria were enrolled at two centers in The Gambia between July 199 and December 199. Following treatment, children were visited at home every 24 hr until a blood film free of asexual parasites was obtained. Genotyping of parasites was used to distinguish recrudescence from new infections. Three days after the start of treatment, 133 (1%) of the -treated children compared with 128 (93.4%) of children treated with had cleared their parasites (P.3). The day 15 cure rate was 93.3% for and 97.7% for (P.13). Within the third and fourth week after initiation of therapy, 2 children treated with and one of the -treated children returned with second malaria episodes (P.1). Genotyping suggested that the majority (19 of 23 [82.%]) of these second episodes were due to new infections, supporting the World Health Organization recommendation that longer follow-up is not relevant for the assessment of drug efficacy. At the two-week follow-up, 28.9% of the treated children but none of the -treated children carried gametocytes (P.1). This study showed that is safe in African children with acute uncomplicated falciparum malaria, clears parasites more rapidly than, and results in fewer gametocyte carriers. More frequent new infections within the third and fourth week following treatment with than treatment with are likely to be due to the short prophylactic effect of. Malaria is estimated to kill between 1.5 and 2.7 million people every year. An additional 3 5 million people suffer annually from malaria attacks but survive. 1 Nine of 1 cases occur in sub-saharan Africa, where malaria is responsible for more disability adjusted life years than any other disease. 2 The main burden of malaria falls on children less than five years of age. The most commonly used treatments, chloroquine and pyrimethamine/sulfadoxine (), have become less effective in the regions where they have been used the longest and are most widely available, such as Southeast Asia. Chloroquine remains useful in selected areas of sub-saharan Africa, but Malawi, Kenya, and Zambia have been forced to replace chloroquine with as a first-line treatment. However, the efficacy of is already compromised in some parts of Africa. High level resistance of Plasmodium falciparum to has been reported in Tanzania and lower levels of resistance have been recorded in other parts of Africa. 3, 4 Thus, alternative antimalarial drugs are needed urgently. Atovaquone in combination with proguanil is effective, but like halofantrine and mefloquine, it is too expensive to be used widely in resource poor areas of the world although a donation program is being planned. 5 Pyronaridine is highly effective, but has yet to be registered outside China. Therefore, great hope is placed on artemisinin derivatives, natural products found in the leafy portions of Artemisia annua (qinghao), a plant used by Chinese herbalists since 18 BC. 7 However, the use of short courses of artemisinin derivatives has been limited by their relatively short half-life, which is associated with recurrence of parasitemia following the clearance of P. falciparum. Thus, artemether, the methyl ether of dihydroartemisinin, has been combined with benflumetol, a novel aryl amino alcohol 38 (class II schizontocide) with a relatively slow onset but a half-life of 4 days, in an oral formulation known as. This drug is undergoing randomized controlled trials in Hainan China, in Bangkok, Thailand, along the Thailand-Myanmar border, in India, and in Ifakara, Tanzania. Following a successful safety trial of in The Gambia, 8 we conducted a randomized controlled trial that compared with. During the safety study of, it had been observed that up to one-third of the children returned during the four-week follow-up period with new infections of P. falciparum. A strategy was therefore needed to treat repeated infections, which were unlikely to be resistant to. To investigate the efficacy of in repeated malaria episodes, we treated children who returned with falciparum malaria with open label (i.e., the investigator and parent or guardian were aware which drug the child received) and followed their response. METHODS The trial design was a randomized, double-blind, controlled efficacy trial that compared with as a treatment for uncomplicated falciparum malaria. The trial was conducted in The Gambia at two centers, one in a semiurban coastal area and the other in a rural region. Children between one and five years of age were eligible for enrollment if they had a parasitemia of more than 5, P. falciparum parasites/ l, a history of fever, lived within 2 km of either trial center, and if a parent or guardian gave informed consent. Children who required parenteral treatment, had been treated within two weeks with, or had a hematocrit

2 RANDOMIZED CONTROLLED TRIAL OF 39 25%, malnutrition, or a chronic disease were excluded. Eligible children were allocated to receive or by computer-generated block randomization (block size: 4). Investigators and patients remained blinded to the treatment code throughout the study. An emergency code was available to the local safety monitor, but was not used during the trial. Children randomized to received one tablet (2 mg of artemether and 12 mg of benflumetol) at, 8, 24, and 48 hr if they weighed less than 15 kg and two tablets at, 8, 24, and 48 hr if they weighed 15 or more kg. Children in the group received half a tablet (12.5 mg of pyrimethamine and 25 mg of sulfadoxine) once if they weighed less than 15 kg and one tablet once if they weighed 15 or more kg, in accordance with Gambian government guidelines. Since and differ in size and color, each child received in addition to the active treatment a placebo with the appearance of the drug they were not randomized to receive (double-dummy technique). Children took the starting dose under direct supervision. Children who vomited the starting dose and replacement dose of their antimalarial treatment received chloroquine intramuscularly. Parents or guardians were asked to give their child paracetamol (15 mg/kg) every hr while he or she remained febrile. Children were discharged home and visited every 24 hr by a fieldworker who obtained a blood film at each visit. The fieldworker also recorded the axillary temperature with a digital thermometer, the use of antipyretics during the previous 24 hr, and any adverse experiences. Visits were continued until a blood film free of asexual P. falciparum had been collected. Children were examined again two weeks after treatment and a blood film was obtained. After the two week follow-up, parents or guardians were encouraged to return with their child if any illness was observed. Symptomatic children with P. falciparum parasitemia of 5,/ l or more were offered re-treatment with. Children whose parents or guardians gave consent received the same regimen and were followed as described earlier. The study was approved by the Medical Research Council (MRC) Laboratories Scientific Coordinating and the Gambia Government/MRC Ethical Committees. The primary trial endpoint was defined as the proportion of children with no detectable P. falciparum by day 4. Secondary endpoints were parasite clearance times, day 15 cure rate, and the number of repeat episodes of malaria within 29 days. Children were considered evaluable if no protocol violation was observed and the required blood films could be obtained. Sample size was calculated at 157 children in each study group (significance level 5%, power 8%) on the assumption that would give a 1% cure rate by day 4 and a 94% cure rate, as suggested by data obtained previously. Laboratory methods. Thick blood films were stained with Giemsa. The level of parasitemia was calculated on the basis of the number of parasites per 2 leukocytes on a thick film. If gametocytes were present, the count was extended to 1, leukocytes and if less than five gametocytes were detected, the count was extended to 2, leukocytes. Blood samples for genotyping were obtained from all children at first presentation and, if applicable, at a second presentation. The DNA was purified from the blood collected as described previously. 9 Briefly, parasites were recovered by centrifugation following lysis of the erythrocytes in.5% saponin, and the DNA was extracted with phenol/ chloroform and recovered by ethanol precipitation. A nested polymerase chain reaction (PCR) protocol was adopted for the analysis of three polymorphic genetic markers from P. falciparum: the three sequence families of the merozoite surface protein-1 (MSP-1) block 2 repeat region, the two sequence families of the MSP-2 repeat region, and the RII region of glutamate-rich protein (GLURP). The PCR products were stained with ethidium bromide and visualized by UV transillumination following electrophoresis on agarose gels in TBE buffer (1 mm Tris, 1 mm boric acid, 5 mm EDTA). Paired samples were run side by side to allow comparison of the pattern of allelic variants. A recrudescent infection was defined as one that showed a complete match in allelic size for all three genes between the first and second samples. If any clone of a polyclonal primary infection was detected during a second episode, it was considered a recrudescence. The hematocrit of the children was measured by microhematocrit centrifugation. Statistical methods. Normally distributed data were compared using the Student s t-test; homogeneity of variance was tested by the F-test. Discrete data were compared by Fisher s exact test or the chi-square test with Yates correction, as appropriate. Sample size was calculated using the conservative method of Casagrande. 1 RESULTS A total of 287 children were recruited into the study between July 199 and January The 144 children randomized to receive were similar to the 143 children treated with in gender, age, weight, and baseline clinical features (Table 1). Parasites of species other than P. falciparum were not detected. The parents or guardians of 57 (19.9%) of the children reported that antimalarial treatment had been given to their child during the previous three months; 34 children (11.8%) had received chloroquine, 1 (5.%), four (1.4%) an unknown antimalarial, and three (1.%) in combination with chloroquine. Prior use of antimalarials was distributed equally between the treatment groups. By day four, 17 children (5.9%) could not be evaluated because they were either lost to follow-up, withdrew their consent, did not comply with the regimen, failed to meet the inclusion criteria, or reported treatment with cotrimoxazole, a potential antimalarial (Figure 1). Two children vomited the starting and replacement dose and received chloroquine injections. One child had received and the other ; both children were excluded from the analysis. Another child presented with fever and vomiting 48 hr after having ingested the full dose of. By this time his parasitemia was 4% less than at enrollment and so he received parenteral quinine. The child was thus a treatment failure and was included in the analysis as such. For the reasons indicated in Figure 1 it was not possible to evaluate 25 children (17.4%) treated with and 15 children (1.5%) treated with two weeks after treatment. Children treated with cleared P. falciparum significantly faster than children treated with. The time to clear 5% of the parasites was 11. hr (95% confidence interval [CI] ) for -treated children

3 4 VON SEIDLEIN AND OTHERS TABLE 1 Characteristics of children treated with or pyrimethamine/sulfadoxine () at the time of enrollment* Patients enrolled Males Median age in years (IQR) Median weight in kg (IQR) Median temperature at baseline, C median (IQR) Median no. of parasites/ l at baseline (IQR) Hepatomegaly Splenomegaly Median hematocrit (IQR) * IQR interquadratile range (3.9%) 3. (2. 4.) 13. ( ) 38.7 ( ) 141, (77, 37,) 21 (14.%) 28 (19.4%) 31. ( ) (58.%) 3. (2. 4.) 12.8 ( ) 38.7 ( ) 18, (,75 34,) 21 (14.7%) 31 (21.7%) 3.2 ( ) compared with 21.1 hr (95% CI ) for -treated children (P.1). By day 4, all children treated with who could be evaluated had cleared their parasites in contrast with nine -treated children who remained parasitemic (Table 2). Therefore, 133 (1%) of the evaluable children treated with and 128 (93.4%) of the evaluable children treated with were free of parasites three days after treatment (relative risk [RR].93, 95% CI.89.98, P.3). If children who were not evaluable are included in the analysis as failures (intention to treat analysis), 133 (92.4%) of 144 -treated children compared with 128 (89.5%) of 143 -treated children are considered cured (RR.73, 95% CI , P.42). Within 24 hr of treatment with, 19 (13.7%) children remained febrile, in contrast with 7 (49.3%) children treated with (P.1; Table 3). This more rapid fever clearance was reflected by a lower requirement for antipyretics among -treated children compared with -treated children (Table 3). Symptoms such as diarrhea, vomiting, or coughing were observed in a similar proportion of children treated with or. Two weeks after treatment, eight (.7%) of 119 -treated children who could be evaluated were parasitemic compared with three (2.3%) of the 128 children treated with (Table 2). Therefore, 111 (93.3%) of the 119 -treated evaluable children and 125 (97.7%) of the 128 -treated evaluable children were free of parasites (RR 2.87, 95% CI , P.13). In the intention to treat analysis, 111 (77.1%) of treated children compared with 125 (87.4%) of 143 treated children were considered cured (RR 1.82, 95% CI , P.3). At the time of enrollment, the median hematocrits (Hcts) of children treated with and children treated with were similar (Table 1). Two weeks after treatment, there was still no significant difference in the median Hct between the children who had received (median Hct 31.9%, interquadratile range [IQR] %) and the children who had been treated with (median Hct 31.8%, IQR %). Prior to therapy, gametocytes, the sexual stage in the life cycle of P. falciparum, were detected in 4.3% of the children (Table 4). Gametocytes were detected in 2.4% of children three days after treatment with and in none of the children two weeks after treatment. In contrast, the proportion of children with gametocytes increased following treatment with : 13.5% of the evaluated children carried gametocytes three days after treatment and 28.9% carried gametocytes two weeks after treatment. In addition to the eight children who were found to be parasitemic at the two-week follow-up, another 2 -treated children returned to the clinic with signs and symptoms of malaria within four weeks of the initial treatment and were found to be infected with P. falciparum. One child treated with returned during this period. It was possible to compare the genotype of the parasite collected during first and second infections in 23 of the 29 repeat malaria episodes that occurred within 29 days of treatment (Table 5). There was genetic heterogeneity in 19 pairs (83%), suggesting new infections as the source of these second episodes. In four paired samples (17%), the alleles under investigation appeared the same: these infections may have been due to recrudescences. Another 13 children presented between 3 and 4 days after the initial treatment with a further episode of malaria, six had been treated with and seven with. A comparison of isolates from enrollment and second episode showed genetic homogeneity in parasites from one child treated with, which could therefore be a recrudescence; the remaining 12 were heterogeneous and likely to have been new infections. We detected in 17 of 3 first episodes of malaria infections and in 13 of 3 second episodes of malaria two or more clones of P. falciparum. Overall, 42 children required therapy for second episodes of malaria. The parents or guardians of 34 of these children agreed that their child could be treated with open-label. Eight parents whose children had received in the preceding four weeks refused to take part in a further study because of the need for a detailed follow-up (Figure 1). The median time between initial therapy and re-treatment was 25 days (range 15 4 days). Within four days of re-treatment, six children could not be evaluated because the parents withdrew their consent for the collection of further blood samples and one child was lost to follow-up. All 27 evaluable children, including 19 children treated previously with, had cleared their parasites within four days of receiving open-label therapy. Two children whose genotyping was consistent with a recrudescence of P. falciparum cleared the parasites after a second course of. Twenty children could be evaluated two weeks after the second treatment; two of the children were found to have a third P. falciparum infection. Four children presented with a third

4 RANDOMIZED CONTROLLED TRIAL OF 41 FIGURE 1. Flow chart of the children, indicating their disposition following enrollment into the study. F/U follow-up. The number of children who could not be evaluated on day 15 includes all children from the start of the trial onwards. episode of malaria between 19 and 25 days after re-treatment. DISCUSSION Rainfall in The Gambia during 199 reached only 4% of the yearly average and the number of malaria cases were lower than expected during the malaria season. Despite enrollment below the target sample size, it was possible to detect important differences between children treated with and those treated with. Children with uncomplicated falciparum malaria cleared parasites faster after treatment with than did children treated with. The rapid parasite clearance time observed in this study is within the range seen with five-day courses of other oral artemisinin derivatives (mean 34 3 hr). 11, 12 Fever is the manifestation of malaria that most frequently causes discomfort. Thus, it was encouraging that children treated with became afebrile significantly faster than children treated with. We measured the requirement for antipy- TABLE 2 Cure rates (absence of parasitemia and symptoms) for children treated with or pyrimethamine/sulfadoxine () on day 4 and day 15* Day 4 ITT Day 4 evaluable patients Day 15 ITT Day 15 evaluable patients * ITT intention to treat analysis; CI confidence interval. P 133/ % ( %) 133/133 1% (97.3 1%) 111/ % ( %) 111/ % ( %) 128/ % ( %) 128/ % ( %) 125/ % ( %) 125/ % ( %)

5 42 VON SEIDLEIN AND OTHERS TABLE 3 Number of children with fever (temperature 37.5 C) and number of children who required antipyretics during the four days following treatment with or pyrimethamine/sulfadoxine () Day 1 Day 2 Day 3 Day 4 Children with fever P 114/144 (79.2%) 19/139 (13.7%) /14 (4.3%) 7/134 (5.2%) 115/143 (8.4%) 7/142 (49.3%) 28/141 (19.9%) 7/141 (5.%) Children requiring antipyretics P 11/144 (7.1%) 7/139 (5.4%) 8/14 (48.%) 49/134 (3.%) 98/143 (8.5%) 89/142 (2.7%) 8/141 (1.%) 58/141 (41.1%) retics by the study children, since their administration may confound fever clearance times. Use of antipyretics, as reported by parents or guardians, was lower in - treated children compared with -treated children. Sequelae of fever such as febrile seizures were not observed in either treatment group. The cure rate on day 15 for children treated with was significantly higher compared with children treated with. After follow-up on day 15, the detection of further malaria episodes depended on parents or guardians returning their children to the clinic. Within 28 days of treatment with, 28 (19%) children had a second episode of malaria compared with only one child treated with. Up to 5% of patients treated with short courses of artemisinin derivatives as monotherapy have second episodes of P. falciparum infections. 7 Only one study, conducted in the People s Republic of China, has compared the combination of artemether with benflumetol with artemether monotherapy directly and found a 1% cure rate on day 28 in individuals treated with the combination compared with a 55% cure rate for individuals treated with artemether alone (Novartis Pharma AG, Basel, Switzerland, unpublished data). To investigate whether second P. falciparum infections were due to recrudescence, a failure to eliminate parasites completely or due to new infections, we compared variants of three highly polymorphic alleles of P. falciparum from first and second malaria episodes. In only four of 28 samples from -treated children collected within 28 days of treatment was homology found between first and second isolates, suggesting that the majority of second malaria episodes were due to new infections. The genotyping method used has limitations. First, if the variation in one of the loci used as a marker is limited or one of the variants is present at high frequency in the parasite population new infections are misclassified as recrudescences. The use of three unlinked markers minimizes this possibility. The markers used in this study are sufficiently polymorphic since we found 2 distinguishable alleles of MSP-1, 21 of MSP-2, and 15 of GLURP. Thus, the chance TABLE 4 Number of children who carried gametocytes within two weeks of treatment with or pyrimethamine/sulfadoxine () Day 1 Day 2 Day 3 Day 4 Day 15 P /14 (4.3%) 7/134 (5.2%) 4/119 (3.4%) 1/41 (2.4%) /125 (%) /143 (4.2%) 8/142 (5.%) 12/134 (9.%) 1/74 (13.5%) 37/128 (28.9%) of encountering two clones with identical genotypes is one in 8,19. This probability is drastically reduced when combinations of more than one clone are present in a sample. Second, it is possible that some lines escape detection by PCR because their concentration in the peripheral circulation is below the detection threshold or because they are sequestered. Once the antimalarial drug concentration decreases below the minimal inhibitory concentration, these lines could multiply and be misclassified as new infections. The lower attack rates for children treated with may be explained by the lack of a prophylactic effect of. Pyrimethamine and sulfadoxine have half-lives of 7.3 days and 8.9 days, respectively, and pyrimethamine concentrations that are suppressive for drug-sensitive P. falciparum isolates remain in the blood for at least two weeks In contrast, the half-life of artemether is hr and the half-life of benflumetol has been estimated between four and six days. 17, 18 More than 28 days after treatment, when a prophylactic effect of becomes unlikely, seven (4.9%) of the -treated children and six (4.2%) of the -treated children had a second episode of malaria. This indicates a similar risk for new infections in each treatment group in the absence of any prophylaxis. The successful spread of resistance may be explained at least in part by the long half-life of. No field cases of malaria resistant to artemisinin derivatives have been reported so far and the short half-life makes a selection process analogous to the development of resistance unlikely. In addition, in patients treated with, the parasites will not become exposed to artemether alone, which during its relatively short half-life is accompanied by benflumentol. The development of benflumetol resistant isolates cannot, however, be excluded by this rationale. In expectation of new infections, we evaluated in the treatment of second episodes of malaria. A relative high number of parents or guardians allowed us to study the effect of a second course of on their children within a relatively short time of the first treatment. None of the children showed evidence of R2 or R3 resistance following a second course of. However, six (17.7%) of 34 children returned with a third episode of malaria, a similar percentage as that following the initial treatment. The frequency of new infections with P. falciparum was surprising considering the low entomologic inoculation rate in the study area. Previous studies at the coastal study site have shown that roughly one in four children receive an infectious bite each year. 19 Therefore, the children with a P. falciparum infection may be at a higher risk for malaria infections upon return to their home compared with other chil-

6 RANDOMIZED CONTROLLED TRIAL OF 43 TABLE 5 Number of children who had second episodes of malaria following treatment with or pyrimethamine/sulfadoxine ()* Second episode Recrudescence New infection Not analyzed Total Reappearance by 2 weeks (days 13 1) Reappearance by 4 weeks (days 17 29) Reappearance after 4 weeks (days 3 4) * Genotyping was used to distinguish recrudescences from new infections. Recrudescence: a complete match in allelic size for the three genes under investigation between the first and second samples dren in the community. These children can be treated with repeated courses of antimalarials or alternatively the children and their homes might be targeted for prophylactic strategies such as the distribution of impregnated bed nets, eradication of nearby mosquito breeding sites, or the detection and treatment of gametocyte carriers. We observed a highly significant difference in the development of gametocytemia between -treated and -treated children. Twenty-nine percent of -treated children carried gametocytes two weeks after treatment, but none of the children treated with carried gametocytes. The effect of and other artemisinin derivatives on gametocytes suggests that this group of drugs may have an effect on malaria transmission. Indeed, some suggestive evidence has been obtained that malaria transmission was reduced in an area of Thailand following the introduction of artemisinin derivatives. 2 The ability of treatment with and artemisinin derivatives to reduce malaria transmission will be determined by the contribution of the treated patients to the overall reservoir of infection. The transmission blocking potential of in different settings and the way in which this drug achieves its effect on gametocytes require further studies. Acknowledgments: We thank Dr. T. Corrah for diligent help as safety monitor during this trial. The help of Lamin Manneh, Gibril Bah, and Nosipho Mtombeni was essential for the completion of this trial. We also acknowledge the support of the Institute of Microbiology and Epidemiology of the Academy of Military Sciences (Beijing, People s Republic of China) and Novartis Pharma A.G. (Basel, Switzerland) for supplying and financial support for this work. Authors addresses: Lorenz von Seidlein, Hopital Albert Schweitzer, Deschapelles, Haiti c/o Agape Flights, Inc., th Street East, Sarasota, FL Shabbar Jaffar and Brian Greenwood, Medical Research Council Laboratories, Fajara, The Gambia and Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom. Kalifa Bojang, Peter Jones, Margaret Pinder, Steven Obaro, Tom Doherty, and Keith McAdam, Medical Research Council Laboratories, Fajara, The Gambia. Michelle Haywood, Department of Medical Parasitology, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom. Georges Snounou, Department of Infection and Tropical Medicine (Lister Unit), Imperial College School of Medicine Northwick Park Hospital, Harrow, Middlesex HA1 3UJ, United Kingdom. Beatrice Gemperli, Insa Gathmann, and Catherine Royce, International Clinical Research, Novartis, Basel, Switzerland. REFERENCES 1. Butler D, Maurice J, O Brien C, Time to put malaria on the global agenda. Nature 38: World Bank, 1993.World Development Report 1993; Investing in Health. New York: Oxford University Press. 3. Ronn AM, Msangeni HA, Mhina J, Wernsdorfer WH, Bygbjerg IC, 199. High level of resistance of Plasmodium falciparum to sulfadoxine-pyrimethamine in children in Tanzania. Trans R Soc Trop Med Hyg 9: Onyiorah E, Boele van Hensbroek M, Jah MS, Greenwood BM, 199. Early clinical failures after pyrimethamine-sulfadoxine treatment of uncomplicated falciparum malaria. Trans R Soc Trop Med Hyg 9: Radloff PD, Philipps J, Nkeyi M, Hutchinson D, Kremsner PG, 199. Atovaquone and proguanil for Plasmodium falciparum malaria. Lancet 347: Ringwald P, Bickii J, Basco L, 199. Randomized trial of pyronaridine versus chloroquine for acute uncomplicated falciparum malaria in Africa. Lancet 347: Meshnick SR, Taylor TE, Kamchonwongpaisan S, 199. Artemisinin and the antimalarial endoperoxides: from herbal remedy to targeted chemotherapy. Microbiol Rev : von Seidlein L, Jaffar S, Pinder M, Haywood M, Snounou G, Gemperli B, Gathmann I, Royce C, Greenwood BM, Treatment of African children with uncomplicated falciparum malaria with a new antimalarial CGP 597. J Infect Dis 17: Viriyakosol S, Siripoon N, Petcharapirat C, Petcharapirat P, Jarra W, Thaithong S, Brown KN, Snounou G, Genotyping of Plasmodium falciparum isolates by the polymerase chain reaction and potential uses in epidemiologic studies. Bull World Health Organ 73: George S, The required sample size and length of a phase III clinical trial. Buyse ME, Staquet MJ, Sylvester RJ, eds. Cancer Clinical Trials: Methods and Practice. Oxford: Oxford University Press, Karbwang J, Bangchang KN, Thanavibul A, Bunnag D, Chongsuphajaisiddhi T, Harinasuta T, Comparison of oral artemether and mefloquine in acute uncomplicated falciparum malaria. Lancet 34: Looareesuwan S, Viravan C, Vanijanonta S, Wilairatana P, Suntharasamai P, Charoenlarp P, Arnold K, Kyle D, Canfield C, Webster K, Randomized trial of artesunate and mefloquine alone and in sequence for acute uncomplicated falciparum malaria. Lancet 339: Goodman Gilman A, Rall T, Nies A., Taylor P, 199. Goodman and Gilman s The Pharmacological Basis of Therapeutics. New York: Pergamon Press. 14. Edstein M, Quantification of antimalarial drugs. II. Simultaneous measurement of dapsone, monoacetyldapsone and pyrimethamine in human plasma. J Chromatogr 37: Edstein M, Quantification of antimalarial drugs. I. Simultaneous measurement of sulphadoxine, N4-acetylsulphadoxine and pyrimethamine in human plasma. J Chromatogr 3: Heligren U, Kihamia CM, Bergqvist Y, Lebbad M, Premji Z,

7 44 VON SEIDLEIN AND OTHERS Rombo L, 199. Standard and reduced doses of sulfadoxinepyrimethamine for treatment of Plasmodium falciparum in Tanzania, with determination of drug concentrations and susceptibility in vitro. Trans R Soc Trop Med Hyg 84: Na-Bangchang K, Karbwang J, Thomas CG, Thanavibul A, Sukontason K, Ward SA, Edwards G, Pharmacokinetics of artemether after oral administration to healthy Thai males and patients with acute uncomplicated falciparum malaria. Br J Clin Pharmacol 37: Navaratnam V, Mansor SM, Chin LK, Mordi MN, Asokan N, Nair NK, Determination of artemether and dihydroartemisinin in blood plasma by high performance liquid chromatography for application in clinical pharmacological studies. J Chromatogr 9: Lindsay S, Campbell H, Adiamah J, Greenwood A, Bangali J, Greenwood B, 199. Malaria in a periurban area of The Gambia. Ann Trop Med Parasitol 84: Price RN, Nosten F, Luxemburger C, ter Kuile FO, Paiphun L, Chongsuphajaisiddhi T, White NJ, 199. Effects of artemisinin derivatives on malaria transmissibility. Lancet 347: