[Surveillance of falciparum malaria susceptibility to antimalarial drugs and policy change in the Comoros]. / Surveillance de la chimiosensibilite du paludisme do à Plasmodium falciparum et changement de politique dans l'Union des Comores.

Between May and June 2001, efficacy of chloroquine was assessed in 5 sentinel sites in the 3 Comoro islands. Among the 183 children, age range between 6 and 59 months, followed up for 14 days, clinical failure rates ranged between 31.2 and 73.1% and the total failure rate (clinical and parasitological) between 50 and 88.5%. Failures were mainly early treatment failures. The Ministry of health, during a consensus meeting decided to change the first line drug and to gather baseline information on the efficacy and the tolerance of the combination artemether-lumefantrine. Between June and September 2004, among the 164 children, age range between 6 and 59 months included, the success rate of the combination was 99.4% in the 3 sites with a follow-up of 28 days. No serious drug related adverse event was reported.

Heterogeneity in the circumsporozoite protein gene of Plasmodium malariae isolates from sub-Saharan Africa.

Polymorphism of the circumsporozoite protein (CSP) of Plasmodium malariae was characterized by comparing gene sequences of twelve field isolates obtained in Yaoundé, Cameroon, Central Africa, and one clinical isolate originating from Côte d'Ivoire, West Africa. The length of the CSP gene ranged from 1266 to 1302 bp. The size polymorphism was due to variation in the number of tandem repeat units. All P. malariae isolates displayed a highly conserved 5' non-repeat region. Seven non-synonymous and two synonymous nucleotide variations were observed in the 3' non-repeat region. In the deduced amino acid sequence the repetitive sequences consisted of a varying number of major (Asn Ala Ala Gly (NAAG); range between 42 and 46 units) and minor (Asn Asp Ala Gly (NDAG) or Asn Asp Gln Gly (NDEG); n = six or seven units) tetrapeptide units. None of the isolates had an identical sequence at nucleotide level. These findings suggest that polymorphism in CSP is essentially limited to the tandem repeat domain.

Molecular epidemiology of malaria in Yaounde, Cameroon. VI. Sequence variations in the Plasmodium falciparum dihydrofolate reductase-thymidylate synthase gene and in vitro resistance to pyrimethamine and cycloguanil.

Pyrimethamine and cycloguanil, the major human metabolite of proguanil, are inhibitors of dihydrofolate reductase that play a key role in the treatment and prevention of chloroquine-resistant Plasmodium falciparum infections in sub-Saharan Africa. Resistance to these antifolate drugs has emerged in some areas of Africa. Earlier molecular studies have demonstrated that point mutations at key positions of the dihydrofolate reductase-thymidylate synthase gene are strongly associated with antifolate resistance. However, whether the same or distinct mutations are involved in the development of resistance to both pyrimethamine and cycloguanil has not been well established in naturally occurring P. falciparum isolates. In this study, the in vitro responses to both antifolate drugs were measured in 42 Cameroonian isolates and compared with the complete sequence of the dihydrofolate reductase domain of the gene (from 34 of 42 isolates) to analyze the genotype that may distinguish between pyrimethamine and cycloguanil resistance. The wild-type profile (n = 11 isolates) was associated with low 50% inhibitory concentrations (IC50s) ranging from 0.32 to 21.4 nanamole for pyrimethamine and 0.60-6.40 nM for cycloguanil. Mutant isolates had at least one amino acid substitution, Asn-108. Only three mutant codons were observed among the antifolate-resistant isolates: Ile-51, Arg-59, and Asn-108. The increasing number of point mutations was associated with an increasing level of pyrimethamine IC50 and, to a much lesser extent, cycloguanil IC50. These results support a partial cross-resistance between pyrimethamine and cycloguanil that is based on similar amino acid substitutions in dihydrofolate reductase and suggest that two or three mutations, including at least Asn-108, may be necessary for cycloguanil resistance, whereas a single Asn-108 mutation is sufficient for pyrimethamine resistance.

Molecular epidemiology of malaria in Yaoundé, Cameroon. III. Analysis of chloroquine resistance and point mutations in the multidrug resistance 1 (pfmdr 1) gene of Plasmodium falciparum.

It has been postulated that chloroquine resistance may be associated with a single point mutation at codon 86 of the Plasmodium falciparum multidrug resistance 1 (pfmdr 1) gene. Using a simple and rapid molecular technique involving polymerase chain reaction and restriction fragment length polymorphism, the frequency of the Asn-to-Tyr mutation associated with chloroquine resistance was established among 129 clinical isolates obtained from indigenous patients in Yaoundé, Cameroon. The results showed that 110 of 129 isolates display a mutant codon. The other clinical isolates had either a pure wild-type Asn-86 codon (n = 12) or mixed Asn/Tyr alleles (n = 7). In vitro drug assays were performed to compare the genotype and phenotype in 102 clinical isolates. Of these isolates, 86 displayed pure Tyr-86 mutant codon; 48 (56%) mutant isolates were chloroquine-resistant (50% inhibitory concentration [IC50] > 100 nM), as expected, but 38 (44%) mutant isolates were chloroquine-sensitive (IC50 < 100 nM). Three chloroquine-resistant isolates and seven chloroquine-sensitive parasites carried a wild-type Asn-86 codon. Mixed alleles were found in six isolates (four chloroquine-sensitive and two chloroquine-resistant isolates). Our results did not confirm previous observations on the possible association between chloroquine resistance phenotype and genotype based on the pfmdr 1 gene.

Molecular epidemiology of malaria in Yaounde, Cameroon I. Analysis of point mutations in the dihydrofolate reductase-thymidylate synthase gene of Plasmodium falciparum.

Resistance to antifolate antimalarial drugs (cycloguanil, a biologically active metabolite of proguanil, and pyrimethamine) is associated with a Ser- to Asn-108 point mutation in the dihydrofolate reductase-thymidylate synthase gene of Plasmodium falciparum. The frequency of this mutation was studied in 127 clinical isolates obtained in Yaounde, Cameroon using a simple and rapid molecular technique based on the polymerase chain reaction and restriction fragment length polymorphism. Of the 127 isolates, pure wild-type Ser-108 codon, pure mutant-type Asn-108 codon, and mixed codons were observed in 66, 55, and six parasites, respectively. The proportion of antifolate-resistant, pure mutant-type codon, with respect to pure wild-type or mixed alleles, was 43% (55 of 127). The results of the molecular assay were compared with those of semimicro isotopic in vitro assay in 34 isolates. All 17 pure Ser-108 isolates and two isolates with mixed alleles were sensitive to both pyrimethamine (50% inhibitory concentration [IC50] < 100 nM) and cycloguanil (IC50 < 50 nM). Fourteen of 15 isolates with the mutant-type Asn-108 codon were resistant to pyrimethamine and cycloguanil. One isolate with Asn-108 showed a slightly elevated pyrimethamine IC50 (78 nM), which was within the sensitive range. This study provides further evidence that antifolate-resistant P. falciparum isolates are already present in Yaounde, Cameroon.

Molecular epidemiology of malaria in Yaounde, Cameroon II. Baseline frequency of point mutations in the dihydropteroate synthase gene of Plasmodium falciparum.

Sulfadoxine-pyrimethamine is one of the alternative antimalarial drugs used to treat chloroquine-resistant Plasmodium falciparum malaria. The molecular target of sulfadoxine, an analog of p-aminobenzoic acid that inhibits the folate biosynthetic pathway, is dihydropteroate synthase (DHPS). The nucleotide sequence of the DHPS gene was determined in 32 clinical isolates obtained in Yaounde, Cameroon, and compared with the sequence of reference clones and Cambodian strains of P. falciparum. Of the 32 Cameroonian isolates, 31 displayed one of the sulfadoxine-sensitive mutation patterns: Ala-436/Ala-437/Ala-581/Ala-613 (n = 20), Ser-436/Gly-437/Ala-581/Ala-613 (n = 6), Ser-436/Ala-437/Ala-581/Ala-613 (n = 4), and Ala-436/Gly-437/Ala-581/Ala-613 (n = 1). One isolate had a sulfadoxine-resistant profile characterized by a double mutation: Phe-436/Ala-437/Ala-581/Ser-613. Although the majority of the isolates had a sulfadoxine-sensitive genetic profile, further studies are needed to correlate the mutation patterns and in vitro and in vivo sulfadoxine sensitivity.

Molecular epidemiology of malaria in Yaounde, Cameroon IV. Evolution of pyrimethamine resistance between 1994 and 1998.

Pyrimethamine, in combination with sulfadoxine, is currently one of the major alternative drugs used for the treatment of chloroquine-resistant Plasmodium falciparum malaria infections in Africa. The mechanism of pyrimethamine resistance has been strongly associated with a single, key point mutation in the dihydrofolate reductase-thymidylate synthase gene, resulting in the substitution of the wild-type allele Ser-108 by either Asn-108 or Thr-108. The pyrimethamine-resistant phenotype and/or genotype were determined in 273 Cameroonian clinical isolates obtained in Yaounde by in vitro assays and polymerase chain reaction-restriction fragment length polymorphism over a 5-year period. The in vitro assays showed that 42% (18 of 43) and 63% (69 of 110) of the isolates obtained in 1994-1995 and 1997-1998, respectively, were resistant to pyrimethamine (50% inhibitory concentration [IC50] > 100 nM). The polymerase chain reaction showed that 43% (55 of 127) and 59% (50 of 85) of the isolates in 1994-1995 and 1997-1998, respectively, had the mutant Asn-108 allele. The pyrimethamine-resistant genotype (Asn-108) corresponded with the pyrimethamine-resistant phenotype (IC50 > or = 100 nM) in a large majority (> 95%) of the isolates. The results of our study suggest an increasing prevalence of pyrimethamine resistance in Yaounde. Our study further suggests that pyrimethamine resistance can be monitored by a technique that can be adopted by malaria research centers in Africa.

Molecular epidemiology of malaria in Yaounde, Cameroon V. analysis of the omega repetitive region of the plasmodium falciparum CG2 gene and chloroquine resistance.

A novel Plasmodium falciparum gene, denoted cg2 gene, has been recently discovered, and a distinct genotype, characterized by 12 point mutations and 3 size polymorphisms, has been shown to be associated with chloroquine resistance in laboratory-adapted parasite strains. One of the polymorphic regions, denoted the omega region, consists of 16 tandem repeat units in chloroquine-resistant strains, while the chloroquine-sensitive strains have either < or = 15 or > or = 17 repeat units. In this study, the in vivo and in vitro responses were compared with the number of repeat units in the omega region of the cg2 gene for 75 Cameroonian isolates determined either by DNA sequencing or agarose gel electrophoresis. The 16-repeat units that characterize the resistant strains were found in 10 chloroquine-sensitive isolates (50% inhibitory concentration [IC50] < 100 nM) and 30 chloroquine-resistant isolates (IC50 > or = 100 nM). Thirty-five isolates (28 chloroquine-sensitive isolates and 7 chloroquine-resistant isolates) displayed < or = 15 or > or = 17 repeat units. Of the 18 patients responding with treatment failure, 15 were infected with parasites carrying 16 repeat units. Twenty-eight patients (11 with isolates carrying 16 repeat units and 17 with isolates carrying < or = 15 or > or = 17 repeat units) showed an adequate clinical response. The sensitivity, specificity, and predictive value were 81% (83%), 74% (61%), and 75% (58%), respectively compared with in vitro (or in vivo) responses. Neither the level of IC50 nor the key P. falciparum multidrug resistance gene 1 (pfmdr 1) allele at position 86 was associated with the number of omega repeat units. Although in vitro and in vivo resistance to chloroquine was statistically associated with the presence of 16 repeat units in the omega region (P < 0.05), the number of omega repeat units did not adequately discriminate patients infected with chloroquine-resistant parasites from those infected with chloroquine-sensitive parasites. Further studies on the cg2 gene are needed to determine whether cg2 gene is a reliable genetic marker for chloroquine resistance.

Molecular epidemiology of malaria in Yaounde, Cameroon. VII. Analysis of recrudescence and reinfection in patients with uncomplicated falciparum malaria.

In an endemic area where malaria transmission is intense and continuous, reappearance of asexual parasites may be ascribed to either recrudescence or reinfection. To distinguish between recrudescence and reinfection after oral treatment with chloroquine, amodiaquine, pyronaridine, sulfadoxine-pyrimethamine, halofantrine, or artesunate, three polymorphic markers (circumsporozoite protein, merozoite surface antigens 1 and 2) from pre-treatment and post-treatment samples were amplified by the polymerase chain reaction, and the in vitro response to chloroquine was determined for comparison. Of 52 paired samples, 22 (42%) were reinfections. Recrudescence occurred more frequently on or before Day 14 (22 of 30 cases, 73%). Except for one case, all reinfections were observed beyond Day 14. The phenotype determination was not sufficiently precise to distinguish between recrudescence and reinfection. Our results suggest that beyond Day 14 (and until Day 42), recrudescence and reinfection cannot be distinguished at our study site unless molecular techniques are used and that some results derived from the polymerase chain reaction need to be compared with the microscopic examination of thick blood smear to exclude gametocyte carriers without asexual parasites after treatment.

Molecular epidemiology of malaria in Yaounde, Cameroon. VIII. Multiple Plasmodium falciparum infections in symptomatic patients.

The extent of genetically distinct parasite populations coinfecting individual human hosts (i.e., multiplicity) was studied by polymerase chain reaction amplification of 3 polymorphic genetic markers, circumsporozoite protein and merozoite surface antigens (MSA) 1 and 2, in symptomatic children and adults and analyzed in relation with age and initial parasitemia. Of the total of 177 DNA samples analyzed (of which 115 were paired pre- and posttreatment samples), 101 (57%) were composed of multiclonal infections, with up to 7 distinguishable parasite populations. Among the 3 polymorphic markers, msa-2 yielded the highest proportion of clinical isolates with multiclonal populations. Patients with multiclonal infections before treatment had, on average, 2.9 genetically distinct parasite populations. The extent of multiplicity decreased significantly (P < 0.05) in recrudescent parasites, but not with reinfections, as compared with the pretreatment samples. Neither age (5-60 years) nor initial parasitemia was correlated with multiplicity. Further studies in different epidemiological settings are required to understand the role of multiclonal Plasmodium falciparum infections in influencing malaria transmission.

In vitro activity of antimalarials against clinical isolates of Plasmodium falciparum in Yaounde, Cameroon.

The in vitro activity of nine antimalarials was determined against 119 fresh clinical isolates of Plasmodium falciparum obtained from symptomatic indigenous patients in Yaounde, Cameroon. using the isotopic semimicrotest. Seventy-four parasites were resistant to chloroquine (mean 50% inhibitory concentration [IC50] 337 nM); 45 were chloroquine-sensitive (mean IC50 35.6 nM). Twenty-five of 58 chloroquine-resistant parasites were resistant to monodesethylamodiaquine, the biologically active metabolite of amodiaquine. None of the chloroquine-sensitive isolates was resistant to monodesethylamodiaquine (IC50 17.3 nM). Pyronaridine, quinine, mefloquine, halofantrine, and artemether were highly active against the chloroquine-sensitive and the chloroquine-resistant isolates. Of the 43 isolates tested, 25 were sensitive to both pyrimethamine and cycloguanil, the biologically active metabolite of proguanil. The in vitro responses of chloroquine and monodesethylamodiaquine, chloroquine and quinine, quinine and mefloquine, mefloquine and halofantrine, artemether and mefloquine or halofantrine, and pyrimethamine and cycloguanil were significantly correlated. The present study suggests that chloroquine resistance is highly prevalent in vitro in Yaounde and that the alternative drugs are generally highly active against the chloroquine-resistant parasites.

In vitro activity of dihydroartemisinin against clinical isolates of Plasmodium falciparum in Yaounde, Cameroon.

The in vitro activities of dihydroartemisinin (the biologically active metabolite of artemisinin derivatives), chloroquine, monodesethylamodiaquine (the biologically active metabolite of amodiaquine), quinine, mefloquine, halofantrine, and pyrimethamine were assessed in 65 African isolates of Plasmodium falciparum from Yaounde, Cameroon using an isotopic microtest. The 50% inhibitory concentration (IC50) values for dihydroartemisinin were within a narrow range from 0.25 to 4.56 nM, with a geometric mean of 1.11 nM (95% confidence interval = 0.96-1.28 nM). Dihydroartemisinin was equally active (P > 0.05) against the chloroquine-sensitive isolates (geometric mean IC50 = 1.25 nM, 95% confidence interval = 0.99-1.57 nM) and the chloroquine-resistant isolates (geometric mean IC50 = 0.979 nM, 95% confidence interval = 0.816-1.18 nM). A significant positive correlation was observed between the responses to dihydroartemisinin and mefloquine (r = 0.662) or halofantrine (r = 0.284), suggesting in vitro cross-resistance. There was no correlation between the responses to dihydroartemisinin and other antimalarial drugs.

Short report: effects of pyronaridine on gametocytes in patients with acute uncomplicated falciparum malaria.

The effects of pyronaridine and chloroquine on mature Plasmodium falciparum gametocytes were compared in 161 patients treated with chloroquine or pyronaridine. Neither pyronaridine nor chloroquine showed gametocytocidal activity. The relative risks of post-treatment gametocytemia after pyronaridine and chloroquine treatment in the presence of chloroquine-resistant isolates were 1.25 and 11.5, respectively, suggesting that the use of chloroquine was associated with a high risk of favoring post-therapeutic gametocytemia in chloroquine-resistant infections.

Cardiac effects of amodiaquine and sulfadoxine-pyrimethamine in malaria-infected African patients.

The cardiac effect of amodiaquine and sulfadoxine-pyrimethamine was studied in adult Cameroonian patients with acute uncomplicated Plasmodium falciparum malaria by electrocardiographic monitoring over the course of 7 days. Clinical and parasitological responses were monitored until Day 14. Bradycardia was observed in 16 of 20 amodiaquine-treated patients on Day 2, which corresponds to the time when maximal cumulative plasma concentration is reached, and in 12 of 20 patients on Day 7. A bradycardic effect lasting several days was not noted in patients treated with sulfadoxine-pyrimethamine. Significantly prolonged P, PQ, QRS, and QTc intervals were recorded on Day 2 after both 30 and 35 mg of amodiaquine base per kilogram of body weight had been administered, but these intervals were not correlated with the plasma monodesethylamodiaquine (main human active metabolite of amodiaquine) level. Electrocardiographic changes after therapy with sulfadoxine-pyrimethamine were minor and transient. All patients had fever and parasite clearance on or before Day 3 and remained free of fever and parasites until Day 14. None of the patients complained of cardiovascular adverse effects during the follow-up. These results suggest the absence of significant cardiac effects of amodiaquine and sulfadoxine-pyrimethamine at usual therapeutic doses, but they should draw the attention of clinicians treating malaria-infected patients who have taken other antimalarial drugs with cardiovascular side effects or those who are under treatment with cardiovascular drugs.

Relationships between malaria prevalence and malaria-related morbidity in school children from two villages in central Africa.

To investigate the relationship between parasite prevalence and malaria-related morbidity, we carried out a comparative study among cohorts of school children from two villages, Dienga, Gabon, and Pouma, Cameroon, both located in malaria-endemic areas. Seven to 17 year-old children attending primary schools were similarly followed-up at each site to evaluate the frequency of malaria attacks. Follow-up involved daily temperature recording (and blood smears in the case of fever) and preparation of blood smears every two weeks. In Pouma, 186 children were followed-up for six months. In Dienga, 228 children were followed-up for nine months. The mean prevalence rate of Plasmodium falciparum infections (as assessed by the blood smears) was twice as high in Pouma compared with Dienga (45.2% versus 26.8%; P < 0.0001), whereas the monthly malaria attack rate (as assessed by the daily surveillance) was twice as high in Dienga compared with Pouma (21.5% versus 41.4%; P = 0.003). The possible implication of several parameters that may differ between the two areas, such as the malaria transmission level, the economical and social status of the inhabitants, the characteristics of infecting parasite strains, and the genetic background of the population, is discussed.

Immune responses against Plasmodium falciparum asexual blood-stage antigens and disease susceptibility in Gabonese and Cameroonian children.

The frequency and level of cellular and humoral responses to seven synthetic peptides from asexual blood stages of Plasmodium falciparum were measured in two cohorts of children living in areas highly endemic for malaria in Gabon and Cameroon. A prospective longitudinal study was conducted for one year in these sites to examine the relationship between specific in vitro immune responses and susceptibility to clinical malaria. Clinical protection was related to high proliferative responses (merozoite surface antigen-1 [MSA-1] and MSA-2 peptides) as well as to elevated antibody levels (schizont extract, MSA-2, and rhoptry-associated protein-1 [RAP-1] peptides) in the village of Dienga, Gabon. Higher response rates of interferon-gamma but lower response rates of tumor necrosis factor-alpha to four and six peptides, respectively, were observed in Dienga than in Pouma that were independent of the older age of the Gabonese children. Age accounted only for the higher prevalence rate in Dienga of the antibody responders to the peptide from Pf155/ring-infected erythrocyte surface antigen (RESA). Our results support the inclusion of epitopes from MSA-1, MSA-2, RAP-1, and Pf155/RESA antigens in a subunit vaccine against malaria, but show that a longitudinal clinical, parasitologic, and immunologic study conducted according to identical criteria in two separate areas may lead to contrasting observations, demonstrating the geographic limitation of the interpretation of such results.

Point mutations in the Plasmodium falciparum cg2 gene, polymorphism of the kappa repeat region, and their relationship with chloroquine resistance.

Based on the available DNA sequence data of the Plasmodium falciparum cg2 gene, we have hypothesized that 3 amino-acid substitutions, His275Gln, Gly281Ala, and His299Gln, may represent the key mutations that confer resistance to chloroquine. The presence of 14 tandemly repeated hexamer units in the kappa region has also been suggested to be indicative of chloroquine resistance. These 2 hypotheses were tested by determining the sequence of DNA fragments containing all 3 codons and kappa repetitive region (approximately 450-basepairs) for 53 randomly selected clinical isolates (obtained in Cameroon in 1994-97) with known response in vivo and/or in vitro to chloroquine. The cg2 genotypes based on the 3 codons and the response in vitro to chloroquine, as well as the number of kappa repeat units and responses in vivo and in vitro to chloroquine, were associated (P < 0.05). cg2 gene mutations were more common in parasites from patients with failure in vivo. However, this difference did not achieve statistical significance (P = 0.055). The sensitivity and specificity of the 3 codons and kappa repeat region to predict the response in vitro to chloroquine ranged between 75% and 85%. The sensitivity and specificity of these genetic markers to predict the response in vivo to chloroquine were of lower values. The kappa repeat region of the clinical isolates is polymorphic but characterized by several conserved features.

Chronotherapy of malaria: improved efficacy of timed chloroquine treatment of patients with Plasmodium falciparum infections.

The effect of routine treatment with chloroquine (10 mg/kg on days 1 and 2 and 5 mg/kg on day 3) on parasitaemia and parasitaemic profile of patients infected with Plasmodium falciparum was studied. As with P. vinckei petteri, the mid-term trophozoites of P. falciparum were the most susceptible stages to chloroquine treatment. It is suggested that, in order to diminish the frequency of drug administration and to lower the risks of chemoresistance developing, treatment should be diversified, using the drug which is most effective on the parasite stages present in the peripheral blood.

Amodiaquine remains effective for treating uncomplicated malaria in west and central Africa.

Many countries in Africa are now confronted with the dilemma of shifting drug policies for uncomplicated falciparum malaria from chloroquine, which has become largely ineffective, to a new first-line drug and amodiaquine is one of the possible options. A multicentre, open-label randomized controlled trial of amodiaquine 30 mg/kg vs chloroquine 25 mg/kg over 3 days was performed in Senegal, Cameroon, Gabon, and Burkina Faso between 1996 and 1998 and patients were followed-up for 14 days. Sensitivity of isolates in vitro and whole blood levels of chloroquine and amodiaquine were also measured. The primary efficacy parameter was parasitological clearance on day 14 (parasitological success). The secondary efficacy parameter was absence of signs/symptoms of malaria on day 14 (clinical success). Among the 364 patients randomized and receiving the assigned treatment (chloroquine n = 185, amodiaquine n = 179), 137 and 139, respectively, reached the primary endpoint. Amodiaquine proved significantly more effective than chloroquine. The summary odds ratio (95% CI) was 7.79 (4.54-13.35) for parasitological success, and 6.3 (3.4-11.68) for clinical success. Sensitivity in vitro and chloroquine blood levels were good predictors of chloroquine failure. Amodiaquine remains effective for treating uncomplicated falciparum malaria in areas of West and Central Africa where chloroquine resistance is prevalent. However, measures should be taken to protect the lifespan of amodiaquine where the drug is introduced for use.

Immune response to Plasmodium falciparum liver stage antigen-1: geographical variations within Central Africa and their relationship with protection from clinical malaria.

Two populations of schoolchildren from Gabon and Cameroon were tested in 1995 for their immunological reactivity to synthetic peptides (LSA-Rep, LSA-J and LSA-CTL) from Plasmodium falciparum liver stage antigen-1 (LSA-1). The prevalence and levels of both cellular (lymphocyte proliferation, tumour necrosis factor alpha (TNF alpha), interferon gamma (IFN gamma), and interleukin-10 (IL-10)) and humoral (immunoglobulin G) responses were determined. Protection from clinical malaria, determined after a prospective 1 year study in both sites, was associated with elevated proliferative responses to LSA-Rep and LSA-CTL in the Gabonese children, as well as with higher antibody levels to both schizont extract and LSA-Rep. The prevalence of peptide-stimulated TNF-alpha secretion was higher in the Cameroonian group, but higher levels of antibodies to LSA-Rep and LSA-J were found in the Gabonese children. The immunological differences observed between children in the 2 study sites are discussed in the context of both epidemiological and individual host factors.