Immunological consequences of intermittent preventive treatment against malaria in Senegalese preschool children

Malaria elimination is now considered a realistic goal for an increasing number of countries [1]. It requires control of the infection in the most at-risk groups, namely pregnant women [2] and children [3]. The distribution of anti-malarial drugs at predetermined regular intervals (Intermittent Preventive Treatment, IPT) to individuals regardless of their malaria status, already implemented during pregnancy, is under clinical evaluation in infants (IPTi, reviewed in [4]), and in preschool (IPTc, [5‐9]) and school-aged children (IPTsc, [7, 10, 11]). Seasonal IPTc (sIPTc) is defined as the administration of IPT to children during the transmission season in locations where malaria transmission is not perennial, mainly in the African Sahelian belt. IPT strategies raise several concerns which are under scrutiny, such as optimal schedule, acceptability, drug resistance, implementation, cost-efficiency, but one question requires an urgent answer-does IPT impair the development of specific immunity? IPT is able to clear a large number of circulating parasites, thus reducing the amount of contacts which are normally required to develop naturally acquired immunity to malaria [12].

Additionally, immuno-suppression has been reported from experimental studies of artemisinin-derivatives [13]. In the case of IPTi, some individual trials provided evidence that treated infants were subsequently, more susceptible to malaria or anaemia, the so-called "rebound effect" [14, 15], but an overall analysis did not show any evidence of rebound [16]. Attention was given mainly to the possible interference between treatment and the infants' response to EPI vaccines, which are delivered at the same time [17‐19]. However, only two studies considered specific anti-Plasmodium immune responses. In Mozambique, sulphadoxine-pyrimethamine (SP) given at the age of three, four, and nine months did not significantly modify the development of naturally acquired antibody (Ab) responses to several Plasmodium falciparum antigens up to 24 months of age [20]. On the other hand, in Ghana, anti-schizont Ab levels were significantly lower in children treated once with SP than in controls [21]. In the latter study, IgG levels were related to the frequency of past infections. Two IPTc trials conducted in children less than five years of age, have demonstrated a lack of a clinical rebound-effect one year after IPT delivery, using SP + artesunate in Senegal [6] and SP in Mali [7]. In Ghana [9], malaria incidence during the post-intervention period was increased by 62% in infants who received six monthly artesunate + amodiaquine, but this rebound was not seen in children aged one year or more at the time of drug administration. Immunological status, known to be closely age-dependent, was not assessed in any of these IPTc trials. Therefore, the objective of the present study was to check whether IPTc had any impact on the anti-Plasmodium IgG response in the Senegalese study [6], eight months after the last IPT delivery. To improve our understanding of the mode of action of IPT [4, 22], epidemiological features of the study children were incorporated as potential confounding factors in a multivariate analysis.

Despite some debate on its efficacy, evidence is growing that IPT is a potentially valuable tool to control malaria in young children [27]. It may be particularly useful in areas with seasonal malaria transmission where drug delivery can be focused on a limited period of time when transmission and malaria risk are maximal. However, the potential negative effect of IPT on the development of natural acquired immunity against the parasite remains an important point to be investigated [28]. Theoretically, IPT could interfere with the anti-Plasmodium immune response via two main distinct mechanisms, either by a direct immunosuppressor effect of the drugs or, indirectly, by reducing contacts between the treated child and Plasmodium thus impairing immunity development. There is already evidence to support the latter mechanism in the case of weekly chemoprophylaxis [29] and this has led to concerns over its widespread use. Unlike some other classes of anti-malarial molecules, including the 4-aminoquinolones [30, 31], neither SP or artesunate is thought to be immunosuppressive. Moreover, several IPTi trials (reviewed by [32]) have shown that SP and artesunate did not have an effect, positive or negative, on the infant response to various childhood vaccines (diphtheria, tetanus, pertussis, polio, hepatitis B, Hib, measles). The most likely mechanism by which IPTc reduces malaria morbidity combines therapeutic (clearing of existing infections) and preventive (allowing immunity development) effects, the prophylactic effect being predominant [33]. IPTi studies have shown that in areas where P. falciparum is still susceptible to SP, a single dose of SP provides infants with a protection lasting from 4-6 weeks [34] to 30-60 days [35] according to the setting, as might be anticipated from the half lives of the component drugs.

The present study showed that children less than five years who had received preventively the combination of SP + artesunate during the rainy season had a slightly lower level of anti-Plasmodium schizont Abs compared to non-treated control children at the beginning of the next transmission season, eight months later. These findings are in agreement with the reduced anti-schizont IgG response measured in Ghanaian infants six months after a single dose of SP given alone [21] although this study differed in several aspects, noticeably in the age of the participants (malaria-naïve infants in Ghana versus 1-6 years old children with some immunity in Senegal) and on the endemicity levels (perennial transmission in rural Ghana with 400 infective bites per person-year [36] versus seasonal transmission in Senegal with 10 infective bites per year [24]).

In a multivariate model, four co-variables significantly influenced anti-Plasmodium IgG levels: age, village, past malaria morbidity and Plasmodium carriage. Age was a major determinant of specific IgG titres, especially among children older than four years. This is in agreement with findings in Northern Senegal where anti-schizont IgG responses increased progressively in one to five-year old children and then remained stable up to nine years of age [37]. Age-associated maturation of the immune system, cumulative impact of mosquito infecting bites and production by the oldest children of long-lived plasma cells [38] may explain the natural influence of age. In accordance with what was observed in Mozambique [20] and in Ghana [21], children who experienced clinical attacks of malaria had higher anti-Plasmodium IgG responses in July 2003 than those who did not become sick. These observations suggest that release into the bloodstream of large numbers of schizont-derived Plasmodium stages during a malaria attack provides substantial antigenic stimulation to the immune effector cells. As reported in the two IPTi antibody surveys previously quoted [20, 21], Plasmodium carriage at the time of blood sampling (July 2003) was associated with a strikingly higher anti-schizont IgG response. Increased humoral responses to various antigens in parasitized children compared to controls is a consistent observation [37, 39], reflecting an increased exposure of the immune system to blood-stage parasites. Similarly, differences in the frequency of infectious anophelines bites [24] may account for the observed significant heterogeneity of specific IgG titres between villages. Surprisingly, parasite carriage seven months before the blood sampling was also strongly and positively linked to a high specific Ab response. Interestingly, parasite rates did not change significantly during the 2003 dry season suggesting that parasites can persist in humans from one transmission season to the other and elicit a steady production of specific Abs through continuous antigenic stimulation. Taken as a whole, it appears that the level of anti-Plasmodium schizont Abs at a given time reflects a cumulative process of specific immunoglobulin production positively related to the level of contact between the host immune system and blood-stage parasites. It underlines the fact that Plasmodium carriage is the key factor for anti-schizont IgG production, irrespective of the preventive treatment received, although the possibility of some contributory effect from the antimalarial drugs used for IPT cannot be completely excluded.

To conclude, administration of SP and artesunate used as sIPTc slightly decreased the humoral reactivity to a crude Plasmodium extract. Specific IgG Abs appeared to increase with the amount of exposure to Plasmodium asexual stages at the time of blood sampling but also to exposure during the previous 10 months. A more accurate analysis of the immune response, looking at different putatively protective isotypes and antigens [40], would be valuable. Previous studies have shown, in The Gambia, that stopping chemoprophylaxis after a period of several years increased the risk of clinical malaria [41]. Monitoring the potential rebound effect in children who will receive IPTc for several consecutive years will also be of great interest [42].