Conceived and designed the experiments: CG CD JJA DLD PLA. Analyzed the data: LQ JJA. Wrote the paper: CG CD PLA. Conducted the field trial and follow-up of participants: CG QB AN MNM MHR MR CM. Performed laboratory analysis: AN MNM RA MHR AB AGM. Contributed to the interpretation of the results and writing of the manuscript: QB RA JJA AGM DJR ES PNLS LS CEC DLD.
DLD was supported by a Pfizer Australia Senior Research Fellowship. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.
The rate of acquisition of naturally acquired immunity (NAI) against malaria predominantly depends on transmission intensity and age, although disentangling the effects of these is difficult. We used chemoprophylaxis to selectively control exposure to
A three-arm double-blind randomized placebo-controlled trial was conducted in 349 infants born to Mozambican HIV-negative women. The late exposure group (LEG) received monthly Sulfadoxine-Pyrimethamine (SP) plus Artesunate (AS) from 2.5–4.5 months of age and monthly placebo from 5.5–9.5 months; the early exposure group (EEG) received placebo from 2.5–4.5 months and SP+AS from 5.5–9.5 months; and the control group (CG) received placebo from 2.5–9.5 months. Active and passive case detection (PCD) were conducted from birth to 10.5 and 24 months respectively. The primary endpoint was time to first or only episode of malaria in the second year detected by PCD. The incidence of malaria during the second year was of 0.50, 0.51 and 0.35 episodes/PYAR in the LEG, EEG and CG respectively (p = 0.379 for the adjusted comparison of the 3 groups). The hazard ratio of the adjusted comparison between the LEG and the CG was 1.38 (0.83–2.28, p = 0.642) and that between the EEG and the CG was 1.35 (0.81–2.24, p = 0.743).
After considerably interfering with exposure during the first year of life, there was a trend towards a higher risk of malaria in the second year in children who had received chemoprophylaxis, but there was no significant rebound. No evidence was found that the age of first exposure to malaria affects the rate of acquisition of NAI. Thus, the timing of administration of antimalarial interventions like malaria vaccines during infancy does not appear to be a critical determinant.
In endemic areas, malaria affects primarily children younger than 5 years of age and pregnant women. Exposure to repetitive
Studies in malaria-naïve transmigrants in Indonesia
Previous trials of malaria chemoprophylaxis and vaccines conducted in Ifakara, Tanzania, suggested that the age of first exposure to
The efficacy provided by continuous and intermittent prophylaxis during the first year of life in these two trials was similar; what differed was the development of NAI, which was impaired in children receiving continuous chemoprophylaxis for most of the first year but was not affected in children receiving IPTi. Given that there is almost no clinical malaria in children younger than 2 months
Two trials of the candidate malaria vaccine Spf66 in the same area in two different age groups further supported this hypothesis. Spf66 had an estimated vaccine efficacy (VE) of 31% (95% CI 0–52%; p = 0.046) in children aged 1 to 4 years
Taken together, data from these four trials posed a fundamental question: is there a critical time window in infancy during which exposure to
The protocol for this trial and supporting CONSORT checklist are available as supporting information; see
The study was conducted at the Centro de Investigação em Saúde de Manhiça (CISM), located in Manhiça, Maputo Province, southern Mozambique. The area has been described in detail elsewhere
CISM runs a demographic surveillance system in its study area and a morbidity surveillance system at Manhiça District Hospital (MDH) and other health posts in the area through which standardized information on all paediatric outpatient visits and admissions to hospital is collected. Recruitment and follow up of study participants were done at the Maragra Health Post (MHP), in the south of the study area, from September 2005 to March 2009.
The protocol was approved by the National Mozambican Ethics Review Committee and the Hospital Clinic of Barcelona Ethics Review Committee. The trial was conducted according to the ICH Good Clinical Practice guidelines and reviewed by a Local Safety Monitor (LSM) and a Data and Safety Monitoring Board (DSMB). The trial was registered in ClinicalTrials.gov (clinical trials identifier NCT00231452).
The study was designed as a double-blind randomized placebo-controlled trial (
Monthly chemoprophylaxis with SP (Fansidar® 500/25 mg) plus Artesunate (AS, Arsumax® 50 mg) or placebo (provided by Roche and Sanofi-Aventis) was administered during different periods of the first year of life according to the randomization group (
Drugs were administered by a field worker according to the following age-based dosing schedule: ½ tablet of SP or placebo and ½ tablet of AS or placebo on the first day and ½ tablet of AS or placebo on the second and third days. Tablets were crushed, mixed with water and administered to the child. If the child vomited, a replacement dose was administered.
During the study, Mozambique's national first line-treatment policy for malaria was SP plus Amodiaquine, which was replaced by SP plus AS in July 2006. Prior to the administration of study drugs the field worker asked the mother/guardian and checked the health card to determine whether the child had received antimalarials during the previous 2 weeks and if so, the study drugs were not administered to avoid repeating a treatment with SP.
Study participants were followed up until age 24 months. Weekly active case detection (ACD) was conducted from birth to approximately age 10.5 months (4 weeks after intervention H) and monthly home visits from 10.5 to 24 months of age. ACD visits were performed at home by field workers, who measured the axillary temperature with a digital thermometer and asked the mother/guardian about history of fever. Children presenting fever (axillary temperature ≥37.5°C) or whose guardians reported history of fever in the preceding 24 h were taken to MHP, where they were examined and parasitaemia and haematocrit were determined. Antimalarials were administered according to procedures explained below if the slide reading was positive for asexual
Additionally, passive case detection (PCD) was carried out at the MHP and MDH through the morbidity surveillance system to monitor attendances to the outpatient clinics and admissions to hospital. Parents/guardians of study participants were asked to take their children to the MHP or MDH should they have any health problem at any time, where they were managed and treated according to Mozambican national guidelines by the attending medical personnel. Standard procedures included identification of the child, measurement of axillary temperature, completion of a standard morbidity questionnaire and physical examination. If the temperature was ≥37.5°C or there was a reported fever in the preceding 24 hours, haematocrit and malaria parasitaemia were assessed. Antimalarials were administered only if the blood slide was positive. As first line antimalarial treatment included SP, study participants who presented with malaria and had received a study intervention within the previous 2 weeks received oral quinine (10 mg/kg/8 h for a minimum of 5 days) for the treatment of their episode. Children seen at the MHP presenting severity criteria were transferred to MDH, where they were reassessed and admitted if necessary. Severe malaria in admitted patients was treated with parenteral quinine for a minimum of 6 doses, followed by SP or SP+AS when the first line treatment changed or continued up to 21 doses if given as monotherapy. Children presenting a haemoglobin (Hb) <10 g/dL on the full blood counts performed during the study cross-sectional samplings, were treated with oral iron for one month.
Safety surveillance of SAEs, including a detailed evaluation of any skin reactions potentially related to SP, was done through the PCD and during study visits. When a child presented with a skin reaction, a specific questionnaire was filled out and pictures of the lesion were taken and assessed by a study physician. SAEs were followed up by a study physician and the LSM and the DSMB reviewed all skin reaction and SAE data. Axillary temperature, weight and length were measured during five cross-sectional surveys at approximately 2.5 months of age (cross-sectional 1, at the intervention A visit), 5.5 months (cross-sectional 2, at the intervention D visit), 10.5 months (cross-sectional 3, 4 weeks after intervention H), 15 months (cross-sectional 4) and 24 months (cross-sectional 5).
Blood samples were collected at delivery, during each cross-sectional survey, at the first clinical malaria episode and one month later. At delivery, the following samples were collected: 10 mL of peripheral venous blood (into an EDTA vacutainer), two blood slides and two drops of blood on filter paper from the mother, 8 mL of blood (into an EDTA vacutainer) and two drops of blood on filter paper from the cord, and a sample of placental tissue. When the delivery occurred outside the maternity, only the mother's sample was collected during the enrolment visit. During the cross-sectional visits, a 1 mL blood sample was collected into EDTA microtainers by finger-prick, blood spots were also collected on filter paper at cross-sectional visits 1 and 5 to determine parasitaemia by PCR and two blood smears were collected at visit 5.
Blood slides were read to quantify parasitaemia following standard quality-controlled procedures at the CISM laboratory. Blood films were air-dried, Giemsa-stained, and examined using a light microscope fitted with a 100× oil immersion lens and a 10× eyepiece. Parasite density was assessed by counting the number of asexual stage parasites until 500 leukocytes or parasites had been counted. Slides were declared negative only after 2000 leukocytes had been counted. Parasite numbers were converted to a count/µL by assuming a standard leukocyte count of 8000/µL. All sides were read by two independent microscopists and a third reading was performed if there was discrepancy in positivity or the ratio of densities from the two readings was more than 1.5 or the absolute difference was >10 parasites/µL. The final result was based on the definitive verdict for positivity or on the geometric mean of the positive densities for positive slides. Haematocrit was measured in heparinised microcapillary tubes with a Hawksley haematocrit reader (Hawksley & Sons Ltd, Lancing, UK) after centrifugation with a microhaematocrit centrifuge. Full blood counts were performed using a Sysmex KX-21N cell counter (Sysmex Corporation, Kobe, Japan). Tissue samples were collected from the maternal side of the placenta and placed into 10% neutral buffered formalin. Biopsies were processed and analyzed following standard procedures to evaluate presence of parasites or pigment in the placenta
Based on results from a previous chemoprophylaxis trial
Randomization was done at CISM using Stata software 7 (College Station, TX, USA) by a statistician who was not involved in other study procedures. The code was kept by the DSMB and released to the investigators once databases had been cleaned and locked after completion of follow-up. Questionnaires were double entered into databases using a program written in Fox Pro (Microsoft Corp., Seattle, WA, USA) at CISM.
The primary case definition of a clinical malaria episode was measured axillary temperature ≥37.5°C or history of fever within the prior 24 hours plus the presence of
The according-to-protocol (ATP) cohort included children who completed > = 75% of the interventions (≥6 of 8 interventions completed), and should there be 2 missing interventions these should not be consecutive and should not be both in the first 3 months (interventions A to C). The Total cohort included children who had received at least one intervention and the intention-to-treat (ITT) cohort included all randomized children.
Analyses were performed according to a predefined analytical plan using Stata/SE 10.1 (College Station, TX, USA). The primary endpoint of the study was the time to first or only episode of clinical malaria (according to the primary case definition) in the second year of follow up detected by PCD in the ATP cohort. The global comparison between the 3 groups and pairwise comparisons of the 3 groups are presented. Secondary endpoints included different case definitions, including multiple episodes of malaria, time to first or only episode of anaemia, total hospital visits and prevalence of parasitaemia and anaemia at different time points. Analyses were also performed in the Total and ITT cohorts and exploratory endpoints included time to first or only episode of clinical malaria during the different intervention periods of the first year of life detected by ACD or PCD in the ATP cohort.
Time at risk was calculated as the number of PYAR from the date or expected date of intervention A until the end of follow-up, migration, death or withdrawal of consent, whichever occurred first. An arbitrary lag period of 28 days was applied after a case of clinical malaria, during which children did not contribute to the time at risk. In the ATP cohort, children who were absent from the study area for >3 months did not contribute to the time at risk after the migration date.
Cox regression was used to evaluate the effect of the study intervention on the risk of first or only clinical malaria or anaemia episode. Negative binomial regression was used to evaluate the effect of the intervention on the incidence of multiple malaria episodes and total hospital visits. Logistic regression was used to compare prevalences. In the ATP cohort, the intervention effect was adjusted for use of insecticide treated nets (ITNs), indoor residual spraying (IRS), exclusive breastfeeding at 5.5 months, birth weight (Kg) and gender. Missing values in continuous covariates were imputed using the prediction from the best available subset of present data (determined by best-subset regression) in order to avoid losing observations and to compare the crude and adjusted models
Continuous variables were compared by ANOVA and categorical variables were compared by Chi-squared test or Fisher's exact test, as appropriate. The global comparison of the interventions was evaluated using a likelihood ratio test and a global p-value for significance. To control for family-wise error, the Bonferroni correction for multiple testing was applied to the p-values from pairwise comparisons between intervention groups. Transformation of child anthropometric data to z-scores were performed using the LMS method
Three hundred and forty-nine children were recruited and randomized into 3 groups (early exposure, late exposure and control).
Mother's variables | Intervention group | p-value | |||
Late exposure | Early exposure | Control | |||
Age (years) |
25.1 (7.1) [102] | 24.3 (7.0) [92] | 24.9 (6.4) [101] | 0.714 |
|
Parity |
Primigravidae | 19 (18.6%) | 23 (25.0%) | 27 (26.2%) | 0.726 |
Multigravidae | 73 (71.6%) | 62 (67.4%) | 68 (66.0%) | ||
Unknown | 10 (9.8%) | 7 (7.6%) | 8 (7.8%) | ||
Haemoglobin (g/dL) |
11.4 (2.4) [80] | 11.3 (2.4) [80] | 11.1 (2.2) [86] | 0.727 |
|
Peripheral |
No | 93 (91.2%) | 81 (88.0%) | 95 (94.0%) | 0.221 |
Yes | 7 (6.9%) | 10 (10.9%) | 3 (3.0%) | ||
Unknown | 2 (1.9%) | 1 (1.1%) | 3 (3.0%) | ||
Peripheral |
No | 73 (71.6%) | 73 (79.3%) | 77 (76.2%) | 0.670 |
Yes | 27 (26.5%) | 18 (19.6%) | 21 (20.8%) | ||
Unknown | 2 (1.9%) | 1 (1.1%) | 3 (3.0%) | ||
Parasite density (parasites/µL) |
7693.5 (20357.9) [7] | 2473.9 (7149.5) [10] | 1079.5 (2974.7) [3] | 0.552 |
|
Placental infection (histology) |
Not infected | 58 (56.9%) | 60 (65.2%) | 72 (71.3%) | 0.072 |
Past Infection | 16 (15.7%) | 15 (16.3%) | 11 (10.9%) | ||
Acute Infection | 0 (0%) | 3 (3.3%) | 1 (1.0%) | ||
Chronic Infection | 0 (0%) | 1 (1.1%) | 1 (1.0%) | ||
Unknown | 28 (27.4%) | 13 (14.1%) | 16 (15.8%) |
: Arithmetic Mean (SD) [n].
: ANOVA.
: n (%).
: Fisher's exact test.
: Geometric Mean (SD).
: Chi-squared test.
Children's variables | Intervention group | ||||
Late exposure (n = 102) | Early exposure (n = 92) | Control (n = 103) | p-value | ||
Gender |
Male | 53 (52.0%) | 44 (47.8%) | 51 (49.5%) | 0.845 |
Female | 49 (48.0%) | 48 (52.2%) | 52 (50.5%) | ||
Birth season | Dry | 37 (36.3%) | 30 (32.6%) | 34 (33.0%) | 0.836 |
Wet | 65 (63.7%) | 62 (67.4%) | 69 (67.0%) | ||
Age at intervention A (months) |
2.6 (0.1) | 2.6 (0.1) | 2.6 (0.1) | 0.658 |
|
Birthweight (Kg) |
2.99 (0.42) | 2.98 (0.46) | 3.02 (0.38) | 0.788 |
|
Exclusive breastfeeding at cross-sectional 2 (5.5 months) |
No | 53 (52.0%) | 56 (60.9%) | 64 (62.1%) | 0.385 |
Yes | 47 (46.1%) | 32 (34.8%) | 36 (35.0%) | ||
unknown | 2 (1.9%) | 4 (4.3%) | 3 (2.9%) | ||
Use of ITNs |
No | 93 (91.2%) | 79 (85.9%) | 93 (90.3%) | 0.449 |
Yes | 9 (8.8%) | 13 (14.1%) | 10 (9.7%) | ||
Indoor residual spraying |
No | 46 (45.1%) | 49 (53.3%) | 64 (62.1%) | 0.050 |
Yes | 56 (54.9%) | 43 (46.7%) | 39 (37.9%) | ||
Num. of clinical malaria episodes before intervention A |
No | 102 (100%) | 91 (98.9%) | 101 (98.1%) | 0.530 |
Yes | 0 (0%) | 1 (1.1%) | 2 (1.9%) | ||
No | 87 (86.1%) | 77 (86.5%) | 87 (85.3%) | 0.969 |
|
Yes | 14 (13.9%) | 12 (13.5%) | 15 (14.7%) | ||
Weight for length z-score at cross-sectional 1 (2.5 months) |
0.73 (1.34) [101] | 0.52 (1.65) [87] | 0.54 (1.31) [99] | 0.527 |
|
Weight for age z-score at cross- sectional 1 (2.5 months) |
0.30 (0.94) [101] | 0.27 (1.26) [88] | 0.44 (1.00) [102] | 0.479 |
|
Length for age z-score at cross- sectional 1 (2.5 months) |
−0.55 (1.15) [101] | −0.51 (1.16) [89] | −0.33 (1.17) [101] | 0.332 |
: n (%).
: Chi-squared test.
: Arithmetic Mean (SD) [n].
: ANOVA.
: Fisher's exact test.
: Log regression model using likelihood ratio test.
Controlling the age of first exposure was successfully achieved with highly efficacious SP+AS chemoprophylaxis, as shown by the incidence of malaria during the intervention periods. The incidence of malaria in the late exposure group during the first intervention period (from intervention A to one month after intervention C) was 0 (0 episodes/23.97 PYAR) and in the early exposure group during the second intervention period (from intervention D to one month after intervention H) was 0.06 (2 episodes/35.72 PYAR).
There were 98 first episodes of clinical malaria meeting the primary case definition during the second year of life in the ATP cohort (36, 34 and 28 in the late exposure, early exposure and control groups respectively), yielding incidences of 0.50, 0.51 and 0.35 episodes per PYAR respectively, which were not significantly different between the 3 groups (p = 0.261 for the crude overall comparison of the 3 groups and p = 0.379 for the adjusted comparison;
Late exposure group (n = 100) | Early exposure group (n = 90) | Control group (n = 102) | p-value |
|||||||
Outcomes | Episodes | PYAR | Incidence | Episodes | PYAR | Incidence | Episodes | PYAR | Incidence | |
|
36 | 72.66 | 0.50 | 34 | 66.58 | 0.51 | 28 | 80.53 | 0.35 | 0.379 |
|
88 | 90.05 | 0.98 | 91 | 80.12 | 1.14 | 79 | 92.59 | 0.85 | 0.833 |
|
27 | 79.18 | 0.34 | 32 | 68.27 | 0.47 | 23 | 84.38 | 0.27 | 0.231 |
|
27 | 79.18 | 0.34 | 31 | 68.33 | 0.45 | 23 | 85.32 | 0.27 | 0.282 |
|
20 | 83.92 | 0.24 | 27 | 71.22 | 0.38 | 21 | 86.38 | 0.24 | 0.244 |
|
13 | 89.10 | 0.15 | 10 | 81.50 | 0.12 | 13 | 92.01 | 0.14 | 0.848 |
|
365 | 69.73 | 5.23 | 334 | 62.40 | 5.35 | 338 | 74.34 | 4.55 | 0.335 |
|
18 | 95.21 | 0.19 | 17 | 85.42 | 0.20 | 18 | 97.00 | 0.19 | 0.910 |
PYAR: person-years at risk.
Adjusted by use of ITNs, indoor residual spraying, exclusive breastfeeding at 5.5 months, birthweight (kg) and gender.
p-value from Cox regression model using Likelihood Ratio Test.
p-value from negative binomial regression model using Likelihood Ratio Test.
Late exposure vs. control group | Early exposure vs. control group | Early vs. late exposure group | |||||||
Outcomes | HR or IRR |
95% CI | p-value |
HR or IRR |
95% CI | p-value |
HR or IRR |
95% CI | p-value |
|
1.38 | 0.83–2.28 | 0.642 |
1.35 | 0.81–2.24 | 0.743 |
0.98 | 0.61–1.59 | 1.000 |
|
1.14 | 0.57–2.26 | 1.000 |
1.23 | 0.62–2.44 | 1.000 |
1.08 | 0.54–2.16 | 1.000 |
|
1.22 | 0.69–2.16 | 1.000 |
1.60 | 0.93–2.75 | 0.273 |
1.31 | 0.77–2.21 | 0.959 |
|
1.23 | 0.70–2.17 | 1.000 |
1.55 | 0.90–2.68 | 0.343 |
1.26 | 0.74–2.15 | 1.000 |
|
0.92 | 0.49–1.71 | 1.000 |
1.47 | 0.82–2.62 | 0.581 |
1.60 | 0.89–2.89 | 0.359 |
|
1.16 | 0.52–2.59 | 1.000 |
0.91 | 0.40–2.11 | 1.000 |
0.79 | 0.34–1.82 | 1.000 |
|
1.15 | 0.92–1.44 | 0.666 |
1.16 | 0.93–1.46 | 0.549 |
1.01 | 0.81–1.27 | 1.000 |
|
1.14 | 0.50–2.62 | 1.000 |
1.19 | 0.52–2.69 | 1.000 |
1.04 | 0.46–2.36 | 1.000 |
95% CI: 95% confidence interval.
Hazard ratio (HR) or incidence rate ratio (IRR) adjusted by use of ITNs, indoor residual spraying, exclusive breastfeeding at 5.5 months, birthweight (kg) and gender.
p-values corrected by Bonferroni.
p-value from Cox regression model using Likelihood Ratio Test.
p-value from negative binomial regression model using Likelihood Ratio Test.
The prevalence of parasitaemia by microscopy at the last cross-sectional sampling (24 to 26 months of age) also showed no differences between the three groups (13.04%, 13.95% and 7.45% in the late exposure, early exposure and control groups respectively, p = 0.302;
Late exposure | Early exposure | Control | p-value | ||
Haemoglobin (g/dL) |
11.4 (2.4) [80] | 11.3 (2.4) [80] | 11.0 (2.2) [88] | 0.569 | |
Moderate or severe anaemia |
No | 86 (93.5%) | 80 (93.0%) | 83 (88.3%) | 0.388 |
Yes | 6 (6.5%) | 6 (7.0%) | 11 (11.7%) | ||
No | 80 (87.0%) | 74 (86.1%) | 87 (92.5%) | 0.302 |
|
Yes | 12 (13.0%) | 12 (13.9%) | 7 (7.5%) | ||
No | 64 (69.6%) | 55 (63.9%) | 55 (58.5%) | 0.290 |
|
Yes | 28 (30.4%) | 31 (36.1%) | 39 (41.5%) | ||
Parasite density (parasites/µL) |
2139.4 (4425.1) [12] | 1959.9 (4788.6) [12] | 6594.8 (19096.2) [7] | 0.535 |
|
Weight for length z-score at cross-sectional 1 |
−0.33 (1.18) [91] | −0.36 (1.23) [86] | −0.42 (1.23) [91] | 0.888 |
|
Weight for age z-score at cross-sectional 1 |
−1.36 (1.14) [93] | −1.32 (1.17) [84] | −1.24 (1.33) [93] | 0.804 |
|
Length for age z-score at cross-sectional 1 |
−1.40 (1.09) [91] | −1.46 (1.06) [86] | −1.19 (1.17) [91] | 0.226 |
: Arithmetic Mean (SD) [n].
: ANOVA.
: n (%).
: Log regression model using likelihood ratio test.
: Geometric Mean (SD).
There were no safety concerns during the trial. There were 62 skin reactions during the intervention period (up to one month from last dose) in the total cohort, but only one (urticaria with pruritus) was assessed to be possibly related to the intervention and all resolved without sequelae. The occurrence of SAEs was similarly distributed among the groups, and none was deemed related to study drugs. There were 5 deaths during the complete follow up, none considered related to the intervention and 3 occurring in children who had not yet received an intervention.
This study used monthly antimalarial chemoprophylaxis to control exposure to the erythrocytic-stage antigens of the
Compliance with study drugs among participants was high, and chemoprophylaxis was well tolerated and highly efficacious, as proven by the near zero incidence of malaria during the different time periods in which children received SP+AS. Therefore, from two to twelve months of age the time of exposure to the parasite was decreased by 50% in the early exposure group and by 30% in the late exposure group compared to the control group. However, we did not find any significant differences in the outcomes between the three different study groups. Indeed, there were no significant differences between the groups in the incidence of clinical malaria, anaemia, total outpatient visits or hospital admissions during the second year of life. Also, there were no differences in the prevalence of parasitaemia or anaemia at the end of follow up.
Although the incidence of malaria during the second year was higher by about 40% in the late exposure and early exposure groups when compared to the control group, and the risk of malaria was highest in the early exposure group for the more specific case definitions, none of these differences reached statistical significance. The sample size had been calculated to detect a relative risk of 2 between the groups, based on results from a previous continuous chemoprophylaxis trial
Our hypothesis was based on results from previous malaria chemoprophylaxis and vaccine trials conducted in Tanzania in infants and young children. Data showed that infants receiving IPTi
Taking together the results of the three trials, data suggest that contact with the parasite is needed at some point during infancy to be able to develop NAI, but that the timing of the exposure does not seem to be a major contributing factor. Clearly, no differences in the incidence of malaria during the second year were seen between the early and late exposure groups, and young infants with relatively immature immune systems and low exposure were able to build NAI as well as older infants. With available data it is not possible to establish the amount of exposure that is needed for the development of NAI, although exposure in our trial and in the Tanzanian IPTi trial varied from 5 to 7 months (from 2 to 12 months of age). Infants in these two trials received comparable challenge during the first year (the incidence of multiple malaria episodes during the first year was 0.33 episodes per PYAR in the control group in the former (data not shown) and 0.43 in the latter
Overall, no evidence was found that the age of first exposure to the parasite during the first year of life affects the rate of acquisition of NAI against malaria. Thus the timing of administration of antimalarial interventions like malaria vaccines during infancy does not appear to be a critical determinant, as infants should be equally able to mount an adequate immune response both during the initial months of life and later. Indeed, recent results from phase I/IIb trials of the RTS,S/AS0 candidate malaria vaccine conducted in endemic areas of Africa support this. In the RTS,S/AS0 trials performed after our trial had started, VE was similar in three trials in which the vaccine was administered to infants at different ages: 10, 14 and 18 weeks of age, staggered with EPI vaccines (VE against infection 65.9, 95% CI 42.6–79.8; p<0.0001)
Secondary objectives of the study included assessment of type and quality of immune responses, oxidative stress markers and host genetic factors, aiming at describing these outcomes according to age of first exposure and correlating them with clinical malaria during the second year of life. Results of these analyses will be presented in other specific papers and will hopefully shed light on the age-dependency of the immune responses and how those correlate with protection or susceptibility against malaria. Any step towards better understanding the development of NAI in infancy will enhance our chances of improving the current preventive tools and strategies against this deadly parasite.
Therefore, from this study we can reasonably conclude that no evidence was found that the age at which the first encounter with the parasite occurs during infancy is a key determinant in the build up of immunity. Secondly, considerable reductions in exposure to parasite antigens in the first year of life do not result in large rebound in the incidence of malaria during the second year and massive reductions in exposure are needed to see a significant impact on the acquisition of NAI.
(DOC)
(PDF)
We thank all children and their families for their participation in the study; the field workers, field supervisors, laboratory staff, data managers and other staff at CISM for their work during the study; Roche and Sanofi-Aventis laboratories for providing the study drugs and placebos; Sònia Tomàs and Patricia García for their work as project managers; Alfons Jiménez for conducting all qPCRs; Laura Puyol and Pau Cisteró for laboratory support; Jaume Ordi for the training and quality control for the placental histology readings; Lázaro Mussacate Quimice for reading all placental histologies; Esperança Sevene, who acted as Local Safety Monitor; and Climent Casals-Pascual and Judy Epstein for significant contribution to initial protocol development and grant applications.