Characteristics of the infants: The 70 infants enrolled in the study were healthy newborns, with an average gestational age of 39.2 ± 2.0 weeks and weighing 3,177 ± 484 grams (Table 1). Only 7.9% of the babies were low birth weight (LBW), although 46.6% of the mothers were PM-positive (PM+) at delivery. The babies were followed longitudinally through the first year of life, with a mean follow-up time of 341 ± 42 days. Babies were seen an average of 8.5 times (range 6–9) during the first year of life.
Table 1
Description of infants residing in Ngali II
Number of infants | 70 |
Gravidity (median, range) | 3 (range 1–12) |
Gestational age at birth (weeks)* | 39.2 ± 2.0 |
Percent full-term births | 89.9% |
Birth weight (grams)* | 3,177 ± 484 |
Percent low-birthweight (< 2,500 g) | 7.9% |
% of women with placental malaria (PM-pos) | 46.6% (27/58) |
Mean number of visits per infant | 8.5 (range 6–9) |
Duration of follow-up (days from birth)* | 341 ± 42 |
Mean number of times malaria-positive by Microscopy (range) | 2.7 ± 2.1 (range: 0 to 7) |
Mean number of times malaria-positive by PCR (range) | 3.9 ± 2.1 (0 to 8) |
*Mean ± SD |
P. falciparum infections during the first year of life: At the first time point (mean 9 ± 2 days), only 3% (3/69) of the newborns were malaria-positive by microscopy and 7% (5/70) by PCR, suggesting few congenital infections (Fig. 1). The number of infants infected with P. falciparum increased rapidly over the first 4 months of life and then remained relatively constant thereafter. Overall, P. falciparum infections were detected in infants 2.7 ± 2.1 (range 0 to 7) and 3.9 ± 2.1 (range 0 to 8) times by microscopy and PCR, respectively, during the year (Table 1).
Time until the first infection: Longitudinal data identified when each baby had their first P. falciparum infection that were i) submicroscopic (i.e., PCR-positive, slide-negative), ii) detected by microscopy, and iii) a symptomatic episode of malaria (Fig. 2). Overall, first infections of 29 babies were submicroscopic and occurred between the 2nd and 6th month of life (Fig. 2 below). Among these 29 babies, 83% (24/29) were malaria-negative at the next time point (~ 1 month later) by microscopy, showing these babies intial infects were transient and self-resolving with very low parasitemia. On the other hand, the first microscopically-detected infection occurred in 86% of the babies (60/70, including some babies who had prior sub-microscopic infection) between 2 to 12 months of life (median of 6 months); while, 14% of the babies were malaria-negative by PCR and microscopy throughout the first year of life. Only 26% (18/70) of the babies developed clinical cases of malaria (i.e., slide-positive for P. falciparum and fever), with the first episode most often occurring after 6–8 months of age.
Decline of maternal IgG: The decline of maternal IgG Ab to 8 malarial antigens (11 recombinant proteins) during the first year of life are shown in Table 2 and Fig. 3. At ~ 1 week of age, babies had high (> 10,000 MFI) levels of maternal IgG Ab to AMA1, EBA-175, (Fig. 3A-C); intermediate (> 1,000 but < 10,000 MFI) Ab levels MSP1-42 (3D7, FVO) and MSP2 (3D7, FC27) (Fig. 3D-G), and very low levels (< 1,000 MFI) to MSP3, RESA, LSA-1 and CSP (Fig. 3H-K). Overall, 80–97% of newborns had maternal IgG Ab to AMA1 (3D7, FVO) and EBA-175 that were detectable for 6 to 8 months, with Ab half-lives ranging from 0.98 to 1.9 months based on data for the first 6 months of life (Table 2). On the other hand, Abs to MSP1-42 and MSP2 declined quickly to background (cut-off) levels by 3 months of age. Thereafter, IgG Ab levels to MSP1-42 and MSP2 quickly increased as the babies produced Abs when they became infected with P. falciparum. Ab half-life estimates for the first 6 months of life was 0.7 months for MSP2, but the half-life to MSP1-42 could not be determined over this period since infant produced IgG to MSP1-42 as early as 3–4 months of age. On the other hand, most infants had very low or no Ab levels to MSP3, RESA, LSA1, and CSP (Fig. 3H-K). As a result, a decrease in maternal IgG over time was not observed, preventing an estimate of Ab half-live.
Table 2
Characteristics of maternal IgG transplacentally-transferred to babies
| Amount of Ab MFI at day 9 ± 2 (Median, 1st, 3rd Quartile) | Percentage of Ab-positive Babies ** | Ab Half-life, ± 95% CI (months)^ | Length of Detectable Maternal IgG (months) |
AMA1 (3D7) | 11,757 (8765, 14108) | 96.9 | 1.46 (1.17, 1.94) | 6 |
AMA1 (FVO) | 12,581 (10218, 14870) | 96.9 | 1.88 (1.4, 2.59) | 8 |
EBA-175 | 10,800 (3635, 12460) | 79.7 | 0.98 (0.83,1.21) | 6 |
MSP1-42 (3D7) | 1,003 (10, 3868) | 35.9 | NC# | NC |
MSP1-42 (FVO) | 4,637 (615, 11207) | 67.2 | NC | 3 |
MSP2 (3D7) | 6,235 (2298, 10,163) | 76.6 | 0.78 (0.67, 0.92) | 3 |
MSP2 (FC27) | 6,935 (2325, 10163) | 81.3 | 0.73 (0.64, 0.86) | NC |
MSP3 | 239 (0, 1482) | 31.3 | 2.41 (1.58, 5.04) | NC |
RESA | 0 (0,0) | 0 | NC | NC |
LSA-1 | 0 (0,218) | 23.4 | NC | NC |
CSP | 0 (0, 352) | 25.0 | 2.39 (1.64, 4.36) | NC |
** Percent of babies at 9 ± 2 days with MFI with Ab above cut-off: AMA1(3D7) 1,970 MFI; AMA1 (FVO) 2,050 MFI; EBA-175 (2,020 MFI; MSP1(3D7) 1,500MFI; MSP1(FVO) 1,800 MFI; MSP2(3D7) 1,980 MFI; MSP2(FC27) 1,880 MFI; MSP3 842 MFI; RESA 390 MFI; LSA-1 380 MFI; CSP 490 MFI. ^based on IgG Ab levels between Day 9 ± 2 after birth until 6 months of age #Not Calculatable, because no consistent decline occurred or Ab were produced prior to 6 months |
Relationship between maternal IgG levels at birth and primary parasitemia: Ab levels at the first visit were compared between babies whose first infections were submicroscopic (n = 18), i.e., very low parasitemia, and those with slide-positive infections (n = 52, range 21–150,000 IE/µl), i.e., higher parasitemia. No difference in median MFI was found (all p values > 0.05), except for one antigen where MFI were higher in the slide-positive group (sFig. 1: end of manscript). Thus, the data suggest that high maternal IgG levels to any the 8 antigens at birth did not help reduce parasitemia to submicroscopic levels.
The IgM antibody response of infants during the first year of life. IgM Ab data were available for 47 babies who had ≥ 7 visits. The results allowed us to answer the following three questions:
Is the parasitemia high enough during the first submicroscopic infection to induce an IgM Ab response? The first infection was submicroscopic in 38% (18/47) of these babies. At that time, relatively few babies had IgM Ab, with i) 35–44% of babies having IgM Abs to AMA-1 (3D7); ii) 27–33% having Ab to AMA-1 (FVO), MSP-1 (FVO), EBA-175, MSP-2 (3D7, FC27); and iii) only 11–22% having Abs to RESA, MSP3, LSA-1 and CSP (Table 3). Since the majority of babies were slide-negative for P. falciparum at the next visit (~ 1 month later), these early transient submicroscopic infections induced an IgM response in only a small proportion of infants. On the other hand, when the first infection was high enough to be detected by microscopy, 68–82% of the babies had IgM to the 8 antigens (Table 3). Thus, submicroscopic infections could induce IgM in some babies, most babies produced IgM to multiple antigens when they had parasitemia detectable by microscopy.
Did the infants produce IgM Ab to all 8 malarial antigens or is the response more restricted? As shown in Table 3, babies with primary submicroscopic infections responded to a restricted repertoire of antigens, showing that some antigens were more immunogenic than others, e.g., 44% of babies with submicroscopic infections produced IgM to AMA1; whereas, only 11% produced IgM to RESA and MSP3. On the other hand, most babies had IgM Abs to most (i.e., 59–82%) of the 8 antigens at the time of their first slide-positive infection (Table 3). Although most babies had Abs to many antigens, variation occurred among the babies, e.g., one baby might have Abs to AMA1, but lack Abs to RESA; whereas, another baby would have Abs to MSP1, but not EBA-175. At the time of the first symptomatic malaria episode, which usually occurred after infants had had several asymptomatic infections, 64–91% of infants had IgM Abs to the 8 malarial antigens (Table 3). Thus, following the first submicroscopic infection the response was quite restricted, but when parasitemia was high enough to be detected by microscopy most babies produced IgM to most of the antigens studied.
Table 3
Percentage of infants positive for IgM antibodies at their first submicroscopic and slide-positive infections, as well as first symptomatic clinical episode
Antigens | Submicroscopic Infection* (n = 18) | Slide- Positive Infections** (n = 31) | First Clinical Episode (n = 11) |
AMA-1 (3D7) | 44.4 (8/18) | 76.9 (30/39) | 73 |
AMA-1 (FV0) | 33.3 (6/18) | 82.1 (32/39) | 91 |
EBA-175 | 27.8 (5/18) | 66.7 (26/39) | 73 |
MSP-1 (3D7) | 22.2 (4/18) | 76.9 (30/39) | 91 |
MSP-1 (FVO) | 33.3 (6/18) | 79.5 (31/39) | 91 |
MSP-2 (3D7) | 27.8 (5/18) | 79.5 (31/30) | 91 |
MSP2 (FC27) | 27.8 (5/18) | 74.4 (29/39) | 82 |
MSP-3 | 11.1 (2/18) | 59.0 (23/39) | 91 |
RESA | 11.1 (2/18) | 71.8 (28/39) | 64 |
LSA-1 | 16.7 (3/18) | 74.4 (29/39) | 91 |
CSP | 22.2 (4/18) | 69.2 (27/39) | 82 |
*PCR-positive, but slide negative. **First slide-positive infection |
Do infants produce IgM Abs before maternal IgG levels “wane,” i.e., reach background levels? To answer this question, the primary IgM response (i.e., the first visit when the MFI was above the cut-off) was compared with the corresponding IgG level to determine if the MFI of maternal IgG was still positive (above cut-off). Although the first IgM response often occurred in babies who either had not received maternal IgG for the antigen or after maternal Ab had waned (dropped below the IgG cut-off), a significant proportion of infants produced their first IgM response while significant amounts of maternal IgG were still present (Table 4). For example, over half the babies had maternal IgG to MSP1, 60.9% to EBA-175, and > 76% had IgG to AMA1 at the time they first produced IgM to these antigens (Table 4). Examples of the production of infant IgM in the presence of maternal IgG are shown in Fig. 4. Thus, babies can produce a humoral response to antigens even when maternal IgG levels remain positive.
Table 4
Percentage of infants who produced an IgM response in the presence of maternal IgG
Antigens | AMA-1 3D7 | AMA-1 FVO | EBA-175 | MSP-1 3D7 | MSP-1 FVO | MSP 2 FC29 | MSP3 |
| 27.3% (12/44) | 34.1% (15/44) | 60.9% (28/46) | 77.8% (35/44) | 76.2% (35/46) | 36.6% (15/41) | 13.3% (6/45) |
Results for RESA, LSA1 and CSP are not included because Ab levels to these antigens were very low and no sustained downward trend in Ab levels was observed over the first few months of life |
The IgG antibody response at first post-natal infection: Because of declining maternal IgG Abs in young babies, it is often difficult to determine when a baby first produces IgG Abs after infection (i.e., a rise in IgG Abs). Accordingly, a longitudinal regression models was created that took the decline of maternal IgG into consideration when assessing if a baby produced IgG upon primary i) submicroscopic infection, ii) slide-positive infection, and iii) when the infant has its first clinical episode of malaria (Table 5). Data included in the analysis were from all 70 infants. Positive regressing results are shown in Table 5, where coefficients with significant p values indicate that an IgG response had occurred. Among the babies with primary submicroscopic infections, IgG Abs were only produced to AMA1 (3D7, FVO). However, when parasitemia were adequate to be detected by microscopy, IgG Abs to 5 antigens were produced, including MSP1 (3D7, FVO), MSP2 (FC27), AMA1 (3D7, FVO), RESA and LSA1, but not to MSP2 (3D7), EBA-175, and CSP. A similar pattern was found in the 18 infants who developed clinical episodes of malaria later in life (Table 5). Taken together, data show that upon initial submicroscopic infections, babies produced IgG Ab to AMA1; however, when parasitemia reached detectable levels by microcopy IgG Ab to most antigens were detected.
Table 5
IgG response upon primary P. falciparum infection(Coefficient, p-value)
| Sub-microscopic* | Slide-Positive | Clinical Episode |
AMA1(3D7) | 0.81; p = 0.036 | 1.17; p = 0.004 | |
AMA1 (FVO) | 0.970; p = 0.019 | 1.00; p = 0.008 | 1.73; p = 0.026 |
EBA-175 | | | |
MSP1 (3D7) | | 2.30; p < 0.0001 | 2.14; p = 0.033 |
MSP1 (FVO) | | 2.04; p = 0.001 | |
MSP2 (3D7) | | | 4.01; p < 0.001 |
MSP2 (FC27) | | 0.97; p = 0.047 | 3.26; p < 0.001 |
MSP3 | | 1.23; p = 0.011 | 1.65; p = 0.048 |
RESA | | 1.31; p < 0.001 | 2.42; p = 0.001 |
LSA1 | | 1.28; p = 0.003 | 1.76; p = 0.023 |
CSP | | | 1.82; p = 0.007 |
*Excluded all babies whose first infection was microscopically-positive and PCR-negative and microscopic-positive and PCR-positive BLANK = negative Coefficient value or non-significant p value |
IgM and IgG Ab responses were short-lived. During the first year of life, IgM and corresponding IgG levels increased one or more times and then quickly declined following recent infection (Fig. 5, sFig. 2). Some babies had second and third peaks of IgM and IgG, which also declined quickly (Fig. 5). In general, primary and secondary IgG responses were similar with minimal or no increase in Ab levels or breadth of the response (sFig. 2). Although IgG Ab levels appeared to be highest at the end of the first year, the response was often still short-lived (e.g., Baby #06 and 68 (Fig. 5B and D). Overall, a true primary versus secondary response with higher IgG levels and broader antigenic specificity was not observed in infants during the first year of life (Fig. 5, sFig. 2 (located at end of manuscript).