Respiratory Syncytial Virus Transplacental Antibody Transfer and Kinetics in Mother-Infant Pairs in Bangladesh

Abstract

Background. Pneumonia is the leading cause of childhood mortality globally. Respiratory syncytial virus (RSV) is the most important viral cause of pneumonia. Maternal serum antibody protects infants from RSV disease. The objective of our study was to characterize RSV antibody levels in mother-infant pairs.

Methods. Serial serum samples were collected from mother-infant pairs in Bangladesh from the third trimester of pregnancy to 72 weeks postpartum and tested using an RSV antibody microneutralization assay. Serologic infection was defined as a 4-fold increase in antibody titer. Maternal antibody half-life was calculated using infant antibody titers from birth to 20 weeks.

Results. The ratio of infant cord blood to maternal serum RSV antibody titers in 149 mother-infant pairs was 1.01 (95% confidence interval [CI], .99–1.03). Maternal RSV antibody titers in the third trimester and at birth were strongly correlated (R = 0.68). Antibody half-life was 38 days (95% CI, 36–42 days). Higher cord blood RSV antibody titers were associated with a lower risk of serologic infection (P = .01) and maintenance of antibody titer above a potentially protective threshold (P < .001).

Conclusions. Efficient transplacental transfer of RSV-specific antibody from mother to the fetus was documented in mother-infant pairs in Asia. Higher cord blood antibody titers were associated with protection from serologic infection.

(See the editorial commentary by Piedra and Munoz on pages 1526–8.)

Pneumonia is the leading cause of mortality in children <5 years old [1]. Respiratory syncytial virus (RSV) is the most important cause of viral pneumonia worldwide, with >95% of the disease burden in resource-limited settings [2]. No vaccine exists to prevent RSV infection, and therapeutic options are limited to supportive care. RSV primarily causes severe disease in infants aged <1 year, although it can cause repeated infections throughout life [3].

Several lines of evidence suggest that RSV serum antibody (Ab) protects against disease in infants [4]. Epidemiologic studies have shown that higher levels of RSV maternal Ab are associated with a decreased risk of infection in young infants [4–6]. Palivizumab, a monoclonal Ab directed against the fusion protein of RSV, reduces RSV hospitalizations when given as prophylaxis to high-risk infants during RSV season [7]. RSV neutralizing Ab titers between 1:64 and 1:256 (log₂ 6.0–8.0) correlate with decreased risk of hospitalization due to RSV infection in US populations [8]. Infants can develop protective, although short-lived, Ab titers in response to natural infection [9].

Maternal Ab transfer across the placenta is highly efficient for many vaccine-preventable diseases [10–12]. Maternal vaccination during pregnancy is a safe and effective strategy to protect infants against neonatal tetanus, diphtheria, and influenza [13, 14]. Maternal Ab transfer and rates of maternal Ab decay vary across populations. Factors such as human immunodeficiency virus (HIV) infection, prematurity, and low birth weight are associated with decreased transplacental Ab transfer efficiency [15–17]. In studies in the United States, the Gambia, and Brazil, the ratio of RSV Ab in cord blood to that in maternal serum is ≥1, confirming that fetal Ab levels can equal or potentially exceed maternal Ab levels [18, 19]. The decay of RSV-specific maternal Ab varies, with the half-life ranging from 21–26 days in the United States and the Netherlands to 79 days in Kenya [20–22].

One RSV fusion protein–based vaccine has been evaluated in postpartum and pregnant women in the United States; this vaccine was not immunogenic but was safe and demonstrated efficient transplacental Ab transfer [20]. With several new RSV vaccines under development, investigators are actively considering RSV vaccination during pregnancy to protect young infants [23]. The natural history of RSV Ab levels from third trimester to the postpartum period has not been previously reported.

South Asia, a region that includes India, Bangladesh, and Nepal, has one of the highest birth rates in the world. Population characteristics include household crowding, use of solid fuel, low birth weight, and malnutrition, all of which are risk factors for severe acute lower respiratory tract infection in children [24]. No prior studies have been conducted to establish the rate of transplacental transfer and the half-life of maternal RSV Ab in a South Asian cohort.

We present RSV Ab levels measured in 149 mother-infant pairs in Dhaka, Bangladesh, followed from the third trimester of pregnancy until week 72 postpartum. We describe factors that influence transplacental Ab transfer and Ab half-life in this South Asian population. These results provide longitudinal information regarding RSV Ab transfer and kinetics to inform vaccine and therapeutics development strategies in resource-limited settings.

METHODS

A randomized clinical trial of maternal immunization with influenza or pneumococcal vaccine was performed in Dhaka, Bangladesh from 2004–2005 [25]. Maternal serum samples were collected in the third trimester of pregnancy, at birth, and at 72 weeks postpartum. Cord blood samples were collected at birth, and infant serum samples were collected at 6, 10, 16, 20, 24, and 72 weeks of age. Residual sera from all time points were available from a subset of 149 mother-infant pairs who had completed follow-up to the end of the 72-week period. These samples were tested for RSV-neutralizing Ab. Sociodemographic and clinical information were collected as part of the parent study. Small-for-gestational-age (SGA) was defined according to US birthweight data [26].

RSV neutralizing Ab titers were measured using a microneutralization assay, using serial 2-fold dilutions of serum against the RSV A2 strain in Hep-2 cells incubated in 96-well plates at 37°C for 72 hours in 5% CO₂. Color development was performed using a RSV-specific mouse monoclonal primary Ab and a horseradish peroxidase–conjugated goat anti-mouse immunoglobulin G (IgG) secondary Ab. Neutralization titer was defined as the minimal dilution of serum resulting in 50% color reduction, compared with a positive control [27]. All Ab levels are expressed as log₂ titers. For our analysis, we considered serologic infection as a 4-fold rise in Ab titer between any 2 time points [27]. We defined a potentially protective threshold titer as a 1:256 (log₂ 8.0) Ab titer [8, 9]. Because it is not possible to separate passively acquired maternal Ab from the infant's actively produced Ab using standard serologic methods, we additionally performed a sensitivity analysis censoring infants at the point where there was a 2-fold rise or no decrease in Ab titer. Infants <6 months of age often do not generate a 4-fold serologic response to RSV infection [22, 28]. Maternal Ab titers decline over time, and lack of a decline could itself indicate acquisition of RSV infection in the infant.

For comparison of baseline variables, a 2-sample t test or a χ² test was used for continuous or categorical measures, respectively. The ratio of infant to maternal Ab titers was calculated using a linear regression model. Examination of the effect of covariates on the Ab ratio was also performed using linear regression. We additionally performed multivariate analysis that included maternal age, maternal education duration, and maternal parity to examine for the effect of potential confounding by age. Linear mixed effects modeling was used to assess the trajectory of infant log₂ Ab titers from birth to 20 weeks to account for within-participant association over time and to evaluate the effect of covariates on Ab half-life. The half-life of Ab titer was calculated as the time at which the predicted Ab titer would decrease by 50% from baseline. A sensitivity analysis for Ab half-life was additionally performed using values from birth to 10 weeks. Survival analysis was performed using Kaplan–Meier estimates to evaluate the effect of covariates on time to rise in Ab titer and reduction below a protective threshold titer. For evaluation of serologic infection, outcomes were censored at the last observed visit if no rise occurred. For evaluation of Ab half-life and reduction in Ab titer below a protective threshold, outcomes were censored at the time when Ab titers quadrupled, at which time it was assumed to indicate potential acquisition of new infection and the end of the ability to measure maternal Ab alone [29, 30]. The median time to reduction below a protective threshold titer was computed using a maximum likelihood model in which exponential Ab decay was assumed.

This study was approved by the institutional review boards at Seattle Children's Hospital and Cincinnati Children's Hospital.

RESULTS

Of the 340 mother-infant pairs in the original clinical trial, serial serum samples from a subset of 149 (44%) were tested for RSV-neutralizing Ab, with 1481 laboratory results. In 9 samples, quantities of sera were insufficient for testing. Baseline sociodemographic and clinical data for the selected and unselected mother-infant pairs are shown in Table 1. Compared with the unselected mother-infant pairs, there were higher rates of nulliparity and lower rates of prematurity in the selected subset. Women were enrolled in the parent study from August 2004 through May 2005, accounting for the uneven distribution of births across seasons. Median maternal age at enrollment was 25 years (range, 18–36 years), with a median maternal education duration of 12 years (range, 0–16 years). Median maternal parity was 1 (range, 0–3). Sixty-two women (42%) delivered via cesarean section. Median birth weight was 3 kg (range, 2–4.9 kg). Fifty infants (34%) were SGA, and 5 (3%) were born at <37 weeks gestation.

Table 1.

Comparison of Baseline Sociodemographic and Clinical Characteristics of 149 Mother-Infant Pairs With and 191 Pairs Without Results of Respiratory Syncytial Virus Antibody (Ab) Tests, Dhaka, Bangladesh

Characteristic	Pairs With Results	Pairs Without Results	P value
Maternal
Age, y	25 (18–36)	25 (18–36)	.42
Education duration, y	12 (0–16)	11 (0–16)	.091
Previous deliveries, no.
0	20 (24)	13 (13)	.046
1	50 (60)	61 (59)
≥2	14 (16)	29 (28)
Type of delivery
Vaginal	87 (58)	92 (49)	.093
Cesarean section	62 (42)	95 (51)
Infant
Sex
Female	65 (44)	85 (45)	.74
Male	84 (56)	102 (55)
Gestational age, wk
≥37	144 (97)	168 (90)	.011
<37	5 (3)	19 (10)
Birthweight, kg	3.0 (2.0–4.9)	3.0 (1.8–5.0)	.59
Small for gestational age	50 (34)	58 (30)	.53
Season of birth
Winter	59 (40)	78 (41)	.25
Spring	27 (18)	26 (14)
Summer	2 (1)	9 (5)
Fall	61 (41)	78 (41)
Infant cord antibody level, log2 titer	11.1 (7.8–14.3)	…

Data are no. (%) of mothers or children or median (range).

Mean maternal Ab titers in the third trimester were correlated with titers at birth and week 72 of the postpartum period (R = 0.68 and R = 0.47, respectively; Figure 1A and 2A and Table 2). Infant Ab titers declined from a peak mean value (±SD) of 11.0 ± 1.4 at birth to a nadir of 6.9 ± 1.6 at 24 weeks, with a rise by 72 weeks (mean [±SD], 9.3 ± 2.1). The ratio of cord blood to maternal Ab titers at birth was 1.01 (95% CI, .99–1.03). Maternal and infant cord blood Ab titers at birth were correlated (R = 0.70; Figure 1B).

Figure 1.

A, Comparison of log₂ maternal respiratory syncytial virus (RSV) antibody (Ab) titers in the third trimester and log₂ maternal RSV Ab titers at birth (R = 0.68). B, Comparison of log₂ maternal RSV Ab titers at birth and log₂ cord blood RSV Ab titers (R = 0.70).

Figure 2.

A, Maternal log₂ respiratory syncytial virus (RSV) antibody (Ab) titers in the third trimester, at birth, and at 72 weeks postpartum. B, Infant log₂ RSV Ab titers at birth and 6, 10, 16, 20, 24, and 72 weeks of age.

Table 2.

Respiratory Syncytial Virus Antibody (Ab) Titers in Mothers and Infants From Third Trimester to 72 Weeks Postpartum

Visit Time	Mother		Infant
	Log2 Ab Titer, Mean ± SD	Geometric Mean Titer	Log2 Ab Titer, Mean ± SD	Geometric Mean Titer
Third trimester	11.1 ± 1.2	2120	…	…
Birtha	10.8 ± 1.3	1760	11.0 ± 1.4	2020
Pospartum week
6	…	…	9.2 ± 1.6	590
10	…	…	8.5 ± 1.5	360
16	…	…	7.8 ± 1.6	230
20	…	…	7.3 ± 1.7	150
24	…	…	6.9 ± 1.6	120
72	11.2 ± 1.1	2360	9.3 ± 2.1	630

^a The ratio of cord blood to maternal Ab titers at birth was 1.01 (95% confidence interval, .99–1.03).

Examination of the impact of covariates on the infant cord blood to maternal Ab ratio revealed no effect of vaginal versus cesarean birth (ratios, 1.00 vs 1.03; P = .14), male versus female sex (ratios, 1.01 vs 1.02; P = .88), primiparity versus multiparity (ratios, 0.99 vs 1.01; P = .59), birthweight >3 kg versus ≤3 kg (ratios, 1.01 vs 1.02; P = .58), SGA vs not SGA (ratios, 1.02 vs 1.01; P = .75), maternal age >25 years vs ≤25 years (ratios, 1.01 vs 1.01; P = .94), or maternal education duration >12 years vs ≤12 years (ratios, 1.01 vs 1.01; P = .91).

When a 4-fold rise in Ab titer was used as a marker for serologic infection, 2 infants were infected by 10 weeks, and 11 were infected by 20 weeks. The number of infants with serologic infection as defined by any rise, a 2-fold rise, or a 4-fold rise in Ab titer at various time points are shown in the first table (Table 5) in the Supplementary Materials.

Linear regression analysis of RSV Ab half-life, in which censoring occurred at a 4-fold rise in Ab titer and values from birth to 20 weeks were used, predicted a daily decrease of 0.026 log₂ titer. On the basis of this finding, antibody half-life was calculated at 38 days (95% CI, 36–42 days) according to the formula 0.026 log₂/day*38 days = 1 log₂ (Table 3 and Figure 2B). When we calculated the Ab half-life after restricting our analysis to values from birth to 10 weeks, we found a half-life of 27 days (95% CI, 25–29 days). Using a lack of decline in Ab level as a marker of serologic infection, the half-life estimate based on values from birth to 20 weeks was 31 days (95% CI, 30–32 days), and the estimate based on values from birth to 10 weeks was 25 days (95% CI, 23–26 days; Table 3). Results from additional sensitivity analyses used to calculate RSV Ab half-life are shown in Table 3.

Table 3.

Sensitivity Analysis of Respiratory Syncytial Virus Neutralizing Antibody (Ab) Half-life, Using Ab Titers From Birth Through 6, 10, 16, or 20 Weeks Postpartum

Censoring Time	6 wk	10 wk	16 wk	20 wk
Achievement of 4-fold rise in Ab titer	23 (21–26)	27 (25–29)	35 (33–37)	38 (36–40)
Achievement of any rise in Ab titer	22 (21–25)	25 (23–26)	30 (28–31)	31 (30–32)
Not censored	23 (21–26)	27 (25–29)	36 (33–38)	40 (37–42)

Data are no. of days (95% confidence interval).

The effect of covariates on cord blood Ab titers and Ab half-life from birth to 20 weeks are shown in Table 4. Longer maternal education duration was associated with a higher cord blood Ab titer (P = .004) but with a shorter Ab half-life (P < .001). We performed further analysis to evaluate whether maternal age was confounding the relationship between maternal characteristics and Ab titers. Among those having their first birth, only 15% were >25 years old, compared with 55% of those with previous deliveries (P = .002). Maternal education duration was significantly associated with maternal age (R = 0.28; P = .001). In multivariate analysis including maternal age, maternal education duration, and parity, cord blood Ab levels increased significantly with increasing maternal education duration (0.82 log₂ increase; P = .025) but not with multiparity (0.65 log₂ increase; P = .091) or older maternal age (0.19 log₂ increase; P = .56).

Table 4.

Effect of Covariates on Mean Respiratory Syncytial Virus (RSV) Antibody (Ab) Titers and RSV Ab Half-life

Covariate	Log2 Ab Titer at Birth, Mean			Ab Half-life, d (95% CI)
	Yes	No	P Valuea	Yes	No	P Valueb
Vaginal birth	10.42	10.88	.066	39 (36–42)	36 (34–40)	.24
Male infant	10.64	10.58	.78	37 (35–40)	38 (35–42)	.65
Any previous children	10.79	10.15	.078	35 (33–37)	51 (43–63)	<.001
Infant birthweight >3 kgc	10.61	10.61	.99	37 (34–41)	38 (36–41)	.72
Small for gestational age	10.44	10.70	.32	38 (35–42)	38 (36–41)	.97
Maternal age >25 yc	11.03	10.39	.012	36 (33–39)	39 (37–42)	.094
Maternal education duration >12 yc	11.12	10.37	.004	33 (31–36)	41 (38–44)	<.001
Infant cord log2 Ab titer >11.1c	11.81	9.47	<.001	32 (30–34)	46 (42–50)	<.001

^a For comparison of the difference in Ab level at birth, by the group indicator.

^b For comparison of the difference in half-life or change in Ab over time (slope), by the group indicator.

^c The median value.

Ab titers from all 149 infants were used to evaluate the time to any rise, 2-fold rise, and 4-fold rise in RSV Ab titer as a marker of serologic infection. Overall, 141 infants (95%) experienced any rise in Ab titer from birth to 72 weeks, 116 (78%) experienced a 2-fold rise, and 92 (62%) experienced a 4-fold rise. Although 98 of 141 infants (70%) had any rise in titer before 72 weeks, only 40 (34%) of 116 infants had a 2-fold rise and 17 (18%) of 92 infants had a 4-fold rise. The effect of covariates on serologic infection is shown in the second table (Table 6) of the Supplementary Materials. Higher baseline log₂ cord blood Ab titers (log₂ titers >11.1) were significantly associated with a decreased risk of a 4-fold rise in titer (hazard ratio [HR], 0.6; 95% CI, .4–.9; P = .011). No effect was observed with higher maternal age or parity, although an increased risk for infection was seen with births in the fall (HR,1.6; 95% CI, 1.0–2.7; P = .039).

On the basis of previous studies evaluating protective titers in infants, we evaluated the time to reduction in Ab level to a titer of less than 1:256 (log₂ Ab titer 8.0). We included only the 147 (99%) infants with titers above 1:256 at birth. Infants were excluded from further analysis when their Ab titer quadrupled, the level at which it was assumed that they potentially acquired new infection. The median time to reduction of the titer below a potentially protective level was 17 weeks (95% CI, 14–20 weeks; Figure 3). The titer in only 13 infants (9%) remained above a potentially protective level by the week 72 visit. Assuming an Ab half-life of 38 days, we estimate that for every 0.5 log₂ increase in cord Ab titer, the duration of protection would increase by 19 days.

Figure 3.

Kaplan–Meier survival curve showing proportion of infants with log₂ antibody titers greater than 1:256 (ie, >8.0) at each visit up until 72 weeks. Infants were censored when they developed a 4-fold rise in titer. The median time to drop below a threshold Ab of 1:256 was 17 weeks (95% confidence interval, 14–20 weeks).

The time to reduction of the titer to below a potentially protective level was not influenced by covariates, including maternal parity (P = .24), maternal age (P = .88), maternal education duration (P = .34), type of birth (P = .53), infant sex (P = .60), prematurity (P = .35), SGA (P = .67), or birth weight (P = .32) (data not shown). Log₂ cord blood Ab titers above median 11.1 were associated with a decreased risk of reduction to below a potentially protective titer (HR, 0.5; 95% CI, .3–.7; P < .001), and being born in the fall was associated with a faster reduction in Ab level relative to that during the winter season (HR, 1.7; 95% CI, 1.1–2.4; P = .011), although spring and summer were not found to have different rates (data not shown). Results were similar when a 2-fold rise instead of a 4-fold rise in titer was used as the definition for serologic infection (data not shown).

DISCUSSION

To our knowledge, this is the first study to examine RSV-neutralizing Ab titers over time in mother-infant pairs followed from the third trimester of pregnancy until 72 weeks postpartum. We found that maternal Ab titers remained stable from third trimester to birth and that transplacental transfer of Ab from mother to infant was efficient. Higher cord blood Ab titers were associated with a decreased risk for serologic infection or with reduction of the titer to a nonprotective level. The half-life of Ab in infants was approximately 1 month. These findings suggest that maternal RSV vaccination during the third trimester may be an effective method to provide RSV-neutralizing Ab to infants in the first months of life.

RSV is generally assumed to cause a self-limited upper respiratory tract infection in healthy adults. In response to natural infection, as opposed to vaccination during clinical trials, both young adults and elderly individuals have been shown to mount a 4-fold increase in Ab levels [31]. It is not known whether pregnancy modifies this serologic response to natural infection. In a RSV purified fusion protein 2 vaccine study, RSV-neutralizing Ab titers in pregnant women after immunization increased by only 1.4-fold (ie, a log₂ Ab titer of 0.5) [20]. We are not aware of prior studies that evaluated the stability of RSV Ab titers during pregnancy and the postpartum period. In our study, we found that pregnant women had protective RSV Ab titers in the third trimester and that these titers remained stable by birth and 72 weeks postpartum.

Transplacental Ab transfer is influenced by a variety of factors, including coinfection with malaria or HIV, prematurity, and SGA [10]. The incidence of SGA in our study was 34%, reflective of the South Asian population. However, we found no effect of SGA on RSV Ab transfer ratio, cord blood Ab titers, or Ab half-life. Malaria and HIV infection are not endemic in our population, and rates of prematurity were too low to be statistically meaningful. In multivariate analysis, we found that maternal education duration was associated with higher cord blood Ab titers. We speculate that longer maternal education duration may be an indicator of higher socioeconomic status or better overall health, which may influence Ab titers.

Ab half-life varies across populations. In developed countries, the half-life of maternal RSV IgG is approximately 1 month [20]. In a study of Kenyan infants, RSV Ab half-life was 79 days, although this study used an enzyme-linked immunosorbent assay (ELISA) as opposed to a RSV microneutralization assay, making direct comparisons difficult [21]. A follow-up study in the same population found that RSV infection in infants elicited a strong neutralizing Ab response but that this response declined to preinfection levels 3–4 months later [32]. A similar result was found in Sao Paolo, Brazil, where maternal Ab was detectable for 3.3 months [18]. We found that that the RSV-neutralizing Ab half-life was 38 days and that a decrease to below a protective titer occurred by a median of 17 weeks. This corresponds to the clinical observation that infants are most vulnerable to developing severe RSV disease between 6 weeks and 6 months of age, a period when maternal Ab is waning and the infant immune response may be incapable of generating a robust neutralizing Ab response to infection [33–35]. Our calculation of half-life was performed by using a stringent definition of serologic infection and by analyzing data from birth to 20 weeks only. When calculation of half-life was limited to values between birth and 10 weeks, we found that the half-life decreased from 38 days to 27 days. This suggests that either the rate of decline in the maternal Ab level decreases after the first 10 weeks or that intercurrent RSV infection occurred between 10 and 20 weeks. To attempt to exclude intercurrent RSV infections as an etiology, we additionally calculated half-life by using a definition in which no decline in Ab titer between 2 time points was considered indicative of serologic infection. We found in this case that the Ab half-life decreased from 38 days to 31 days, implying that the differential rate of the decrease in Ab titer may be due to intercurrent RSV infection.

In our study, the only factor associated with both protection from serologic infection and maintenance of a protective Ab titer was a higher baseline cord blood Ab level. Prior studies using ELISA have shown a blunting of antibody response to the fusion protein by the presence of maternal antibody in younger infants [28]. Epidemiologic studies in the United States showed higher cord blood RSV Ab levels are associated with protection from RSV infection in young infants [4]. This finding emphasizes the fact that if maternal vaccination has the ability to increase cord blood Ab titer, vaccination may be an effective means of decreasing severe RSV infection in infants.

We chose to study the time it took for the Ab level to fall below a titer of 1:256, which we selected for the purpose of analyses, although there is no clearly defined protective titer. Studies performed in cotton-rat models found that a titer of 1:380 was protective against lung infection [36], and titers of 1:256 have been used as a protective cutoff in other studies in infants [8, 9]. There is also variability in titers across neutralization assays used by various institutions. We found that our assay provides values within the appropriate range, using a panel of reference sera from the National Institutes of Health (data are shown in the final table of the (Table 7) Supplementary Materials).

Limitations of our study include the lack of virologic diagnosis for RSV illness in mothers and infants. We used serologic responses as a proxy for RSV infection in this population, although a 4-fold increase in Ab titer may not be a sensitive indicator of RSV infection in young infants, who are unable to generate a robust humoral response to infection. Moreover, we found that only 11 infants achieved a 4-fold rise in Ab titer by 20 weeks, while 75 achieved any rise by the same time point. This suggests that any rise in titer is perhaps a more appropriate definition in the infant population and should be considered in vaccine trials moving forward. Our results are consistent with previous studies, including the original study by Glezen et al, and the more recent study from Kenya, both of which found that higher cord blood Ab levels were protective from RSV infection [4, 21]. Other limitations of our study include the use of sera only from mother-infant pairs who completed a full 72 weeks of follow-up, indicating a potential source of selection bias. The baseline characteristics of included and excluded mother-infant pairs were largely similar, although mothers included in the study were more likely to be having their first child and to deliver at full term. An additional limitation is the lack of sample collection across a full calendar year. All births in our study occurred during a period of low RSV infection incidence in Dhaka, as described in a separate surveillance study of acute respiratory illnesses in children conducted during the same period [37]. RSV infection occurs in highly seasonal epidemics, and variation in maternal Ab levels have been reported, with peaks in Ab titers corresponding to these epidemic periods [38]. Finally, we do not have data on the kinetics of RSV Ab titers between 24 and 72 weeks; it is possible that infant Ab titers decline further or increase transiently after infection during this period.

We conclude that RSV-neutralizing Ab titers are stable in pregnant and postpartum women and transfer efficiently across the placenta in mother-infant pairs in Bangladesh. Maternal RSV immunization in the third trimester has the potential to provide protective levels of passive neutralizing Ab levels to infants in the first few weeks of life and should be given priority as a strategy to prevent severe RSV disease in very young infants.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.