1 INTRODUCTION

Since the identification of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in December 2019, much has been learned about the effects of its associated disease, COVID-19, on pregnancy and perinatal outcomes. The results are sobering with pregnant individuals having higher rates of intensive care unit admission, mechanical ventilation, extracorporeal membrane oxygenation, and death compared with their non-pregnant counterparts with COVID-19.1-4 In addition, severe COVID-19 in pregnancy has been associated with elevated rates of preterm delivery and neonatal morbidities.1-5 The higher severity of illness in pregnant individuals underscores the need to maximize vaccine uptake in this population. Recommendations to prioritize pregnant and lactating individuals for vaccination are further supported by observational data on vaccine safety6-11 and efficacy.12, 13

Maternal vaccination has the potential to transfer immunity to the newborn and infant.14 At the time of writing, severe illness due to COVID-19 in infants is less frequent than in adults.15-17 However, the emergence of more transmissible and virulent variants coupled with widespread vaccination of adults and older children has created uncertainty in the future trajectory of the disease in young children and infants. COVID-19 vaccines have been evaluated in a stepwise manner by age, establishing safety and efficacy first in adults, then adolescents, and now in pediatric age groups. However, there are no clinical trials actively recruiting infants under age 6 months of age and the immunogenicity of new vaccine platforms in infants is unknown.

The paradigm of maternal immunization to enhance protection of infants is well established for infections such as pertussis and influenza.18-20 SARS-CoV-2-binding and neutralizing antibodies are present in infant cord blood and breast milk following natural maternal infection21-23 and transfer of maternal immunity following COVID-19 vaccination is an active research area. In this review, we synthesize the available research, highlight knowledge gaps, and outline factors that should be considered and reported when studying the transfer of maternal immunity following COVID-19 vaccination. To set the evidence in context, we briefly review background data on the transfer of maternal immunity via the placenta and breast milk and immunologic correlates of protection induced by vaccination.

2 SOURCES

We conducted a literature search of PubMed for articles published between January 1, 2019, and September 12, 2021, using combinations of the following search terms: COVID-19, SARS-CoV-2, vaccine, BNT162b2, mRNA-1273, ChAdOx1-S, breast milk, human milk, lactation, pregnancy, newborn, infant, immunoglobulin, antibody, umbilical cord blood, and immunogenicity. We updated the search on October 2, 2021. We included prospective or retrospective cohort studies, cross-sectional studies, and case series published in English or French. We conducted an additional search of the John Hopkins University COVID-19 Maternal and Child Health, and Nutrition Scientific Repository.24 We examined the reference lists of relevant articles to identify sources not captured in the electronic literature search.

3 TRANSFER OF MATERNAL IMMUNITY VIA THE PLACENTA

Neonates and young infants do not have a fully developed immune system and are unable to mount an efficient humoral immune response to infectious agents.25 Vulnerability to infection during the first months of life is mitigated by transfer of maternal immunity via the placenta and through breast milk.25 Transplacental passage of antibodies, principally immunoglobulin G (IgG), to the fetal circulation begins as early as 13 weeks gestation and increases exponentially as pregnancy progresses, with the largest transfer accomplished in the third trimester.26 It has been demonstrated that fetal IgG rises from approximately 10% of the maternal concentration at 17- to 22-weeks gestation, to 50% at 28 to 32 weeks, and reaches concentrations of 20–30% higher than maternal concentrations at 29- to 41-weeks gestation.26 Accordingly, there is reduced transplacental IgG passage to preterm infants.26 Chronic maternal infections that can injure the placenta, such as malaria and HIV, may impair IgG transplacental passage to the fetus.26 In general, maternal IgG in the infant declines over the first 6 months of life, although the kinetics vary by antigen specificity.27

With regard to the transplacental passage of vaccine-induced antibodies, the optimal gestational window for vaccination is an important consideration. Achieving maximum protection for the at-risk pregnant individual without compromising protection for the infant during the first months of life may be challenging. For example, a recent systematic review and meta-analysis found that influenza vaccination later in pregnancy, at least 15 weeks before delivery, was associated with higher cord and newborn blood antibody levels than earlier maternal vaccination.28 On the contrary, given the risk that influenza imposes on pregnant individuals and the fetus, vaccinating earlier in pregnancy provides a longer duration of protection to the mother and may be beneficial if preterm delivery occurs.

4 TRANSFER OF MATERNAL IMMUNITY THROUGH BREAST MILK

After birth, breast milk plays a critical role in transferring maternal immunity to the infant.25 The composition of breast milk is dynamic and varies within and between mothers, as well as over the duration of lactation.29 Colostrum is synthesized and secreted during the first 48 hours postpartum; it is rich in proteins and immune factors such as antibodies, cytokines, and leukocytes. This is followed by transitional milk, which shifts in composition to a primary source of nutrition and is secreted until approximately 2 to 3 weeks postpartum. Finally, mature milk appears several weeks postpartum with further increases in carbohydrates and lipids and decreases in proteins, including antibodies.29

Mucosal barriers close at birth preventing passage of breast milk antibodies into the neonatal circulation.25 Thus, in contrast to the systemic immunity conferred by IgG transferred via the placenta, breast milk antibodies provide mucosal immunity, acting as a first line of defense by binding to pathogens and toxins, preventing them from entering cells at mucosal surfaces.30 Secretory IgA (sIgA) is the principle immunoglobulin in breast milk comprising 80 to 90% of breast milk immunoglobin.30 The two IgA subclasses in humans are IgA1 and IgA2; IgA1 is present in the respiratory tract, saliva, serum, and skin, whereas IgA2 is present in the intestine.31 Secretory IgA is relatively resistant to enzymatic degradation in the infant gut. In addition to its role in mucosal immunity, it can prevent the pro-inflammatory response elicited by the activation of other immune defense mechanisms.30 Multiple studies have found increased concentrations of pathogen-specific sIgA and neutralizing antibodies in breast milk following maternal immunization (reviewed in32). Data also suggest exclusive breastfeeding following maternal influenza vaccination is associated with decreased respiratory tract infections during the first 6 months of life.20, 33

In addition to sIgA, secretory IgM (sIgM) and IgG are transferred via breast milk, albeit to a lesser extent.25 Breast milk IgG comprises approximately 10% of total immunoglobulin initially postpartum but increases over time. In certain infections, such as HIV and cytomegalovirus, IgG may have a greater neutralizing capacity than sIgA.34, 35 sIgM plays a role in opsonization of Gram-negative bacteria and, similar to sIgA, acts to prevent colonization and cell entry by pathogenic microorganisms.36 Other protective factors in breast milk include lactoferrin, oligosaccharides, interleukin-10, and epidermal growth factor.32

5 IMMUNOLOGIC CORRELATES OF PROTECTION INDUCED BY VACCINATION

Before discussing data on transfer of maternal antibodies via the placenta and breast milk, it is important to emphasize that there is currently no validated correlate of protection against SARS-CoV-2 infection or COVID-19 illness.37 Vaccines, like natural infection, protect via multiple mechanisms.38 Single components, such as antibody response, are easier to measure than cellular immune responses and have, therefore, been the focus of most research in this area.38 Several published studies suggest a relationship between SARS-CoV-2-specific neutralizing antibody concentrations and vaccine efficacy.39, 40 Animal studies and natural infection cohorts add further support to antibodies as a correlate of protection against SARS-CoV-2 infection.41, 42 However, “protective thresholds” have not been established and likely vary by end point (i.e., protection from infection vs. protection from severe disease or death).37 In addition, correlates of protection may be specific to a particular vaccine platform or a specific vaccine.37 Thresholds could differ by viral variant, infectious inoculum, patient population, timing of sample collection, specimen type, and neutralization assay.37, 38 Finally, waning of antibody concentrations over time is frequently observed after vaccination; vaccine-induced memory may be important for protection at later time points but is difficult to measure.43

6 TRANSPLACENTAL SARS-COV-2-SPECIFIC ANTIBODY TRANSFER FOLLOWING MATERNAL COVID-19 VACCINATION

We retrieved six prospective case series/cohort studies that evaluated transplacental passage of SARS-CoV-2-specific antibodies following maternal vaccination in 212 mother-infant dyads.44-49

One of the first studies to investigate the immunogenicity of COVID-19 vaccines in pregnancy prospectively followed 30 individuals who received two doses of an mRNA vaccine (BNT162b2 or mRNA-1273), 3 to 4 weeks apart at variable time points during gestation as well as 22 individuals with SARS-CoV-2 infection during pregnancy.45 Among vaccine recipients, nine individuals delivered during follow-up and maternal and infant cord blood were collected for analyses. Receptor-binding domain (RBD) IgG concentrations were approximately 25% higher in infant cord blood than maternal serum at the time of delivery. Moreover, neutralizing antibody concentrations in cord blood in vaccinated individuals suggested efficient transplacental passage. Both cord blood and maternal serum antibody concentrations were higher in vaccinated individuals compared with individuals with natural SARS-CoV-2 infection; however, this may be partly explained by differences in timing between dose 2 and delivery compared with COVID-19 symptom onset and delivery (median 21 vs. 41 days, respectively).

In a second prospective case series of pregnant health-care workers, transplacental passage of COVID-19 vaccine-elicited antibodies was evaluated in 16 paired maternal and infant cord blood samples.47 All participants received two doses of BNT162b2, 3 weeks apart, with the first dose given between gestation week 29 and 36. IgG directed against the S1 subunit of the spike protein (anti-S1) was detected in 100% of maternal and infant cord blood samples. The mean cord-to-maternal anti-S1 IgG ratio was 1.28 +/- 0.798. The number of weeks between the first or second dose and delivery was positively correlated with antibody concentrations in cord blood (i.e., earlier vaccination was associate with higher antibody concentrations). A similar pattern was observed with the mean cord-to-maternal anti-S1 IgG ratio; higher ratios correlated with earlier vaccination.

Transplacental passage of COVID-19 vaccine-elicited IgG and IgM was assessed in a third study of 27 parturient health-care workers who delivered 28 infants (one set of twins) at a single medical center.49 Seventy-four percent of participants received two vaccine doses (64% BNT162b2) before delivery with a mean (standard deviation, SD) latency of 6 (3) weeks. IgG and IgM directed against the SARS-CoV-2 RBD were detected in 26/27 and 15/27 maternal plasma samples at the time of delivery, respectively. IgG was detected in 25/28 infant cord blood specimens. The negative samples were from infants of participants who received their first vaccine dose less than 3 weeks before delivery. IgM was not detected in any cord blood samples. The mean (SD) maternal-to-cord blood IgG transfer ratio was 1 (0.6); two vaccine doses and increased latency from vaccination to delivery were associated with an increased IgG transfer ratio.

A fourth study compared cord blood anti-spike IgG concentrations in parturient individuals (n = 64) who had been vaccinated with two doses of BNT162b2 to 11 parturient individuals who had SARS-CoV-2 infection during pregnancy.48 The median placental transfer ratio of anti-spike IgG was 0.77. Significantly higher anti-spike IgG concentrations were found in cord blood from vaccinated versus recovered individuals (p < 0.0001). Like the previously described study,45 this difference may be partly explained by time elapsed since second vaccination and delivery (mean (SD) 21.7 (11.0) days) and positive SARS-CoV-2 PCR test and delivery (mean (SD) 92.5 (75.8) days).

In one of the largest studies to date on this topic, SARS-CoV-2-binding IgG and IgM concentrations in paired maternal and infant cord blood samples were evaluated in 86 recipients of BNT162b2 and 65 individuals with SARS-CoV-2 infection during pregnancy.44 The first vaccine dose was administered at mean (SD) 34.5 (7.5) weeks gestation followed by a second dose 3 weeks later. Similar cord blood anti-S1 and RBD IgG concentrations were observed in vaccine recipients and those with natural infection. By contrast, cord blood IgG directed against the spike protein S2, and N subunits were higher in those with natural infection. IgG to S1, S2, and RBD but not N was detected in cord blood as early as 2 weeks after a single COVID-19 vaccine dose. There was no evidence of IgM transplacental passage.

Finally, infant cord blood antibody concentrations were evaluated in 10 vaccinated individuals at delivery as part of a larger study evaluating several components of mRNA COVID-19 vaccine (BNT162b2 or mRNA-1273) immunogenicity in pregnant and lactating individuals.46 All were vaccinated in the third trimester, and 9/10 individuals received two vaccine doses before delivery. Anti-S1 and RBD IgG were detected in all infant cord blood samples; lower concentrations were observed in the individual who received only one vaccine dose. Neutralizing antibody concentrations were detected in 8/10 cord blood samples; the undetectable samples were from the individual who received only one vaccine dose and an individual who received the second dose only 7 days before delivery. Consistent with this latter observation, in the overall cohort, there was a significant improvement in passage of anti-S1 IgG, but not RBD IgG, with longer times between vaccine dose 2 and delivery.

In summary, SARS-CoV-2-specific binding and neutralizing antibodies have been detected in infant cord blood of recipients of mRNA COVID-19 vaccines. Antibody concentrations appear to be at least as high in infant cord blood as in the maternal circulation, and at least two vaccine doses during pregnancy are likely needed to reach maximal concentrations. The optimal timing of vaccination and the magnitude of antibody transplacental passage following infection compared with vaccination remain unclear. Data on transplacental passage of antibodies for maternal vaccines administered in the pre-conception period and the first trimester are not currently available since most of these individuals are still currently pregnant, but will be of high importance in optimizing future immunizations.

7 BREAST MILK SARS-COV-2-SPECIFIC ANTIBODIES FOLLOWING COVID-19 VACCINATION

Our search identified 17 studies evaluating breast milk SARS-CoV-2 binding and neutralizing antibodies following maternal COVID-19 vaccination (Table 1).45, 46, 48, 50-63

TABLE 1. Studies of SARS-CoV-2-specific antibodies in breast milk following vaccination First author (publication year) Country Number of lactating participants (number of breast milk samples) Administered vaccine Time of vaccine administration Methodology and parameter analyzed Time of sampling Main antibody-related findings Adverse effects in infants Baird JK (2021)50 USA 7

BNT162b2 (n = 4)

mRNA−1273 (n = 3)

2 doses, 3–4 weeks apart

Not reported Anti-S1 RBD IgG and IgA detected by ELISA

Breast milk collected before vaccination; 1-, 4-, 7-, 11-, 14-, and 20-days post dose 1; and 1-, 4-, 7-, 11-, and 14-day post dose 2

N = 1 sample 80 days post dose 1

Anti-S1 RBD IgA and IgG increased by day 7 post dose 1, declined before dose 2, and sharply increased after dose 2. Elevated vs. baseline at day 11 post dose 2

IgA and IgG remain elevated at day 80 in the 1 participant who provided a sample

IgG levels ~2 x higher than IgA (reverse pattern of most studies)

Milk pasteurization resulted in <7% decline in antibody levels

Not reported Calil V (2021)51 Brazil 20 (170) CoronaVac (inactivated vaccine): 2 doses, 4 weeks apart Nursing period of mean (SD) 11.2 (8.7) months Breast milk anti-S1 IgA detected by ELISA

Breast milk collected before dose 1 and weekly for 7 weeks

4 months after dose 1 (n = 10)

Breast milk IgA rapidly increased after dose 1, declined before dose 2, and peaked 1 and 2 weeks after dose 2

5/10 individuals had IgA above seroconversion threshold at 4 months after dose 1

No significant adverse effects reported Charepe N (2021)63 Portugal 14 BNT162nB Breastfeeding duration 12–24 months for 57% of participants and 2–12 months for 43% Blood and breast milk anti-spike IgG, IgA, and IgM detected by ELISA Blood and milk collected 1–3 weeks after dose 1 and 2

100% of blood samples positive for anti-spike IgG, IgA, and IgM. IgG was the dominant Ig

Anti-spike IgA detected in 36% and 21% of breast milk samples after dose 1 and 2, respectively (opposite pattern of most studies)

Anti-spike IgG detected in 7% and 43% of breast milk sample after dose 1 and 2, respectively

No IgM was detected in breast milk after dose 1 or 2

Higher breast milk IgG in milk correlated with longer breastfeeding time

1 infant showed increased somnolence Collier AY (2021)45 USA 16

BNT162b2 (n = 11)

mRNA−1273 (n = 5)

2 doses, 3–4 weeks apart

Not reported Serum and breast milk anti-RBD IgA and IgG detected by ELISA; neutralization determined with a pseudo virus neutralizing antibody assay Breast milk collected median (IQR) 26 (18–31) days after dose 2

Median breast milk IgG concentration ~2 x higher following infection vs. vaccination

Median breast milk IgA concentration ~75 x higher after infection vs. vaccination

SARS-CoV−2 neutralization concentration lower for breast milk following vaccination than after infection

IgG and IgA following vaccination higher in serum than in breast milk. Serum IgG and IgA concentrations higher following vaccination than following infection (opposite trend of breast milk)

Not reported Esteve-Palau E (2021)52 Spain 33 (93)

BNT162b2:

2 doses, 3 weeks apart

Mean (SD) 17.5 (10.1) months postpartum Not reported Blood and breast milk collected 2 weeks after dose 1 (T1), 2 weeks after dose 2 (T2), and 4 weeks after dose 2 (T3)

Median IgG serum: milk pairs at each time point:

T1 519:1 AU/mL

T2 239:1 AU/mL

T3 248:1 AU/mL

Correlation between serum and breast milk IgG r = 0.7

IgA not assessed

Not reported Gray KJ (2021)46 USA 31 (57)

BNT162b2 (n = 16)

mRNA−1273 (n = 15)

2 doses, 3 weeks apart

Median (IQR) 7.3 (3.8–10.8) months after delivery Anti-RBD and anti-spike detected by ELISA Blood and breast milk collected before vaccination (T1), at the time of dose 2 (T2) and 2–6 weeks after dose 2 (T3)

IgA, IgG, and IgM levels significantly increased at T2 and T3 vs. T1

IgA and IgM did not significantly increase at T3 vs. T2. Further increases in IgG from T2 to T3

Reactogenicity symptom score in maternal participants positively correlated with breast milk IgG and IgA concentrations

Not reported Guida M (2021)53 Italy 10 (15)

BNT162b2

2 doses, 3 weeks apart

Mean (SD) 11 (5.1) months breastfeeding Anti-S1 RBD antibodies Elecsys immunoassay Blood and breast milk collected 1 day before dose 2 (T1) and 7 days after dose 2 (T2)

Anti-S1antibodies detected in all serum and breast milk samples at T2

Serum concentrations ~100 x higher than breast milk

Not reported Jakuszko K (2021)54 Poland 32 (64) (+28 non-vaccinated participants as controls)

BNT162b2

2 doses, 3 weeks apart

Infant mean (SD) age 8.8 (6.8) months SARS-CoV−2-specific IgA (anti-S1), IgM (anti-N), and IgG (anti-S1) detected by ELISA Blood and breast milk collected 8 days after dose 1 and 7 and 21 days after dose 2

Breast milk IgG and IgA correlated to serum IgG and IgA

IgM was not detected any breast milk samples

Highest breast milk IgG and IgA concentrations on day 7 after dose 2 with a decrease on day 21

As reported by mothers: 1 infant with behavior change and increased tearfulness after the first dose, 1 infant with sleeplessness after the second dose Juncker HG (2021)55 Netherlands 26 (366)

BNT162b2

1 dose (n = 6), 2 doses (n = 20)

Breastfeeding for median (IQR) 7 (5–9) months Breast milk anti-S1 IgA and serum anti-S1 IgG detected by ELISA

Blood collected before vaccination, 15–17 days post dose 1, pre-dose 2 and 15–17 days after dose 2

Breast milk collected before vaccination and 3, 5, 7, 9, 11, 13, and 15 days after both doses

Breast milk anti-S1 IgA showed biphasic pattern: concentration increased ~day 5–7 post dose 1, peaked ~day 12–15 post dose 1, declined and increased sharply after dose 2

Serum anti-S1 IgG detected ~day 7 after dose 1 and increased thereafter

Not reported Kelly JC (2021)56 USA 5 (29)

BNT162b2

2 doses, 3 weeks apart

Infant age range at dose 1: 1 to 24 months SARS-CoV−2-specific anti-spike IgA and IgG detected by ELISA Breast milk collected before vaccination, then on day 1 and 7 following dose 1 and 2

SARS-CoV−2-specific IgA significantly elevated vs. baseline starting 2 weeks after dose 1, through day 20–29 with a gradual decline after dose 2

SARS-CoV−2-specific IgG significantly elevated vs. baseline starting

around day 20 after dose 1 and sustained to day 40

Not reported Lechosa-Muniz C (2021)57 Spain 110 (+ 23 non-vaccinated participants as the control)

BNT162b2, 2 doses (n = 70)

mRNA−1273, 2 doses (n = 20)

ChAdOx1-S, 1 dose (n = 20)

Mean (SD) infant age at dose 1: 15.9 (11.2) months

Serum and breast milk SARS-CoV−2-specific IgA and IgG detected by ELISA

Serum and breast milk anti-S1 (RBD) IgG detected by CLIA

Blood and breast milk collected 30 days after dose 2 (or dose 1 if ChAdOx1-S received)

Mean serum and milk IgA and IgG concentrations higher in participants who received BNT162b2 or mRNA−1273 vs. ChAdOx1-S (1 dose)

IgA and IgG correlated in recipients of BNT162b2 or mRNA−1273

Anti-S1 IgG detected in 100% of vaccine recipients serum

Serum anti-S1 IgG higher in recipients of BNT162b2 or mRNA−1273 vs. ChAdOx1-S (1 dose)

Serum anti-S1 IgG ~100 higher than breast milk

As reported by mother: 1 breastfed infant irritable x 24 hours Low JM (2021)58 Singapore 14 (66) (+ 6 COVID−19 recovered and 9 healthy participants as controls)

BNT162b2

2 doses, 3 weeks apart

Mean (SD) 9.0 (3.8) months postpartum Anti-S1 IgA and IgG detected by ELISA

Breast milk collected before vaccination (T1), 1–2 days after dose 1 (T2), 7–10 days after dose 1 (T3), 3–7 days after dose 2 (T4), and 4–6 weeks after dose 2 (T5)

Breast milk anti-S1 and RBD IgA significantly higher than controls at T4 with decline at T5

Breast milk anti-S1 and RBD IgG significantly higher than controls at T4, maintained at T5

No adverse effects Nir O (2021)48 Israel 30

BNT162b

2 doses

Dose 2 mean (SD) 21.7 (11.0) days before delivery Anti-RBD IgG detected by ELISA In hospital following delivery

Anti-RBD IgG detected in 100% of serum, 98% of cord blood, 96.4% of neonatal dried blood spot, and 100% of breast milk samples

Ratio of median serum-to milk IgG 5.3:1

Not reported Perl SH (2021)59 Israel 84 (504)

BNT162b2

2 doses, 3 weeks apart

Mean infant age at time of vaccine dose 1: 10.32 (7.31) months

SARS-CoV−2-specific IgA detected by EUROIMMUN AG Anti-SARS-CoV−2 S Kit

SARS-CoV−2-specific IgG detected by Elecsys Anti–SARS-CoV−2 S serology assay and read on the Cobas e801 analyzer

Breast milk collected before vaccination then weekly x 6 weeks starting 2 weeks after dose 1

62% of milk samples contained SARS-CoV−2-specific IgA 2 weeks after dose 1

86% of milk samples contained SARS-CoV−2-specific IgA 1 week after dose 2 (week 4)

IgA remained elevated at week 6

SARS-CoV−2-specific IgG increased at week 4 after dose 2 when IgG was detected in 92% of samples. IgG detected in 97% of samples at week 5 and 6

Fever and symptoms of upper respiratory tract infections in 4 infants; ¼ admitted to hospital for evaluation and antibiotic treatment Ramireza DSR (2021)60 Spain 98 (93) (+24 non-vaccinated group as controls)

BNT162b2 (n = 92)

mRNA−1273 (n = 6)

2 doses, 3–4 weeks apart

Mean breastfeeding duration at study entry 6.5 months

SARS-CoV−2-specific IgA detected by ELISA

Anti-S1 (RBD) IgG detected with CLIA

Blood and breast milk collected on day 14 after dose 2

Serum anti-S1 (RBD) IgG neutralizing concentrations (>560.9 BAU/mL) detected in 100% of vaccinated participants

Serum anti-S1 IgM detected in 22.5% of samples

Breast milk anti-S1 IgA detected in 89%, IgM in 0, and IgG in 100% of participants

Serum IgG ~200 x higher vs. breast milk

IgA concentrations in breast milk also lower than serum; value not reported

IgG and IgA breast milk concentrations significantly higher in participants breastfeeding ≥24 months vs. <24 months

Not reported Schwartz A (2021)61 Israel 61

BNT162b2

2 doses, 3 weeks apart

Not reported

Blood IgG and breast milk IgA and IgG detected by ELISA

Infant dried blood spot IgA and IgG detected by ELISA; neutralization determined with a pseudo virus neutralizing antibody assay

Blood and breast milk collection times not reported

Infant dried blood spot samples collected; timing not reported

Infant saliva collected immediately after breastfeeding and at 30, 90 and 150 minutes after breastfeeding

IgG detected in 100% of maternal serum and breast milk samples

Serum and breast milk IgG positively correlated (r = 0.514)

sIgA detected in 15% of breast milk samples

18/47 milk samples neutralized SARS-CoV−2 infectivity

IgG detected in saliva of 3/5 infants

IgG not detected in any infant dried blood samples

Not reported Valcarce V (2021)62 USA 21 (63)

BNT162b2 (n = 14)

mRNA−1273, (n = 7)

2 doses

Mean (SD) 6.8 (4.8) months postpartum Serum and breast milk SARS-CoV−2-specific IgA and IgG detected by ELISA Blood and breast milk collected before vaccination (T1), 16–30 days post dose 1 (T2), and 7–10 days post dose 2 (T3)

Breast milk SARS-CoV−2-specific IgA detected in 85% of participants; IgA increased significantly over follow-up

Breast milk SARS-CoV−2-specific IgG detected in 100% of participants by 7–10 days post dose 2; IgG increased significantly over follow-up

Breast milk IgA higher than IgG at all time points

Plasma SARS-CoV−2-specific IgA ~2 x higher at T1-3 vs. milk

Plasma SARS-CoV−2-specific IgG ~2–50 x higher at T1-3 vs. milk

Not reported Abbreviations: CLIA, chemiluminescent microparticle immunoassay; ELISA, enzyme-linked immunosorbent assay; Ig, immunoglobulin; IQR, interquartile range; n, number; RBD, receptor binding domain; SD, standard deviation; sIg, secretory immunoglobulin; T, time.

Similar to the temporal pattern observed in serum following natural infection, in the reviewed studies breast milk IgA concentrations tended to increase rapidly, with significant elevations compared with baseline as early 1–2 weeks following a single vaccine dose and remain elevated for many weeks following the second vaccine dose.46, 56, 59, 62 Breast milk IgG was generally detected at later time points, often requiring two doses to be significantly increased over baseline, but IgG appeared to persist in breast milk longer than IgA.