Maternal SARS-CoV-2 IgG provides limited protection for their infants

At present, there are still many ambiguous reports about the perinatal infection of infants born to mothers infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) 1-3 . The dynamic characteristics of infantile serum antibodies born to mother with SARS-CoV-2 has not been well described 4-6 . In this study, we analyzed the seroconversion of 27 newborns born to 26 pregnant women infected with SARS-CoV-2. The SARS-CoV-2 IgG positive rate of parturients was 80.8%, and half of their infants obtained maternal IgG. IgG transfer rates were 18.8% and 81.8% in those infants whose mother infected less and more than 2 weeks before delivery respectively. In the rst two months of life, the IgG level of infants dropped sharply to one tenth of that at birth. The IgM was conrmed positive in 53.8% of mothers and negative in all infants. These results suggest that maternal IgG has limited protection for infants, which may help us to improve vaccination strategies in future.


Introduction
Pregnant women seem to be more vulnerable to coronavirus infected disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection due to the characteristic immune responses during pregnancy. It has been found that more than 13% admitted pregnant women were symptomatically with SARS-CoV-2 infection 9 . Vertical transmission of SARS-CoV- 2 has not yet been detected, whereas perinatal transmission has been suspected in a few cases [10][11][12] . Due to the limitations of viral nucleic acid detection, SARS-CoV-2 serological antibody's detection plays an important role in the diagnosis of pregnant women and their newborns 13 .
It is well known that when a woman was infected with a particular virus during pregnancy, the fetus can obtain speci c maternal IgG through placental transport, which plays a passive immune role to protect the baby from virus infection after birth 13 . Changes of the level of baby's speci c antibodies can also prompt the intrauterine exposure risk. While the maternal gestational age (GA) at infected time and the duration of infection before delivery may affect their babies' seroconversion. The dynamic changes of antibodies against SARS-CoV-2 were different from those of ordinary respiratory viruses [16][17][18] . The acquisition, maintenance time and protective effect of maternal antibody in infants are still unclear 19 . Here, we described the serodynamic results of pregnant women with SARS-CoV-2 infection and their infants delivered in Zhongnan Hospital of Wuhan University in Wuhan, China.

Results
Objects and characteristics. Between January 27, 2020 and May 10, 2020, 26 pregnant women were con rmed to be infected with SARS-CoV-2 by laboratory evidences, and their 27 babies were tested for IgM and IgG antibodies against SARS-CoV-2 by chemiluminescence method. The positive rate of viral pneumonia-like in computed tomography (CT), SARS-CoV-2 seroconversion and viral nucleic acid test were 88.4% (23/26), 80.8% (21/26) and 34.6% (9/26) respectively in parturient women, while all infants were negative in SARS-CoV-2 nucleic acid test at birth. The age range of the mothers was 22-41 years, and the range of gestational age at admission was 31 +6 to 41 +1 weeks. There were 5 pregnant women with SARS-CoV-2 infection in the second trimester and 21 cases infected in the third pregnancy. 13 pregnant women had COVID-19-like symptoms such as fever and/or cough before delivery. Of the 27 infants, 21 were full-term, the other 6 were preterm including 1 pair of twins. The infants were normal at birth with Apgar scores all over 7. All infants were separated from their mothers immediately and were not breastfed before SARS-CoV-2 antibodies testing. The median time from onset of SARS-CoV-2 infection to delivery of parturients women was 10.5 days (1-107 days), and the primary seroconversion of IgG and IgM were 80.8% (21/26) and 53.9% (14/26) respectively. (Table 1).
Detection and seroconversion of mothers and infants. Wuhan was the epicenter of SARS-CoV-2 outbreak, and from February to June in 2020, all parturient women need to undergo the throat swab SARS-CoV-2 polymerase chain reaction with reverse transcription (RT-PCR) detection and lung CT examination before delivery. Since serological testing had not been widely used until March 2020, serum SARS-CoV-2 antibody detection was carried out in two stages: mothers who delivered before March 2020 received antibody testing in the follow-up stage after delivery, and pregnant women who gave birth after March 2020 were detected antibody before delivery.
Eleven pregnant women underwent serological testing 1-8 days before delivery. During pregnancy, they had positive of SARS-CoV-2 nucleic acid test or lung CT examination that showed lung viral pneumonialike changes. The median time from infection to delivery was 70 days (6-107 days), and the median time from infection to antibody detection was 69 days (16-99 days). All those mothers were IgG positive (11/11, 100%), 63.6% cases were IgM positive (7/11). The IgG was positive in 9 (9/11, 81.8%) infants born to this group mothers. Fifteen mothers who were con rmed with SARS-CoV-2 infection before delivery with symptoms, or without symptoms but with lung viral pneumonia-like changes, underwent antibodies test after delivery. The median time from infection to delivery was 4.5 days (1-15 days), and the median time from infection to antibody detection was 64.5 days (36-81 days). Among this group of mothers, 40% (6/15) were IgM positive, and 66.7% (10/15) were IgG positive. Of their 16 infants, only 2 cases (12.5%, 2/16) were IgG positive. From these results, we speculate that maternal SARS-CoV-2 seroconversion rate was related to the infection duration of pregnant women before delivery.  (Table 1). Of the 22 infants born to 21 IgG-positive mothers, only 11 (50.0%) were IgG positive.We found that only half of infants obtained maternal IgG from SARS-CoV-2 infected mothers, regardless of the gestational age at which the mothers were infected.
Factors related to infantile acquisition of maternal IgG. To further understand the serodynamic changes of IgG and IgM in mothers and infants, we compared the correlation between maternal antibody level and infant serum conversion. As shown in gure 1a and 1b, there was no correlation between duration of maternal infection time and the titer of serum IgM and IgG . Because all detection was performed 16 days after maternal infection, these results implied that although the IgM titer indicated a downward trend, maternal IgG and IgM rose rapidly to high levels during the observation period and maintained at a platform stage. And the levels of IgG were positive correlated to that of IgM (p=0.0035) (Figure 1c). We also compared the correlation of the maternal serum antibodies titer with that of their infants, and found that the serum IgG titer of infants was positive correlated to that their mothers (p=0.01) (Figure 1d). These data demonstrated that infants could acquired more IgG from those mothers with higher titer of serum IgG. A similar rule was found in the positive correlation between the serum antibody level of infants and the time of infection of mothers before delivery ( Figure 1d).
According to the time from onset of SARS-CoV-2 infection to delivery of mothers, we divided this group of data into 2 groups which were within 14-day and more than 14-day. The IgG seroconversion rates of mothers were 66.7% (10/15) and 100% (11/11), while the infants' IgG positive rates were 18.8% (3/16) and 81.8% (9/11), respectively (p value=0.002). Then we analyzed the relationship between the maternal and infantile IgG antibody titers and found that the two showed a positive correlation (Figure 1e). The maternal IgG antibody titer was used to predict the positive of infantile IgG after birth (>1s/co), with a cutoff value of 8.22 s/co, which had a sensitivity of 84.3% and a speci city of 93.3% (Figure 1f).
Effects of mothers with and without symptoms on IgG conversion in infants. It has been suggested that asymptomatic individuals had a weaker immune response to SARS-CoV-2 infection in non-pregnant population 20 . The seroconversion rate of mothers with symptoms before delivery was signi cantly higher than that of mothers without symptoms, and the IgM seroconversion rate of asymptomatic mothers was lower. However, there was no signi cant difference of IgG and IgM titer between these two groups of data ( Table 2). The dynamic characteristics of maternal IgG and IgM titer against SARS-CoV-2 of those with and without symptoms were consistent with reported above. While whether mothers had symptoms did not affect the IgG conversion rate of their infants.
Serodynamic characteristics of maternal IgG. In this study, four mothers were performed twice quanti cation tests of serum antibodies (Figure 2a). They all had symptoms of fever or cough at the beginning of the onset with typical COVID-19 pneumonia ndings of chest CT images. The median interval time between detection and onset of two tests were 26.5 days (17-46 days) and 72 days (62-91 days). The average serum titer of IgG was 11.1 s/co and 8.3 s/co, and the average titer of IgM was 8.5 s/co and 13.6 s/co respectively. The antibodies were also detected twice in their 4 infants on the rst day and the day of 31-63 after birth (Figure 2b). The IgM levels of all infants were below threshold, and the average IgG levels were 9.2 s/co and 1.0 s/co respectively. Infantile IgG levels decreased sharply in the rst two months of life, accounting for only 10.7% of the titer at birth. These results show that after two to three months of infection with SARS-CoV-2, the IgM levers of these 4 mothers with COVID-19 gradually increased while most IgG (3/4, 75.0%) showed a downward trend. This is not on par with the dynamic regularity of antibodies in other viral infection. Many similar results have reported that the serum antibodies dynamic of this novel virus surprised all of us 17,[21][22][23] .We also found that without breast feeding, the maternal protective effect in infants was rapidly eliminated naturally within two months after birth (Table 3). It is worth noting that this rate of decline was beyond expectations.

Discussion
In this study, we report antibody responses to SARS-CoV-2 in 26 pregnant women with COVID-19 and the seroconversion of their 27 babies. The probability of infantile acquisition of maternal antibody was related to the infection time before delivery and the serum antibody concentration of their mothers. The cutoff value was 8.22 s/co in mother, which had a sensitivity of 84.3% and a speci city of 93.3% for predicting postnatal infantile IgG positive (>1 s/co). In the rst two months of life, the IgG level of infants dropped sharply, only one tenth of that at birth. The IgM positive rate was 53.8% in mothers and negative in all infants. These results imply that maternal IgG provided limited protection for their infants, which may be helpful to improve vaccination strategy in the future.
The impacts of maternal SARS-CoV-2 infection on the fetus may include vertical transmission, abnormal intrauterine growth, abortion and stillbirth 3,24-26 . Due to the role of placental barrier, maternal SARS-CoV-2 IgG in infants can be used to observe whether they has acquired the passive immunity, while the increase of fetal IgM can be used to judge whether there is intrauterine exposure. At present, there are few studies focusing on serodynamics of infants born to mothers with SARS-CoV-2 infection during postpartum period. Different from previous case reports, our data show that no positive results of IgM were found in infants born to mothers with SARS-CoV-2, regardless of the mother's clinical symptoms, antibody titer and infection during the second or third trimester 1,5 . These results were further supported by polymerase chain reaction with reverse transcription (RT-PCR) detection of throat swabs, suggesting that the probability of intrauterine vertical transmission caused by maternal infection of SAR-CoV-2 during pregnancy is very rare in non critical cases 6 .
There are many factors that affect infants to obtain maternal passive immunity. Despite the abundance of antibodies produced by prenatal infection of mothers, infants did not obtain the IgG titer in proportion to their mothers due to the short time period between maternal infection and delivery. This is consistent with our expectations, in which the longer the mother infected the virus before delivery, the higher the seroconversion rate of the maternal IgG, and the greater the probability of infants obtaining maternal antibody.
Mothers who had been infected for more than two weeks give their babies more adequate antibody titers. However, the longer the intrauterine exposure was, the more likely it was to affect the development of the fetus, which constitutes other high-risk factors 6 . The maternal IgG transfer e ciency was also depended on the mother's immune response to SARS-CoV-2. The IgG titer of infants born to high titer IgG mothers also increased. The duration of effective concentration of maternal SARS-CoV-2 IgG in infants is related to vaccination procedures for infants and their risk of infection after birth. It is generally believed that maternal antibodies disappear gradually in 12-18 months after birth. However, the maternal protective effect of against SARS-CoV-2 in infants was rapidly eliminated naturally after birth. These results can help us understand the risk of infantile vulnerability and improve vaccination procedures in the future 27,28 . Based on the fact that the titer of maternal antibody in infants' serum decreased faster than expected, we believe that it may be an appropriate time to vaccinate a safe and reliable SARS-CoV-2 vaccination in 2 months after birth.
In summary, the infection time and antibody titer of mothers before delivery affected the effect of infants obtaining maternal IgG, and the passive immunity lasted for a short time and disappeared in 2 months after birth. Our study has some limitations. This study is a retrospective study and no comparison was made between the antibody titer of the mother before delivery and the antibody titer of the baby at birth.
And whether the speci c antibody in the baby can play a protective effect needs further observation, and more data are needed for the following studies.

Declarations
Con ict of interest statement: No con ict of interest.
Author Contributions: X Wang, P Yang, J Zheng, P Liu and D Zhao conceptualized the study design. X Wang, P Yang, C Wei, Y Zhang and J Guo recruited the patients, collected specimens, collected demographic, clinical, and laboratory data; X Wang, J Zheng, P Liu and P Yang interpreted the laboratory result; P Liu, J Zheng, Y Zhang and D Zhao analyzed the data; and X Wang, P Yang, J Zheng and D Zhao wrote the initial drafts of the manuscript; All authors read and approved the nal report and have no con icting issues with the contributions listed herein. 2. Diagnosis of pregnant women SARS-Cov-2 infection. According to the supply of detection kit, maternal antibody detection was divided into two stages. Wuhan was the epidemic center of SARS-CoV-2, all pregnant women suspected with SARS-CoV-2 infection would be examined the viral RNA by polymerase chain reaction with reverse transcription (RT-PCR) before delivery. Since the serological test had not be widely used until March 2020, the primary time of SARS-CoV-2 antibody detection was performed as: 11 cases were tested for antibody 1-8 days before delivery. They were con rmed to have nucleic acid or lung CT positive during pregnancy. The time from infection to delivery was 70 days (6-107 days), and the median time from infection to antibody detection was 69 days (16-99 days). After delivery, 15 mothers underwent antibody test, but the infection was con rmed before delivery (with or without symptoms, but lung CT was positive). The time from infection to delivery was 4.5 days (1-15 days), and the median time from infection to antibody detection was 64.5 days (36-81 days). In the seventh edition of the guidelines, serological test results were used as the basis for etiological diagnosis.     predicting the sensitivity and speci city of maternal antibodies transferring into infants. Pearson test was used to analyze the correlation between the two groups, P < 0.05 was considered to be statistically signi cant.

Figure 2
Dynamic changes of serum antibodies to SARS-CoV-2 in 4 COVID-19 mothers and their 4 infants. Antibody concentration ≥1.0s/ Co was de ned as positive. Solid lines mean levels of IgG, dotted lines mean levels of IgM. M, cases of mothers; Inf, cases of infants. a, antibodies levels of mothers; the serum antibodies detection interval of mother 1 was on the day of 27 and 91, mother 2 on the day of 16 and 71, mother 3 on the day of 17 and 64, and mother 4 on the day of 26 and 73 after onset. b, antibodies levels of infants; the serum antibodies detection interval of four infants was on the day at birth and the day of 63, 63, 46 and 31 after birth, the same day with their mothers respectively. All 4 infants were formula fed.