Severe acute respiratory syndrome coronavirus 2 seroprevalence and longitudinal antibody response following natural infection in pregnancy: A prospective cohort study

Abstract

Background:

Antenatal care provides unique opportunities to assess severe acute respiratory syndrome coronavirus 2 seroprevalence and antibody response duration after natural infection detected during pregnancy; transplacental antibody transfer may inform peripartum and neonatal protection. We estimated seroprevalence and durability of antibodies from natural infection (anti-nucleocapsid immunoglobulin G) among pregnant people, and evaluated transplacental transfer efficiency.

Objective and design:

We conducted a cross-sectional study to measure severe acute respiratory syndrome coronavirus 2 seroprevalence, and a prospective cohort study to longitudinally measure anti-nucleocapsid immunoglobulin G responses and transplacental transfer of maternally derived anti-nucleocapsid antibodies.

Methods:

We screened pregnant people for the seroprevalence study between 9 December 2020 and 19 June 2021 for anti-nucleocapsid immunoglobulin G in Seattle, Washington. We enrolled anti-nucleocapsid immunoglobulin G positive people from the seroprevalence study or identified through medical records with positive reverse transcription polymerase chain reaction or antigen positive results in a prospective cohort between 9 December 2020 and 9 August 2022.

Results:

In the cross-sectional study (N = 1284), 5% (N = 65) tested severe acute respiratory syndrome coronavirus 2 anti-nucleocapsid immunoglobulin G positive, including 39 (60%) without prior positive reverse transcription polymerase chain reaction results and 42 (65%) without symptoms. In the prospective cohort study (N = 107 total; N = 65 from the seroprevalence study), 86 (N = 80%) had anti-nucleocapsid immunoglobulin G positive results during pregnancy. Among 63 participants with delivery samples and prior anti-nucleocapsid positive results, 29 (46%) were anti-nucleocapsid immunoglobulin G negative by delivery. Of 34 remaining anti-nucleocapsid immunoglobulin G positive at delivery with paired cord blood, 19 (56%) had efficient transplacental anti-nucleocapsid immunoglobulin G antibody transfer. Median time from first anti-nucleocapsid immunoglobulin G positive to below positive antibody threshold was 19 weeks and did not differ by prior positive reverse transcription polymerase chain reaction status.

Conclusions:

Maternally derived severe acute respiratory syndrome coronavirus 2 antibodies to natural infection may wane before delivery. Vaccines are recommended for pregnant persons to reduce severe illness and confer protection to infants.

Keywords

antibody natural infection pregnancy SARS-CoV-2 seroprevalence transplacental transfer

Introduction

Antenatal care offers a unique opportunity to assess severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) seroprevalence among pregnant people, including those with previously unknown infection.¹ Prior SARS-CoV-2 seroprevalence studies among pregnant people have been primarily cross-sectional, often focused at delivery in the hospital, and have potentially missed pregnant people infected earlier in pregnancy whose antibody response has waned by the time of delivery, including those with asymptomatic infection or mild disease.^2
–4

Natural infection with SARS-CoV-2 during pregnancy may provide some protection against infection during the peripartum period in pregnant people and their infants, but longitudinal immunological responses across the pregnancy-postpartum continuum have not been well characterized. COVID-19 vaccines are recommended for people who are pregnant, recently pregnant (including those who are lactating), trying to become pregnant, or who might become pregnant in the future, regardless of prior infection status.^5,6 In addition, while COVID-19 vaccines are approved in the United States for children as young as 6 months of age, they are not currently available or being studied in infants aged < 6 months.⁷ Prospective data on antibody responses following infection during pregnancy and evaluation of transplacental transfer of antibodies to neonates has the potential to provide important information on the duration of maternal and neonatal immunity following natural infection during pregnancy.⁸

We conducted a SARS-CoV-2 seroprevalence study among pregnant people in the Seattle, Washington metropolitan area, examined longitudinal SARS-CoV-2 anti-nucleocapsid (anti-N) immunoglobulin G (IgG) antibody responses as a marker of immune response to natural infection among participants with evidence of natural infection, and measured transplacental transfer of maternally derived anti-N antibodies. We hypothesized that 5% of pregnant women would have SARS-CoV-2 infection based on anti-N IgG antibody levels and models of SARS-CoV-2 prevalence in the region⁹ and the presence of anti-N IgG antibody will be durable through 4 months post-infection based on prior respiratory syncytial virus studies conducted in pregnancy.^10
–12

Materials and methods

Study setting and participants

Seroprevalence study

Pregnant people aged ⩾ 18 years seeking antenatal care at 14 affiliated clinics, or admitted to three labor and delivery units, at University of Washington (UW)-affiliated medical centers were eligible for participation in the seroprevalence study (see Supplemental Table 1). Participants were excluded if they were < 18 years old, not receiving antenatal care at a UW study site, a non-pregnant adult, did not have a blood sample collected during routine antenatal care, or a sample that lacked the blood volume needed to conduct SARS-CoV-2 anti-N IgG testing. Healthcare providers obtained informed oral consent to screen blood samples collected from pregnant people receiving antenatal care during the period 9 December 2020 to 19 June 2021 for anti-N IgG antibodies to SARS-CoV-2. Healthcare providers offered participation to pregnant people whenever possible. Samples for screening were derived from residual samples collected for routine clinical care at sites affiliated with two medical centers or collected specifically for screening after consent was obtained at sites affiliated with one medical center between 9 December 2020 and 30 June 2021. Individuals who did not have a blood sample available for antibody testing but provided oral consent for screening were not screened in the seroprevalence study. History of positive SARS-CoV-2 PCR results and COVID-19 disease severity and symptoms were abstracted from the electronic medical record for people who screened positive for anti-N IgG.

Prospective cohort study

Pregnant people with evidence of prior SARS-CoV-2 infection (with either documented positive reverse transcription polymerase chain reaction (RT-PCR) positive results via medical record reviews during pregnancy or within 6 months before pregnancy or with anti-N IgG positive results from the seroprevalence study) during the period 9 December 2020 through 9 August 2022 were eligible to enroll in a cohort study evaluating longitudinal SARS-CoV-2 anti-N IgG responses; follow-up samples included in this analysis were collected from 8 February 2021 through 23 February 2023. Rapid antigen results were used for diagnostics beginning in January 2022 due to regional PCR test shortages; therefore, beginning in January 2022, pregnant people who self-reported a positive rapid antigen test for SARS-CoV-2 infection were also eligible to enroll. Participants in the prospective cohort study, identified as having a history of SARS-CoV-2 infection either from the seroprevalence study or via medical records, provided written informed consent prior to participating in the cohort study. Participants in the prospective cohort study provided consent for collection of cord blood samples. Participants were excluded if they were < 18 years old, not receiving pregnancy care at a UW study site, or did not have plans to deliver at a UW-affiliated medical center. For participants identified through the seroprevalence study, the anti-N IgG result served as the enrollment sample. All participants were scheduled for follow-up blood sample collection at 1, 2, 3 months post-enrollment and delivery (including maternal and cord blood). Additional blood samples were collected at 1–2, 2–4, and 6 months postpartum if sample collection dates did not fall within the post-enrollment sample collection windows. All blood samples were tested for SARS-CoV-2 anti-N IgG antibodies. COVID-19 disease severity and symptoms were abstracted from the electronic medical record and classified as asymptomatic, mild, severe, or critical; disease severity was reported for the initial infection for individuals who were known to have multiple SARS-CoV-2 infections.¹³ Participants were also asked about the presence of symptoms during surveys administered at enrollment interviews; individuals who did not self-report symptoms during the survey and had no symptoms recorded in the medical records were classified as asymptomatic.

Laboratory methods

SARS-CoV-2 anti-N IgG serology

Samples collected for SARS-CoV-2 serology were tested using the Abbott Architect chemiluminescent immunoassay (CMIA), an automated qualitative test designed to detect anti-N IgG to SARS-CoV-2 (Abbott, Abbott Park, Illinois, USA) at UW. This assay has high sensitivity (100%, 17 days post-infection) and specificity (> 99.9%) in the 2 months after SARS-CoV-2 infection.^14
–16 Samples with an Abbott index ⩾ 1.4 were considered positive per manufacturer recommendations.¹⁷

Exposure and outcome variables

The primary exposures were SARS-CoV-2 diagnosis, symptoms, and disease severity; the primary outcomes were presence and durability of SARS-CoV-2 anti-N IgG, and transplacental transfer of maternally derived anti-N IgG antibodies.

Statistical analysis

Transplacental transfer ratios were calculated as the Abbott index from cord blood collected at delivery divided by the Abbott index from maternal blood collected at delivery with a ratio of ⩾ 1 considered as efficient transplacental transfer.¹⁸ Vaccination status based on medical record abstraction and/or self-report was classified as follows: partial with one dose of a messenger RNA (mRNA) vaccine, fully vaccinated if two doses of an mRNA vaccine or one dose of a viral vector vaccine, and boosted if three doses of an mRNA vaccine (or at least one dose plus a viral vector vaccine) or two doses of viral vector vaccine. Wilcoxon rank sum tests were used to compare distributions of continuous variables. Prevalence ratios were calculated to assess potential sources of bias in people who consented versus declined screening, and also between those who were screened versus did not have a blood sample collected, in the seroprevalence study. We assessed co-factors of SARS-CoV-2 anti-N IgG positive results, which were calculated using generalized linear models (GLMs) with Poisson family and log link. Kaplan–Meier analysis was used to calculate time to anti-N IgG Abbott below the positive threshold (index < 1.4); differences in curves by PCR status were assessed using the log-rank test. Anti-N IgG Abbott index results were log₁₀ transformed to reduce skewness, and general estimating equations (GEEs) with a Gaussian link and robust standard errors were constructed to measure the rate of change in log₁₀ anti-N IgG response over time since first anti-N IgG positive result, time since first RT-PCR positive result or antigen positive result, and time since infection (first RT-PCR positive, antigen positive, or anti-N IgG positive result). GEE models also assessed whether the average monthly rate of change in log₁₀ anti-N IgG response since the first positive result differed by potential covariates, including pregnancy status, trimester of infection (among pregnant people), presence of symptoms, and vaccination status. A multivariable model was constructed with all covariates significant at p < 0.1.

Sample size

Seroprevalence study: With a sample size of 1263 pregnant people, we have 1.22% precision to detect seroprevalence based on anti-N IgG results of 5% (95% confidence interval (CI): 3%–7%) by the end of pregnancy (i.e. at the time of delivery admission); an additional 21 people were enrolled by healthcare providers after the target sample size was reached; therefore, the overall sample size was 1284.

Prospective cohort study: We calculated a sample size of 50 pregnant people with evidence of SARS-CoV-2 would be required to measure correlates of infection during pregnancy (to be presented in a future article) with effect sizes of 2.5–3.4 for a range of co-factors for infection with prevalence ranging 10%–40%, assuming 80% power, alpha = 0.05, and two-sided testing; however, we continued to accrue additional infections in the cohort due to the convenience sample of pregnant people with virologic evidence of SARS-CoV-2 infection available during the study time frame and reported to the study by healthcare providers. We adhered to the STROBE guidelines in preparation of the article.

Results

SARS-CoV-2 seroprevalence study

Overall, we identified 8627 pregnant people who received antenatal care or delivered at UW enrollment sites during the seroprevalence study enrollment period (see Figure 1, Supplementary Table 1); 2685 pregnant people (31%) were offered participation for the study (2213 consented and 472 declined). Only 31% of all people seeking care were offered participation due to individual providers’ inability to incorporate consent into clinical care. Among the 2213 who consented to blood screening for SARS-CoV-2 anti-N IgG, 1284 (58%) had blood samples tested. A total of 42% (929/2213, 42%) individuals provided consent for screening but did not have a blood sample available for antibody testing and were not screened in the seroprevalence study. There were some significant differences in pregnant people who consented to screening and had samples tested by age, race, and ethnicity (p < 0.05, see Supplementary Table 2); notably Black race was more frequently reported among those who declined screening (19%) than those who consented to screening (7%). Among those who consented, Black race was less frequently reported among those who did not have blood tested (1%) than those who were screened (8%). Among pregnant people screened for anti-N IgG, the median age was 32 years (interquartile range (IQR) 29–36) with a median gestational age at screening of 16 weeks (IQR 11–38; see Table 1). SARS-CoV-2 seroprevalence among pregnant people screened was 5% (65/1284), of whom 60% (n = 39) did not have a prior RT-PCR positive test result during pregnancy documented in their medical record. Monthly seroprevalence was significantly different over time (p = 0.03), peaking in December 2020 at 8.2% prior to the World Health Organization variants of concern becoming dominant in the area, and was lowest in April 2021 before the Delta variant became dominant (see Supplementary Figure 1).¹⁹ There were significant differences in seroprevalence by race and ethnicity (see Table 1). Among those who tested negative for anti-N IgG, 57% were White compared with 32% of those testing positive for anti-N IgG. Compared to White pregnant people, seroprevalence was at least three times as high among those who were Black, Native Hawaiian or Pacific Islander, American Indian or Alaska Native, or identified as another race (p < 0.05 for all). Seroprevalence was also three times as high among pregnant people of Hispanic versus non-Hispanic ethnicity. Older pregnant people were less likely to test anti-N IgG positive with each year increase in age associated with 8% lower seroprevalence (prevalence ratio (PR): 0.92, 95% CI for PR: 0.87–0.96). Among 65 pregnant people identified in the seroprevalence study with anti-N IgG positive results, 23 (35%) had symptoms (21 mild, 2 severe), 2 (3%) of whom were hospitalized for COVID-19. RT-PCR dates were available for 22 of 26 pregnant people with a RT-PCR positive result, with a median time between RT-PCR positive date and blood collection date for anti-N IgG of 7 weeks (IQR 4–15); 4 people were RT-PCR positive before pregnancy.

Figure 1.

Flowchart of pregnant people enrolled in prospective cohort study with a history of SARS-CoV-2 infection (from seroprevalence study or identified through the medical system) as having a prior positive RT-PCR or antigen test.

Table 1.

Baseline characteristics of pregnant people screened through seroprevalence study for SARS-CoV-2 anti-N IgG, by anti-N IgG screening result.

	Screened for anti-N IgG^a (n = 1284)	Anti-N IgG-^a (n = 1219)	Anti-N IgG+^a (n = 65)	PR^b (95% CI)	p-value
	n (%) or median (IQR)			PR^b (95% CI)	p-value
Age^c	32 (29–36)	32 (29–36)	29 (26–34)	0.92 (0.87–0.96)	0.001
Gestational age (weeks) at screening^c,d	16 (11–38)	15 (11–38)	18 (11–38)	1.00 (0.99–1.02)	0.75
Race
American Indian/Alaska Native	17 (1)	15 (1)	2 (3)	3.99 (1.01–15.68)	0.048
Asian	254 (20)	250 (21)	4 (6)	0.53 (0.18–1.54)	0.25
Black	101 (8)	88 (7)	13 (20)	4.36 (2.26–8.44)	< 0.0001
Native Hawaiian/Pacific Islander	31 (2)	26 (2)	5 (8)	5.47 (2.21–13.54)	< 0.0001
White	712 (55)	691 (57)	21 (32)	REF
Other	112 (9)	94 (8)	18 (28)	5.45 (3.00–9.90)	< 0.0001
Not reported	57 (4)	55 (4)	2 (3)	—	—
Hispanic ethnicity	138/1239 (11)	120/1175 (10)	18/64 (28)	3.12 (1.86–5.23)	< 0.0001

SARS-CoV-2: severe acute respiratory syndrome coronavirus 2; Anti-N IgG: anti-nucleocapsid immunoglobulin G; PR: prevalence ratio; CI: confidence interval; IQR: interquartile range.

Anti-N IgG tested using Abbott Architect chemiluminescent immunoassay (CMIA), a semi-qualitative test designed to detect anti-nucleocapsid (anti-N) IgG of SARS-CoV-2. IgG+ if Abbott index ⩾ 1.4.

PR, prevalence ratio comparing anti-N IgG+ versus anti-N IgG-.

Per 1 unit increase in the model.

Missing gestational age for 10 people (9 IgG– and 1 IgG+ people).

Median Abbott index among 65 pregnant people with anti-N IgG positive results (Abbott index ⩾ 1.4) and available Abbott index in the seroprevalence study was 3.18 (IQR 2.06–5.00). Median Abbott index was significantly higher among those with COVID-19 symptoms reported in the medical record than without (median 4.39, IQR 3.18–5.42 versus median 2.49, IQR 1.92–4.30, respectively; p = 0.02) and among those with RT-PCR positive results than without (median: 4.19, IQR 2.87–5.42 versus 2.49, IQR 1.89–4.16, respectively; p = 0.02). Fifteen participants received a COVID-19 vaccine (14 fully vaccinated, 1 partially vaccinated) prior to initial anti-N IgG positive test, but no differences in anti-N IgG Abbott index were detected between people who were and were not vaccinated (data not shown).

Prospective cohort study

We enrolled 107 pregnant people with evidence of prior SARS-CoV-2 infection in the prospective cohort study; 23 (26%) from the seroprevalence study and 84 (74%) identified through medical records (see Figure 1). Median age was 32 years (IQR 30–35) and median gestational age at first blood sample collection was 25 weeks (IQR 16–37). Among 91 pregnant people with RT-PCR positive or antigen positive results and available test dates, median time between identification of positive result and first blood draw was 6.4 (IQR 4.4–15.5). Less than 5% of participants identified as either non-binary or did not report their gender; all other participants identified as women. Overall, 21 (20%) were anti-N IgG negative, all with prior RT-PCR positive results. Among 86 of 107 (80%) people with anti-N IgG positive results, most (n = 61, 71%) also had a prior RT-PCR positive result, 10 (14%) positive by anti-N IgG alone, and 15 (17%) also had a prior positive antigen test but no prior RT-PCR positive result.

Pregnant people with SARS-CoV-2 anti-N IgG positive results

Of 107 pregnant people enrolled in the prospective cohort, 86 (80%) had anti-N IgG positive results documented at any time during the study (see Table 2). Among 80 people who were pregnant or in delivery when the first anti-N IgG positive result was collected, the median gestational age was 25 weeks (IQR 17–37). Most (n = 61, 76%) had a prior positive RT-PCR result (5 recorded prior to pregnancy (median 8 weeks before; range 0.6–24 weeks), data not shown); 64 (75%) reported having symptoms, and 13 (15%) had neither reported symptoms or a prior RT-PCR positive result (data not shown). Among the 64 people with symptoms reported, 96% (n = 62) had mild symptoms and 4% (n = 2) had severe symptoms, both of whom were hospitalized for COVID-19. Median Abbott index at the first anti-N IgG positive sample was 3.21 (IQR 2.07–5.12, range (data not shown) 1.43–9.87), and was not significantly different between those with (3.72, IQR 2.33–5.12) and without (2.84, IQR 1.97–5.00) prior RT-PCR positive results (p = 0.30; data not shown). Median Abbott index at the first anti-N IgG positive sample was also not significantly different between those who were symptomatic versus asymptomatic (3.73 versus 2.7, p = 0.12; data not shown). Among 56 pregnant people with known dates for their prior RT-PCR positive result, the median time from RT-PCR positive result to first available anti-N IgG positive result was 6 weeks (IQR 4–10; data not shown).

Table 2.

Baseline characteristics of pregnant people enrolled in the prospective cohort with a history of SARS-CoV-2 infection and positive anti-N IgG.

	Anti-N IgG+ (N = 86)	Anti-N IgG+ with ⩾ 1 follow-up anti-N IgG result (N = 68)
	n (%) or Median (IQR)
Age (years)	32 (30–35)	32 (30–35)
Gestational age at first blood collection (weeks)^a	25 (16–37)	21 (15–33)
Prior RT-PCR+ result	61 (71)	45 (66)
Abbott Index at first anti-N IgG+^b	3.21 (2.07–5.12)	3.39 (2.16–4.92)
COVID-19 severity
Asymptomatic	22 (26)	21 (31)
Mild	62 (72)	46 (68)
Critical	0 (0)	0 (0)
Severe	2 (2)	1 (1)
Hospitalized for COVID-19	2 (2)	1 (1)
Vaccination status at first anti-N IgG+^c
None	35 (41)	29 (43)
Partial	2 (2)	2 (3)
Fully	18 (21)	12 (18)
Boosted	31 (36)	25 (37)

SARS-CoV-2: severe acute respiratory syndrome coronavirus 2; Anti-N IgG: anti-nucleocapsid immunoglobulin G; IQR: interquartile range; RT-PCR: reverse transcription polymerase chain reaction.

Three people (one with ⩾ 1 follow-up) whose first blood collection was after delivery.

Abbott index at first anti-N IgG+ ⩾ 1.4.

Vaccination status was classified as partial with one dose of an mRNA vaccine or missing vaccine type; fully vaccinated if two doses of an mRNA vaccine, two doses with missing vaccine type, or one dose of a viral vector vaccine; and boosted if three vaccine doses of any type (three mRNA vaccines, two doses plus a viral vector vaccine, or three doses with one or missing vaccine types) or two doses of viral vector vaccine.

Among 68 pregnant people with an anti-N IgG positive result and at least one subsequent blood sample, the median time from the first positive anti-N IgG result to anti-N IgG below the positive threshold was 19 weeks (95% CI: 13–24) and was similar regardless of whether they had a prior RT-PCR positive result (median 19 weeks for both; see Figure 2(a)). Median time from infection to anti-N IgG below the positive threshold was 26 weeks (95% CI: 21–32; see Figure 2(b)).

Figure 2.

Time to anti-N IgG below positive threshold among pregnant people with serologic evidence of prior SARS-CoV-2 infection. (a) Time from anti-N IgG+ to anti-N IgG below positive threshold (n = 68). (b) Time from infection to anti-N IgG below positive threshold (n = 85).

Overall, the average monthly rate of change was –0.018 log₁₀ Abbott index per month since first blood collection (see Figure 3(a)), –0.025 since RT-PCR positive or antigen positive result (see Figure 3(b)), and –0.024 since first positive result (see Figure 3(c)). During pregnancy, there was on average a 0.124 decline in the change in log₁₀ Abbott index per month since the first positive result compared to a 0.009 increase during the postpartum period (p < 0.001). The average monthly decline was significantly larger among those who were symptomatic than asymptomatic (–0.065 versus –0.002, respectively; p < 0.001). In a multivariable model that included pregnancy status and symptomatic infection, significant differences in the rate of change remained (p < 0.001 for both symptoms and pregnancy status). There were no differences in the rate of change in log₁₀ index since first positive result by trimester of infection or vaccination status (results not shown).

Figure 3.

Abbott anti-N IgG index and timing among pregnant people with a history of SARS-CoV-2 infection among those with ⩾ 2 samples available. (a) Time since first anti-N IgG+ (n = 68). (b) Time since RT-PCR+ (or antigen +) (n = 71). (c) Time since infection (n = 81).

Anti-N IgG at delivery and transplacental antibody transfer

Maternal blood samples were collected at delivery from 63 people with prior anti-N IgG positive results; 34 (54%) remained anti-N IgG positive and 29 (46%) were anti-N IgG negative at delivery. Paired maternal and cord blood were collected from 59 of 63 participants at delivery. Delivery samples for people with paired cord blood were collected at a median of 10 weeks after the first anti-N IgG positive result (n = 57) and a median 16 weeks after the RT-PCR positive result (n = 39). Among 34 participants with anti-N IgG positive results at delivery and available cord blood results, most (n = 32, 94%) cord blood samples were also anti-N IgG positive (see Figure 4). However, just over half (n = 19, 56%) had evidence of efficient transplacental transfer of anti-N IgG with transfer ratio of ⩾ 1. The corresponding median IgG index for maternal blood and cord blood samples above the positive threshold was 3.10 (IQR 1.85–4.84) and 3.58 (IQR 2.89–4.71), respectively. Among participants remaining anti-N IgG positive at delivery, the median placental transfer ratio of maternally derived anti-N IgG was 1.06 (IQR 0.82–1.56). Maternal samples collected at delivery that were anti-N IgG negative were collected at a median of 20 weeks after the initial anti-N IgG positive result (n = 27) and 22 weeks after the prior RT-PCR positive result (n = 33).

Figure 4.

IgG index among paired maternal cord blood samples among participants with anti-N IgG positive results at delivery (n = 34 pairs).

Discussion

SARS-CoV-2 seroprevalence among 1284 pregnant people was 5% in the Seattle metropolitan area between December 2020 and June 2021, with highest prevalence of ~8% in December 2020 to January 2021. Over half of pregnant people identified with serological evidence of natural SARS-CoV-2 infection did not have a prior RT-PCR or antigen positive test documented in their medical record. Among all people with a prior SARS-CoV-2 infection enrolled in the cohort study, 15% were both asymptomatic and unaware they were previously infected with SARS-CoV-2. While our estimates of asymptomatic (and mild) infections among pregnant people with RT-PCR+ results are similar to other pregnant cohorts,^20,21 our assessment of anti-N IgG enabled us to characterize longitudinal antibody responses among people with less severe infections that may be missed in other seroprevalence SARS-CoV-2 studies conducted in pregnancy. In our study, we found that pregnant people with serologic evidence of SARS-CoV-2 were more likely to be Black, Native Hawaiian/Pacific Islander, or American Indian/Alaska Native race or Hispanic ethnicity; these observed disparities concur with results from prior studies of SARS-CoV-2 infection conducted among pregnant people King County, WA (location of Seattle metropolitan area)²² and in other parts of the United States.^3,22
–24 We found anti-N IgG responses waned over time following natural infection; these results concur with other studies of waning anti-N IgG levels in non-pregnant cohorts, particularly those using the Abbott Architect platform.^25,26 Over half of pregnant people with anti-N IgG antibodies in pregnancy remained anti-N IgG positive at delivery, with over 90% of cord blood samples also remaining anti-N IgG positive. Over half of participants remaining anti-N IgG positive at delivery had evidence of efficient transplacental transfer of maternal anti-N IgG with ratio of cord to maternal anti-N antibody ratio of ⩾ 1. Assuming that pregnant people who were anti-N IgG negative at delivery would have had corresponding negative cord samples, we estimate that about half of people who had prior evidence of anti-N IgG antibodies efficiently transferred anti-N IgG transplacentally. These results concur with prior studies demonstrating correlations between maternal IgG and cord blood IgG concentrations.³ Our findings have important implications for conveying information about potential protection following natural infection during (or before) pregnancy. Natural infection may contribute to decisions to delay or decline vaccination under the assumption that natural infection provides sufficient immunity against maternal reinfection, infant infection, and COVID-19 morbidity and mortality. Median time to a decrease in anti-N IgG below the threshold for positive was 4.4 months after anti-N IgG was first detected and 6.0 months after testing positive by RT-PCR or antigen. While this duration was similar to some studies conducted among non-pregnant cohorts²⁷ and shorter than others,²⁸ it corresponds to one-third of pregnant people with previous SARS-CoV-2 infection no longer having positive anti-N IgG antibodies by the time of delivery—a critical time (including for late pregnancy) for transplacental transfer of protective antibodies to infants.

Some studies of transplacental transfer of maternal IgG to SARS-CoV-2 suggest efficiency is greatest when infection occurs during the second trimester, while others have found transfer efficiency is higher in the third trimester.²⁹ Differences in these studies may be attributed to duration of infection prior to delivery, with potential for antibody waning if infection occurs early in pregnancy, and insufficient time to mount a robust immune response if infection occurs too close to delivery. Alternatively, efficiency of antibody transfer may be related to severity of disease, or disease severity and timing of infection. Continued collection of data on potential correlates of infant protection from severe illness by maternally derived antibodies (whether from natural infection or vaccination) will be important to evaluate the degree of protection afforded by maternal immunity.

Our study had several strengths. The seroprevalence study included a large catchment area (medical facilities that capture > 6200 deliveries annually) and a similar racial and ethnic distribution to King County and the general Seattle metropolitan area. Since pregnant people are generally healthy, our estimate may be representative of general population prevalence.³⁰ Characterization of antibody responses over time following natural infection includes pregnant people who may have had undiagnosed, mild and/or asymptomatic infection identified through the seroprevalence study. Similarly, we collected cord blood and measured transplacental antibody transfer of anti-N IgG in a cohort of pregnant people that included infections that were less severe. Finally, we were able to longitudinally characterize maternal infection for at least 6 months, throughout pregnancy and the early postpartum period.

Our study also has some limitations. While the Abbott Architect anti-N IgG sensitivity is high in the weeks to months immediately after infection, the assay preferentially detects low affinity antibodies known to wane^25,31 and it may fail to detect individuals infected with SARS-CoV-2 whose anti-N IgG response waned below the positive threshold prior to sample collection.²⁶ Even with detectable levels of anti-N IgG, the degree of potential protection offered by anti-N (versus anti-spike or other immune markers) to neonates or young infants is unknown. Estimates of > 50% of pregnant people identified as having serologic evidence of natural infection without a known prior infection history were based on medical record review; it is possible that some participants may have had a prior positive RT-PCR or antigen test that was not documented in their medical record. During later periods of enrollment for the prospective cohort, a self-reported positive antigen test was considered evidence of prior infection, which could introduce bias not present earlier in the enrollment period. While we were unable to determine the timing of infection for 10% of women without a prior RT-PCR or antigen positive result, the duration of anti-N IgG response at or above the Abbott index threshold were similar among those with and without prior RT-PCR or antigen positive results, which suggest our results were robust despite this limitation. Pregnant people with serologic evidence of SARS-CoV-2 were captured when different dominant variants were circulating in the area. Thus, their antibody responses are likely based on a variety of variants; however, we were unable to characterize responses by variant. In addition, following the circulation of Omicron variants in late 2021 and early 2022, the median time between identification of RT-PCR positive or antigen positive result and first blood draw was short (5.0 weeks) compared to 11.4 weeks prior to widespread circulation of Omicron. Finally, differences in age, race, and ethnicity of participants who consented versus declined and those who did and did not have blood samples available for testing may have biased the results; similarly, there may be differences in people who were and were not offered participation in the seroprevalence study resulting in selection bias. We anticipate SARS-CoV-2 seroprevalence may be biased toward the null due to the larger proportion of participants of non-White race and Hispanic ethnicity who declined blood screening for anti-N IgG, and these groups are more likely to have become infected with SARS-CoV-2 during the study period.

Conclusion

In conclusion, we found that anti-N IgG levels following natural infection during pregnancy wane over time, with only about half of pregnant people remaining with anti-N IgG positive results at delivery and slightly over half of these individuals had evidence of efficient transplacental anti-N antibody transfer to their infants. More information is needed on the timing and severity of maternal natural infection provided by anti-N IgG, as well as other immune markers, to better understand the transfer and potential for protection to infants. Anti-N IgG levels may not indicate a sustained immunological response. These data further support the use of vaccines during pregnancy among people, even among those who have previously been infected with SARS-CoV-2.^5,6

Supplemental Material

sj-docx-1-whe-10.1177_17455057231190955 – Supplemental material for Severe acute respiratory syndrome coronavirus 2 seroprevalence and longitudinal antibody response following natural infection in pregnancy: A prospective cohort study

Supplemental material, sj-docx-1-whe-10.1177_17455057231190955 for Severe acute respiratory syndrome coronavirus 2 seroprevalence and longitudinal antibody response following natural infection in pregnancy: A prospective cohort study by Alison L Drake, Jaclyn N Escudero, Morgan C Aurelio, Erica A Wetzler, Sascha R Ellington, Lauren B Zapata, Romeo R Galang, Margaret C Snead, Krissy Yamamoto, Carol C Salerno, Barbra A Richardson, Alexander L Greninger, Alisa B Kachikis, Janet A Englund and Sylvia M LaCourse in Women's Health

Footnotes

Acknowledgements

This study is partially funded (contract 75D30120C09610) by the US Centers for Disease Control and Prevention (CDC), which also provided technical assistance related to analysis and interpretation of data and writing the report. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the US Centers for Disease Control.

Declarations

ORCID iDs

Morgan C Aurelio

Lauren B Zapata

Margaret C Snead

Supplemental material

Supplemental material for this article is available online.

References

McClymont

Elwood

Sekirov

, et al. Population SARS-CoV-2 seroprevalence using antenatal serum samples in British Columbia, Canada. J Obstet Gynaecol Can 2021; 43(11): 1242–1243.

Sharma

Singh

Hillman

, et al. Seroprevalence of SARS-CoV-2 antibodies among first-trimester pregnant women during the second wave of the pandemic in India. Int J Gynaecol Obstet 2022; 158: 494.

Flannery

Gouma

Dhudasia

, et al. Assessment of maternal and neonatal cord blood SARS-CoV-2 antibodies and placental transfer ratios. JAMA Pediatr 2021; 175(6): 594–600.

Centers for Disease Control and Prevention. COVID-19 vaccination for pregnant people to prevent serious illness, deaths, and adverse pregnancy outcomes from COVID-19. Health Alert Network (HAN), 2021, https://emergency.cdc.gov/han/2021/han00453.asp

Centers for Disease Control and Prevention. COVID-19 vaccines while pregnant or breastfeeding, https://www.cdc.gov/coronavirus/2019-ncov/vaccines/recommendations/pregnancy.html (accessed 13 June 2022).

The American College of Obstetrics and Gynaecology. ACOG and SMFM recommend COVID-19 vaccination for pregnant individuals, https://www.acog.org/news/news-releases/2021/07/acog-smfm-recommend-covid-19-vaccination-for-pregnant-individuals

Fleming-Dutra

Wallace

Moulia

, et al. Interim recommendations of the Advisory Committee on Immunization Practices for use of Moderna and Pfizer-BioNTech COVID-19 vaccines in children aged 6 months–5 years — United States, June 2022. MMWR Morb Mortal Wkly Rep 2022; 71: 859–868.

Shook

Atyeo

Yonker

, et al. Durability of anti-spike antibodies in infants after maternal COVID-19 vaccination or natural infection. JAMA 2022; 327(11): 1087–1089.

WA situation report 1: COVID-19 transmission across Washington State, 2020, https://iazpvnewgrp01.blob.core.windows.net/source/2021-02/reports/pdf/COVID-19_transmission_patterns_across_Washington_State.pdf

10.

Chu

Steinhoff

Magaret

, et al. Respiratory syncytial virus transplacental antibody transfer and kinetics in mother-infant pairs in Bangladesh. J Infect Dis 2014; 210(10): 1582–1589.

11.

Falsey

Singh

Walsh

EE.

Serum antibody decay in adults following natural respiratory syncytial virus infection. J Med Virol 2006; 78(11): 1493–1497.

12.

Ochola

Sande

Fegan

, et al. The level and duration of RSV-specific maternal IgG in infants in Kilifi Kenya. PLoS ONE 2009; 4(12): e8088.

13.

Lokken

Walker

Delaney

, et al. Clinical characteristics of 46 pregnant women with a severe acute respiratory syndrome coronavirus 2 infection in Washington State. Am J Obstet Gynecol 2020; 223(6): 911.e1–911.e14.

14.

Alfego

Sullivan

Poirier

, et al. A population-based analysis of the longevity of SARS-CoV-2 antibody seropositivity in the United States. EClinicalMedicine 2021; 36: 100902.

15.

Bryan

Pepper

Wener

, et al. Performance characteristics of the Abbott Architect SARS-CoV-2 IgG assay and seroprevalence in Boise, Idaho. J Clin Microbiol 2020; 58(8): e00941-20.

16.

Interiano

Muze

Turner

, et al. Longitudinal evaluation of the Abbott Architect SARS-CoV-2 IgM and IgG assays in a pediatric population. Pract Lab Med 2021; 25: e00208.

17.

Abbott Architect SARS-CoV-2 lgG user guide, 2021, https://manuals.plus/abbott/architect-sars-cov-2-lgg-manual#axzz7ZnPIihwS

18.

Beharier

Plitman Mayo

Raz

, et al. Efficient maternal to neonatal transfer of antibodies against SARS-CoV-2 and BNT162b2 mRNA COVID-19 vaccine. J Clin Invest 2021; 131(13): e150319.

19.

UW Medicine. UW Virology COVID-19 dashboard: sequencing information, https://depts.washington.edu/labmed/covid19/#sequencing-information

20.

Buckley

Mills

Paul

, et al. SARS-CoV-2 antibody response among women infected during pregnancy. Am J Perinatol 2022; 39(7): 707–713.

21.

Edlow

Collier

, et al. Assessment of maternal and neonatal SARS-CoV-2 viral load, transplacental antibody transfer, and placental pathology in pregnancies during the COVID-19 pandemic. JAMA Netw Open 2020; 3(12): e2030455.

22.

King County. COVID-19 race and ethnicity dashboard, 2023, https://kingcounty.gov/depts/health/covid-19/data/race-ethnicity.aspx

23.

Gonik

Alonso

Gonik

SARS-CoV-2 seroprevalence in Florida Department of Health in Palm Beach County obstetric clinics: a cross-sectional study during the first pandemic surge. Am J Perinatol 2023; 40: 912–916.

24.

Zambrano

LDE

Ellington

Strid

, et al. Characteristics of symptomatic women of reproductive age with laboratory-confirmed SARS-CoV-2 infection by pregnancy status—United States, January 22–October 3, 2020. MMWR Morb Mortal Wkly Rep 2020; 69: 1641–1647.

25.

Kerr

Allen

Hughes

, et al. SARS-CoV-2 anti-nucleocapsid assay performance in healthcare workers at baseline and 6 months. Ir J Med Sci 2022; 191(3): 1089–1092.

26.

Van Elslande

Oyaert

Ailliet

, et al. Longitudinal follow-up of IgG anti-nucleocapsid antibodies in SARS-CoV-2 infected patients up to eight months after infection. J Clin Virol 2021; 136: 104765.

27.

Gerhards

Thiaucourt

Kittel

, et al. Longitudinal assessment of anti-SARS-CoV-2 antibody dynamics and clinical features following convalescence from a COVID-19 infection. Int J Infect Dis 2021; 107: 221–227.

28.

Mariam

Chopikyan

Kasparova

, et al. Kinetics of anti-nucleocapsid IgG response in COVID-19 immunocompetent convalescent patients. Sci Rep 2022; 12(1): 12403.

29.

Surinach-Ayats

Carbonell-Estarellas

Becerra-Hervas

, et al. [Epidemiology and characteristics of SARS-CoV-2 infection in the newborn and pregnant woman. Transplacemental transfer of immunoglobulins]. Enferm Infecc Microbiol Clin. Epub ahead of print 13 May 2022. DOI: 10.1016/j.eimc.2022.04.009.

30.

Angulo

Finelli

Swerdlow

DL.

Reopening society and the need for real-time assessment of COVID-19 at the community level. JAMA 2020; 323(22): 2247–2248.

31.

Theel

Johnson

Kunze

, et al. SARS-CoV-2 serologic assays dependent on dual-antigen binding demonstrate diverging kinetics relative to other antibody detection methods. J Clin Microbiol 2021; 59(9): e0123121.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB