Shadow of a Pandemic: Persistence of Prenatal SARS-CoV-2 Antibodies in Newborn Blood Spots

Abstract

To investigate COVID-19 surveillance among pregnant women, the California Genetic Disease Screening Program conducted a screening performance and seroprevalence evaluation of maternal SARS-CoV-2 antibodies detected in banked newborn dried blood spots (DBS). We obtained seropositive results for 2890 newborn DBS from cohorts in 2020 and 2021 using Enable Bioscience's Antibody Detection by Agglutination-PCR (ADAP) assay for SARS-CoV-2 antibodies. To infer maternal infection, we linked 312 women with a known laboratory-confirmed COVID-19 episode with their newborn's DBS SARS-CoV02 antibody result. Among 2890 newborns, we detected 453 (15.7%) with SARS-CoV-2 antibodies in their DBS. Monthly snapshot statewide seroprevalence among neonates was 12.2% (95% CI 10.3-14.1%, n =1156) in December 2020 and 33.3% (95% CI 29.1-37.4%, n = 26) in March 2021. The longest time recorded from COVID-19 infection to a seropositive neonatal result was 11.7 months among the 312 mothers who had an available SARS-CoV-2 PCR test result. Approximately 94% (153/163) of DBS were seropositive when a known maternal infection occurred earlier than 19 days before birth. The estimated relative sensitivity of DBS to identify prevalent maternal infection was 85.1%, specificity 98.5% and PPV 99.2% (n = 312); the sensitivity was lowest during the December 2021 surge when many infections occurred within 19 days of birth. Fifty pre-pandemic specimens (100% seronegative) and 23 twin-pair results (100% concordant) support an intrinsic specificity and PPV of ADAP approaching 100%. Maternal infection surveillance is limited by a time lag prior to delivery, especially during pandemic surges.

Keywords: COVID-19; SARS-CoV-2; antibody testing; dried blood spots; newborn; newborn screening; positive predictive value; pregnancy; prenatal; seroprevalence; serosurveillance.

Figure 1

Adjusted Poisson prevalance estimates by cohort. Graphical results from a Poisson prevalence analysis showing whisker plots of least squares means as central circles and upper and lower 95% confidence interval bars for (A) maternal age group and (B) newborn race-ethnicity by the three cohorts: 1 October 2020 pilot, 2 December 2020 snapshot, and 3 March 2021 snapshot.

Figure 2

Maternal COVID-19 episode date: months before and after birth date. We plotted the months before and after birth of the mother–newborn matched data against the ADAP antibody results with all pandemic cohorts combined. Result indicates the dimensionless signal of ADAP assay ΔCt as the Ct value difference between the sample and a blank control (ΔCt = Ct_blank − Ct_sample). The assay cutoff was established as the 99 percentile of healthy control’s dried blood spot ΔCt or serum sample. Scattergram (A) shows seropositive and seronegative results above and below the ΔCt result cutoff of 1.8. The linear regression lines were fitted to the following models: ADAP result(detected: Yes) = 6.049449 − 0.316978 * Months-before-birth, ADAP result(detected: No) = −0.796007 − 0.09239 * Months-before-birth. Scattergram (B) shows the same data identifying the cohort responsible for each data point: 1 October 2020 pilot, 2 December 2020 snapshot, and 3 March 2021 snapshot. NBS DBS from March 2021 found inside the polygon were tested for antibodies to nucleocapsid protein (NP); all with positive. No other DBS were tested for NP antibodies. (C) shows the months a COVID-19 episode was identified prior to prenatal specimen collection by SARS-CoV-2 antibodies detected in serum by ADAP. These 7 banked serum specimens where the only ones we could identify where the prenatal specimen was collected after or around the date of a known COVID-19 episode. The one prenatal serum specimen collected 5 months after an episode of COVID-19 was identified is the same pregnant woman linked to the DBS result listed in (A,B) over 9 months before delivery.