Quantitating SARS-CoV-2 neutralizing antibodies from human dried blood spots

Abstract

In the earliest days of the COVID-19 pandemic, the collection of dried blood spots (DBS) enabled public health laboratories to undertake population-scale seroprevalence studies used to estimate rates of SARS-CoV-2 exposure. With SARS-CoV-2 seropositivity levels now estimated to exceed 94% in the United States, attention has turned to using DBS to assess neutralizing antibodies within cohorts of interest. With this goal in mind, we generated contrived DBS (cDBS) and whole blood-derived DBS from convalescent and vaccinated individuals and subjected DBS eluates to a battery of assays, including a SARS-CoV-2 multiplexed microsphere immunoassay (MIA), a receptor binding domain (RBD)-human ACE2 inhibition assay (iACE2), a cell-based pseudovirus neutralization assay, and real-time PCR-based surrogate neutralization assay (NAB-Sure). The DBS results were benchmarked against paired serum samples tested in a clinically validated SARS-CoV-2 plaque reduction neutralization titer (PRNT) assay. The results of an 8-plex MIA and NAB-Sure assays demonstrated highly significant correlations with PRNT values when evaluated with a panel of 86 paired serum-DBS samples. Both the MIA and NAB-Sure are adaptable to automated liquid handlers for high-throughput capacity. While neutralizing assays were limited to the ancestral SARS-CoV-2 WA1, this study nonetheless represents an important proof of concept demonstrating the potential utility of DBS as a biospecimen type for use in assessing immunity to SARS-CoV-2 at the community and population levels.IMPORTANCESARS-CoV-2 variants of concern continue to circulate globally and remain a serious health threat to large segments of the population. From a public health standpoint, identifying vulnerable communities based on immune status is critical in terms of vaccine booster recommendations. In this report, we investigated the utility of dried blood spots (DBS) as a biospecimen type from which to estimate SARS-CoV-2 neutralizing antibody titers. Using contrived and whole blood-derived DBS, we demonstrate that SARS-CoV-2 neutralizing antibodies are readily measurable in DBS eluates and correlate with plaque reduction neutralization titer (PRNT) values from paired serum samples. Moreover, several of the methods used to estimate SARS-CoV-2 neutralizing antibodies in DBS eluates are adaptable to high-throughput platforms.

Keywords: COVID-19; antibody; human; neutralizing; serology.

Fig 1

Generation of contrived DBS (cDBS) eluates. (A) Schematic showing the generation of contrived DBS (cDBS), DBS punches, and eluate collection.

Fig 2

Correlation between PRNT and MIA with cDBS eluates. (Top) Workflow for the 8-plex MIA. (Bottom) Subset of Panel D cDBS eluates (n = 15) underwent testing via 8-plex MIA and were compared to paired serum samples tested by PRNT. Pearson correlation coefficient of MFI values and SARS-CoV-2 neutralizing activity (%) for the following antigens: (A) FLT; (B) FLS; (C) S1; (D) RBD; and (F) N. The Pearson r values and P-values are shown as insets within each panel.

Fig 3

Compatibility of cDBS eluates with the RVP neutralizing assay. (Top) Workflow for the RVP neutralizing assay. (Bottom) (A) Neutralizing titers associated with CC 12.3 with IC₅₀ shown as the text. (B) Correlation between PRNT (% neutralization) and RVP (% neutralization) with matched Panel D serum samples. (C) Correlation between PRNT (% neutralization) generated with Panel D serum versus RVP generated (% neutralization) from matched cDBS eluates. (D) Comparison of RVP neutralization of Panel D (n = 15) and Panel E (n = 7) cDBS eluates during serial dilutions. Asterisks indicate a significant difference between groups by two-way ANOVA, where *P = 0.02, **P = 0.009, ***P = 0.0002, and ****P < 0.0001. For (B), Pearson r and P-value are shown as a panel inset. For (C), Spearman r_s and P-value are shown as a panel inset. DBS eluates were generated by incubating the punches in DMEM-HEPES at room temperature for 2 hours with agitation.

Fig 4

Correlation between Luminex-based ACE2 inhibition and SARS-CoV-2 neutralization activity. (Top) Workflow for the ACE2 inhibition (iACE2) assay. (A) Examination of the iACE2 activity in Panel D cDBS eluates (n = 15) versus Panel E eluates (n = 7). In (A), asterisks indicate a significant difference between groups by Welch’s t-test, where ***P = 0.0006. (B) Correlation of iACE2 values from Panel D cDBS eluates, as shown in (A), to SARS-CoV-2 neutralizing activity (%) from paired serum samples analyzed via PRNT. cDBS eluates were evaluated in triplicate; shown are average values. The Spearman r_s and P-value are shown as a panel inset for (B). The dotted line on (B) represents the average iACE2 activity for the SARS-CoV-2 negative cohort (Panel E) and indicates the threshold for positive responses.

Fig 5

NAB-Sure exhibits high concordance with PRNT. (Top) Workflow for the NAB-Sure assay. (A) Spearman’s correlation of NAB-Sure NT₅₀ (log₁₀) from authentic DBS sample eluates (n = 11) compared to PRNT paired serum. DBS samples were evaluated in duplicate; shown are average values. (B) Spearman correlation comparing NAB-Sure NT₅₀ (log₁₀) for cDBS eluates and paired serum NT₅₀ (log₁₀) from PRNT for 14 Panel D samples performed in triplicate; shown are average values. For (A) and (B), Spearman’s r_s and P values are shown as panel insets.

Fig 6

SARS-CoV-2 neutralizing titers in cDBS eluates from vaccinated individuals. (A) cDBS eluates from Panels H, D, and E were rank ordered from high to low (left to right) based on the RBD_WT MFI determined in the 8-plex MIA. (B) Comparison of RVP neutralization % of Panels H, D, and E during serial dilutions. (C) Comparison of the iACE2 activity in Panels H, D, and E. (D) Comparison of NT₅₀ (log₁₀) from NAB-Sure between Panel H and Panel D. Two of the highest Panel H eluates were diluted 1:2 prior to the NAB-Sure assay to achieve NT₅₀ (log₁₀) values. For (B), asterisks indicate a significant difference between groups by two-way ANOVA, where *P = 0.02, **P = 0.009, ***P = 0.0002, and ****P < 0.0001. For (C) and (D), asterisks indicate a significant difference between groups using Welch’s t-test, where ****P < 0.0001.

Fig 7

SARS-CoV-2 neutralizing titers in cDBS eluates from venous whole-blood cDBS. (A) Spearman’s correlation of 8-plex MIA RBD_WT from venous blood DBS sample eluates (n = 86) compared to PRNT NT₅₀ paired serum. (B) Spearman’s correlation of NAB-Sure NT₅₀(log₁₀) from venous blood DBS sample eluates (n = 83) compared to PRNT NT₅₀ (log₁₀) paired serum.