Development and Implementation of Dried Blood Spot-Based COVID-19 Serological Assays for Epidemiologic Studies

Abstract

Serological surveillance studies of infectious diseases provide population-level estimates of infection and antibody prevalence, generating crucial insight into population-level immunity, risk factors leading to infection, and effectiveness of public health measures. These studies traditionally rely on detection of pathogen-specific antibodies in samples derived from venipuncture, an expensive and logistically challenging aspect of serological surveillance. During the COVID-19 pandemic, guidelines implemented to prevent the spread of SARS-CoV-2 infection made collection of venous blood logistically difficult at a time when SARS-CoV-2 serosurveillance was urgently needed. Dried blood spots (DBS) have generated interest as an alternative to venous blood for SARS-CoV-2 serological applications due to their stability, low cost, and ease of collection; DBS samples can be self-generated via fingerprick by community members and mailed at ambient temperatures. Here, we detail the development of four DBS-based SARS-CoV-2 serological methods and demonstrate their implementation in a large serological survey of community members from 12 cities in the East Bay region of the San Francisco metropolitan area using at-home DBS collection. We find that DBS perform similarly to plasma/serum in enzyme-linked immunosorbent assays and commercial SARS-CoV-2 serological assays. In addition, we show that DBS samples can reliably detect antibody responses months postinfection and track antibody kinetics after vaccination. Implementation of DBS enabled collection of valuable serological data from our study population to investigate changes in seroprevalence over an 8-month period. Our work makes a strong argument for the implementation of DBS in serological studies, not just for SARS-CoV-2, but any situation where phlebotomy is inaccessible. IMPORTANCE Estimation of community-level antibody responses to SARS-CoV-2 from infection or vaccination is critical to inform public health responses. Traditional studies of antibodies rely on collection of blood via venipuncture, an invasive procedure not amenable to pandemic-related social-distancing measures. Dried blood spots (DBS) are an alternative to venipuncture, since they can be self-collected by study participants at home and do not require refrigeration for shipment or storage. However, DBS-based assays to measure antibody levels to SARS-CoV-2 have not been widely utilized. Here, we show that DBS are comparable to blood as a sampling method for antibody responses to SARS-CoV-2 infection and vaccination over time measured using four distinct serological assays. The DBS format enabled antibody surveillance in a longitudinal cohort where study participants self-collected samples, ensuring the participants' safety during an ongoing pandemic. Our work demonstrates that DBS are an excellent sampling method for measuring antibody responses whenever venipuncture is impractical.

Keywords: COVID-19; SARS-CoV-2; antibodies; dried blood spot; serology; seroprevalence.

FIG 1

Validation of DBS in anti-S serological assays. Paired DBS and plasma samples (n = 39) from previously SARS-CoV-2-infected individuals were compared in (A) the anti-S IgG ELISA and (C) the Ortho COV2T assay. DBS samples (n = 100) from individuals without previous SARS-CoV-2 infection were analyzed by (B) the anti-S IgG ELISA and (D) the Ortho COV2T assay. Cutoffs for the assays are denoted by dashed lines. Linear regression comparing IgG levels between sample types (A and C) is depicted by a solid line with 95% confidence intervals (CI).

FIG 2

Validation of DBS in anti-N serological methods. Paired DBS and plasma samples (n = 39) from previously SARS-CoV-2-infected individuals were compared in (A) the anti-N IgG ELISA. DBS samples (n = 100) from individuals without previous SARS-CoV-2 infection were analyzed by (B) the anti-N IgG ELISA. Validation of DBS on the Roche N assay was performed on DBS samples derived from study participants in Round 2 of the EBCOVID study. (C) Samples considered SARS-CoV-2-seropositive (n = 33) from the anti-S IgG DBS ELISA were analyzed by the Roche N assay. (D) Samples considered SARS-CoV-2-seronegative (n = 99) by the Ortho COV2T assay were analyzed by the Roche N assay. Cutoffs for the assays are denoted by dashed lines. Linear regression comparing IgG levels between sample types is depicted by a solid line with 95% CI.

FIG 3

Durable SARS-CoV-2 antibody responses to S can be detected by DBS. Paired plasma (solid lines) and DBS samples (dashed lines) from 10 COVID-19 convalescent plasma donors (L1–L10) sampled longitudinally between 0 and 246 days from their first donation were analyzed by (A) the anti-S IgG ELISA and (B) the Ortho CoV2T assay. In 3A, the positivity cutoff for plasma is OD450 ≥ 0.095 (solid gray line) and DBS is OD450 ≥ 0.32 (dotted, shaded line). In 3B, the positivity for both plasma and DBS is S/Co ≥1 (dotted, shaded line).

FIG 4

SARS-CoV-2 antibody responses to N wane over time and can be detected by DBS. Paired plasma (solid lines) and DBS samples (dashed lines) from 10 COVID-19 convalescent plasma donors (L1–L10) sampled longitudinally between 0 and 246 days from their first donation were analyzed by (A) the anti-N-IgG ELISA and (B) the Roche N assay. In 4A, the positivity cutoff for both plasma and DBS is OD450 ≥ 0.32 (dotted, shaded line). In 4B, the positivity cutoff for plasma is S/Co ≥ 1 (solid gray line) and DBS is S/Co ≥ 0.045 (dotted, shaded line).

FIG 5

Vaccine-elicited SARS-CoV-2 antibody kinetics can be detected by DBS. (A–B) Plasma and DBS generated from 12 SARS-CoV-2 S-vaccinated individuals sampled before their first dose, after their first dose, and after their second dose were analyzed using the Ortho COV2T assay. DBS from these same individuals were analyzed by the anti-S IgG DBS ELISA as (C) OD₄₅₀ values or (D) endpoint titers. DBS from 4 other vaccinated individuals sampled weekly after their first, second, and third doses were analyzed by the anti-S IgG DBS ELISA as (E) OD₄₅₀ value or (F) endpoint titer. Solid line in (F) represents positivity cutoff. Dashed lines in (E–F) denote days when additional doses of vaccine were administered.

FIG 6

Schematic of the East Bay COVID (EBCOVID) study. There were 2 phases to the study: (A) the screening phase, where participants were recruited and screened for eligibility for inclusion into our study, and (B–D) the sampling phase, where study participants were invited to provide biospecimens, including DBS, in 3 separate rounds.

FIG 7

Testing algorithm and results from EBCOVID Round 2. (A) Schematic of the testing algorithm used for Round 2 of the EBCOVID study. (B) Round 2 EBCOVID results comparing DBS reflexed according to the testing algorithm on the anti-S IgG ELISA and the Roche N assays. IS, insufficient sample.

FIG 8

Testing algorithm and results from EBCOVID Round 3. (A) Schematic of the testing algorithm used for Round 3 of the EBCOVID study. (B) Round 3 EBCOVID results comparing DBS reflexed according to the testing algorithm on the Ortho CoV2T (Ortho S), the anti-S-IgG ELISA, and the Roche N assays. IS, insufficient sample.