Sensitive Detection of SARS-CoV-2–Specific Antibodies in Dried Blood Spot Samples

Dried blood spot (DBS) samples can be used for the detection of severe acute respiratory syndrome coronavirus 2 spike antibodies. DBS sampling is comparable to matched serum samples with a relative 98.1% sensitivity and 100% specificity. Thus, DBS sampling offers an alternative for population-wide serologic testing in the coronavirus pandemic.

from PCR testing. All participants were anonymized, and SARS-CoV-2 PCR status was recorded as positive or unknown.
For DBS collection, we collected capillary blood samples onto forensic-grade 226 DBS cards (Ahlstrom Munksjo, https://www.ahlstrom-munksjo.com) by using finger-prick lancets (4,8). We stored DBS cards at room temperature in individual sample bags with desiccant. Concomitantly, we collected venous blood from volunteers and separated serum by using centrifugation at 9,700 × g for 5 min at room temperature. Laboratory analysis was blinded to PCR status, and we reported SARS-CoV-2-specific antibody results as positive, negative, or equivocal.
To elute antibody from DBS cards, we isolated individual preperforated DBS spots by using a sterile pipette tip and placed them into a universal tube at a ratio of 1 spot to 250 µL 0.05% phosphatebuffered saline (PBS)-Tween 20 (PBS-T) (PBS, xoid; Tween-20; Sigma-Aldrich, https://www.sigma aldrich.com). We briefly vortexed and incubated tubes overnight at room temperature. We then harvested DBS eluate into a microtube and centrifuged it at 10,600 × g for 10 min at room temperature. We stored eluate at 4°C for <14 days in accordance with standard protocols (4). We quantified total IgG, IgA, and IgM concentrations in matched serum and DBS eluate, plus pre-August 2019 DBS samples, with nephelometry by using the automated COBAS 6000 (Roche, https://www.roche.com).
We performed statistical analyses by using Prism 8 (GraphPad, https://www.graphpad.com) and assessed correlations between continuous data by using Spearman's rank test (p<0.05 was considered statistically significant). We assessed DBS sample ELISA performance, relative to the serum assay, by calculating the comparative sensitivity, specificity, and positive and negative predictive values, with 95 % CIs.
We assessed the agreement between DBS and serum ELISA results by determining the Cohen κ coefficient and Bland-Altman mean-difference. We performed quantification of total immunoglobulin concentrations in serum and DBS eluate. We observed 7-to 11-fold reduction in mean immunoglobulin concentration (IgG, IgA, and IgM) in DBS eluate compared with matched serum (Table 1). Matched serum and DBS titration curves showed the detection of SARS-CoV-2 S glycoprotein antibodies in both serum and DBS eluate with the limits of detection and the optimal detection dilution indicated (1:10 for DBS eluate and 1:100 for serum). PCR-positive matched samples showed higher responses, whereas pre-August 2019 DBS samples were negative across all dilutions (Figure 1).

Conclusions
We show that DBS samples can be used for the detection of SARS-CoV-2-specific antibodies with results comparable to serum samples, supporting the findings of recent preliminary studies (11,12). Although individual laboratories should optimize DBS-derived antibody detection, considering dilution-factor and cutoff thresholds for their relevant downstream assay, these results demonstrate that DBS sampling could complement venipuncture for serologic assessments, such as seroprevalence studies, during the COVID-19 pandemic.
A current limitation of antibody assays is the necessity for venipuncture by skilled phlebotomists; DBS sampling overcomes this limitation and introduces the opportunity for wider population-level testing and improved surveillance in groups at heightened risk for infection. For example, DBS could be delivered using postal services (4) to patients with chronic conditions, the immunocompromised, and the elderly, all of which are groups disproportionately affected by CO-VID-19 (13). Furthermore, the DBS method is simple and inexpensive (4), which could enhance sampling in low-and middle-income countries, among groups where venipuncture is culturally unacceptable or in a geographically dispersed population.