Decrease in serum antibodies to SARS-CoV-2 in convalescent plasma donors over time.

To the Editor

Convalescent plasma, collected from donors who have previously recovered from viral illnesses, has been used to treat patients with a variety of emerging infectious diseases. This product has become more prominent during the current COVID‐19 pandemic, including with the recent emergency use authorization from the Food and Drug Administration, but the longevity of SARS‐CoV‐2 antibodies in donated plasma is unknown. Historically, convalescent plasma has been used to reduce mortality in a number of illnesses, such as SARS, H5N1, H1N1, Ebola, and others.1, 2 A limited number of prospective trials have been published or are pre‐publication regarding treatment of COVID‐19 with convalescent plasma, many of which are small but report conflicting data spanning from no benefit to statistical significance in specific patient groups but not overall, and the uncertainty of effectiveness is reflected in a large review as well.3, 4, 5 Post‐infection antibody titers vary and may remain constant or wax and wane over time independent of sex and age.6, 7 Our institution collected convalescent plasma throughout the current COVID‐19 pandemic, and recorded data points including the Signal‐to‐Cutoff ratio (S/C ratio) of antibodies to SARS‐CoV‐2 in these units. This letter aims to describe trends in the S/C ratio observed in donor units over the past 4 months to ultimately improve donor collection practices and yield units with sufficient antibody to qualify as convalescent plasma.

Plasma donors self‐referred to our donor center by completing an online form, advertised on the blood center website, and were eligible to donate 14 days after resolution of symptoms. This form asked for demographic data points and for information regarding their COVID‐19 disease history, and donors were collected if they had laboratory evidence of positive PCR for SARS‐CoV‐2 virus or antibodies against the virus, regardless of disease severity or presence of symptoms. After passing a routine donation screen, donor plasma was collected by apheresis and subsequently tested for antibodies via the EuroImmun SARS‐CoV‐2 IgG ELISA (Lubeck, Germany) for semiquantitative analysis of the presence of antibodies against the S1 domain of the spike protein of SARS‐CoV‐2. Donors were allowed to donate plasma up to once a week for 4 weeks, then asked to space further donations to every 4 weeks thereafter. A 1:320 dilution was performed at each donation and this reactivity was used to distinguish suitable convalescent plasma units, as used in one clinical trial of treatment with convalescent plasma as compared to standard plasma. Donor plasma was not used as convalescent plasma if it was not positive at the 1:320 dilution, and those donors were asked not to donate further convalescent plasma.

This institution overall had 262 individual donors, of which 38 qualified for analysis. Donors were included in this data set if they successfully completed at least three convalescent plasma donations over a minimum of 30 days. This was set to allow for enough data points over time to establish a trend in S/C ratios. Donors included had an age range of 19‐75 years at time of first donation, with an average age of 48.2 years (median: 51 years). The last day of symptoms is marked as day 0, which was self‐reported from donors on their recruitment form. While the donor forms did not allow for assessment of severity of symptoms, all donors did have some degree of symptoms and none reported an asymptomatic disease. Over the course of this data set, two donors became ineligible to donate due to low S/C ratio values, at 96 and 133 days. These donors were 41 and 48 years old, respectively. All donor plasma showed a decrease in the antibody S/C ratio from day 1 to their most recent donation, which varies from 64 to 139 days (Figure 1). Collected products overall showed a decrease in S/C ratio in all time periods and more significant decreases over time. Specifically, for products collected on days 29 to 56 days, 57% of products showed a decrease in S/C ratio from that donorʼs baseline, as compared to 78% of products with a decrease from 57 to 84 days and 100% of products with a decrease from 85 to 112 days and 113 to 140 days (Figure 2). The average S/C ratio decreased over time for each additional month after symptoms resolved. Donors had an average S/C ratio of 10.3 from days 1 to 28, and this decreased to 8.1 for days 29 to 56, 6.5 for days 57 to 84, 5.1 for days 85 to 112, and 4 for days 113 to 140. In the group that donated at 4 months or later post‐recovery (113 days‐140 days), these donors had an average decrease in S/C ratio from baseline of 47% (mean 53.6%).

FIGURE 1

Change in S/C ratio over time for each individual donor. Each data point represents the S/C ratio calculated from an individual convalescent plasma donation, and each color represents an individual donor followed longitudinally. Each timepoint was calculated using the day of donation and the donor's self‐reported last day of symptoms [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 2

Percent change in S/C ratio as grouped by uniform time periods since resolution of symptoms. Each 28‐day period includes all convalescent plasma units donated that fall within a specific time interval. The percent change was calculated from the donor's baseline S/C ratio value as defined as the first unit of convalescent plasma donated. The days included in each time group were calculated from the last day of donor symptoms [Color figure can be viewed at wileyonlinelibrary.com]

Overall, this data set suggests that antibody S/C ratios decrease over time, and most pronounced at 12 weeks or later using the EuroImmun assay. This assay has been shown to be comparable in sensitivity and specificity to other major antibody detection assays, including the Ortho‐Clinical Diagnostics assay. Many donors also showed stable or increasing levels during the first eight to 11 weeks after donation. Based on our data, donor centers may yield more convalescent plasma units if each collection is optimized to maximize volume collected and donors are collected more frequently closer to their disease recovery, ideally within 4 to 8 weeks, as opposed to many collections over a long period of time. This would facilitate increased antibody titers in collected plasma and reduce collections of plasma that is unable to be used as convalescent plasma. We believe our findings will aid other collection centers in the development of convalescent plasma collection algorithms to optimize SARS‐CoV‐2 antibody titers.

CONFLICT OF INTEREST

The authors disclose no conflicts of interest.