Diminished Neutralization Capacity of SARS-CoV-2 Omicron BA.1 in Donor Plasma Collected from January to March 2021

ABSTRACT The 50% plaque reduction neutralization assay (PRNT50) has been previously used to assess the neutralization capacity of donor plasma against wild-type and variant of concern (VOC) severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Emerging data suggest that plasma with an anti-SARS-CoV-2 level of ≥2 × 104 binding antibody units/mL (BAU/mL) protects against SARS-CoV-2 Omicron BA.1 infection. Specimens were collected using a cross-sectional random sampling approach. For PRNT50 studies, 63 previously analyzed specimens by PRNT50 versus SARS-CoV-2 wild-type, Alpha, Beta, Gamma, and Delta were analyzed by PRNT50 versus Omicron BA.1. The 63 specimens plus 4,390 specimens (randomly sampled regardless of serological evidence of infection) were also tested using the Abbott SARS-CoV-2 IgG II Quant assay (anti-spike [S]; Abbott, Chicago, IL, USA; Abbott Quant assay). In the vaccinated group, the percentages of specimens with any measurable PRNT50 versus wild-type or VOC were wild type (21/25 [84%]), Alpha (19/25 [76%]), Beta (18/25 [72%]), Gamma (13/25 [52%]), Delta (19/25 [76%]), and Omicron BA.1 (9/25 [36%]). In the unvaccinated group, the percentages of specimens with any measurable PRNT50 versus wild type or VOC were wild-type SARS-CoV-2 (16/39 [41%]), Alpha (16/39 [41%]), Beta (10/39 [26%]), Gamma (9/39 [23%]), Delta (16/39 [41%]), and Omicron BA.1 (0/39) (Fisher's exact tests, vaccinated versus unvaccinated for each variant, P < 0.05). None of the 4,453 specimens tested by the Abbott Quant assay had a binding capacity of ≥2 × 104 BAU/mL. Vaccinated donors were more likely than unvaccinated donors to neutralize Omicron when assessed by a PRNT50 assay. IMPORTANCE SARS-CoV-2 Omicron emergence occurred in Canada during the period from November 2021 to January 2022. This study assessed the ability of donor plasma collected earlier (January to March 2021) to generate any neutralizing capacity against Omicron BA.1 SARS-CoV-2. Vaccinated individuals, regardless of infection status, were more likely to neutralize Omicron BA.1 than unvaccinated individuals. This study then used a semiquantitative binding antibody assay to screen a larger number of specimens (4,453) for individual specimens that might have high-titer neutralizing capacity against Omicron BA.1. None of the 4,453 specimens tested by the semiquantitative SARS-CoV-2 assay had a binding capacity suggestive of a high-titer neutralizing capacity against Omicron BA.1. These data do not imply that Canadians lacked immunity to Omicron BA.1 during the study period. Immunity to SARS-CoV-2 is complex, and there is still no wide consensus on correlation of protection to SARS-CoV-2.

T he use of convalescent plasma to treat patients infected with emerging respiratory viruses, including severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1) and avian influenza, has been a topic of study for decades (1)(2)(3). Since the start of the SARS-CoV-2 pandemic, convalescent plasma was identified as a potential therapeutic candidate for clinical trials (4,5). Those clinical trials identified mixed efficacy of convalescent plasma and the potential for early use of high-titer convalescent plasma in immunocompromised patients infected with SARS-CoV-2 (6)(7)(8)(9)(10). A revision of the U.S. Food and Drug Administration (FDA) emergency use authorization (EUA) for the use of COVID-19 convalescent plasma identified immunocompromised individuals as clinical trial candidates for high-titer COVID-19 convalescent plasma (11). Work on convalescent plasma has also led to further studies on protective immunity to SARS-CoV-2 (12)(13)(14)(15) and has informed our incomplete understanding of the correlation of protection against SARS-CoV-2 (16,17). Earlier convalescent plasma qualification approaches relied on low-throughput culture-based 50% plaque reduction neutralization (PRNT 50 ) assays (18). Other, more rapid and easier-to-utilize approaches, including virus-like particle (VLP), competition assays, and enzyme-linked immunosorbent assays, were also used to identify high-titer plasma and study immune responses in individuals previously infected with SARS-CoV-2 (8,12,15,19).
The Abbott SARS-CoV-2 IgG II Quant assay (Abbott anti-spike [S]; Abbott, Chicago, IL, USA; here referred to as the Abbott Quant assay) is a high-throughput assay that is simpler to operationalize than PRNT 50 . This assay generates semiquantitative results which can be converted into binding antibody units [BAU] per milliliter (20). A prior study noted that a cutoff of 7.1 Â 10 3 BAU/mL might be used to screen for neutralizing high-titer plasma against wild-type, Alpha, Beta, Gamma, and Delta SARS-CoV-2 (12). High-throughput semiquantitative technologies enable researchers to screen large numbers of plasma donations for unique specimens that might contain high-titer anti-SARS-CoV-2 neutralizing plasma against wild-type and variant of concern (VOC) SARS-CoV-2 (21).
Assessments of neutralizing capacity of plasma or serum may be impacted by local and temporal factors. Prior to the emergence of Omicron, less than 10% of Canadians were estimated to have been naturally infected with SARS-CoV-2 (31,32). Until January to March 2021, most infections in Canada were likely due to wild-type or Alpha SARS-CoV-2 (33). Vaccination campaigns were initiated in December 2020, with 96% of all Canadian blood donors showing evidence of measurable antibodies to anti-spike (S) by August 2021 (34). Canadian Blood Services was able to determine donor vaccination status most effectively for the time from January to March 2021 (12,13).
This study used PRNT 50 to determine the neutralizing capacity of vaccinated and unvaccinated donor plasma collected from January to March 2021against Omicron BA.1. This study also used the Abbott Quant assay to screen a larger number of donor plasma specimens collected from this time period for individual specimens potentially containing high-titer neutralizing capacity against Omicron BA.1.

RESULTS
All specimens were from donors with an anti-S or an anti-RBD serological signal. Study specimens were subsamples of a larger repeated cross-sectional design with random cross-sectional sampling. Previously, 65 specimens were analyzed by PRNT 50 (wildtype, Alpha, Beta, Gamma, and Delta SARS-CoV-2) (13) as well as the Abbott Quant assay (12). All specimens previously tested by PRNT 50 had evidence of an anti-S or anti-receptor binding domain (RBD) signal (with or without anti-N) (12). Sixty-three specimens had sufficient sample volume to be tested by PRNT 50 for Omicron SARS-CoV-2. For the 63 specimens, anti-N profiles, Abbott Quant assay results, donor vaccination histories, and PRNT 50 (wild type, Alpha, Beta, Gamma, Delta, and Omicron SARS-CoV-2) are presented in Tables 1 to 4.
Neutralization of wild-type and VOC SARS-CoV-2 in vaccinated versus unvaccinated donors. Since different cell lines were used to understand the neutralizing capacity of donor plasma against Omicron SARS-CoV-2, median PRNT 50 results were not compared directly. Instead, the numbers of specimens producing any neutralizing antibodies (e.g., $20) were compared within vaccinated and unvaccinated groups.
Small numbers of specimens for individuals with a vaccine history and an anti-N signal (possible evidence of a past SARS-CoV-2 infection) led to the combination of data from donors vaccinated with an anti-N signal ( Table 1) and donors vaccinated without an anti-N signal ( Table 2). Data from unvaccinated donors with an anti-N signal (Table 3) and unvaccinated donors without an anti-N signal (Table 4) were also combined.
When neutralizing capacity was measured by PRNT 50 , plasma from vaccinated donors was more likely than plasma from unvaccinated donors to neutralize VOCs (including Omicron BA.1) and wild type. For wild-type neutralization, the proportions were vaccinated ( Assessment of residual specimens from January, February, and March 2021 using the Abbott Quant assay. In addition to the 63 specimens tested by PRNT 50 for Omicron and the Abbott Quant Assay, 4,390 randomly sampled specimens were tested by the Abbott Quant assay (n = 4,453). The monthly distribution of these 4,453 specimens collected in 2021 was 1,499 in January, 1,465 in February, and 1,489 in March. The BAU per milliliter values of these 4,453 specimens are presented in Fig. 1. None of the BAU per milliliter values reached a level of 2 Â 10 4 BAU/mL.

DISCUSSION
For the period from January to March 2021, plasma collected from vaccinated Canadian blood donors was more likely to have measurable neutralizing antibodies a Median Abbott Quant assay values for this group were 3 Â 10 3 BAU/mL (25th percentile to 75th percentile, 1 Â 10 2 to 6 Â 10 3 BAU/mL). NA, not available.

SARS-CoV-2 Neutralization Screening Microbiology Spectrum
(measured by PRNT 50 against wild type, Alpha, Beta, Gamma, Delta, and Omicron BA.1) than plasma from unvaccinated blood donors. In the unvaccinated group, none of the plasma specimens had measurable PRNT 50 titers versus Omicron BA.1. As previously noted, specimens were collected when seroprevalence to SARS-CoV-2 was ,10% and when most Canadians with a history of SARS-CoV-2 infection would have been infected with wild-type or Alpha SARS-CoV-2 (32,33). Only a minority (8%) of vaccinated donors in this study claimed to be fully vaccinated (12,32), and only 2% of Canadians had received two doses of a SARS-CoV-2 vaccine (35). Wastewater studies    a Median Abbott Quant assay values for this group were 0 BAU/mL (25th percentile to 75th percentile, 0 to 1 Â 10 1 ). patients hospitalized from January to April 2020 with no vaccination history (n = 24) or 1 dose of vaccine (n = 20) also exhibited reduced neutralization against BA.1 compared to wild type using a pseudovirus assay (43). A small number of specimens collected from Austrian patients with ancestral infection (March and April 2020 [n = 10]) had reduced neutralization of BA.1, using a focus-forming neutralization assay (44). None of the specimens screened with the Abbott Quant assay had a value of $2 Â 10 4 BAU/mL, which has been previously associated with high-titer plasma against Omicron BA.1 (28). This is not unexpected, as convalescent plasma collected during earlier waves of the pandemic may have reduced efficacy against Omicron subvariants as they arise (45). However, this finding does not imply that the donors tested lacked protection against SARS-CoV-2 disease and death. Immunity to SARS-CoV-2 is complex and involves neutralizing antibodies, binding antibodies, antibody-dependent cellular cytotoxicity (46), complex mechanisms of cell-mediated immunity (47), and elements of innate immunity (48). Due to this complexity, there is still no wide consensus on correlations of protection to SARS-CoV-2 (16,17). Apart from a potential role as a cutoff for high-titer convalescent plasma by convalescent plasma trials (28,29), there is also no international consensus on the protective utility of the binding antibody value of $2 Â 10 4 BAU/mL (30,49).
A full year of the pandemic would need to pass before the Canadian population developed high BAU per milliliter values. A larger Canadian seroprevalence study (10,000 to 40,000 specimens/month) first identified median BAU/mL levels of $2 Â 10 4 BAU/mL in February of 2022 after the emergence of Omicron. However, the low frequency of anti-N and high frequency of anti-S in the population suggests that high BAU per milliliter values were being driven by COVID-19 vaccination programs rather than natural infection (50). This study does not discriminate between the impacts of boosters or new bivalent vaccines. However, it is important to note the benefit of SARS-CoV-2 vaccines in reducing disease burden and death in the Canadian population, even in an environment dominated by Omicron (51)(52)(53). The rollout of SARS-CoV-2 vaccines in Canada can be seen as a success story, with 85% of Canadians receiving at least one dose and 82% receiving a primary series by 11 September 2022. However, some Canadians expressed antivaccine sentiments, lacking understanding of vaccines and herd immunity (54), and vaccine-hesitant individuals often expressed a preference for natural immunity (55).
This study has several additional caveats. Different cell culture conditions were used for wild type, Alpha, Beta, Gamma, and Delta than for Omicron. To account for this, the study focused on identifying the presence or absence of any neutralizing antibody capacity against SARS-CoV-2 VOCs. This study included a small number of specimens for the time from January 2021 to March 2021 used for PRNT 50 (13). Due to the time taken to develop Omicron BA.1 PRNT 50 assays, this study did not assess donor plasma for neutralization against later sublineages of BA.1, BA.2, BA.3, BA.4, BA.5, or recombinants that have circulated in Canada (56). It is also important to acknowledge that donor-declared histories of vaccination may be confounded by recall bias and may be incomplete (57). The collection of vaccination histories, as approved in the study ethics proposal, was also limited to the specimens used for PRNT 50 and not linked to data broadly tested with the Abbott Quant assay.
Although this work relies on specimens collected early in the pandemic, it does have applicability to understanding humoral immunity in individuals who are partially vaccine hesitant (receiving less than a full series of wild-type SARS-CoV-2 vaccine) or completely vaccine hesitant (relying on immunity from an earlier infection with wildtype or Alpha SARS-CoV-2). Those individuals may have impaired humoral protection against Omicron BA.1 SARS-CoV-2 infection. Therefore, even in populations with high rates of SARS-CoV-2 infection, vaccination (including boosting with monovalent or bivalent vaccines) is an important strategy in reducing the burden of severe disease and death (58,59). This protection is broad and ensures the safety of adults and children in the population from outcomes including intensive care admission and death, even when Omicron is dominant (51).

MATERIALS AND METHODS
Ethical considerations. Institutional ethics board clearance for this project was received from the University of Alberta and the following institutions: Canadian Blood Services and Sinai Health, Toronto (Mount Sinai Hospital).
CIHR Correlates of Immunity study participants and samples. Canadian Blood Services collects retention EDTA plasma (Becton Dickson [BD], Mississauga, ON, Canada) specimens as previously described (12,13,32,60). As previously described, this was a repeated cross-sectional design with random cross-sectional sampling of all available retention samples (n = 1,500/month) for a 12-month period from January, February, and March of 2021 (total n = 4,500) (20). Samples were then anonymized, aliquoted, transported to test sites, and then stored (240 to 280°C) (12). A total of 4,453 retention specimens were available from January (n = 1,499), February (n = 1,465), and March (n = 1,489) for testing with the Abbott Quant assay.
Donor SARS-CoV-2 vaccination history and linking to specific specimens. During the donation screening process, all donors were asked if they received a SARS-CoV-2 vaccine in the past 3 months. This was standard practice at Canadian Blood Services, did not collect information on the vaccine producer, and was not linked to provincial vaccination records. Donor vaccine information focused on donors with specimens linked to PRNT 50 neutralization assays (12).
Specimens chosen for SARS-CoV-2 neutralization testing. Specimens assessed for antibody neutralizing capacity of wild type and variant (Alpha, Beta, Gamma, Delta, and Omicron BA.1) were previously selected using a published tiered testing approach (12,13).
Definitions of evidence of anti-N positivity. Serological evidence of anti-N positivity was defined as the presence of an anti-N signal by at least one of the Abbott Architect anti-N SARS-CoV-2 IgG assay or the Sinai Health anti-N assay (see previous publication [12]).
PRNT 50 assays: wild type and variants of concern. Selected EDTA plasma specimens were used in PRNT 50  . Culture conditions for wild type, Alpha, Beta, Gamma, and Delta followed the experimental conditions previously described (12,61). For Omicron PRNT 50 , experimental conditions varied only in that PRNT 50 plates were incubated for 3 days prior to fixation with crystal violet-formaldehyde solution for at least 1 h. After rinsing with distilled water (dH 2 O), plates were air-dried, and plaques were counted on a lightbox (for the detailed PRNT 50 procedure, please see Valcourt et al. [61] and Lin et al. [12]). The Omicron virus stock was a clinical isolate passaged in Vero E6 and TMPRSS2 cells, and next-generation sequencing (NGS) was used to confirm the Omicron BA.1 sequence.
SARS-CoV-2 antibody testing using the Abbott Quant assay. We tested 4,453 randomly selected retention specimens by using the Abbott Quant assay (Abbott Laboratories, Chicago, IL, USA) as per the manufacturer's guidelines and as previously described (12). These specimens were not subjected to prior stratification based on anti-N, anti-RBD, or anti-S. Semiquantitative values (units per milliliter) generated by the Abbott Quant were converted to BAU per milliliter as described in a prior analysis (12,20).
Data storage and statistical analysis. A study identification number was assigned by the information technology team at Canadian Blood Services. All samples were labeled with a study identification number, and all data were stored with this number. Researchers did not have access to the donor-identifying data. Data were stored using a password-protected Microsoft Excel (Redmond, WA, USA) spreadsheet. Descriptive data (median, 25th percentile, and 75th percentile), Fisher's exact test (two-sided), odds ratios, and 95% CIs were calculated with GraphPad Prism (version 9.2.0; GraphPad Software, Inc., San Diego, CA, USA) was used to analyze data. PRNT 50 values were assessed for the presence (yes or no) of any measurable neutralizing response against wild type and variant (Alpha, Beta, Gamma, Delta, BA.1 Omicron) SARS-CoV-2.

ACKNOWLEDGMENTS
Canadian Blood Services staff and leadership were instrumental in supporting this project. The operations of the University of Alberta biosafety level 3 (BSL3) laboratory were supported by M. Desaulniers.
The Abbott Architect SARS-Cov-2 IgG assay kit costs were supported by Abbott Laboratories, Abbott Park, Illinois. Abbott analyzers used by Canadian Blood Services were supplied by the COVID-19 Immunity Task Force (CITF). Publication charges for this manuscript were funded by the CITF. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Methodology