Previous SARS-CoV-2 infection or a third dose of vaccine elicited cross-variant neutralising antibodies in vaccinated solid-organ transplant recipients

Abstract

Objectives: The SARS-CoV-2 pandemic poses a great threat to global health, particularly in solid organ transplant recipients (SOTRs). A 3-dose mRNA vaccination protocol has been implemented for the majority of SOTRs, yet their immune responses are less effective compared to healthy controls (HCs).

Methods: We analyzed the humoral immune responses against the vaccine strain and variants of concern (VOC), including the highly mutated-omicron variant in 113 SOTRs, of whom 44 had recovered from COVID-19 (recovered-SOTRs) and 69 had not contracted the virus (COVID-naïve). In addition, 30 HCs, 8 of whom had recovered from COVID-19, were also studied.

Results: Here, we report that three doses of the mRNA vaccine had only a modest effect in eliciting anti-viral antibodies against all viral strains in the fully vaccinated COVID-naive SOTRs (n = 47). Only 34.0% of this group of patients demonstrated both detectable anti-RBD IgG with neutralization activities against alpha, beta, and delta variants, and only 8.5% of them showed additional omicron neutralizing capacities. In contrast, 79.5% of the recovered-SOTRs who received two doses of vaccine demonstrated both higher anti-RBD IgG levels and neutralizing activities against all VOC, including omicron.

Conclusion: These findings illustrate a significant impact of previous infection on the development of anti-SARS-CoV-2 immune responses in vaccinated SOTRs and highlight the need for alternative strategies to protect a subset of a lesser-vaccine responsive population.

Keywords: SARS‐CoV‐2; neutralising antibody; solid‐organ transplant recipient; vaccine; variant of concern.

Figure 1

Vaccination‐induced anti‐RBD IgG antibodies in SOTRs and HCs. Sera from recovered‐SOTR (n = 44) who received two doses of vaccines and sera from COVID‐naïve (n = 69) who received 2 (n = 39) or three doses (n = 47) of vaccines, along with samples from recovered‐HCs (n = 8) and COVID‐naive HCs (n = 22) were tested for IgG antibodies against SARS‐CoV‐2 antigens (RBD, S1 and nucleocapsid) by multiplexed magnetic bead‐based assay. Levels of antinucleocapsid IgG antibodies were used for confirmation of COVID infection. Positivity (> 700 MFI) of assay, pre‐et by manufacturer, is denoted as a horizonal line. **** was referred to P < 0.0001.

Figure 2

Neutralising capacities of sera from vaccinated SOTRs or HCs against the vaccine and alpha, beta and delta variants. (a) Multiplexed neutralisation assays were used to determine the degree of inhibition (%) by sera on binding of indicated viral S1 proteins to ACE2 receptor. A horizonal line (8% inhibition at 1:200‐fold dilution) representing the positive neutralising cut‐off was derived from the results of Supplementary figure 2. The qualitative analysis (% inhibition) is defined as 100 × (1 − sample value /negative control value). (b) Linear regression analyses were carried out to determine relationships between the levels of anti‐RBD IgG antibodies (X axis) and % ACE2 inhibition (Y axis). In the recovered‐SOTRs series, levels of anti‐RBD IgG were found strongly correlated with the degree (%) of inhibition on binding of vaccine strain (R ² = 0.60; P < 0.0001), alpha (R ² = 0.59; P < 0.0001), beta (R ² = 0.51; P < 0.0001) and delta (R ² = 0.48; P < 0.0001) to ACE2. In the COVID‐naïve SOTRs series, the correlations between these two events were even stronger, with vaccine strain (R ² = 0.80; P < 0.0001), alpha (R ² = 0.79; P < 0.0001), beta (R ² = 0.75; P < 0.0001) and delta (R ² = 0.69; P < 0.0001), respectively. **** was referred to P < 0.0001.

Figure 3

Neutralising capacities of sera from vaccinated SOTRs and HCs against omicron. (a) Omicron‐neutralisation of COVID‐naïve SOTRs (median IC₅₀ = 0.3) was significantly lower than that of recovered‐SOTRs (median IC₅₀ = 193, P = 0.0027) or COVID‐naïve HCs (median IC₅₀ = 414, P < 0.0001). (b) Correlation tests between omicron‐neutralisation with vaccine‐ and delta‐neutralisation. In the recovered‐SOTRs series, omicron‐neutralisation strongly correlated with both vaccine strain‐ (Spearman r = 0.870, P < 0.0001) and delta strain‐neutralisation (r = 0.869, P < 0.0001). In the COVID‐naïve SOTRs series, omicron‐neutralisation was modestly correlated with both vaccine strain‐ neutralisation (Spearman r = 0.326, P = 0.104) and delta strain‐neutralisation (Spearman r = 0.376, P = 0.058). ** was referred to P = 0.0027 and **** was referred to P < 0.0001.