Third Early "Booster" Dose Strategy in France of bnt162b2 SARS-CoV-2 Vaccine in Allogeneic Hematopoietic Stem Cell Transplant Recipients Enhances Neutralizing Antibody Responses

Abstract

Immunocompromised individuals generally fail to mount efficacious immune humoral responses following vaccination. The emergence of SARS-CoV-2 variants of concern has raised the question as to whether levels of anti-spike protein antibodies achieved after two or three doses of the vaccine efficiently protect against breakthrough infection in the context of immune suppression. We used a fluorescence-based neutralization assay to test the sensitivity of SARS-CoV-2 variants (ancestral variant, Beta, Delta, and Omicron BA.1) to the neutralizing response induced by vaccination in highly immunosuppressed allogeneic HSCT recipients, tested after two and three doses of the BNT162b2 vaccine. We show that neutralizing antibody responses to the Beta and Delta variants in most immunocompromised HSCT recipients increased after three vaccine doses up to values similar to those observed in twice-vaccinated healthy adults and were significantly lower against Omicron BA.1. Overall, neutralization titers correlated with the amount of anti-S-RBD antibodies measured by means of enzyme immunoassay, indicating that commercially available assays can be used to quantify the anti-S-RBD antibody response as a reliable surrogate marker of humoral immune protection in both immunocompetent and immunocompromised individuals. Our findings support the recommendation of additional early vaccine doses as a booster of humoral neutralizing activity against emerging variants, in HSCT immunocompromised patients. In the context of Omicron circulation, it further emphasizes the need for reinforcement of preventive measures including the administration of monoclonal antibodies in this high-risk population.

Keywords: COVID-19 vaccine; HSCT immunocompromised patients; SARS-CoV-2; neutralizing antibody; variant of concern.

Figure 1

Immunofluorescence labeling of SARS-CoV-2 variant delta infection in Calu-3 cells and neutralization by patient serum dilutions. Representative examples of individuals from the three groups are shown, including ImmunoSupp-V (double- then triple-vaccinated allogeneic HSCT recipients), ImmunoComp-V (double-vaccinated immunocompetent healthcare workers), and ImmunoComp-C (immunocompetent non-vaccinated convalescent individuals).

Figure 2

Antibody-mediated neutralization effectiveness of sera from the three groups tested against SARS-CoV-2 variants D614G, Beta, Delta, and Omicron (BA.1). NT₅₀ represents the serum dilution resulting in 50% virus neutralization. Neutralization assay was performed using serum samples obtained from double- or triple-BNT162b2 vaccinated immunocompromised allogeneic HSCT recipients (ImmunoSupp-V), non-vaccinated immunocompetent convalescent individuals (ImmunoComp-C), and double-BNT162b2-vaccinated healthy immunocompetent individuals (ImmunoComp-V). Negative titers were handled as 0.5. Statistical significance was calculated by two-tailed, paired Student’s t tests. Asterisks indicate p-values as **: p < 0.01, and ***: p < 0.001. NS: Not significantly different.

Figure 3

Changes in serum-neutralizing effectiveness against variants D614G, beta, delta, and Omicron (BA.1) after a third dose of BNT162b vaccine compared to post-second vaccine dose in allogeneic HSCT recipients from the ImmunoSupp-V group. The experiments were performed in triplicate. Asterisks indicate p-values as **: p < 0.01 and ***: p < 0.001.

Figure 4

Relationship between serum neutralization titers in cell culture and the amounts of anti-S-RBD antibodies detected in the same sera from individuals in the ImmunoSupp-V (double- or triple-BNT162b2 vaccinated immunocompromised allogeneic HSCT recipients) and ImmunoComp-V (double-BNT162b2 vaccinated healthy immunocompetent individuals) groups. Anti-S-RBD titers were plotted against the corresponding NT₅₀ for variants D614G, Beta, Delta, and Omicron BA.1. Correlations between NT₅₀ and anti-S-RBD titers were calculated using Spearman’s correlation, and p-values are indicated for each graph.