Reduced durability of hybrid immunity to SARS-CoV-2 in immunocompromised children

Abstract

Background: In endemic COVID-19, immunocompromised children are vulnerable until vaccinated but the optimal primary vaccination regime and need for booster doses remains uncertain.

Methods: We recruited 19 immunocompromised children (post-solid organ transplantation, have autoimmune disease or were on current or recent chemotherapy for acute lymphoblastic leukemia), and followed them from the start of primary vaccination with BNT162b2 mRNA SARS-CoV-2 until 1-year post-vaccination. We investigated the quality of vaccine immunogenicity, and longevity of hybrid immunity, in comparison to healthy children.

Results: Immunocompromised children failed to produce T cell and memory B cell (MBC) responses reaching thresholds of protection after 2 doses; a third dose however improved both responses. Initially robust hybrid immunity demonstrated significantly more decline in T cell and MBC responses in immunocompromised compared to healthy children, to levels below the protective threshold by month 12.

Discussion: Immunocompromised children may benefit from a 3-dose primary vaccination regime, with yearly or twice-yearly booster doses for sustained immunity.

Keywords: COVID-19; T cells; adaptive immunity; correlates of protection; memory B cells; vaccine durability; vaccine immunogenicity.

Figure 1

Immunocompromised children have reduced immunogenicity to two vaccine doses, but this is improved upon after dose 3. (A) Schematic of study schedule for healthy (grey) and immunocompromised (blue) children, all aged 5-12 years, in the MARkers of Vaccine Efficacy and Longevity in SARS-CoV-2 (MARVELS) study up to 12 months post vaccination. Healthy children (n = 116) were given two doses of monovalent 10mcg BNT162b2 on day 0 and 21 of the study. Immunocompromised children (n = 19) were given the same 2 doses, and a third dose of 10mcg BNT162b2 2 months after dose 2 as part of their primary vaccination regime, according to physician discretion. Venous blood was drawn at day 10 after dose one, 6 weeks post dose 2 only for immunocompromised children who received 3 doses of vaccine, and 3, 6 and 12 months after the last dose of primary vaccination. Children who acquired natural SARS-CoV-2 infection (both symptomatic and asymptomatic) or received a booster dose before each sampling time point were excluded from comparison for that time point. Thus, for the post dose 1 time point, n = 110 healthy and n = 16 immunocompromised, for the post dose 2 time point, n = 60 healthy and n = 7 immunocompromised, and for the post dose 3 time point, n = 3 immunocompromised. For the post-dose 2 and pre-dose 3 time points for immunocompromised children, data from 4 children who received a 2-dose primary vaccination regime and 15 children who received a 3-dose regime were combined (marked with #). (B) Anti-Spike (S) IgG titers post dose 1 and post dose 2 of vaccine. (C) Titers of antibodies that neutralized 50% of Wuhan-Hu-1 S protein RBD binding to ACE2, as measured using surrogate virus neutralization test (sVNT₅₀). (D) Titers of antibodies that neutralized 50% of indicated variants, as measured using pseudovirus neutralization test (pVNT₅₀), for post dose 2 time point. (E) Percentage of S+ memory B cells (MBCs) out of total B cells. (F) S-reactive T cell responses measured by post-stimulation interferon (IFN)-γ. (G) Correlation of total immunosuppression score with S-reactive T cell responses as measured by post-stimulation interferon-γ 10 days after dose 1 of vaccine. (H) pVNT₅₀ before and after dose 3 for healthy and immunocompromised children. (I) Titers of antibodies that neutralized 50% of indicated variants, as measured using pseudovirus neutralization test (pVNT₅₀), for before and after dose 3. (J) S-reactive T cell responses before and after dose 3. For all box-and-whisker graphs, the top and bottom boundaries of the boxes indicate upper and lower quartiles, respectively, the center line indicates median and the whiskers represent the range. For all panels, a two-tailed Mann-Whitney U test was used for comparisons between groups. Ns, not significant, *P ≤ 0.05. The schematic was created in BioRender.com.

Figure 2

Longevity of hybrid immunity is reduced in immunocompromised children compared to healthy children. Immunological parameters of healthy and immunocompromised children with hybrid immunity, at month 3, 6 and 12 after completion of primary vaccination regime. 12 immunocompromised children had hybrid immunity at month 3 after having received 3 doses of vaccine and acquired one to two episodes of natural infection, and 54 healthy children. Children who acquired natural SARS-CoV-2 infection (both symptomatic and asymptomatic) or received a booster dose before month 6 and 12 were excluded from comparison for that time point. Thus, for month 6, n = 10 immunocompromised and n = 54 healthy, and for month 12, n = 7 immunocompromised and n = 17 healthy. (A) Anti-S IgG titers of healthy and immunocompromised children with hybrid immunity at month 3, with month 6 and month 12 titers for those who did not encounter further antigenic exposures. (B) sVNT₅₀ against Wuhan-Hu-1. (C) pVNT₅₀ against variants of concern. (D) S+ MBCs out of total B cells. (E) S-reactive T cell responses measured by post-stimulation IFN-γ. For all box-and-whisker graphs, the top and bottom boundaries of the boxes indicate upper and lower quartiles, respectively, the center line indicates median and the whiskers represent the range. For all panels, a two-tailed Mann-Whitney U test was used for comparisons between groups. Ns, not significant, *P ≤ 0.05. (F) Proportion of subjects who had Omicron BA.2 pVNT50 titers above the threshold of protection, between month 3, when hybrid immunity was first attained, and month 12. (G) Proportion of subjects who had S+ MBC responses above the threshold of protection. (H) Proportion of subjects who had S-reactive T-cell responses. Comparison of Kaplan Meier curves was analyzed with Logrank (Mantel-Cox) test.