Humoral and cellular immune correlates of protection against COVID-19 in kidney transplant recipients

Abstract

As solid organ transplant recipients are at high risk of severe COVID-19 and respond poorly to primary SARS-CoV-2 mRNA vaccination, they have been prioritized for booster vaccination. However, an immunological correlate of protection has not been identified in this vulnerable population. We conducted a prospective monocentric cohort study of 65 kidney transplant recipients who received 3 doses of BNT162b2 mRNA vaccine. Associations among breakthrough infection (BTI), vaccine responses, and patient characteristics were explored in 54 patients. Symptomatic COVID-19 was diagnosed in 32% of kidney transplant recipients during a period of 6 months after booster vaccination. During this period, SARS-CoV-2 delta and omicron were the dominant variants in the general population. Univariate Analyses identified the avidity of SARS-CoV-2 receptor binding domain binding IgG, neutralizing antibodies, and SARS-CoV-2 S2-specific interferon gamma responses as correlates of protection against BTI. No demographic or clinical parameter correlated with the risk of BTI. In multivariate analysis, the risk of BTI was best predicted by neutralizing antibody and S2-specific interferon gamma responses. In conclusion, T cell responses may help compensate for the suboptimal antibody response to booster vaccination in kidney transplant recipients. Further studies are needed to confirm these findings.

Keywords: COVID-19; breakthrough cases; immunogenicity; kidney transplantation; mRNA vaccination.

Figure 1

Study flow chart.

Figure 2

Occurrence of symptomatic BTI in relation to incidence of COVID-19 in the Belgian population. The black line represents the proportion of kidney transplant recipients who remained free of SARS-CoV-2 BTI during the study period. The dotted lines are 95% confidence intervals. Day 0 is the date of the first vaccine dose for each patient. Patients were followed-up until the first day of administration of the fourth vaccine dose (February 1, 2022). Each cross represents a patient lost to follow-up. Numbers of patients with BTI, lost to follow-up, and at risk of BTI are indicated in the lower part of the graph (black numbers: before the second vaccine dose; gray: between the second and third dose of vaccine; green: after the third vaccine dose). The gray line represents daily confirmed COVID-19 cases per million people 7 days rolling average in Belgium. Day 0 for the gray line is the median date of the first vaccination of the study population (March 10, 2021). Proportions of infections with VOC are presented in the upper part of the figure (alpha in green, delta in orange, omicron BA.1 in gray and omicron BA.2 in pink, and others in white). BTI, breakthrough infection

Figure 3

Immune response to second and third COVID-19 mRNA vaccinations in kidney transplant recipients. (A) Immune responses, including SARS-CoV-2 receptor binding domain (RBD)-specific binding IgG titer (RBD IgG) and avidity (RBD IgG avidity), SARS-CoV-2 Wuhan neutralizing antibody titer (Wuhan NT₅₀) and S1 (S1 Interferon gamma) and S2 (S2 Interferon gamma) domain-specific Interferon gamma-producing cells, were measured one month after the 2nd dose (D2D28) and one month after the 3rd dose (D3D28) of vaccine among kidney transplant recipients naive of symptomatic BTI at D3D28 (n = 53). Each data point represents a sample. Samples were available for 51 patients at D2D28 and 53 patients at D3D28. The lower limit of detection (LLOD) is 5.4 IU/mL for RBD binding IgG and 77 IU/mL for Wuhan NT₅₀. Twenty-one patients had detectable RBD binding IgG at D2D28 and 27 at D3D28. One patient had detectable neutralizing antibodies at D2D28 and 24 at D3D28. RBD IgG avidity was only measured for samples with RBD binding IgG levels > 5.4 IU/mL. Statistical comparisons were made using Wilcoxon signed-rank test. RBD: receptor binding domain. Wuhan NT₅₀: 50% neutralizing antibody titer against Wuhan strain. S1 or S2 Interferon gamma: S1 or S2 specific PMBC responses. ∗∗ P < .01; and ∗∗∗∗ P < .0001. (B) Spearman correlation between S1 and S2 IFNγ response measured at D2D28 and D3D28. The dotted line indicates the function f(X)=Y.

Figure 4

Univariate analysis of risk factors of symptomatic breakthrough infection in kidney transplant recipients. Univariate logistic regression was used to assess the relationship between immunological, demographic, and clinical parameters and the occurrence of symptomatic BTI. Obesity was defined as BMI > 30 kg/m². Transplantation rank was divided into first transplantation and second or more transplantation. Immunological parameters and absolute lymphocyte counts were log10 transformed before analysis. RBD: receptor binding domain. Wuhan NT₅₀: 50% neutralizing antibody titer against Wuhan strain. S1 or S2 Interferon gamma: S1 or S2-specific PMBC responses. eGFR: estimated glomerular function rate; OKT3: muromumab; ATG: anti-thymoglobulin. Ly count: absolute lymphocyte count.

Figure 5

Immune response to booster COVID-19 vaccination in kidney transplant recipients with and without symptomatic BTI. Humoral and cellular immune responses were measured 28 days after a third dose of BNT162b2 COVID-19 vaccine in 53 kidney transplant recipients who developed (red symbols) or did not develop (blue symbols) symptomatic SARS-CoV-2 BTI after vaccination. Black bars indicate geometric means. Variables were compared with Wilcoxon—Mann--Whitney test. ∗ P < .05; ∗∗ P < .01. (A) S1-specific binding IgG (S1 IgG), S2-specific binding IgG (S2 IgG), S1-specific binding IgG antibody–dependant cellular phagocytosis (S1 IgG ADCP), and S2-specific binding IgG antibody –dependant cellular phagocytosis (S1 IgG ADCP). (B) SARS-CoV-2 receptor binding domain (RBD)-specific binding IgG titer (RBD IgG) and avidity (RBD IgG avidity), SARS-CoV-2 Wuhan neutralizing antibody titer (Wuhan NT₅₀). (C) S1-specific cells producing Interferon gamma (S1 Interferon gamma) and S2-specific cells producing Interferon gamma (S2 Interferon gamma). (D) Pearson correlation matrix of immune response parameters. Correlation coefficients are color coded. Circle size indicates the absolute value of corresponding correlation coefficients. All correlations were statistically significant, except those marked by a cross instead of a circle.

Figure 6

Risk BTI according to neutralizing antibody against Wuhan and S2 T cell response. (A) Patients were categorized on the basis of median neutralizing antibody titer and S2 Interferon gamma spot frequency in low responders (gray symbols; low neutralizing antibody titer and low S2 IFNγ spot frequency), medium responders (red symbols; high neutralizing antibody titer or high S2 IFNγ spot frequency) and high responders (green symbols; high neutralizing antibody titer and high S2 IFNγ spot frequency). Circled symbols indicate patients with BTI. Dotted line indicates the median S2 IFNγ response (16.1 SFC/10⁶ PBMC). Black line indicates the median neutralizing antibody titer (77 IU/mL). (B) Geometric mean percentage and 95% CI of patients with BTI in the 3 patient categories. ∗∗: P < .01; ns: not significant. Fisher test. (C) Odds ratio and 95% CI of risk of BTI in patients with low neutralizing antibody titer with high or low S2 IFNγ spot frequency (NT₅₀- S2 IFNγ+) and in patients with low S2 IFNγ spot frequency with high or low neutralizing antibody titer (NT₅₀+ S2 IFNγ-).