Lung Transplant Recipients Immunogenicity after Heterologous ChAdOx1 nCoV-19—BNT162b2 mRNA Vaccination

(1) Background: High immunosuppressive regimen in lung transplant recipients (LTRs) hampers the immune response to vaccination. We prospectively investigated the immunogenicity of heterologous ChAdOx1 nCoV-19-BNT162b2 mRNA vaccination in an LTR cohort. (2) Methods: Forty-nine COVID-19 naïve LTRs received a two-dose regimen ChAdOx1 nCoV-19 vaccine. A subset of 32 patients received a booster dose of BNT162b2 mRNA vaccine 18 weeks after the second dose. (3) Results: Two-doses of ChAdOx1 nCoV-19 induced poor immunogenicity with 7.2% seropositivity at day 180 and low neutralizing capacities. The BNT162b2 mRNA vaccine induced significant increases in IgG titers with means of 197.8 binding antibody units per milliliter (BAU/mL) (95% CI 0–491.4) and neutralizing antibodies, with means of 76.6 AU/mL (95% CI 0–159.6). At day 238, 32.2% of LTRs seroconverted after the booster dose. Seroneutralization capacities against Delta and Omicron variants were found in only 13 and 9 LTRs, respectively. Mycophenolate mofetil and high-dose corticosteroids were associated with a weak serological response. (4) Conclusions: The immunogenicity of a two-dose ChAdOx1 nCoV-19 vaccine regimen was very poor in LTRs, but was significantly enhanced after the booster dose in one-third of LTRs. In immunocompromised individuals, the administration of a fourth dose may be considered to increase the immune response against SARS-CoV-2.

Heterologous vaccination (i.e., primary course with two different COVID-19 vaccines or the use of a third booster dose of a different COVID-19 vaccine) has been shown to induce strong immune responses, inducing improved humoral and cellular immune responses in the general population and immunocompromised individuals [8][9][10]. Based on currently available evidence, both the European Medicines Agency and the European Centre for Disease Prevention and Control recommend a heterologous vaccination course against COVID-19, either in the primary course or as a booster [11].
However, data regarding the immunogenicity of a third vaccine dose and heterologous vaccination in LTRs are scarce. In the present study, we aimed at evaluating the immune response to the two-dose regimen of ChAdOx1 nCoV-19 vaccine in a prospective cohort of LTRs up to 6 months after the first administration. In addition, we investigated the 28-day impact of the heterologous third dose (BNT162b2 mRNA vaccine) on the immune response in a subset of LTR participants.

Study Participants
Fifty-four LTRs were vaccinated against COVID-19 with a two-dose regimen ChA-dOx1 nCoV-19 vaccine (AZD1222, AstraZeneca, Cambridge, UK) on 7 March 2021 (day 0). The second dose was administered 12 weeks later, on 6 June 2021 (day 84). The third heterologous dose of BNT162b2 mRNA COVID-19 vaccine (Comirnaty ® , Pfizer/BioNtech, Mainz, Germany) was administered 18 weeks after the second dose, on 7 October 2021 (at day 210). Figure 1 summarizes the STROBE flow-chart of the study cohort whose main demographic characteristics are recapitulated in Table 1.  Five LTRs with previous SARS-CoV-2 infection (documented by a positive PCR and/or positive antibodies against the nucleocapsid (NCP)) were excluded. Forty-nine Five LTRs with previous SARS-CoV-2 infection (documented by a positive PCR and/or positive antibodies against the nucleocapsid (NCP)) were excluded. Forty-nine LTRs without previous SARS-CoV-2 infection were considered COVID-19-naïve before the study and enrolled in the present ongoing prospective and non-interventional clinical trial, approved by an ethical commission (extension of the study protocol of the CRO-VAX-HCP study, EudraCT registration number: 2020-006149-21, CE Mont-Godinne 01/2021) [12]. Among them, 32 received the third booster dose at day 210, and 28 pursued the investigation up to day 238.
All vaccinated and enrolled LTRs provided written informed consent prior to data and specimen collection. Demographic and clinical data were retrieved from their electronic medical record at baseline.
A total of 291 sera were aliquoted and stored at −80 • C until analysis. Sera at days 84 and 210 were collected prior to the administration of the second ChadOx1 nCov-19 and booster BNT162b2 dose, respectively.

Analytical Procedures for Evaluation of Immunogenicity
Total antibodies against the SARS-CoV-2 Nucleocapsid (NCP) (Elecsys ® Anti-SARS-CoV-2 NCP qualitative ECLIA, Roche Diagnostics, Machelen, Belgium) were measured to document a previous infectious episode. According to the manufacturer's instructions, the positivity threshold cutoff was set to 1.
Anti-SARS-CoV-2 responses were assessed and performed on a VITROS ® 5600 integrated system (Ortho Clinical Diagnostics, New Jersey, USA). All samples were analyzed using anti-SARS-CoV-2 IgG quantitative immunoassays (Ortho Clinical Diagnostics) against the trimeric spike S1 protein. The results are expressed as binding antibody units per milliliter (BAU/mL), and results < 2 and >4000 BAU/mL (lower and upper limits of quantification) were rounded to 2 and 4000 BAU/mL for statistical analyses. The positivity cut-off was set to ≥17.8 BAU/mL as reported by the manufacturer.
The neutralizing capacity of antibodies (NAbs) was estimated as previously described [13]. Briefly, NAbs inhibiting the binding of the receptor-binding domain of the surface spike to the human angiotensin-converting enzyme 2 receptor were measured by performing a SARS-CoV-2 surrogate virus neutralization test (sVNT) using the iFlash-2019 nCoV NAbs assay (YHLO Biotech Co., Shenzhen, China). The positivity cut-off was set to ≥10 AU/mL according to the manufacturer's instructions.
All serum samples (n = 291) were assessed using a pseudovirus neutralization test (pVNT) for wild-type SARS-CoV-2 (referred to as SARS-CoV-2 spike protein carrying the original D614 genotype). Samples with positive NAbs for the wild-type virus (n = 45) were also assessed for the Delta and the Omicron variants. A sample was considered negative if the half maximal inhibitory concentration (IC50) value of that sample was less than 1:20 dilution [14,15].

Statistical Analyses
Quantitative data regarding the serological response are expressed in log 10 (geometric mean ± 95% confidence interval (CI)) and descriptive statistics were used to analyze the demographic and clinical data. Statistical significance across longitudinal collection time-points was assessed on raw data by the Kruskal-Wallis test followed by Dunn's multiple comparison tests. The Mann-Whitney test was performed to evaluate the impact of immunosuppressive regimen on anti-SARS-CoV-2 IgG titers. A chi-square test or a Fisher's exact test was used to determine if there was a significant relationship between two variables. A p-value < 0.05 was considered statistically significant. Statistical analyses were performed using GraphPad Prism (v 9.2.0 for Windows, GraphPad Software, San Diego, CA, USA).

Demographic and Clinical Characteristics of Lung Transplant Recipients
The baseline characteristics of the study population are summarized in Table 1. Among the 49 included participants, 24 (48.9%) were women (median age = 63, interquartile range (IQR) 58-66) and 25 (51.1%) were men (median age = 67, IQR 62-68). The initial respiratory diseases that led to lung transplant were chronic obstructive pulmonary disease in 34 (69.4%) LTRs and pulmonary fibrosis in 7 (14.3%) LTRs, while rarer respiratory diseases (cystic fibrosis, lymphangioleiomyomatosis, chronic lung allograft dysfunction and pulmonary hypertension) accounted for 16.3% of the total. The median time between transplantation and the first dose of vaccine was 3.8 years (IQR 1.8-6.1). The immunosuppressive regimen included corticosteroids and calcineurin inhibitors (tacrolimus or cyclosporine) in all patients (n = 49) and the addition in most patients an antimetabolite (mycophenolate mofetil [n = 25] or azathioprine [n = 11]) or everolimus (n = 8) ( Figure 1 and Table 1).

Seroconversion and Neutralizing Capacity in Vaccinated LTRs
Regarding anti-SARS-CoV-2 IgG titers, no significant modification was observed regarding anti-SARS-CoV-2 IgG titers after two doses of ChadOx1 nCov-19 vaccine in LTRs (i.e., from day 0 to 180) (Figure 2a). At day 180, the serological response rate was 7.2% ( Table 2). Two LTRs (4.0%) were considered responders after the first dose, with IgG titers of 87. 6 Table 2. Anti-SARS-CoV-2 IgG and NAbs titers, and proportion of seropositive LTRs at each collection time point. For sVNT NAbs titers, no significant difference was observed after the two-dose regimen of ChAdOx1 nCoV-19 vaccine in LTRs (i.e., from day 0 to 180). Before vaccination (day 0), one LTR presented positive NAbs titers (14.4 AU/mL), while negative having anti-nucleocapsid titers. This was considered a false-seropositive patient, probably due to interference because NAbs titers did not progress throughout the follow-up. At day 112, five LTRs (11.4%) displayed detectable NAbs titers, with discordant results for two LTRs, which had negative anti-SARS-CoV-2 IgG titers (Figure 2d, black lines and Table 2). None of them had developed previous SARS-CoV-2 infection, as witnessed by negative antinucleocapsid titers at day 0 and NAbs titers remaining close to the negativity cut-off with 11.4 and 12.2 AU/mL, respectively. The third dose led to a significant increase in mean NAbs titers at day 238 compared with primary vaccination course time points (days 0, 28, and 84), and increased by 9.5-fold between day 210 and 238. At day 238, nine LTRs (32.2%) displayed positive NAbs titers with a mean of 229 AU/mL (95% CI 0-490.8) (Figure 2d). At day 238, the two IgG-seropositive LTRs after the primary vaccination course showed similar trends in NAbs titers compared to their IgG titer evolution (Figure 2c,d, blue lines).

LTR Participants
The results of the pseudovirus neutralization test showed that four patients ( Table 3 reports the rate of seropositive LTRs with neutralization capacities or com plete non-responders after the heterologous ChAdOx1 nCoV-19-BNT162b2 mRNA va cine according to demographic and clinical characteristics of the participants. , assessed by pseudovirus neutralization test (pVNT) (geometric mean ± 95% CI). Statistical analysis by Kruskal-Wallis test followed by a post hoc Dunn's test provided non-significant results. (b) Paired Delta and Omicron SARS-CoV-2 neutralizing capacities of wild-type-positive samples (n = 45) assessed by pVNT (geometric mean ± 95% CI). Dotted black lines represent the positivity cut-offs for neutralization (>20 dilution titer, i.e., 1.3 in log 10 ). *** p < 0.001 wild-type vs. Delta and Omicron by the Kruskal-Wallis test followed by a post hoc Dunn's test. D1 and D2 correspond to the first and second doses of ChAdOx1, respectively; B corresponds to the booster dose, BNT162b2. Table 3 reports the rate of seropositive LTRs with neutralization capacities or complete non-responders after the heterologous ChAdOx1 nCoV-19-BNT162b2 mRNA vaccine according to demographic and clinical characteristics of the participants. Table 3. Demographic and clinical characteristics of LTRs with stratification by antibody and neutralization responses after vaccination scheme completion. Seropositive LTRs with neutralization capacities assessed by sVNT were evaluated at day 238 (n = 9) as well as non-responders (n = 19). Seroneutralization against wild-type SARS-CoV-2 and the Delta and Omicron variants was recorded for LTRs with positive titers for least at one collection time point.

Discussion
This study highlights that the two-dose regimen of ChAdOx1 nCoV-19 vaccine induces poor immunogenicity in LTRs, as witnessed by the low neutralizing capacities against wildtype and specific SARS-CoV-2 variants. Indeed, 92.8% of the included LTRs failed to mount detectable antibody response following this vaccination regimen. A third heterologous dose with the BNT162b2 vaccine, administered in a subset of LTRs, elicited higher rates of both seroconversion and positive neutralizing antibodies against SARS-CoV-2, observed in 32.2% of triple-vaccinated LTRs.
The immunosuppressive regimen appears to be a key player in preventing the development of immune response. Among SOTRs, LTRs are distinguished by a heavier immunosuppressive regimen, potentially explaining the discrepant response to vaccination observed in our study population compared with other studies. For instance, Schmidt et al. reported 35.3% seropositivity in SOTRs after completion of the two-dose ChAdOx1 nCoV-19 vaccine versus 7.2% of LTRs in our study [8]. Our data nevertheless corroborate previous study results [7,21,22], as only LTRs treated with a low daily dose of corticosteroids reached seropositivity with neutralization capacities (for the wild-type, Delta, and Omicron variants), and immunosuppressive regimens including either mycophenolate mofetil or high-dose corticosteroids were associated with poorer vaccine-induced immune response. However, the threshold for effective antibody protection following SARS-CoV-2 vaccination remains undetermined.
Facing the weak immune response after the completion of a primary vaccination course with vectored or mRNA vaccine in SOTRs, booster doses (third and even a fourth) have been proposed by national and international competent authorities [23,24] to reach and maintain higher immunogenicity levels. Previous studies reported up to 50% seropositivity 4 weeks after a third booster dose in distinct SOTR cohorts [25][26][27][28]. The seropositivity rate and NAbs titers increased more than four-fold following the booster dose in our specific patient cohort compared with the two-dose regimen of ChAdOx1 nCoV-19 vaccine. While the neutralization capacity of antibodies is increasingly considered a predictor of vaccination efficacy [29,30], both seroneutralization techniques, namely sVNT and pVNT, are poorly correlated with each other and with quantitative anti-SARS-CoV-2 immunoassays [13,15,18,31].
Our study has several limitations. Firstly, the national vaccination strategy negatively impacted the recruitment of a two-dose ChAdOx1 nCoV-19 vaccinated control group, although information about immunogenicity after a two-dose ChAdOx1 nCoV-19 regimen in the general population can be found in several previous studies [9,10,17,[32][33][34][35][36]. Secondly, our study design prevented us from comparing the effectiveness of a heterologous versus homologous booster dose in LTRs. Thirdly, we did not investigate the cellular immune response, while it has been shown that heterologous vaccination leads to a stronger CD4+ T cells response in the general population as well as in SOTRs [8,9], and vaccineinduced T cell responses are of similar magnitude to those seen after natural infection in immunocompetent individuals [37]. In a recent Letter to the Editor from Havlin et al., emergence of cellular response was reportedly detected in 47% of LTRs (n = 15) after the third vaccine dose [38]. High-affinity NAbs are closely regulated by T follicular helper cells, which are essential for the control of viral infections and vaccine responses by mediating the interaction between T and B cells [39][40][41].

Conclusions
We showed that a two-dose regimen of ChAdOx1 nCoV-19 vaccine provides poor immunogenic effects on humoral response in LTRs. Our data, specifically collected in LTRs, indicated that a third heterologous dose of BNT162b2 (booster) elicited higher titers of IgG and NAbs after homologous two-dose ChAdOx1 nCoV-19 vaccine. Nevertheless, the rate of three-dose non-responders remained substantial at day 238 with 67.8% of seronegative LTRs in our study population and was even more important regarding Omicron SARS-CoV-2 neutralizing capacities. These results support the strategy of administering a fourth SARS-CoV-2 vaccine dose in immunocompromised patients, a decision that was implemented in January 2022 by the Superior Health Council of Belgium.