Evaluation of Immunogenicity to Three Doses of the SARS-CoV-2 BNT162b2 mRNA Vaccine in Lung Transplant Patients

The aim of the study was to explore the humoral and T-cell response in lung transplant (LuT) patients. Two-time points were considered, before (T0) and after (Tpost) the third dose of the BNT162b2 mRNA vaccine, comparing LuT with healthy donors (HD). LuT patients showed a lower serologic response against SARS-CoV-2 compared with HD at both time-points (p = 0.0001 and p = 0.0011, respectively). A lower percentage of IFNγ+orIL2+orTNFα+CD4+ and CD8+ T-cells LuT patients was observed in LuT patients compared with HD at T0 (CD4+: p = 0.0001; CD8+: p = 0.0005) and Tpost (CD4+: p = 0.0028; CD8+: p = 0.0114), as well as in the percentage of IFNγ+IL2+TNFα+CD4+ T-cells (T0: p = 0.0247; Tpost: p = 0.0367). Finally, at Tpost, a lower percentage of IFNγ+IL2+TNFα+ CD8+ T-cells in LuT patients compared with HD was found (p = 0.0147). LuT patients were stratified according to the lowest cut-off value for the detection of a humoral response (4.81 BAU/mL) at T0, into responder (R) and non-responder (NR) groups. In the R group, no differences in the percentage of IFNγ+or IL2+orTNFα+ and IFNγ+IL2+TNFα+CD4+ and CD8+ T-cells compared with HD at both time-points were observed. Otherwise, in the NR group, lower percentages of IFNγ+IL2+TNFα+CD4+ T-cells compared with the R group (T0: p = 0.0159; Tpost: p = 0.0159), as well as compared with the HD, at both time-points, were observed (T0: p = 0.0064; Tpost: p = 0.0064). These data seem to confirm that some LuT patients can mount cellular responses even in the absence of a positive humoral response (>33.8 BAU/mL), although this cellular response is dysfunctional and partially detrimental.


Introduction
Every year, 4500 lung transplants (LuT) are carried out over the world, with about 2000 in Europe and just over 100 in Italy [1].
Several studies have shown a higher risk of severe COVID-19 (Coronavirus Disease) among immunocompromised patients, but the impact on solid organ transplant (SOT) patients remains debated [2][3][4]. It is well known that LuT patients are at high risk of severe COVID-19 because the respiratory system is the main target of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection [5,6]. Furthermore, LuT patients may

Study Participants and Surveillance
An observational, monocentric, prospective study including adult LuT patients enrolled between October 2021 and March 2022 was conducted. As a control group, we enrolled healthy donors (HD), matched for age and gender. The inclusion criteria for both LuT patients and HD were: (I) being >18 years old, (II) having already received two mRNA BNT162b2 vaccine doses of anti-SARS-CoV-2, and (III) no evidence of previous SARS-CoV-2 infection.
Both groups were tested for the humoral and T-cell response at two time-points, two months before (T0) and after (Tpost) the third vaccine dose administration. Specifically, at T0, blood samples were taken on the same day as the third dose of the mRNA BNT162b2 vaccine, before the administration of the latter.
All LuT patients were evaluated for demographics, comorbidities, basic laboratory findings, months elapsed from transplant, immunosuppression therapy, and SARS-CoV-2 vaccination timing.

Determination of SARS-CoV-2 Anti-S and Anti-N
For each group, evaluation of the humoral response was performed at both time-points. The serum samples collected were evaluated with The DiaSorin Liaison SARS-CoV-2 TrimericS IgG (DiaSorin S.p.A, Saluggia, VC, Italy) chemiluminescence immunoassay (CLIA), which detects SARS-CoV-2 spike S1/S2 protein specific IgG antibody levels. The performance of the sensitivity and specificity were reported according to the manufacturer's instructions. The levels of anti-SARS-CoV-2 IgG antibodies were expressed using the World Health Organization International Standard (NIBSC code. 20/268) binding antibody unit (BAU/mL). A positive serologic response was defined as having detectable IgG antibodies against SARS-CoV-2 over 33.8 BAU/mL. The lower and upper detection limits of the assay were 4.81 BAU/mL and 2080 BAU/mL, respectively. To confirm no previous SARS-CoV-2 infection, the presence of the anti-N SARS-CoV-2 antibody was determined using the KT-1032 EDITM Novel Coronavirus COVID-19 IgG Enzyme Linked Immunosorbent Assay (ELISA) Kit (Epitope Diagnostics, Inc. 7110 Carroll Rd, San Diego, CA, USA) and it was performed according to the manufacturer's instructions. The average value of the absorbance of the negative control was less than 0.25, and the absorbance of the positive control was not less than 0.30.

Stimulation of T-Cell Using SARS-CoV-2 Peptides Libraries
To evaluate the T-cell specific response, peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation using Ficoll-Paque PLUS (Sigma-Aldrich, Saint Louis, MO, USA) and then cryopreserved in cell recovery media containing 10% dimethyl sufoxide (DMSO) supplemented with heat inactivated Fetal Calf Serum (FCS). Finally, PBMCs were stored at −196 • C until they were used.
The T-cell specific response was assessed using multiparametric flow cytometry after overnight stimulation with SARS-CoV-2 peptide libraries, as previously shown [14]. Briefly, 1 × 10 6 PBMCs resuspended in 200 µL of RPMI were incubated overnight at 37 • C and 5% CO 2 . For each patient, a negative (unstimulated condition) and positive control (phytohemagglutinin (PHA) condition, 5 µg/mL) was also included. Brefeldin A at a final concentration of 5 µg/mL was added in the culture after 1 h of incubation.
After overnight stimulation, PBMCs were incubated for 30 min with Fixable Viability Dye and thy were washed in stain buffer (SB) containing 1% FCS. PBMCs were stained with the appropriate mix of monoclonal antibodies (Pacific Blue-conjugated anti-CD45, APC-conjugated anti-CD8 and APC-Cy7-conjugated anti-CD4) and incubated in darkness at 4 • C for 20 min. Then, the red blood cells were lysed using the lysing solution (BD Biosciences, Franklin Lakes, NJ, USA), in darkness at room temperature for 20 min (BD Biosciences, Franklin Lakes, NJ, USA). Fix/Perm solution (BioLegend, San Diego, CA, USA) was used prior to intracellular staining (FITC-conjugated anti-IFNγ, PerCp-Cy5.5conjugated anti-TNFα, and PE-Cy7-conjugated anti-IL2), according to the manufacturer's instructions. The cells were washed once in Perm wash solution (BioLegend, San Diego, CA, USA) according to the manufacturer's instructions. All the antibodies were from BioLegend. Finally, the cells were fixed in Phosphate-Buffered Saline (PBS) containing 0.5% formaldehyde (Sigma-Aldrich, St. Louis, MO, USA).
The stained samples were acquired using the MACSQuant Flow Cytometer (Miltenyi Biotec, Bergisch Gladbach, Germany) and were analyzed using FlowJo™ v10.8.1 software. The gating strategy for analysis of the antigen-specific T-cells is illustrated in Figure 1. The cytokine background obtained from the negative condition (unstimulated) was subtracted from the stimulated ones. As previously shown [15], the co-expression of cytokines was analyzed via Boolean gating using FlowJo™ v10.8.1. The display and analysis of the different cytokine combinations was performed with SPICE v6.1.
We labelled T-cells producing at least one of the three cytokines as IFNγ+orIL2+orTNFα+ and those that simultaneously producing all three cytokines as IFNγ+IL2+TNFα+. We labelled T-cells producing at least one of the three cytokines as IFN +orIL2+orTNF + and those that simultaneously producing all three cytokines as IFN +IL2+TNF +.

Statistical Analysis
Statistical analyses were performed using GraphPad Prism v.9 for macOS. Twotailed p ≤ 0.05 was considered statistically significant. Data are represented as median with interquartile range (IQR). The nonparametric comparative Mann-Whitney test and the nonparametric Kruskal-Wallis's test with Dunn's post-test were used for comparing the medians between groups. Longitudinal evaluation of anti-S levels and percentages of T cells producing IFN +orIL2+orTNF + and IFNγ+IL2+TNFα+ was performed using the nonparametric Wilcoxon test. Correlations between quantitative data were assessed using the non-parametric Spearman test. Linear correlation was evaluated using the regression test. Distributions of different cytokine combinations were performed by the nonparametric Wilcoxon rank test using SPICE, distributed by the National Institute of Allergy and Infectious Diseases, NIH.

Statistical Analysis
Statistical analyses were performed using GraphPad Prism v.9 for macOS. Two-tailed p ≤ 0.05 was considered statistically significant. Data are represented as median with interquartile range (IQR). The nonparametric comparative Mann-Whitney test and the nonparametric Kruskal-Wallis's test with Dunn's post-test were used for comparing the medians between groups. Longitudinal evaluation of anti-S levels and percentages of T cells producing IFNγ+orIL2+orTNFα+ and IFNγ+IL2+TNFα+ was performed using the nonparametric Wilcoxon test. Correlations between quantitative data were assessed using the non-parametric Spearman test. Linear correlation was evaluated using the regression test. Distributions of different cytokine combinations were performed by the nonparametric Wilcoxon rank test using SPICE, distributed by the National Institute of Allergy and Infectious Diseases, NIH.

Characteristics of Study Population, Clinical Presentation, and Laboratory Findings
The demographic and clinical features of the study population are reported in Table 1. Nine LuT adult patients (two female and seven male) with a median age (IQR) of 56 (46-62) years and nine HD (two female and seven male) with a median age (IQR) of 50 (47-58) Vaccines 2022, 10, 1642 5 of 13 years were enrolled. Among the LuT patients, the underlying disease that led to the lung transplant was emphysema (4/9, 44.4%), idiopathic pulmonary fibrosis (2/9, 22.2%), histiocytosis X (2/9, 22.2%), or bilateral congenital bronchiectasis (1/9, 11.1%). Overall, 45% (4/9) of LuT received a bilateral lung transplant, whereas the remaining 55% (5/9) received a monolateral lung transplant. As reported in Table 1, the median time elapsed from the transplant was 24 months. The antirejection chronic immunosuppressive therapy was a combination of a calcineurin inhibitor, mycophenolate mofetil (MMF), and prednisone. All the LuT patients were receiving low-dose steroid and calcineurin inhibitors treatments at both vaccination time-points, while 66% (6/9) were also treated with mycophenolate ( Table 1).  Overall, all LuT patients had at least one comorbidity and the most common were arterial hypertension (55%), diabetes mellitus (44%), cardiopathy (22%), and dyslipidemia (11%) ( Table 1). No graft rejection in LuT patients was observed during the study period.
No adverse events among LuT patients and HD after receiving the third dose of the BNT162b2 mRNA vaccine were observed.
Finally, the median value (IQR) in days between the second dose of the vaccine and T0 for LuT patients and HD were 177 (175-182) and 179 (177-185), respectively.

Vaccination-Induced Humoral Response
Overall, all of the enrolled LuT patients and HD had a negative SARS-CoV-2 N-protein IgG serology test at both time-points.
At both time-points, the R group showed a statistically significant higher IgG bodies against the SARS-CoV-2 titer compared with the NR one (T0: 16 Figure 2B). No significant differences between t group and HD at both time-points were observed ( Figure 2B). Conversely, the NR g showed significant lower anti-S titer compared to the HD at both time-point (T0: 4.   (Figure 2A).

Evaluation of T-Cell Response against S and N Peptide Libraries
Overall, all the enrolled LuT patients and HD had no cytokine production after N peptide library stimulation at both time-points.
After S peptide library stimulation assessed by multiparametric flow cytometry, theTcell specific response showed an uneven T-cell subset distribution in LuT patients, and this was different between T0 and Tpost ( Figure 3A). Conversely, in the HD group, at both time-points, a heterogeneous distribution of T-cell cytokines producers was observed ( Figure 3A).

Evaluation of T-Cell Response against S and N Peptide Libraries
Overall, all the enrolled LuT patients and HD had no cytokine production after N peptide library stimulation at both time-points.
After S peptide library stimulation assessed by multiparametric flow cytometry, theT-cell specific response showed an uneven T-cell subset distribution in LuT patients, and this was different between T0 and Tpost ( Figure 3A). Conversely, in the HD group, at both time-points, a heterogeneous distribution of T-cell cytokines producers was observed ( Figure 3A). The IFN +orIL2+orTNF +CD4+ and CD8+ T-cell response was evaluated in the LuT and HD groups at two time-points: at T0 and Tpost. Longitudinal evaluation in the LuT group was performed using the Wilcoxon test. Differences between LuT and HD were evaluated using the nonparametric Mann-Whitney test. (C) The IFNγ+IL2+TNFα+ CD4+ and CD8+ T-cell response was evaluated in the LuT and HD groups at two time-points: at T0 and at Tpost. Longitudinal evaluation in the LuT group was performed using the Wilcoxon test. The differences between LuT and HD were evaluated using the nonparametric Mann-Whitney test. T0: before the third dose of vaccine; Tpost: two  IFNγ+orIL2+orTNFα+CD4+ and CD8+ T-cell response was evaluated in the LuT and HD groups at two time-points: at T0 and Tpost. Longitudinal evaluation in the LuT group was performed using the Wilcoxon test. Differences between LuT and HD were evaluated using the nonparametric Mann-Whitney test. (C) The IFNγ+IL2+TNFα+ CD4+ and CD8+ T-cell response was evaluated in the LuT and HD groups at two time-points: at T0 and at Tpost. Longitudinal evaluation in the LuT group was performed using the Wilcoxon test. The differences between LuT and HD were evaluated using the nonparametric Mann-Whitney test. T0: before the third dose of vaccine; Tpost: two months after the third dose; LuT: lung transplants; HD: healthy donors; ****: p < 0.0001; ***: 0.0001 < p < 0.001; **: 0.001 < p < 0.01; * 0.05 < p < 0.01. Finally, in LuT patients, no differences in the longitudinal evaluation of IFNγ+orIL2+or TNFα+ and IFNγ+IL2+TNFα+ T-cells were found ( Figure 3B,C, respectively).

Discussion
In this observational, monocentric, and prospective study, we investigated the im munogenicity before and after the third dose of BNT162b2 mRNA vaccine in LuT adul patients, evaluating both the humoral and specific T-cell response.
In line with other studies involving different SOT patients [16,17], the first main re sult of our study was a low serologic response against SARS-CoV-2 both before and afte the third dose of BNT162b2 mRNA vaccine in LuT patients compared with HD. Specifi cally, in line with Tobudic et al. [18], before the third dose, almost all of the LuT patient showed an anti-S antibody titer under the positivity cut-off value (33.8 BAU/mL). Con versely, after the third dose, we observed a clear tendency towards a better humoral re sponse, although it was still significantly lower compared with HD. Overall, our data are

Discussion
In this observational, monocentric, and prospective study, we investigated the immunogenicity before and after the third dose of BNT162b2 mRNA vaccine in LuT adult patients, evaluating both the humoral and specific T-cell response.
In line with other studies involving different SOT patients [16,17], the first main result of our study was a low serologic response against SARS-CoV-2 both before and after the third dose of BNT162b2 mRNA vaccine in LuT patients compared with HD. Specifically, in line with Tobudic et al. [18], before the third dose, almost all of the LuT patients showed an anti-S antibody titer under the positivity cut-off value (33.8 BAU/mL). Conversely, after the third dose, we observed a clear tendency towards a better humoral response, although it was still significantly lower compared with HD. Overall, our data are in line with other authors [18][19][20], showing an almost complete lack of anti-SARS-CoV-2 antibody response in LuT patients before the booster dose. On the other hand, as reported in a recent work by Catry et al. [21], the third dose induces an increase in the seroconversion rate, underlining its potential benefit, at least within two months from vaccine boosting.
Regarding immunosuppressive therapy, a reduced immune response was mostly reported in patients treated with MMF or rituximab [22][23][24]. MMF, one of the key immunomodulators used in anti-rejection regimens for SOT patients, was associated with weak antibody responses to COVID-19 vaccines [5,9,25] and after conventional vaccination [23]. However, stratifying all the enrolled LuT patients according to MMF therapy, no differences in the anti-SARS-CoV-2 antibody and specific T-cells response were observed. Demographic factors, time since transplantation, and differences in immunosuppressive regimens employed, could probably explain our result.
Although the first line of protection against SARS-CoV-2 includes pre-existing antibodies, induced by vaccination or infection, great safeguards can also be attributed to the T-cell response [16,26]. Indeed, as shown by Agrati et al. [27], in immunocompetent subjects, the BNT162b2 mRNA vaccine is able to elicit a coordinated spike-specific T-cell response characterized by the production of all Th1 cytokines, with IFNγ correlating with both TNFα and IL2. Cellular immunity after COVID-19 vaccination probably plays a contributing role in controlling SARS-CoV-2 infection, even in the absence of a humoral response [22]. Given this, we performed a broad characterization of the functional profiles of specific CD4+ and CD8+ T-cells in LuT patients, and compared the observed findings with HD.
In the LuT patients, we observed an imbalance in T-cell subset distribution at both time-points. Conversely, in the HD group, at both time-points, a heterogeneous distribution of T-cell subsets that produced cytokines was observed.
Overall, in LuT patients, we observed a lower percentage of IFNγ+orIL2+orTNFα+ and IFNγ+IL2+TNFα+ T-cell response compared with HD. Otherwise, a higher percentage of IFNγ+IL2-TNFα-CD8+ T-cells in LuT patients compared with HD was observed. This is in line with the results from Picchianti-Diamanti et al. [19], showing the production of only one cytokine by T-cells in fragile patients and suggesting a potential disfunction in T-cell response in frail subjects.
Next, stratifying LuT patients according to the lowest cut-off value of detection of humoral response (4.81 BAU/mL) at T0, into R and NR groups, no significant differences in the percentage of IFNγorIL2orTNFα and IFNγ+IL2+TNFα+ T-cell response in the R group compared with HD at both time-points were observed. Otherwise, a lower percentage of IFNγorIL2orTNFα and IFNγ+IL2+TNFα+ T-cell response in the NR group compared with the R group and HD at both time-points was observed. These data seem to confirm that some LuT patients can mount cellular responses even in the absence of a positive humoral response (>33.8 BAU/mL), although this cellular response is dysfunctional, and possibly partially detrimental. Further studies with a higher number of LuT patients are needed to confirm our findings.
Nevertheless, as reported by Callaghan et al. [28], in a real-world vaccine effectiveness study in SOT patients, SARS-CoV-2 vaccines reduced the risk of death from COVID-19 compared with unvaccinated SOT patients, although the level of vaccine-enabled protection in SOT recipients was markedly less than that observed in the general population.
Our study detected no breakthrough infection in LuT patients up to 2 months from the third dose. This could be explained by a T-cell response in the absence of humoral responses, and by more conscious use of non-pharmacological protective measures. Indeed, strong evidence has shown that routine masking protects against the spread of respiratory viruses, including SARS-CoV-2 [29]. However, more research is needed to determine the benefit of universal masking for SOT patients.
Our study has some limitations, such as the small sample size due to the peculiarity and high sensibility of lung transplants, as well as the absence of an in-depth investigation into the influence of immunosuppressive therapy on cellular responses. Moreover, we used S-peptide from the Wuhan SARS-CoV-2 strain for the stimulation of T-cells, although the Omicron variant was the dominant circulating strain. However, in terms of the humoral response, Kumar et al. [30] showed that three doses of mRNA vaccine resulted in poor neutralizing responses against the Omicron variant in LuT patients, suggesting the same trend for T-cell response. Further studies are needed to evaluate specific T-cell response against the Omicron variant.
In summary, our data indicate that LuT patients receiving three doses of the SARS-CoV-2 mRNA vaccine had lower humoral and cellular immune responses compared with HD. Overall, our data underline the possibility that some LuT patients can mount cellular responses even in the absence of seroconversion, even if they are lower, dysfunctional, and possibly partially detrimental. Given this, as suggested for other fragile subjects by several authors [20,[30][31][32][33], early intervention with monoclonal antibody treatment (such as tixagevimab and cilgavimab) could reduce the risk of COVID-19 severity in LuT patients and could be a useful preventive strategy.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.