These authors have contributed equally to this work and share first authorship
These authors have contributed equally to this work and share last authorship
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Among heart transplant (HT) recipients, a reduced immunological response to SARS-CoV-2 vaccination has been reported. We aimed to assess the humoral and T-cell response to SARS-CoV-2 vaccination in HT recipients to understand determinants of immunogenicity. HT recipients were prospectively enrolled from January 2021 until March 2022. Anti-SARS-CoV-2-Spike IgG levels were quantified after two and three doses of a SARS-CoV-2 vaccine (BNT162b2, mRNA1273, or AZD1222). Spike-specific T-cell responses were assessed using flow cytometry. Ninety-one patients were included in the study (69% male, median age 55 years, median time from HT to first vaccination 6.1 years). Seroconversion rates were 34% after two and 63% after three doses. Older patient age (
The clinical management of heart transplant (HT) recipients during the ongoing COVID-19 (coronavirus disease 2019) pandemic has been challenging, as these patients are at high risk of severe clinical impairment and adverse outcomes upon infection with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) (
Recent studies have reported a reduced humoral response to SARS-CoV-2 vaccination in SOT recipients (
Here, we report quantification of the humoral and T-cell response after a second and third dose of a COVID-19 vaccine in a consecutive cohort of heart transplant patients seen at a large transplant center. We also report determinants of vaccine response in this cohort.
From January 2021 until March 2022, we enrolled HT recipients that presented to the HT outpatient clinic of the University Heart & Vascular Center Hamburg, a large tertiary care center. Clinical variables including age, sex, date of transplantation, immunosuppressive medications, renal function
Participants had previously received two doses of the mRNA-based vaccines BNT162b2 (Pfizer-BioNTech) and mRNA-1273 (Moderna), or the viral vector-based AZD1222 (AstraZeneca) vaccine. After each the second and the third vaccination, anti-SARS-CoV-2 IgG concentrations in the blood serum and, in a subset of patients, spike-specific T-cell responses were assessed during routine ambulatory follow-up visits. Vaccinations had been administered by the patients’ primary care physicians or by specialized vaccination centers in accordance with the German prioritization guidelines and recommendations of the standing vaccination committee (STIKO) (
Overview of the study population.
We assessed the vaccine-specific humoral response after a median of 42 days (interquartile range [IQR] 29.0–98.8) after the second vaccination (“pre-booster”) and 39.5 days (28.0–62.0) after the third vaccination (“post-booster”). The DiaSorin LIAISON XL anti‐SARS‐CoV‐2 TrimericS IgG ChemiLuminescent ImmunoAssay (sensitivity 99.4%, specificity 99.8%) (
The spike-specific T-cell response was assessed using an activation-induced marker assay (AIM) similar to previous studies (
In accordance with a previous study on T-cell immunity after SARS-CoV-2 vaccination using a similar assay (
Continuous variables are presented as median with interquartile range (25th percentile to 75th percentile), and categorical variables as absolute numbers (relative frequencies). Pearson’s Chi-squared test, the Wilcoxon rank sum test and Fisher’s exact test were used to investigate the effect of the type of vaccine, vaccination regimen, and immunosuppressive agents on the immune response.
We analysed the association of several non-modifiable characteristics with antibody levels and seroconversion rates, namely patient age at first vaccine dose, sex, the timespan between vaccination and serological measurements, and timespan from HT to the first vaccination, and performed multivariable analyses (logistic and Tobit regression analyses) to identify determinants of seroconversion. Further, we used a Tobit regression model to account for values below the limit of detection of the assay used (<4.81 BAU/mL, which affected
Antibody concentrations were log-transformed for linear and Tobit regression analysis. For logistic regression models, we used the manufacturer’s threshold for antibody positivity (≥33.8 BAU/mL) to differentiate between positive and negative antibody responses, as described above. The effect of immunosuppressive agents on seroconversion rates was assessed in multivariable logistic regression analyses adjusting for age at first vaccine dose, sex, and an interaction effect between the two. The effect of prednisolone use on IgG concentrations was also studied adjusting for the timespan from HT to vaccination (in addition to age and sex) in a Tobit linear regression model since prednisolone is often included in immunosuppressive regimens in the first years after HT.
A two-tailed
Of 99 patients screened, 8 patients were excluded due to a SARS-CoV-2 infection prior to the first serologic assessment, resulting in a total of 91 HT recipients to be included in the study. Sixty-three patients were male (69%) and 28 female (31%). Median age was 55 years (IQR 48.5–61) and median time from HT to first vaccination was 6.1 years (1.6–13.2). Seventy-seven patients (85%) were treated with calcineurin inhibitors (62 [68%] with tacrolimus and 15 [16%] with ciclosporin), 41 (45%) with mycophenolate, 67 (74%) with everolimus and 49 (54%) with low-dose prednisolone (generally 5 mg per day). Forty-one patients (45%) were on dual, 48 (53%) on triple, and 2 (2%) on quadruple immunosuppressive therapy. All patients on a triple therapy regimen except for one received prednisolone. The most common immunosuppressive regimen was everolimus combined with tacrolimus, with or without prednisolone (24 patients [26%] and 21 patients [23%], respectively). A history of diabetes was reported in 27 patients (30%), and median eGFR was 49.0 mL/min (IQR 34.0–69.0) (
Baseline characteristics.
Overall, N = 91 |
Sex | |||
---|---|---|---|---|
Female, N = 28 |
Male, N = 63 |
|||
Clinical characteristics | ||||
Patient age at first vaccine dose [years] | 55 [48.5, 61] | 54 [41.8, 60] | 55 [50.5, 62] | |
Time from HT to first dose [years] | 6.1 [1.6, 13.2] | 4.3 [1.7, 10.6] | 7.2 [1.5, 13.2] | |
Time between first and second dose [days] | 42.0 [35.0, 42.0] | 42.0 [38.5, 42.2] | 41.0 [35.0, 42.0] | |
Time between second dose and first antibody measurement [days] | 42.0 [29.0, 98.8] | 37.0 [27.0, 94.8] | 44.5 [32.5, 94.2] | |
Time between third vaccination and second antibody measurement [days] | 39.5 [28.0, 62.0] | 32.5 [24.2, 75.8] | 41.0 [32.8, 48.2] | |
History of type 2 diabetes mellitus | 27 (30%) | 5 (18%) | 22 (35%) | |
Estimated glomerular filtration rate (eGFR [mL/min]) | 49.0 [34.0, 69.0] | 46.5 [33.2, 62.0] | 52.0 [34.0, 69.0] | |
Immunosuppressive therapy | ||||
Everolimus use | 67 (74%) | 22 (79%) | 45 (71%) | |
Cyclosporine use | 15 (16%) | 4 (14%) | 11 (17%) | |
Mycophenolate mofetil use | 41 (45%) | 8 (29%) | 33 (52%) | |
Prednisolone use | 49 (54%) | 16 (57%) | 33 (52%) | |
Tacrolimus use | 62 (68%) | 23 (82%) | 39 (62%) | |
CNI regimen | ||||
Regimen containing CNI | 77 (85%) | 27 (96%) | 50 (79%) | |
CNI-free regimen | 14 (15%) | 1 (3.6%) | 13 (21%) | |
Drug combinations | ||||
Tacrolimus + Everolimus + Prednisolone | 24 (26%) | 11 (39%) | 13 (21%) | |
Tacrolimus + Everolimus | 21 (23%) | 8 (29%) | 13 (21%) | |
Tacrolimus + Mycophenolate mofetil + Prednisolone | 8 (8.8%) | 1 (3.6%) | 7 (11%) | |
Tacrolimus + Mycophenolate mofetil | 7 (7.7%) | 2 (7.1%) | 5 (7.9%) | |
Tacrolimus + Everolimus + Mycophenolate mofetil + Prednisolone | 2 (2.2%) | 1 (3.6%) | 1 (1.6%) | |
Cyclosporine A + Mycophenolate mofetil + Prednisolone | 5 (5.5%) | 2 (7.1%) | 3 (4.8%) | |
Cyclosporine A + Mycophenolate mofetil | 4 (4.4%) | 1 (3.6%) | 3 (4.8%) | |
Cyclosporine A + Everolimus + Prednisolone | 3 (3.3%) | 1 (3.6%) | 2 (3.2%) | |
Cyclosporine A + Everolimus + Mycophenolate mofetil | 1 (1.1%) | 0 | 1 (1.6%) | |
Cyclosporine A + Everolimus | 2 (2.2%) | 0 | 0 | |
Everolimus + Mycophenolate mofetil + Prednisolone | 7 (7.7%) | 0 | 7 (11%) | |
Everolimus + Mycophenolate mofetil | 7 (7.7%) | 1 (3.6%) | 6 (9.5%) |
Median [IQR] or Frequency with number (%); Missing data excluded.
Continuous variables with few values and/or few different values are shown as categorical.
HT heart transplantation; CNI calcineurin inhibitor.
All patients screened received at least two SARS-CoV-2 vaccinations. Most participants (79%) received BNT162b2 as their first and second vaccine doses, while 11% received two doses of mRNA-1273. Of the 9 patients (9.9%) vaccinated with a first dose of AZD1222 (AstraZeneca), only 6 received a second dose of AZD1222, whereas the other 3 were switched to BNT162b2 as the second vaccine. The median time span between the two primary vaccinations was 42 days (35.0, 42.0). Regarding the third vaccination, more than two thirds (68%) received a third dose of an mRNA vaccine matching the primary vaccination (homologous vaccine regimen), while 32% were switched from BNT162b2 to mRNA-1273 or
SARS-CoV-2 vaccine regimens. BNT162b2 Tozinameran, Pfizer-BioNTech; mRNA-1273 Spikevax, Moderna; AZD1222 Vaxzevria, AstraZeneca.
More than half of the patients (52%) reported no (solicited or unsolicited) vaccination-associated adverse event whatsoever. Systemic reactogenicity was reported by 29.2%, including fatigue (16%), fevers and chills (5.5%), headaches (4.4%) and myalgia (3.3%), and local reactogenicity in the form of pain at the injection site by 27%. There were no vaccine-related adverse events requiring professional medical attention in our cohort.
After two vaccine doses, a positive humoral response could be detected in 31 out of 82 patients (37.8%), and the median antibody concentration was 74.6 BAU/mL (14.9–358.0). After three doses, the median antibody concentration was 553.0 BAU/mL (80.1–1,400.0), and the number of participants with seroconversion rose to 44 out of 70 in which antibody concentrations were measured (62.9%). A third vaccine dose nearly doubled the probability of a positive humoral response (
Anti-Spike IgG and SI after two and three doses. SI Stimulation index.
The spike-specific T-cell response was measured in a subset of 49 patients: In 18 patients after two vaccine doses, and in 39 patients after three doses (data after both two and three doses were available for 8 patients). A positive T-cell response was observed in 9 out of 18 patients after two doses (50%), compared to 29 out of 39 patients (74%) after three doses. Interestingly, out of these 29 patients with a detectable T-cell response, 8 (28%) did not show a humoral response (
Humoral and spike-specific T-cell response for the whole study population.
Overall, N = 91 |
Sex |
|
||
---|---|---|---|---|
Female, N = 28 |
Male, N = 63 |
|||
Humoral response | ||||
Anti-SARS-CoV-2 spike IgG after 2nd dose [BAU/mL] |
74.6 [14.9, 358.0] | 113.0 [45.7, 234.0] | 59.2 [14.4, 473.0] | 0.520 |
Seroconversion |
31/82 (38%) | 14/26 (54%) | 17/56 (30%) | 0.041 |
Anti-SARS-CoV-2 spike IgG after 3rd dose [BAU/mL] |
553.0 [80.1, 1,400.0] | 675.0 [131.0, 1,400.0] | 456.0 [77.9, 1,332.5] | 0.535 |
Seroconversion |
44/70 (63%) | 16/22 (73%) | 28/48 (58%) | 0.247 |
Spike-specific T-cell response | ||||
SI after 2nd dose (n = 18) | 2.2 (1.0, 7.6) | 2.7 (1.1, 7.7) | 2.0 (1.0, 6.1) | 0.5 |
Positive response after 2nd dose ( |
9/18 (50%) | 4/7 (57%) | 5/11 (45%) | >0.9 |
SI after 3rd dose (n = 39) | 5 (2, 12) | 6 (3, 15) | 4 (2, 11) | 0.3 |
Positive response after 3rd dose ( |
29/39 (74%) | 12/14 (86%) | 17/25 (68%) | 0.3 |
Median [IQR] or Frequency with number (%); Missing data excluded.
Wilcoxon rank sum test; Fisher’s exact test; Pearson’s Chi-squared test; Wilcoxon rank sum exact test.
Anti-SARS-CoV-2 spike IgG calculated without non-measurable patients.
Anti-SARS-CoV-2 spike IgG ≥33.8 BAU/mL.
Continuous variables with few values and/or few different values are shown as categorical.
SI, stimulation index.
In multivariable logistic regression analyses, higher patient age at vaccination was identified as a predictor of lower seroconversion rates (odds ratio [OR] 0.95, 95% confidence interval [CI] 0.91–0.99,
Using a pre-defined threshold for severe renal impairment of an eGFR <30 mL/min, we could not detect a significant influence of eGFR on log-transformed anti-Spike IgG levels after adjusting for age and timespan from HT to vaccination in a Tobit regression model. Of note, only 13 patients in our cohort had an eGFR below this threshold. While a history of type 2 diabetes mellitus (T2DM) was reported in 27 patients (30%), the median age in this group was higher compared to non-diabetic HT recipients (60 years [IQR 54.5–63] vs. 54 years [41–60],
The vaccine types, and whether a homologous or heterologous scheme was used, did not influence the humoral response to the second or third vaccine dose, respectively. These results persisted in a logistic regression analysis comparing seropositivity rates after homologous and heterologous vaccine schemes, adjusting for age and sex (OR 1.24, 95% CI 0.41–3.87,
In a multivariable logistic regression model adjusting for age, sex and the interaction effect between both these variables, prednisolone intake was associated with lower seroconversion rates after both two (OR 0.12, 95% CI 0.03–0.38,
Logistic regression results for the association between the two-dose or three-dose antibody positivity outcome and exposure to immunosuppressive drug use, adjusted for age at first vaccination and sex as an interaction term.
Antibody positivity after two vaccine doses (n = 82) | ||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Predictors | OR |
|
OR |
|
OR |
|
OR |
|
OR |
|
OR |
|
OR |
|
Age at first vaccine dose | 1.04 (0.98–1.11) | 0.240 | 1.03 (0.97–1.10) | 0.303 | 1.04 (0.98–1.11) | 0.210 | 1.03 (0.97–1.10) | 0.322 | 1.01 (0.95–1.09) | 0.656 | 1.04 (0.98–1.11) | 0.172 | 1.03 (0.98–1.10) | 0.266 |
Male sex | 791.10 (9.48–138251.09) | 0.006 | 749.86 (8.31–141907.54) | 0.007 | 958.61 (10.61–185645.85) | 0.005 | 867.15 (8.97–175602.99) | 0.006 | 194.23 (1.61–51378.07) | 0.043 | 1,386.33 (14.29–292694.91) | 0.004 | 964.21 (10.69–186659.88) | 0.005 |
Age at first vaccine dose * Male sex | 0.86 (0.78–0.94) | 0.001 | 0.87 (0.78–0.94) | 0.002 | 0.86 (0.78–0.94) | 0.001 | 0.86 (0.78–0.94) | 0.002 | 0.89 (0.80–0.97) | 0.012 | 0.86 (0.78–0.93) | 0.001 | 0.86 (0.78–0.94) | 0.002 |
Everolimus use | 2.05 (0.62–7.60) | 0.255 | ||||||||||||
Cyclosporine A use | 0.67 (0.12–3.29) | 0.633 | ||||||||||||
Mycophenolate use | 0.37 (0.12–1.09) | 0.078 | ||||||||||||
Prednisolone use | 0.12 (0.03–0.38) | <0.001 | ||||||||||||
Tacrolimus use | 2.35 (0.69–9.14) | 0.188 | ||||||||||||
Use of any calcineurin inhibitor | 3.12 (0.59–26.35) | 0.224 |
Antibody positivity after three vaccine doses (n = 70) | ||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Predictors | OR |
|
OR |
|
OR |
|
OR |
|
OR |
|
OR |
|
OR |
|
Age at first vaccine dose | 1.00 (0.93–1.06) | 0.913 | 1.01 (0.94–1.07) | 0.817 | 0.99 (0.93–1.06) | 0.870 | 1.00 (0.94–1.07) | 0.883 | 0.99 (0.92–1.05) | 0.707 | 0.99 (0.93–1.06) | 0.867 | 1.00 (0.93–1.06) | 0.920 |
Male sex | 1.17 (0.01–101.76) | 0.944 | 1.48 (0.01–138.00) | 0.867 | 1.17 (0.01–103.15) | 0.945 | 1.62 (0.01–155.78) | 0.837 | 1.00 (0.01–103.15) | 1.000 | 1.18 (0.01–103.84) | 0.943 | 1.18 (0.01–102.10) | 0.943 |
Age at first vaccine dose * Male sex | 0.99 (0.91–1.07) | 0.742 | 0.98 (0.91–1.07) | 0.692 | 0.99 (0.91–1.07) | 0.750 | 0.98 (0.91–1.07) | 0.717 | 0.99 (0.91–1.08) | 0.806 | 0.99 (0.91–1.07) | 0.735 | 0.99 (0.91–1.07) | 0.731 |
Everolimus use | 3.10 (1.01–10.01) | 0.051 | ||||||||||||
Cyclosporine A use | 1.18 (0.28–5.52) | 0.828 | ||||||||||||
Mycophenolate use | 0.41 (0.13–1.19) | 0.106 | ||||||||||||
Prednisolone use | 0.34 (0.11–0.95) | 0.046 | ||||||||||||
Tacrolimus use | 0.82 (0.27–2.45) | 0.730 | ||||||||||||
Use of any calcineurin inhibitor | 0.87 (0.23–3.10) | 0.832 |
OR, odds ratio.
Antibody positivity: anti-SARS-CoV-2 spike IgG concentration ≥33.8 BAU/mL.
Regarding the spike-specific T-cell responses, no association between either of the non-modifiable predictors mentioned above or the immunosuppressive regimen and the SI after three vaccine doses could be established.
This report details the humoral and cellular immune response to up to three SARS-CoV-2 vaccinations in a large, consecutive cohort of HT patients. We observed seroconversion rates of 34% and 63% and a T-cell response in 50% and 74% after two and three vaccine doses, respectively. Higher age and shorter time since transplantation were identified as predictors of seroconversion, while there was no association with vaccine type and type of immunosuppressive therapy.
While all approved SARS-CoV-2 vaccines have repeatedly and thoroughly proven to be safe in use (
We identified several non-modifiable predictors for an impaired humoral response, including higher patient age, and a shorter timespan from HT to vaccination. This is consistent with previous reports (
While antibody production represents a major mechanism of vaccine-induced immunogenicity, eliciting a T-cell response is considered important for long-term protection against severe disease after SARS-CoV-2 vaccination (
Percentage of positive vs. negative T-cell responses in relation to antibody response after 3rd vaccine dose.
T-cell response after three vaccine doses in relation to humoral response.
Overall, |
Spike-specific T-cell response |
|
||
---|---|---|---|---|
Negative, |
Positive, |
|||
Antibody positivity | 0.3 | |||
Negative | 13 (33%) | 5 (50%) | 8 (28%) | |
Positive | 26 (67%) | 5 (50%) | 21 (72%) |
n (%).
Fisher’s exact test.
There are several advantages inherent to the single-center nature of the data, such as the consecutive enrollment of study participants, complete data capture and the homogeneous management regimen. On the other hand, the small sample size limits statistical power and generalizability of the findings. The small case number of COVID-19 infections of participants during the study period (see
In our study, the activation-induced marker assay for the assessment of the specific T-cell response was only applied in a subset of patients, which might have influenced conclusions on the interaction between cellular and humoral immunity. While rates of reported SARS-CoV-2 infection during the study period were low in our population, we did not assess antibody levels to SARS-CoV-2 nucleocapsid protein and thus cannot exclude undetected or asymptomatic infections prior to sample acquisition.
Despite ISHLT-recommended SARS-CoV-2 vaccination schedules, a significant proportion of HT recipients exhibit insufficient humoral and T-cell responses. Patient age and time since transplantation predict lower immunogenicity, but inhibitory effects of prednisolone (within 3-drug immunosuppressive regimens) on antibody production may be attenuated through booster vaccination. More data on the effect of immunosuppressive agents on immune response is warranted to improve management of this exceptionally vulnerable group of patients.
The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by The Ethics Committee of the Hamburg Chamber of Physicians (Ethikkommission der Ärztekammer Hamburg). The patients/participants provided their written informed consent to participate in this study.
FM and SK: Research design, data collection, analysis, and interpretation, manuscript preparation. KB: Statistical analysis, data visualization, and interpretation, critical revision of the manuscript. PK, HR, and SB: Critical revision of the manuscript. PD: Data collection, data analysis and interpretation. ML: Data collection and interpretation. MB, FB, NF, PB, CK, and AB: Critical revision of the manuscript. CM and MR: Research design, data collection, analysis, and interpretation, manuscript preparation.
SK received scholarship funding from the German Academic Scholarship Foundation and the German Centre for Cardiovascular Research (DZHK) and travel support from the International Society for Heart and Lung Transplantation. PK receives research support for basic, translational, and clinical research projects from the European Union, British Heart Foundation, Leducq Foundation, Medical Research Council (UK), and German Centre for Cardiovascular Research, from several drug and device companies active in atrial fibrillation. PK was partially supported by European Union BigData@Heart (grant agreement EU IMI 116074), AFFECT-AF (grant agreement 847770), and MAESTRIA (grant agreement 965286). PK has received honoraria from several pharmaceutical and medical device companies in the past, but not in the last 3 years. PK is listed as inventor on two patents held by University of Birmingham (Atrial Fibrillation Therapy WO 2015140571, Markers for Atrial Fibrillation WO 2016012783). All outside the submitted work. NF reports grants from Biotronik. All outside the submitted work. PB received funding from the German Research Foundation. All outside the submitted work. AB has received honoraria, consultancy fees and/or research support from Abbott, Abiomed, AstraZeneca, BerlinHeart, Medtronic (unrelated to the submitted work). SB has received speaker fees from Medtronic, Pfizer, Roche, Novartis, SiemensDiagnostics (unrelated to the submitted work). CM receives research funding from the German Center for Cardiovascular Research (DZHK) within the Promotion of women scientists’ program, the Deutsche Stiftung fuer Herzforschung and the Dr. Rolf Schwiete Stiftung and has received Honoraria from AstraZeneca, Novartis, Heinen & Loewenstein, Boehringer Ingelheim/Lilly, Bayer, Pfizer, Sanofi, Aventis, Apontis, Abbott (unrelated to the submitted work).
The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at:
AIM, Activation-induced marker assay; AZD1222, Vaxzevria (AstraZeneca, Cambridge, United Kingdom); BAU/mL, Binding antibody units per milliliter; BNT162b2, Tozinameran (Pfizer, New York City, USA; BioNTech, Mainz, Germany); CD, Cluster of differentiation; CI, Confidence interval; COVID-19, Coronavirus disease 2019; EDTA, Ethylenediaminetetraacetic acid; eGFR, Estimated glomerular filtration rate; HT, Heart transplantation; IgG, Immunoglobulin G; IQR, Interquartile range; ISHLT, International Society for Heart and Lung Transplantation; mAB, Monoclonal antibody; mRNA, Messenger ribonucleic acid; mRNA1273, Spikevax (Moderna, Cambridge, USA); OR, Odds ratio; PrEP, Pre-Exposure Prophylaxis; SARS-CoV-2, Severe acute respiratory syndrome coronavirus-2; SI, Stimulation index; SOT, Solid organ transplant; STIKO, Standing vaccination committee (Germany), “Ständige Impfkommission”; T2DM, Type 2 diabetes mellitus.