This is a prospective, observational, single-center study comparing the kinetics of the anti-SARS-CoV-2 antibodies and Interferon-Gamma (IFN-γ) release in patients receiving a kidney transplant who received the ChAdOx1 vaccine as the primary vaccine schedule according to the maintenance immunosuppressive regimen prescribed on the first day after transplantation.

This is a post-hoc analysis of a prospective study previously published comparing the kinetics of SARS-CoV-2 immune response in dialysis patients and those undergoing a kidney transplant.7 For the primary study, the investigators calculated a sample size of 114 individuals to be included in each group. For this post hoc analysis, we selected those (n = 89) who received the ChAdOx1 nCoV-19 as primary schedule for vaccination, aiming to review the effect of the type of maintenance immunosuppression on humoral and cellular kinetics.

We included patients who received the ChAdOx1 vaccine as the primary schedule and had undergone kidney transplantation in our center from a living or deceased donor. We excluded patients with a previous diagnosis of COVID-19, negative SARS-CoV-2 IgG serology on inclusion, those living with HIV, and those currently undergoing cancer treatment. All patients were followed up for 12-months from the date of inclusion. A blood sample was collected immediately before the transplant and after 1-, 3-, 6-, and 12-months for immunological analysis. In each study visit, the patients were argued about vaccine boosters received and COVID-19 diagnosis.

The eligible population signed an informed consent form. The study was complied with the Good Clinical Practice guidelines and the Declaration of Helsinki and was approved by the Institutional Review Board of the Federal University of São Paulo (approval number: 5.056.135).

The main objective of the study was to compare the kinetics of anti-SARS-CoV-2 IgG titers and neutralizing antibodies, and the Interferon-Gamma (IFN-γ) release in de novo kidney transplant recipients receiving mycophenolate or azathioprine as part of the immunosuppressive regimen over a 12-month follow-up period. Secondary assessments were conducted at 1-, 3-, and 6-months.

Other objectives were to assess the number, type, and time of additional vaccine doses, relevant clinical outcomes such as COVID-19 incidence, death, all-cause hospitalization, and transplant outcomes such as cytomegalovirus infection, delayed graft function, acute rejection, and graft loss. These outcomes were also analyzed and compared between the groups at 1-, 3-, 6-, and 12-months.

The humoral immune was assessed by the kinetics of the SARS-CoV-2, IgG titers, and neutralizing antibodies in blood samples collected at five study visits: screening (immediately before the transplant), and study visits after 1-, 3-, 6-, and 12-months of inclusion.

We used the AdviseDx SARS-CoV-2 IgG II test (Abbott Laboratories, Sligo, Ireland; lower limit of positivity 50 AU/mL) to evaluate IgG antibodies targeting the receptor-binding domain (RBD) of the S1 subunit of the SARS-CoV-2 Spike protein.8 We used the cPass™ SARS-CoV-2 test (GenScript Laboratory, Rijswijk, The Netherlands; positivity limit 30 %) to evaluate the neutralizing activity of anti-SARS-CoV-2 antibodies.9

Cell-mediated immunity was assessed using the QuantiFERON SARS-CoV-2 RUO Starter Set (QIAGEN), which measures Interferon-Gamma (IFN-γ) release in response to SARS-CoV-2-specific antigens. Whole blood was collected into three tubes (Nil, Ag3, and Mitogen), incubated, and processed according to the manufacturer’s instructions. Results were interpreted as positive, negative, or indeterminate.10

In our country, the COVID-19 vaccination was managed by the National Immunization Program, and the public health system applied all vaccines. The KTRs have been prioritized for receiving vaccine doses and boosters since May 2021. During this study, all patients received at least one vaccine booster by the public health system. Our study did not intervene on vaccination policies of these patients.

The ChAdOx1 nCoV-19 is a non-human adenovirus vector vaccine given in two doses, with an interval of four to twelve weeks, developed by AstraZeneca.11 The study used to register the vaccine included 32,451 adult volunteers to receive the ChAdOx1 nCoV-19 vaccine or placebo, in a 2:1 ratio. The efficacy in preventing infection was 74 %, and 83.5 % in people over 65-years of age.11

According to the standard institutional protocol, all kidney transplant recipients undergo intravenous induction therapy, which includes 1 g of methylprednisolone administered intraoperatively and a single dose of rabbit Anti-Thymocyte Globulin (rATG) at 3 mg/kg on the first postoperative day. This study did not involve any interventions or modifications to the established immunosuppressive regimen. Maintenance therapy comprises prednisone 0.5 mg/kg QD tapered to 5 mg QD until the end of the first month and tacrolimus 0.05‒0.1 mg/kg BID for all patients. The third drug consists of mycophenolate sodium 720 mg BID or azathioprine 2 mg/kg QD, selected based on donor characteristics (standard or expanded criteria donor) and the recipient's immunological risk profile.

Categorical variables were expressed as percentages and compared using the Chi-Square test. Numerical variables were evaluated for normality using the Kolmogorov-Smirnov test. Quantitative variables were presented as a median and interquartile range for non-parametric variables and as a mean and standard deviation for parametric variables. Differences between groups were analyzed using the Mann-Whitney test.

We performed a linear regression using Generalized Estimated Equations (GEE) with pairwise comparison adjusted by Bonferroni to compare the time and group effect on anti-SARS-CoV IgG titers and neutralizing antibodies over time.12 Due to the small number of negative and indeterminate results, neutralizing antibody data were dichotomized into positive and non-positive.

Statistical analyses were performed using the statistical package SPSS version 29 (IBM Corp. Released 2022. IBM SPSS Statistics for Windows, Version 29.0, Armonk, NY, USA: IBM Corp.). We considered a p < 0.05 as statistically significant.

Between December 13, 2021 and April 05, 2022, 115 patients were enrolled, but 26 were excluded due to other primary vaccine regimens (n = 23), SARS-CoV-2 IgG negative at screening (n = 2), or receiving mTOR inhibitor as maintenance immunosuppressive therapy (n = 1), resulting in 89 patients included in this study (Fig. 1). They all had a blood sample collected at screening and a positive anti-SARS-CoV-2 IgG serology.

Fig. 1.

Flowchart of the study population.

The median age was 47.1-years, most were male (52.8 %) and white (54 %). The most common etiology of Chronic Kidney Disease (CKD) was undetermined (30 %), followed by chronic glomerulonephritis (27 %), diabetes mellitus (21.3 %), autosomal dominant polycystic kidney disease (7.0 %), urological diseases (5.6 %) and hypertension (4.5 %).

Comorbidities such as hypertension and diabetes mellitus were present in 80.9 % and 22.5 % of patients, respectively. Hemodialysis was the most common therapy of CKD (83.1 %), with median duration of 29.2-months. Of 89 patients, 51 received mycophenolate sodium and 38 received azathioprine as maintenance immunosuppressive regimen. In both groups, the median HLA mismatches in loci A, B and DR was 2, and panel-reactive antibody was 27.5 % and 7.9 %, respectively. Of all, 2.2 % of patients received a retransplantation (Table 1).

The median donor age was 50-years, 48 % were male and 50 % were white. Kidneys from a deceased donor were transplanted in 76 % of the patients; 30 % expanded criteria donors, with a median KDPI of 55 %, final creatinine of 1.41 mg/dL, and cold ischemia time of 22.8 h. In the MPS group, donors were older (54.0 vs. 38.0 years; p < 0.001) and filled the expanded criteria in 98.0 % vs. 68.4 % (p < 0.001). The proportion of male donors (60.8 % vs. 44.7 %; p = 0.133) and white donors (54.9 % vs. 57.9 %; p = 0.761) were similar between the groups.

As this study was conducted during the COVID-19 pandemic and transplant patients were prioritized for boosters, most patients (n = 69; 77.5 %) had already received at least one booster at inclusion (Tables 2 and 3). Of these, 63 received three doses, and 6 patients received four doses, with no difference in the distribution between groups. The time elapsed from the last dose of vaccine administered to inclusion in the study was the same for both groups (90-days; p = 0.990), with the BNT162b2 vaccine being preferentially administered in both groups (51 % and 65.8 %) (Tables 2 and 3). About the other vaccines, 23.6 % of patients received ChAdOx1 nCoV-19, 18 % received CoronaVac and 1.1 % Ad26.COV2.S (Tables 2 and 3). Among the patients who received the third and fourth doses of vaccine before the inclusion, the BNT162b2 vaccine was the preferred in both groups (67.6 % and 82.8 %; p = 0.273) and (60 % and 100 %; p = 0.741) (Table 3).

This study followed the national immunization policy against COVID-19 during the pandemic, and patients continued to receive boosters during the study follow-up and the vaccines available then. During the 12-month follow-up period, the study patients received 114 additional vaccine doses, 61 in the MPS group and 53 in the AZA group (p = 0.195). The distribution of boosters was very similar between the groups (Fig. 2).

Fig. 2.

Time of booster and number of accumulated doses in both groups after inclusion. Each point represents the accumulated number of additional vaccine doses in patients receiving mycophenolate (blue) or azathioprine (orange).

Most of the boosters were received until M3, with the AZA group receiving the most doses in this period (62.2 % vs. 37.8 %; p = 0.028). At M6, the MPS group received more doses than the AZA group (46.7 % vs. 27 %; p = 0.068). In addition, the vaccine of choice for both groups during the study was BNT162b2 (Table 4). The time between the last dose and the study visit was similar in both groups, except at M3, with 102 vs. 55 days in the MPS and AZA groups, respectively (p = 0.007) (Table 4).

Anti-SARS-CoV-2 IgG antibody titers were assessed at screening and in months-1, −3, −6, and −12 and compared according to the groups. As summarized in Table 5 and Fig. 3, IgG titers in the AZA group were numerically but not statistically superior compared to patients receiving MPS at all study visits (p = 0.286). In both groups, IgG titers varied similarly over time.

Fig. 3.

Evolution of anti-SARS-CoV-2 IgG antibody titers. Means and 95 % CI for anti-SARS-CoV-2 IgG antibody titers by group and study visit. Test used: AdviseDx SARS-CoV-2 IgG II test, Abbott Laboratories; lower limit of positivity 50 AU/mL.

Fig. 3 shows the means and confidence intervals of the anti-SARS-CoV-2 IgG antibody titers by group, by study visit, and the descriptive levels of the linear models with longitudinal random effects. In this model, the effects of three components are evaluated: time, group, and interaction between group and time (Fig. 3).

We performed a linear regression by GEE (Generalized Estimated Equations) with pairwise comparison adjusted by Bonferroni to compare the groups over time. A total of 417 measures were evaluated across all the study visits: 234 in the mycophenolate group and 183 in the azathioprine group. Throughout the study period, the mean antibody titers were similar in the mycophenolate group (11,611.2 IU/mL; 95 % CI: 8878.6‒14,343.7) and the azathioprine group (13,060.1 IU/mL; 95 % CI: 10,237.1–15,883.1) (p = 0.470; adjusted by Bonferroni) (Fig. 3; Table 6).

Considering the entire study population, time affected the mean antibody titers, mainly due to the drop in the mean antibodies in both groups at visit M1. Antibody titers recovered at visit M3, and the means stabilized at visits M6 and M12 (Fig. 3; Table 6).

Analyzing the effect of the interaction between time and the type of immunosuppressive regimen, there was no difference between the groups (p = 0.286; Table 6).

The neutralizing activity of antibodies between MPS and AZA groups at Screening (0.764 vs. 0.924; p = 0.006) and M1 (0.817 vs. 0.957; p = 0.004) was significantly different, but it was similar in M3, M6, and M12 (Table 7; Fig. 4). In the AZA group, the mean neutralizing activity remained stable over time, but there was a significant increase in the neutralizing activity of patients in the MPS group.

Fig. 4.

Evolution of neutralizing antibodies. Means and 95 % CI for anti-SARS-CoV-2 IgG neutralizing antibodies by group and study visit. Test used: AdviseDx SARS-CoV-2 IgG II test, Abbott Laboratories; lower limit of positivity 50 AU/mL.

We performed a linear regression by GEE (Generalized Estimated Equations) with a pairwise comparison to compare the groups over time. A total of 404 measures were evaluated over all the study visits: 224 in the mycophenolate group and 180 in the azathioprine group. Considering all the measurements, on average, the neutralizing activity of the antibodies was higher in the AZA group at 0.082 (95 % CI 0.021 to 0.143; p = 0.008, adjusted by Bonferroni).

Considering the entire study period, there was an increase in antibody-neutralizing activity, with an average gain of 0.175 (95 % CI 0.089 to 0.260; p = 0.001, adjusted by Bonferroni) between the screening visit and M12.

There was also a group and time effect with a reduction of −0.137 (95 % CI −0.241 to −0.033; p = 0.017, adjusted by Bonferroni) reduction in the difference in neutralizing antibody activity at 12-months in the AZA group compared to the MPS group, primarily due to increased neutralizing activity in the AZA group (Table 8).

Neutralizing antibodies were high in both groups and at all study visits (Fig. 5). At screening, positive neutralizing activity was higher in the AZA group (83.3 % vs. 97.3 %; p = 0.038). However, during the follow-up and boosters received by both groups, this difference was attenuated: M1 (88.9 % vs 97.2 %; p = 0.155), M3 (88.6 % vs. 97.3 %; p = 0.138), M6 (97.7 % vs. 100 %; p = 0.351), and M12 (97.7 % vs. 100 %; p = 0.383) (Fig. 5).

Fig. 5.

Distribution of neutralizing antibody responses by group and study visit. Incidence of positive and non-positive neutralizing antibodies between groups. M, Month. Test used: cPass™ SARS-CoV-2 test, GenScript Laboratory; positivity threshold 30 %.

As described on Methods, all patients included in this study had a positive anti-SARS-CoV-2 antibody detected at screening. But, on the day of the transplant, only 4.2 % of patients in the MPS group and 15.8 % in the AZA group had a positive cellular immunity test (p = 0.171). After one month, IFN-γ detection increased in both groups, but maintaining the advantage in the AZA group (13.3 % vs. 27.8 %; p = 0.266). On the following visits, the detection of IFN-γ release were similar in both groups: M3 (11.4 % vs. 13,5 %; p = 0.813), M6 (4.7 % vs. 5.4 %; p = 0.779), and M12 (20.5 % vs. 20.6 %; p = 0.807) (Fig. 6). Indeterminate results predominated in all study visits in both groups: screening (91.7 % vs. 78.9 %), M1 (68.9 % vs. 58.3 %), M3 (52.3 % vs. 56.8 %), M6 (58.1 % vs. 64.9 %), and M12 (20.5 % vs. 26.5 %).

Fig. 6.

Positive results of QuantiFERON IFN-γ release by group and study visits.

During the 12-month follow-up period, we also collected exploratory data on relevant clinical outcomes. We observed 18 cases of COVID-19 in the MPS group and 11 cases in the AZA group (p = 0.527), occurring earlier after transplantation in the MPS group (45 vs. 107 days; p = 0.003). There were three deaths, all of them in the MPS group – one due to urinary tract infection, and two due to COVID-19, at day 27 and day 50 after transplantation. All-cause hospitalizations also occurred more frequently in the MPS (68.6 % vs. 36.8 %; p = 0.003).

Regarding transplant outcomes, the MPS group had a higher incidence of delayed graft function (60.8 % vs. 36.8 %; p = 0.025) and cytomegalovirus infection (56.9 % vs. 13.2 %; p < 0.001). Acute rejection (7.8 % vs. 7.9 %; p = 0.993) and graft loss (9.8 % vs. 2.6 %; p = 0.182) were similar between the groups.