key: cord-0878783-qd0hoofo authors: Liefeldt, Lutz; Glander, Petra; Klotsche, Jens; Straub-Hohenbleicher, Henriette; Budde, Klemens; Eberspächer, Bettina; Friedersdorff, Frank; Halleck, Fabian; Hambach, Pia; Hofmann, Jörg; Koch, Nadine; Schmidt, Danilo; Schrezenmeier, Eva; Seelow, Evelyn; Weber, Ulrike; Zukunft, Bianca; Eckardt, Kai-Uwe; Choi, Mira; Bachmann, Friederike; Waiser, Johannes title: Predictors of Serological Response to SARS-CoV-2 Vaccination in Kidney Transplant Patients: Baseline Characteristics, Immunosuppression, and the Role of IMPDH Monitoring date: 2022-03-18 journal: J Clin Med DOI: 10.3390/jcm11061697 sha: 4f11bcac79412e683bda47d088781ace2646b031 doc_id: 878783 cord_uid: qd0hoofo Immunosuppression increases the risk of severe coronavirus disease 2019 (COVID-19). Morbidity and mortality of this disease in kidney transplant patients are higher than in the general population. As the vaccination response of transplant patients is weak, serological monitoring was performed. In this cohort study, we analyzed the determinants of vaccination response. All patients had no history of COVID-19. With anti-spike IgG monitoring, 148 responders and 415 non-responders were identified. We compared both groups using multivariate analyses of the cohort and a sub-cohort of mycophenolic-acid-treated patients. We investigated the influence of patient characteristics, immunosuppression, and erythrocyte inosine monophosphate dehydrogenase (IMPDH) activity. In responders, the time after transplantation was longer (13.5 vs. 8.5 years), the glomerular filtration rate was higher (56.9 vs. 47.8 mL/min/1.73 m(2)), and responders were younger (53.0 vs. 57.4 years). Heterologous vaccination was more effective than homologous vaccination. Calcineurin inhibitors plus mycophenolate reduced the seroconversion rate. No seroconversion was observed in belatacept patients. In mycophenolate-treated patients, IMPDH activity was a significantly better predictor of response than mycophenolate dose (AUC 0.84 vs. 0.62, p < 0.001). Immunosuppression strongly affects vaccine response. Modifications to immunosuppression should be considered in order to facilitate this response. Erythrocyte IMPDH activity can be used to guide mycophenolate treatment. Recipients of solid organ transplants are at an increased risk of severe disease and death after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection [1, 2] . The responsiveness of kidney transplant patients (KTPs) to SARS-CoV-2 vaccination is severely limited by maintenance immunosuppression [3] [4] [5] [6] . The response rate in heart and liver transplant recipients appears to be higher [7] . Consequently, given the low response rates of KTPs after two vaccine applications, three [8] [9] [10] [11] or more [12] vaccinations may be necessary to achieve a positive serological response. Most KTPs receive triple maintenance immunosuppression consisting of tacrolimus or cyclosporine, mycophenolic acid (MPA), and low-dose steroids. Previous studies have demonstrated that treatment with MPA is a negative predictor of vaccination success [3, 6, 13] , which is also observed in autoimmune diseases, such as systemic lupus erythematosus [14] . MPA dosing usually follows a standard regimen adapted by individual clinical events, such as rejection on one side and gastrointestinal symptoms, hematopoietic side effects, and opportunistic infections on the other side. Because of the well-known dose-response variability of MPA, therapeutic drug monitoring (TDM) has been established, either as MPA blood level or inosine 5 -monophosphate dehydrogenase (IMPDH) monitoring [15, 16] . We recently described that IMPDH activity measurement in erythrocytes is a useful strategy for longitudinal monitoring of MPA treatment [17] . The robustness and reproducibility of this method led to routine IMPDH monitoring of all MPA-treated patients in our clinic. IMPDH monitoring has displaced MPA trough level monitoring, as well as the measurement of IMPDH in mononuclear cells. Given that an optimal understanding of vaccination efficacy is essential to protect KTPs against severe coronavirus disease 2019 (COVID- 19) , we investigated the response rate to SARS-CoV-2 vaccination in a single-center cohort of renal allograft recipients to identify the factors associated with vaccination response, including the potential value of IMPDH activity measurement. We searched for all adult (≥18 years) kidney allograft recipients with a functioning graft at Charité-Universitätsmedizin Berlin using our electronic patient record system, TBase [18] . All patients with "complete" vaccination according to the protocol of the manufacturer with authorized vaccines were included, i.e., two applications in the case of Comirnaty ® (BioNTech Manufacturing GmbH, Mainz, Germany), Spikevax ® (Moderna Biotech, Madrid, Spain), and Vaxzevria ® (AstraZeneca AB, Södertälje, Sweden); or one injection in the case of COVID-19 Vaccine Janssen ® (Janssen-Cilag International NV, Beerse, Belgium). Combinations of different vaccines were allowed. Further inclusion criteria were an anti-spike IgG antibody test ≥28 days after the last vaccination, no clinical history of COVID-19, and a negative anti-nucleocapsid antibody test. The patients underwent negative serological tests before vaccination. Third vaccine applications led to the exclusion of patients. Positive results for anti-spike IgA only were not classified as seroconversion. The deadline for the database retrieval was 31 August 2021. We used an anti-SARS-CoV-2 enzyme-linked immunosorbent assay (ELISA) to detect IgG antibodies against the S1 domain of the SARS-CoV-2 spike (S) protein in serum according to the manufacturer's instructions (Anti-SARS-CoV-2-ELISA (IgG), EI 2606-9601 G, EUROIMMUN Medizinische Labordiagnostika AG, Lübeck, Germany). Processing and measurement were done using the fully automated Immunomat (Institut Virion\Serion GmbH, Würzburg, Germany). The results were determined by comparing the obtained signals of the patient samples with the previously obtained cutoff value of the calibrator. As suggested by the manufacturer, samples with a cutoff index ≥ 1.1 were considered positive. To exclude patients who had previously acquired COVID-19, we simultaneously measured antibodies against the nucleocapsid (N protein) using an electrochemiluminescence immunoassay (ECLIA, Elecsys Anti-SARS-CoV-2, 09203079190, Roche Diagnostics GmbH, Mannheim, Germany). The results were determined by comparing the obtained signals of the patient samples with the previously obtained cutoff value of the calibrator. As suggested by the manufacturer, samples with a cutoff index ≥ 1.0 were considered positive. Erythrocyte IMPDH activity was measured as previously described [17] . Briefly, lithium heparin blood was drawn immediately before morning drug intake (pre-dose). Blood was stored at room temperature until cell extracts were prepared (within 24 h). Whole blood (250 µL) was transferred to 15 mL polystyrene tubes filled with 2.5 mL phosphatebuffered saline. After centrifugation at 1200× g for 10 min at 20 • C, the supernatant was discarded, and the packed cells were washed with another 2.5 mL of phosphate-buffered saline and centrifuged under the same conditions. The sediment was lysed with 3 mL HPLC-water, and 1 mL aliquots were frozen at −20 • C for later use. IMPDH activity in erythrocytes was measured in the supernatant of thawed cell lysates. We used identical incubation conditions and the previously described HPLC method for mononuclear cells [16] . Xanthosine 5 -monophosphate production was normalized to the hemoglobin concentration. The internal quality control samples were included in each analytical run. The interassay variability for the three different activity levels (1713, 831, and 231 pmol XMP/h/mg hemoglobin) was 4.9%, 5.7%, and 11.3%, respectively. The most recent measurements under steady-state conditions before vaccination were used for the analysis. Descriptive analyses were performed to describe the parameters of interest using absolute and relative frequencies for categorical data and means with standard deviations or medians with ranges for continuously distributed parameters, as appropriate. The vaccination response rate was analyzed using a multilevel mixed-effects logistic regression model to account for the clustering of patients at the three sites of Charité-Universitätsmedizin Berlin. The medical therapies were divided into mutually exclusive groups. Grand mean coding was performed to analyze the association between vaccination response and therapies and to compare a single therapy with the mean vaccination response across all therapies. Reference-level coding was performed for the other categorical predictor variables. Calcineurin inhibitor (CNI) levels under treatment with tacrolimus or cyclosporine A were combined by calculating the percentage of deviation from the base 6 ng/mL for tacrolimus and 80 ng/mL for cyclosporine A. Receiver operating characteristic (ROC) analysis was used to evaluate the predictive ability of MPA dose, lymphocyte count, and IMPDH activity in relation to response to SARS-CoV-2 vaccination in patients treated with MPA. The area under the ROC curve (AUC) was compared using the statistical test proposed by DeLong et al. [19] . ROC analyses were used to estimate the cutoff for IMPDH using the Youden index. The resulting cutoff value maximizes the sum of the sensitivity and specificity to predict the response to vaccination. All statistical analyses were performed using STATA version 12.1. The algorithm used to establish the cohort is shown in Figure 1 . Based on the availability of serological, vaccination, and medication data in our electronic patient record system, we identified 148 responders and 415 non-responders to standard SARS-CoV-2 vaccinations. The baseline characteristics of the study cohort are summarized in Table 1 . The responders were younger and had better graft function. The interval between transplantation and vaccination was longer in responders. The gender distribution was similar between responders and non-responders. The application of two doses of BioNTech-Pfizer vaccine was dominant in our cohort. A small subset of the cohort was administered a heterologous vaccination (mRNA after vector vaccine). The patients had been administered Vaxzevria ® , and due to increasing safety concerns for the second application, an mRNA-based vaccine had been chosen. Among the non-responders, the majority received three-fold immunosuppression, whereas responders were similarly often two-and three-fold immunosuppressed. Tacrolimus in combination with MPA is the most common immunosuppressive regimen. Altogether, 49.5% of patients received immunosuppressive doses of glucocorticoids (≥5 mg prednisolone equivalent daily). Multivariate analyses were performed to determine the influence of baseline characteristics, vaccine types, immunosuppressive combinations, eGFR, lymphocyte counts, and CNI trough levels on response to SARS-CoV-2 vaccination (Table 2) . Higher age increased the risk of non-response to vaccination, whereas longer time after transplantation significantly increased the serological response rate. The selection of vaccine(s) influenced the success of vaccination. In our cohort, we found the highest immunogenicity with heterologous vaccine combinations. Immunosuppressive combinations containing MPA (tacrolimus + MPA ± steroid, cyclosporine + MPA ± steroid, and belatacept + MPA ± steroid) significantly decreased the odds of serological response after SARS-CoV-2 vaccination. In contrast, the absence of MPA or replacement of MPA with other immunosuppressants increased the probability of response. A higher eGFR, higher lymphocyte count, and lower CNI trough levels were also associated with higher response rates. However, steroids did not have convincing effects on humoral immune reactions after vaccination. For a detailed analysis of the determinants of vaccination response in patients receiving MPA, we analyzed a sub-cohort of 492 patients with MPA (Table 3) . 1 Odds ratio for increase by one unit, except for lymphocytes (by 100 units) and CNI trough levels (%deviation, by 5%). 2 AstraZeneca-BioNTech-Pfizer (n = 28), AstraZeneca-Moderna (n = 13). 3 Deviation in % from base 6 ng/mL for tacrolimus and 80 ng/mL for cyclosporine A trough levels. 3 Deviation in % from base 6 ng/mL for tacrolimus and 80 ng/mL for cyclosporine A. As before, younger age and a longer interval between transplantation and vaccination increased the probability of response in this subgroup. In addition, male gender had a positive impact on the response rate. The inclusion of MPA dose and IMPDH activity measurements in the analysis led to the loss of significance of the vaccine(s), most of the immunosuppressive combinations (except MPA + Tac + steroid), and lymphocyte counts with respect to the serological response rate. Higher eGFR and lower CNI trough levels moderately improved the humoral response to vaccination. Markers of MPA exposure, i.e., a lower MPA dose and lower erythrocyte IMPDH activity, were associated with significantly increased response rates. The correlation between IMPDH activity and MPA dose was weak (r = 0.17, p = 0.015). ROC analyses were performed to compare the influence of MPA dose and IMPDH activity on the prediction of response to SARS-CoV-2 vaccination (Figure 2 ). The need for better vaccination strategies led us to analyze our kidney transplant patients with respect to their humoral response after standard SARS-CoV-2 vaccination. For this purpose, we identified clearly defined responders and non-responders based on the detection of anti-spike IgG antibodies and sought factors with a significant influence in multivariate analyses. We focused on serological results in our cohort because of the broad availability of data and the lack of more sophisticated assays, such as neutralizing antibody assays and cellular immunity assays, in the majority of patients. Although the value of serological response after SARS-CoV-2 vaccination was initially questioned, German health authorities have accepted and recommended this monitoring in immunocompromised patients, including following solid organ transplantation [20] . The spectrum of assays used to monitor the immunological response to SARS-CoV-2 vaccination is broad, and recent studies have suggested significant differences between humoral and cellular immunity [21] . The question of protection against SARS-CoV-2 infection in relation to the extent of humoral responses after vaccination was not the focus of our analysis but is an important topic for future studies. Our serological approach is supported by the increasing burden of breakthrough infections [22, 23] and data on the kinetics of antibody responses over time (neutralizing antibodies and total antibody concentrations) for the estimation of the duration and degree of protection provided by vaccines [24, 25] . As expected, several patient characteristics have an impact on the serological response after vaccination: older patients, more recently grafted patients, and those with poorer kidney graft function are at a higher risk of non-response. Age and lower GFR were also predictive of fatal outcomes in a Brazilian cohort used to establish a tool for the early prediction of COVID-19-associated death after kidney transplantation [26] , and immunosenescence likely played a role in both observations [27] . Beyond patient age, the intensity of early immunosuppression with high rates of triple immunosuppression, higher MPA exposure, higher CNI trough levels, and various induction regimens with monoclonal or polyclonal antibodies (data not shown) favored non-response to vaccination. The lower rate of antibody formation after vaccination in patients with lower lymphocyte counts is in accordance with data from a recent French study [8] . Soon after the approval of mRNA-and vector-based SARS-CoV-2 vaccines, the concept of heterologous vaccination with both types of vaccines was discussed based on retrospective data from non-renal patients [28] . Data from a prospective study of healthcare workers at our institution further supports this concept [29] . The significantly higher response rate of KTPs vaccinated with heterologous combinations in our cohort is in accordance with the data from non-renal patients. Patients receiving immunosuppressive combinations including MPA had significantly lower response rates to SARS-CoV-2 vaccination than those without MPA. Among patients treated with MPA, 18.7% were responders, compared to 78.9% of patients without MPA. The magnitude of the problem was first described by Segev et al. in May 2021 using a social-media-based study of solid organ recipients [13] and subsequently confirmed in other studies [3, 6] . The combination of belatacept and MPA in not a single case enabled B lymphocytes to mount a measurable antibody response after vaccination in our cohort. In a French study, even the administration of a third dose of BNT162b2 mRNA COVID-19 vaccine did not improve immunogenicity in KTPs treated with belatacept [30] . The negative effect of proliferation inhibition with MPA on vaccination response led us to perform a more detailed analysis of the sub-cohort of MPA-treated patients. In the multivariate analysis, age, time interval between transplantation and vaccination, and GFR remained significant predictors of response, whereas vaccine combinations and most immunosuppressive regimens lost their influence. The significant disadvantage of female patients probably reflects a higher susceptibility to MPA effects. The MPA dose as a marker of short-term exposure and IMPDH activity as a marker of medium-term MPA exposure clearly dominated the results of the multivariate analysis. In a head-to-head comparison between MPA dose and IMPDH activity, the latter proved to be a significantly better predictor of serological response after complete SARS-CoV-2 vaccination in KTPs. These data suggest that dose modification of MPA is a strategy to improve response rates after vaccination, supporting the conclusions of other studies [31] . Our study had some limitations. This was a retrospective analysis based solely on serological monitoring after only two vaccine applications. The cohort represents a broad spectrum of patient characteristics, vaccines, and immunosuppressive regimens, rendering the analysis a real-world study. In contrast, the classification of patients into two response groups resulted from a strict selection process. The number of patients with respect to lymphocyte counts was low (62.9%), and IMPDH monitoring was incomplete (348 of 492 patients). After 31 August 2021 (deadline for the database retrieval), several patients were vaccinated a third time (usually without prior reduction in immunosuppression) or even a fourth time (usually with reduced immunosuppression). The former did not significantly improve the response rate and therefore has not been included in this analysis. The results of the latter are still not available (short followup). In summary, elderly patients with low GFR are at high risk of non-response to standard SARS-CoV-2 vaccination. Strong immunosuppression or overimmunosuppression should be avoided to ensure successful vaccination. MPA and belatacept are highly potent in preventing humoral immune responses to vaccination. Heterologous vaccination strategies elicited the highest response rates in our cohort. Tailoring MPA-based immunosuppression using IMPDH monitoring may be a promising strategy for successful vaccination against SARS-CoV-2 after kidney transplantation. Another strategy might be the complete perivaccination cessation of MPA, as already done in non-renal patients [32] and planned in a prospective study with KTPs in Israel [33] . Our data suggest a risk-based stratification to obtain a sufficient immune response to SARS-CoV-2 vaccination and an effective immunosuppression in KTPs: First, given the observed lack of response in patients treated with belatacept, the drug should not be initiated unless patients are immunized against SARS-CoV-2. Second, in patients already treated with belatacept, a temporary switch to another immunosuppressive strategy before the first vaccination attempt is warranted. If this is impossible, a passive vaccination strategy might be advisable. Third, in patients treated with dual immunosuppression (e.g., CNI plus steroids), an attempt to achieve a vaccination response without prior modification of immunosuppression is warranted. Fourth, in patients treated with MPA, dose reduction and the best available TDM appear to be important for optimizing the vaccination response. In addition, given the association of dose reductions or discontinuation of mycophenolate with an increased risk of acute rejection and inferior patient and graft survival [34] , a precise TDM may increase safety and should be the basis of such maneuvers. To our knowledge, this is the first report to show that erythrocyte IMPDH activity measurements are helpful in predicting vaccination responsiveness in MPA-treated patients. Given the reduced mortality of Omicron and new therapeutic options for COVID-19, one should balance the risks of reduction in immunosuppression for successful vaccination against risks of SARS-CoV-2 infection of vulnerable patients. Informed Consent Statement: Written informed consent was obtained from all patients. This study was conducted in accordance with the guidelines of Good Clinical Practice (GCP) and the Declaration of Helsinki. The data presented in this study are available on request from the corresponding author. The data are not publicly available. Factors associated with COVID-19-death in 17 million patients using OpenSAFELY COVID-19-related mortality in kidney transplant and dialysis patients: Results of the ERACODA collaboration Low immunization rates among kidney transplant recipients who received 2 doses of the mRNA-1273 SARS-CoV-2 vaccine Impaired humoral and cellular immunity after SARS-CoV-2 BNT162b2 (tozinameran) prime-boost vaccination in kidney transplant recipients Impaired humoral immunity to SARS-CoV-2 BNT162b2 vaccine in kidney transplant recipients and dialysis patients Humoral and cellular immunity to SARS-CoV-2 vaccination in renal transplant versus dialysis patients: A prospective, multicenter observational study using mRNA-1273 or BNT162b2 mRNA vaccine Cellular and humoral immune response after mRNA-1273 SARS-CoV-2 vaccine in liver and heart transplant recipients Three doses of an mRNA Covid-19 vaccine in solid-organ transplant recipients B and T Cell responses after a third dose of SARS-CoV-2 vaccine in kidney transplant recipients Cellular and humoral immune responses after three doses of BNT162b2 mRNA SARS-Cov-2 vaccine in kidney transplant Antibody response after a third dose of the mRNA-1273 SARS-CoV-2 vaccine in kidney transplant recipients with minimal serologic response to 2 doses Antibody response to a fourth dose of a SARS-CoV-2 vaccine in solid organ transplant recipients: A case series Antibody Response to 2-Dose SARS-CoV-2 mRNA Vaccine Series in Solid Organ Transplant Recipients Impact of distinct therapies on antibody response to SARS-CoV-2 vaccine in systemic lupus erythematosus Personalized Therapy for Mycophenolate: Consensus Report by the International Association of Therapeutic Drug Monitoring and Clinical Toxicology Improved assay for the nonradioactive determination of inosine 5 -monophosphate dehydrogenase activity in peripheral blood mononuclear cells Inosine 5 -monophosphate dehydrogenase activity for the longitudinal monitoring of mycophenolic acid treatment in kidney allograft recipients TBase-An Integrated Electronic Health Record and Research Database for Kidney Transplant Recipients Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach STIKOEmpfehlung zur COVID-19-Impfung bei Personen mit Immundefizienz und die dazugehörige wissenschaftliche Begründung Cellular immunity predominates over humoral immunity after homologous and heterologous mRNA and vector-based COVID-19 vaccine regimens in solid organ transplant recipients SARS-CoV-2 infection after full vaccination in kidney transplant recipients Risk of breakthrough SARS-CoV-2 infections in adult transplant recipients Antibody titers and protection against a SARS-CoV-2 infection Correlates of protection against symptomatic and asymptomatic SARS-CoV-2 infection Development and validation of a simple web-based tool for early prediction of COVID-19-associated death in kidney transplant recipients Aging and the immune response to organ transplantation Heterologous prime-boost vaccination with ChAdOx1 nCoV-19 and BNT162b Safety, reactogenicity, and immunogenicity of homologous and heterologous prime-boost immunisation with ChAdOx1 nCoV-19 and BNT162b2: A prospective cohort study Weak antibody response to three doses of mRNA vaccine in kidney transplant recipients treated with belatacept Intensity of mycophenolate mofetil treatment is associated with an impaired immune response to SARS-CoV-2 vaccination in kidney transplant recipients Temporary hold of mycophenolate augments humoral response to SARS-CoV-2 vaccination in patients with rheumatic and musculoskeletal diseases: A case series Immunosuppression reduction when administering a booster dose of the BNT162b2 mRNA SARS-CoV-2 vaccine in kidney transplant recipients without adequate humoral response following two vaccine doses: Protocol for a randomised controlled trial (BECAME study) Tolerability of mycophenolate sodium in renal transplant recipients The authors sincerely thank Charlotte E.L. Hammett and Annika Jens for their technical assistance in the vaccination process. The authors declare no conflict of interest.