Weaning Immunosuppressant in Patients with Failing Kidney Grafts and The Outcomes: A Single-Center Retrospective Cohort Study

An immunosuppressant weaning protocol in failing allografts has not yet been established. Maintaining immunosuppressants would preserve residual renal function (RRF) and prevent graft intolerance syndrome and sensitization but would increase the risks of infection and malignancy. In this study, graft failure cases after kidney transplantation in a single center were reviewed retrospectively. The outcome differences in all-cause mortality, infection-related hospitalization, cancer, graft intolerance syndrome, re-transplantation, and RRF duration between the immunosuppressant maintaining and weaning groups 6 months after graft failure were compared. Among the weaning group, the outcome differences according to low-dose steroid use were also compared at 6 and 12 months. In a total of 131 graft failure cases, 18 mortalities, 42 infection-related hospitalizations, 22 cancer cases, 11 graft intolerance syndrome cases, and 28 re-transplantations occurred during the 94-month follow-up. Immunosuppressant maintenance significantly decreased the patient survival rate 6 months after graft failure compared with weaning (log-rank P = 0.008) and was an independent risk factor for mortality, even after adjustments (hazard ratio, 3.01; P = 0.025). Infection-related hospitalization, graft intolerance syndrome development, and re-transplantation were not affected by the immunosuppressant weaning protocol. Among the immunosuppressant weaning group, low-dose steroid maintenance at 6 and 12 months helped preserved RRF (P = 0.008 and P = 0.003, respectively).

has several advantages, such as reduction of graft intolerance syndrome occurrence 15,16 , maintenance of residual renal function (RRF) 17 , and prevention of sensitization from allografts, which could affect the outcomes of the next KT 18,19 . On the contrary, immunosuppressant maintenance could elevate the risk of infection and cardiovascular events 16,20 . Additionally, long-term use of immunosuppressants may be associated with the development of cancer 21,22 and secondary adrenal insufficiency 23 . Therefore, a balance between the advantages and disadvantages of the use of immunosuppressants is critical in patients with failing grafts to improve their survival and the outcomes of the next KT.
However, only a few studies have been conducted in patients with failing grafts, the majority of which are retrospective cohort studies with a small number of participants and performed in non-Asian populations 16,20,24 . In this study, we aimed to explore the evidence on when and how immunosuppressants should be weaned in recipients with failing grafts.

Materials and Methods
Study subjects. This retrospective study was conducted in a single tertiary hospital. Among a total of 2,121 KTs performed in Seoul National University Hospital from January 1984 to January 2016, 465 patients lost their allograft function permanently. Allograft failure was defined as the requirement of maintenance renal replacement therapy owing to deteriorated allograft function. We excluded recipients who were aged under 19 years at the time of transplantation (n = 110), had allograft failure or mortality before 1999 (as their data could not be extracted from the electronic medical records) (n = 119), graft failure within a month after transplantation (n = 8), mortality within a month after graft failure (n = 54), and those who were lost to follow-up after transplantation (n = 43). After the exclusion of these 334 patients, a total of 131 patients were analyzed in this study ( Fig. 1). This study was approved by the Institutional Review Board (IRB) of Seoul National University Hospital (IRB No. 1805-108-947) and performed in accordance with the recent guideline of the Declaration of Helsinki. Written consent was waived by the IRB because of the retrospective nature of the study with minimal risk to the study subjects. clinical parameters. Electronic medical records were reviewed retrospectively. Data on the underlying disease, including diabetes and hypertension, cause of ESRD, donor type, and history of transplantation were gathered. In addition, we reviewed the duration of graft functioning and cause of allograft failure. Information on immunosuppressive treatment, including the use of steroids, before and after allograft failure was obtained.
Definitions. We divided our subjects into two groups according to their immunosuppressant use 6 months after allograft failure as follows: immunosuppressant maintaining and immunosuppressant weaning groups. The immunosuppressant maintaining group consisted of patients receiving single steroid therapy with an equivalent dose of ≥10 mg per day of prednisolone and those using more than two kinds of concurrent immunosuppressants, including low-dose steroid with calcineurin inhibitors (CNIs) or antimetabolites, 6 months after graft failure. The immunosuppressant weaning group consisted of patients whose immunosuppressive treatment was discontinued and those receiving single steroid therapy with an equivalent dose of <10 mg per day of prednisolone (low-dose steroid therapy) 6 months after graft failure. To assess the beneficial effect of low-dose steroid use, we further subdivided the immunosuppressant weaning group according to the maintenance of low-dose steroid therapy 6 and 12 months after allograft failure as follows: steroid stopped group and steroid maintaining group. outcomes. We reviewed the following outcome events classified into three groups: 1) all-cause mortality, 2) preferred immunosuppressant withdrawal outcomes (infection-related hospitalization and cancer occurrence), and 3) preferred immunosuppressant maintenance outcomes (graft intolerance syndrome development, re-transplantation, and RRF duration, defined as the duration of diuretic therapy after graft failure).
Weaning immunosuppressants and its impact on clinical outcomes. The weaning protocol varied among the patients. CNIs were weaned before antimetabolites in 42 (32.1%) patients, antimetabolites before CNIs in 62 (47.3%), and both CNIs and antimetabolites simultaneously in 24 (18.3%). In most cases, the steroid was weaned last, except in 1 patient wherein CNIs were weaned last.
In the comparison between the immunosuppressant maintaining and weaning groups 6 months after graft failure, there was no significant difference in the baseline characteristics, clinical outcomes, renal replacement therapy modality after graft failure, and duration of patient survival, diuretic use, and follow-up after graft failure.  Table 2).
The survival analysis using the Kaplan-Meier curves and log-rank test also showed significantly lower survival rates in the immunosuppressant maintaining group than in the immunosuppressant weaning group (log-rank P = 0.008). Moreover, the elevated mortality risk remained even after adjustment for sex, age at the time of graft failure, donor type, presence of diabetes or hypertension, re-transplantation, and dialysis modality after graft www.nature.com/scientificreports www.nature.com/scientificreports/ failure (adjusted hazard ratio, 3.01; 95% confidence interval, 1.15-7.88; P = 0.025). However, there were no significant differences in the other preferred immunosuppressant withdrawal outcomes, such as infection-related hospitalization (log-rank P = 0.914), nor in the preferred immunosuppressant maintaining outcomes, such as graft intolerance syndrome occurrence (log-rank P = 0.445) and re-transplantation (log-rank P = 0.838), between the two groups (Fig. 3).
Effects of low-dose steroid maintenance 6 and 12 months after graft failure. The subgroup analysis among the immunosuppressant weaning group was conducted according to the duration of steroid therapy. Among the 109 patients in whom immunosuppressants were weaned 6 months after graft failure, 71 (54.2%) stopped taking steroids, while 38 (34.9%) continued taking them in low dose. Twelve months after graft failure, 91 (69.5%) patients stopped taking steroids, while 18 (16.5%) still received low-dose steroid therapy (Fig. 2). There was no significant difference in the outcomes after graft failure, including all-cause mortality, infection-related hospitalization, post-transplantation cancer occurrence, graft intolerance syndrome occurrence, and re-transplantation between the steroid weaning and maintaining groups both 6 and 12 months after graft failure. The steroid weaning group had a higher incidence of nephrectomy due to graft intolerance syndrome both 6 and 12 months after graft failure than the steroid maintaining group, although the difference was not significant (P = 0.158 and P = 0.351, respectively). The duration of diuretic therapy was longer in the steroid maintaining group both 6 (P = 0.008) and 12 months after graft failure (P = 0.003) ( Table 3).

Discussion
In this study, we discovered that maintaining immunosuppressants 6 months after graft failure elevated the risk of all-cause mortality approximately three-fold compared with weaning immunosuppressants, even after adjusting for other confounding factors, although the other outcomes were not significantly affected. Conversely, we suggest that maintaining low-dose steroids until 12 months after graft failure could preserve RRF, which was based on the duration of diuretic therapy. Based on these findings, CNIs and antimetabolites may be weaned within 6 months after graft failure, and low-dose steroids may be maintained up to 12 months after graft failure for survival improvement and RRF preservation in patients who have lost their allograft function. Maintaining immunosuppressants after graft failure has both advantages and disadvantages. It influences the outcomes of patients with failing grafts in both positive and negative aspects. An appropriate immunosuppressant weaning protocol is important to balance its positive and negative effects and consequently improve the overall outcomes of patients with allograft failure. There are only a few recommendations and guidelines regarding immunosuppressive therapies in patients with failing grafts 3,23,25 . In addition, there has been no definite immunosuppressant weaning protocol in KT recipients until recently. The consensus from currently available guidelines and recommendations suggest weaning immunosuppressants 6 months after graft failure, especially in patients with minimal RRF. It is recommended to taper steroids carefully and gradually while monitoring patients' symptoms for adrenal insufficiency and rejection 25 . Our study findings also support the recommendation of early withdrawal of immunosuppressants within 6 months after graft failure. However, our data suggest that maintaining steroids in low doses (equivalent dose of <10 mg per day of prednisolone) up to 12 months can be beneficial in preserving RRF.
In our study, immunosuppressant maintenance 6 months after graft failure increased the risk of all-cause mortality even after adjusting for other confounding factors. In a previous study that used USRDS data, the main cause of mortality in patients with graft failure was cardiovascular problems and infections 8 . In other studies, www.nature.com/scientificreports www.nature.com/scientificreports/ increased risks of infection and hospitalization were associated with immunosuppressant maintenance after graft failure 16,20 . The study also showed that the main cause of mortality was cardiovascular complications (44.4%), followed by infections (22.2%). However, the other outcomes related to the adverse effects of immunosuppressants, including infection-related hospitalization and post-KT cancer occurrence, did not significantly differ between the immunosuppressant maintaining and weaning groups. This discrepancy between our results and those of previous studies might be attributed to the small number of both infection-related hospitalization (32.1%) and study subjects, especially in the immunosuppressant maintaining group 6 months after graft failure; there were only 22 (16.8%) patients included. Further, among the 42 patients who had infection-related hospitalization in this study, 16 (38.1%) had more than two hospitalizations. As the survival analysis was conducted using the first hospitalization data, the increased risk of infection in immunosuppressant maintenance might have been devaluated. The rates of graft intolerance syndrome occurrence and nephrectomy, which are considerable side effects of immunosuppressant weaning, did not differ between the immunosuppressant weaning and maintaining groups in this study.
One of the strengths of this study is that a subgroup analysis was conducted among the immunosuppressant weaning group according to low-dose steroid maintenance 6 and 12 months after graft failure. In the subgroup analysis, the steroid maintaining group 6 and 12 months after graft failure showed a significantly longer use of diuretics, which was also interpreted as the duration of RRF. It is well known that RRF is important in improving the survival and quality of life of incidence dialysis patients with naïve kidneys 8,9,13,26 . The importance of RRF on survival has also been reported in peritoneal dialysis patients with graft failure 17 . However, in this study, no survival benefit was observed in the steroid maintaining group, although RRF was preserved longer in this group than in the immunosuppressant weaning group. Additionally, the rate of nephrectomy due to graft intolerance syndrome was lower in the steroid maintaining group 6 and 12 months after graft failure than in the immunosuppressant weaning group although the difference was not significant (P = 0.158 and P = 0.351, respectively). Furthermore, maintaining low-dose steroids did not increase the adverse outcomes of mortality, infection-related hospitalization, and post-KT cancer occurrence. Therefore, maintaining low-dose steroids until 12 months after graft failure may have beneficial effects on RRF without increasing adverse events.
None of the patients had nephrectomy due to graft intolerance syndrome in the steroid maintaining group 12 months after graft failure, although this number did not significantly differ with that in the steroid stopped group (n = 9 patients, 9.9%). To determine the risk factor for graft intolerance syndrome in our study patients, we compared the immunosuppressant weaning protocol between the patients with (n = 11, 8.4%) and without graft intolerance syndrome (n = 120, 91.6%). There was no significant difference in the immunosuppressant and steroid weaning protocols between the two groups (Table S1). As the number of graft intolerance syndrome cases in our study was relatively too small, we could not conclude whether immunosuppressant weaning could increase the risk for graft intolerance syndrome after graft failure.
This study has a few limitations. Among the 465 patients with graft failure, we could only investigate 131 (28.2%) patients in this study. Further, the number of analyzed patients and events was small, which might have lessened the statistical power. The dose of CNI or antimetabolites after graft failure was not included in the analysis which could be a confounding factor. The direct residual urine output could not be assessed; instead, RRF was considered based on the duration of diuretic therapy after graft failure since diuretics were only prescribed when the patients had residual urine output in the studied center. We could also not assess the effects of immunosuppressant weaning on allosensitization, owing to the retrospective nature of the study. Only 19% of the patients with graft failure had sensitization data both before and after the graft failure. However, this study also has strengths, one of which is that it was conducted in a single-center where the study population received relatively homogeneous medical management. Further, the various outcomes of the patients with graft failure (i.e., preferred immunosuppressant withdrawal and maintaining outcomes) were reviewed in detail. By conducting a subgroup analysis according to steroid use, the possible beneficial effect of low-dose steroid  www.nature.com/scientificreports www.nature.com/scientificreports/ maintenance could be reported. Nevertheless, larger prospective cohort studies are needed to establish evidence for optimizing and personalizing immunosuppressant weaning protocols to improve the outcomes of patients with graft failure.

conclusion
Considering the increasing number of patients with allografts and their improving survival, balancing the favorable and adverse effects of immunosuppressants is crucial in patients with graft failure to improve their outcomes and prepare them for their next KT. Along with current recommendations, our data suggest that immunosuppressants should be tapered within 6 months after graft failure. However, patients with RRF can benefit from low-dose steroid maintenance until 12 months after graft failure without increasing adverse outcomes. Further larger prospective cohort studies are required to establish evidence for developing immunosuppressant weaning protocols in graft failure patients.

Data availability
All produced data are available as upon request.