Inclusion criteria included patients who had been followed-up for more than 10 years after KT; no history of rejection; estimated glomerular filtration rate (GFR) more than 50 mL/min/1.73 m² via modification of diet in renal disease (MDRD) with renal function change within 10% of baseline value in the past year; no immunosuppressive agent change during the past year; and proteinuria less than 0.3 protein (g)/creatinine (g) in spot urine. This study was approved by the Institutional Review Board (KC11OIS0917) of the Seoul St. Mary’s Hospital, and all the patients provided written informed consent. All clinical investigations were conducted according to the principles expressed in the Declaration of Helsinki. The study is registered at www.ClinicalTrials.gov (Identifier: NCT01592253).

We started patients on 2 or 3 mg/day of SRL and reduced CNI dose to 50% of the baseline value at the same time. After 2 weeks, we stopped CNIs entirely and titrated the SRL dose to maintain the blood levels between 3 and 8 ng/mL. We discontinued other immunosuppressive agents such as enteric-coated mycophenolate sodium, azathioprine, or mizoribine till 3 months after conversion, but maintained low-dose steroids [12].

We evaluated the safety of SRL conversion using allograft function (estimated GFR via MDRD) at 2, 6, 12, and 24 weeks after conversion. Acute rejection was suspected if serum creatinine increased to more than 30% of the baseline value after SRL conversion. Rejection was confirmed via biopsy. Graft loss was defined as declining graft function that required dialysis, retransplantation, or death with functioning graft. We measured trough level of CNIs at baseline and 2 weeks after conversion. We checked SRL levels and performed laboratory tests including complete blood count, blood chemistry, urinalysis, and urine chemistry at 2, 6, 12, and 24 weeks after conversion.

We sampled peripheral blood before and 6 months after conversion to SRL. Peripheral blood mononuclear cells (PBMCs) were isolated, and cell cultures were performed as described previously [13]. Flow cytometric study of PBMCs was performed to analyze human intracellular cytokine production. The PBMCs were stimulated with phorbol myristate acetate and ionomycin in the presence of GolgiStop (BD Biosciences, San Diego, CA, USA) for 4 hours. For surface staining, cells were stained with combinations of the following monoclonal antibodies: CD4-PE/Cy7 (BioLegend, San Diego, CA, USA); CD45RAFITC (BD Pharmingen, San Diego, CA, USA); CD8-APC (Pharmingen), and CD25-APC (Pharmingen). Staining for chemokine receptor CCR7 was performed using monoclonal antibodies to it. Cells were washed, fixed, permeabilized, and stained to detect intracellular cytokines with monoclonal antibodies specific to interferon γ (IFN-γ, eBioscience); interleukin 4 (IL-4, eBioscience); IL-17 (FITC, eBio64DEC17, IgG1, κ; eBioscience); Foxp3 (FITC, PCH101, IgG2a, κ; eBioscience); IFN-γ (PE, B27, IgG1, κ; Pharmingen), and IL-10-APC (JES3-19F1, IgG2a, κ; Pharmingen). Appropriate isotype controls were used for gate setting for measuring cytokine expression. Cells were analyzed on a FACS Calibur flow cytometry system (BD Biosciences).

During each outpatient clinic visit, we examined adverse events that developed after SRL conversion. These included mucositis, folliculitis, dermatitis, or any signs of infection. We monitored the following: development of proteinuria as spot urine protein-to-creatinine ratio (PCR), diabetes mellitus via fasting glucose level, dyslipidemia via fasting lipid level or lipid lowering agents, leukopenia, anemia, thrombocytopenia, increasing aspartate or alanine aminotransferase, and requirement of antihypertensive medication. We defined the patient’s situation as tolerable when the patient had successfully received SRL for 6 months without any serious adverse events.

Statistical analysis was performed using SPSS version 18.0 (SPSS Inc., Chicago, IL, USA). The comparison of numerical variables at baseline, 2, 6, 12, and 24 weeks from conversion was performed using the Friedman test. The comparison of numerical variables before and 24 weeks after conversion was performed using the Wilcoxon signed-rank test. The results are presented as mean ± standard deviation or standard error, counts and percentages, or median with ranges depending on the data type. The p values < 0.05 were considered significant.

A total of 45 KT patients met the inclusion criteria in Seoul St. Mary’s hospital. Among them, 14 patients (31%) agreed to participate in this study, and they were ultimately enrolled. The mean duration from KT to initiation of the study was 14.7 ± 2.9 years (median, 13.5; range, 12 to 19.5). The mean MDRD-GFR was 72.2 ± 20.2 mL/min/1.73 m². Only one patient was highly sensitized (her flow cytometric crossmatch was positive for B cells), so she had taken plasmapheresis three times before KT. Other baseline patient characteristics are presented in Table 1. Of the 14 patients, eight (57%) were receiving cyclosporine (CsA) with a median dosage of 150 mg (range, 75 to 175); the mean blood trough level was 95.8 ± 58.4 ng/mL. The remaining patients (n = 6, 43%) were receiving tacrolimus (TAC) with a median dosage of 2 mg (range, 2 to 4); the mean blood trough level was 4.2 ± 2.1 ng/mL. The patients’ immunosuppressive regimens are also shown in Table 1.

The 6-month conversion success rate was 64.2% (n = 9), while 35.8% of patients (n = 5) ceased receiving SRL. Renal allograft function at baseline and 2, 6, 12, and 24 weeks after SRL conversion was not significantly different (71.2 ± 17.6, 71.0 ± 17.1, 72.0 ± 16.7, 71.2 ± 17.6, and 67.7 ± 14.6 mL/min/1.73 m² via MDRD-GFR respectively), indicating good graft function preservation after SRL conversion (p = 0.115) (Fig. 1). There was no acute rejection or graft loss after SRL conversion. Fig. 2 shows trough levels of TAC and CsA before and 2 weeks after SRL conversion and SRL trough levels at 2, 6, 12, and 24 weeks after SRL conversion. The 50% dosage reduction of CsA or TAC caused the trough level to decrease by 48.7% and 27.2%, respectively (Fig. 2A and 2B). Levels of SRL were variable in the early period of conversion but gradually stabilized and reached the target level (3 to 8 ng/mL) by 12 weeks post-conversion (mean trough level, 7.4 ng/mL). The levels were not different statistically (p = 0.919), but showed decreasing pattern throughout the study periods (Fig. 2C).

Figs. 3 and 4 show the immunologic changes in six of the nine successfully converted patients. Three of them had received CsA, while the remaining three had received TAC before SRL conversion.

The percentage of CD4⁺ T_reg increased at 24 weeks after SRL conversion (6.3% ± 3.2% vs. 8.6% ± 5.4%, p = 0.028) (Fig. 3A). There were no differences in the percentages of Th1 (23.4% ± 9.8% vs. 24.1% ± 10.4%, p = 0.917), Th2 (3.5% ± 4.2% vs. 1.3% ± 1.2%, p = 0.6), or Th17 (1.3% ± 0.6% vs. 2.1% ± 1.1%, p = 0.249) (Fig. 3B-3D). The percentage of CD8⁺ T_naïve (44.0% ± 13.8% vs. 39.8% ± 9.0%, p = 0.462), CD8⁺ central memory T cells (TCM) (27.1% ± 10.5% vs. 30.6% ± 12.7%, p = 0.753) or CD8⁺ effector memory T cells (T_EM) (13.0% ± 9.3% vs. 15.3% ± 4.2%, p = 0.6) also did not alter after SRL conversion (Fig. 4A-4C). The percentage of IFN-γ producing CD8⁺ T_naïve (59.7% ± 26.1% vs. 50.2% ± 41.8%, p = 0.6) and T_CM (57.4% ± 15.1% vs. 38.1% ± 24.7%, p = 0.075) did not change (Fig. 4D and 4E), but the percentage of IFN-γ producing CD8⁺ T_EM reduced after conversion (58.5% ± 18.9% vs. 37.9% ± 25.4%, p = 0.046) (Fig. 4F). No significant difference was seen between B cell subsets. The percentage of memory B cells (22.6% ± 11.5% vs. 30.9% ± 11.1%, p = 0.345), mature B cells (39.1% ± 9.8% vs. 35.1% ± 11.3%, p = 0.293), and immature B cells (5.1% ± 3.6% vs. 9.7% ± 9.3%, p = 0.345) did not change 24 weeks after SRL conversion.

Total cholesterol, low-density lipoprotein, and triglyceride levels increased after SRL conversion (178 ± 22 mg/dL vs. 223 ± 58 mg/dL, p < 0.001; 97 ± 28 mg/dL vs. 149 ± 45 mg/dL, p < 0.001; and 94 ± 38 mg/dL vs. 114 ± 57 mg/dL, p = 0.018; respectively). Urine PCR also increased after conversion (0.08 ± 0.08 vs. 0.39 ± 0.40, p = 0.026). Of the 14 patients, 11 (79.6%) experienced mucocutaneous adverse events. Oral mucositis (n = 7), dermatitis (n = 4), facial and leg edema (n = 3), folliculitis (n = 1), acne (n = 1), and diarrhea (n = 1) were all recorded. Most events occurred within 6 weeks after conversion and led to withdrawal. The main cause of discontinuation was intolerable oral mucositis (n = 3). Other causes were facial dermatitis (n = 1) and diarrhea (n = 1). Development of leukopenia, anemia, thrombocytopenia, abnormality of liver function tests, increase of fasting glucose, or infections was not observed.