Abstract

Objective To assess the long-term safety and effectiveness of tacrolimus as maintenance therapy in patients with lupus nephritis (LN) receiving treatment in real-world clinical settings in Japan.

Methods An open-label, noncomparative, observational, prospective postmarketing surveillance study was conducted in 1395 patients with LN receiving maintenance treatment with tacrolimus at 278 medical institutions across Japan over a period of 10 years. Tacrolimus continuation rate and cumulative incidence of adverse drug reactions (ADRs), relapse, progression to renal failure, and progression to dialysis were calculated using Kaplan-Meier analysis.

Results Safety data were available for 1355 patients, almost half (49.3%) of whom remained on tacrolimus for the full 10 years of follow-up. A significant reduction in mean (SD) daily oral corticosteroid dose was observed from 16.0 (9.7) mg/day at 4 weeks after initiation of tacrolimus treatment to 7.2 (4.4) mg/day at year 10 (P < 0.001). The most frequently reported serious ADRs were infections (reported for 131 [9.7%] patients). Except for infections, no marked increase in the incidence of any other ADRs was seen over time, including renal impairment, malignant tumors, and cardiac dysfunction. Renal function was generally well maintained over the 10 years of follow-up. At year 10, cumulative rates of relapse, renal failure, and dialysis were 44.5%, 12.2%, and 4.5%, respectively.

Conclusion Tacrolimus was effective and generally well tolerated as maintenance therapy for LN in a large cohort of patients in Japan followed for 10 years, almost half of whom remained on therapy for the entire duration of follow-up. (ClinicalTrials.gov: NCT01410747)

Key Indexing Terms:

Systemic lupus erythematosus (SLE) is a chronic multisystem autoimmune disease, predominantly affecting women of reproductive age.1,2 Asian populations are also disproportionately affected,1-3 with a reported prevalence of SLE in Japan of up to 38 per 100,000.3 SLE has a heterogeneous presentation, with clinical manifestations ranging from skin rashes to major organ involvement.1 Lupus nephritis (LN) is one of the most severe manifestations of SLE,4 and is associated with increased morbidity and mortality in patients with SLE.4-6 Approximately 5% to 20% of patients with LN develop end-stage kidney disease within 10 years of diagnosis.4 LN appears to occur more frequently in patients with SLE of Asian or African descent than in other ethnic groups.5,7-10 LN also tends to be more severe and associated with poorer outcomes in these patient populations, and LN constitutes an important cause of chronic renal failure in Asia.10-13

Treatment of LN aims to prevent disease flares and preserve kidney function, thereby limiting disease progression and preventing the need for dialysis and transplantation.12,14 Current treatment guidelines generally recommend corticosteroids in combination with cyclophosphamide (CYC) or mofetil mycophenolate (MMF) as induction therapy in patients with LN, followed by ongoing maintenance therapy with low-dose oral corticosteroids combined with MMF or azathioprine (AZA).12,14

The calcineurin inhibitor, tacrolimus, is approved for the treatment of LN in Japan and other Asian countries. Tacrolimus blocks T-cell activation, thus suppressing autoantibody production and attenuating glomerular deposition of immune complexes.15,16 Tacrolimus has also demonstrated direct effects on podocytes, including stabilization of the actin cytoskeleton and inhibition of podocyte apoptosis, which may contribute to preservation of kidney function in patients with LN.16 Tacrolimus, in combination with corticosteroids, has been shown to be effective and well tolerated as induction and maintenance treatment in patients with LN in short-term clinical trials.17-22 Tacrolimus also appears effective and well tolerated when used in combination with MMF and corticosteroids as a multitarget therapy in patients with LN, particularly in those with severe or refractory disease.23-28 However, robust data regarding the long-term safety and effectiveness of tacrolimus as maintenance therapy in patients with SLE who have LN are lacking.

A Study to Evaluate the Safety and Efficacy of Tacrolimus for Lupus Nephritis Under Actual Use Situations (TRUST)29 was a postmarketing surveillance study undertaken in Japan to assess the long-term safety and effectiveness of tacrolimus as maintenance therapy in patients with LN receiving treatment in the real-world clinical setting. Results of a planned 5-year interim analysis have been reported previously.29 This paper presents the results of the final analysis at 10 years of follow-up.

METHODS

Study design and patients. This open-label, noncomparative, observational, prospective postmarketing surveillance study was conducted in patients with LN who started maintenance treatment with tacrolimus at 278 medical institutions across Japan between January 2007 and January 2010 (ClinicalTrials.gov: NCT01410747). The study design has been reported in detail previously.29 Briefly, eligible patients were registered into a central data collection system and were followed at planned visits for up to 10 years (data cut-off date: January 2020).

Surveillance was conducted under conditions of routine clinical practice in accordance with the ordinance for Good Post-Marketing Study Practice issued by the Japanese Ministry of Health, Labor and Welfare (MHLW). The study protocol and all subsequent amendments were submitted to the MHLW, and written agreement was obtained from all participating institutions prior to patient enrollment. As patient data were anonymized, it was not necessary to obtain informed consent from patients. This approach was compliant with Japanese regulations for postmarketing surveillance studies.

Tacrolimus treatment. Patients initiated treatment with tacrolimus (immediate-release formulation; Astellas Pharma, Inc., Tokyo, Japan) at an individualized dose and were then maintained on tacrolimus with dose adjustment, if required, according to clinical signs and symptoms and tacrolimus blood trough concentrations (determined according to local standard practice). The treating physician judged whether concomitant prednisolone was necessary.

Data collection. Patient data were collected at the time of treatment initiation (baseline) and then at planned visits over the 10 years of follow-up at 4, 12, and 28 weeks, 12, 18, and 24 months, and annually thereafter. Safety, steroid dose reduction, and disease relapse were evaluated during tacrolimus exposure. Progression to renal failure and dialysis were evaluated at 10 years, regardless of whether a patient was exposed to tacrolimus at the time.

Safety. Adverse events (AEs) were recorded by system organ class and preferred terms from the Japanese edition of the Medical Dictionary for Regulatory Activities (MedDRA/J; version 24.0). Adverse drug reactions (ADRs) were defined as any AE considered to be at least possibly related to treatment with tacrolimus based on the assessment of the treating physician or for which assessment of causality was missing. An ADR was considered serious if it resulted in death, was life-threatening, required hospitalization or prolongation of existing hospitalization, or resulted in persistent or significant disability/incapacity.

ADRs considered to be of special interest during treatment with tacrolimus were infections, renal disorders, glucose tolerance disorders, neuropsychiatric disorders, cardiac dysfunction, pancreatic dysfunction, malignancy (including lymphoma), worsened interstitial pneumonia, and menstrual disorders. Renal and hepatic impairment and cardiac function were rated as normal or abnormal as defined in the study protocol, with renal failure defined as a change in renal function meeting stage 2 or higher chronic kidney disease (Supplementary Data S1, available with the online version of this article).

Effectiveness. Effectiveness was assessed by cumulative rates of relapse, and progression to renal failure and dialysis, based on physician assessment according to local standard clinical practice using the definition of renal failure provided in the study protocol (Supplementary Data S1, available with the online version of this article). Changes from baseline for routine clinical variables of renal function were also assessed, including serum creatinine levels and estimated glomerular filtration rate (eGFR), calculated as described in Matsuo et al.30

Statistical analysis. Data were summarized descriptively; categorical variables are shown as n (%) and continuous variables as mean (SD). The tacrolimus continuation rate and cumulative incidence rates of ADRs, serious ADRs, relapse, progression to renal failure, and progression to dialysis were calculated by survival analysis using the Kaplan-Meier method. In these analyses, patients who dropped out or were lost to follow-up without events of interest were censored at subsequent assessments. Changes in variables of renal function from baseline were evaluated using the Wilcoxon signed-rank test. All analyses were performed using SAS statistical software (version 9.4; SAS Institute); missing data were not imputed.

RESULTS

Patients. Data were collected for a total of 1395 patients with LN receiving maintenance treatment with tacrolimus (Figure 1). Of these, 1355 patients were included in the safety analysis set (40 patients were excluded because of missing safety data). The effectiveness analysis set comprised 1353 patients, 1140 of whom were included in the renal prognosis analysis set. Reasons for exclusion from these analysis sets are shown in Figure 1.

Figure 1.

Patient disposition, adapted with permission from Takeuchi et al.29a A total of 18 patients had > 1 reason for exclusion from the renal prognosis set. These patients were only counted once in the overall total of excluded patients. CRF: case report form.

Patient baseline demographics and clinical characteristics have been reported previously.29 In brief, patients were predominantly female (84.9%), with a mean (SD) age of 38.3 (13.6) years (Supplementary Table S1, available with the online version of this article). Supplementary Table S1 also includes patient immunologic and clinical activity at baseline. Mean (SD) duration of SLE was 9.3 (8.1) years and of LN was 6.7 (7.0) years. Renal biopsies were available for 900 patients (66.4%) at the time of treatment initiation. Based on biopsy-proven pathologic type according to the International Society of Nephrology/Renal Pathology Society classification of LN,31 most patients had class IV or V disease.

Mean (SD) duration of follow-up was 2479.7 (1445.0) days (median 3653 days). Two-thirds of all patients (66.6%) were followed for ≥ 5 years. Follow-up biopsies were performed in 72 (11.6%) cases, with pathologic changes confirmed in 36 (5.8%) cases (Supplementary Table S2, available with the online version of this article).

Treatment. The mean daily dose of tacrolimus and mean tacrolimus blood concentrations over the 10 years of follow-up are shown in Supplementary Figure S1 (available with the online version of this article). The mean (SD) daily dose of tacrolimus remained stable from year 1 and was 2.6 (0.8) mg/day at year 10. Mean (SD) tacrolimus blood concentrations increased slightly over time, from 4.0 (2.6) ng/mL at week 4 to 5.4 (2.8) ng/mL at year 10. Almost half of all patients remained on tacrolimus for 10 years (Figure 2). At year 10, the Kaplan-Meier estimated tacrolimus continuation rate was 49.3%.

Figure 2.

Tacrolimus continuation rate over the 10 years of follow-up (safety analysis set).

Tacrolimus was discontinued in 588 of the 1355 patients included in the safety analysis set over the 10 years of follow-up (43.4%). The most common reasons for treatment discontinuation were onset of AEs (242 [17.9%] patients), unchanged/worsened symptoms (200 [14.8%] patients), symptoms improved (56 [4.1%] patients), and patient request (65 [4.8%] patients). Of the 200 patients with unchanged/worsened symptoms, 135 were switched to other therapies; cyclosporine was discontinued in 40 patients, CYC in 32 patients, MMF in 25 patients, mizoribine in 19 patients, AZA in 12 patients, and rituximab in 5 patients. Of the patients who discontinued tacrolimus because of onset of AEs, 40.9% discontinued treatment during the first 28 weeks.

Most patients (94.6%) were receiving corticosteroids at baseline. At baseline, the mean (SD) daily dose of oral corticosteroids (prednisolone equivalent) was 17.3 (12.0) mg/day. The mean (SD) daily oral corticosteroid dose decreased significantly from 16.0 (9.7) mg/day at 4 weeks after initiation of tacrolimus treatment to 7.2 (4.4) mg/day at year 10 (P < 0.001; Supplementary Figure S1, available with the online version of this article). Of 434 patients receiving corticosteroids at year 10, 2 (0.5%) were withdrawn from corticosteroid treatment after 10 years.

Safety. A total of 2409 ADRs were reported in 821 (60.6%) patients over the 10 years of follow-up. The most commonly reported ADRs (ie, occurring in ≥ 2% of patients) were hypertension (occurring in 7.7% of patients), nasopharyngitis (5.4%), herpes zoster and diarrhea (both 4.6%), upper respiratory tract inflammation (4.1%), bronchitis (3.8%), increased blood creatinine (3.7%), hyperuricemia (3.2%), diabetes mellitus (3.1%), pharyngitis (3%), gastroenteritis and renal impairment (both 2.9%), cystitis (2.4%), constipation (2.3%), and pneumonia (2.1%). The cumulative incidence of ADRs and serious ADRs of special interest over the 10 years of follow-up are shown in Figure 3. There was no marked increase in incidence of any ADR or serious ADR over time.

Figure 3.

Cumulative incidence of (A) adverse drug reactions, and (B) serious adverse drug reactions over the 10 years of follow-up (safety analysis set). No serious adverse drug reactions of menstrual disorders were reported.

ADRs and serious ADRs of special interest are summarized in the Table. Among the 1355 patients in the safety analysis set, infections were reported in 390 patients (28.8%), renal impairment in 156 patients (11.5%), neuropsychiatric disorders in 94 patients (6.9%), impaired glucose tolerance in 92 patients (6.8%), cardiac dysfunction in 34 patients (2.5%), malignant tumors (lymphomas) in 33 patients (2.4%), menstrual disorders in 13 patients (1%), and pancreatic dysfunction in 5 patients (0.4%). No patients reported worsening of interstitial pneumonia over the 10 years of follow-up.

Table.

Incidence of ADRs and serious ADRs of special interest reported by MedDRA/J preferred terms occurring in ≥ 2% of patients over the 10 years of follow-up (safety analysis set; N = 1355).

The most frequently reported serious ADRs of special interest were infections, which were reported in 131 patients (9.7%). Except for 1 case of adenocarcinoma, all malignant tumors were classed as serious ADRs. Malignant tumors reported by more than a single patient were lymphoma (4 patients), colon cancer, lung adenocarcinoma, and papillary thyroid cancer (each in 3 patients), and cervical carcinoma and malignant lung neoplasm (both in 2 patients). The annual incidence of malignant tumors over the 10 years of follow-up ranged from 0.1% to 1%. Serious ADRs of cardiac dysfunction were observed in 13 patients (1%); events reported by more than a single patient were angina pectoris (4 patients), acute myocardial infarction, cardiac failure, and myocardial infarction (each in 2 patients). The incidence of ADRs of cardiac dysfunction according to time of onset during follow-up was 0.7% during the first 28 weeks, 0.1% from 28 weeks up to end of year 1, and ranged from 0.2% to 1% from 1 year until the end of follow-up.

The safety profile of tacrolimus was consistent across different LN pathologic classes, although some differences in the incidence of major ADRs, such as infections, were observed, with the highest incidence of infections occurring in those with type IV-G+V disease (10 [43.5%] patients; Supplementary Table S3, available with the online version of this article).

Forty-eight patients died over the 10 years of follow-up (3.5%). Twenty-two deaths were considered related to tacrolimus: pneumonia (3 patients), sudden death (2 patients), and Pneumocystis jirovecii pneumonia, staphylococcal endocarditis, staphylococcal pneumonia, angina pectoris, recurrent angioimmunoblastic T-cell lymphoma, death, gastric ulcer hemorrhage, cardiac failure, prostate cancer, large intestine perforation, hematemesis, urosepsis, disseminated intravascular coagulation, lung adenocarcinoma, large cell lung cancer, ovarian cancer, and pancreatic neuroendocrine tumor (1 patient for each).

Effectiveness. Kaplan-Meier estimated cumulative rates of relapse and progression to renal failure and dialysis are shown in Figure 4. In the effectiveness analysis set, the cumulative rate of relapse at 10 years was 44.5% and overall patient survival rate was 96.5%. In the renal prognosis analysis set, the cumulative rates of progression to renal failure and dialysis at 10 years were 12.2% and 4.5%, respectively. When analyzed by LN pathologic class, progression to renal failure and dialysis (renal prognosis analysis set) and incidence of disease relapse (effectiveness analysis set) were generally consistent across pathologic classes (Supplementary Table S4A, available with the online version of this article). The 10-year effectiveness data by continuation or discontinuation of tacrolimus are given in Supplementary Table S4B. However, progression to renal failure and dialysis was more common in patients with type IV-G+V disease. Serum creatinine levels and eGFR over the 10 years of follow-up are shown in Figure 5. Serum creatinine levels remained relatively stable from year 1 through year 10. For eGFR, a significant decrease from baseline was observed from week 4 through end of follow-up (P < 0.001). Mean (SD) eGFR was 80.9 (29.8) mL/min/1.73 m2 at baseline, 79.3 (27.9) mL/min/1.73 m2 at week 4, and 68.3 (23.0) mL/min/1.73 m2 at year 10.

Figure 4.

Cumulative rates of relapse (effectiveness analysis set) and progression to renal failure and dialysis (renal prognosis analysis set) over the 10 years of follow-up.

Figure 5.

Mean (SD) (A) serum creatinine level and (B) eGFR over the 10 years of follow-up (effectiveness analysis set). eGFR: estimated glomerular filtration rate.

The number of patients progressing to the different stages of chronic kidney disease during the observation period is described in Supplementary Table S5 (available with the online version of this article) and the need for additional immunosuppressive therapy and specific therapy administered is given in Supplementary Table S6.

DISCUSSION

This is the first 10-year study that we know of that assesses the long-term safety and effectiveness of tacrolimus as maintenance treatment in routine clinical practice settings for patients with LN, a patient population for whom such data are notably lacking. In this study, a total of 1355 patients with LN receiving maintenance treatment with tacrolimus were followed for 10 years, including assessment of renal survival. To our knowledge, this is the largest cohort of patients with LN receiving maintenance treatment to be followed for this length of time. Our results show tacrolimus is well tolerated and effective as LN maintenance therapy over a period of 10 years, confirming and extending the findings of an interim analysis conducted after 5 years of follow-up.29 The high rate of treatment continuation observed in the present study after 10 years of follow-up is particularly noteworthy. Almost half of all patients (49.3%) remained on tacrolimus for the full 10 years of follow-up. The previously reported continuation rate at 5 years of follow-up was 62.7%.29

The type and incidence of ADRs observed over the 10 years of follow-up in this study were in line with the known safety profile of tacrolimus, and no unexpected safety concerns were reported. Infections were the most common ADR of special interest, occurring in 28.8% of patients. As previously observed in the interim analysis after 5 years of follow-up,29 ADRs tended to develop early in the course of tacrolimus treatment. Whereas some differences according to LN pathologic class were observed in the incidence of infections at 10 years of follow-up, such differences should be interpreted with caution owing to confounding factors. Except for infections, there was no marked increase in incidence of any of the ADRs of interest over time, including renal impairment, malignant tumors, and cardiac dysfunction. Of note, a significant reduction in mean (SD) daily corticosteroid dose was observed over the 10 years of follow-up, from 17.3 (12.0) mg/day at baseline, to 10.6 (5.3) mg/day at 1 year, and 7.2 (4.4) mg/day at 10 years. Other studies have also shown tacrolimus to have steroid-sparing effects,32,33 which may be expected to reduce the likelihood of adverse effects associated with long-term corticosteroid treatment. The apparent lack of adverse effects on ovarian function is also noteworthy, given the increased risk of premature ovarian failure associated with use of CYC in patients with SLE.34,35

The current results extend from the previously published findings of the interim analysis conducted after 5 years of follow-up, which showed a cumulative relapse rate of 30.6% and cumulative progression rates of 6.6% for renal failure and 1% for dialysis.29 The present analysis showed that the cumulative rate of relapse over the 10 years of follow-up was 44.5% and overall patient survival rate was 96.5%. The cumulative rate of progression to renal failure over 10 years of follow-up was 12.2%, whereas the cumulative rate of progression to dialysis was 4.5%. Further, renal function was found to be well maintained over the entire follow-up period. This is of note, as concerns have been raised that long-term use of calcineurin inhibitors may be associated with potential nephrotoxic effects in patients who have undergone kidney transplant.36,37 Our findings support the results of randomized clinical studies that demonstrate the safety and efficacy of tacrolimus as induction or maintenance therapy in patients with LN.17-22 Our findings are also consistent with those of other postmarketing surveillance studies undertaken in Japan that demonstrate the safety and effectiveness of tacrolimus-containing immunosuppressive regimens for the treatment of other conditions, such as rheumatoid arthritis and myositis-associated interstitial pneumonia.38,39

The observed 10-year rates of overall and renal survival in this study compare favorably with data from other studies of long-term outcomes in cohorts of patients with LN. In a retrospective analysis of 186 Japanese patients with LN, Kaplan-Meier estimates of 10-year overall and renal survival were 95.7% and 94.3%, respectively.40 In the Mycophenolate Mofetil Versus Azathioprine for Maintenance Therapy of Lupus Nephritis (MAINTAIN) Nephritis trial, which compared AZA and MMF as LN maintenance therapy, data were available for 92 patients at 10 years of follow-up.41 Of these patients, 5 had died over the 10 years of follow-up (5.4%), 4 had progressed to end-stage kidney disease (4.3%), and 41 had experienced relapse (44.5%).41 Of note, only one-third of patients remained on their assigned maintenance therapy (AZA or MMF) at 10 years.41 This is notably lower than the observed tacrolimus continuation rate of 49.3% over the 10 years of follow-up in the present study. Although care should be taken when comparing results of studies of heterogeneous design in different patient cohorts and at different points in time, our findings support use of tacrolimus as long-term maintenance therapy in patients with LN.

The limitations of this study have been reported previously.29 In brief, these include the lack of a control group to provide comparative data, the absence of a strict protocol-defined definition of relapse, and the fact that results may not be generalizable to other ethnicities. In addition, it is possible that patients who tolerated and were responding to tacrolimus were more likely to continue treatment, which could be a potential source of bias. Study findings, including the incidence of ADRs, could be influenced by factors other than use of tacrolimus (eg, concomitant medication), and there may be potential confounding variables. Additionally, we could not determine whether safety data and/or AE-related discontinuations occurred in the context of moderate/high-dose corticosteroids. During the observation period, safety, steroid dose reduction, and disease relapse were evaluated during tacrolimus exposure, whereas progression to renal failure and dialysis were evaluated regardless of tacrolimus exposure, meaning that the effect of tacrolimus on these outcomes could be overestimated. Further, the influence of treatments after tacrolimus discontinuation was not considered. Data were not collected on nonrenal activity, which may account for prolonged use of corticosteroids. Finally, data were collected under conditions that do not reflect recent changes in the LN treatment landscape; for example, the approval of belimumab in Japan for the treatment of adult patients with SLE who have inadequately responded to conventional treatments. Despite these limitations, this study provides valuable data concerning the long-term safety and effectiveness of tacrolimus for the treatment of LN.

In conclusion, tacrolimus was found to be effective and generally well tolerated as maintenance therapy for LN in a large cohort of patients in Japan followed for 10 years. Almost half of all patients remained on tacrolimus for the full 10 years of follow-up, and renal function was found to be well maintained in these patients over time. Our findings add to the evidence supporting the utility of tacrolimus for the clinical management of patients with LN.

ACKNOWLEDGMENT

The authors wish to thank all the investigators and medical institutions involved in this postmarketing surveillance study. Medical writing support was provided by Jennifer Coward, PhD, for Lumanity, funded by Astellas Pharma, Inc. Interim data from this study were published as: Takeuchi T, Wakasugi N, Uno S, Makino H. Long-term safety and effectiveness of tacrolimus in patients with LN: 5-year interim postmarketing surveillance study in Japan (TRUST). J Rheumatol 2021;48:74-81.

Footnotes

This work was supported by Astellas Pharma. Editorial support was funded by Astellas Pharma.

All authors report nonfinancial support from Astellas during the development of the manuscript. TT has received grants from AbbVie, Chugai Pharmaceutical, and Mitsubishi-Tanabe Pharma; consulting fees from AbbVie, Astellas Pharma, Chugai Pharmaceutical, Eli Lilly Japan, and Mitsubishi-Tanabe Pharma; and honoraria for speakers from AbbVie, Chugai Pharmaceutical, Eli Lilly Japan, and Mitsubishi-Tanabe Pharma. NW, TH, and SU are employees of Astellas. HM has served on advisory boards for Boehringer Ingelheim and Travere Therapeutics.

Accepted for publication January 31, 2024.Copyright © 2024 by the Journal of Rheumatology

This is an Open Access article, which permits use, distribution, and reproduction, without modification, provided the original article is correctly cited and is not used for commercial purposes.