Successful Discontinuation of Glucocorticoid Treatment after Administration of a Calcineurin Inhibitor for Nephrotic Syndrome in a Patient with Lupus Nephritis

Abstract

Systemic lupus erythematosus was diagnosed in a patient at 43 years old. When proteinuria recurred at 57 years old, the first kidney biopsy was performed, and class IV-G (A)+V lupus nephritis was diagnosed. The prednisolone dose was increased to 40 mg/day, and cyclosporine A was introduced. After 1 year, proteinuria had decreased to 0.1 g/day. Prednisolone was discontinued three years later, and cyclosporine A was continued. Thereafter, proteinuria did not reoccur. At 67 years old, a second kidney biopsy showed complete remission of lupus nephritis. Cyclosporine A enabled permanent discontinuation of glucocorticoids in a patient with lupus nephritis.

Introduction

Lupus nephritis (LN) is an inflammation of the kidneys caused by systemic lupus erythematosus (SLE), affecting every organ and tissue. SLE had a very poor prognosis until the introduction of glucocorticoids, which dramatically improved the outcome (1). However, cardiovascular disease and infection associated with long-term glucocorticoid use have become newer complications that contribute to mortality (2). Immunosuppressive drugs, including calcineurin inhibitors (CNIs), hydroxychloroquine, mycophenolate mofetil, and biologic therapies, such as belimumab, are indicated as glucocorticoid-sparing agents in severe cases of SLE (3-5); however, in patients with LN, discontinuation of glucocorticoid treatment is considered difficult after long-term use.

We herein report a patient with LN who has been glucocorticoid-free for more than 20 years since the introduction of the CNI cyclosporine A (CyA) and switching to tacrolimus (Tac).

Case Report

A 67-year-old Japanese woman with a 24-year history of LN was admitted to our hospital for the evaluation of increased proteinuria. At 43 years old, the patient had been diagnosed with SLE because of butterfly rash, a fever, arthritis, proteinuria (3.6 g/day), and positive tests for antinuclear antibody and double-stranded DNA antibody (ds-DNA Ab). Treatment had been started with prednisolone (PSL, 40 mg/day) and azathioprine (50 mg/day), and the proteinuria had subsequently subsided.

However, at 49 years old, the patient developed aseptic bone necrosis, so the PSL dose was gradually reduced. At 52 years old, the PSL dose was reduced to 5 mg/day, but proteinuria subsequently increased to 1.2 g/day. Therefore, the dose was increased to 40 mg/day, and the proteinuria subsided. At 57 years old, the PSL dose was reduced to 25 mg, but proteinuria rose again to 3.7 g/day, and the urinary sediment contained 1 to 5 erythrocytes per high-power field. At this time, creatinine was 0.8 mg/dL; ds-DNA Ab, 42 IU/mL (normal value, <10 IU/mL); total complement activity (CH50), 26 U/mL (normal value, >30 U/mL); C3, 61 mg/dL (reference range, 86-160 mg/dL); and C4, 18 mg/dL (reference range, 17-45 mg/dL) (Table). Mizoribine were used in combination with PSL but was not effective in PSL reduction. The first kidney biopsy was therefore performed (Fig. 1).

Table.

Laboratory Data at the First and Second Kidney Biopsies.

Laboratory data		At 1st biopsy		At 2nd biopsy		Normal
WBC [/μL]		9,000		4,800		3,400-9,200
Hb [g/dL]		11.9		9.1		13.0-17.0
Plt [/μL]		21.9×104		18.6×104		13-36×104
Alb [g/dL]		2.7		3.1		3.9-5.2
UN [mg/dL]		18		50		8-21
Cr [mg/dL]		0.8		1.7		0.65-1.06
eGFR [mL/min/1.73m2]		57.3		23.9
IgG [mg/dL]		1,739		2,237		870-1,700
IgA [mg/dL]		573		423		110-410
IgM [mg/dL]		62		82		33-190
CH50 [U/mL]		26		47		23-46
C3 [mg/dL]		61		90		86-160
C4 [mg/dL]		18		23		17-45
ANA		(+)		(+)		(-)
Anti-dsDNA Ab [IU/mL]		42		22.1		<20
proteinuria [g/day]		3.0		0.1		<0.3
U-RBC (/HPF)		1-5		1-5		<5
β2MG [μg/L]		442		864		14-329

Figure 1.

Clinical course over the past 30 years from the onset of lupus nephritis to the present day. LN: lupus nephritis, PSL: prednisolone, CyA: cyclosporine A, Tac: tacrolimus, AZA: azathioprine, ds-DNA Ab: double-stranded DNA antibody

Results of the first kidney biopsy

A light microscopic examination of a biopsy specimen revealed global sclerosis in 3 out of 16 glomeruli. More than 50% of the glomeruli were involved. Endocapillary proliferative lesions were seen, including wire-loop lesions in the intracapillary space that were filled with homogenous and rigid material (Fig. 2A). A subepithelial spike formation was also noted. Chronic inactive lesions with adhesions were seen in some areas. Tubular atrophy and tubulointerstitial fibrosis occupied approximately 20% of the total renal cortex. The arterioles and interlobular arteries showed mild sclerotic changes.

Figure 2.

Kidney biopsy results. (a) The first kidney biopsy. A: Periodic acid-Schiff (PAS) stain (light microscopy, LM). B: Periodic acid methenamine silver (PAM)-Masson stain (LM). C: Immunoglobulin G stain (immunofluorescence, IF). D: Ciq stain (IF). E: Electron microscopy (EM) revealed massive subendothelial electron-dense deposits (EDD; large arrow) and EDD in the subepithelial region (small arrow). F: EM revealed mesangial EDD (arrow). (b) The second kidney biopsy. A: Masson trichrome stain (LM) revealed that tubular atrophy and tubulointerstitial fibrosis occupied approximately 80% of the renal cortex. B: Masson trichrome stain (LM) showed widespread striped fibrosis as the main feature. C: PAM-Masson stain (LM). D: PAS stain showed severe hyalinosis and stenosis of the interlobular artery and arteriole. E: PAM-Masson stain (LM) showed that preserved glomeruli were intact. F: IgG stain (IF) was negative for IgG. J: EM showed no EDD. (c) The third kidney biopsy. A: PAM-Masson stain showed that preserved glomeruli were intact. B: PAS stain showed that the vascular lesions appeared to have improved compared to the second biopsy.

Immunofluorescence showed positivity for IgG, IgA, IgM, C3, and C1q (the so-called “full house”) in the mesangial area, peripheral intracapillary area, and subepithelial area. An analysis of the IgG subclasses showed similar levels of positivity for IgG1, IgG2, IgG3, and IgG4. Immunofluorescence was negative for C4. Electron microscopy revealed massive subendothelial and mesangial electron-dense deposits, but microfibrillar structures with a fingerprint pattern were not detected. Electron-dense deposits were mainly found in the subendothelial and mesangial areas but also partly in the subepithelial region in less than 10% of the glomerular basement membrane. Class IV-G (A) LN was diagnosed according to the classification of the International Society of Nephrology/Renal Pathology Society (6).

Subsequent clinical course

After the first biopsy, the PSL dose was increased to 40 mg/day, and treatment was started with CyA under the diagnosis of nephrotic syndrome. After 1 year, proteinuria decreased to 0.1 g/day. The PSL dose was gradually reduced because of exacerbation of aseptic bone necrosis. After three years, PSL was discontinued, but CyA was continued. Thereafter, proteinuria did not reoccur. When the patient was 67 years old, the creatinine level increased to 1.7 mg/dL. The patient's blood pressure was 130/76 mmHg, and she had no symptoms of a fever, rash, or edema. The estimated glomerular filtration rate was 23.9 mL/min/1.73 m²; C-reactive protein, 0.9 mg/dL; C3, 90 mg/dL; C4, 23 mg/dL; CH50, 47 U/mL; and ds-DNA Ab fell to less than 10.0 IU/mL and remained in the normal range after the change to Tac. Urinary protein excretion was 0.1 g/day, and the urinary sediment contained 1 to 5 erythrocytes per high-power field (Table). At this time, a second kidney biopsy was performed.

Results of the second kidney biopsy

A light microscopic examination of a biopsy specimen containing 42 glomeruli revealed global sclerosis of 25 glomeruli. Preserved glomeruli were intact, but the interlobular artery and arteriole showed severe hyalinosis and stenosis. Tubular atrophy and tubulointerstitial fibrosis occupied approximately 80% of the total renal cortex, and widespread striped fibrosis was the main feature. Immunofluorescence microscopy was negative for all immunoglobulins, including IgG, IgA, and IgM, and all complements, including C3, C1q, and C4. Electron microscopy showed no electron-dense deposits (Fig. 2B). Complete remission of LN and cyclosporine-induced chronic nephropathy were diagnosed.

Subsequent clinical course

After the second biopsy, CyA was discontinued, and Tac was started at a dose of 2 mg/day and subsequently continued at a dose of 1 mg/day. Ten years later, proteinuria was 0.5 to 1.0 g/day without glucocorticoid treatment, and a third kidney biopsy was performed. Twenty-two of 28 glomeruli showed global sclerosis, and tubulointerstitial fibrosis appeared to be slightly more advanced than in the second biopsy due to long-term administration by Tac, but the small renal artery lesion appeared to have improved compared to the second biopsy over a 10-year period by adjusting the dosage of Tac. Complete remission of LN was confirmed because no immune deposits were seen on immunofluorescence or electron microscopy studies (Fig. 2C), and blood test findings, such as complement and ds-DNA Ab, remained intact (Fig. 1).

Discussion

Long-term exposure to a glucocorticoid is consistently linked with a number of serious complications, such as aseptic bone necrosis, osteoporosis, cardiovascular disease and infections, and stroke, all of which have been reported to contribute to increased mortality (3,7). Large observational studies support the notion that glucocorticoid-mediated toxicity is largely dependent on the dose and exposure time (8). Therefore, reducing the dose of a glucocorticoid or stopping it altogether is essential to avoid complications.

CyA, a CNI, has been reported to be effective in patients with LN and to contribute to a decrease in the glucocorticoid dose (9). Tac, another CNI, has also been reported to be effective at inducing remission in LN. Tac may bring about greater reductions in proteinuria than other LN treatments, an effect that has been linked to better long-term outcomes (10,11).

Zavada et al. reported that CyA was as effective as cyclophosphamide in a trial of sequential induction and maintenance treatment in patients with proliferative LN and a preserved renal function (12). However, no report has been published wherein the use of a CNI allowed long-term glucocorticoid treatment to be discontinued, and to our knowledge, our report is the first to describe a patient who achieved glucocorticoid-free remission after starting treatment with a CNI.

Goswami et al. conducted a retrospective observational study on discontinuation of glucocorticoid therapy in 148 patients with SLE and reported that prescription of an immunosuppressive agent lengthened the flare-free survival. However, their paper did not specify the type of immunosuppressive drug administered (13).

In the present case, proteinuria of 0.5 to 1.0 g/day has persisted since glucocorticoid discontinuation. However, we believe that this proteinuria is not due to LN, as immunofluorescence and electron microscopy of kidney biopsy specimens do not show any immunodeposits that suggest activity, such as LN, and blood test findings, such as complement and ds-DNA Ab, are normal.

In this case, PSL was effective for SLE/LN at the time, but relapse occurred with dose reduction; the combination of azathioprine or mizoribine was not effective in reducing the PSL dose. A CNI was initiated as the next treatment modality, allowing PSL reduction and even discontinuation. The complication associated with the use of CNIs was the development of severe vascular lesions and associated tubular lesions. However, with a dosage adjustment, ds-DNA Ab, complement, and urinary occult blood, all of which are therapeutic indicators of LN, remained within reference limits, avoiding the need for additional PSL treatment.

The findings of several reports on the mechanism of action of CNIs in LN and the differences in the mechanism of action between CyA and Tac have proven useful. CyA and Tac bind to a cytoplasmic immunophilin [cyclophilin for CyA and FK-binding protein 12 (FKBP12) for Tac] to inhibit calcium signaling by blocking the calcium-dependent enzyme calcineurin to exert an immunosuppressive effect (14). Regarding the non-immune effects of these agents, Faul et al. found that CyA blocks calcineurin-mediated dephosphorylation of synaptopodin in renal podocytes, stabilizes the actin cytoskeleton of podocytes, and decreases proteinuria (15,16). Tac is also known to have the same effects. Tac blocks both the binding of FKBP12 to transient receptor potential-cation channel (TRPC) 6 and the phosphatase activity of calcineurin. In contrast, CyA inhibits calcineurin without blocking the binding of FKBP12 to TRPC6 channels (17). This difference in their mechanism of action may explain the protective effect of Tac on podocytes (18).

Compared to the findings at the second renal biopsy, proteinuria had increased to 0.5-1.0 g/day 10 years later. This change was attributed to the increased number of sclerotic glomeruli, which contributed to hemodynamic changes associated with excessive blood flow overload to the residual glomeruli, as there were no abnormalities associated with immune deposits in the residual glomeruli.

In conclusion, we described a case of LN that recurred each time the glucocorticoid dose was reduced but subsided after the introduction of CNIs, allowing this patient with LN to finally discontinue glucocorticoid treatment.

Limitations

Tac is covered by insurance for lupus nephritis, but CyA is not; however, CyA was used to treat glucocorticoid-dependent nephrotic syndrome in the present case. The treatment was performed with the approval of the Institutional Review Board (approval number 1487-B).

Some rheumatology specialists have asked why mycophenolate mofetil (MMF) and intravenous cyclosporin A (IVCY) were not used. MMF and IVCY are now common treatments that achieve remission in SLE, but they were not available 15 to 20 years ago, at the time of treatment of the present case. Azathioprine and mizoribine were used in combination with PSL but were not effective in PSL reduction. CNIs were used because CNIs were the only immunosuppressive agents available.

The present report was produced in conformity with the Declaration of Helsinki, and the patient gave her consent for this case report to be published.

The authors state that they have no Conflict of Interest (COI).