Treatment of pure membranous lupus nephropathy with prednisone and azathioprine: an open-label trial

Treatment of pure membranous lupus nephropathy with prednisone and azathioprine: an open-label trial

Treatment of Pure Membranous Lupus Nephropathy With Prednisone and Azathioprine: An Open-Label Trial Chi Chiu Mok, MD, MRCP, King Yee Ying, MB, MRCP, ...

68KB Sizes 0 Downloads 51 Views

Treatment of Pure Membranous Lupus Nephropathy With Prednisone and Azathioprine: An Open-Label Trial Chi Chiu Mok, MD, MRCP, King Yee Ying, MB, MRCP, Chak Sing Lau, MD, FRCP, Cheuk Wan Yim, MB, MRCP, Woon Leung Ng, MB, MRCP, Woon Sing Wong, MB, FRCP, and Tak Cheung Au, MB, FRCP ● Background: The aim of this study was to report the outcome of pure membranous lupus nephropathy treated with prednisone and azathioprine (AZA). Methods: Consecutive patients with pure membranous lupus glomerulonephritis (World Health Organization [WHO] Va and Vb) from 4 regional hospitals were recruited for an open-label treatment trial consisting of prednisone and AZA. Remission status was evaluated at 12 months. Maintenance treatment with low-dose prednisone and AZA was continued indefinitely for those who achieved remission. Factors predictive of initial renal remission and subsequent relapse were studied by statistical analyses. Results: Thirtyeight patients (31 women and 7 men) were studied. The mean age was 35.0 ⴞ 9.2 years, and the duration of systemic lupus erythematosus was 48.5 ⴞ 59 months. Seventeen (45%) patients had WHO class Va lupus nephritis, whereas 21 (55%) had class Vb disease. Two patients withdrew from the protocol because of idiosyncratic reactions to AZA. At 12 months, 24 (67%) patients achieved complete remission (CR), 8 (22%) achieved partial remission (PR), and 4 (11%) were treatment resistant. Patients who achieved CR or PR were maintained on low-dose prednisone and AZA. Over a mean follow-up period of 90.4 ⴞ 59 months, 6 (19%) patients had relapse of nephritis (proteinuric flare in 4 and nephritic flare in 2). The cumulative risk of renal relapse was 12% at 36 months and 16% at 60 months. No particular clinical variables were found to predict renal remission or relapses. Over a mean follow-up of 90 months, 13% of patients had decline of creatinine clearance by 20%, but none had doubling of serum creatinine. Renal outcome was not significantly worse in patients presenting with nephrotic syndrome. Treatment generally was well tolerated. Conclusion: A combination of prednisone and AZA is reasonably effective for the initial treatment of pure membranous lupus nephritis. Severe adverse effects are uncommon. The additional efficacy of AZA in comparison with prednisone alone has to be confirmed with randomized, controlled trials. Am J Kidney Dis 43:269-276. © 2004 by the National Kidney Foundation, Inc. INDEX WORDS: Glomerulonephritis; nephritis; prognosis; cytotoxic; immunosuppressive; lupus erythematosus.

R

ENAL DISEASE is a major cause of morbidity and mortality in patients with systemic lupus erythematosus (SLE). Among the various World Health Organization (WHO) histologic classes of lupus nephritis, membranous glomerulopathy (class V) accounts for only 8% to 20% of patients with biopsy-confirmed lupus nephritis in recently published large international series.1-5 Membranous lupus nephropathy was formerly subclassified into 4 groups: pure membranous nephropathy without (Va) or with mesangial hypercellularity (Vb), membranous nephropathy with segmental endocapillary proliferation and/or necrosis (Vc), and membranous nephropathy with diffuse endocapillary proliferation and/or necrosis (Vd). However, it was recognized that the long-term prognosis of lupus membranous nephropathy depends on the degree of glomerular inflammation and proliferative changes on renal biopsy findings.6-8 Patients with Vc and Vd disease have a clinical course and prognosis similar to the proliferative class III or class IV lupus nephritis. The WHO classification was recently revised, with Vc and Vd reclassified into classes III or IV depending on active necro-

tizing lesions. Class V retains only the subclasses Va and Vb under the category “diffuse membranous glomerulonephritis.”9 The optimal treatment of membranous lupus nephropathy is virtually unknown because there have been extremely few controlled studies in this condition. Uncontrolled experience has shown that azathioprine (AZA), chlorambucil, cyclosporin A, and fludarabine, with and without corticosteroid, are effective.10-15 An open study found that oral corticosteroid and the sequential

From the Department of Medicine, Tuen Mun Hospital; Department of Medicine, Princess Margaret Hospital; Department of Medicine, Queen Mary Hospital; and Department of Medicine, United Christian Hospital, Hong Kong, SAR, China. Received August 18, 2003; accepted in revised form October 21, 2003. Address reprint requests to Chi Chiu Mok, MD, MRCP, Associate Consultant in Rheumatology, Department of Medicine & Geriatrics, Tuen Mun Hospital, Tsing Chung Koon Rd, New Territories, Hong Kong, China. E-mail: [email protected] © 2004 by the National Kidney Foundation, Inc. 0272-6386/04/4302-0021$30.00/0 doi:10.1053/j.ajkd.2003.10.029

American Journal of Kidney Diseases, Vol 43, No 2 (February), 2004: pp 269-276

269

270

MOK ET AL

use of oral cyclophosphamide (CYC) and AZA alleviated proteinuria and preserved renal function in 90% of patients with membranous lupus nephropathy and nephrotic syndrome.16 A retrospective analysis of 19 SLE patients with Va and Vb disease showed that combining chlorambucil and methylprednisolone was more effective than methylprednisolone alone in inducing remission and preserving renal function over a mean observation period of 83 months.11 An ongoing randomized, controlled trial initiated by the National Institutes of Health (NIH) reported that either pulse CYC or cyclosporin A was more effective than prednisone alone in terms of remission and proteinuria reduction at 12 months.17 Thus, it appears that a combination of corticosteroid and cytotoxic agents is a better treatment strategy than corticosteroid alone for membranous lupus nephropathy. More aggressive treatment is warranted for patients with persistent nephrotic syndrome or declining renal function. In the current study, we evaluated the outcome of a cohort of consecutive Chinese SLE patients with membranous lupus nephropathy (WHO Va and Vb) treated with prednisone and AZA in 4 regional hospitals in Hong Kong. PATIENTS AND METHODS The Queen Mary, Tuen Mun, United Christian, and Prince Margaret Hospitals are 4 large regional hospitals in Hong Kong that serve a total population of at least 2 million. Between 1986 and 2001, consecutive patients with renal biopsy–proven pure membranous lupus glomerulonephritis (WHO Va and Vb) were recruited into an open-label treatment study consisting of prednisone and AZA. Renal biopsy samples were examined routinely by light microscopic, immunofluorescence, and electron microscopic studies. Patients with Vc and Vd lupus nephritis according to the old WHO classification were excluded for analysis.

Treatment Protocol All the patients studied were treated initially with prednisone (0.8 to 1 mg/kg/d) or equivalent for 6 to 8 weeks, after which the dosage was tapered by 5 mg/wk to a maintenance dose of 5 to 10 mg/d. AZA was initiated at 1 mg/kg/d and targeted to a dose of 2 mg/kg/d. The exact dosage depends on patients’ tolerability, white cell counts, and the presence of intercurrent infection. Hypertension, if present, was treated initially with either ␤-blockers or calcium channel blockers. Angiotensin-converting enzyme inhibitors were used only at the discretion of physicians if proteinuria persisted after 12 months’ immunosuppressive treatment. Persistent hyperlipidemia was treated with the hydroxymethylglutaryl coenzyme A (HMG CoA) reductase inhibitors.

At 12 months of therapy, patients were assessed for their remission status (see definitions below). For those who remitted completely or partially, AZA and low-dose prednisone (ⱕ10 mg/d) was continued indefinitely as maintenance. For those who withdrew from the treatment protocol or who were resistant to initial treatment, alternative modalities were given according to individual physicians’ decisions. The progress of our patients was traced until the latest follow-up visits, and the following outcome was obtained: relapse of nephritis, loss of 20% of creatinine clearance, doubling of serum creatinine, and infective and thromboembolic complications.

Criteria for Remission and Relapse Complete remission (CR), partial remission (PR), and nonremission (NR) were defined as follows. CR was defined as stabilization or improvement in serum creatinine level with reduction of proteinuria to less than 1 g/d and persistent normalization of the C3 level for at least 6 months. PR was defined as stabilization or improvement in serum creatinine level with persistent reduction of proteinuria (if nephrotic range at baseline, a ⱖ50% decrease in proteinuria but ⬍3 g/d; if nonnephrotic at baseline, a decrease to ⱕ50% of the pretreatment value but ⬎1 g/d) for at least 6 months. NR referred to deterioration of serum creatinine level exclusive of other causes (such as sepsis, nephrotoxic agents, overdiuresis, and renal vein thrombosis), an increase in proteinuria, or a reduction in proteinuria but not to the extent of CR or PR. A proteinuric renal flare was defined as an increase in proteinuria to more than 2 g/d, with or without deterioration in serum creatinine level, after a CR, or doubling of proteinuria, with or without deterioration in renal function, in patients who achieved PR. A nephritic flare was defined as an increase or recurrence of active urinary sediments (ⱖ10 per high-power field), with or without a concomitant increase in proteinuria or deterioration of serum creatinine level.

Assessment of Disease Activity of SLE Disease activity of SLE was measured using the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI), a validated tool that has been shown to be sensitive to change.18 The SLEDAI scores at the time of renal biopsy and at 12 months after immunosuppressive treatment were obtained retrospectively and compared.

Statistical Analysis Unless otherwise specified, values in this study were expressed as mean ⫾ SD. Comparison of categorical data between 2 groups was made by the ␹2 test, and Yates’ continuity correction was performed when the frequency was small. Comparison of continuous data between 2 groups was achieved by the Students’ t test. When normal distribution or equal variance could not be assumed, the Mann Whitney rank sum test was used instead. Paired data before and after treatment were compared using the paired Students’ t test for continuous variables and Wilcoxon’s matched pair test for categorical variables.

OUTCOME OF PURE MEMBRANOUS LUPUS NEPHRITIS

Logistic regression was used to analyze the factors associated with CR 12 months posttreatment. Timing and the cumulative risks of renal flares were studied by KaplanMeier analysis. Cox regression was used for identification of factors predicting renal flares. Covariates included for the regression analyses included the following: age at the time of renal biopsy, sex, SLE duration, WHO Va/Vb histology, initial renal parameters (proteinuria, presence of urinary casts, serum creatinine, creatinine clearance, serum albumin, nephrotic syndrome), presence of hypertension, hypocomplementemia, autoantibodies (anti–double-stranded DNA [dsDNA], anti-Ro, anti-La, anti-nuclear ribonucleoprotein [nRNP], anti-Sm), and remission status (CR/PR) at 12 months (for Cox regression only). A backward stepwise procedure was adopted based on a likelihood ratio test, with P value greater than 0.10 for removal and P value less than 0.05 for entry of variables. Statistical significance was defined as a P value of less than 0.05, 2 tailed. All statistical analyses were performed on a computer using the SPSS program, version 8.0 (SPSS, Chicago, IL) for Windows 98.

RESULTS

Thirty-eight patients were studied. All were ethnic Chinese and fulfilled at least 4 of the American College of Rheumatology criteria for the classification of SLE.19 Table 1 shows their demographic and clinical manifestations at the time of renal biopsy. There were 31 women and 7 men, and the mean age was 35.0 ⫾ 9.2 years. The mean duration of SLE was 48.5 ⫾ 59 months. Twenty-two (58%) patients were nephrotic. Seven (18%) patients had received AZA previously because of extrarenal manifestations (hematologic in 3, dermatologic in 2, neuropsychiatric in 1, and gastrointestinal in 1). Seventeen (45%) patients had WHO class Va lupus nephritis, whereas 21 (55%) had class Vb disease. All patients were treated with oral prednisone (mean initial daily dose, 0.85 ⫾ 0.24 mg/kg and duration of treatment, 7.1 weeks) and AZA (mean dose, 1.72 ⫾ 0.43 mg/kg/d). Two patients withdrew from the protocol because they had idiosyncratic reactions to AZA shortly after commencement of the drug (hypersensitivity in 1 and agranulocytosis in the other). Table 2 shows the renal parameters of our patients at start of protocol and at 12 months posttreatment. There was significant improvement in proteinuria, creatinine clearance, serum albumin, C3 levels, and the SLEDAI scores. According to our remission criteria, 24 (67%) patients achieved CR, 8 (22%) achieved partial remission, and 4 (11%) were treatment resistant.

271 Table 1. Clinical Characteristics of 38 Patients With Pure Membranous Lupus Nephropathy Women (%) Age (y) SLE duration (mo) Body weight (kg) WHO Va histology (%) WHO Vb histology (%) Extrarenal features at time of nephritis Musculoskeletal (%) Mucocutaneous (%) Neuropsychiatric (%) Serositis (%) Hematologic (%) Previous treatment history Prednisone (%) AZA (%) HCQ (%) CYC (%) Anti-Ro (%) Anti-La (%) Anti-Sm (%) Anti-nRNP (%)

31 (82) 35.0 ⫾ 9.2 48.5 ⫾ 59 52.0 ⫾ 8.9 17 (45) 21 (55) 15 (39) 19 (50) 2 (5) 4 (11) 9 (24) 25 (66) 7 (18) 8 (21) 2 (5) 22 (58) 1 (31) 2 (5) 14 (37)

Abbreviation: HCQ, hydroxychloroquine.

Three patients with NR remained nephrotic. Logistic regression was performed to identify factors associated with CR at 12 months after treatment (data not shown). Unfavorable factors for CR were younger age, Vb histology, higher serum C3, increasing proteinuria, nephrotic syndrome, and lower serum albumin levels, but none of these reached statistical significance. Patients who achieved either CR or PR (n ⫽ 32) were maintained on low-dose prednisolone (7.0 ⫾ 2.5 mg/d) and AZA (mean daily dose, 75.9 ⫾ 25.4 mg). The mean duration of AZA maintenance in these patients was 77.1 months at the time of data analysis. Over a mean follow-up period of 90.4 ⫾ 57 months, 6 (19%) patients had relapse of nephritis (proteinuric flare in 4 and nephritic flare in 2). The cumulative risk of renal relapse was 0%, 12.1%, and 15.6% at 12, 36, and 60 months, respectively. Renal biopsy performed in 4 patients showed focal proliferative nephritis (class III) in 1, diffuse proliferative nephritis (class IV) in 2 (both had nephritic flares), and membranous nephritis (class Vb) in the remaining patient. Reinduction therapies (CYC in 3 with relapse of proliferative nephritis and highdose methylprednisolone in the remaining 3) were given for these patients, and all achieved remission (CR in 5 and PR in 1). After successful

272

MOK ET AL Table 2.

Renal and Serologic Parameters at 0 and 12 Months After Treatment

Proteinuria (g/d) Urinary casts (%) Serum creatinine (mg/dL) Creatinine clearance (mL/min) Serum albumin (g/dL) Nephrotic syndrome (%) Hypertension (%) Anti-dsDNA (%) Depressed serum C3 (%) Depressed serum C4 (%) SLEDAI scores

0 mo

12 mo

P Value

3.91 ⫾ 2.2 5 (14) 73.0 ⫾ 16 79.6 ⫾ 28 28.2 ⫾ 6.8 21 (58) 4 (11) 19 (53) 24 (67) 12 (33) 11.5 ⫾ 4.3

1.19 ⫾ 1.7 0 (0) 73.0 ⫾ 13.7 91.5 ⫾ 17.7 38.4 ⫾ 6.4 3 (8) 5 (14) 10 (28) 13 (36) 6 (17) 3.0 ⫾ 2.6

⬍0.001 0.03 0.90 0.03 ⬍0.001 ⬍0.001 0.32 0.008 0.005 0.06 ⬍0.001

NOTE. To convert serum creatinine in mg/dL to ␮mol/L, multiply by 88.4; creatinine clearance in mL/min to mL/s, multiply by 0.01667; serum albumin in g/dL to g/L, multiply by 10.

reinduction, they were maintained again on lowdose prednisone and AZA. The 4 patients who were initially refractory to corticosteroid and AZA treatment received salvage therapy with other cytotoxic agents (CYC in 1, cyclosporin A in 2, and mycophenolate mofetil in 1) and all achieved CR. Two of these patients received angiotensin-converting enzyme inhibitors in addition. The outcome of our cohort of patients with pure membranous lupus nephritis is summarized in Fig 1. Cox regression was used to analyze the factors predicting renal relapse after initial remission (CR or PR) to prednisone and AZA (data not shown). Unfavorable factors for renal flares included Vb histology, presence of anti-nRNP, and

CR at 12 months posttreatment, but, again, none of these were statistically significant. At the end of study, 5 (13%) patients had loss of 20% of creatinine clearance, but none had doubling of serum creatinine level. No particular clinical variables were shown to be associated with renal function decline. Table 3 shows the renal outcome of the patients with regard to whether nephrotic syndrome was present at the onset of nephritis. Patients presenting with nephrotic syndrome had a lower rate of CR and higher rate of NR, but the difference was not significant. The proportion of patients with loss of creatinine clearance was similar to those without nephrotic syndrome at the time of renal biopsy. Thromboembolic events and treatment-related

Fig 1. Outcome of patients with pure membranous lupus nephritis treated with prednisone and AZA. Abbreviations: P, prednisone; CSA, cyclosporin A; MMF, mycophenolate mofetil; HD, high dose.

OUTCOME OF PURE MEMBRANOUS LUPUS NEPHRITIS Table 3.

273

Renal Outcome of Pure Membranous Lupus Nephropathy With Regard to the Presence of Nephrotic Syndrome at Nephritis Onset Nephrotic Syndrome

Remission status at 12 mo Complete remission (%) Partial remission (%) Nonremission (%) Loss of 20% of creatinine clearance (%)

complications are summarized in Table 4. Two patients had idiosyncratic reactions to AZA, whereas 3 patients had corticosteroid-induced avascular necrosis of the hips. Severe infective complications, malignancies, and sustained amenorrhea were not reported. Five patients had thromboembolic events (arterial thrombosis in 4, and venous thrombosis in 1) during the course of follow-up. Among these patients, 3 were positive for the IgG-anticardiolipin antibodies, and 1 was initially treatment refractory with persistent nephrotic syndrome. None of our patients had renal vein thrombosis. DISCUSSION

This is an open-label study of the therapeutic efficacy of prednisone and AZA in pure or minimally proliferative membranous lupus nephropathy. Our results show that a combination of prednisone and AZA is reasonably effective for the initial treatment and maintenance of pure membranous lupus nephritis. Renal remission occurred in 90% of patients in the first year of Table 4. Thromboembolic Events and TreatmentRelated Complications in Our Patients (n ⴝ 38) Complications

No. (%)

Thromboembolic events Cerebrovascular accident Cardiovascular disorder Amaurosis fugax Pulmonary embolism Avascular necrosis of the hips Agranulocytosis (AZA) Leucopenia Hepatitis AZA hypersensitivity Herpes zoster Severe infection Malignancy

2 (5) 1 (3) 1 (3) 1 (3) 3 (8) 1 (3) 2 (5) 0 (0) 1 (3) 1 (3) 0 (0) 0 (0)

Yes (n ⫽ 21)

No. (n ⫽ 15)

P Value

13 (62) 5 (24) 3 (14) 3 (14)

11 (73) 3 (20) 1 (7) 2 (13)

— — 0.71 1.00

treatment. Extrarenal disease activity also improved significantly after treatment. With a mean follow-up of almost 8 years, only a small proportion of patients lost 20% of their creatinine clearance. No patients had doubling of serum creatinine. Relapse of nephritis was infrequent, and all patients could be reinduced to remission with salvage therapies. Serious adverse effects related to the treatment protocol were uncommon. The overall prognosis of lupus nephritis in southern Chinese patients is similar to that reported in the series with white patients, with a 10-year renal survival rate of 81%.1 Pure membranous lupus nephropathy was associated with a much better renal outcome when compared with the diffuse proliferative type of lupus nephritis. However, the prognosis of lupus nephritis is much worse in certain ethnic groups such as the African- and Hispano-Americans.20,21 Thus, the current protocol of prednisone and AZA combination may not be as effective in these ethnic groups. The natural history of lupus membranous nephropathy is uncertain because patients often were treated with immunosuppression for either renal or extrarenal disease. Compared with proliferative lupus nephritis, membranous nephropathy runs a more indolent course. A number of studies in the past decade have reported the outcome of lupus membranous nephropathy.8,11,12,16,22-24 These are summarized in Table 5. Comparison among these studies is confounded by the differences in ethnicity of patients, baseline severity of renal disease, treatment protocols, duration of follow-up, and whether the Vc and Vd subclasses were included for analysis. In general, the 10-year renal survival rate exceeds 72%, which is better than that of proliferative

274

MOK ET AL Table 5.

Study

Current study

Prognosis of Membranous Nephropathy in SLE

No.

Nephrotic at Presentation

Histologic Subclasses

38

58%

All Va/Vb

64%

Treatment Protocol

Follow-Up

Outcome

Pred ⫹ AZA

90.4 mo

All Va/Vb

No treatment (24%) Pred ⫾ CYC/ CSA/AZA (76%)

6.9 y

CR (67%), PR (22%), NR (17%) 19% renal flares 13% patients had 20% decline in CrCl Renal survival at 5, 10 y ⫽ 97%, 88% 20% doubling of Cr, 9% ESRF, 49% renal flares Steroid did not affect outcome; multivariate predictors for renal function deterioration were thrombosis and relapse of nephritis Improvement in proteinuria in all patients

Mercadal et al,22 2002

66 (47% blacks)

Hallegua et al,12 2000

10 (50% Asians/ Hispanics/ blacks) 19

NS

NS

Pred ⫹ CSA

24.8 mo

All

Va/Vb (79%) Vc (21%)

Pred alone (42%) MP ⫹ CHL (58%)

114 mo

Chan et al,16 1998

20

All

All Va/Vb

Pred ⫹ oral CYC (6/12) followed by AZA

74 mo

Sloan et al,8 1996

79

NS

Va/Vb (46%) Vc/Vd (54%)

Pred ⫾ CYC/ AZA/apheresis

4.3 y

Radhakrishnan et al,23 1994

10

All

Va/Vb (70%) Vc (30%)

23-43 mo

Pasquali et al,24 1993

42

64%

Va/Vb (62%) Vc/Vd (38%)

Pred ⫹ CSA (80%) CSA alone (20%) Pred alone (33%) Pred ⫹ cytotoxic agents (67%)

Moroni et al,11 1998

6y

More remission (CR/PR) and less relapses with combination treatment Doubling of creatinine/ ESRF (38% in Pred alone group; 9% in Pred ⫹ CHL group) At 12 mo, CR (55%), PR (35%) Renal function stable in all, 40% renal relapses Va/Vb better prognosis; 10-y renal survival (72% for Va/Vb, 20-49% for Vc/Vd) Increasing Cr predicted ESRF in Vc/Vd but not in Va/Vb CR (60%), PR (20%), NR (20%) Serum creatinine not significantly elevated 10-y renal survival 93% CR (38% in Va/Vb; 6% in Vc/Vd) 19% with thrombotic complications

Abbreviations: Pred, prednisone; CrCl, creatinine clearance; Cr, creatinine; CSA, cyclosporin A; MP, methylprednisolone; CHL, chlorambucil; ESRF, end-stage renal failure; NS, not stated.

lupus nephritis. Patients with Vc and Vd histology have worse outcome in terms of remission rate, relapses, and renal function deterioration than those with Va and Vb disease. The renal outcome of our patients appears to be relatively

good. This can be for several reasons. First, only patients with Va and Vb membranous nephropathy were recruited. Second, nephrotic syndrome occurred in just 58% of the patients at presentation, and none had abnormal serum creatinine.

OUTCOME OF PURE MEMBRANOUS LUPUS NEPHRITIS

Third, all patients were maintained on long-term low-dose prednisone and AZA. Finally, the follow-up time of 90 months may not be long enough for membranous nephropathy to evolve into renal failure. Few clinical predictors of renal function deterioration in membranous lupus nephropathy have been identified. This can be related to the relatively small sample size in various studies. Apart from the histologic subtypes (Va/Vb v Vc/Vd), increasing serum creatinine level on presentation was reported to be an independent predictor for end-stage renal failure in 1 study,8 but the predictive value was no longer significant when only Va and Vb disease was considered. Although patients with more severe proteinuria were at higher risk of thromboembolic and vascular complications, whether the amount of proteinuria at presentation might predict renal function decline has not been confirmed with multivariate data. In our cohort of patients who received a relatively uniform treatment protocol, the degree of proteinuria, serum creatinine level, and nephrotic syndrome at presentation did not significantly correlate with remission, relapse, or renal function deterioration. Other clinical characteristics such as age, sex, presence of hypertension, hypocomplementemia, and serum albumin levels were also not found to be associated with renal outcome. Eighty percent of our patients with thromboembolic events on follow-up had actually achieved remission after treatment, and the events were probably related to the presence of the antiphospholipid antibodies. Pure membranous lupus nephropathy remains a therapeutic enigma. From the limited information in the literature, it appears that a combination of corticosteroid and cytotoxic agents is better than corticosteroid alone.11,17 AZA is a reasonable initial choice of cytotoxic therapy because its toxicity profile has been well documented in renal transplant recipients. In our study, long-term maintenance with low-dose AZA (1 to 2 mg/kg/d for a mean of 77 months) was not associated with significant toxicities such as malignancies and severe infections. Idiosyncratic reactions to AZA might have been predicted by measuring the thiopurine methyltransferase activity before treatment.25 Moreover, AZA is beneficial in other concomitant manifestations of SLE such as neuropsychiatric, hematologic, dermato-

275

logic, and serosal disease, as well as a corticosteroid-sparing agent. This is shown in our study by the significant improvement in disease activity scores and seroconversion of anti-dsDNA after treatment. Thus, prednisone and AZA may be considered for the initial treatment of pure membranous lupus nephropathy. More aggressive cytotoxic therapy is indicated in blacks and Hispanics, as well as in patients with impaired renal function at nephritis onset or those who are resistant to AZA treatment. Randomized, controlled trials certainly are needed to evaluate whether the efficacy of prednisone/AZA combination is better than prednisone alone in the treatment of this condition. REFERENCES 1. Mok CC, Wong RW, Lau CS: Lupus nephritis in Southern Chinese patients: Clinicopathologic findings and long-term outcome. Am J Kidney Dis 34:315-323, 1999 2. Huong DL, Papo T, Beaufils H, et al: Renal involvement in systemic lupus erythematosus: A study of 180 patients from a single center. Medicine (Baltimore) 78:148166, 1999 3. Donadio JV Jr, Hart GM, Bergstralh EJ, Holley KE: Prognostic determinants in lupus nephritis: A long-term clinicopathologic study. Lupus 4:109-115, 1995 4. Yang LY, Chen WP, Lin CY: Lupus nephritis in children—A review of 167 patients. Pediatrics 94:335-340, 1994 5. Neumann K, Wallace DJ, Azen C, et al: Lupus in the 1980s: III. Influence of clinical variables, biopsy, and treatment on the outcome in 150 patients with lupus nephritis seen at a single center. Semin Arthritis Rheum 25:47-55, 1995 6. Adler SG, Johnson K, Louie JS, Liebling MR, Cohen AH: Lupus membranous glomerulonephritis: Different prognostic subgroups obscured by imprecise histologic classifications. Mod Pathol 3:186-191, 1990 7. Schwartz MM, Kawala K, Roberts JL, Humes C, Lewis EJ: Clinical and pathological features of membranous glomerulonephritis of systemic lupus erythematosus. Am J Nephrol 4:301-311, 1984 8. Sloan RP, Schwartz MM, Korbet SM, Borok RZ: Long-term outcome in systemic lupus erythematosus membranous glomerulonephritis. Lupus Nephritis Collaborative Study Group. J Am Soc Nephrol 7:299-305, 1996 9. Churg J, Bernstein J, Glassock RJ: Renal Disease. Tokyo, Ikagu-Shoin, 1995, pp 151-179 10. Barnett EV, Dornfeld L, Lee DB, Liebling MR: Longterm survival of lupus nephritis patients treated with azathioprine and prednisone. J Rheumatol 5:275-287, 1978 11. Moroni G, Maccario M, Banfi G, Quaglini S, Ponticelli C: Treatment of membranous lupus nephritis. Am J Kidney Dis 31:681-686, 1998 12. Hallegua D, Wallace DJ, Metzger AL, Rinaldi RZ, Klinenberg JR: Cyclosporine for lupus membranous nephritis: Experience with ten patients and review of the literature. Lupus 9:241-251, 2000

276

13. Tam LS, Li EK, Szeto CC, et al: Treatment of membranous lupus nephritis with prednisone, azathioprine and cyclosporin A. Lupus 10:827-829, 2001 14. Boumpas DT, Tassiulas IO, Fleisher TA, et al: A pilot study of low-dose fludarabine in membranous nephropathy refractory to therapy. Clin Nephrol 52:67-75, 1999 15. Kolasinski SL, Chung JB, Albert DA: What do we know about lupus membranous nephropathy? An analytic review. Arthritis Rheum (Arthritis Care Res) 47:450-455, 2002 16. Chan TM, Li FK, Hao WK, et al: Treatment of membranous lupus nephritis with nephrotic syndrome by sequential immunosuppression. Lupus 8:545-551, 1999 17. Austin HA, Vaughan EM, Balow JE: Lupus membranous nephropathy: Randomized controlled trial of prednisone, cyclosporine and cyclophosphamide. J Am Soc Nephrol 11:81A, 2000 (abstr) 18. Gladman DD, Goldsmith CH, Urowitz MB, et al: Sensitivity to change of 3 systemic lupus erythematosus disease activity indices: International validation. J Rheumatol 21:1468-1471, 1994 19. Tan EM, Cohen AS, Fries JF, et al: The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 25:1271-1277, 1982

MOK ET AL

20. Dooley MA, Hogan S, Jennette C, Falk R: Cyclophosphamide therapy for lupus nephritis: Poor renal survival in black Americans. Glomerular Disease Collaborative Network. Kidney Int 51:1188-1195, 1997 21. Alarcon GS, McGwin G Jr, Bartolucci AA, et al: Lupus in minority populations, nature versus nurture. Systemic lupus erythematosus in three ethnic groups. IX. Differences in damage accrual. LUMINA Study Group. Arthritis Rheum 44:2797-2806, 2001 22. Mercadal L, Montcel ST, Nochy D, et al: Factors affecting outcome and prognosis in membranous lupus nephropathy. Nephrol Dial Transplant 17:1771-1778, 2002 23. Radhakrishnan J, Kunis CL, D’Agati V, Appel GB: Cyclosporine treatment of lupus membranous nephropathy. Clin Nephrol 42:147-154, 1994 24. Pasquali S, Banfi G, Zucchelli A, Moroni G, Ponticelli C, Zucchelli P: Lupus membranous nephropathy: Longterm outcome. Clin Nephrol 39:175-182, 1993 25. Stolk JN, Boerbooms AM, de Abreu RA, et al: Reduced thiopurine methyltransferase activity and development of side effects of azathioprine treatment in patients with rheumatoid arthritis. Arthritis Rheum 41:1858-1866, 1998