The Official Journal of the
National Kidney Foundation
AJKD
VOL 44, NO 3, SEPTEMBER 2004
American Journal of Kidney Diseases
REVIEW
Immunosuppressive Treatment for Idiopathic Membranous Nephropathy: A Systematic Review Annalisa Perna, StatSciD, Arrigo Schieppati, MD, Javier Zamora, PhD, Giovanni A. Giuliano, DiplStat, Norbert Braun, MD, and Giuseppe Remuzzi, MD ● Background: This study aims to assess whether immunosuppression is beneficial in the treatment of idiopathic membranous nephropathy (IMN). Methods: We reviewed randomized controlled trials (RCTs) addressing the effect of immunosuppression on histologically proven IMN in adults with nephrotic syndrome followed up for at least 6 months. The literature was searched according to Cochrane Collaboration guidelines. Four therapeutic classes were considered: (1) steroids (alone), (2) alkylating agents (alone or in combination with steroids), (3) calcineurin inhibitors (alone or in combination with steroids), and (4) antiproliferative agents (alone). Results: Eighteen RCTs were selected (1,025 patients). Overall, no differences were found between immunosuppressive treatment and placebo or no treatment. For oral glucocorticoid therapy, no beneficial effect was observed. For alkylating agents, a beneficial effect was observed on complete remission (relative risk [RR], 2.37; 95% confidence interval [CI], 1.32 to 4.25; P ⴝ 0.004). Excluding patients with particularly well-preserved renal function, there was no significant difference in complete remission. Cyclophosphamide therapy resulted in a lower rate of adverse-event discontinuations compared with chlorambucil (8 versus 21 discontinuations, respectively; RR, 2.34; 95% CI, 1.25 to 4.39; P ⴝ 0.008). For calcineurin inhibitors, no difference was observed. For antiproliferative agents, a paucity of data did not allow a conclusion. Conclusion: The meta-analysis failed to show a long-term effect of treatment on patient and/or renal survival. There is weak evidence that the immunosuppressive regimen increased the remission rate. The review of safety showed a higher number of discontinuations for adverse events in immunosuppressive-treatment groups and that cyclophosphamide had fewer side effects than chlorambucil. Am J Kidney Dis 44:385-401. © 2004 by the National Kidney Foundation, Inc. INDEX WORDS: Idiopathic membranous nephropathy (IMN); adult; nephrotic syndrome (NS); immunosuppressive therapy; systematic review.
Editorial, p. 562
From Mario Negri Institute for Pharmacological Research; Nephrology and Dialysis Unit, Ospedali Riuniti di Bergamo, Italy; and Section of Nephrology and Hypertension, University of Tuebingen, Germany. Received January 14, 2004; accepted in revised form May 10, 2004. This review is excerpted from a Cochrane Review that will be published in The Cochrane Library 2004, Issue 4 (http:// www.update-software.com/cochrane). Cochrane Reviews are regularly updated as new evidence emerges in response to
I
DIOPATHIC MEMBRANOUS nephropathy (IMN) is the most common form of nephrotic syndrome (NS) in adults.1 The disease shows a benign or indolent course in the majority
comments and criticisms, and The Cochrane Library should be consulted for the most recent version of the Review. Address reprint requests to Annalisa Perna, StatSciD, Clinical Research Centre for Rare Diseases Aldo e Cele Daccò, Mario Negri Institute for Pharmacological Research, Via Gavazzeni 11, 24125 Bergamo, Italy.Email address:
[email protected] © 2004 by the National Kidney Foundation, Inc. 0272-6386/04/4403-0001$30.00/0 doi:10.1053/j.ajkd.2004.05.020
American Journal of Kidney Diseases, Vol 44, No 3 (September), 2004: pp 385-401
385
386
of patients, with a rate of spontaneous complete or partial remission of NS as high as 30% or more.2 However, a non-negligible proportion of patients (30% to 40%) progress toward end-stage renal failure in 5 to 15 years.3 Several therapeutic regimens have been proposed; namely, oral glucocorticoids, alkylating agents (chlorambucil and cyclophosphamide), and, lately, cyclosporine.4 The uncertain risk-benefit ratio of corticosteroid and cytotoxic therapy in the treatment of membranous nephropathy and the lack of definitive evidence that cytotoxic drugs alter the longterm course of the disease make optimal therapy difficult to determine. Because almost all studies involved at most only a few tens of patients, data were combined through different approaches,5-7 but yielded conflicting conclusions and still maintained uncertainty on the most appropriate treatment. Couchoud et al5 combined 8 randomized controlled trials (RCTs) of glucocorticoid and cytotoxic and antiproliferative immunosuppressive agents, including 526 randomized patients, and found that with 2 surrogate criteria, such as “improvement in proteinuria” and “impairment of renal function,” there was a tendency in favor of immunosuppressive treatment, but no effect on the 2 absolute criteria of “complete remission” and “renal death.” Hogan et al6 performed a pooled analysis of 35 studies of 1,815 subjects (of whom 475 were included in RCTs)8-13 on corticosteroid and alkylating therapy, combining retrospective and prospective studies, both controlled and uncontrolled. Complete remission of NS was observed more frequently with the use of alkylating agents, combining retrospective and prospective studies, both controlled and uncontrolled. Complete remission of NS was observed more frequently with the use of alkylating agents compared with no treatment or corticosteroids, which were no better than no treatment. Corticosteroids or alkylating agents did not improve renal survival. Imperiale et al7 included 5 prospective studies, 4 RCTs, and 1 non-RCT in which cytotoxic agents were compared with steroids, placebo, or symptomatic treatment. Response to therapy was assessed by means of short-term responses, achieving complete or partial remission of proteinuria. The investigators found a beneficial effect of cytotoxic agents on these short-term responses among the 228 subjects
PERNA ET AL
enrolled, of whom 202 were included in RCTs.11-14 Conversely, they were unable to address the question of the effects of this therapy on long-term renal function. The 2 more recent meta-analyses6,7 based their search strategy on MEDLINE, 1968 to 1993. By December 15, 2003, we developed a new and updated bibliographic search.15 This sensitive approach allowed the identification of 1 more relevant trial before 1994 on a comparison between alkylating agents and steroids in 71 randomized patients16 not identified in the previous meta-analyses, 1 trial of azathioprine,17 1 trial of alkylating agents,18 and 8 more randomized trials from 1994 to December 15, 2003; 5 published19-23 as full articles and 3 published24-26 in abstract form, regarding approximately 400 more new patients. Apart from the use of steroids alone compared with placebo or no treatment, for which the new search strategy did not show additional trials since the publication of the metaanalysis by Hogan et al,6 we considered that there was a substantial increase in available new data for an updated systematic review of the effect of different immunosuppressive agents on need for dialysis, death, and remission of NS, as well as on changes in 24-hour urinary protein excretion, glomerular filtration rate (GFR), and serum creatinine level. METHODS
Trial Inclusion Criteria Data were searched for from reports of RCTs, published as full articles, in abstract form, or partially/totally unpublished, addressing the effect of immunosuppressive agents on patients with NS. Studies should have a planned follow-up of at least 6 months. Both parallel and the first period (of at least 6 months) of cross-over randomized studies could be eligible for the review. In case of only 1 report published as a full article, interim data not followed by a subsequent publication were considered suitable for inclusion. In the case of more than 1 report published as full articles on the same patients, the primary reference showing final results was chosen for the meta-analysis, excluding both interim and follow-up reports. For results published in abstract form, the first investigator was contacted to determine whether a publication was published recently, in press, or submitted to a medical journal or, alternatively, whether unpublished material could be released for the present metaanalysis. Because of limitations of reporting, abstract data were included only if an implementation through additional adequate information was provided by the investigators. The selected patients were adult subjects with IMN who had NS. The definition of adult was according to the single
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY
studies’ inclusion criteria: usually the cutoff value of 18 years was considered, but some trials referred to a lower cutoff value.8-10,12,22 The diagnosis of IMN was histologically proven. Secondary forms of membranous nephropathy had to be excluded. NS assessment relied on that chosen by the investigators in the single studies. It must be said that this definition was heterogeneous. In trials that included a minority of non-nephrotic subjects, when possible, analyses were restricted to nephrotic patients only. In the absence of an explicit NS definition, a cutoff value for urinary protein excretion greater than 3.5 g/24 h was used.
Search Techniques Electronic databases, MEDLINE, EMBASE (1966 to December 15, 2003), and the Cochrane Central Register of Controlled Trials (issue 4, 2003) were searched, regardless of the article language, with the Cochrane Renal Group trial coordinator using the optimally sensitive strategy developed for the Cochrane Collaboration for the identification of RCTs.15 The following medical subject heading terms and text words were used: glomerulonephritis, membranous, membranous nephropathy, membranous glomerulopathy, MGN, extramembranous glomerulopathy, membranous glomerulonephropathy, idiopathic membranous glomerulonephritis, and idiopathic membranous nephropathy. Reference lists of retrieved reports were checked. Abstract books of the American Society of Nephrology, the European Dialysis and Transplantation Society, and the Gesellschaft für Nephrologie (Germany, Switzerland, and Austria) were systematically hand searched. Electronic search in the Journal of Nephrology was supplemented by hand searching from 1989 to February 1997. Furthermore, several investigators of RCTs in the field of nephrology were contacted since 1994 for identifying planned and ongoing RCTs with the aim to set up and update a “Prospective Registry of RCTs in Nephrology.”27 Other sources for identifying ongoing studies were the metaRegister of Current Controlled Trials Ltd28 and The Lancet29 Web site specifically devoted to RCT protocols.
Data Collection The review was undertaken by 6 reviewers (A.P., A.S., J.Z., G.A.G., N.B., and G.R.). The search strategy was independently performed by 2 reviewers (A.P. and G.A.G.). Two independent reviewers (A.P. and J.Z.) inspected all articles identified in their abstract form. For studies that could possibly have been RCTs or in case of disagreement between the 2 reviewers, the full article was obtained. In turn, the same reviewers, who were not blinded to authorship or journal, compiled an ad hoc questionnaire and independently revised these articles. The questionnaires then were cross-checked, and disagreements were resolved by consensus or with a third reviewer (A.S.). The following general descriptive information was extracted from each trial: number, age, and sex of participants; baseline renal function evaluated by means of 24-hour urinary protein excretion, serum creatinine level, creatinine clearance, and GFR; number of participants assigned to each intervention; and definite and surrogate end points. Outcomes were evaluated at the last available follow-up. For articles published in
387
abstract form,24-26 it was possible to retrieve all relevant additional information by contacting the first investigator25 and considering, when available, a later publication of the trial as full article.30 In 1 case,24 data from the abstract referred to final results and therefore were used for the review, supplemented with some unpublished material. For the other study,26 the abstract reported preliminary data. Although no substantial differences occurred compared with the subsequent unpublished data, it was decided to use the latter because it was defined “final” by the investigators. First, trials were inspected for 3 principal aspects of the randomization process: (1) generation of allocation sequence, (2) allocation concealment, and (3) blinding. The first and second aspects refer to the prevention of foreknowledge of treatment assignment.31 The latter refers to measures taken to mask group identity after allocation. Following the method of Shultz et al,31 the following approaches to the generation of allocation sequences were considered adequate: computer, random number table, shuffled cards or tossed coins, and minimization. The following allocation concealment strategies were considered adequate: central randomization; numbered or coded containers; or sequentially numbered, opaque, or sealed envelopes.31 Furthermore, it was assessed whether the analysis was by intention to treat and the completeness of follow-up. The following Cochrane quality checklist was compiled: (1) allocation concealment: adequate/unclear/inadequate; (2) blinding: blinding of investigators (yes/no/not stated), blinding of participants (yes/no/not stated), blinding of outcome assessors (yes/no/not stated), blinding of data analysis (yes/no/not stated); (3) intention-to-treat: (A) yes; specifically reported by investigators that intention-to-treat analysis was undertaken and this was confirmed on study assessment; (B) yes; not stated, but confirmed on study assessment; (C) no; not reported and lack of intention-to-treat analysis confirmed on study assessment (patients who were randomized were not included in the analysis because they did not receive the study intervention, withdrew from the study, or were not included because of protocol violation); (D) no; stated, but not confirmed on study assessment; and (E) not stated; unable to be determined on study assessment; and (4) completeness of follow-up: percentage of participants lost to follow-up or with no data for the primary outcome of effectiveness. Furthermore, 2 additional important aspects regarding the diagnosis of IMN32 were evaluated in the questionnaire: whether the article described the diagnostic criteria used in the trial and if the diagnosis was confirmed by using immunohistochemical microscopy or electron microscopy. A recent methods meta-analysis evaluating studies on the prognosis of IMN32 discovered that several RCTs were not of higher quality compared with some nonrandomized ones. Some of the RCTs evaluated by Marx and Marx32 are included in the present review.8-10,14,33
Types of Interventions The following 4 classes of immunosuppressive treatments were identified: (1) glucocorticoids (alone), (2) alkylating agents (alone or in association with glucocorticoids), (3) calcineurin inhibitors (alone or in association with glucocorticoids), and (4) antiproliferative agents (alone).
388
Control groups were administered placebo or no treatment other than supportive therapy. The following interventions, nonimmunologic in nature, were excluded: drugs aimed to reduce proteinuria through inhibition of the renin-angiotensin system (eg, angiotensinconverting enzyme [ACE] inhibitors), antialdosterone drugs (eg, spironolactone), nonsteroidal anti-inflammatory drugs (eg, indomethacin), and drugs aimed to correct dyslipidemia (eg, statins, diet). Conversely, it was attempted to consider the possible confounding effect of these nonimmunosuppressive treatments as concomitant medications used in the studies included in the review.
Outcome Measures One or more of the following outcome measures for efficacy were considered: (1) definite end points: death or end-stage renal failure that requires initiation of dialysis therapy or renal transplantation; and (2) surrogate end points: partial remission; complete remission; final proteinuria, measured as grams of protein per 24 hours; final serum creatinine level, measured as micromoles per minute; and final GFR, measured as milliliters per minute per 1.73 m2. We adopted the same definitions of partial and complete remission of NS as those used by the study investigators. Usually, the most frequent definitions adopted were: partial remission, proteinuria with protein between 0.2 and 2 g/24 h and a serum creatinine level of 2.0 mg/dL or less (ⱕ180 mol/L); and complete remission, proteinuria with protein less than 0.2 g/24 h and a serum creatinine level of 2.0 mg/dL or less (ⱕ180 mol/L). We found the definitions heterogeneous, and although this outcome measure can be considered an acceptable “definite” outcome measure in IMN, we decided to include this outcome among “surrogate,” and not “definite,” end points. The number of patients at risk usually was the same for surrogate and hard end points and for safety; however, occasionally, for the evaluation of surrogate end points (ie, complete and/or partial remission), it may differ.
Statistical Analysis Data were quantitatively combined by 2 independent reviewers (A.P. and G.A.G.) through a meta-analysis undertaken using Review Manager 4.2. The calculated relative risk (RR) with 95% confidence interval (CI) for dichotomous data and a weighted mean difference for continuous data were determined. A random-effects model was used for analyses of both dichotomous and continuous data. Additionally, robustness of the model chosen was evaluated by using the fixed-effects model. When analyzing unfavorable events (death, need for dialysis therapy or transplantation, or adverse events), an RR less than 1 favors the experimental treatment, whereas an RR greater than 1 favors the control treatment. In the case of favorable events (partial or complete remission), an RR less than 1 favors the control treatment, whereas an RR greater than 1 favors the experimental treatment. Statistical heterogeneity across studies was explored by using chi-squared test for heterogeneity, automatically performed by Review Manager. Because this is a low-power test, the ␣ level was set at 0.10. Two main sources of heterogeneity across trials were considered a
PERNA ET AL
priori: trial quality and characteristics of study participants. Furthermore, it can be argued that although recommended by the Cochrane Collaboration, the inclusion of totally or partially unpublished material can be questionable because it is not possible to verify unpublished trial data. To address this issue, studies providing partially unpublished data were both included and excluded for sensitivity analysis. Additionally, the following potential confounding factors were considered: inclusion/exclusion criteria regarding baseline renal function (preserved/declining), unbalances in main baseline characteristics, and use of concomitant medication(s), such as ACE inhibitors or statins.
RESULTS
Figure 1 shows progress through the different stages of this systematic review. From MEDLINE, 753 citations were retrieved; from EMBASE, 254 citations; and from the Cochrane Central Register of Controlled Trials, 60 citations. Overall, 943 potentially relevant publications were identified, discarding duplications. Subsequently, 896 articles were excluded, being nonrandomized, retrospective series, review articles, and so on. Forty-seven studies were evaluated in a closer inspection: 16 studies were discarded because they were non-RCTs, and 6 other RCTs were excluded34-39 (Fig 1; Table 1). Twenty-five trial reports were considered eligible to be included in the meta-analysis: 4 of these subsequently were excluded, being duplications or interim/early reports of main publications12,40-42 (Fig 1; Table 1). Overall, 18 RCTs8-11,13,14,16-26,33 published as full articles or in abstract form were identified and included in the present review. The total number of patients randomized in the included trials was 1,025. Ten studies provided data on 1 or more outcomes for the comparison between immunosuppressive treatment versus placebo or no treatment. Of these, 3 trials tested the effect of steroids alone8-10; 4 trials, of alkylating agents alone or in combination with steroids11,13,26,33; 1 trial, of azathioprine17; and 3 trials, of cyclosporine.21,24,26 Five studies14,16,18,19,25 compared 2 different classes of immunosuppressive treatments: 4 studies evaluated the effect of adding an alkylating agent14,16,18,19 and 1 study, of adding cyclosporine,25 to therapy based on steroids. The remaining 3 studies20,22,23 compared the effect of 2 specific immunosuppressive treatments within the class of alkylating agents. Main trial and patient characteristics of the RCTs are listed in Table 2. The median average follow-up period
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY
Fig 1. Flow diagram of the systematic review.
389
390
PERNA ET AL Table 1.
Excluded Studies
Study Identification
Reason for Exclusion
Ambalavanan et al34
Randomized crossover study of cyclosporine v ACE inhibitor; excluded because the first crossover period was ⬍ 6 mo (3 mo) RCT of chlorambucil, azathioprine, and placebo, excluded because it was possible to isolate the NS subgroup for 1 of the treatment groups only RCT; excluded because it was not possible to identify patients with NS RCT; excluded because it was not possible to identify patients with NS RCT; excluded because all participants were children RCT; excluded because the control group was administered ACE inhibitor RCT; excluded because it was an early report of Ponticelli, 198933 RCT; excluded because it was an early report of Ponticelli, 198933 RCT; although at first sight, this report seemed to be the primary reference, Ponticelli, 1989,33 reported more complete results with an increased number of patients of the same protocol; therefore, Ponticelli, 1984,12 was considered an interim analysis of Ponticelli, 198933 RCT; this is a longer follow-up report corresponding to the Ponticelli 198933 study
Lagrue et al35
Medical Research Council Working Party36 Nand et al37
Tejani et al38 Yao et al39
Imbasciati et al40
Ponticelli et al41
Ponticelli et al12
Ponticelli et al42
reported in the trials was 24 months (range, 12 to 60 months). Method Quality of Included Studies Reporting of trial method details was unsatisfactory and incomplete for the majority of trials (Table 3). Nine9-11,14,17,22,24-26 of the 18 studies included reported, or additional information was provided, on adequate allocation concealment. In the remaining 9 trials, it was reported that the trial was randomized, but no information on the method was
given. Three trials8,9,17 only reported blinding of both investigators and participants. Three trials11,24,26 reported, or additional information was provided, on the blinding of outcome assessors or data analysts. Intention-to-treat analysis was confirmed in 7 trials,9,10,14,17,20,24,33 whereas it was not undertaken in 6 trials11,13,16,22,23,26 and was unclear in the remaining 5 trials.8,18,19,21,25 Information regarding completeness of follow-up was reported or deducible in all trials, ranging from 81% to 100%. Comparison of Immunosuppressive Regimens Versus Placebo or No Treatment We analyzed the efficacy of all combined immunosuppressive regimens versus placebo or no treatment. Some unbalances were observed in the proportion of males, serum creatinine levels, and proteinuria at randomization, which were greater in the treatment versus control groups. Fourteen deaths and 49 cases of end-stage renal failure were reported. Seven patients died in the immunosuppressive groups versus 7 patients in the control groups (RR, 1.00; 95% CI, 0.34 to 2.90; P ⫽ 0.99), whereas 23 patients required dialysis therapy in the experimental group versus 26 patients in the control group (RR, 0.74; 95% CI, 0.38 to 1.44; P ⫽ 0.37). Overall, 30 patients reached the combined end point of death, dialysis therapy, or transplantation in the immunosuppressive groups versus 32 patients in the control groups (RR, 0.82; 95% CI, 0.49 to 1.40; P ⫽ 0.47). There was no heterogeneity between trials (P ⫽ 0.38; Fig 2). No significant differences were observed in the number of partial (RR, 1.25; 95% CI, 0.83 to 1.88; P ⫽ 0.29), complete (RR, 1.36; 95% CI, 0.82 to 2.24; P ⫽ 0.23; Fig 3), or complete or partial remissions of proteinuria (RR, 1.30; 95% CI, 0.86 to 1.96; P ⫽ 0.21), as far as in final proteinuria, final serum creatinine level, and final GFR (data not shown). There were 17 discontinuations caused by adverse events in immunosuppressive treatment. The most frequently experienced adverse events were leukopenia (5 events), cushingoid features (5 events), and gastric disorders (2 events).
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY Table 2.
Reference
Immunosuppressive treatment v no treatment Steroids v no treatment Cameron et al9 Cattran et al10 Coggins8 Alkylating agents v no treatment Braun et al26 Donadio et al11 Murphy et al13 Ponticelli et al33 Cyclosporine v no treatment Braun et al*26 Cattran et al21 Pisoni et al*24 Azathioprine v no treatment Silverberg17 Two different classes of immunosuppressive treatments Steroids v alkylating agents Ahmed et al19 Falk et al18 Pahar et al16 Ponticelli et al14 Cyclosporine v steroids Cattran et al25 Cyclosporine v alkylating agents Braun et al*26 Two different immunosuppressive treatments within the same class Chlorambucil v cyclophosphamide Branten et al20 Ponticelli et al22
Reichert23
391
Summary of Included Studies
No. of Patients
Mean Follow-Up (mo)
Mean Age at Randomization (y)
Sex (% men)
Declining Renal Function at Baseline
103 158 72
52 48 23
45 45 39
84 66 58
No No No
Not mentioned Not mentioned Not mentioned
53 22 40 81
60 12 24 60
44 44 43 43
72 77 65 78
No No No No
Yes, no confounding effect Not mentioned Not mentioned Not mentioned
66 17 21
60 21 12
44 42 49
52 82 81
No Yes Yes
Yes, no confounding effect Yes, no confounding effect Yes, no confounding effect
9
12
43
67
No
Not mentioned
20 26 71 92
15 24 46 48
35 45 34 47
80 62 69 64
No Yes No No
Not mentioned Yes, no confounding effect Not mentioned Not mentioned
51
18
48
82
No
Yes, no confounding effect
75
60
43
61
No
Yes, no confounding effect
32 95
26 36
52 49
94 69
Yes No
18
12
47
94
Yes
Not mentioned Discouraged, but not prohibited; no confounding effect Yes, no confounding effect
*Partially unpublished data provided by the study investigators.
Use of ACE Inhibitors During Follow-Up
392
PERNA ET AL Table 3.
Quality Assessment of Included Studies
Reference
Allocation Concealment
Blinding of Investigators
Blinding of Participants
Blinding of Outcome Assessor(s)
Blinding of Data Analysis
ITT Analysis*
n/N Lost to Follow-Up
Ahmed et al19 Branten et al20 Braun et al26 Cameron et al9 Cattran et al10 Cattran et al21 Cattran et al25,30 Coggins et al,8 1979 Donadio et al,11 1974 Falk et al,18 1992 Murphy et al,13 1992 Pahari et al,16 1993 Pisoni et al,24 2000 Ponticelli et al,33 1989 Ponticelli et al,14 1992 Ponticelli et al,22 1998 Reichert et al,23 1994 Silverberg et al,17 1976
Not stated Not stated Adequate Adequate Adequate Not stated Adequate Not stated Adequate Not stated Not stated Not stated Adequate Not stated Adequate Adequate Not stated Adequate
Not stated Not stated No Yes No No No Yes No Not stated Not stated Not stated No No No No Not stated Yes
Not stated Not stated No Yes No Yes Yes Yes No Not stated Not stated Not stated No No No No Not stated Yes
Not stated Not stated No Not stated Not stated Not stated Not stated Not stated No Not stated Not stated Not stated No Not stated Not stated Not stated Not stated Not stated
Not stated Not stated No Not stated Not stated Not stated Not stated Not stated No Not stated Not stated Not stated No Not stated Not stated Not stated Not stated Not stated
5 2 3 1 1 5 5 5 3 5 3 3 1 1 1 3 3 2
0/20 0/32 18/97 7/103 27/158 0/17 2/51 5/72 3/22 2/26 1/40 10/71 1/21 3/81 1/92 5/95 0/18 0/9
Abbreviation: ITT, intention to treat. *1 ⫽ yes, specifically reported by investigators that ITT analysis was undertaken, and this was confirmed on study assessment; 2 ⫽ yes, not stated, but confirmed on study assessment; 3 ⫽ no, not reported and lack of ITT analysis confirmed on study; 4 ⫽ no, stated, but not confirmed on study assessment; 5 ⫽ not stated, unable to be determined on study assessment.
Comparison of Each Class of Immunosuppressive Regimen Versus Placebo or No Treatment Baseline serum creatinine level was greater in steroid-treated patients compared with control groups, whereas increased baseline proteinuria was observed in groups treated with alkylating agents or cyclosporine versus placebo or no treatment. No beneficial effect on total mortality or endstage renal failure was observed in patients treated with glucocorticoids (combined end point: RR, 0.88; 95% CI, 0.39 to 1.97; P ⫽ 0.75; Fig 2), alkylating agents (combined end point: RR, 0.56; 95% CI, 0.18 to 1.70; P ⫽ 0.3; Fig 2), or cyclosporine (combined end point: RR, 0.93; 95% CI, 0.32 to 2.71; P ⫽ 0.89; Fig 2) compared with placebo or no treatment. There was no heterogeneity between trials. There were no deaths or end-stage renal failure within the trial period in the only trial of azathioprine. Glucocorticoids had no effect on partial or complete remission.
Alkylating agents showed a significant effect on complete remission (RR, 2.37; 95% CI, 1.32 to 4.25; P⫽ 0.004; Fig 3) and final proteinuria (weighted mean difference, ⫺2.36 g/24 h; 95% CI, ⫺4.27 to ⫺0.46; P ⫽ 0.02), but not on partial remission (RR, 1.22; 95% CI, 0.63 to 2.35; P ⫽ 0.56) or complete or partial remission (RR, 1.55; 95% CI, 0.72 to 3.34; P ⫽ 0.27). Final renal function was similar in the 2 groups (data not shown). In cyclosporine-treated patients, no beneficial effect was observed. The only trial of azathioprine showed no difference for any of the chosen end points. Comparison of 2 Different Classes of Immunosuppressive Regimens Information was available for 3 kinds of comparisons: steroids versus alkylating agents, cyclosporine versus steroids, and cyclosporine versus alkylating agents. Baseline characteristics were similar. For each of the 3 comparisons, no clear effect was seen as far as hard end points are concerned.
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY
Fig 2.
Outcome: end-stage renal failure or death.
393
394
PERNA ET AL
Fig 3.
Outcome: complete remission.
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY
Alkylating agents induced more remissions than steroids (complete: RR, 1.89; 95% CI, 1.34 to 2.67; P ⫽ 0.0003; Fig 3; complete or partial: RR, 1.45; 95% CI, 1.16 to 1.81; P ⫽ 0.001). The only trial reporting adverse events14 showed 4 discontinuations caused by adverse events, including leukopenia, pneumonia, and gastric discomfort, in the alkylating-agent group versus 1 discontinuation caused by renal vein thrombosis in the steroid-treated group. This difference was not statistically significant. Cyclosporine induced more remissions than steroids, which did not reach statistical significance (partial: RR, 3.70; 95% CI, 0.89 to 15.44; P ⫽ 0.07; complete or partial: RR, 3.01; 95% CI, 0.95 to 9.52; P ⫽ 0.06). Similarly, partial remissions were more frequent with cyclosporine than alkylating agent therapy (partial: RR, 1.68; 95% CI, 1.06 to 2.65; P ⫽ 0.03; complete or partial: RR, 1.46; 95% CI, 0.97 to 2.20; P ⫽ 0.07). Comparison of 2 Alkylating Agents We compared the relative efficacy of the 2 alkylating agents, chlorambucil versus cyclophosphamide. Proportions of males and baseline renal function were similar. There was no significant difference in the incidence of death and need for dialysis therapy or transplantation or in the combined end point. Within the class of alkylating agents, there were no differences for complete, partial, or complete or partial remission or for final proteinuria, serum creatinine level, or GFR. Cyclophosphamide treatment resulted in a statistically significantly lower rate of discontinuations caused by adverse events compared with chlorambucil (8 versus 21 discontinuations, respectively; RR, 2.34; 95% CI, 1.25 to 4.39; P ⫽ 0.008; Fig 4). Three discontinuations were caused by leukopenia in cyclophosphamidetreated patients versus 14 discontinuations in the chlorambucil group. Sensitivity Analysis To assess the robustness of meta-analytic results, a sensitivity analysis was conducted. Two main sources of heterogeneity across trials were considered: trial quality and characteristics of participating patients. First, it was performed with the analysis excluding trials providing no information on allocation concealment and blind-
395
ing.13,16,18-20,23 This first sensitivity analysis did not substantially change the results referred to the analyses versus no treatment. Modifications were noted for the comparison of steroids versus alkylating agents. The statistically significant increase in complete or partial remissions observed with alkylating agent therapy was not sustained further after the exclusion of trials by Ahmed et al19 and Pahari et al.16 Such discrepancy can be explained because of the reduced power to detect the difference (42% and 16%, respectively). It can be argued that although recommended by the Cochrane Collaboration, inclusion of totally or partially unpublished material can be questionable because data from abstracts and not published data were not subject to peer review and validation. To address this issue, the 2 studies24,26 that provided unpublished data in addition to those published were excluded for sensitivity analysis. No significant changes were observed for the comparison of immunosuppressive treatment versus placebo or no treatment. For the comparison of alkylating agents versus placebo or no treatment, by excluding the study of Braun et al,26 the difference in complete remissions was magnified. This can be considered an example of publication bias.43 No changes occurred for the comparison of cyclosporine versus placebo or no treatment after exclusion of the study by Pisoni et al.24 It was found that baseline renal function evaluated by serum creatinine level differed across studies. Therefore, inclusion and/or exclusion criteria of the studies were checked to assess whether they systematically varied. Some trials included patients with particularly well-preserved renal function (ie, average serum creatinine ⬍ 1.1 mg/dL [97 mol/ L]). It was decided to exclude this kind of trial.14,22,33 This approach resulted in several substantial changes. When reanalyzing the effect of alkylating agents versus placebo or no treatment, the difference in rates of complete remission was no more statistically significant. Conversely, exclusion of trials with inclusion and/or exclusion criteria leading to the enrollment of patients with worse renal function18,20,21,23,24 did not change the results. It was thought that an additional potential confounding factor can be represented by the use of concomitant medication, such as ACE inhibitors and/or statins. A few more recent trials reported this information (Table 1). It therefore was considered
396
PERNA ET AL
Fig 4.
Outcome: temporary or permanent discontinuation caused by adverse events.
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY
that an additional source of heterogeneity can be seen in the time frame of the study. It is always difficult to define a clear cutoff value, but it is acceptable to identify the turning point in the seminal observation of Lewis et al44 on the beneficial effect of ACE inhibition, published in 1993. After this date, use of this important concomitant medication is discussed and, relying on the assessment of investigators, no major confounding effect occurred. In 1 case, only a publication after 1993 did not mention the use of ACE inhibitors20: removal of this study did not affect the results. It is much more difficult to identify a similar cutoff value for the use of statins. Few studies are available45,46 and are too recently published to allow meaningful analyses. DISCUSSION
Objectives of this review of the treatment of IMN are to address the important question, which is controversial in nephrology, of whether immunosuppressive treatment is effective and safe. There is weak evidence that in patients with IMN and NS, treatment with an immunosuppressive regimen may be associated with an increased rate of remission. The safety of immunosuppressive treatments must be taken into account. Meta-analysis showed a statistically significant number of discontinuations caused by adverse events in the treatment group, mainly leukopenia and cushingoid features. The meta-analysis has been unable to show an effect of immunosuppression on mortality and need for dialysis therapy. Responses to treatment were observed in relatively short-term studies and with a low incidence of these events, which possibly prevented an observation of an effect on these definite end points. We attempted to minimize publication bias, including unpublished data and non–English language sources, and by inspecting prospective trial registries. Six ongoing and/or unpublished trials were identified.47-52 Through the Prospective Registry,27 we knew that Stegeman et al47 were performing a trial (group I, ACE inhibitor; group II, prednisolone; group III, control; total projected sample size, 150 patients), the protocol of which was published in 1994 before randomization of the first patient. However, no additional information was available for this trial. Through Current Controlled Trials28 and per-
397
sonal communication, we became aware that Mathieson et al of the University of Bristol, on behalf of the UK Renal Association, are studying patients with declining renal function (20% decline in GFR). Patients are randomized to 1 of 3 groups: cyclosporine versus alternating months of prednisolone and chlorambucil versus supportive therapy. The aim is to include 120 patients in total (40 patients in each group). Another recent ongoing National Institutes of Health study is a 24-week, randomized, double-blind, placebocontrolled trial of the effect of a new immunosuppressive drug, a genetically engineered antibody called h5G1.1-mAb, in reducing proteinuria in patients with IMN.48 Another study of prednisone alone versus prednisone plus cyclophosphamide versus prednisone plus cyclosporine was mentioned by Austin et al.49,50 No additional information could be retrieved. A study by Jha et al51 has a pending classification: it is an RCT performed in 89 subjects on the comparison of steroids versus cyclophosphamide. Finally, another ongoing trial was identified through the Cochrane renal group, in which “early” versus “rescue” immunosuppressive therapy (cyclophosphamide plus steroids) were compared.52 It was attempted to examine the possible role of publication bias by using funnel plots, but an adequate number of RCTs cannot be identified. Are there treatments among those considered to be definitely discarded? The answer to this question can be given mainly by comparing each of the identified immunosuppressive regimens with placebo or no treatment. The role of steroids alone, evaluated in 3 old RCTs,8-10 remains highly uncertain for both definite and surrogate end points. Overall, this kind of therapy at the schedules used in the trials included in the review (ie, high-dose short-term or moderate doses for 6 months) fails to show a beneficial effect. Furthermore, the meta-analysis showed a greater incidence of discontinuations caused by adverse events found under steroids. Even data for azathioprine were not promising, but it must be said that results relied on a single trial17 that enrolled only 9 patients overall, resulting in lack of power to detect even important differences. Clear indication on the use of a specific kind of treatment should be based on definite unambiguous end points, such as death or need for
398
dialysis therapy.32 Complete remission of proteinuria could not be considered a hard end point because it was not always sustained and frequently was followed by relapses.32 There is evidence based on surrogate end points that alkylating agents significantly increased the number of remissions and decreased final proteinuria compared with placebo or no treatment or in comparison with steroids alone. These results can represent the premises for long-term well-designed studies of this kinds of treatment. Among alkylating agents, there was a tendency of cyclophosphamide to be associated with fewer adverse effects compared with chlorambucil. However, this does not mean that cyclophosphamide can be considered a safe agent, having much toxicity. For patients with rapidly deteriorating renal function, salvage therapy with alkylating agents may be useful, although it is not yet clear how long this salvage therapy would keep the patient off dialysis therapy. These conclusions, which confirmed previous results observed in a pooled analysis of randomized and nonrandomized studies,53 are weakened by the lack of robustness arisen after the sensitivity analysis. Particularly, it was shown that 1 of the sources of heterogeneity found across studies can be identified in the different degree of renal impairment of patients at the time of randomization. Substituting 1989 data from Ponticelli et al33 with those from the 10-year follow-up42 deserves an additional comment. This change did not substantially affect the overall comparison between all immunosuppressive treatments with placebo or no treatment. When focusing on the class of alkylating agents, evidence of a tendency of improvement in renal survival under treatment arose. The selection of this longer follow-up report likely accounted for an additional amount of variability. In the present review, no remarkable differences were observed between cyclosporine and placebo. This information was based on 3 trials21,24,26 involving 104 randomized patients. Although Cattran et al25,30 reported that cyclosporine induced more partial remissions than steroids, and Braun, in his partially unpublished data, showed that cyclosporine also was superior to alkylating agents in increasing partial remissions, it must be said that the superiority of cyclosporine in comparison to placebo or no
PERNA ET AL
treatment cannot be clearly shown. Furthermore, these results arose from only a single study and consequently need to be proven by additional evidence. Although it was not possible to adequately address the possible confounding effect of the use of ACE inhibitors and/or statins, it likely did not have a major role because studies not reporting this information were randomized. This systematic review, focusing on immunosuppressive therapy, did not investigate the role of ACE inhibitors. However, the importance of ACE inhibitors in reducing proteinuria and delaying the progression of renal disease should be emphasized. Conversely, it is unlikely their role will be assessed in a specifically designed trial because ACE inhibitors are administered routinely to all proteinuric patients, including those with IMN. We decided to perform a review addressing broadly focused questions. Because of the paucity of trials, subgroup analyses allowing for variations in interventions; ie, different doses, different routs of administration, different lengths of follow-up, and in populations, ie, different races, cannot be carried out. Furthermore, when interpreting study results, some unbalances in baseline characteristics, which occurred for some comparisons, should be taken into account. Particularly, because level of proteinuria is known to be a predictor of outcome, it can be of relevance that patients treated with alkylating agents or cyclosporine had greater proteinuria than controls: this could have induced a bias against seeing an effect of treatment. The underlying pathogenic mechanism(s) of membranous nephropathy should be better understood to allow the discovery of new selective drugs for the treatment of this condition. Very recently, the safety and efficacy profile of a monoclonal antibody to B-cell antigen CD-20, rituximab, was evaluated,54,55 giving favorable results to be studied further. Encouraging results also were reported with adrenocorticotropic hormone,56 pentoxiphylline,57 and mycophenolate mofetil.58 The meta-analyses of Couchoud et al5 and Imperiale et al7 suggest the need for welldesigned RCTs of the long-term benefits and risks in the categories of agents considered in their articles. Hogan et al6 emphasized the impor-
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY
tance of identifying patients who are likely to have progressive disease and who may benefit from treatment, understanding the contribution of prognostic variables in the long term. Hogan et al6 concluded that “we must await an era of new therapeutic agents that interfere with etiology or pathogenesis of this disease before a truly effective therapy is available.” Nine years after this report, there is still high uncertainty regarding the beneficial effect of these immunosuppressive agents, even after increasing the number of RCTs available. This remarkable finding indicates that the conclusions of Hogan et al6 possibly also apply 9 years later. It therefore is mandatory to confirm the encouraging results observed in novel alternative immunosuppressive regimens by means of well-designed RCTs. ACKNOWLEDGMENT This systematic review was prepared under the aegis of the Cochrane Collaboration, an international organization that aims to help people make well-informed decisions about health care by preparing, maintaining, and promoting the accessibility of systematic reviews of the effects of health care interventions. The authors thank Lisa A. Tjosvold, who helped perform the electronic search; Luciana Tammuzzo, who hand searched the Journal of Nephrology; and principal investigators who provided additional information or clarification (Daniel C. Cattran, Peter Mathieson, Roberto Pisoni, and Teut Risler); Ruth Mitchell, the Cochrane Trials Search coordinator who provided us with the Cochrane library search strategy and relevant information; and Narelle Willis, the Cochrane Renal Review Group coordinator, for her help and support.
REFERENCES 1. Cameron JS: Membranous nephropathy and its treatment. Nephrol Dial Transplant 7:S72-S79, 1992 (suppl 1) 2. Schieppati S, Mosconi L, Perna A, et al: Prognosis of untreated patients with idiopathic membranous nephropathy. N Engl J Med 329:85-89, 1993 3. Honkanen E, Tornroth T, Grohagen-Riska C: Natural history, clinical course and morphological evolution of membranous nephropathy. Nephrol Dial Transplant 7:S35-S41, 1992 (suppl 1) 4. Glassock RJ: The therapy of idiopathic membranous glomerulonephritis. Semin Nephrol 11:138-147, 1991 5. Couchoud C, Laville M, Boissel JP: Treatment of membranous nephropathy. Nephrol Dial Transplant 9:469470, 1994 6. Hogan SL, Muller KE, Jennette JC, Falk RJ: A review of therapeutic studies of idiopathic membranous glomerulopathy. Am J Kidney Dis 25:862-875, 1995 7. Imperiale TF, Goldfarb S, Berns JS: Are cytotoxic agents beneficial in idiopathic membranous nephropathy? A meta-analysis of the controlled trials. J Am Soc Nephrol 5:1553-1558, 1995
399
8. Coggins CH, for the Collaborative Study of the Adult Idiopathic Nephrotic Syndrome: A controlled study of shortterm prednisone treatment in adults with membranous nephropathy. N Engl J Med 301:1301-1306, 1979 9. Cameron JS, Healy MJR, Adu D: The Medical Research Council trial of short-term high-dose alternate day prednisolone in idiopathic membranous nephropathy with nephrotic syndrome in adults. Q J Med 274:133-156, 1990 10. Cattran DC, Delmore T, Roscoe J, et al: A randomized controlled trial of prednisone in patients with idiopathic membranous nephropathy. N Engl J Med 320:210-215, 1989 11. Donadio JV, Holley KE, Anderson CF, Taylor WF: Controlled trial of cyclophosphamide in idiopathic membranous nephropathy. Kidney Int 6:431-439, 1974 12. Ponticelli C, Zucchelli P, Imbasciati E, et al: Controlled trial of methylprednisolone and chlorambucil in idiopathic membranous nephropathy. N Engl J Med 310:946950, 1984 13. Murphy BF, McDonald I, Fairley KF, Kincaid-Smith PS: Randomised controlled trial of cyclophosphamide, warfarin and dipyridamole in idiopathic membranous glomerulonephritis. Clin Nephrol 37:229-234, 1992 14. Ponticelli C, Zucchelli P, Passerini P, Cesana B, Italian Idiopathic Membranous Nephropathy Treatment Study Group: Methylprednisolone plus chlorambucil as compared with methylprednisolone alone for the treatment of idiopathic membranous nephropathy. N Engl J Med 327:599603, 1992 15. Robinson KA, Dickersin K: Development of a highly sensitive search for the retrieval of reports of controlled trials using PubMed. Int J Epidemiol 31:150-153, 2002 16. Pahari DK, Das S, Dutta BN, Banerjee D: Prognosis and management of membranous nephropathy. J Assoc Physicians India 41:350-351, 1993 17. Silverberg DS, for the Western Canadian Glomerulonephritis Study Group: Controlled trial of azathioprine in the nephrotic syndrome secondary to idiopathic membranous glomerulonephritis. CMAJ 115:1209-1210, 1976 18. Falk RJ, Hogan SL, Muller KE, Jenette C: Treatment of progressive membranous glomerulopathy. Ann Intern Med 116:438-445, 1992 19. Ahmed S, Rahman M, Alam MR, et al: Methyl prednisolone plus chlorambucil as compared with prednisolone alone for the treatment of idiopathic membranous nephropathy. A preliminary study. Bangladesh Ren J 13:5154, 1994 20. Branten AJW, Reichert LJM, Koene RAP, Wetzels JFM: Oral cyclophosphamide versus chlorambucil in the treatment of patients with membranous nephropathy and renal insufficiency. Q J Med 91:359-366, 1998 21. Cattran DC, Greenwood C, Ritchie S, et al: A controlled trial of cyclosporine in patients with progressive membranous nephropathy. Kidney Int 47:1130-1135, 1995 22. Ponticelli C, Altieri P, Scolari F, et al: A randomized study comparing methylprednisolone plus chlorambucil versus methylprednisolone plus cyclophosphamide in idiopathic membranous nephropathy. J Am Soc Nephrol 9:444450, 1998 23. Reichert LJM, Husmans FTM, Assmann K, Koene RAP, Wetzels JFM: Preserving renal function in patients with membranous nephropathy: Daily oral chlorambucil
400
compared with intermittent monthly pulses of cyclophosphamide. Ann Intern Med 121:328-333, 1994 24. Pisoni R, Grinyo JM, Salvadori M, et al: Cyclosporine versus conservative therapy in patients with idiopathic membranous nephropathy (IMN) and deteriorating renal function: Results of the CYCLOMEN trial. J Am Soc Nephrol 11:95A, 2000 (abstr) 25. Cattran DC, Pohl M, Maxwell D, et al: Randomized controlled trial of cyclosporine (D) vs. placebo (P) in steroid resistant, nephrotic patients (NS) with membranous nephropathy (SR MGN). J Am Soc Nephrol 9:85A, 1998 (abstr) 26. Braun N, Erley C, Benda N, et al: Therapy of membranous glomerulonephritis with nephrotic syndrome. 5 Years follow-up of a prospective, randomised multi-centre study. Nephrol Dial Transplant 10:967, 1995 (abstr) 27. Tognoni G, Mallick N: (Another) registry of clinical trials? J Nephrol 6:117, 1993 28. Current Controlled Trials Ltd. The metaRegister of Controlled Trials (mRCT). Available at: http://www. controlled-trials.com. Accessed: November 24, 2003 29. Protocol Review, The Lancet. Available at: http:// www.thelancet.com/authorinfo/Protocol⫹review. Accessed: November 24, 2003 30. Cattran DC, Appel GB, Hebert LA, et al: Cyclosporine in patients with steroid-resistant membranous nephropathy: A randomised trial. Kidney Int 59:1484-1490, 2001 31. Schulz KF, Chalmers I, Grimes DA, Altman DG: Assessing the quality of randomization from reports of controlled trials published in obstetrics and gynecology journals. JAMA 272:125-128, 1994 32. Marx BE, Marx M: Prognosis of idiopathic membranous nephropathy: A methodologic meta-analysis. Kidney Int 51:873-879, 1997 33. Ponticelli C, Zucchelli P, Passerini P, et al: A randomized trial of methylprednisone and chlorambucil in idiopathic membranous nephropathy. N Engl J Med 320:8-13, 1989 34. Ambalavanan S, Fauvel J-P, Sibley RK, Myers BD: Mechanism of the antiproteinuric effect of cyclosporine in membranous nephropathy. J Am Soc Nephrol 7:290-298, 1996 35. Lagrue G, Bernard D, Bariety J, Druet P, Guenel J: Traitement par le chlorambucil et l’azathioprine dans les glomérulonéphrites primitives: Résultats d’une étude “controlée”. J Urol Nephrol 1975;81:655-672. 36. Medical Research Council Working Party: Controlled trial of azathioprine and prednisone in chronic renal disease. Br Med J 2:239-241, 1971 37. Nand N, Das R, Jaswal TS: Evaluation of efficacy of methylprednisolone, prednisolone and chlorambucil in idiopathic glomerulonephritis. J Indian Med Assoc 95:163-165, 1997 38. Tejani A, Suthanthiran M, Pomrantz A: A randomized controlled trial of low-dose prednisone and cyclosporin versus high-dose prednisone in nephrotic syndrome of children. Nephron 59:96-99, 1991 39. Yao X, Chen H, Wang Q, et al: Cyclosporin A treatment for idiopathic membranous nephropathy. Chin Med J 114:1305-1308, 2001
PERNA ET AL
40. Imbasciati E, Cagnoli L, Case N, et al: Studio controllato di un trattamento a mesi alterni di steroidi e clorambucil nella nefropatia membranosa e nella glomerulosclerosi focale. Valutazione preliminare dei risultati. Minerva Nefrol 27:571-575, 1980 41. Ponticelli C, Zucchelli P, Imbasciati E, et al: Controlled trial of monthly alternated courses of steroid and chlorambucil for idiopathic membranous nephropathy. Proc Eur Dial Transplant Assoc 19:717-723, 1983 42. Ponticelli C, Zucchelli P, Passerini P, et al: A 10-year follow-up of a randomized study with methylprednisolone and chlorambucil in membranous nephropathy. Kidney Int 48:1600-1604, 1995 43. Dickersin K, Min YI: NIH clinical trials and publication bias. On-line J Curr Clin Trials [serial on line]. Apr 28, 1993; doc 50 44. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD, for the Collaborative Study Group: The effect of angiotensinconverting-enzyme inhibition on diabetic nephropathy. N Engl J Med 329:1456-1462, 1993 45. Rayner BL, Byrne MJ, van Zyl Smit R: A prospective clinical trial comparing the treatment of idiopathic membranous nephropathy and nephrotic syndrome with simvastatin and diet, versus diet alone. Clin Nephrol 46:219-224, 1996 46. Olbricht CJ, Wanner C, Thiery J, Basten A, for the Simvastatin in Nephrotic Syndrome Study Group: Simvastatin in nephrotic syndrome. Kidney Int Suppl 71:S113-S116, 1999 47. Stegeman CA, De Zeeuw D, De Jong PE: A randomized multicentre trial to compare the effect of symptomatic antiproteinuric treatment with an ACE inhibitor to an immunosuppressive regimen with corticosteroids and of no treatment on the course of membranous nephropathy. J Nephrol 7:294-300, 1994 48. National Institute of Diabetes and Digestive and Kidney Diseases: A phase II randomized, double-blind, placebo-controlled study of the effect of h5G1.1-mAb on the reduction of proteinuria in patients with idiopathic membranous glomerulopathy. Available at: http://clinicalstudies.info. nih.gov/detail/A_2000-DK-0126.html. Accessed: June 29, 2001 49. Austin AH: Membranous nephropathy. Ann Intern Med 116:672-682, 1992 50. Austin AH, Vaughan EM, Boumpas DT, Balow JE: Randomized trial of pulse cyclophosphamide and prednisone vs. prednisone alone in idiopathic membranous nephropathy. J Am Soc Nephrol 7:1327, 1996 (abstr) 51. Jha V, Kohli HS, Sud K, Gupta KL, Joshi K, Sakhuja V: A randomized controlled trial of steroids and cyclophosphamide in adults with idiopathic membranous nephropathy. J Am Soc Nephrol 213A, 2001 (abstr) 52. Branten AJ, du Buf Vereijken PW, Wetzel JF: Defining the optimal time of treatment start in patients with membranous nephropathy: A randomized study. J Am Soc Nephrol 668A, 2002 (abstr) 53. Schieppati A, Perna A, Remuzzi G: Recent developments in the management of membranous nephropathy. Exp Opin Invest Drugs 6:521-532, 1997
IMMUNOSUPPRESSION IN IDIOPATHIC MEMBRANOUS NEPHROPATHY
54. Remuzzi G, Chiurchiu C, Abbate M, Brusegan V, Bontempelli M, Ruggenenti P: Rituximab for idiopathic membranous nephropathy. Lancet 360:923-924, 2002 55. Ruggenenti P, Chiurchiu C, Brusegan V, et al: Rituximab in idiopathic membranous nephropathy: A one-year prospective study. J Am Soc Nephrol 14:1851-1857, 2003 56. Berg AL, Nilsson-Ehle P, Arnadottir M: Beneficial effects of ACTH on the serum lipoprotein profile and glomer-
401
ular function in patients with membranous nephropathy. Kidney Int 56:1534-1543, 1999 57. Ducloux D, Bresson-Vautrin C, Chalopin J: Use of pentoxiphylline in membranous nephropathy. Lancet 357: 1672-1673, 2001 58. Choi MJ, Eustace JA, Gimenez LF, et al: Mycophenolate mofetil treatment for primary glomerular diseases. Kidney Int 61:1068-1114, 2002
