A Randomized Study Comparing Methylprednisolone Plus

Chlorambucil Versus Methylprednisolone Plus

Cyclophosphamide in Idiopathic Membranous Nephropathy

CLAUDIO PONTICELLI,* PAOLO ALTIERI,t FRANCESCO SCOLARI,�PATRIZIA PASSERINI,* DARIO ROCCATELLO,� BRUNO CESANA,”

PATRIZIA MELIS,t BRUNELLA VALZORIO,� MAURO SASDELLI,1

SONIA PASQUALI,# CLAUDIO POZZI,** GIUSEPPE PICCOLI,� ANTONIO LUPO,tt

SIRO SEGAGNI,� FRANCESCO ANTONUCCI,5� MAURO DUGO,111’

MARILENA MINARI,11 ALFIO SCALIA,� LUCIANO PEDRINI,***

GABRIELE PISANO,ttt CLAUDIO GRASSI,t� MARCO FARINA,� andROBERTO BELLAZZI�’1111*Dj�,jsjone di Nefrologia e Dialisi IRCCS Ospedale Maggiore Milano; �OspedaIe Brotzu Cagliari; �Ospedale

Regionale Spedali Civili Brescia; ‘�Ospedale Giovanni Bosco Torino; ‘Laboratorio Epidemiologico IRCCS

Ospedale Maggiore Milano; �‘Ospedale Civile Arezzo; #Ospedale Malpighi Bologna; **Ospedale Civile

Lecco; “Ospedale Regionale Civile Maggiore Verona; �Fondazione Clinica del Lavoro Pavia; �Ospedale

Provinciale Feltre; 1�Ospedale Provinciale Castelfranco Veneto; t��Ospedale Regionale Parma; ##Ospedale

Zonaie Uboldo Cernusco sul Naviglio; ***Ospedale Provinciale Civile Sondrio; �Ospedale Provinciale

Magenta; �Ospedale Predabissi Melegnano; �Ospedale Provinciale Maggiore Lodi; and hi�Ospedale

Provinciale Civile Vigevano.

Abstract. To assess whether chborarubucil or cycbophospha-

mide may have a better therapeutic index in patients with

idiopathic merubranous ncphropathy, we compared two regi-

mens based on a 6-mo treatment, alternating every other month

methylpnednisobone with chborambucib or mcthylprcdnisobone

with cycbophosphamide. Patients with biopsy-proven membra-

nous ncphropathy and with a ncphrotic syndrome were ran-

domized to be given methylprednisobonc (1 g intravenously for

3 consecutive days followed by oral ructhylpnednisobone, 0.4

mg/kg per d for 27 d) alternated every other month either with

chlonambucil (0.2 rug/kg per d for 30 d) or cycbophosphamide

(2.5 mgfkg per d for 30 d). The whole treatment lasted 6 mo:

3 mo with corticosteroids and 3 mo with one cytotoxic drug.

Among 87 patients followed for at beast 1 yr. 36 of 44 (82%;

95% confidence interval [CI], 67.3 to 91 .8%) assigned to

methylprcdnisobone and chborambucib entered complete on par-

tial remission of the ncphrotic syndrome, versus 40 of 43

(93%; 95% CI, 80.9 to 98.5%) assigned to mcthylpnednisobonc

and cycbophosphamide (P = 0. 1 16). Of patients who attained

remission of the ncphrotic syndrome, 1 1 of 36 in the chloram-

bucib group (30.5%) and 10 of 40 in the cycbophosphamidc

group (25%) had a relapse of the nephrotic syndrome between

6 and 30 mo. The reciprocal of plasma creatininc improved in

the cohort groups followed for I yr for both treatment groups

(P < 0.01) and remained unchanged when compared with

basal values in the cohort groups followed for 2 and 3 yr. Six

patients in the chborambucib group and two in the cycbophos-

phamide group did not complete the treatment because of side

effects. Four patients in the chlorambucil group but none in the

cycbophosphamide group suffered from herpes zoster. One

patient per group developed cancer. It is concluded that in

ncphrotic patients with idiopathic merubranous nephropathy

both treatments may be effective in favoring remission and in

preserving renal function for at least 3 yr. (I Am Soc Nephrob

9: 444-450, 1998)

There is now controlled evidence that a 6-mo course with

methylprednisobonc and chbonambucil may favor remission of

Received June 16, 1997. Accepted September 6. 1997.Correspondence to Dr. Claudio Ponticelli, Divisione di Nefrologia e Dialisi.

Ospedale Maggiore Policlinico IRCCS, via Commenda 15. 20122 Milano,

Italy.

1046-6673/0903-0444$03.00/0Journal of the American Society of Nephrology

Copyright 0 1998 by the American Society of Nephrology

the nephrotic syndrome (1-3) and may significantly improve

the 10-yr kidney survival rate in patients with idiopathic mem-

branous nephropathy (4). Randomized (5) and nonrandomized

(6,7) trials showed that this regimen can improve kidney func-

tion and reduce protcinunia also in patients with established

renal insufficiency.

Only a few nephrobogists, however, arc familiar with

chborarubucil; the majority have experience with cycbophosph-

amidc. Although cycbophosphamide and chborambucil both

have the capacity to contribute alkyl groups to biologically

Idiopathic Membranous Nephropathy 445

vital macromolecules such as DNA, there are differences be-

tween the two drugs in dosing, metabolism, pharmacokinetics,

and elimination. Thus, the results of treatment with either

chborambucil or cycbophospharuide might be different.

To evaluate whether these alkylating agents had similar

effects when alternated with corticosteroids in idiopathic mem-

branous nephropathy, we organized a multicenter, controlled

trial. Patients with biopsy-proven membranous nephropathy

and ncphrotic syndrome were randomly assigned to receive a

6-mo regimen with ructhylprednisobone alternated with

chborambucil or with the same doses of mcthylprednisobone

alternated with cycbophospharuide.

Materials and MethodsPatients

Patients of either sex. between the ages of 14 and 65 with a

biopsy-proven ruembranous nephropathy and a nephrotic syndrome

(defined as proteinuria exceeding 3.5 g/d and plasma albumin con-

centrations of <25 gIL), were considered for the study. The criteria

for exclusion were a plasma creatinine > I .7 mg per deciliter (150

�moLfL); previous treatments with corticosteroids, irumunosuppres-

sive drugs, or cycbosporine; a positive serum test for anti-DNA anti-

bodies (and antinuclear antibodies in many patients) or hepatitis B

antigen or hepatitis C virus antibodies; a positive Veneral Disease

Research Laboratory test: or low serum concentrations of the corn-

plement component C3 or C4. A clinical diagnosis of diabetes mel-

litus. malignancy. systemic lupus erythematosus, infections, or expo-

sure to drugs that could induce membranous nephropathy were also

criteria for exclusion.

Renal Histology

The pathologist at each study center examined each renal biopsy

specimen by light. immunofluorescence. and electron microscopy.

Gbomerular stages were classified according to the system of Ehren-

reich ci’ al. (8). Mesangial glornerular sclerosis and interstitial fibrosis

with tubular atrophy were classified as present or absent. No fob-

low-up biopsies were performed.

Study Design

This was an open-label multicenter study. The protocol was ap-

proved by the institutional review board of the coordinating center,

and each patient gave informed written consent. At the coordinating

center, patients were assigned consecutively to one of the two treat-

ruent regimens, according to a center-stratified random order. A

sample size of approximately 50 patients per group was necessary to

demonstrate a decrease of about 25% of response from a baseline of

80% with ruethylprednisobone plus chborambucil (2) at power of about

0.80 with a two-tailed test (alpha = 0.05). Patients were examined at

least every month during the first 6 ruo. every 2 mo until the end of

the first year, and every 3 to 6 ruo thereafter. At each visit, plasma

creatinine concentrations and 24-h urinary protein excretion rates

were measured, and the patient was questioned about symptoms and

possible side effects of therapy. Leukocyte, erythrocyte, and platelet

counts were measured every 7 to 10 d during administration of the

cytotoxic drug, and repeated at each visit.

In our previous experience with patients selected with the samecriteria, only one of 39 untreated patients had a spontaneous remission

of the nephrotic syndrome within 6 mo (4). Therefore, we preferred to

start early, without a run-in period, a potentially beneficial treatment

to prevent complications of the nephrotic syndrome. Patients assigned

to the chlorambucil group received three cycles of treatment with

methylprednisobone ( I g given intravenously on 3 consecutive days

and then 0.4 mg/kg body wt per d given orally for 27 d, in a single

morning dose), and each cycle was followed by 1 mo of treatment

with chborambucil (0.2 mg/kg per d, orally). Patients assigned to the

cyclophospharuide group received the same three cycles of methyl-

prednisobone, but chlorambucil was replaced by oral cycbophospha-

mide at a dose of 2.5 mg/kg body wt per d. If a patient’s leukocyte

count decreased below 5000 per cubic millimeter (5 X b0’�1L), the

dose of the cytotoxic agent was reduced by 50%. and it was discon-

tinued for the remainder of that cycle if the leukocyte count was less

than 3000 per cubic millimeter (3 X b0#{176}IL).The total duration of

treatment, therefore, was 6 mo for both groups; 3 ruo with the same

doses of methylprednisobone and 3 mo with either of the two cytotoxic

drugs. Steroids were completely stopped at the end of the study

period. The use of angiotensin-converting enzyme inhibitors was

discouraged but not prohibited. Hypolipemic drugs were permitted.

Clinical response was defined as a complete or partial remission.

Complete remission was defined as a reduction in the rate of urinary

protein excretion to �0.2 g/d for at least I wk. and partial remission

as a reduction in the rate of protein excretion to between 0.2 1 and 2

g/d for at least 1 wk, with a normal plasma creatinine concentration.

Relapse of the nephrotic syndrome was defined as an increase in

proteinuria to more than 3.5 g/d for at least I wk in patients who had

a remission. Stable worsening was defined as an increase in plasma

creatinine concentration of at least 50% over the baseline value.

Statistical Analyses

Cumulative probability of complete/partial remission, complete

remission alone, and thereafter of relapses was estimated according to

Kaplan and Meier (9). Because assumption of the hazard functions

proportionality cannot be considered fulfilled as suggested by the log

(-log) survival plots, survival curves have been shown for illustrative

purposes only; � tests have been performed on the cumulative

proportion of patients with complete or partial remission. A further

index of efficacy was obtained as the percentage of the remission time

(partial or complete) on the total follow-up for each patient. The two

treatments were then compared by the Wilcoxon rank sum test.

The reciprocal of the plasma creatinine and proteinuria values for

the two treatments were compared by mixed-factorial ANOVA for

repeated measurements (10), with treatment as a fixed, “between-

patient” factor at two levels and time as a fixed, “within-patient”

factor at 2 1 levels (every 6 mo). The Geisser-Greenhouse ( I I ) pro-

cedure for correction of the degrees of freedom for statistical signif-

icance of the within-patient terms was followed. In addition, a trend

analysis has been carried out by means of orthogonal polynomial

contrasts to assess the presence of a linear increasing/decreasing trend

(first degree), of an increasing and then decreasing (or vice versa)

trend (second degree). of an increasing-decreasing-increasing or de-

creasing-increasing-decreasing trend (third degree) of a mean fluctu-

ating pattern (degree higher than the third). Indeed, the shape of an

overall pattern could have therapeutic impact.

ResultsAmong 97 patients who satisfied the criteria for randomiza-

tion, two were wrongly randomized due to the presence of

underlying neoplasias (discovered, respectively, I wk and 2 mo

after randomization). Of the remaining 95 patients, 50 were

assigned to receive ructhylprednisobone and chlorarubucil,

whereas 45 were randomized to ruethylprednisobone and cy-

clophosphamide. However, two patients did not present at the

446 Journal of the American Society of Nephrobogy

follow-up visit and a 5 1 -yr-old woman died because of a

deep-vein thrombosis with acute renal failure and cardiac

shock 3 mo after the diagnosis of membranous nephropathy,

before treatment was started. There was no difference at pre-

sentation between the two groups in the main clinical, biobog-

ical, and histological features (Table I).

Four patients in the chborambucil group and one in the

cycbophosphamide group, who completed the treatment, did

not present at the follow-up visit and were considered lost to

follow-up after the sixth month. All of the other patients were

followed for at beast 1 yr (range, 12 to 78 mo). The median

period of follow-up was 36 mo (range, 12 to 78 mo) in the

chborambucil group and 42 mo (range, 12 to 72 mo) in the

cycbophosphamide group, with a very similar distribution.

Of the 87 patients followed for at least 1 yr. 76 (87.3%; 95%

CI, 78.5 to 93.5%) entered a complete (28 patients) or partial

remission (48 patients). The presence or absence of interstitial

fibrosis or mesangial sclerosis and the stage of gborucrubar

lesions at renal biopsy did not influence the likelihood nemis-

sion of ncphrotic syndrome. However, if one takes into account

the complete remission only this was significantly related to

the presence (8.7%; 95% CI, 10.7 to 28.0%) on absence

(40.6%; 95% CI, 28.5 to 53.6%) of mcsangial gbomenubar

sclerosis (P = 0.005). Of 44 patients assigned to methylpred-

nisobone and chlorarubucil, 36 (82%; 95% CI, 67.3 to 91.8%)

entered complete ( 12 patients) or partial remission (24 patients)

of protcinunia as a first event. Of the 43 patients given mcdi-

ylprednisobonc and cyclophosphamide, 40 (93%; 95% CI, 80.9

to 98.5%) entered complete (16 patients) or partial remission

(24 patients) as a first event. The difference between the two

groups was not significant (P = 0. 1 1 6). The curves of the

probability of a partial on a complete remission after treatment

were similar in the two study groups (Figure 1).

Eleven patients in the chborarubucib group and 10 in the

cycbophosphamide group had a relapse of the ncphrotic syn-

drome. All of the relapses occurred between the sixth and the

thirtieth month (Figure 2). Four patients were retreated. In the

chborambucil group, two patients were retreated with steroids

and chborambucil. One did not respond, and the other attained

partial remission. In the cycbophosphamidc group, one patient

was retreated with steroids and cycbophosphamide and had

complete remission. Another patient was treated with steroids

and chlorambucib and had partial remission. The median of the

percentage of time spent without nephrotic syndrome was 60%

of the whole period of observation for the entire study group,

without significant difference between the group given

chlorambucil (58.6%) and that given cycbophosphamide

(62.5%, P = 0.534). We considered separately patients with

complete follow-ups of 12, 24, and 36 mo, respectively, to

have more precise estimates on a greater number of patients for

the I - and 2-yr analyses. For both of the treatment groups, there

was a significant decrease in proteinuria when compared with

the basal value for the patients with a complete follow-up of

12, 24, and 36 mo (P = 0.0001). Instead, differences between

the two treatment groups turned out not to be statistically

significant. After an early relevant decrease of protcinuria at 6

mo, there was a fluctuating pattern of approximately 2 g/d,

similar in the two treatment groups (Figure 3) as confirmed by

the significance (P < 0.0001) of the orthogonal polynomials

higher than the third degree.

The reciprocal of plasma creatinine was significantly (at

least P < 0.0055) higher at 6 and 12 ruo for both treatment

Table 1. Characteristics of patients at the start of treatment with ructhylprcdnisolonc plus chborambucil on

ructhylprcdnisolonc plus cycbophospharuide”

CharacteristicMP + Chlorambucil

(�7 = 50)MP + Cyclophosphamide

(Ii = 45)

Age (median) 50 48

Range 18 to 65 17 to 55

Sex (MIF) 37/13 29/16

Plasma creatinine (mg/dl)” 1 .06 ± 0.27 1 .04 ± 0.27

Urinary protein excretion (g/d) 7.96 ± 5.19 6.85 ± 3.51

Hypertensive patients (ACE inhibitors) 15 (1) 14 (2)

Stage of gbomerular lesions at renal biopsy

lorIl 32 27

IIIorIV 18 18

Mesangial sclerosis

yes 14 12

no 36 33

Tububointerstitial lesions

yes 11 13

no 39 32

a Plus-minus values are means ± SD. Data for two patients (one per group) wrongly randomized are not included in this table. MP,methylprednisolone; ACE, angiotensin-converting enzyme.

h To convert values for creatinine to micromoles per liter, multiply by 88.4.

100

80

Partial + complete remiselon

0 6 12 18 24 30 36 42 48 54 60

months

Figure 1. Cumulative probability of obtaining partial or complete

remission of the nephrotic syndrome or complete remission alone as

a first event in patients given methylprednisobone plus chlorambucil

(- - -) and in patients given methylprednisolone plus cyclophospha-

mide (-). Data were calculated every 6 mo.

%

�

20 -

�o 6 12 18 24 30 36 42 48 54 60

Idiopathic Membranous Nephropathy 447

% at the sixth month because of leukopenia, which reversed

spontaneously. Two patients developed pneumonia after 5 and

3 mo of therapy, respectively. Both of them recovered but

continued without chlorambucil. One patient developed ane-

mia and thrombocytopenia with hypoplasia at the bone marrow

aspiration after I mo of chborambucib. He stopped treatment

with chborambucil and recovered. Another patient stopped

chborarubucil treatment because of nausea after a few days.

Two patients (4.5%) stopped treatment in the cycbophospha-

mide group, one because of nausea and vomiting after a few

days. The other patient stopped treatment after 2 ruo because of

a cerebral transient ischcmic attack. One patient per group

developed glucose intolerance during treatment, which re-

versed after completion of treatment. Four patients in the

chlorambucil group, but none in the cycbophosphamide group,

developed herpes zoster during treatment. In the long-term, a

40-yr-old woman treated with chlorambucib had irreversible

amenorrhea. In the chlorarubucil group, a man 58 yr old

developed a laryngeal carcinoma and underwent a laryngec-

tomy after 48 ruo. In the cyclophosphamide group, a 58-yr-old

man was recognized to have a prostatic carcinoma 66 mo after

randomization. Both patients are still alive after 1 2 and 6 mo,

respectively.

Figure 2. Cumulative probability of relapse of the nephrotic syndrome

after complete or partial remission in patients given methylpred-

nisolone plus chborambucil (- - -) and in patients given methylpred-

nisobone plus cyclophosphamide (-).

groups (without a statistically significant difference between

them) when compared with baseline, followed by a mild de-

crease to values similar to baseline, at least for patients fob-

bowed for 3 yr (Figure 4). One patient in the chborarubucil

group had worsening of renal function and had to be submitted

to regular dialysis after 42 mo of follow-up. Two patients in

the cycbophosphamide group worsened, and one of them had to

be submitted to regular dialysis after 36 mo of follow-up

(Table 2).

Six patients ( 12%) did not complete treatment in the

chborambucil group. Two of them did not receive chborambucil

DiscussionIn this study, a 6-mo course with methybprcdnisolonc and

chlorarubucil favored the remission of the ncphrotic syndrome

in patients with idiopathic ruembranous ncphnopathy. After

treatment, the mean proteinunia significantly decreased com-

pared with baseline at any time point considered. Eighty-two

percent of patients given mcthylprednisobonc and chborambucil

attained complete or partial remission of the ncphrotic syn-

droruc as a first event, a proportion similar to the 8 1 % (2) and

the 75% (3) found, respectively, in two previous different

controlled studies of Italian patients selected with the same

criteria and treated with methylprednisobone and chlonarubucil

for 6 mo. In the study presented here, some 30% of patients had

a relapse of the nephrotic syndrome, which occurred between

months the sixth and the thirtieth month.Similar results were observed in the group randomized to

ruethylprednisobone and cycbophosphamide. The mean de-

crease in protcinuria was similar to that seen in patients given

chborarubucil at any month cohort investigated. Ninety-three

percent of patients entered remission of the ncphrotic syn-

drome as a first event, and 24% had a relapse of the nephrotic

syndrome between the sixth and the twenty-fourth month.

However treated, patients spent 60% of the period of observa-

tion without nephrotic syndrome. The percentage of time in

remission was similar in the two study groups and was almost

the same as the 58% found in a previous trial with methyl-

prcdnisobonc and chborambucib (4). These data show that either

chborambucil or cycbophosphamide alternated with methyl-

prednisolone can result in a sustained remission of the ne-

phrotic syndrome in most patients. This result may have a

clinical impact if one considers that ncphrotic patients arc at a

greater risk of coronary heart disease (1 2) and thrombotic

events, particularly when they have membranous ncphnopathy

36

months months months

6 12 18 24 30

months months months

448 Journal of the American Society of Nephrology

>.(5

0)ci

:�Ca,0

0.

Figure 3. Changes in daily urinary protein excretion (mean ± SD) in patients followed for 12, 24, and 36 mo (�, methylprednisolone plus

chboranthucil; A, methylprednisobone plus cycbophospharuide).

Figure 4. Changes in the reciprocal of plasma creatinine levels (mean ± SD) in the cohort groups followed for 12, 24, and 36 mo (�,

methylprednisobone plus chlorambucil: A, methylprednisolone plus cyclophosphamide).

Table 2. Clinical status at the last follow-up visit

Status .Chborambucil(44 patients)

Cycbophosphamide(43 patients)

Follow-up (median) 36 42

Complete remission 12 (27.2%) 16 (37.2%)

Partial remission 24 (54.5%) 24 (55.8%)Stable 7 (15.9%) 1 (2.3%)

Worsened 1 (2.3%) 2 (4.6%)

Mean proteinunia (g/d) 2. 1 1 ± 2.89 1 .69 ± 2.36

Mean plasma creatinine 1 .25 ± 1 .37 1 .32 ± I .72

(mg/db)

as the underlying disease ( I 3, 14). Moreover, persistent pro-

teinunia may make them susceptible to impaired renal function

in the long-term ( 15,16).

The second outcome measure of this trial was the effect of

treatment on renal function, as assessed by the reciprocal of

plasma creatinine over time. In both groups, the mean recip-

rocal of plasma creatinine improved in the 12-mo cohort and

remained stable in either group in the 24- and 36-mo cohorts.

The follow-up period of this study was short, but it has been

shown that following the reciprocal of plasma creatinine even

for a short time may predict the long-term prognosis in patients

with membranous nephropathy ( 1 7). Although these data

should be extrapolated with caution, one may speculate that

these patients should have little risk of developing renal insuf-

ficiency even in the bong-term. This view is reinforced by

previous results showing that patients who attained a complete

remission (1 8) or even partial remission ( I 9) of the ncphrotic

syndrome were unlikely to develop any deterioration of renal

function in the long-term follow-up.

The use of corticosteroids and alkylating agents may expose

these patients to side effects. Several patients had to stop

treatment because of adverse effects. In all cases, patients with

side effects that developed during treatment recovered corn-

pletcly after stopping treatment and adequate management.

The rate of adverse events leading to discontinuation in pa-

ticnts given chborarubucib was 1 2%, similar to the 9.5% (2) and

9% (3) that we observed in two previous controlled trials with

methylprednisobone and chborarubucil. A smaller number of

patients given cycbophospharuide had to stop treatment because

of side effects. Moreover, none of them suffered from herpes

Idiopathic Membranous Nephropathy 449

zoster, which occurred in four patients given chborarubucil. The

main concern with the usc of cytotoxic drugs is the potential

risk of neoplasia in the long-term. Leukemia and, more rarely,

lyruphoma (20,2 1) arc the most frequent types of cancer that

develop in patients given chborambucil, whereas bladder can-

cen and lymphoreticular tumors (22,23) arc the most frequent

malignancies in patients treated with cycbophospharuide. None

of our patients suffered from any of these tumors. It is possible,

however, that alkylating agents may expose them to other

forms of neoplasia. In immunosuppressed patients, the hazard

of neoplasia is roughly related to the cumulative dose and to

the duration of cytotoxic therapy (24-26). Whether short-term

administration of an alkylating agent increases the risk of

cancer in patients with membranous ncphropathy is still un-

clear. The problem is further complicated by the fact that

membranous ncphropathy may be the first clue of an underly-

ing cancer (27,28). Even in this study, two patients were

recognized to have malignancy only after randomization. One

patient received steroids for only a few days, whereas the other

patient did not receive any treatment for ruerubranous nephrop-

athy. On the other hand, we observed one case of prostatic

cancer 5 yr after treatment in the group given chborarubucil and

one case of laryngeal cancer 4 yr after treatment in the group

given cyclophosphamide. To evaluate whether these cancers

represented a chance event or a complication of treatment, we

compared the cumulative risk of cancer in patients given meth-

ylprcdnisobone and chborambucil for 6 mo with that of a

general white population. To increase the number of patients at

risk (in addition to the patients of this study), 79 other patients

treated with steroids and chborambucil who participated to two

previous controlled trials in rucmbranous ncphropathy (3,4)

were considered. Therefore, we were able to collect informa-

tion about 662 patients/years treated with mcthylprcdnisobone

and chborambucil for 6 mo. In this population, we observed

three cases of cancer: one in the present study and two cases in

the two previous controlled studies. Thus, the cumulative risk

of developing cancer in our patients given mcthylprednisobonc

and chbonambucil was 4.53/1000 patients per year (95% CI,

0.935 to 1 3.24 1 ). This hazard is similar to that observed in a

general white population, being the average annual incidence

of primary cancer among white population (4.33/1000 for men

and 3.40/1000 for women) (29). We do not have controlled

data with the usc of cyclophospharuide other than those re-

ported in this trial, clearly too a small number of patients to

draw any conclusion.

In conclusion, this controlled trial confirms that the altenna-

tion of steroids and cytotoxic drugs for 6 mo favors remission

of the nephrotic syndrome and protects from renal dysfunction.

No significant difference could be observed between patients

given chborarubucil or cycbophosphamidc. On the basis of these

results, either of these cytotoxic drugs may be used in mem-

branous ncphnopathy whenever the clinician feels that a par-

ticular patient should be treated.

AcknowledgmentsWe are indebted to the following physicians who participated in

this study: G. Banfi (Ospedale Maggiore Milano), R. Maiorca (Uni-

versity of Brescia), D. Bizzarri (Ospedale Civile Arezzo), and P.

Zucchelli (Ospedale Malpighi Bologna). This work was supported in

part by a grant from Ospedabc Maggiore di Milano.

References1 . Ponticelli C, Zucchelli P, Imbasciati E, Cagnoli L, Pozzi C,

Passerini P, Grassi C, Liruido D, Pasquali S. Volpini T, Sasdelli

M, Locatelli F: Controlled trial of methylprednisolone and

chborarubucil in idiopathic membranous nephropathy. N EngI

J Med 310: 945-950, 1984

2. Ponticelbi C, Zucchelli P. Passerini P, Cagnoli L, Cesana B, Pozzi

C, Pasquali 5, Imbasciati E, Grassi C, Redaelli B, Sasdelli M,

Locatelbi F: A randomized trial of methylprednisolone and

chlorambucil in idiopathic membranous nephropathy. N Engl

J Med 320: 8-13, 1989

3. Ponticelbi C, Zucchelli P. Passerini P. Cesana B, and the Italian

Idiopathic Membranous Nephropathy Treatment Study Group:

Methylprednisolone plus chborambucil as compared with meth-

ylprednisolone alone for the treatment of idiopathic membranous

nephropathy. N Engl J Med 327: 599-603, 1992

4. Ponticelli C, Zucchelli P. Passerini P. Cesana B, Locatelli F,

Pasquali 5, Sasdelli M, Redaelli B, Grassi C, Pozzi C, Bizzarri D,

Banfi G: A 10-year follow-up of a randomized study with meth-

ylprcdnisobone and chlorambucil in membranous nephropathy.

Kidney bnt 48: 1600-1 604, 1995

5. Reichert U, Huysmans FTM, Assman K, Koene RAP, Wetzels

JFM: Preserving renal function in patients with membranousnephropathy: Daily oral chborambucil compared with intermit-

tent monthly pulses of cycbophosphamide. Ann bntern Med I 21:

328-333, 1994

6. Brunkhorst R, Wrenger E, Koch KM: Low-dose prednisobone/

chlorarubucib therapy in patients with severe membranous gb-

merubonephritis. Cli,z bnt’est 72: 277-280, 1994

7. Mathieson PW, Turner AN, Maidment CGH, Evans DI, Rees Al:

Prednisobone and chborambucil treatment in idiopathic membra-

nous nephropathy with deteriorating renal function. Lancet 2:

869-872, 1988

8. Ehrenreich T, Churg I: Pathology of ruerubranous nephropathy.

In: Pathology Annual, edited by Sommers SC, New York, Apple-

ton-Century-Crofts, I 986, pp 145-186

9. Kaplan EL, Meier P: Nonparametnic estimation from incomplete

observations. J Am Stat Assoc 53: 457-481, 1958

10. Kirk RE: Experimental Design: Procedures for the Behavioral

Sciences, 2nd Ed., Monterey, CA, Brooks-Cole, 1982, pp 489-

505

1 1 . Geisser 5, Greenhouse SW: An extension of Box’s results on the

use of the F distribution in multivariate analysis. Ann Math Stat

29: 885-891, 1958

I 2. Ordonez ID, Hiatt RA, Killebrew El, Fireman BH: The increasedrisk of coronary heart disease associated with nephrotic syn-

drorue. Kidney In! 44: 638-642, 1993

13. Bellomo R, Atkins RC: Membranous nephropathy and thrombo-embolism: Is prophylactic anticoagulation warranted? Nephron

63: 249-253, 1993

14. Sarasin FP, Schifferli IA: Prophylactic oral anticoagulation in

nephrotic patients with idiopathic ruembranous nephropathy.

Kidney but 45: 578 -585, 1994

15. Remuzzi G, Bertani T: Is glomerulosclerosis a consequence of

altered gbomerular permeability to macromolecules? Kidney bnt

38: 384-394, 1990

16. Burton C, Harris KPG: The role ofproteinuria in the progression

of chronic renal failure. Am J Kidney Dis 6: 765-77 1 , 1996

450 Journal of the American Society of Nephrology

17. Davison AM, Cameron IS, Kerr DN, Ogg CS, Wilkinson RW: The

natural history of renal function in untreated idiopathic membranousgbomembonephnitis in adults. Cli,i Nephrol 22: 61-67, 1984

18. Passenini P. Pasquali 5, Cesana B, Zucchelli P. Ponticelli C:

Long-term outcome of patients with membranous nephropathy

after complete remission of proteinuria. Nephrol Dial Transplant

4: 525-529, 1989

19. Ponticelli C, Passenini P. Altieri P. Locatelli F, Pappalettera M:Remissions and relapses in idiopathic membranous nephropathy.

Nephrol Dial Transplant 7[Suppl I]: 85-90, 1992

20. Berk PD, Goldberg ID, Silverstein MN, Weinfeld A, Donovan

PB, Ellis IT, Landaw SA, Laszbo I, Najean Y: Increased mci-dence of acute leukemia in polycythemia vera associated withchborambucil therapy. N EngI J Med 304: 441-447, 1981

2 1 . Kahn MF, Arlet I, Bboch-Michel H, Caroit M, Chaouat Y, Renier

IC: Leuc#{233}miesaigues apr#{232}straitement par agents cytotoxiques

en rheumatologie: I 9 observations chez 2006 patients. Nouv

Presse Med 8: 1393-1397, 1979

22. Wall RL, Clausen KP: Carcinoma of the urinary bladder inpatients receiving cycbophosphamide. N EngI J Med 293: 271-

273, 1975

23. Kinlen LI, Sheil AGR, Peto I, Doll R: Collaborative United

Kingdom-Australasian study of cancer in patients treated with

immunosuppressive drugs. Br Med J 2: 1461-1466, 1979

24. Smith AG, Prentice AG, Lucie NP, Browning ID, Dagg IH,

Rowan RM: Acute myelogenous leukemia following cytotoxic

therapy: Five cases and a review. Q J Med 5 1: 227-240, 1982

25. Grunwald HW, Rosner F: Acute leukemia and immunosuppres�

sive drug use. Arch Intern Med 139: 461-466. 1979

26. Baltus JAM, Boersrua 1W, Hartman AP, Vandenbroucke IP: The

occurrence of malignancies in patients with rheumatoid arthritis

treated with cycbophosphamide: A controlled retrospective fol-

low-up. Ann Rheum Dis 42: 368-373, 1983

27. Alpers CE, Cotran RS: Neoplasia and glomerular injury. Kidney

but 30: 465-473, 1986

28. Burstein DM, Korbet SM, Schwartz MM: Membranous gbomer-

ulonephnitis and malignancy. Am J Kidney Dis 22: 5-10, 1993

29. Fraurueni IF, Devasa 55, Hoover RN, Kinlen Li: Epidemiology

of cancer. In: L�ancer Principles 38 Practice of Oncology, 4th

Ed., edited by Devita VT, Hellman 5, Rosenberg SA, Philadel-

phia, I. B. Lippincott, 1993, pp 150-181