Pediatr Nephrol (2005) 20:10–14
DOI 10.1007/s00467-004-1615-9
EDITORIAL COMMENTARY
J. C. Davin · M. P. Merkus
Levamisole in steroid-sensitive nephrotic syndrome of childhood:
the lost paradise?
Received: 27 October 2003 / Revised: 6 July 2004 / Accepted: 12 July 2004 / Published online: 17 September 2004

IPNA 2004
Abstract Among the different drugs used for sparing
steroids in steroid-sensitive nephrotic syndrome (SSNS)
with frequent relapses and steroid dependency, levamisole
is the least toxic and the least expensive. However, it is
neither approved for this indication nor widely used in
Europe. This may be explained by the difficulty in ob-
taining levamisole in some countries and the lack of good
quality evidence for its effectiveness. Evidence is limited
to three clinical trials that all suffered from methodolog-
ical limitations. Statistical synthesis of these trials showed
that levamisole reduces the risk of a relapse during treat-
ment (relative risk 0.60, 95% confidence interval 0.45–
0.79). From the available information, no conclusions can
be drawn on the steroid-sparing effect, the long-term ef-
ficacy, and safety, as well as possible differences in ef-
ficacy in different subgroups of SSNS patients. The con-
firmation of a favorable effect of levamisole on the re-
duction of the frequency of relapses and on sparing ster-
oids in an adequately powered, double-blind, placebo-
controlled, randomized, multi-center clinical trial will
promote consensus on the place of levamisole in the
treatment of SSNS of childhood. Follow-up should be at
least 1 year to evaluate long-term efficacy and side ef-
fects. If the results of such a clinical trial confirm the
beneficial effects of levamisole in nephrotic syndrome,
this may allow registration for this indication and interest
companies other than Jansen-Cilag, which only recently
has decided to stop its production.
Keywords Idiopathic nephrotic syndrome · Frequent
relapsers · Steroid-dependent nephrotic syndrome ·
Levamisole
J. C. Davin ())
Pediatric Nephrology Unit,
Emma Children’s Hospital/Academic Medical Center,
9 Meibergdreef, 1105 AZ Amsterdam Z-O, The Netherlands
e-mail: j.c.davin@amc.uva.nl
M. P. Merkus
Center for Pediatric Clinical Epidemiology,
Emma Children’s Hospital/Academic Medical Center,
Amsterdam, The Netherlands
The incidence of nephrotic syndrome (NS) varies from
2 to 7 new patients per 100,000 children [1]. In about
80%, the etiology is minimal change nephrotic syndrome
(MCNS) [1], probably caused by an aberrant immune
response with a type 2 cytokine bias [2]. Of children with
MCNS, 95% are steroid sensitive (SSNS), but 70% of
SSNS patients relapse [3]. Of the latter, 50%–60% are
estimated to experience relapses as soon as steroid ther-
apy is stopped or when the prednisone dosage is de-
creased [3] and are therefore exposed to side effects of
long-term steroid therapy, including growth failure, hy-
pertension, infections, and osteoporosis.
To reduce steroid toxicity, immunosuppressive agents
have been proposed for frequently relapsing NS patients.
Cyclophosphamide and cyclosporine are the two most
widely used. Cyclophosphamide has been known to ef-
fectively reduce the frequency of relapses since the early
1970s [4, 5]. Most SSNS patients do not relapse on cy-
closporine [3, 6], which is widely used when cyclophos-
phamide has failed. Although effective, repeated or long-
term use of these immunosuppressive agents may be as-
sociated with serious toxic effects such as carcinogenesis
and sterility, or nephrotoxicity and hypertension [3, 7].
Since the early 1980s levamisole has also been
used. Levamisole is an antihelminthic agent that has
immunomodulatory properties. Experiments conducted
in Brown Norway rats suggest that levamisole may act by
augmenting the type 1 response and reciprocally by
down-regulating the type 2 response by selective induc-
tion of gene transcription of the key cytokines (inter-
leukin-18) [8].
Since the first publication of Tanphaichitr et al. [9],
several studies have suggested that levamisole reduces the
frequency of relapses and diminishes steroid doses in
steroid-dependent patients [10, 11, 12, 13, 14, 15, 16, 17,
18, 19, 20, 21, 22, 23, 24, 25, 26, 27], both as a first
alternative for steroids and after failure of cyclophos-
phamide or cyclosporine [21, 27]. However, only three of
these studies were randomized controlled trials (RCTs)
[10, 14, 15]. A recent systematic review and meta-anal-
ysis of RCTs of non-corticosteroid treatment in childhood

NS summarized these [6]. Since then no other RCT on
this topic has been published. Table 1 presents the char-
acteristics and results of these RCTs. Levamisole was
administered for 4 [14], 6 [10], or 12 [15] months.
Overall, levamisole appeared more effective in main-
taining remission during treatment than prednisone alone.
A statistical synthesis of the three RCTs yielded a reduced
risk of a relapse with levamisole compared with predni-
sone alone (relative risk 0.60, 95% confidence interval
0.45–0.79) during administration [6]. Results regarding a
sustained effect of treatment after levamisole withdrawal
are not consistent: one trial reported 87% of levamisole
patients relapsed within 3 months of cessation of treat-
ment [14], whereas a sustained effect after stopping le-
vamisole treatment was reported by another [10].
All three RCTS were found to have methodological
limitations. Two RCTs [10, 15] suffered from unclear
allocation concealment, no blinding of patients and in-
vestigators, insufficient statistical power, and inadequate
definitions of a relapse. Although the British Association
of Pediatric Nephrology (BAPN) trial [14] had adequate
allocation concealment, was double blinded and placebo
controlled, its definition of a relapse (proteinuria for at
least 3 days without considering hypoalbuminemia) risks
confusing transient and spontaneously resolving protein-
uria with clinical relapses necessitating the use of ster-
oids. This could have been avoided if it had been recorded
whether the exit relapse prompted a further course of
steroids. Since this was not done, an evaluation of a ste-
roid-sparing effect was not possible [14]. In addition,
more than half of the patients in the BAPN study had
already received alkylating agents, complicating an eval-
uation of levamisole as a first-step alternative treatment
for steroids. A recent retrospective study [27] suggested
that levamisole may be more effective in patients who
received previous alkylating therapy. None of the trials
reported important side effects.
In conclusion, evidence to date is limited to the ob-
servation that levamisole reduces the risk of a relapse
during treatment. However, this meta-analysis is ham-
pered by limitations of the primary data [6]. Since it has
been shown that meta-analyses of low-quality trials may
overestimate treatment effect [28, 29], the reported rela-
tive risk may be inflated. No conclusions can be drawn on
the long-term efficacy and safety and possible differences
in efficacy in different subgroups of SNSS patients (low-
versus high-dose steroid dependency, with versus without
previous steroid-sparing agent).
This lack of evidence may explain at least partly why
levamisole is not more widely used. Only in France, ac-
cording to national protocol, is levamisole systematically
used as a first alternative drug to steroids. A recent in-
quiry on the use of levamisole in frequently relapsing
SSNS conducted by e-mail showed that among non-
French members of the European Society for Pediatric
Nephrology (ESPN) (n=182), 56% (n=26) of responders
(n=46) used levamisole (J.C. Davin, personal communi-
cation). However, since levamisole users may be more
inclined to answer than non-users, the actual proportion
11
may be even lower. These numbers should be interpreted
with some caution because of the low response rate
(25%).
Also, the difficulty in obtaining the drug may con-
tribute to its low user rate. The procedure for obtaining the
drug in Europe varies from country to country and since
there is no official registration for this indication, its
availability is influenced by local rules. In France, le-
vamisole is supplied by hospital pharmacies according to a
complex procedure of temporary authorization of use. In
Italy, the drug can be obtained directly from the pharma-
ceutical company for compassionate use. In The Nether-
lands, before the year 2004, the clinician had to supple-
ment a conventional prescription with a special form
specifying the indication in order to obtain reimbursement
from insurances. In 2004, the levamisole distributor Jan-
sen-Cilag decided to provide levamisole for free in The
Netherlands, making any special procedure unnecessary.
In Canada, levamisole is not distributed any longer by
pharmaceutical societies and therefore cannot be obtained.
In Australia, levamisole is only approved for colonic can-
cer and therefore cannot be obtained from the Australian
Pharmaceutical Benefits Scheme, funded by the Federal
Government. The drug has to be obtained from hospital
pharmacies with special approval and is paid for by the
hospitals. Remarkably, in contrast to Europe, levamisole is
very expensive in Australia (the cost of one 50-mg tablet is
U.S. $ 3.79 versus U.S. $ 0.65 in Belgium). Manufacturing
costs are unlikely to be responsible given the distribution
for free in The Netherlands. The latest information is that
very soon levamisole will not be available any longer in
any country, since Jansen-Cilag has only recently decided
to stop its production. This decision has been made be-
cause of the low rate of use of levamisole (personal
communication Jansen-Cilag to the first author).
Given that the majority of children with SSNS enter
long-term remission, it is highly desirable to minimize
harmful side effects of medication to prevent relapses and
shorten the period of morbidity. With respect to drug
toxicity levamisole should be the drug of choice. Le-
vamisole is generally well tolerated. Its main side effect is
neutropenia, which necessitates a close follow-up of the
leukocyte count. However, leukocytosis is always re-
versible after withdrawing the drug. Other rarely reported
side effects during levamisole treatment for NS included
vasculitis, liver toxicity, and convulsions, but these were
also always reversible after withdrawal of treatment [30,
31, 32].
In contrast, cyclophoshamide is associated with serious
infections, hair loss, and cystitis [6]. There are also con-
cerns regarding adverse effects on fertility and an in-
creased risk of malignancy. A recent retrospective cohort
study of adults suffering from end-stage renal failure from
childhood (mean follow-up since first renal replacement
therapy 15.5 years, range 0.2–30 years) reported that use
of cyclophosphamide at cumulative doses of 100–200 mg/
kg was associated with an elevated risk of cancer (relative
risk 4.26, 95% confidence interval 1.78–10.20) [33]. Of
course, the immunosuppression used for transplantation is
 12

Table 1 Characteristics of randomized controlled trials (RCTs) evaluating levamisole in steroid-sensitive nephrotic syndrome (SSNS) in childhood (CI confidence interval)
Reference Patients Design n Levamisole group Control group Definitions Follow-up n (%) relapses at Relative risk
follow-up (95% CI)
BAPN 1991 Steroid- Placebo- 61 n=31 n=30 Relapse: proteinuria 16 weeks Levamisole 17/31 0.63 (0.45–0.90)
[14] dependent con- 3+ >3 days, confirmed by (55%)
SNSS trolled, albumin/creatinine >2 mg/mg
double- or protein/creatinine >200 mg/
blinded mmol
RCT Placebo 26/30 (87%)
Levamisole Placebo ad for Remission: protein-free early 6 months Levamisole 27/31 0.90 (0.78–1.05)
2.5 mg/kg ad for 16 weeks morning on 3 consecutive (87%)
16 weeks days, by Albustix recording
of 0/trace
Placebo 29/30 (97%)
Prednisone Prednisone
8 weeks progres- 8 weeks progres-
sively reduced sively reduced
Dayal et al. Previous RCT 37 n=23 n=14 Relapse: edema, proteinuria 12 months Prior relapsers 0.57 (0.31–1.05)
1994 [15] relapse(s) >1 g/m2 per day and hypo-
albuminemia <2.5 g/dl
Levamisole 9/22 (41%)
Levamisole No treatment for Remission: no edema and Control 10/14 (71%)
2–3 mg/kg twice a 12 months oliguria, 24-h urine protein
week for 12 months excretion <1 g/m2 per day
and serum albumin >2.5 g/dl
Rashid et al. Steroid RCT 40 n=20 n=20 Relapse: not defined 6 months Levamisole 6/20 (30%) 0.50 (0.23–1.07)
1996 [10] dependent Control 12/20 (60%)
and fre- Levamisole Prednisone Remission: disappearance of 44 weeks Levamisole 11/20 0.61 (0.40–0.93)
quently 2.5 mg/kg ad for 6 months proteinuria (<4 mg/m2 per (55%)
relapsing 6 months hour) for at least 3 consecutive
SNSS days under prednisone treat-
ment of 60 mg/m2 for 8 weeks
Control 18/20 (90%)
Prednisone Progressively
6 months, progres- reduced
sively reduced

not comparable to that used for SNSS. However, even if
cellular mutations induced by cyclophosphamide are not
sufficient to lead to cancer by themselves, they may fa-
cilitate the development of cancer in case of a ‘second hit’
later in life [34]. An increased risk of oligo- or azoo-
spermia has been reported even for cumulative cyclo-
phosphamide doses less than 200 mg/kg [7].
Cyclosporine carries the risk of nephrotoxicity. Other
adverse drug reactions comprise gingival hypertrophy,
hirsutism, and hypertension [6]. Finally, since the finan-
cial cost of cyclosporine is more than 10 times that of
levamisole, a successful levamisole treatment will also
allow a considerable cost reduction.
To conclude, the confirmation in an adequately pow-
ered, double-blind, placebo-controlled, randomized, multi-
center clinical trial of a favorable effect of levamisole on
reduction of the frequency of relapses and on sparing
steroids will promote consensus on the role of levamisole
in the treatment of SSNS of childhood. As a consequence
this may allow its official registration for this indication,
increase its use, and stimulate new financial interest from
pharmaceutical companies. Follow-up should be at least
1 year to evaluate long-term efficacy and side effects.
Ideally, such a trial should also be adequately powered to
allow subgroup analyses for different grades of steroid
dependency and previous steroid-sparing agents. Such a
clinical trial is currently being designed by an inter-
national collaboration between France, The Netherlands,
and Belgium. Colleagues from other countries are also
cordially invited to join.
The disappearance of an attractive drug from the
market because of lack of sufficient financial profit raises
the following question: should society leave the respon-
sibility of therapeutic choices to private initiatives driven
only by financial interests? The solution to this problem
has already been found in some countries where the
production of medical drugs is directly controlled by the
state.
Acknowledgement The authors are very grateful to Thomas Martin
Barratt and Jonathon Craig for providing their personal expert view
on levamisole treatment and clinical trials in SSNS. They also
thank Jonathon Craig for his critical appraisal of the manuscript.
They thank all members of the ESPN for participating in the e-mail
investigation, with particular thanks to Giulana Lama, Albert
Bensman, Elisabeth Hodson, Thierry Schuurmans, Guido Filler,
and Aicha Merouani for providing data on the availability of le-
vamisole in their countries.
References
1. ISKDC (1978) The primary nephrotic syndrome in children:
prediction of histopathology from clinical and laboratory
characteristics at time of diagnosis. A report of the International
Study of Kidney Disease in Children. Kidney Int 13:159–165
2. Mathieson PW (2003) Immune dysregulation in minimal
change nephropathy. Nephrol Dial Transplant 18 [Suppl 6]:26–
29
3. Niaudet P (2004) Steroid-sensitive nephrotic syndrome in
children. In: Avner ED, Harmon WE, Niaudet P (eds) Pediatric
13
nephrology. Lippincott Williams and Wilkins, Philadelphia,
pp 543–556
4. Chiu J, McLaine PN, Drummond KN (1973) A controlled
prospective study of cyclophosphamide in relapsing, corti-
costeroid-responsive, minimal-lesion nephrotic syndrome in
childhood. J Pediatr 82:607–613
5. Barratt TM, Soothill JF (1970) Controlled trial of cyclophos-
phamide in steroid-sensitive relapsing nephrotic syndrome of
childhood. Lancet II:479–482
6. Durkan A, Hodson E, Willis N, Craig J (2001) Non-cortico-
steroid treatment for nephrotic syndrome in children. Kidney
Int 59:1919–1926
7. Latta K, Schnakenburg C von, Ehrich J (2001) A meta-analysis
of cytotoxic treatment for frequently relapsing nephrotic syn-
drome in children. Pediatr Nephrol 16:271–282
8. Szeto CC, Gillespie KM, Mathieson PW (2000) Levamisole
induces interleukin 18 and shifts TH1/TH2 balance. Immu-
nology 100:217–224
9. Tanphaichitr P, Tanphaichitr D, Sureeratanan J, Chatasingh S
(1980) Treatment of nephrotic syndrome with levamisole.
J Pediatr 96:490–493
10. Rashid HU, Ahmed S, Fatima N, Khanam A (1996) Levamisole
in the treatment of steroid dependent or frequent relapsing
nephritic syndrome in children. Bangladesh Ren J 15:6–8
11. Niaudet P, Drachman R, Gagnadoux MF, Broyer M (1984)
Treatment of idiopathic nephrotic syndrome with levamisole.
Acta Paediatr Scand 73:637–641
12. La Manna A, Polito C, Del Gado R, Foglia AC (1988) Le-
vamisole in children’s idiopathic nephrotic syndrome. Child
Nephrol Urol 9:200–202
13. Drachman R, Schlesinger M, Alon U, Mor J, Etzioni A, Shapira
H, Ohali M, Drukker A (1988) Immunoregulation with le-
vamisole in children with frequently relapsing steroid respon-
sive nephrotic syndrome. Acta Paediatr Scand 77:721–726
14. BAPN (British Association for Paediatric Nephrology) (1991)
Levamisole for corticosteroid-dependent nephrotic syndrome in
childhood. Lancet 337:1555–1557
15. Dayal U, Dayal AK, Shastry JC, Raghupathy P (1994) Use of
levamisole in maintaining remission in steroid-sensitive neph-
rotic syndrome in children. Nephron 66:408–412
16. Meregalli P, Bianchetti MG, Imoberdorf G, Lutschg J, Rey-
mond D, Oetliker OH (1994) Levamisole in children with
frequently recurring idiopathic nephrotic syndrome. Schweiz
Med Wochenschr 124:801–805
17. Mongeau JG, Robitaille PO, Roy F (1988) Clinical efficacy of
levamisole in the treatment of primary nephrosis in children.
Pediatr Nephrol 2:398–401
18. Mancini ML, Rinaldi S, Rizzoni G (1974) Treatment of par-
tially corticosteroid-sensitive nephrotic syndrome with le-
vamisole. Pediatr Nephrol 8:788
19. Bagga A, Sharma A, Srivasta RN (1997) Levamisole therapy in
corticosteroid-dependent nephrotic syndrome. Pediatr Nephrol
11:415–417
20. Kemper MJ, Amon O, Timmermann K, Altrogge H, Muller-
Wiefel DE (1998) The treatment with levamisole of frequently
recurring steroid-sensitive idiopathic nephrotic syndrome in
children. Deutsch Med Wochenschr 123:239–243
21. Tenbrock K, Mller-Berghaus J, Fuchshuber A, Michalk D,
Querfeld U (1998) Levamisole treatment in steroid-sensitive
and steroid-resistant nephrotic syndrome. Pediatr Nephrol 12:
459–462
22. Ginevri F, Trivelli A, Ciardi MR, Ghiggeri GM, Parfumo F,
Gusmano R (1996) Protracted levamisole in children with
frequent-relapse nephrotic syndrome. Pediatr Nephrol 10:550
23. Fu LS, Chi CS (2000) Levamisole in steroid-sensitive nephrotic
syndrome children with steroid-dependency and/or frequent
relapses. Acta Paediatr Taiwan 41:80–84
24. Alsaran K, Grisaru S, Stephens D, Arbus G (2001). Levamisole
vs. cyclophosphamide for frequently-relapsing steroid-depen-
dent nephrotic syndrome. Clin Nephrol 56:289–294
25. Donia AF, Amer GM, Ahmed HA, Gazareen SH, Moustafa FE,
Shoeib AA, Ismail AM, Khamis S, Sobh MA (2002) Levami-
14
sole: adjunctive therapy in steroid-dependent minimal change
nephrotic children. Pediatr Nephrol 17:355–358
26. Alshaya HO, Kari JA (2002) Levamisole treatment in steroid
sensitive nephrotic syndrome. Saudi Med J 23:1101–1104
27. Abeyagunawardena AS, Dillon MJ, Rees L, Hoff W van’t,
Trompeter RS (2003) The use of steroid-sparing agents in
steroid-sensitive nephrotic syndrome. Pediatr Nephrol 18:919–
924
28. Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M,
Tugwell P, Klassen TP (1998) Does quality of randomised trials
affect estimates of intervention efficacy reported in meta-
analyses? Lancet 352:609–613
29. Schulz KF, Chalmers I, Hayes RJ, Altman DG (1995) Empir-
ical evidence of bias: dimensions of methodological quality
associated with estimates of treatment effects in controlled
trials. JAMA 273:408–412
30. Barbano G, Ginevri F, Ghiggeri GM, Gusmano R (1999) Dis-
seminated autoimmune disease during levamisole treatment of
nephrotic syndrome. Pediatr Nephrol 13:602–603
31. Bulugahapitiya DT (1997) Liver toxicity in a nephrotic patient
treated with levamisole. Arch Dis Child 76:289
32. Palcoux JB, Niaudet P, Goumy P (1994) Side effects of le-
vamisole in children with nephrosis. Pediatr Nephrol 8:263–264
33. Coutinho HM, Groothoff JW, Offringa MO, Gruppen MP,
Heymans HSA (2001) De novo malignancy after paediatric
renal replacement therapy. Arch Dis Child 85:478–483
34. Knudson AG (1996) Hereditary cancer: two hits revisited.
J Cancer Res Clin Oncol 122:135–140