In the Literature

Corticosteroids in IgA Nephropathy

Claudio Ponticelli and Francesco Locatelli

C orticosteroid use in patients with immunoglobulin A

nephropathy (IgAN) remains controversial. The KDIGO
(Kidney Disease: Improving Global Outcomes) guideline
mean daily urine protein excretion of 1.37 (SD, 1.08) g/d
versus 2.36 (SD, 1.67) g/d in the placebo arm (P < 0.001).
The strengths of this study include the blinded nature of
recommended use of renin-angiotensin-aldosterone system the trial, proper use of the Data and Safety Monitoring
(RAAS) inhibitors as the initial approach, followed by a Board, blinded adjudication of events, and adequate run-in
course of corticosteroids in patients with IgAN who have an period to optimize the use of RAAS inhibitors and ensure
estimated glomerular filtration rate (eGFR) > 50 mL/min/ adequate blood pressure control.
1.73 m2 and proteinuria with protein excretion > 1 g/d.1 However, there are also limitations: participants
The VALIGA (Validation Study of the Oxford Classification randomly assigned to methylprednisolone treatment were
given the same dose regardless of kidney function, which
Commentary on Lv J, Zhang H, Wong MG, et al; TESTING could have had implications for adverse effects. In this
Study Group. Effect of oral methylprednisolone on clinical study, it appears there was no difference in side effects
outcomes in patients with IgA nephropathy: the TESTING observed in patients with eGFRs > 50 or <50 mL/min/
randomized clinical trial. JAMA. 2017;318(5):432-442. 1.73 m2. However, no information for side effects was
provided for participants randomly assigned to the meth-
of IgAN) retrospective study2 and 2 randomized controlled
ylprednisolone arm who had been previously treated
trials (RCTs)3,4 reported that corticosteroids improved pro-
(probably unsuccessfully) with corticosteroids or other
teinuria and protected kidney function in the long term. A
immunosuppressive drugs and another 11 participants
meta-analysis confirmed the efficacy of corticosteroids, but
in the methylprednisolone arm who had a baseline
raised concerns about a significant risk for gastrointestinal
eGFR < 30 mL/min/1.73m2. Both prior use of immu-
tract adverse events.5 The recently reported TESTING
nosuppressants and reduced kidney function
(Therapeutic Evaluation of Steroids in IgA Nephropathy
(eGFR < 30 mL/min/1.73m2 ) increase susceptibility
Global) RCT assessed the long-term efficacy and safety of oral
to side effects of corticosteroids.8,9 Thus, further analysis
methylprednisolone compared to matching placebo, on a
might inform the rate and severity of side effects in these
background of routine RAAS inhibitor therapy in patients
participants. We believe that patients who failed to
with IgAN.6 An increased rate of serious side effects of cor-
respond to previous treatments and those with decreased
ticosteroids was also found by TESTING.6
kidney function should be stratified at randomization to a
What Does This Important Study Show? high-risk group and analyzed because their results may not
TESTING evaluated the safety and effectiveness of methyl- be generalized to patients without prior treatment and/or
prednisolone in Chinese patients6 with biopsy-proven IgAN, preserved kidney function. A therapeutic regimen that may
eGFRs between 20 and 120 mL/min/1.73 m2 (estimated be effective and relatively safe in patients with normal or
with the use of the CKD-EPI [Chronic Kidney Disease mildly decreased kidney function may be very toxic in
Epidemiology Collaboration] creatinine equation),7 and patients with advanced deterioration of kidney function.
proteinuria with protein excretion > 1 g/d. After a 2-month Thus, these latter high-risk patients might receive corti-
run-in period, 262 participants met the eligibility criteria and costeroid therapy, but only after careful screening for their
were randomly assigned to receive placebo or methylpred- clinical conditions and assessment of at least partial
nisolone, 0.6 to 0.8 mg/kg, for 2 months, which was reversibility of kidney function. In addition, the oral
reduced by 8 mg every month over 6 to 12 months. Because corticosteroid doses should be reduced, infectious foci
of an excess of serious side effects in the corticosteroid arm in should be eradicated, and the clinical conditions should be
comparison to the placebo arm (20 vs 4; P < 0.001), the trial frequently monitored. In immunosuppressed patients,
was interrupted after a median follow-up of 25 months, concurrent antibiotic treatment may prevent infection or
when 86% of participants had completed study treatment. avoid the transformation of a trivial infection into a severe
Most side effects occurred in the first 3 months when the complication.
methylprednisolone dosage was highest. At interruption, the
primary composite end point (50% decrease in eGFR, end- How Does This Study Compare to Previous
stage kidney disease, or death due to kidney disease) Studies?
occurred in 5.9% of participants in the methylprednisolone Three open-label RCTs, smaller than TESTING, have eval-
arm versus 15.9% in the placebo arm (P = 0.02). In addition, uated the role of corticosteroids in IgAN. In an Italian
the methylprednisolone arm had a lower mean annual rate of multicenter study, treatment included an intravenous
eGFR decline (−1.79 mL/min/1.73 m2) compared to the methylprednisolone pulse given for 3 consecutive days at
placebo arm (−6.95 mL/min/1.73 m2; P = 0.03) and lower the beginning of months 1, 3, and 5, followed by oral

160 AJKD Vol 71 | Iss 2 | February 2018

In the Literature

prednisone, 0.5 mg/kg, every other day for 6 months, and azathioprine. However, malignancies developed in 2
approximately half the dose of TESTING. Participants had participants who received cyclophosphamide and
biopsy-proven IgAN, serum creatinine concentrations azathioprine.
≤ 1.5 mg/dL, and proteinuria with protein excretion of 1 Not all participants enrolled in the STOP-IGAN trial
to 3 g/d.3 In comparison to TESTING, there were impor- were at high risk. Rather, patients with proteinuria with
tant differences. All 86 participants were white and had protein excretion > 3.5 g/d or progressive loss of eGFR
serum creatinine concentrations ≤ 1.5 mg/dL. The study were excluded, while participants with proteinuria with
was not discontinued, so median follow-up was 8 years. protein excretion < 1 g/d and mild eGFR decline (mean,
Ten-year renal survival was significantly better in the 4.7 mL/min/1.73 m2 in 3 years, or 1.57 mL/min/
corticosteroid group (n = 12) than the control group 1.73 m2 per year in patients given supportive treatment)
(n = 7; 97% vs 53%; P = 0.0003). Median daily prote- were included. A 3-year follow-up for patients with such a
inuria significantly decreased in the corticosteroid arm slow progression was likely insufficient to detect any
participants, but increased in the control group. One beneficial effect of corticosteroids. Apart from ethnic dif-
steroid-treated participant developed diabetes; no other ferences, it is difficult to compare this study with TESTING
serious side effects were reported. because of the different inclusion criteria and use of
Another Italian multicenter trial randomly assigned cyclophosphamide and azathioprine in the German study.
97 participants with IgAN, proteinuria with protein At any rate, both the STOP-IGAN and TESTING trials had
excretion > 1g/d, and eGFRs > 50 mL/min/1.73 m2 to follow-up durations too short to allow firm conclusions
ramipril alone or to a combined therapy of ramipril and about the efficacy and long-term safety of corticosteroids.
a 6-month course of prednisone (0.8-1 mg/kg/d for 2 While TESTING showed an increased risk for side effects,
months, followed by monthly dose reductions of particularly in the first 3 months of treatment, longer
0.2 mg/kg/d during the next 4 months).4 Combined follow-up would be needed to assess the effects of treat-
treatment increased the probability of not reaching the ment on kidney function. It is a pity that the follow-up of
primary outcome measure of doubling of serum creat- participants already enrolled in TESTING was stopped
inine concentration or end-stage kidney disease (85.2% because it is unlikely that these participants would develop
vs 52.1%; P = 0.003) and significantly reduced pro- other steroid-related side effects after methylprednisolone
teinuria. Adverse events were mild in both groups. The treatment discontinuation, but might still have derived
treatment was somewhat similar to that used in treatment benefit.
TESTING, but no patients with eGFRs < 50 mL/min/
1.73 m2 were included, mean follow-up was 96 What Are the Implications for Nephrologists?
months, and all participants were white. From a clinical perspective, one may wonder whether
In the German STOP-IGAN (Supportive Versus Immu- the benefits of corticosteroids outweigh the risk for side
nosuppressive Therapy for the Treatment of Progressive IgA effects in patients with IgAN. The available RCTs
Nephropathy) trial, 162 participants with persistent pro- reported contradicting conclusions. Two studies
teinuria with protein excretion > 0.75 g/d and concluded that the corticosteroid regimens were effec-
eGFRs ≥ 60 mL/min/1.73m2 were assigned to supportive tive and well tolerated,3,4 whereas 2 other trials high-
care with RAAS blockers or supportive care plus 3 methyl- lighted the risk for steroid-related side effects.6,10 It is
prednisolone pulses at the start of months 1, 3, and 5 and not surprising that corticosteroids may increase the risk
oral prednisolone, 0.5 mg/kg, every 48 hours.10 Partici- for adverse events, but the high rate of side effects re-
pants with eGFRs between 30 and 59 mL/min/1.73 m2 ported is unexpected. However, in the STOP-IGAN
were given oral prednisolone, 40 mg/d, tapered to 10 mg/ study, not only glucocorticoids, but also cyclophos-
d over the first 3 months of the study, plus cyclophospha- phamide and azathioprine were used. These agents can
mide, 1.5 mg/kg/d, for 3 months, then azathioprine, clearly increase the risk for serious adverse events,
1.5 mg/kg/d, during months 4 through 36. At the end of 3 whereas there is no evidence that azathioprine may add
years of follow-up, 14 participants in the immunosup- benefit when combined with corticosteroids.11 TESTING
pression group had full clinical remission versus 4 in the used high-dose oral methylprednisolone in individuals
supportive care group (P = 0.001). However, there was no already at high risk for side effects, namely with
difference between the 2 groups in the decrease in eGFRs < 30 mL/min/1.73 m2 and failure to respond to
eGFRs > 15 mL/min/1.73 m2 (28% vs 26%). More pa- previous immunosuppression. The true benefit-risk
tients receiving corticosteroid monotherapy than those also balance of corticosteroid use in this high-risk IgAN
given cyclophosphamide and azathioprine had remission of group remains uncertain in the absence of testing its use
proteinuria, hematuria, or both. This result was expected in with concurrent strategies to mitigate anticipated
view of the different eGFRs in the 2 groups. Serious adverse adverse events, such as the use of prophylactic antibi-
events occurred with similar frequency in the 2 study arms otics for serious infectious complications.
(29 vs 33), even considering participants receiving corti- We believe that patients with eGFRs > 30 mL/min/
costeroids alone and those also given cyclophosphamide 1.73 m2 and proteinuria with protein excretion > 1 g/d

AJKD Vol 71 | Iss 2 | February 2018 161

 In the Literature

merit treatment with corticosteroids. To reduce the risk for References

side effects, we prefer to use the protocol based on pulse 1. KDIGO clinical practice guideline for glomerulonephritis. Kid-
methylprednisolone and alternate-day prednisone because ney Int Suppl. 2012;2:139-274.
such a schedule proved to be well tolerated not only in 2. Tesar V, Troyanov S, Bellur S, et al. Corticosteroids In IgA
patients with IgAN,3 but also in patients with minimal nephropathy: a retrospective analysis from the VALIGA Study.
change disease12 and membranous nephropathy.13 How- J Am Soc Nephrol. 2015;26(9):2248-2258.
ever, we recommend the following guidelines before us- 3. Pozzi C, Andrulli S, Del Vecchio L, et al. Corticosteroid effec-
tiveness in IgA nephropathy: long-term results of a randomized,
ing corticosteroids in IgAN. (1) Carefully ascertain the
controlled trial. J Am Soc Nephrol. 2004;15(1):157-163.
patient’s history and clinical conditions to screen for a 4. Manno C, Torres DD, Rossini M, Pesce F, Schena FP.
history of infection, immunosuppressive therapy, or Randomized controlled clinical trial of corticosteroids plus
severely decreased kidney function, which would then ACE-inhibitors with long-term follow-up in proteinuric IgA ne-
prompt a different personalized treatment. (2) Give short- phropathy. Nephrol Dial Transplant. 2009;24(12):3694-3701.
acting corticosteroid daily by a single administration be- 5. Cheng J, Zhang X, Zhang W, He Q, Tao X, Chen J. Efficacy and
tween 7 AM and 9 AM to mimic the circadian rhythm of safety of glucocorticoids therapy for IgA nephropathy: a meta-
analysis of randomized controlled trials. Am J Nephrol.
cortisol and reduce side effects.14,15 (3) When possible,
2009;30(4):315-322.
alternate-day regimens should be used for long-term 6. Lv J, Zhang H, Wong MG, et al; TESTING Study Group. Effect
treatments. (4) Patients should stop smoking and follow of oral methylprednisolone on clinical outcomes in patients with
a low-calorie and low-sodium diet to reduce the risk for IgA nephropathy: the TESTING randomized clinical trial. JAMA.
diabetes, hypertension, and obesity. (5) Patients already 2017;318(5):432-442.
exercising should continue regular physical activity to 7. Stevens LA, Claybon MA, Schmid CH, et al. Evaluation of the
prevent myopathy, obesity, infection, and cardiovascular Chronic Kidney Disease Epidemiology Collaboration equation
for estimating the glomerular filtration rate in multiple ethnic-
disease. In nonexercising individuals, an exercise program
ities. Kidney Int. 2011;79(5):555-562.
may be initiated under supervision of a trainer with 8. Coppo R. Corticosteroids in IgA nephropathy: lessons from
guidance. These suggestions in conjunction with close recent studies. J Am Soc Nephrol. 2017;28(1):25-33.
monitoring and a good physician-patient relationship may 9. Sarcina C, Tinelli C, Ferrario F, et al. Changes in proteinuria and
help reduce the risk for corticosteroid-related side effects. side effects of corticosteroids alone or in combination with
Future interventions such as use of encapsulated budeso- azathioprine at different stages of IgA nephropathy. Clin J Am
nide, a corticosteroid that mainly works in the distal ileum, Soc Nephrol. 2016;11(6):973-981.
10. Rauen T, Eitner F, Fitzner C, et al. Intensive supportive care plus
may be beneficial.16
immunosuppression in IgA nephropathy. N Engl J Med.
Article Information 2015;373(23):2225-2236.
11. Pozzi C, Andrulli S, Pani A, et al. Addition of azathioprine to
Authors’ Full Names and Academic Degrees: Claudio Ponticelli, corticosteroids does not benefit patients with IgA nephropathy.
MD, and Francesco Locatelli, MD. J Am Soc Nephrol. 2010;21(10):1783-1790.
Authors’ Affiliations: Ospedale Maggiore Milano, Milan (CP); and 12. Imbasciati E, Gusmano R, Edefonti A, et al. Controlled trial of
Ospedale Alessandro Manzoni, Lecco, Italy (FL). methylprednisolone pulses and low dose oral prednisone for
Address for Correspondence: Claudio Ponticelli, MD. E-mail: the minimal change nephrotic syndrome. Br Med J.
[email protected]. 1985;291(6505):1305-1308.
13. Ponticelli C, Zucchelli P, Passerini P, Cesana B. Methylpred-
Support: None.
nisolone plus chlorambucil as compared with methylpredniso-
Financial Disclosure: Dr Locatelli has consulted on IgAN for lone alone for the treatment of idiopathic membranous
Pharmalink. Dr Ponticelli declares that he has no relevant financial nephropathy. The Italian Idiopathic Membranous Nephropathy
interests.
Treatment Study Group. N Engl J Med. 1992;327(9):599-603.
Other Disclosures: Dr Ponticelli was a member of the steering 14. Debono M, Price JN, Ross RJ. Novel strategies for hydrocor-
committee of the Italian IgA study. Dr Locatelli was a member of tisone replacement. Best Pract Res Clin Endocrinol Metab.
the steering committee of the NEFIGAN study and the Italian IgA 2009;23(2):221-232.
study. 15. Cutolo M, Seriolo B, Pizzorni C, et al. Use of glucocorticoids
Peer Review: Received August 27, 2017 in response to an invitation and risk of infections. Autoimmun Rev. 2008;8(2):153-155.
from the journal. Editorial input from an Associate Editor and a 16. Fellstr€
om BC, Barratt J, Cook H, et al. Targeted-release
Deputy Editor. Accepted in revised form October 18, 2017. budesonide versus placebo in patients with IgA nephrop-
Publication Information: © 2017 by the National Kidney Founda- athy (NEFIGAN): a double-blind, randomised, placebo-
tion, Inc. Originally published online December 1, 2017 with doi controlled phase 2b trial. Lancet. 2017;389(10084):2117-
10.1053/j.ajkd.2017.10.004 2127.

162 AJKD Vol 71 | Iss 2 | February 2018