Review

Management of steroid-resistant nephrotic syndrome in children and adolescents

Kjell Tullus, Hazel Webb, Arvind Bagga

More than 85% of children and adolescents (majority between 1–12 years old) with idiopathic nephrotic syndrome Lancet Child Adolesc Health 2018 show complete remission of proteinuria following daily treatment with corticosteroids. Patients who do not show Published Online remission after 4 weeks’ treatment with daily prednisolone are considered to have steroid-resistant nephrotic October 17, 2018 syndrome (SRNS). Renal histology in most patients shows presence of focal segmental glomerulosclerosis, minimal http://dx.doi.org/10.1016/ S2352-4642(18)30283-9 change disease, and (rarely) mesangioproliferative glomerulonephritis. A third of patients with SRNS show mutations Nephrology Unit, Great in one of the key podocyte genes. The remaining cases of SRNS are probably caused by an undefined circulating Ormond Street Hospital for factor. Treatment with calcineurin inhibitors (ciclosporin and tacrolimus) is the standard of care for patients with Children, Great Ormond Street, non-genetic SRNS, and approximately 70% of patients achieve a complete or partial remission and show satisfactory London, UK (K Tullus MD, long-term outcome. Additional treatment with drugs that inhibit the renin–angiotensin axis is recommended for H Webb BSc) andDivision of Nephrology, Indian Council of hypertension and for reducing remaining proteinuria. Patients with SRNS who do not respond to treatment with Medical Research Advanced calcineurin inhibitors or other immunosuppressive drugs can show declining kidney function and are at risk for end- Center for Research in stage renal failure. Approximately a third of those who undergo renal transplantation show recurrent focal segmental Nephrology, All India Institute glomerulosclerosis in the allograft and often respond to combined treatment with plasma exchange, rituximab, and of Medical Sciences, New Delhi, India (Prof A Bagga MD) intensified immunosuppression. Correspondence to: Dr Kjell Tullus, Nephrology Unit, Introduction prednisolone at a dose of 60 mg/m² daily for 4 weeks. Great Ormond Street Hospital for Nephrotic syndrome is diagnosed based on a triad of Others recommend treatment for 8 weeks.2–4 Several Children, Great Ormond Street, symptoms: severe proteinuria (>1 g/m² per day), centres, including the Great Ormond Street Hospital for London WC1N 3JH, UK [email protected] hypoalbuminaemia (albumin <2·5 g/dL), and oedema. Children, London, UK, administer three intra­venous Most children older than 1 year (generally until 12 years pulses of methylprednisolone (500 mg/m²) before old) are diagnosed with idiopathic nephrotic syndrome regarding patients as resistant.5 Steroid resistance most based on these features and little else. A secondary cause often occurs during initial treatment with prednisolone (eg, systemic lupus erythematosus and Henoch- (initial resistance), but can also occur during treatment Schönlein purpura) is rare. for a relapse, in a patient who had previously responded Idiopathic nephrotic syndrome presents as an acute to treatment with steroids or with a second-line drug disease, with substantial proteinuria and increasing (late resistance). Steroid resistance is an important deter­ oedema. A smaller proportion of patients have a slower minant of future risk for end-stage renal disease. and more atypical onset, sometimes over several months or even years. Atypical features include onset in infancy or adolescence, symptoms suggestive of an inflam­ Key messages matory kidney disease, renal failure, hypertension, and • About 10–15% of children with idiopathic nephrotic syndrome who do not show macroscopic haematuria. complete remission of proteinuria following 4 weeks’ treatment with corticosteroids More than 85% of patients with idiopathic nephrotic are considered to have steroid-resistant nephrotic syndrome syndrome respond (ie, complete remission of proteinuria • For 30% of patients with steroid-resistant nephrotic syndrome, the condition results and normal serum albumin) following treatment with from a genetic cause; for the remainder, the disease is probably caused by a circulating 1 prednisolone. Response to prednisolone is an important factor prognostic indicator for survival of kidney function. • Renal histology shows minimal change disease and focal segmental glomerulosclerosis Although many patients with steroid-sensitive nephrotic (FSGS) in most patients syndrome have frequent relapses or steroid dependence, • Approximately 50–70% of patients with steroid resistance show complete or partial the long-term outlook for kidney function is favourable. remission following treatment with a calcineurin inhibitor (either ciclosporin or The main long-term problem in these patients is the risk tacrolimus) of side-effects from prolonged treatment with cortico­ • Although additional treatment with mycophenolate mofetil or rituximab is steroids and other immuno­suppressive medications. considered in children who are resistant to treatment with steroids and calcineurin Patients who do not respond to prednisolone are inhibitors, their efficacy to induce remission is low considered to have steroid-resistant nephrotic syndrome • Patients with a genetic FSGS or those who do not show complete or partial remission (SRNS). The medical community has not yet reached a following treatment with calcineurin inhibitors have high risk of end-stage renal failure consensus regarding the length of time prednisolone • Patients with late resistance and absence of a genetic cause show high risk of recurrent should be given before regarding a patient as steroid FSGS in the renal allograft resistant. We recommend that steroid resistance should • Intensification of treatment with ciclosporin or tacrolimus, and combined treatment be considered in patients who do not show complete with rituximab and plasma exchange might prevent or treat recurrent FSGS effectively remission of proteinuria despite treatment with www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 1 Review

Congenital nephrotic syndrome refers to infants with slit diaphragm (NPHS1, NPHS2, CD2AP, and PLCE1); onset of nephrotic syndrome before 3 months of age. cytoskeleton (ACTN4, MHY9, MYO1E, INF2, and actin Most of these patients have mutations involving the regulatory genes); glomerular basement membrane and podocyte proteins (nephrin, podocin, and WT1) and do matrix proteins (LAMB2, ITGA3, and COL4A3–5); not respond to treatment with steroids and other mitochondrial proteins (COQ2, COQ6, and ADCK4); medications. The course of illness is progressive and nuclear proteins (WT1, LMX1B, NUP93, NUP107, most patients require renal replacement treatment in the NUP205, and SMARCAL1); and other intracellular first decade. We have not included the clinical manage­ proteins (TRPC6, SCARB2, APOL1, DGKE, CUBN, ment of these children in this Review. and GAPVD1) with a wide spectrum of illness. Nephrotic syndrome might also be associated with Cause of SRNS syndromic features with mutations in specific genes— Genetic causes eg, Denys-Drash syndrome and Frasier syndrome (WT1); The medical community has learnt much about the Pierson syndrome (LAMB2); nail-patella syndrome genetic causes for SRNS over the past decades. Mutations (LMX1B); Epstein syndrome, Sebastian syndrome, and in more than 70 genes encoding key podocyte proteins related illnesses (MYH9); MELAS syndrome and Leigh are recognised to cause the illness. Despite recognition syndrome (mitochondrial genes); Galloway-Mowat syn­ of an increasing number of genetic causes, only drome (WDR73); and Schimke dysplasia (SMARCAL1), 30% of patients with sporadic SRNS show a defined as reviewed by Bierzynska and colleagues6 and Preston mutation.6 and colleagues.7 Genes associated with the occurrence of SRNS are broadly classified as involving: structural elements of the Circulating factor The probable cause for most patients with non-genetic 8,9 A SRNS is thought to be a circulating factor. Circum­ stantial evidence exists that makes this theory quite probable, but it has been elusive to define the circulating factors responsible for SRNS.10 A large proportion of children with non-genetic SRNS relapse quickly after kidney transplantation: these patients often respond to plasma exchange or immune adsorption. Small animals infused with patient plasma, whole or its fractions, also develop proteinuria. In 1975, a vascular permeability factor was described.11 Other important suggestions of circulating factors include haemopexin, interleukin-13, cardiotrophin-like cytokine-1, and soluble urokinase-type plasminogen activator receptor.12–15 None of these sug­ gested factors have been confirmed by independent research groups so far.

B Kidney biopsy Renal histology is an important tool for diagnostic and prognostic categorisation. Biopsies should be examined by light, immunofluorescence, and electron microscopy to define their histological features. An ade­quate biopsy should have approximately 25 glomeruli, especially when evaluating lesions that are focal or segmental. Biopsies with fewer glomeruli have low diagnostic accuracy. Common histological diagnoses include focal seg­ mental glomerulosclerosis (FSGS) in 35–55% of patients, minimal change disease in 25–40% of patients, and idiopathic mesangioproliferative glomerulonephritis in 10–15% of patients (figure 1).16 In about 20% of patients, the histology shows membranous nephropathy, Figure 1: Kidney biopsy images of patients with SRNS immunoglobulin A nephropathy, or proliferative glome­ (A) Glomerulus in minimal change disease appear normal on light microscopy. rulonephritis. We do not discuss their management in Tubules and interstitium are normal. Haematoxylin and eosin (H&E), magnification 40×. (B) Focal segmental glomerulosclerosis. Segmental sclerosis this Review. is noted in a perihilar location with hyalinosis. Evidence of tubular atrophy and Data from the 21-nation multiethnic PodoNet Registry interstitial fibrosis is also present. H&E, magnification 20 ×. have provided important information on the course and

2 www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 Review

outcome of a large cohort of children with SRNS.4 Clinical features were similar in patients with minimal Panel 1: Clinical tests for children with steroid-resistant nephrotic syndrome change disease, FSGS, and mesangioproliferative glome­ Main tests rulonephritis. Repeat biopsies, in 171 patients, found that • Blood creatinine, electrolytes, protein, albumin, transaminases the diagnosis changed from minimal change disease to • Full blood count FSGS in 26 of 47 (55%) cases, and from mesangio­ • 24 h urine protein excretion, spot urine protein to creatinine ratio (or albumin to proliferative glomerulonephritis to FSGS in 16 of 33 creatinine ratio) (48%) cases.4 • Varicella antibody status, measles immunoglobulin G (if no history of the measles, Patients with SRNS mostly require treatment with mumps, and rubella vaccination) calcineurin inhibitors that can cause nephrotoxicity, which is characterised histologically by microvascular Additional tests ischaemia and striped interstitial fibrosis.17 • Streptococcal antibodies, complement C3, C4 (if suspected lupus nephritis) • Antinuclear antibody, anti-double stranded DNA antibody (if suspected lupus Evaluation nephritis) Treatment of SRNS should start with a re-evaluation of • Hepatitis C and B serology, HIV antibody the diagnosis. Does the child have idiopathic nephrotic • Kidney biopsy: light, immunofluorescence, and electron microscopy syndrome with a probable biopsy diagnosis of minimal Genetic studies (targeted exome sequencing) indications change disease or FSGS, or are there atypical features? • Congenital (onset<3 months of age), infantile-onset nephrotic syndrome, family Evaluation should include a detailed investigation of history of steroid resistance, suspected syndromic forms, resistance to calcineurin current metabolic status and an underlying cause inhibitors, and before transplantation (eg, proliferative glomerulonephritis; panel 1). Patients require regular monitoring for severity of proteinuria and blood concentrations of creatinine and albumin. Child with typical nephrotic syndrome: Experts recommend a renal biopsy in all patients to treatment with prednisolone for 4 weeks counsel parents and guide treatment. Screening for variations in known genes for SRNS is important, especially in infants with initial resistance, familial or Remission achieved No response: syndromic FSGS, non-response to treatment with start CNI—aim to treat for calcineurin inhibitors, and before transplantation. 3–6 months and send genetic screen (for primary steroid resistant) Treatment Although the aim of treatment in SRNS is the Manage as steroid-sensitive Response to CNI: complete No response to CNI achievement of complete remission, the occurrence of nephrotic syndrome remission even partial remission is satisfactory.18,19 Complete remission is defined as presence of trace or negative proteinuria (by dipstick) or spot urine protein to Genetic screen negative: Genetic screen positive: consider creatinine ratio less than 0·2. Patients are considered to consider adding in MMF or stopping immunosuppression be in partial remission if they show proteinuria level of rituximab 1–2+ (urine protein:urine creatinine from 0·2–2), serum albumin more than 2·5 g/dL, and no oedema. Persistence Remission achieved: No response to increased High chance of development of of 3–4+ proteinuria (urine protein:urine creatinine more decreased chance of develop- immunosuppression ESRD: consider supportive than 2), albumin less than 2·5 g/dL, or oedema ment of ESRD and high chance measures constitutes non-response. of relapse Calcineurin inhibitors (ciclosporin and tacrolimus) with low-dose prednisolone are recommended as the Figure 2: Flow chart schematically outlining clinical management for a child with steroid-resistant nephrotic syndrome main treatment for children with SRNS (figure 2). CNI=calcineurin inhibitor. MMF=mycophenolate mofetil. ESRD=end-stage renal disease. Evidence-based role for treatment with other drugs, including cyclophosphamide, mycophenolate­ mofetil, and rituximab, is scarce. The optimal duration of occurrence of partial remission following treatment, treatment with calcineurin inhibitors is not clear. most experts do not recommend that patients with a Guidelines from Kidney Disease Improving Global confirmed mutation in podocyte genes receive Outcomes recommend a minimum duration of immunosuppressive medi­cations.19 12 months;2 in practice, treatment is often continued for 24–36 months. Calcineurin inhibitors The question whether children with a proven genetic Although the immunosuppressive effect of calcineurin cause should receive immunosuppressive treatment is inhibitors is through inhibition of T lymphocytes, a being debated. Although anecdotal reports suggest direct effect of ciclosporin on the podocyte actin www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 3 Review

Dose Monitoring Side-effects Tacrolimus 0·05–0·1 mg/kg in two divided doses Drug concentrations, creatinine Nephrotoxicity, hypertension, hyperglycaemia, orally; trough 4–8 ng/mL electrolytes, glucose, seizures, diarrhoea, hypomagnesemia transaminases Ciclosporin 4–6 mg/kg per day in two divided doses Drug concentrations, creatinine Nephrotoxicity, hypertension, hypertrichosis, orally; trough 80–150 ng/mL electrolytes, glucose, gingival hyperplasia transaminases Mycophenolate 600 mg/m² in two divided doses orally; Area under the curve Abdominal pain, diarrhoea, anorexia. Leucopenia mofetil* maximum dose 1200 mg/m² measurements, full blood counts, and raised transaminases are uncommon creatinine, transaminases Cyclophosphamide† 500–750 mg/m² intravenously every month Blood counts Alopecia, marrow suppression, vomiting, for 6 months; coadministration with sodium haemorrhagic cystitis, risk of systemic infections. mercaptoethane sulphonate, and Long-term risks: cancer, infertility ondansetron Rituximab* 375–750 mg/m² (maximum 1000 mg) Hepatitis B screen, CD19 count, Monitoring of allergic reactions during infusion, intravenously; 2–4 infusions, 1–2 weeks apart full blood count immunosuppression, risk of infections Enalapril and 0·2–0·5 mg/kg per day for enalapril Creatinine, electrolytes, caution if Dry cough, hyperkalaemia, hypotension, anaemia, ramipril and 1·25–6 mg/m² per day for ramipril hypovolaemia or hypotension nephrotoxicity

Drugs options are combined with alternate day prednisolone: 1·5 mg/kg on alternate days for 2–4 weeks, 1·25 mg/kg in the subsequent 2–4 weeks, 1 mg/kg for 3 months, and 0·5–0·75 mg/kg for 9–12 months. *Scarce evidence for efficacy of these drugs. †Oral cyclophosphamide is not recommended for steroid resistance. Intravenous cyclophosphamide limited to centres where cost is a major factor.

Table 1: Available drug options for the treatment of steroid-resistant nephrotic syndrome

cyto­skeleton has been suggested.20 Several prospective of cytochrome P450, respectively. The main CYP3A studies have been done with calcineurin inhibitors for inhibitors are azoles (fluconazole, ketoconazole, and patients with SRNS. Treatment with ciclosporin has been voriconazole), calcium channel blockers (diltiazem shown to be effective in inducing complete or partial hydrochloride and verapamil hydrochloride), protease remission in placebo-controlled studies, and in com­ inhibitors (ritonavir, saquinavir, and indinavir sulphate), parison with intravenous cyclo­phosph­amide.21 A macrolides (clarithromycin and erythromycin), and grape­ randomised controlled trial of 131 children with SRNS fruit and pomegranate juices. Clinically effective inducers compared the efficacy of 12-month treatment with of the enzymatic pathway are rifampicin, rifabutin, tacrolimus with six intravenous pulses of cyclo­phosph­ phenytoin, phenobarbital, and some non-nucleoside amide.22 Complete remission was reported in 33 of 63 reverse transcriptase inhibitors (efavirenz and nevirapine). (52%) patients receiving tacrolimus and in 9 of 51 (18%) Therapeutic monitoring of calcineurin inhibitor concen­ patients receiving cyclophosphamide. Respective partial tration is necessary in patients receiving concomitant res­ponses were 19 of 63 (30%) and 19 of 61 (31%). treatment with these drugs. Another trial, funded by the US National Institutes of The most important side-effect of calcineurin inhibitors Health (NIH), showed a 46% combined complete and is the potential risk on long-term renal function. partial response with ciclosporin.18 Although the medical community has not reached a The probability of complete or partial response to consensus on the duration of treatment, most experts treatment with calcineurin inhibitors is similar in consider weaning off calcineurin inhibitors after patients with minimal change disease (46·5%), 2–3 years.2,27 The risk of relapse on stopping treatment is mesangio­­­­proliferative glomerulonephritis (38·0%), and high, but some children can be weaned off the medication FSGS (39·0%).4,19 Although most patients with SRNS and maintain remission. Treatment with a calcineurin who respond to treatment stay in remission, relapses are inhibitor increases the risk of acute kidney injury, not uncommon. Most of these relapses respond to particularly if the child is depleted intravascularly. addition of daily corticosteroid treatment (table 1). Temporary discontinuation of treatment could be Therapies with tacrolimus and ciclosporin show similar considered during such episodes. Cosmetic side-effects, efficacy.23 Treatment with tacrolimus might be preferred including hypertrichosis and gingival hyperplasia, because of lower risk of relapses and fewer side-effects.24 are common with ciclosporin. Glucose intolerance and 12 hour trough levels are used for monitoring treatment diarrhoea are potential risks with tacrolimus. Both with ciclosporin and tacrolimus; target levels vary between medications have a small risk of neurotoxicity, mani­ 80–150 ng/mL for ciclosporin and 4–8 ng/mL for festing with headache and occasional seizures. tacrolimus.24–26 Calcineurin inhibitors are metabolised by cytochrome P450 enzymes (CYP3A). Therefore, their Mycophenolate mofetil concentrations are affected by inhibitors or inducers of Mycophenolate mofetil, which works by selective in­ this pathway, resulting in increased or reduced amounts hibition of DNA replication in T and B lymphocytes, is

4 www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 Review

considered less effective than calcineurin inhibitors.23,26 especially in patients with minimal change disease.19 A However, an NIH trial in children and adults (<40 years of review of data from multiple case series shows that age) with steroid-resistant FSGS did not show a statistically treatment with intravenous cyclophosphamide results in significant difference in treatment response between the complete and partial remission in 30–35% of patients.22,27 combination of oral dexamethasone and mycophenolate Cyclophosphamide can result in severe side-effects, mofetil compared with ciclosporin.18 A report by Sinha including haemorrhagic cystitis, bone marrow depres­ and colleagues26 showed that an early switch (at 6 months) sion with leucopenia, and alopecia. The risk of azo­ from tacrolimus to mycophenolate mofetil was not ospermia is higher in pubertal than pre-pubertal boys effective in sustaining remission in patients with SRNS. and is related to cumulative exposure to the drug; the risk Studies in paediatric transplant recipients suggest the of female infertility is lower than for male patients.36 The importance of measuring area under the curve dose of intravenous cyclophosphamide used for SRNS is (mycophenolic acid concentration over time) for considerably lower than that associated with infertility. monitoring treatment with mycophenolate mofetil.28 The proposed area under the curve measurement for patients Rituximab with steroid-sensitive nephrotic syndrome is between Rituximab, a monoclonal anti-CD20 antibody, is standard 30 mg/h per litre and 60 mg/h per litre.29 Information on treatment for children with frequently relapsing or target blood concentrations of mycophenolate­ mofetil for steroid-dependent nephrotic syndrome, particularly patients with SRNS is scarce, and possibly targeting when other immunosuppressive drugs have not achieved higher concentrations than those currently in clinical longstanding full remission.37–39 Apart from action on practice might be associated with better response to lymphocytes, rituximab might have off-target effects mycophenolate mofetil treatment. The side-effect profile on lipid metabolism in podocytes by binding to of mycophenolate mofetil makes it preferable to sphingomyelin phosphodiesterase acid-like 3b protein calcineurin inhibitors, since mycophenolate mofetil does and regulating acid sphingomyelinase activity. The first not have adverse effects on renal function. Gastro­ children with SRNS to be treated with rituximab were intestinal disturbances are important; leucopenia and reported by Bagga and colleagues,37 in 2007. Three patients other adverse effects are less common. had a complete response and two had a partial response. Patients who do not show remission following treat­ Further case series confirmed these findings, but the ment with calcineurin inhibitors pose a great challenge overall response to treatment was less impressive.39,40 An for management of SRNS. Few reports suggest that the international cohort of 27 children showed response in combined use of calcineurin inhibitors, mycophenolate 18 patients (67%), but only six patients had complete mofetil, and prednisone might be effective in inducing response.41 Similar figures were found in other reports, partial or complete remission in a small proportion with complete remission in 27% and 29% of patients.25,42 of these patients.30,31 However, the intense immuno­ The first, and to our knowledge only, randomised, suppression associated with combined treatment could controlled study in children with SRNS included predispose patients to serious infectious complications. 31 patients who were non-responsive to calcineurin inhibitors and steroids. All patients continued pred­ Cyclophosphamide nisolone and the calcineurin inhibitors, whereas 16 also Cyclophosphamide, an alkylating agent, has been widely received two doses (375 mg/m²) of rituximab. No used in patients with steroid-sensitive nephrotic syn­ statistically significant reduction of proteinuria was drome. Early studies in SRNS did not show benefit with measured at 3 months.43 Despite limitations, many use of oral cyclophosphamide and steroids com­pared paediatric nephrology centres use intravenous rituximab with treatment with steroids only.32,33 Therefore, oral in selected patients of steroid and calcineurin inhibitors- cyclo­phosphamide is not recommended for manage­ resistant nephrotic syndrome. The dose and frequency ment of SRNS.2,26 of administration is not defined. A range of doses An aggressive regime, proposed by Tune and Mendoza,34 from 375 mg/m² (single dose) to 750 mg/m² per dose combined multiple pulses of intravenous methylpredni­ (two doses given 7–14 days apart) are used.44 solone (30 mg/kg per dose, initially given every other day, Rituximab is well tolerated but might cause infusion and then tapered to weekly and monthly dosing), oral reactions of varying severity.45 These reactions can be prednisone (2 mg/kg every other day), and oral cyclo­ controlled by premedication with steroids and anti­ phosphamide.­ The efficacy of this protocol was variable, histamines, and reduced rate of infusion. Bone marrow with remission varying from 20–50%. These regimens suppression with neutropenia has been described.46 have substantial adverse events and are not recommended. The risks for prolonged hypogammaglobulinaemia­ are Intravenous cyclophosphamide is more effective than controversial and, although monitoring blood concen­ oral treatment,35 and due to its low cost is still used in tration of immunoglobulins has been suggested, most resource-limited countries. Treatment involves six-dose centres do not administer intravenous immunoglobulins, treatment with intravenous cyclophosphamide (once a even if they find low immunoglobulin concentrations in month) and tapering alternate day prednisolone, the blood.47 www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 5 Review

or prevent glomerular fibrosis (inhibitors of transforming Trial Study design and indication growth factor β, fresoli­mumab, microRNA, pirfenidone, NCT02592798 Safety and efficacy of abatacept (every 28 days Phase 2 placebo-controlled; minimal rosiglitazone, pioglitazone­ hydrochloride, and paclitaxel).54 for four cycles) in steroid-resistant and change disease, FSGS calcineurin-resistant illness (adults and Some of these strategies are being examined children 6–17 years) prospectively in phase 1–4 studies (table 2). However, NCT02394106 Ofatumumab in steroid-resistant and Phase 2, double-blind, placebo-controlled; none of these drugs are currently recommended for calcineurin inhibitor-resistant nephrotic minimal change disease, FSGS, treatment of patients with SRNS. syndrome (children 2–18 years) mesangioproliferative glomerulonephritis NCT02382874 Adipose-derived mesenchymal stromal cells Phase 1 open label pilot study on safey Ofatumumab (children 2–14 years) and efficacy; multiresistant FSGS The first use of ofatumumab, a humanised anti-CD20 NCT00098020 Efficacy of 6 months’ therapy with isotretinoin Phase 2 open label pilot study on safey (children ≥16 years and adults) with resistant and efficacy; minimal change disease, monoclonal antibody, was reported in five children who nephrotic syndrome FSGS, HIV-associated disease were resistant to cyclophosphamide, mycophenolate mofetil, calcineurin inhibitors, and rituximab.55 The FSGS=focal segmental glomerulosclerosis. results were exceptionally positive with all patients going Table 2: Ongoing clinical trials of steroid-resistant nephrotic syndrome in children and adolescents into remission with normal blood concentration of albumin 6-weeks after treatment. Two other studies that reported on nine children have shown that four patients A small but increased risk of viral infections is present achieved complete remission, two had partial remission, after rituximab treatment. Cerebral infections with two did not respond, and one patient had a severe JC virus have been described in adult patients and also infusion reaction.56,57 reactivation of hepatitis B.48,49 A severe respiratory illness, rituximab-associated lung injury, develops in a very small Abatacept group of children.50 Infections with Pneumocystis jiroveci Abatacept (cytotoxic T lymphocyte-associated antigen have also been reported.51 Previous treatment can result 4-immunoglobulin fusion protein, CTLA-4-Ig) is a co- in antichimeric antibodies, which reduce the effect of stimulation inhibitor targeting B7–1 (CD80) on antigen- rituximab and might result in anaphylactic reactions. presenting cells, including podocytes. In 2013, treatment Rituximab-induced depletion of CD19 B cells is taken with this drug was reported to induce partial or total as a sign of effective treatment and usually lasts for remission in four patients with recurrent and one with 6–8 months.52 Data from most studies suggest that primary FSGS.58 The authors also showed that abatacept relapses of nephrotic syndrome are uncommon in stabilises β1-integrin activation in podocytes in vitro, patients with B lymphocyte depletion. Therapeutic reducing proteinuria in patients with B7–1-positive response does, however, last several months, even after glomerular disease. These findings have not been repopulation of B cells.42 Recent data also suggest that replicated in other patients treated with abatacept or with recovery of memory B cells might predict the occurrence belatacept, a drug that binds B7–1 with even higher of a relapse.53 affinity than abatacept.59,60 Although treatment with calcineurin inhibitors is effective in most patients with SRNS, the precise duration Adrenocorticotropic hormone of treatment is not defined. Also, a few patients continue to Administration of adrenocorticotropic hormone was one relapse while on treatment with calcineurin inhibitors and of the first therapies for nephrotic syndrome.61 Since show features of steroid or calcineurin inhibitor toxicity. clinical and experimental evidence suggested that Treatment with two doses of rituximab has been shown to adrenocorticotropic hormone has antiproteinuric, lipid- enable weaning off treatment with calcineurin inhibitors lowering, and renoprotective properties,62 the drug was and prednisolone. A report from Ito and colleagues39 reintroduced as treatment for nephrotic syndrome. showed that treatment with rituximab resulted in a Hogan and colleagues63 treated 24 adult patients with 71% reduction in relapse, with withdrawal of calcineurin steroid-resistant or steroid-dependent FSGS with adreno­ inhibitors in 95% and corticosteroids in 74% patients. corticotropic hormone and achieved remission in seven More information is required regarding the frequency of (29%) patients. Adrenocorticotropic hormone is proposed redosing and potential long-term toxicities. to have actions beyond those of corticosteroids via anti- inflammatory mechanisms, or directly on podocytes via Other treatments the melanocortin receptor.62,64 The treatment of steroid-resistant and calcineurin inhibitor-resistant nephrotic syndrome is challenging. Adalimumab Newer drugs are available that target immune-mediated Since tumour necrosis factor is upregulated in human inflammation and glomerular damage (ofatumumab, and experimental models of FSGS, monoclonal adrenocorticotropic hormone, abatacept, adalimumab, antibodies have been used to inhibit this pathway.65 In a and saquinavir); inhibit action of permeability factors phase 1 trial on novel therapies for resistant FSGS (galactose and inhibitors of soluble urokinase receptor); (FONT), adalimumab was well tolerated, and after

6 www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 Review

16 months of follow-up, four of ten patients showed of proteinuria by approximately 15%. However, patients stable kidney function and reduced proteinuria.66 Studies receiving dual treatment are at risk of hypotension, are required to confirm the benefits of adalimumab in hyperkalaemia, and impaired kidney function; therefore, slowing the progression of FSGS. the combination is not recommended.74 The DUET study,75 which concluded in 2017, showed that sparsentan, Lipoprotein apheresis a novel drug with dual blockade of the angiotensin II 11 children with steroid-resistant and ciclosporin- receptor and endothelin 1, was more efficient than resistant nephrotic syndrome were treated with irbesartan in decreasing proteinuria in SRNS. lipoprotein apheresis three times per week, for 3 weeks, and then once a week for 6 weeks. Prednisone was given Infection prophylaxis at a dose of 1 mg/kg per day during the last 6 weeks and Children with SRNS have increased risk of infections, then tapered.67 Five of these children had complete due to urinary loss of immunoglobulins and effect of remission and two had partial remission. Those children immunosuppressive treatment. An important compli­ with complete response showed normal kidney function cation is spontaneous bacterial peritonitis, often caused during a median follow-up of 4·4 (range 4·0–11·1) years, by Streptococcus pneumoniae or by gram-negative bacteria. whereas patients with partial and no response progressed Although few paediatric nephrology centres advocate the to end-stage renal failure. Further studies are ongoing to use of prophylactic oral penicillin, most do not follow determine the efficacy of this treatment. this practice.68 All patients with SRNS should receive the pneumococcal vaccine. Symptomatic treatment Viral infections, especially varicella zoster virus, can be A comprehensive review on symptomatic treatments for a serious problem in immunocompromised patients.68 If children with SNRS was published in 2016.68 a child has substantial exposure to chickenpox (within the preceding 48 h of appearance of rash), immunity Oedema against varicella should be checked. If not immune, Oedema is a cardinal feature of nephrotic syndrome and varicella zoster immunoglobulin is given as soon as often a debilitating symptom. A bodyweight increase by possible, but up to 10 days after exposure. Confirmed 25% from the baseline (before treatment) is not unusual chickenpox while on prednisolone or immunosuppressive in children with nephrotic syndrome. However, a treatment, or within 3 months of stopping, should be relatively large group of children show only mild oedema, treated with acyclovir for 7 days. Intravenous treatment is despite having chronically low concentrations of serum considered in all patients, especially if unwell, and albumin, sometimes below 10 g/L. usually for at least the first 48 h of treatment. Initial treatment consists of restriction of intake of sodium and of fluids to two-thirds of maintenance.69 Thrombosis Careful use of diuretics including furosemide, meto­ Children with ongoing nephrotic syndrome have an lazone, and spironolactone is helpful.70 Many children with increased risk of both venous and arterial thrombosis. active nephrotic syndrome have a loss of fluid from the Prophylactic aspirin is advised, but is not widely used.68 vascular compartment to the extracellular space and have Physicians should have a high index of suspicion for low intravascular volume. Aggressive diuretic treatment in thrombosis. Prompt evaluation and treatment with these patients carries a risk of hypovolaemia and acute heparin followed by oral anticoagulants is recommended. kidney injury.71 Despite conservative management, a small group of children continue to have substantial oedema Hyperlipidaemia resulting in discomfort. These children need repeated Patients with SRNS often show markedly raised blood infusions of human albumin together with intravenous concentrations of cholesterol and triglycerides.76 The furosemide. abnormal lipid profile might contribute to later cardiovascular morbidity. Adult patients with nephrotic Residual proteinuria syndrome are treated with statins, with satisfactory Children with SRNS who show partial or no response to improvement of lipid concentrations. Treatment with calcineurin inhibitors continue to show proteinuria. statins is not routinely used in children with SRNS and Studies in adults, as well as children, show that treatment prospective studies are required.77 with angiotensin-converting enzyme (ACE) inhibitors reduces the degree of proteinuria by 30–40% in a Vitamin D dose-dependent and time-dependent manner and, Children with SRNS have urinary losses of thus, is recommended.72,73 Angiotensin receptor blockers vitamin D-binding protein, which might lead to reduced (valsartan and irbesartan) are advised for patients bone density and increased risk of fractures. A study on showing adverse effects to ACE inhibitors, mainly cough. supplementation with vitamin D and calcium did not Combined treatment with an ACE inhibitor and show any benefits on bone mineral density, but did show angiotensin receptor blocker results in further reduction an increase in the risk of hypercalciuria.78 www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 7 Review

Results from various case series show that almost Panel 2: Prevention and management of recurrent focal segmental 30–50% patients with idiopathic SRNS with end-stage glomerulosclerosis after kidney transplantation kidney disease who receive an allograft can develop Prevention of recurrence recurrent FSGS. Patients with recurrent FSGS are at risk • Ensure period of normalised serum albumin before transplantation of delayed graft function and graft loss, approaching • Genetic screening of recipient if living-related donor 30–50% at 5 years.80 Although the pathogenesis is • Intravenous rituximab (375 mg/m²; one or two doses) 2 weeks before transplantation unclear, disease recurrence is attributed to circulating • Plasma exchanges (begin 7–10 days before transplantation) permeability factors. Recurrent proteinuria occurs from • Induction with intravenous methylprednisolone and basiliximab; hours to days after the transplant, and is associated with triple immunosuppression progressive hypoalbuminaemia and effacement of • Post-transplantation proteinuria screen: daily for 1 week; weekly for 4 weeks; podocyte foot process under electron microscopy. every 3 months for first year; and every 6 months thereafter Features associated with increased risk of recurrent FSGS include onset of disease before 15 years of age, Treatments for recurrence non-genetic (immune) form of the disease, mesangial • Plasma exchange (8–12 sessions for response; might require long-term for maintenance proliferation on renal biopsy, and rapid progression to of remission; 1–1·5 volume exchanges) on alternate days for 7–10 sessions, then taper end-stage kidney disease.81,82 and continue until proteinuria subsides A report on 150 patients with nephrotic syndrome who • Methylprednisolone (10–15 mg/kg intravenously, three doses daily) underwent renal transplantation showed that the risk of • Switch to high-dose ciclosporin, targeting trough concentrations of 250–300 ng/mL recurrence was highest (26 of 28, 93%) in patients who for 2–3 weeks; alternatively, continue tacrolimus at higher dose were initially corticosteroid responsive, but showed late

• Rituximab(1–6 doses used; early use and normal serum albumin at onset predict resistance; intermediate (26 of 86, 30%) in those with response): 375 mg/m², two doses 1 week apart; avoid plasma exchange for 36 h initial resistance, but no known mutations; and none in following treatment; monitor CD19 depletion those with mutations affecting key podocyte genes.83 A • Other treatments (considered for refractory patients; variable efficacy): low-density panel of antibodies against seven glomerular antigens lipoprotein apheresis, and oral cyclophosphamide (8–12 weeks) to replace (CD40, PTPRO, CGB5, FAS, P2RY11, SNRPB2, and mycophenolate mofetil. APOL2) is reported to predict post-transplant FSGS recurrence with 92% accuracy; findings of this report need confirmation.84 Thyroid disease Children with SRNS, especially congenital nephrotic Treatment recommendations syndrome, lose large amounts of thyroid-binding protein All patients with FSGS who are about to undergo renal in urine. Therefore, the amount of thyroid hormones in transplantation should be advised about the possibility of such patients should be monitored.68 recurrence of disease. Combined treatment with 1–2 doses of rituximab and pretransplant plasmapheresis Prognosis is likely to benefit recipients at high risk of recurrent Children with SRNS are at risk of progressing to end- FSGS. Since rituximab is cleared from the body by stage kidney disease. One study showed renal survival at plasma exchange, an interval of 36–48 h between rituxi­ 10 years to vary between 54·6% and 100%, depending on mab infusion and the exchange is recommended.85 renal histology and treatment. Important influencing Strategies for management of children with docu­ factors for unsatisfactory outcome were a proven genetic mented recurrent disease are not based on randomised cause, renal histology, response to immunosuppressive trials, but on observational reports and clinical expertise. treat­ment, and duration of follow-up.79 Data from a large Typically plasma exchange or immunoadsorption in multicentre registry showed that 10-year survival free of combination with high-dose ciclosporin (trough concen­ end-stage kidney disease was 29% in patients with a trations 250–300 ng/mL for 3 weeks) and cyclophos­ genetic diagnosis, 52% in those with FSGS, and 79% in phamide (2–2·5 mg/kg per day for 3 months) is used. minimal change disease.19 The 10-year renal survival With these strategies, 70% of children with recurrent in the group that responded to immunosuppression FSGS achieve complete or partial remission.86,87 Experts was 94%, compared with 72% with partial remission, and have reported remission of proteinuria in 64% patients 43% with non-response to treatment.19 with the use of rituximab (2–6 doses of 375 mg/m², administered once every 1–2 weeks), in conjunction with Recurrence in transplantation plasma exchange and post-transplantation­ immuno­ Several paediatric nephrology centres advocate nephrect­ suppression.88 The median time to best clinical response omy before accepting the child for transplantation.­ This was 2 months (range 0·6–12·0 months). On stepwise procedure is used to reduce the problems related to regression, normal serum albumin concentration at continuous proteinuria, which can make the monitoring recurrence (suggesting mild FSGS) and young age of any recurrence of the disease in the transplanted predict a favourable response to treatment.88 In a review kidney easier. on patients with recurrent FSGS treated with rituximab

8 www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 Review

5 Deschenes G, Vivarelli M, Peruzzi L. Variability of diagnostic Search strategy and selection criteria criteria and treatment of idiopathic nephrotic syndrome across European countries. Eur J Pediatr 2017; 176: 647–54. We did multiple searches on PubMed and Cochrane over the 6 Bierzynska A, McCarthy HJ, Soderquest K, et al. Genomic and past 10 years for studies published until Jan 1, 2018, with clinical profiling of a national nephrotic syndrome cohort advocates a precision medicine approach to disease management. different combinations of the search terms “nephrotic Kidney Int 2017; 91: 937–47. syndrome”, “steroid resistant treatment”, “calcineurin 7 Preston R, Stuart HM, Lennon R. Genetic testing in inhibitor”, “ciclosporin”, “tacrolimus”, “mycophenolate steroid-resistant nephrotic syndrome: why, who, when and how? Pediatr Nephrol 2017; published online Nov 27. DOI:10.1007/s00467- mofetil”, “rituximab”, “outcome”, “prognosis”, “kidney 017-3838-6 (preprint). transplantation”, and “relapse after transplantation”. Studies 8 Dogra S, Kaskel F. Steroid-resistant nephrotic syndrome: were chosen based on the quality and amount of information. a persistent challenge for pediatric nephrology. Pediatr Nephrol 2017; 32: 965–74. Only studies published in Eglish were used. 9 Davin JC. The glomerular permeability factors in idiopathic nephrotic syndrome. Pediatr Nephrol 2016; 31: 207–15. 10 Savin VJ, Sharma M, Zhou J, et al. Multiple targets for novel infusions and plasma exchange, the 12 patients who therapy of FSGS associated with circulating permeability factor. achieved remission had received the infusions statistically Biomed Res Int 2017; 2017: 6232616. 11 Lagrue G, Branellec A, Blanc C, et al. A vascular permeability factor significantly earlier than the 13 patients who did not in lymphocyte culture supernants from patients with nephrotic respond (mean 100·5 days [SD 95·4] from onset of syndrome. II. Pharmacological and physicochemical properties. Biomedicine 1975; 23: 73–75. recurrence vs 468·1 days [379·8]).89 A systematic review of 12 Konigshausen E, Sellin L. Circulating permeability factors in 423 patients with recurrent FSGS showed that patients primary focal segmental glomerulosclerosis: a review of proposed treated with prompt plasma exchanges within 2 weeks of candidates. Biomed Res Int 2016; 2016: 3765608. recurrence show likelihood of remission.86 The role of 13 Lennon R, Singh A, Welsh GI, et al. Hemopexin induces nephrin-dependent reorganization of the actin cytoskeleton in lipid apheresis in achieving remission in patients with podocytes. J Am Soc Nephrol 2008; 19: 2140–49. recurrent FSGS is under investigation. Guidelines for 14 Wei C, Trachtman H, Li J, et al. Circulating suPAR in two cohorts of prevention and management of recurrent FSGS in primary FSGS. J Am Soc Nephrol 2012; 23: 2051–59. 15 Sinha A, Bajpai J, Saini S, et al. Serum-soluble urokinase receptor allograft recipients are summarised in panel 2. levels do not distinguish focal segmental glomerulosclerosis from other causes of nephrotic syndrome in children. Kidney Int 2014; Future directions in research 85: 649–58. 16 Nammalwar BR, Vijayakumar M, Prahlad N. Experience of renal The biggest challenge, as we see it, is to define the biopsy in children with nephrotic syndrome. Pediatr Nephrol 2006; presumed circulating factors that cause the disease in 21: 286–88. most patients with SRNS. The nature of this factor has 17 Sinha A, Sharma A, Mehta A, et al. Calcineurin inhibitor induced nephrotoxicity in steroid resistant nephrotic syndrome. been investigated for several decades without success. Indian J Nephrol 2013; 23: 41–46. The precise effect of this circulating factor on podocyte 18 Gipson DS, Trachtman H, Kaskel FJ, et al. Clinical trial of focal biology also needs to be clarified. We suspect that the segmental glomerulosclerosis in children and young adults. Kidney Int 2011; 80: 868–78. circulating factor is in some way related to the immune 19 Trautmann A, Schnaidt S, Lipska-Zietkiewicz BS, et al. system, since all currently effective treatments are with Long-term outcome of steroid-resistant nephrotic syndrome in immunosuppressive medications. children. J Am Soc Nephrol 2017; 28: 3055–65. 20 Shen X, Jiang H, Ying M, et al. Calcineurin inhibitors ciclosporin A The number of genes causing SRNS will undoubtedly and tacrolimus protect against podocyte injury induced by puromycin continue to rise until all have been found. Patients with aminonucleoside in rodent models. Sci Rep 2016; 6: 32087. SRNS who have an underlying genetic disorder are 21 Hodson EM, Wong SC, Willis NS, Craig JC. Interventions for idiopathic steroid-resistant nephrotic syndrome in children. unlikely to respond to medications and might be future Cochrane Database Syst Rev 2016; 10: CD003594. candidates for genetic therapy, including genomic 22 Gulati A, Sinha A, Gupta A, et al. Treatment with tacrolimus and editing. prednisolone is preferable to intravenous cyclophosphamide as the initial therapy for children with steroid-resistant nephrotic Contributors syndrome. Kidney Int 2012; 82: 1130–35. All authors wrote the first draft of one part of the Review. The draft was 23 Hodson EM, Willis NS, Craig JC. Interventions for idiopathic reviewed by all authors iteratively, until all three approved of the final steroid-resistant nephrotic syndrome in children. manuscript. Cochrane Database Syst Rev 2010; 10: CD003594. Declaration of interests 24 Choudhry S, Bagga A, Hari P, Sharma S, Kalaivani M, Dinda A. Efficacy and safety of tacrolimus versus ciclosporin in children with We declare no competing interests. steroid-resistant nephrotic syndrome: a randomized controlled trial. References Am J Kidney Dis 2009; 53: 760–69. 1 Lombel RM, Gipson DS, Hodson EM. Treatment of steroid-sensitive 25 Gulati A, Sinha A, Jordan SC, et al. Efficacy and safety of treatment nephrotic syndrome: new guidelines from KDIGO. with rituximab for difficult steroid-resistant and -dependent Pediatr Nephrol 2013; 28: 415–26. nephrotic syndrome: multicentric report. Clin J Am Soc Nephrol 2010; 2 Lombel RM, Hodson EM, Gipson DS. Treatment of steroid-resistant 5: 2207–12. nephrotic syndrome in children: new guidelines from KDIGO. 26 Sinha A, Gupta A, Kalaivani M, Hari P, Dinda AK, Bagga A. Pediatr Nephrol 2013; 28: 409–14. Mycophenolate mofetil is inferior to tacrolimus in sustaining 3 Nourbakhsh N, Mak RH. Steroid-resistant nephrotic syndrome: past remission in children with idiopathic steroid-resistant nephrotic and current perspectives. Pediatric Health Med Ther 2017; 8: 29–37. syndrome. Kidney Int 2017; 92: 248–57. 4 Trautmann A, Bodria M, Ozaltin F, et al. Spectrum of steroid-resistant 27 Gulati A, Bagga A, Gulati S, Mehta KP, Vijayakumar M. and congenital nephrotic syndrome in children: the PodoNet registry Management of steroid resistant nephrotic syndrome. cohort. Clin J Am Soc Nephrol 2015; 10: 592–600. Indian Pediatr 2009; 46: 35–47. www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 9 Review

28 Filler G, Alvarez-Elias AC, McIntyre C, Medeiros M. 51 Sato M, Ito S, Ogura M, et al. Atypical Pneumocystis jiroveci The compelling case for therapeutic drug monitoring of pneumonia with multiple nodular granulomas after rituximab for mycophenolate mofetil therapy. Pediatr Nephrol 2017; 32: 21–29. refractory nephrotic syndrome. Pediatr Nephrol 2013; 28: 145–49. 29 Gellermann J, Weber L, Pape L, Tonshoff B, Hoyer P, Querfeld U. 52 Colucci M, Carsetti R, Cascioli S, et al. B cell reconstitution after Mycophenolate mofetil versus ciclosporin A in children with rituximab treatment in idiopathic nephrotic syndrome. frequently relapsing nephrotic syndrome. J Am Soc Nephrol 2013; J Am Soc Nephrol 2016; 27: 1811–22. 24: 1689–97. 53 Colucci M, Carsetti R, Cascioli S, Serafinelli J, Emma F, Vivarelli M. 30 El-Reshaid K, El-Reshaid W, Madda J. Combination of B cell phenotype in pediatric idiopathic nephrotic syndrome. immunosuppressive agents in treatment of steroid-resistant Pediatr Nephrol 2018; published online Sept 28. DOI:10.1007/s00467- minimal change disease and primary focal segmental 018-4095-z (preprint). glomerulosclerosis. Ren Fail 2005; 27: 523–30. 54 Belingheri M, Moroni G, Messa P. Available and incoming 31 Wu B, Mao J, Shen H, et al. Triple immunosuppressive therapy in therapies for idiopathic focal and segmental glomerulosclerosis in steroid-resistant nephrotic syndrome children with tacrolimus adults. J Nephrol 2018; 31: 37–45. resistance or tacrolimus sensitivity but frequently relapsing. 55 Basu B. Ofatumumab for rituximab-resistant nephrotic syndrome. Nephrology Carlton 2015; 20: 18–24. N Engl J Med 2014; 370: 1268–70. 32 [No authors listed]. Prospective, controlled trial of 56 Bonanni A, Rossi R, Murtas C, Ghiggeri GM. Low-dose ofatumumab cyclophosphamide therapy in children with nephrotic syndrome. for rituximab-resistant nephrotic syndrome. BMJ Case Rep 2015; Report of the International Study of Kidney Disease in Children. 2015: bcr2015210208. Lancet 1974; 304: 423–27. 57 Wang CS, Liverman RS, Garro R, et al. Ofatumumab for the 33 Tarshish P, Tobin JN, Bernstein J, Edelmann CM. treatment of childhood nephrotic syndrome. Pediatr Nephrol 2017; Cyclophosphamide does not benefit patients with focal segmental 32: 835–41. glomerulosclerosis. A report of the International Study of Kidney 58 Yu CC, Fornoni A, Weins A, et al. Abatacept in B7–1-positive Disease in Children. Pediatr Nephrol 1996; 10: 590–93. proteinuric kidney disease. N Engl J Med 2013; 369: 2416–23. 34 Tune BM, Mendoza SA. Treatment of the idiopathic nephrotic 59 Delville M, Baye E, Durrbach A, et al. B7–1 blockade does not syndrome: regimens and outcomes in children and adults. improve post-transplant nephrotic syndrome caused by recurrent J Am Soc Nephrol 1997; 8: 824–32. FSGS. J Am Soc Nephrol 2016; 27: 2520–27. 35 Bajpai A, Bagga A, Hari P, Dinda A, Srivastava RN. Intravenous 60 Garin EH, Reiser J, Cara-Fuentes G, et al. Case series: CTLA4-IgG1 cyclophosphamide in steroid-resistant nephrotic syndrome. therapy in minimal change disease and focal segmental Pediatr Nephrol 2003; 18: 351–56. glomerulosclerosis. Pediatr Nephrol 2015; 30: 469–77. 36 Latta K, Von SC, Ehrich JH. A meta-analysis of cytotoxic treatment 61 Mateer FM, Weigand FA, Greenman L, Weber CJ Jr, Kunkel GA, for frequently relapsing nephrotic syndrome in children. Danowski TS. Clinical results and effects of protenuria in one Pediatr Nephrol 2001; 16: 271–82. hundred and six instances. AMA J Dis Child 1957; 93: 591–603. 37 Bagga A, Sinha A, Moudgil A. Rituximab in patients with the 62 Lieberman KV, Pavlova-Wolf A. Adrenocorticotropic hormone steroid-resistant nephrotic syndrome. N Engl J Med 2007; 356: 2751–52. therapy for the treatment of idiopathic nephrotic syndrome in 38 Benz K, Dotsch J, Rascher W, Stachel D. Change of the course of children and young adults: a systematic review of early clinical steroid-dependent nephrotic syndrome after rituximab therapy. studies with contemporary relevance. J Nephrol 2017; 30: 35–44. Pediatr Nephrol 2004; 19: 794–97. 63 Hogan J, Bomback AS, Metha K, et al. Treatment of idiopathic 39 Ito S, Kamei K, Ogura M, et al. Survey of rituximab treatment for FSGS with adrenocorticotropic hormone gel. childhood-onset refractory nephrotic syndrome. Clin J Am Soc Nephrol 2013; 8: 2072–81. Pediatr Nephrol 2013; 28: 257–64. 64 Gong R. Leveraging melanocortin pathways to treat glomerular 40 Nakagawa T, Shiratori A, Kawaba Y, Kanda K, Tanaka R. Efficacy of diseases. Adv Chronic Kidney Dis 2014; 21: 134–51. rituximab therapy against intractable steroid-resistant nephrotic 65 Trachtman H, Vento S, Herreshoff E, et al. Efficacy of galactose and syndrome. Pediatr Int 2016; 58: 1003–08. adalimumab in patients with resistant focal segmental 41 Prytula A, Iijima K, Kamei K, et al. Rituximab in refractory glomerulosclerosis: report of the font clinical trial group. nephrotic syndrome. Pediatr Nephrol 2010; 25: 461–68. BMC Nephrol 2015; 16: 111. 42 Sinha A, Bhatia D, Gulati A, et al. Efficacy and safety of rituximab in 66 Peyser A, Machardy N, Tarapore F, et al. Follow-up of phase I trial children with difficult-to-treat nephrotic syndrome. of adalimumab and rosiglitazone in FSGS: III. Report on the FONT Nephrol Dial Transplant 2015; 30: 96–106. study. BMC Nephrol 2010; 11: 2. 43 Magnasco A, Ravani P, Edefonti A, et al. Rituximab in children with 67 Hattori M, Chikamoto H, Akioka Y, et al. A combined low-density resistant idiopathic nephrotic syndrome. J Am Soc Nephrol 2012; lipoprotein apheresis and prednisone therapy for steroid-resistant 23: 1117–24. primary focal segmental glomerulosclerosis in children. 44 Leandro MJ, Edwards JC, Cambridge G, Ehrenstein MR, Am J Kidney Dis 2003; 42: 1121–30. Isenberg DA. An open study of B lymphocyte depletion in systemic 68 McCaffrey J, Lennon R, Webb NJ. The non-immunosuppressive lupus erythematosus. Arthritis Rheum 2002; 46: 2673–77. management of childhood nephrotic syndrome. 45 Kamei K, Ogura M, Sato M, Ito S, Ishikura K. Infusion reactions Pediatr Nephrol 2016; 31: 1383–402. associated with rituximab treatment for childhood-onset 69 Kapur G, Valentini RP, Imam AA, Mattoo TK. Treatment of severe complicated nephrotic syndrome. Pediatr Nephrol 2018; 33: 1013–18. edema in children with nephrotic syndrome with diuretics alone— 46 Kamei K, Takahashi M, Fuyama M, et al. Rituximab-associated a prospective study. Clin J Am Soc Nephrol 2009; 4: 907–13. agranulocytosis in children with refractory idiopathic nephrotic 70 Bockenhauer D. Over- or underfill: not all nephrotic states are syndrome: case series and review of literature. created equal. Pediatr Nephrol 2013; 28: 1153–56. Nephrol Dial Transplant 2015; 30: 91–96. 71 Rheault MN, Zhang L, Selewski DT, et al. AKI in children 47 Sellier-Leclerc AL, Belli E, Guerin V, Dorfmuller P, Deschenes G. hospitalized with nephrotic syndrome. Clin J Am Soc Nephrol 2015; Fulminant viral myocarditis after rituximab therapy in pediatric 10: 2110–18. nephrotic syndrome. Pediatr Nephrol 2013; 28: 1875–79. 72 Bagga A, Mudigoudar BD, Hari P, Vasudev V. Enalapril dosage in 48 Delbue S, Ferraresso M, Elia F, et al. Investigation of steroid-resistant nephrotic syndrome. Pediatr Nephrol 2004; polyomaviruses replication in pediatric patients with nephropathy 19: 45–50. receiving rituximab. J Med Virol 2012; 84: 1464–70. 73 Yi Z, Li Z, Wu XC, He QN, Dang XQ, He XJ. Effect of fosinopril in 49 Kronbichler A, Windpessl M, Pieringer H, Jayne DRW. children with steroid-resistant idiopathic nephrotic syndrome. Rituximab for immunologic renal disease: what the nephrologist Pediatr Nephrol 2006; 21: 967–72. needs to know. Autoimmun Rev 2017; 16: 633–43. 74 Makani H, Bangalore S, Desouza KA, Shah A, Messerli FH. 50 Bitzan M, Anselmo M, Carpineta L. Rituximab B-cell depleting Efficacy and safety of dual blockade of the renin-angiotensin antibody associated lung injury (RALI): a pediatric case and system: meta-analysis of randomised trials. BMJ 2013; 346: f360. systematic review of the literature. Pediatr Pulmonol 2009; 44: 922–34.

10 www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 Review

75 Komers R, Gipson DS, Nelson P, et al. Efficacy and safety of 84 Delville M, Sigdel TK, Wei C, et al. A circulating antibody panel for sparsentan compared with irbesartan in patients with primary focal pretransplant prediction of FSGS recurrence after kidney segmental glomerulosclerosis: randomized, controlled trial design transplantation. Sci Transl Med 2014; 6: 256ra136. (DUET). Kidney Int Rep 2017; 2: 654–64. 85 Puisset F, White-Koning M, Kamar N, et al. Population 76 Arneil GC. The nephrotic syndrome. Pediatr Clin North Am 1971; pharmacokinetics of rituximab with or without plasmapheresis in 18: 547–59. kidney patients with antibody-mediated disease. 77 Querfeld U, Kohl B, Fiehn W, et al. Probucol for treatment of Br J Clin Pharmacol 2013; 76: 734–40. hyperlipidemia in persistent childhood nephrotic syndrome. 86 Kashgary A, Sontrop JM, Li L, et al. The role of plasma exchange in Report of a prospective uncontrolled multicenter study. treating post-transplant focal segmental glomerulosclerosis: Pediatr Nephrol 1999; 13: 7–12. asystematic review and meta-analysis of 77 case-reports and 78 Banerjee S, Basu S, Sengupta J. Vitamin D in nephrotic syndrome case-series. BMC Nephrol 2016; 17: 104. remission: a case-control study. Pediatr Nephrol 2013; 28: 1983–89. 87 Ponticelli C. Recurrence of focal segmental glomerular sclerosis 79 Inaba A, Hamasaki Y, Ishikura K, et al. Long-term outcome of (FSGS) after renal transplantation. Nephrol Dial Transplant 2010; idiopathic steroid-resistant nephrotic syndrome in children. 25: 25–31. Pediatr Nephrol 2016; 31: 425–34. 88 Araya CE, Dharnidharka VR. The factors that may predict response 80 Leca N. Focal segmental glomerulosclerosis recurrence in the renal to rituximab therapy in recurrent focal segmental glomerulosclerosis: allograft. Adv Chronic Kidney Dis 2014; 21: 448–52. a systematic review. J Transplant 2011; 2011: 374213. 81 Saeed B, Mazloum H, Askar M. Spontaneous remission of 89 Sakai K, Takasu J, Nihei H, et al. Protocol biopsies for focal post-transplant recurrent focal and segmental glomerulosclerosis. segmental glomerulosclerosis treated with plasma exchange and Saudi J Kidney Dis Transpl 2011; 22: 1219–22. rituximab in a renal transplant patient. Clin Transplant 2010; 82 Vinai M, Waber P, Seikaly MG. Recurrence of focal segmental 24 (suppl 22): 60–65. glomerulosclerosis in renal allograft: an in-depth review. © 2018 Elsevier Ltd. All rights reserved. Pediatr Transplant 2010; 14: 314–25. 83 Ding WY, Koziell A, McCarthy HJ, et al. Initial steroid sensitivity in children with steroid-resistant nephrotic syndrome predicts post-transplant recurrence. J Am Soc Nephrol 2014; 25: 1342–48.

www.thelancet.com/child-adolescent Published online October 17, 2018 http://dx.doi.org/10.1016/S2352-4642(18)30283-9 11