Introduction

The Wilms’ tumor suppressor gene 1 (WT1) encodes a transcription factor involved in kidney and gonadal development [1]. Mutations in the WT1 gene result in the occurrence of diffuse mesangial sclerosis (DMS), focal segmental glomerulosclerosis (FSGS), or membranoproliferative glomerulonephritis [2], which may appear isolated or in combination with other features, such as in the Denys–Drash and Frasier syndromes [1]. WT1 mutations have been found in a small percentage of children with isolated or syndromal steroid-resistant nephrotic syndrome (SRNS) [3–5], but have not been described in patients with steroid-sensitive nephrotic syndrome (SSNS) [6].

It is commonly assumed that the nephrotic syndrome (NS) in patients with WT1 mutations is unresponsive to therapy [3]; persistence of the NS or of nephrotic-range proteinuria (PU) with progression to end-stage renal disease (within <5 years) has been typically found in all patients described thus far [7]. We here describe 3 children (2 with SRNS, 1 with nephrotic-range proteinuria) with focal segmental glomerulosclerosis (FSGS) on renal biopsy and WT1 mutations, who responded to intensified therapy consisting of oral cyclosporin A (CyA) in combination with intravenous and oral prednisone (Pred).

Patients and methods

Patients were retrospectively identified and data collected from medical records. In all cases, results of WT1 screening had become available after patients had shown a favorable response to therapy. The glomerular filtration rate (GFR) was calculated according to Schwartz et al. [8].

Patient 1

Current age 16.9 years, WT1 gene mutation: intron 9: 1228 +5 G>A, normal female phenotype, karyotype 46XX, normal genital development. Proteinuria (PU) was first diagnosed at the age of 1 year. At the age of 3 years, overt NS developed with a PU of 6 g/g (protein/creatinine in urine), serum albumin (s-Alb) 24 g/l, s-creatinine 0.24 mg/l. A kidney biopsy showed FSGS. Standard therapy with oral Pred (60 mg/m2/day for 4 weeks) was unsuccessful and the patient was switched to intensified therapy. After 1 year of therapy with CyA and Pred, s-Alb was normal, PU 0.5 to 1 g/g. Therapy was further augmented with an ACE inhibitor and an angiotensin receptor blocker (started at the age of 8 and 12 years, respectively). After discontinuing therapy at the age of 14 years and 4 months, the NS relapsed (PU 4.8 g/g, s-Alb 28 g/l); after resuming CyA/Pred, PU decreased and s-Alb normalised (Fig. 1a). A slow deterioration of the GFR has been observed since the age of 13 (Fig. 1b) with a total follow-up time of >15 years.

Fig. 1
figure 1

a Clinical course of patient 1. Dots indicate the urine protein/creatinine ratio (g/g), thin bars the time-averaged mean urine protein/creatinine ratio and thick bars the trough level of cyclosporin A (CyA) in serum. ACE angiotensin-converting enzyme inhibitor therapy, ARB angiotensin receptor blocker therapy. b Glomerular filtration rate (GFR) of patient 1. Dots indicate the calculated GFR; the line is a calculated mean GFR over time

Full size image

Patient 2

Current age 11.6 years, WT1 gene mutation: exon 9; 1147 T>C:383 Phe>Leu, male phenotype, karyotype 46 XY, left cryptorchidism necessitating orchidopexy. PU was diagnosed accidentally at the age of 3 years (2 g/g creatinine, s-Alb 45 g/l, s-creatinine 0.19 mg/l) and a renal biopsy was performed, which showed FSGS. Intensified therapy (CyA/Pred) was started, resulting in a decrease in PU to 0.5 g/g creatinine. An ACE inhibitor was started at the age of 3.5 years and an angiotensin receptor blocker at the age of 8.9 years. After discontinuing therapy at the age of 9 years, a relapse of PU was observed, with an increase from 0.7 g/g to 2.6 g/g. A second course of CyA/Pred reduced PU to 0.2–0.3 g/g creatinine (Fig. 2a). The patient has maintained a normal GFR thus far (Fig. 2b) with a total follow-up time of >8 years.

Fig. 2
figure 2

a Clinical course of patient 2. Dots indicate the urine protein/creatinine ratio (g/g), thin bars the time-averaged mean urine protein/creatinine ratio and thick bars the trough level of CyA in serum. b GFR of patient 2. Dots indicate calculated GFR; the line is a calculated mean GFR over time. ACE angiotensin-converting enzyme inhibitor therapy, ARB angiotensin receptor blocker therapy

Full size image

The intensified regimen in patients 1 and 2 consisted of intravenous pulses of methylprednisolone (750 mg/m2 given on 3 consecutive days), oral CyA started at a dose of 150 mg/m2 divided into two doses and initially adjusted to trough levels >100 ng/ml, and oral prednisone given for a total of 24 weeks at a starting dose of 30 mg/m2/day tapered to 10 mg/m2/48 h. During follow-up, monotherapy with CyA was aimed at the lowest possible dose (but adjusted to increases in PU) and stopped if possible (apparent remission). Relapses after discontinuation of therapy were treated with a new course of intensified therapy.

Patient 3

Current age 8 years and 3 months, WT1 gene mutation: exon 9 c.1180C>T (p.R394W), normal female phenotype, karyotype 46XX. PU was diagnosed at the age of 5.3 years (4.2 g/24 h, PU 8.4 g/g, s-Alb 24 g/l, s-creatinine 0.6 mg/dl). There was no reduction in proteinuria after 4-week oral administration of prednisolone 60 mg/m2/day. A renal biopsy revealed typical lesions of FSGS. Intensified therapy included intravenous methylprednisolone (six pulses of 300 mg/m2 on alternate days given over 2 weeks combined with oral prednisolone 60 mg/m2 between pulses) and oral CyA (Neoral) 150 mg/m2/day. In addition, enalapril (2.5 mg/day) and felodipine (5 mg/day) were given. PU decreased progressively to 0.1 g/g after 1 year and Pred was discontinued (Fig. 3a).

Fig. 3
figure 3

a Clinical course of patient 3. Dots indicate the urine protein/creatinine ratio (g/g), thin bars the time-averaged mean urine protein/creatinine ratio and thick bars the trough level of CyA in serum. b GFR of patient 3. Dots indicate calculated GFR; the line is a calculated mean GFR over time

Full size image

No relapses were recorded during 2.5 years of follow-up. The patient is presently being treated with enalapril (5 mg) and CyA (30 mg bid) with levels of C0 from 50 to 70 ng/ml and of C2 from 370 to 440 ng/ml. GFR has remained normal (Fig. 3b).

Discussion

We report long-term observations in 3 children with FSGS associated with WT1 mutations. In 2 cases, the NS was unresponsive to 4 weeks’ treatment with the standardized regime of oral prednisone and was therefore considered steroid-resistant [9]. In the remaining case (patient 2), the presence of nephrotic-range proteinuria in combination with FSGS on renal biopsy was considered an indicator of a poor response to standard oral steroid therapy, as has been shown for both children and adults [10]. However, all children showed a favorable response to an intensified therapy consisting of CyA in combination with induction therapy, using both intravenous and oral Pred given for a total of 6 and 12 months (patients 1 and 2, and patient 3 respectively). As shown both by short-term response and during long-term follow-up, this treatment resulted in clinical remission of the NS and/or significant reduction in proteinuria. These observations suggest an inverse relation of proteinuria to CyA/Pred dosing (Figs. 1a, 2a). However, it cannot be decided from these retrospective data whether this was due to increased dosing of CyA—as reflected by the higher trough levels—or Pred, which was given as induction therapy if a relapse occurred.

This therapy seemed not to be accompanied by a significant loss of renal function, which remained supranormal over many years (in patients 1and 2). Whether the long-term natural course of FSGS (with the inevitable deterioration of GFR) could be significantly improved by this form of treatment cannot be answered; however, recent retrospective data [11] have demonstrated a favorable effect of intensified CyA/Pred treatment on outcome in most children with steroid-resistant NS without known genetic mutations.

Our observations suggest that glomerular diseases in selected patients with mutations in genes regulating renal structural development may respond to combination therapy with CyA and corticosteroids. This could result from non-immunological mechanisms. It has recently been shown that the antiproteinuric properties of CyA can be explained by the direct effects of calcineurin signaling on podocytes (stabilization of the actin cytoskeleton) rather than by suppression of T lymphocytes [12]. Similarly, glucocorticoids have direct effects on human podocytes in vitro, such as upregulation of nephrin expression [13]. Apart from the patients with WT1 mutations described here, response to therapy in patients with known mutations in genes encoding podocyte proteins has been documented only in patients with PCLE1 mutations; among 7 patients with PLCE1 mutations and early-onset nephrotic syndrome associated with DMS, 2 patients had a sustained response to Pred or CyA therapy [14]. This could suggest that a favorable response to therapy in SRNS may not be confined to patients with FSGS, which was the histological finding in all our patients. In addition, patients with Alport syndrome, another glomerular disease associated with congenital structural abnormalities, have been shown to respond to treatment with CyA [15, 16]. While these findings suggest the direct effects of CyA and glucocorticoids on podocyte function, additional therapy with angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) may have contributed to the favorable clinical course of the patients described here. However, relapses of proteinuria were observed after discontinuation of CyA (while treatment with ACE inhibitors and ARBs was continued), suggesting that maintenance of remission was critically dependent on CyA.

WT1, one of the WT suppressor genes is located on chromosome 11p13 and encodes a transcription factor of the zinc finger family; four different transcripts (WT1 isoforms) are produced by alternative splicing [17]. Mutations in the WT1 gene in patients with SRNS seem to be restricted to exons 8 and 9 [7]. WT1 mutations in patients with SRNS are associated with FSGS, diffuse mesangial sclerosis (DMS) or membranoproliferative glomerulonephritis [2] on biopsy. Interestingly, the same WT1 mutation (1228 +5 G>A) may be associated with different phenotypes and glomerular diseases in the same family, as shown by Denamur et al. who found transmission of this mutation from a phenotypically normal (karyotype XX) mother with proteinuria and FSGS to a child with the Denys–Drash syndrome (karyotype XY) and DMS [18]. Analysis of a large cohort of pediatric patients with WT1 mutations and SRNS revealed that all patients with isolated SRNS were phenotypically female, whereas males had additional genitourinary tract malformations [7]. This corresponds to our observations in 3 patients.

All mutations found in our patients have been described previously. The splice site mutation detected in patient 1 (1228 +5 G>A) is as a frequent finding in patients with Frasier syndrome (FS), which is typically associated with FSGS [18–20]. In normal XX females (such as patient 1), FS may occur as an isolated glomerulopathy, i.e. FSGS without WT and with normal gonadal development [5]; however, most patients develop ESRD during the second decade of life. The mutation found in patient 2 (1147 T>C:383 Phe>Leu) has been described previously in a male patient with DMS as the only abnormality; this patient developed ESRD at the age of 6 [21]. Our observations indicate that this mutation can also be associated with FSGS and a more benign clinical course, possibly induced by early effective therapy. The mutation c.1180C>T present in the third patient is the most common mutation found in patients with the Denys–Drash syndrome. However, the glomerulopathy in these patients occurs typically in the form of DMS, which is unresponsive to medication and leads to ESRD in early childhood [17]. In our case, FSGS was present and remission of the NS without progression of renal disease could be achieved.

Interestingly, this mutation has been shown to alter the ability of WT1 to regulate downstream target genes, resulting in glomerulosclerosis in animal experiments [22]. After renal differentiation, WT1 is highly expressed only in podocytes, where it has been shown to be a key regulator of podocyte functions, including the expression of nephrin [23] and podocalyxin [24]. Further analysis of gene regulation by WT1 in cultured podocytes and transgenic mice seems of scientific value in elucidating the pathogenesis of podocyte dysfunction in these disorders [25].

Our data suggest that the detection of WT1 mutations in patients with SRNS/FSGS should not be an obstacle to intensified treatment with CyA and glucocorticoids. Whether beneficial effects of this treatment are confined to certain mutations and/or can be extended to other histological variants of the NS can only be answered by additional studies.