GB Haycock

Dr Cheng, Members of the Hong Kong Paediatric Society, ladies and gentlemen: let me begin by thanking you for the honour you have done me in inviting me to give the 25th C Elaine Field Lecture. It is a pleasure not only to pay tribute to Dr Field herself but also to join the ranks of the distinguished paediatricians who have preceded me over the past 25 years.

Introduction and Definitions

I have chosen as my subject the management of the steroid sensitive nephrotic syndrome, because it is one of the commoner potentially serious renal diseases of children and because there have been some interesting developments in the last few years, particularly in the treatment of children in the more severely affected subgroups. In order to understand clearly what follows, it is necessary to agree on standard definitions of the disease and its subtypes: although arbitrary in origin, these have achieved a measure of consensus over the past few decades and are helpful in ensuring that when evaluating the results of published reports of new (and old) treatments we are indeed comparing like with like. The definitions used by most paediatric nephrologists are based on those put forward three decades ago by members of the multinational, collaborative research group known as the International Study of Kidney Disease in Childhood (ISKDC).

Steroid sensitive

The disease is labelled steroid sensitive (or steroid responsive) if there is response within 8 weeks of starting a standard course of prednisone (or prednisolone) therapy.

Steroid resistant

Failure to respond to standard steroid therapy within 8 weeks.

Response to treatment (remission)

Disappearance of proteinuria (negative or trace on testing with Dipstix or sulphosalicylic acid) for three consecutive days.

Relapse

Recurrence of proteinuria at the ++ level or greater for three consecutive days, or recurrence of proteinuria at any level with hypoalbuminaemia and oedema.

Standard steroid therapy

The ISKDC recommended that the initial episode of the disease be treated with prednisone in a dose of 60 mg/m²/day in 2 or 3 divided daily doses for 28 days, followed by 40 mg/m²/day as a single morning dose for 3 consecutive days every week (e.g. Monday, Tuesday, Wednesday) for 4 weeks Relapses were to be treated with 60 mg/m²/day until response occurred (see definition of response above), followed by 4 weeks of intermittent treatment as described in the preceding sentence. This regimen is still widely used, with the modification that most people now give the intermittent part of the treatment as 40 mg/m²/dose as a single morning dose on alternate days for 4 weeks rather than 3 consecutive days of each week.

Frequent relapses

The disease is described as frequently relapsing if two or more relapses occur within 6 months of completing the initial course of steroid treatment, or if three or more relapses occur within any 12 month period.

Infrequent relapses

Relapses occurring less frequently than the above.

Steroid dependent

Two or more consecutive relapses occurring during steroid treatment or within two weeks of stopping treatment. Children with steroid dependent disease require continuous, or nearly continuous, steroid treatment to keep them in remission.

Steroid Treatment of the Initial Episode

Several studies have questioned whether the above recommendations are optimal. A report by the ISKDC itself¹ showed that treating children who relapsed early with a slightly more aggressive steroid regime significantly extended the time to next relapse. A Japanese study² comparing standard ISKDC treatment of the initial episode with an extended (6 months) course of prednisone found that the latter was significantly more effective in preventing relapses during the first 6 months (76% relapse free vs. 41%), and even more so at protecting against frequent relapsing or steroid dependent status at last follow up (6% vs. 36%). A multicentre study from Germany³ reported that shortening the standard ISKDC treatment of the initial attack halved the likelihood of children remaining relapse-free at one and two years, from approximately 40% to 20%. A large study from a single centre in Poland,⁴ comparing three different steroid regimes, showed marked benefit from a 6 month course of steroid for the initial attack, 50% of patients so treated remaining in remission at 24 months compared with 21% and 27% of children treated for 2 and 3 months respectively. This was achieved without any apparent increase in the incidence or severity of side effects, and indeed the group treated for 6 months had the lowest total cumulative dose of steroid though it did not reach statistical significance over the course of the first 2 years, which included the initial treatment period, because of the reduced need for treatment of relapses. My reading of the published evidence therefore leads me to recommend that the first episode of nephrotic syndrome be treated with 60 mg/m²/day, given as a single morning dose, for 4 weeks, after which alternate day treatment is commenced at 40 mg/m²/dose for 4 weeks, tapering over the subsequent 4 months. A recent study from India⁵ failed to show any advantage of divided daily doses over single morning dosage in the time to remission, and an advantage of morning administration of steroids is that suppression of the pituitary-adrenal axis is minimized. Failure to respond to this regime within 4 to 8 weeks is an indication for renal biopsy, bearing mind that 7% of those who responded to standard ISKDC treatment did so during the second 4 weeks.

Treatment of Relapses with Steroids

There is very little evidence-based information to guide the clinician in the rational treatment of relapses. Even following a 6 month initial course of steroids, about 50% of patients can be expected to relapse one or more times, with a smaller proportion becoming frequent relapsers or steroid dependent.^2-4 A commonly used protocol is to give prednis(ol)one at 60 mg/m²/day until response occurs, followed by about a month of alternate day or tapering daily treatment. If relapses are infrequent, this approach works well and is well tolerated, because the child has time to recover from the side effects of each course of steroids before the next one is given. In frequently relapsing and, especially, steroid dependent children adverse effects of treatment are invariably seen in proportion to the time spent on high dose steroids and other approaches should be considered. The first of these is to attempt to achieve long term remission by giving maintenance, low dose, alternate day steroid therapy. One way of introducing this is to treat the latest relapse conventionally, but to halt the tapering of the alternate day part of the regime at a dose 5 mg higher than that at which the last relapse occurred. For example, if the patient relapsed at a dose of 10 mg on alternate days, maintenance treatment is initially set at 15 mg every other day, with no attempt to reduce the dose further until several months of continuous remission, unless side effects make this unacceptable I am unaware of any controlled trials of this form of treatment, but some children undoubtedly tolerate it well and can be successfully weaned off steroids over a long period of time (such as 1-3 years). Preliminary results suggest that deflazacort, a synthetic glucocorticoid, may be more effective than prednis(ol)one in maintaining remission with less adverse effects on bone mineral content,⁶ but this observation requires confirmation before deflazacort can be recommended as the glucocorticoid of choice in childhood nephrotic syndrome.

Alkylating Agents

There is no doubt that alkylating agents can be effective in the treatment of steroid sensitive nephrotic syndrome, as supported by numerous published studies. An important and influential report⁷ demonstrated that a two week course of cyclophosphamide in a dose of 3 mg/kg/day was more effective than no treatment in preventing relapses, and an eight week course was more effective still, with three quarters of patients in steroid-free remission one year after treatment. Grupe et al reported that chlorambucil, in a mean cumulative dose of 16.9 mg/kg, was even more effective with no relapses occurring within one year (actuarial) and two years (actuarial) of treatment in 10 children.⁸ However, as stated in the text of this paper, a cumulative dose of only 6 mg/kg is thought to be hazardous in terms of future fertility in male patients, so it is likely that the apparent superior efficacy of chlorambucil over cyclophosphamide may simply have been due to the fact that a very large amount of the drug was given. In a study of 13 children, Padilla & Brem showed not only that alkylating agents were effective in preventing relapses but also that the consequent reduction in steroid dosage led to striking improvement in linear growth in both boys and girls.⁹ In a controlled study of 58 nephrotic children, Vallo et al reported that intravenous nitrogen mustard (mechlorethamine) was significantly more effective than oral cyclophosphamide in preventing relapses over a 5 year follow up period.¹⁰ It is at least possible, and in my opinion likely, that at least some of this difference is due to non-compliance in some children receiving cyclophosphamide this does not reduce the importance or value of the observation.

Alkylating agents are extremely toxic drugs, which explains the reluctance of many paediatricians to use them in non-malignant diseases. Their major side effects include infertility, the induction of secondary malignancies and in the case of cyclophosphamide haemorrhagic cystitis carrying the risk of bladder fibrosis. Moreover, the early studies cited above may have overestimated their effectiveness. A more recent trial by Ueda et al suggested that only about 25% of steroid dependent children achieved long term steroid free remission after either 2 or 3 months cyclophosphamide treatment.¹¹ The inconsistency between this finding and the more favourable results reported by Barratt et al, among others,⁷ almost certainly lies in the different mix of cases treated. Early studies did not clearly distinguish between different categories of relapsing children (see definitions above), and included children who were not steroid dependent. Not surprisingly, children with less severe disease respond better to most forms of treatment than those more severely affected. The balance of evidence indicates that only a minority of truly steroid dependent children, the group that is in most need of steroid-sparing treatment, will gain lasting benefit from alkylating agents. However, in this minority the magnitude of the benefit may be large, with many becoming permanently relapse-free and losing their dependency on steroids, alkylating agents are therefore still used in my department in selected individuals. For reasons given above, we now usually use nitrogen mustard according to the regime described by Vallo et al. The use of modern, powerful anti-emetic drugs such as ondansetron has transformed the acceptability of this treatment to the patients. An interesting recent report suggests that class II HLA antigen status may influence the result of treatment with alkylating agents.¹² Children who were negative for DR7 were much more likely to experience a good result than those who were DR7 positive. If confirmed, this observation may make it possible to make a more informed decision about whether to use these drugs in individual patients.

Levamisole

Most drugs that are effective in the nephrotic syndrome are immunosuppressive (although all have other actions). Levamisole is the exception, being a stimulator of the immune system. Following the publication of several anecdotal reports a randomised, prospective, double blind controlled trial was conducted by the British Association for Paediatric Nephrology.¹³ This showed a significant effect in delaying relapse, with about 40% of treated patients remaining in remission at 100 days compared with less than 10% of controls. The follow up period in this study was short, and almost all patients went on to relapse. There seems no doubt that levamisole has a modest effect on children with steroid sensitive nephrotic syndrome, but it very rarely achieves the primary objective of treatment, which is to induce a prolonged drug free remission. Colleagues in India, where the disease is very common advise me that the drug has a useful steroid sparing effect in that children maintained on levamisole and alternate day steroids often need a lower dose of steroids than they would if they were on prednisolone alone. A prospective study of this mode of therapy has not been done: the results would be interesting, but it would be difficult to conduct a study with sufficient power to detect what is probably a relatively small effect.

Cyclosporin A

The powerful immunosuppressive agent cyclosporin A has been shown in several studies to be capable of inducing and sustaining remission in children with steroid sensitive nephrotic syndrome. A study from the Great Ormond Street Hospital for Sick Children in London is representative.¹⁴ In general, about 75% of children can be weaned off their steroids under cover of low dose cyclosporin A without immediate relapse. However, up to 50% of these will relapse during the first two years of treatment, and almost all relapse soon after treatment is stopped. In other words, the effect of this drug is to substitute cyclosporin A dependence for steroid dependence. Cyclosporin A is nephrotoxic, and should only be prescribed by physicians who are familiar with its use in transplantation and who have access to facilities to monitor blood levels of the drug regularly. However, in a large and so far unpublished series we have followed all nephrotic patients on long term cyclosporin A with serial, usually annual, renal biopsies, m some cases for more than 5 years uninterrupted treatment. At the dosage used (usually 4-6 mg/kg/day in two divided doses), with the blood concentration of the drug at the lower end of the anti-rejection range, almost no evidence of chronic cyclosporin A nephrotoxicity was found and we believe that the drug has an important role in steroid dependent children who have failed to respond to an alkylating agent. The subjective benefit to the children and their families is attested to by the fact that they are nearly always most reluctant to consider stopping cyclosporin A for fear of having to go back on steroid treatment.

Opinions are divided as to whether an alkylating agent or cyclosporin A should be tried first in steroid dependent children. One study has shown that the probability of remission at one year is almost exactly the same in children treated with cyclophosphamide for 8 weeks and in those treated with cyclosporin A for 12 months.¹⁵ After stopping the cyclosporin A, however, almost all the children in this arm of the study relapsed. Thus, although alkylating agents do not induce steroid free remissions in all steroid dependent patients, the remissions that it does cause are much more stable than those produced by cyclosporin A. I therefore prefer to use an alkylating agent first, and to reserve cyclosporin A for children in whom the alkylating agent is unsuccessful. A case could be made that in children who are DR7 positive the order should be reversed, but I would like to see independent confirmation of the association between DR7 status and the response to alkylating agents before altering my current policy.

Other Drugs

The immunosuppressive agent tacrolimus (FK506) is similar in its actions to cyclosporin A, but perhaps more powerful. It is certainly more expensive. A preliminary report suggests that it may be effective in steroid resistant nephrotic syndrome.¹⁶ At least one trial is under way to evaluate tacrolimus in steroid dependent patients who have relapsed on both alkylating agents and cyclosporin A, but as yet there is no evidence that tacrolimus is more effective than cyclosporin in the nephrotic syndrome. A few reports have appeared of the use of other experimental drugs, including mizorbine,¹⁷ but the results so far are inconsistent and they certainly have no place in the routine treatment of this non-fatal, but often very disabling, condition. It is to be hoped that improved understanding of the pathophysiology of the disease, which has proved unexpectedly refractory to investigation, will eventually pave the way for more effective, more specific and less toxic treatments than are available at present.

References

1. International Study of Kidney Disease in Childhood. Nephrotic syndrome in children: a randomized trial comparing two steroid regimens in steroid-responsive patients who relapse early. J Pediatr 1979;95:239-43.

2. Ueda N, Chihara M, Kawaguchi S, Niinomi Y, Nonoda T, Matsumoto J, et al. Intermittent versus long-term tapering prednisolone for initial therapy in children with idiopathic nephrotic syndrome. J Pediatr 1988;112:122-6.

3. Arbeitsgemeinschaft fur Padiatrische Nephrologie. Short versus standard prednisone therapy for initial treatment of idiopathic nephrotic syndrome in children. Lancet 1988;1:380-3.

4. Ksiazek J, Wyszynska T. Short versus long initial prednisone treatment in steroid sensitive nephrotic syndrome in children. Acta Paediatr 1995;84:889-93.

5. Ekka BK, Bagga A, Srivastava RN Single- versus divided-dose prednisolone therapy for relapses of nephrotic syndrome. Pediatr Nephrol 1997;11:597-9.

6. Broyer M, Terzi F, Lehnert A, Gagnadoux M-F, Guest G, Niaudet P. A controlled study of deflazacort in the treatment of idiopathic nephrotic syndrome Pediatr Nephrol 1997;11:418-22.

7. Barratt TM, Cameron JS, Chantler C, Ogg CS, Soothill JF. Comparative trial of 2 weeks and 8 weeks cyclophosphamide in steroid-sensitive relapsing nephrotic syndrome of childhood Arch Dis Child 1973;48:286-90.

8. Grupe WE, Makker SP, Ingelfinger JR Chlorambucil treatment of frequently relapsing nephrotic syndrome. N Engl J Med 1976;295:746-9.

9. Padilla R, Brem AS. Linear growth of children with nephrotic syndrome: effect of alkylating agents. Pediatrics 1989;84:495-9.

10. Vallo A, Rodriguez-Soriano J, Quintela J. Comparative effects of cyclophosphamide and nitrogen mustard in treatment of idiopathic nephrotic syndrome. Pediatr Nephrol 1993;7(Suppl):C33.

11. Ueda N, Kuno K. Ito S. Eight and 12 week courses of cyclophosphamide in nephrotic syndrome. Arch Dis Child 1990;65:1147-50.

12. Konrad M, Mytilineos J, Ruder H, Opelz G, Scharer K. HLA-DR7 predicts the response to alkylating agents in steroid-sensitive nephrotic syndrome. Pediatr Nephrol 1997;11:16-9.

13. British Association for Paediatric Nephrology. Levamisole for corticosteroid-dependent nephrotic syndrome in childhood. Lancet 1991;337:1555-7.

14. Hulton S-A, Neuhaus TJ, Dillon MJ, Barratt TM. Long-term cyclosporin A treatment of minimal-change nephrotic syndrome of childhood. Pediatr Nephrol 1994;8:401-3.

15. Ponticelli C, Edefonti A,Ghio L, Rizzoni G, Rinaldi S, Gusmano R, et al. Cyclosporin versus cydophosphamide for patients with steroid-dependent and frequently relapsing idiopathic nephrotic syndrome: a multicentre randomized controlled trial. Nephrol Dial Transplant 1993;8:1326-32.

16. McCauley J, Shapiro R, Ellis D, Igdal H, Tzakis A, Starzl TE. Pilot trial of FK 506 in the management of steroid-resistant nephrotic syndrome. Nephrol Dials Transplant 1993;8:1286-90.

17. Hamasaki T, Mon M, Kinoshita Y, Saeki T, Sakano T. Mizorbine in steroid-dependent nephrotic syndrome of childhood. Pediatr Nephrol 1997;1l:625-7.