Main

Nephropathic cystinosis is a rare autosomal recessive disorder characterized by intracellular storage of free cystine in various organs, predominantly the kidney(1–3). Untreated patients develop end-stage renal failure at an average of 9 y(4, 5). There is evidence that treatment with cysteamine, a cystinedepleting agent, slows the progression of renal glomerular disease, although it does not prevent the development of Fanconi syndrome(4, 6, 7).

Most children with nephropathic cystinosis develop severe growth retardation(2, 6, 8, 9). Cysteamine treatment may prevent growth retardation to a certain extent when administered from early infancy onward(4, 6), but does not induce catch-up growth in children who are already growth-retarded(7). rhGH treatment has been proven to be effective in promoting growth in short children with chronic renal failure for a treatment period of up to 5 y(10, 11). There is evidence that rhGH also markedly improves growth velocity in children with nephropathic cystinosis(10, 12, 13). However, concern has been raised that rhGH therapy might induce an accelerated rise of serum creatinine levels, resulting in the anticipated need for renal replacement therapy(14). This study analyzes the course of renal function in 36 short children with nephropathic cystinosis before dialysis entered into a trial of rhGH treatment.

METHODS

Patients. An open-label European multicenter study was initiated by the principle investigators (O.M., M.B., W.G.v.H.) in 1990 to evaluate the efficacy and safety of rhGH treatment in short children with nephropathic cystinosis(13). Sixty-four patients, 44 on conservative treatment, 9 on dialysis, and 11 with renal allografts have been included in this study, fulfilling the following inclusion criteria:1) growth retardation defined as a height SDS for chronologic age of≤ 2.0 and/or a height velocity SDS for chronologic age ≤0 using the first Zurich longitudinal study asreference (15);2) age >2 y; 3) first clinical signs of puberty present for no more than 1 y; 4) at least three separate height measurements during the previous year performed at the clinic of the responsible investigator; 5) cysteamine treatment, when used, had been started at least 1 y before rhGH and was continued during the study period;6) normal thyroid function levels or, if hypothyroid, and had been started on thyroid hormone treatment at least 1 y before rhGH; 7) no diabetes mellitus according to WHO criteria; 8) absence of severe renal osteodystrophy; and 9) written informed consent of patients and parents. The study protocol was approved by the Ethics Committee of the University of Heidelberg and the local Ethics Committees of the participating centers.

To analyze the effect of rhGH on the course of renal function, this evaluation was restricted to 36 patients on conservative treatment with normal(n = 5) or reduced (n = 31) renal function (i.e. calculated CCR < 80 mL (min × 1.73 m2)-1 with a follow-up period of more than 1 y. Baseline clinical data are shown in Table 1.

Table 1 Baseline clinical characteristics
Full size table

To evaluate the influence of concomitant cysteamine treatment on the course of renal function before and after start of rhGH therapy, two subgroups were defined. The arbitrary separation into group A and B was based on the assumption that cysteamine is more effective if given early in the course of the disease and continuously over a long period of time.

Group A. Group A (no or short-term cysteamine treatment;n = 21) included children without cysteamine therapy at any time(n = 6), late onset of treatment (age > 3 y), and/or a short duration of treatment (<2 y time) (n = 15). Mean age at start of cysteamine was 4.3 ± 2.3 y, and mean duration of cysteamine treatment was 1.8 ± 1.7 y.

Group B. Group B (cysteamine treatment; n = 15) included children with the start of cysteamine therapy before 3 y of age and with duration of treatment for at least 2 y before rhGH. Mean age at start of cysteamine was 1.8 ± 0.6 y, and mean duration of cysteamine treatment 5.7 ± 2.1 y. The median prescribed cysteamine dosage was 45 (35-75) mg(kg × d)-1 for those patients who received cysteamine. There were insufficient data on leukocyte cystine concentrations to allow analysis of compliance.

Study medication. rhGH (Genotropin®, Pharmacia & Upjohn, Stockholm, Sweden) was administered at a weekly dose of 1 IU/kg in daily evening s.c. injections.

Physical and laboratory assessments. Standard anthropometry was performed at three monthly intervals. At each visit to the clinic, blood samples were taken for measurement of serum creatinine and urea. The measurements were performed by autoanalyzer in the individual centers. GFR was estimated by CCR according to the method of Schwartz et al.(16), using the formula: CCR = height [cm] ×F/serum creatinine [mg/dL]. The factor F varied between 0.45 and 0.55 as reported by the individual centers. In addition, CIN measurements (optional) were performed (using standard techniques) in some centers (Heidelberg, London, Lyon, and Paris) in 12 children at baseline and after 1 y of rhGH treatment. A 24-h urine sample was used for assessment of calcium and protein excretion. Urinary protein excretion was measured by the Biuret method.

To analyze the progression of renal failure before rhGH treatment, the participating centers reported all available serum creatinine values of their patients determined before the start of rhGH. The individual rise in serum creatinine levels was compared with that of 116 control cystinotic patients(who did not receive cysteamine or rhGH) using standards of serum creatinine class values given by Manz and Gretz(5). In addition, the change of calculated CCR per year was compared with the change of calculated CCR in 1) noncystinosis control patients with CRF without rhGH treatment taken from the European Study on Dietary Intervention(17) (age 9.0 ± 4.2 y) and 2) short noncystinotic patients with CRF treated with rhGH within the German Study Group on Growth Hormone Treatment in Chronic Renal Failure(10) (age 8.3 ± 3.4 y). The majority of patients of the “control” groups suffered from renal hypoplasia or dysplasia.

Statistical analysis. Values are expressed as mean ± SD, if not indicated otherwise. For comparison of the rise in serum creatinine in rhGH treated and control patients the progression charts of Manz and Gretz(5) were used. The number of patients progressing within a certain range (0.0-50.0 centile and 50.0-100.0 centile) was counted and analyzed in a cross table by Fisher's exact test according to the method proposed by Gretz(18).

Comparisons between groups were performed by ANOVA, followed by Duncan's multiple rank test. Longitudinal within-group differences were assessed by one-way ANOVA for repeated measurements. Statistical significance was accepted for p < 0.05.

RESULTS

Growth. During the first treatment year mean height velocity of the 36 patients increased from 4.1 ± 1.6 cm/y before the start of rhGH to 8.8 ± 2.5 cm/y, and height SDS improved within 1 y from -4.2± 1.0 to -3.3 ± 1.0. In nine patients treated for 4 y or more, mean height SDS improved by 1.7 ± 0.9 SDS within 4 y. A detailed analysis of complete growth data of the total study population (including dialysis and transplant patients) will be published separately.

Rise of serum creatinine concentration. The individual rise in serum creatinine values with increasing age was plotted into the progression charts of Manz and Gretz(5) (Fig. 1,A and B). In the subgroup (A) without cysteamine treatment, the number in the different centile ranges of the progression charts did not differ from the historical control group (Table 2; p = 0.12), either at the start of rhGH treatment, or at the time of the last observation. In contrast, in the cysteamine-treated group (B), only 2 out of 15 patients had an early rise in serum creatinine at baseline (Table 2; p = 0.0062), and there was no change with rhGH treatment. The individual curves of the total group with rhGH treatment showed a significantly (p = 0.0064) later rise in serum creatinine levels than the historical control group (Table 2). Serum urea concentration remained stable during the first study year (9.2 ± 4.2versus 7.8 ± 4.8 mmol/L)

Figure 1
figure 1

Rise in serum creatinine levels of patients with no or short-term cysteamine treatment (panel A) and of patients with long-term cysteamine treatment (panel B) plotted on the creatinine centiles of a historical control group (-, 5th, 50th, and 95th centile)(5), - - - - -, pretreatment period; -, treatment period. Statistical evaluation seen from data of Table 2.

Full size image
Table 2 Number of patients treated with rhGH with an early (0-50th centile) or late (50-100th centile) rise in serum creatinine levels in comparison with the median rise in an untreated control group (n = 116)(5, 18)
Full size table

GFR. Mean CCR decreased during the year before rhGH treatment as well as during the study period (Fig. 2). The decline of CCR was compared with the recently published data of noncystinotic patients without rhGH treatment from the European Study on Dietary Intervention(17), as well as to the data of rhGH-treated short children with chronic renal failure from the German Study Group(10). The absolute loss of CCR during the year before rhGH treatment was higher in the present study than during the pretreatment period in the other two studies, whereas the change of CCR during the 3-y study period did not differ significantly in the three studies (Fig. 2). In nine patients with an observation period of at least 4 y, mean loss of GFR was 12.8 mL (min × 1.73 m2)-1 within 4 y.

Figure 2
figure 2

Course of CCR in patients with nephropathic cystinosis before and during rhGH treatment (▾-▾) in control patients with CRF without rhGH treatment taken from the European Study on Dietary Intervention (○- - - -○)(17), and in short noncystinotic patients with CRF treated with rhGH within the German Study Group on Growth Hormone Treatment in Chronic Renal Failure (▵-▵)(10). Data are given as mean ± SEM.

Full size image

CIN values were obtained in 12 patients at start and after 1 y of rhGH treatment (Table 3). Serum creatinine in these patients was 107.8 ± 51.2 μmol/L at the start and 135.3 ± 66.3 μmol/L after the first treatment year. Calculated CCR overestimated the “true” GFR (CIN) by 13%. However, the mean change of GFR/y measured by both methods did not differ significantly.

Table 3 Inulin clearance, calculated CCR, and serum creatinine values in 12 patients with nephropathic cystinosis before and after 1 y of rhGH treatment
Full size table

In the same patients, urinary protein excretion was 3.8 ± 2.0 g(m2 × d)-1 at the start and 1.9 ± 1.0 g (m2× d)-1 after 1 y of rhGH treatment. Urinary calcium per creatinine excretion was 0.85 ± 0.54 mmol/mmol of creatinine at the start and 0.71 ± 0.44 after 1 y of rhGH treatment (normal range<0.74 mmol/mmol of creatinine).

DISCUSSION

Preliminary data from previous studies(10–12) and the present study(13) on the use of rhGH in short patients with nephropathic cystinosis that have shown a beneficial effect on growth(10, 11) are encouraging. However, there is concern that rhGH treatment may accelerate the progression of CRF and may induce an early need for renal replacement therapy(14). The present study provides strong evidence that treatment with rhGH for up to 5 y does not accelerate the progression of CRF in a large cohort of children with nephropathic cystinosis at various stages of renal function.

Data from a historical control group of patients with nephropathic cystinosis suggests that patients develop chronic renal failure at a mean age of 3 y and reach end-stage renal failure at a mean age of 9 y(4, 5). Treatment with cysteamine postpones the need for renal replacement therapy by some years(4, 6, 7, 19), but only if treatment is started before the age of 2 y and the patients remain compliant. Cysteamine treatment also appears to improve growth in these patients(4, 6). However, despite early and appropriate cysteamine treatment, many children do not grow adequately, and statural height falls below the third centile(7).

In 1992, Andersson et al.(14) reported their experience with 2 y of treatment with rhGH in three short children with nephropathic cystinosis. They reported an accelerated rise of serum creatinine. This is not confirmed by our investigation. Our data demonstrate that at baseline the serum creatinine values of noncysteamine-treated patients(group A) did not differ from the historical control group. The rise in serum creatinine followed the centiles of Manz and Gretz(5). In addition, the point at which serum creatinine began to rise more sharply did not coincide with the start of rhGH treatment. Patients on long-term cysteamine treatment (group B) had a later increase in serum creatinine than did those with short-term or no cysteamine treatment. The rise of serum creatinine was again parallel to the centiles. Although, in this multicenter trial, there were insufficient data on leukocyte cystine concentrations to evaluate compliance to cysteamine therapy, this finding suggests a positive effect of cystine-depleting agents in preserving renal function. It confirms the observation of Schneider et al.(20) who also used the serum creatinine class values of Manz and Gretz(5). In our study this effect was independent of rhGH treatment.

It is conceivable that a more contemporary control group would demonstrate a much better renal survival than the historical control group. However, this trial of rhGH was undertaken in a selection of severely affected cystinotic patients (i.e. the shortest). It is possible that this selected group has a more rapid decline in renal function compared with unselected patients. Therefore, it seems justified to compare our patients with the historical control group of Manz and Gretz(5).

The annual change of CCR in the present investigation was compared with that in noncystinotic patients with CRF, either treated or nontreated with rhGH. The comparison (Fig. 2) showed a higher loss of CCR in the cystinotic patients during the 1-y run-in period than in the other two groups, whereas the change of CCR was only slightly higher during the 2 y of rhGH treatment. In addition, the mean loss of 13 mL (min× 1.73 m2)-1 within 4 y in the cystinotic patients is comparable to the mean loss of GFR in noncystinotic children with CRF treated with rhGH in the American Study of Fine et al.(11). These observations provide further evidence that rhGH does not lead to an accelerated deterioration of renal function.

We have confirmed in this study the well known observation that CCR overestimates true GFR (as determined by CIN). However, the change of GFR/y measured by both methods was comparable in the 12 patients in whom it was evaluated by CIN and CCR in parallel during the first treatment year. It might be argued that measurement of creatinine is not appropriate for assessment of renal function in patients treated with rhGH, because rhGH increases height and disproportionately muscle mass(21) and by this the endogenous source of creatinine. However, an increased creatinine production rate can cause an over-but not an underestimation of the rise in serum creatinine. The Schwartz formula(16) for estimation of CCR is based on the assumption that an increase in height (numerator) is balanced by an increase in muscle mass reflected by the serum creatinine concentration (denominator). Again, disproportionate increase of muscle mass by rhGH treatment would lead to an underestimation and not to an overestimation of CCR.

The observation that rhGH did not lead to a rapid deterioration of renal function is in line with observations in animals and humans with CRF. Although an rhGH-induced increase in GFR followed by glomerular sclerosis was reported in rodents without CRF(22), no hyperfiltration was noted in rats(23) or humans(24) with CRF. Recent clinical experience in short-term controlled(25, 26) and uncontrolled(10, 11) studies in children with CRF with observation periods for up to 5 y also do not support the view that rhGH results in an accelerated rise in serum creatinine and decline of GFR.

Rodent experiments have further evidenced that rhGH treatment concomitantly with 1,25-(OH)2D3 treatment leads to hypercalciuria and eventually to renal damage(27). Patients with nephropathic cystinosis often present with hypercalciuria(28). In addition, almost all patients in the present study received 1,25-(OH)2D3 treatment for metabolic bone disease. It is therefore important to note that no increase of urinary calcium excretion was observed during the 1st y of rhGH treatment. Concomitantly, no increase of urinary protein excretion, a factor highly correlated with the decline of renal function(17) was noted. These results are in line with the observation in noncystinotic children with CRF treated concomitantly with growth hormone and 1,25-(OH)2D3(29).

In conclusion, the present study provides strong evidence that rhGH treatment in short children with nephropathic cystinosis does not accelerate the deterioration of renal function. Therefore, the continuation of controlled prospective studies appears to be justified in those cystinotic patients who urgently need treatment of growth failure.