- Email: [email protected]
Partial reversibility of cisplatin nephrotoxicity in children
Partial reversibility of cisplatin nephrotoxicity in children P e n e l o p e R. Brock, MD, Dimitri E. Koliouskas, PhD,* T. Martin Barratt, FRCP, Elizabeth Y e o m a n s , SRN, a n d Jon Pritchard, FRCP From the Departments of Haematology and Oncotogy and Paediatric Nephrology, Hospital for Sick Children and Institute of Child Health, London, England
To evaluate the long-term renal toxicity of cisplatin, 40 children who had been without treatment at least 18 months (range 18 months to 7 years) were observed. In all the children, glomerular filtration rate (GFR) was estimated from the plasma clearance of chromium 51-1abeled ethylenediaminetetraacetic acid, both at the end of treatment and at a median follow-up of 2 years 6 months after treatment was stopped (range 18 months to 7 years). In 21 children, serum magnesium level was also measured at follow-up. Median a g e at diagnosis was 15 months (range 13 days to 13 years 8 months), and median cumulative dose of cisplatin was 500 m g / m 2 (range 120 to 1860 mg/m2). In 22 of 24 children with an end-oftreatment GFR Of <80 ml/min per 1.73 m 2, the median improvement in GFR at follow-up was 22 ml/mln per 1.73 m 2 (range 2 to 56 ml/min per 1.73 m2). Hypomagnesemia was found in 6 of 21 children and was Independent of GFR. No significant correlation was found between improvement in renal function and total cisplatin dose, age, gender, tumor type, or associated nephrotoxlc medication. We conclude that most children have some recovery from cisplatin glomerular toxicity, especially if d a m a g e is not severe, but that h y p o m a g n e s e m i a may persist. (J PEDIATR1991;118:531-4)
Nephrotoxicity is the most important dose-limiting side effect of cisplatin. 1, 2 In 1985, Womer et al. 3 showed that creatinine clearance and serum creatinine concentration were inadequate measures of renal toxic effects in children treated with cisplatin. They used estimated glomerular filtration rate from the plasma clearance of chromium 5 I-labeled ethylenediaminetetraacetic acid and found reduced renal function in most cases. The aim of this follow-up study was to determine whether, in patients receiving cisplatin, there is recovery of renal function with time. The effects of
other nephrotoxic agents administered to these children as part of their treatment were included in the analysis. METHODS The study population consisted of 40 children treated with cisplatin, 60 to 100 m g / m 2 per course, as a component of multiagent chemotherapy. Between 1979 and 1988, a total of 164 children with a solid tumor received such treatment at our institution. No child died of nephrotoxic effects, 51Cr-EDTA
Supported in part by the British Council, Het Belgische Werk Tegen Kanker, and the Leukaemia Research Fund (Dr. Brock), and by the Imperial Cancer Research Fund (Dr. Pritchard). Submitted for publication Aug. I, 1990; accepted Oct. 29, t990. Reprint requests: Penelope R, Brock, MD, Paediatric Oncology, University Hospital Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium. *Current address: Paediatric Oncology, Aghias Sofias 9, 546 23 Thessaloniki, Greece.
9/20/26388
GFR
chromium 5 I-labeled ethylenediaminetet raacetic acid Glomerular filtration rate
and there were 55 survivors in January 1990. Parental consent was obtained for follow-up G F R measurements in 40 children. There was no reason to believe that the remaining 15 differed significantly in any respect. The diagnoses of the study population were neuroblastoma (27 patients), germ cell tumor (8), hepatoblastoma (3), and osteogenic sarcoma (2). Median age at diagnosis was 15 months (range 13 days
531
532
Brock et al.
The Journal of Pediatrics April 1991
n=23
---q
n=15
n=2
II
m
End-of-treatment
[~1
GFR
F o l l o w - u p GFR
GFR>80ml/min per 173mz GFR 60-80mr/rain per 1.73m2 GFR<60mt/min per 1.73m2
Figure. Glomerular filtration rate (GFR) measured by plasma clearance of StCr-EDTA in 40 children treated with cisplatin. Each line represents one patient. Median follow-up time from end of treatment, 2 years 6 months. Asterisk indicates patient with acute melphalan toxic effects.
to 13 years 8 months). There were 25 boys. Cisplatin was given intravenously for 15 to 30 minutes with hydration at 3 L / m 2 every 24 hours and diuresis with mannitol. The interval between courses was 3 to 4 weeks. The median cumulative dose of cisplatin was 500 m g / m 2 (range 120 to 1860 mg/m2). Chemotherapy regimens were as follows: neuroblastoma---cisplatin, cyclophosphamide, vincristine, teniposide-etoposide, and high-dose melphalan; germ cell tumor--cisplatin, Neomycin, and vinblastine-etoposide;hepatoblastoma--cisplatin and doxorubicin; and osteosarcoma---cisplatin, doxorubicin, and methotrexate. Thirteen children received high doses of melphalan as consolidation treatment for stage 4 neuroblastoma.4 Gentamicin was given to 18 children for a median of 17 days (range 7 to 34 days); serum levels were monitored, and in
no case did they reach toxic levels. 5 A further eight children also received at least one dose (1 mg/m 2) of furosemide. No child had a nephrectomy or underwent flank irradiation. The GFR was estimated during treatment and at follow-up from the plasma clearance of 51Cr-EDTA, by using a single-compartment analysis.6,7 Additional measurements were carried out in 27 children; the number of additional measurements ranged from one to five (median two). Each child's GFR was measured at least 18 months after treatment was stopped. When later measurements were undertaken, the most recent figure was used in analyzing the results. The median longest follow-up time after treatment was 2 years 6 months (range 18 months to 7 years). Serum magnesium and calcium values were obtained in 21 children at their most recent follow-up. Normal electrolyte values were taken from Clayton et al. 8 A gradual fall in concentration of both magnesium and calcium occurs during early childhood, - 2 SD from the mean being 0.75 mmol/L at 1 month and 0.7 mmol/L at 5 years for magnesium and 2.4 mmol/L at 1 month and 2.225 mmol/L at 5 years for calcium. Blood pressure was assessed according to the age-specific percentiles of Park and Menard 9 in the younger children, and on the basis of age and length according to Andre et al} ~ in the older children. The Spearman rank test was used to evaluate the relationship between mutually dependent variables, the Mann-Whitney U test for differences between medians, and the sign test for differences between matched samples. Two-tailed probabilities were calculated. For differences between proportions, the chi-square test was used. The null hypothesis was rejected at values of p <0.05. RESULTS
End-of-treatment GFR. Median end-of-treatment GFR was 74 ml/min per 1.73 m 2 (range 13 to 184 ml/min per 1.73 m2): 16 children had a GFR of >80 ml/min per 1.73 m 2, 13 had a GFR of between 60 and 80 ml/min per 1.73 m z, and 11 had a GFR of <60 ml/min per 1.73 m 2. No significant correlation was found between GFR and total cisplatin dose, age, gender, tumor type, or associated nephrotoxic medication. One child, treated early in the series (1980), had a sudden drop in GFR, from 72 ml/min per 1.73 m 2 after six courses of cisplatin (700 m g / m 2) to 15 ml/min per 1.73 m 2 after a further two courses, without apparent reason. In this child grade 4 ototoxic effects also developed. 11 At follow-up 9 years from diagnosis, his height and weight were at the 10th and 3rd percentiles, respectively, and his blood pressure was normal. His condition is managed conservatively, and his latest GFR was 27 ml/min per 1.73 m 2. Recovery of glomerular renal function. Median GFR at
Volume 118 Number 4, Part 1
follow-up was 90 ml/min per 1.73 m 2 (range 27 to 135 ml/ min per 1.73 m2). Of the 16 children with an end-of-treatment GFR of >80 ml/min per 1.73 m 2, 12 maintained a GFR of >80 ml/min per 1.73 m 2. The GFR fell to 60 to 80 ml/min per 1.73 m 2 in three children, and one child had progressive renal failure immediately after receiving highdose melphalan therapy. Of the 24 children who had a GFR of <80 ml/min per 1.73 m 2 at the end of treatment, all but two children showed recovery to a variable extent (Figure). The median change in GFR was +22 ml/min per 1.73 m 2 (range - 7 to +56 ml/min per 1.73 m2). A recovery GFR of >80 ml/min per 1.73 m 2 was measured in 11 children. Of these 11, the endof-treatment GFR was between 60 and 80 ml/min per 1.73 m 2 in eight and <60 ml/min per 1.73 m 2 in three. In the two remaining children a reduction in GFR of 2 and 7 ml/min per 1.73 m 2, resPectively, was observed. A Statistically significant improvement in GFR was found at 1-, 2-, and 4-year foll0w-up with resPect to the end-of-treatment GFR (p <0.05). However, there was no significant correlation between the end-of-treatment GFR and the increment subsequently achieved. Nevertheless, children with an end-of-treatment G F R of between 60 and 80 ml/min per 1.73 m z had a significantly better chance of regaining a GFR of 80 ml/min per 1.73 m 2 at a median follow-up time of 2 years 6 months than those whose end-oftreatment GFR was <60 ml/min per !.73 m 2 (p<0.01). No significant correlation was found between the increment in GFR at follow-up and the cumulative dose of cisplatin; however, the improvement in GFR has not yet reached a plateau. The children in whom repeat GFR measurements were made after treatment was stopped showed no further deterioration after the end of treatment and a gradual improvement. Infants. In 17 children the diagnosis was made before the age of 1 year; in 10 of them the end-of-treatment GFR was <80 ml/min per 1.73 m 2. All GFR results had improved at follow-up, with a median increase of 21 ml/min per 1.73 m 2 (range 2 to 55 ml/min per 1.73 m2). The follow-up GFRs of these 17 infants were compared with those of the other 23 children; no significant correlation with age was found. Renal tubular function, Of 21 c.hildren, six had hypomagnesemia at follow-up. No correlation was found with GFR. All 21 had a normal serum calcium concentration. Blood pressure. Only one child had transient hypertension, concomitantly with hydrocephalus and precocious puberty; the cause was not tumor related, and his most recent GFR was 76 ml/min per 1.73 m z. High-dose melphalan therapy. The follow-up GFRs of the 13 children who received high doses of melpha!an were compared with those of the children who had not received melphalan; no significant correlation was found. One child,
Partial reversibility o f cisplatin nephrotoxicity
533
however, had severe renal failure after receiving melphalan. At the end of treatment with cisplatin, he had a GFR of 125 ml/min per 1.73 m2; he subsequently received consolidation therapy with high intravenous doses of melphalan. On the day after receiving melphalan, his serum creatinine concentration rose above normal levels, and 3 days later his GFR was 15 ml/min per 1.73 m 2. His chronic renal failure has since been successfully treated by kidney transplantation. DISCUSSION None of the children treated with cisplatin died of nephrotoxic effects, but two of the survivors of advanced neuroblastoma have renal failure. We consider that the severity of the renal toxic reaction in one child was caused by melphalan and not cisplatin. More than half the children ended cisplatin treatment with a GFR by 51Cr-EDTA clearance of <80 m!/min per 1.73 m 2, which we assumed to be the lower limit of normal, although no data on normal rates as determined by this technique are available in children. Some recovery of renal function has been noted in adults treated with cisplatin.12, 13 According to Winberg 14 and Aperia et al., a5 GFR corrected for surface area, measured by endogenous creatinine clearance and inulin clearance, respectively, reaches adult values between the ages of 1 and 2 years. Of the 17 infants in the study, 10 had a GFR of <80 ml/min per 1.73 m e at the end of treatment, and all of them showed improvement at follow-up. However, we found no correlation between improvement in GFR at follow-up and age. That a number of children had isolated hypomagnesemia years after receiving cisplatin confirms the findings of Sheldon et al. 16 Our observations show that cisplatin nephrotoxicity varies widely in children. More than half the hildren have impaired glomerular function at the end of treatment, but almost all improve eventually. We recommend that GFR be measured by the 51Cr-EDTA or inulin method in all children receiving cisplatin, before treatment, at a cumulative dose of 450 mg/m 2, and thereafter at every additional dose of 100 mg/m 2. We now discontinue cisplatin if the GFR falls below 60 mt/min per 1.73 m 2 or is less than three fourths of the - 2 SD value for age in infants. 14 We prefer to give cisplatin as a 24-hour continuous infusion.172~ Serum aminoglycoside levels should be kept within the normal range. We believe that every child should be monitored with renal function studies until his or her GFR has recovered to 80 ml/min per 1.73 m 2. Magnesium supplementation may be necessary for many years. We thank Angela Bowmanfor invaluable information, William Wilson for technical assistance, and Hans Daniels for the statistical analysis.
534
Brock et al.
The Journal of Pediatrics April 1991
REFERENCES
1. Prestayko AW, D'Aoust JC, Issell BF, Crooke ST. Cisplatin (cis-diamminedichloroplatinum If). Cancer Treat Rev 1979; 6:17-39. 2. Krakhoff IH. Nephrotoxicity of cis-dichlorodiammineplatinum (II). Cancer Treat Rep 1979;63:1523-5. 3. Womer RB, Pritchard J, Barratt TM. Renal toxicity of cisplatin in children. J PEDJATR 1985;106:659-63. 4. Pritchard J, Germond S, Jones DH, de Kraker J. High dose melphalan improves the prognosis in advanced neuroblastoma: results of a randomised study of "ENSG-I?' Br J Cancer 1990;61:158. 5. Haas A, Anderson L, Lad T. The influence of aminoglyeosides on the nephrotoxicity of cis-diamminedichloroplatinum in cancer patients. J Infect Dis 1983;147:3631 6. Chantler C, Garnett ES, Parsons V, Veall N. Glomerular filtration rate with measurement in man by the single injection method using 51Cr EDTA. Clin Sci 1969;37:169-80. 7. Chantler C, Barratt TM. Estimation of glomerular filtration rate from plasma clearance of 51chromium edetic acid. Arch Dis Child 1972;47:613-7. 8. Clayton BE, Jenkins P, Round JM. Paediatric chemical Pathology: clinical tests and reference ranges. Oxford, England: Blackwell, 1980. 9. Park MK, Menard SH. Normative oscillometric blood pressure values in the first 5 years in an office setting. Am J Dis Child 1989;143:860-4. 10. Andr6 JL, Deschamps JP, Gueguen R, Jacques J. Etude de la tension art+rielle chez l'enfant et l'adolescent. Influence des modalit+s de mesure--distribution des valeurs fr6quentes et r6f6rence fi l'fige et fi la taille. Arch Mal Coeur 1981;74: 3-13. 11. Brock P, Pritchard J, Bellman S, Pinkerton CR. Ototoxicity of
12.
13.
14.
15.
16.
17.
18. 19.
20.
high-dose cis-platinum in children. Med Pediatr Oncol 1988;16:368-9. Hansen SW, Groth S, Daugaard G, Rossing N, Rorth M. Long-term effects on renal function and blood pressure of treatment with cisplatin, vinblastine, and bleomyein in patients with germ cell cancer. J Clin Oncol 1988;6:1728-31. Hamilton CR, Bliss JM, Horwieh A. The late effects of eisplatinum on renal function. Eur J Cancer Clin Oncol 1989;25:185-9. Winberg J. The 24-hour true endogenous creatinine clearance in infants and children without renal disease. Acta Paediatr 1959;48:443-52. Aperia A, Broberger O, Thodenius K. Development of renal control of salt and fluid homostasis during the first year of life. Acta Paediatr Scand 1975;64:393-8. Sheldon W, Welch R J, Bonham JR, Pearson ADJ, Craft AW. Hypomagnesaemia following treatment of childhood cancer with cisplatinum. Ann Clin Biochem 1987;24(suppl 1):85-8. Salem P, Khalyl M, Jarboury K, Hashimi L. Cis-diamminedichloroplatinum (II) by 5-day continuous infusion. A new dose schedule with minimal toxicity. Cancer 1984;53:83740. Posner MR, Ferrari L, Belliveau JF, et al. A phase I trial of continuous infusion cisplatin. Cancer 1987;59:15-8. Forastiere AA, Belliveau JF, Goren MP, Vogel WC, Posner MR, O'Leary GP. Pharmacokinetic and toxicity evaluation of five-day continuous infusion versus intermittent bolus cisdiamminedichloroplatinum (II) in head and neck cancer patient s. Cancer Res 1988;48:3869-74. Castello MA, Dominici C, Clerico A. A pilot study of 5-day continuous infusion of high-dose eisplatin and pulsed etoposide in childhood solid tumors. Am J Pediatr Hematol Oncol t988;10:103-8.
To evaluate the long-term renal toxicity of cisplatin, 40 children who had been without treatment at least 18 months (range 18 months to 7 years) were observed. In all the children, glomerular filtration rate (GFR) was estimated from the plasma clearance of chromium 51-1abeled ethylenediaminetetraacetic acid, both at the end of treatment and at a median follow-up of 2 years 6 months after treatment was stopped (range 18 months to 7 years). In 21 children, serum magnesium level was also measured at follow-up. Median a g e at diagnosis was 15 months (range 13 days to 13 years 8 months), and median cumulative dose of cisplatin was 500 m g / m 2 (range 120 to 1860 mg/m2). In 22 of 24 children with an end-oftreatment GFR Of <80 ml/min per 1.73 m 2, the median improvement in GFR at follow-up was 22 ml/mln per 1.73 m 2 (range 2 to 56 ml/min per 1.73 m2). Hypomagnesemia was found in 6 of 21 children and was Independent of GFR. No significant correlation was found between improvement in renal function and total cisplatin dose, age, gender, tumor type, or associated nephrotoxlc medication. We conclude that most children have some recovery from cisplatin glomerular toxicity, especially if d a m a g e is not severe, but that h y p o m a g n e s e m i a may persist. (J PEDIATR1991;118:531-4)
Nephrotoxicity is the most important dose-limiting side effect of cisplatin. 1, 2 In 1985, Womer et al. 3 showed that creatinine clearance and serum creatinine concentration were inadequate measures of renal toxic effects in children treated with cisplatin. They used estimated glomerular filtration rate from the plasma clearance of chromium 5 I-labeled ethylenediaminetetraacetic acid and found reduced renal function in most cases. The aim of this follow-up study was to determine whether, in patients receiving cisplatin, there is recovery of renal function with time. The effects of
other nephrotoxic agents administered to these children as part of their treatment were included in the analysis. METHODS The study population consisted of 40 children treated with cisplatin, 60 to 100 m g / m 2 per course, as a component of multiagent chemotherapy. Between 1979 and 1988, a total of 164 children with a solid tumor received such treatment at our institution. No child died of nephrotoxic effects, 51Cr-EDTA
Supported in part by the British Council, Het Belgische Werk Tegen Kanker, and the Leukaemia Research Fund (Dr. Brock), and by the Imperial Cancer Research Fund (Dr. Pritchard). Submitted for publication Aug. I, 1990; accepted Oct. 29, t990. Reprint requests: Penelope R, Brock, MD, Paediatric Oncology, University Hospital Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium. *Current address: Paediatric Oncology, Aghias Sofias 9, 546 23 Thessaloniki, Greece.
9/20/26388
GFR
chromium 5 I-labeled ethylenediaminetet raacetic acid Glomerular filtration rate
and there were 55 survivors in January 1990. Parental consent was obtained for follow-up G F R measurements in 40 children. There was no reason to believe that the remaining 15 differed significantly in any respect. The diagnoses of the study population were neuroblastoma (27 patients), germ cell tumor (8), hepatoblastoma (3), and osteogenic sarcoma (2). Median age at diagnosis was 15 months (range 13 days
531
532
Brock et al.
The Journal of Pediatrics April 1991
n=23
---q
n=15
n=2
II
m
End-of-treatment
[~1
GFR
F o l l o w - u p GFR
GFR>80ml/min per 173mz GFR 60-80mr/rain per 1.73m2 GFR<60mt/min per 1.73m2
Figure. Glomerular filtration rate (GFR) measured by plasma clearance of StCr-EDTA in 40 children treated with cisplatin. Each line represents one patient. Median follow-up time from end of treatment, 2 years 6 months. Asterisk indicates patient with acute melphalan toxic effects.
to 13 years 8 months). There were 25 boys. Cisplatin was given intravenously for 15 to 30 minutes with hydration at 3 L / m 2 every 24 hours and diuresis with mannitol. The interval between courses was 3 to 4 weeks. The median cumulative dose of cisplatin was 500 m g / m 2 (range 120 to 1860 mg/m2). Chemotherapy regimens were as follows: neuroblastoma---cisplatin, cyclophosphamide, vincristine, teniposide-etoposide, and high-dose melphalan; germ cell tumor--cisplatin, Neomycin, and vinblastine-etoposide;hepatoblastoma--cisplatin and doxorubicin; and osteosarcoma---cisplatin, doxorubicin, and methotrexate. Thirteen children received high doses of melphalan as consolidation treatment for stage 4 neuroblastoma.4 Gentamicin was given to 18 children for a median of 17 days (range 7 to 34 days); serum levels were monitored, and in
no case did they reach toxic levels. 5 A further eight children also received at least one dose (1 mg/m 2) of furosemide. No child had a nephrectomy or underwent flank irradiation. The GFR was estimated during treatment and at follow-up from the plasma clearance of 51Cr-EDTA, by using a single-compartment analysis.6,7 Additional measurements were carried out in 27 children; the number of additional measurements ranged from one to five (median two). Each child's GFR was measured at least 18 months after treatment was stopped. When later measurements were undertaken, the most recent figure was used in analyzing the results. The median longest follow-up time after treatment was 2 years 6 months (range 18 months to 7 years). Serum magnesium and calcium values were obtained in 21 children at their most recent follow-up. Normal electrolyte values were taken from Clayton et al. 8 A gradual fall in concentration of both magnesium and calcium occurs during early childhood, - 2 SD from the mean being 0.75 mmol/L at 1 month and 0.7 mmol/L at 5 years for magnesium and 2.4 mmol/L at 1 month and 2.225 mmol/L at 5 years for calcium. Blood pressure was assessed according to the age-specific percentiles of Park and Menard 9 in the younger children, and on the basis of age and length according to Andre et al} ~ in the older children. The Spearman rank test was used to evaluate the relationship between mutually dependent variables, the Mann-Whitney U test for differences between medians, and the sign test for differences between matched samples. Two-tailed probabilities were calculated. For differences between proportions, the chi-square test was used. The null hypothesis was rejected at values of p <0.05. RESULTS
End-of-treatment GFR. Median end-of-treatment GFR was 74 ml/min per 1.73 m 2 (range 13 to 184 ml/min per 1.73 m2): 16 children had a GFR of >80 ml/min per 1.73 m 2, 13 had a GFR of between 60 and 80 ml/min per 1.73 m z, and 11 had a GFR of <60 ml/min per 1.73 m 2. No significant correlation was found between GFR and total cisplatin dose, age, gender, tumor type, or associated nephrotoxic medication. One child, treated early in the series (1980), had a sudden drop in GFR, from 72 ml/min per 1.73 m 2 after six courses of cisplatin (700 m g / m 2) to 15 ml/min per 1.73 m 2 after a further two courses, without apparent reason. In this child grade 4 ototoxic effects also developed. 11 At follow-up 9 years from diagnosis, his height and weight were at the 10th and 3rd percentiles, respectively, and his blood pressure was normal. His condition is managed conservatively, and his latest GFR was 27 ml/min per 1.73 m 2. Recovery of glomerular renal function. Median GFR at
Volume 118 Number 4, Part 1
follow-up was 90 ml/min per 1.73 m 2 (range 27 to 135 ml/ min per 1.73 m2). Of the 16 children with an end-of-treatment GFR of >80 ml/min per 1.73 m 2, 12 maintained a GFR of >80 ml/min per 1.73 m 2. The GFR fell to 60 to 80 ml/min per 1.73 m 2 in three children, and one child had progressive renal failure immediately after receiving highdose melphalan therapy. Of the 24 children who had a GFR of <80 ml/min per 1.73 m 2 at the end of treatment, all but two children showed recovery to a variable extent (Figure). The median change in GFR was +22 ml/min per 1.73 m 2 (range - 7 to +56 ml/min per 1.73 m2). A recovery GFR of >80 ml/min per 1.73 m 2 was measured in 11 children. Of these 11, the endof-treatment GFR was between 60 and 80 ml/min per 1.73 m 2 in eight and <60 ml/min per 1.73 m 2 in three. In the two remaining children a reduction in GFR of 2 and 7 ml/min per 1.73 m 2, resPectively, was observed. A Statistically significant improvement in GFR was found at 1-, 2-, and 4-year foll0w-up with resPect to the end-of-treatment GFR (p <0.05). However, there was no significant correlation between the end-of-treatment GFR and the increment subsequently achieved. Nevertheless, children with an end-of-treatment G F R of between 60 and 80 ml/min per 1.73 m z had a significantly better chance of regaining a GFR of 80 ml/min per 1.73 m 2 at a median follow-up time of 2 years 6 months than those whose end-oftreatment GFR was <60 ml/min per !.73 m 2 (p<0.01). No significant correlation was found between the increment in GFR at follow-up and the cumulative dose of cisplatin; however, the improvement in GFR has not yet reached a plateau. The children in whom repeat GFR measurements were made after treatment was stopped showed no further deterioration after the end of treatment and a gradual improvement. Infants. In 17 children the diagnosis was made before the age of 1 year; in 10 of them the end-of-treatment GFR was <80 ml/min per 1.73 m 2. All GFR results had improved at follow-up, with a median increase of 21 ml/min per 1.73 m 2 (range 2 to 55 ml/min per 1.73 m2). The follow-up GFRs of these 17 infants were compared with those of the other 23 children; no significant correlation with age was found. Renal tubular function, Of 21 c.hildren, six had hypomagnesemia at follow-up. No correlation was found with GFR. All 21 had a normal serum calcium concentration. Blood pressure. Only one child had transient hypertension, concomitantly with hydrocephalus and precocious puberty; the cause was not tumor related, and his most recent GFR was 76 ml/min per 1.73 m z. High-dose melphalan therapy. The follow-up GFRs of the 13 children who received high doses of melpha!an were compared with those of the children who had not received melphalan; no significant correlation was found. One child,
Partial reversibility o f cisplatin nephrotoxicity
533
however, had severe renal failure after receiving melphalan. At the end of treatment with cisplatin, he had a GFR of 125 ml/min per 1.73 m2; he subsequently received consolidation therapy with high intravenous doses of melphalan. On the day after receiving melphalan, his serum creatinine concentration rose above normal levels, and 3 days later his GFR was 15 ml/min per 1.73 m 2. His chronic renal failure has since been successfully treated by kidney transplantation. DISCUSSION None of the children treated with cisplatin died of nephrotoxic effects, but two of the survivors of advanced neuroblastoma have renal failure. We consider that the severity of the renal toxic reaction in one child was caused by melphalan and not cisplatin. More than half the children ended cisplatin treatment with a GFR by 51Cr-EDTA clearance of <80 m!/min per 1.73 m 2, which we assumed to be the lower limit of normal, although no data on normal rates as determined by this technique are available in children. Some recovery of renal function has been noted in adults treated with cisplatin.12, 13 According to Winberg 14 and Aperia et al., a5 GFR corrected for surface area, measured by endogenous creatinine clearance and inulin clearance, respectively, reaches adult values between the ages of 1 and 2 years. Of the 17 infants in the study, 10 had a GFR of <80 ml/min per 1.73 m e at the end of treatment, and all of them showed improvement at follow-up. However, we found no correlation between improvement in GFR at follow-up and age. That a number of children had isolated hypomagnesemia years after receiving cisplatin confirms the findings of Sheldon et al. 16 Our observations show that cisplatin nephrotoxicity varies widely in children. More than half the hildren have impaired glomerular function at the end of treatment, but almost all improve eventually. We recommend that GFR be measured by the 51Cr-EDTA or inulin method in all children receiving cisplatin, before treatment, at a cumulative dose of 450 mg/m 2, and thereafter at every additional dose of 100 mg/m 2. We now discontinue cisplatin if the GFR falls below 60 mt/min per 1.73 m 2 or is less than three fourths of the - 2 SD value for age in infants. 14 We prefer to give cisplatin as a 24-hour continuous infusion.172~ Serum aminoglycoside levels should be kept within the normal range. We believe that every child should be monitored with renal function studies until his or her GFR has recovered to 80 ml/min per 1.73 m 2. Magnesium supplementation may be necessary for many years. We thank Angela Bowmanfor invaluable information, William Wilson for technical assistance, and Hans Daniels for the statistical analysis.
534
Brock et al.
The Journal of Pediatrics April 1991
REFERENCES
1. Prestayko AW, D'Aoust JC, Issell BF, Crooke ST. Cisplatin (cis-diamminedichloroplatinum If). Cancer Treat Rev 1979; 6:17-39. 2. Krakhoff IH. Nephrotoxicity of cis-dichlorodiammineplatinum (II). Cancer Treat Rep 1979;63:1523-5. 3. Womer RB, Pritchard J, Barratt TM. Renal toxicity of cisplatin in children. J PEDJATR 1985;106:659-63. 4. Pritchard J, Germond S, Jones DH, de Kraker J. High dose melphalan improves the prognosis in advanced neuroblastoma: results of a randomised study of "ENSG-I?' Br J Cancer 1990;61:158. 5. Haas A, Anderson L, Lad T. The influence of aminoglyeosides on the nephrotoxicity of cis-diamminedichloroplatinum in cancer patients. J Infect Dis 1983;147:3631 6. Chantler C, Garnett ES, Parsons V, Veall N. Glomerular filtration rate with measurement in man by the single injection method using 51Cr EDTA. Clin Sci 1969;37:169-80. 7. Chantler C, Barratt TM. Estimation of glomerular filtration rate from plasma clearance of 51chromium edetic acid. Arch Dis Child 1972;47:613-7. 8. Clayton BE, Jenkins P, Round JM. Paediatric chemical Pathology: clinical tests and reference ranges. Oxford, England: Blackwell, 1980. 9. Park MK, Menard SH. Normative oscillometric blood pressure values in the first 5 years in an office setting. Am J Dis Child 1989;143:860-4. 10. Andr6 JL, Deschamps JP, Gueguen R, Jacques J. Etude de la tension art+rielle chez l'enfant et l'adolescent. Influence des modalit+s de mesure--distribution des valeurs fr6quentes et r6f6rence fi l'fige et fi la taille. Arch Mal Coeur 1981;74: 3-13. 11. Brock P, Pritchard J, Bellman S, Pinkerton CR. Ototoxicity of
12.
13.
14.
15.
16.
17.
18. 19.
20.
high-dose cis-platinum in children. Med Pediatr Oncol 1988;16:368-9. Hansen SW, Groth S, Daugaard G, Rossing N, Rorth M. Long-term effects on renal function and blood pressure of treatment with cisplatin, vinblastine, and bleomyein in patients with germ cell cancer. J Clin Oncol 1988;6:1728-31. Hamilton CR, Bliss JM, Horwieh A. The late effects of eisplatinum on renal function. Eur J Cancer Clin Oncol 1989;25:185-9. Winberg J. The 24-hour true endogenous creatinine clearance in infants and children without renal disease. Acta Paediatr 1959;48:443-52. Aperia A, Broberger O, Thodenius K. Development of renal control of salt and fluid homostasis during the first year of life. Acta Paediatr Scand 1975;64:393-8. Sheldon W, Welch R J, Bonham JR, Pearson ADJ, Craft AW. Hypomagnesaemia following treatment of childhood cancer with cisplatinum. Ann Clin Biochem 1987;24(suppl 1):85-8. Salem P, Khalyl M, Jarboury K, Hashimi L. Cis-diamminedichloroplatinum (II) by 5-day continuous infusion. A new dose schedule with minimal toxicity. Cancer 1984;53:83740. Posner MR, Ferrari L, Belliveau JF, et al. A phase I trial of continuous infusion cisplatin. Cancer 1987;59:15-8. Forastiere AA, Belliveau JF, Goren MP, Vogel WC, Posner MR, O'Leary GP. Pharmacokinetic and toxicity evaluation of five-day continuous infusion versus intermittent bolus cisdiamminedichloroplatinum (II) in head and neck cancer patient s. Cancer Res 1988;48:3869-74. Castello MA, Dominici C, Clerico A. A pilot study of 5-day continuous infusion of high-dose eisplatin and pulsed etoposide in childhood solid tumors. Am J Pediatr Hematol Oncol t988;10:103-8.