- Email: [email protected]
Urinary Loss of Erythropoietin after Intravenous Versus Subcutaneous Epoetin-beta in Preterm Infants
CLINICAL AND LABORATORY OBSERVATIONS
Urinary Loss of Erythropoietin after Intravenous Versus Subcutaneous Epoetin-beta in Preterm Infants JULIANE LANGER, BSC, MICHAEL OBLADEN, MD,
AND
CHRISTOF DAME, MD
Hematopoietic and non-hematopoietic effects of recombinant erythropoietin (Epo) given to preterm infants are controversially discussed. Because renal loss of Epo was significantly higher after intravenous versus subcutaneous Epoetin-beta administration, we suggest a reconsideration of whether subcutaneous recombinant Epo is more efficient and safer because of lower peaks of circulating Epo. (J Pediatr 2008;152:728-30)
ow erythropoietin (Epo) levels are the rationale for the treatment of anemia of prematurity with recombinant human Epo (rhEpo). Although early rhEpo treatment reduces the frequency and volume of blood transfusions, the recent Cochrane review indicates a very limited overall benefit, because most infants already underwent transfusion before entry into Epo trials.1 Furthermore, a significantly higher incidence of retinopathy stage ⱖ3 was observed in infants treated with rhEpo.1 Thus, the question arises whether peaks in circulating Epo or urinary loss of Epo contribute to these issues. Although some data on the pharmacokinetics of rhEpo in the circulation or cerebrospinal fluid have been reported,2 the question on its urinary loss after intravenous (IV) or subcutaneous (SC) treatment has not been addressed systematically.
L
METHODS This longitudinal study included 20 very low birth weight (VLBW) infants born at a gestational age ⬍31 ⫹ 0 weeks. Treatment with 250 IU/kg Epoetin-beta (NeoRecormon, diluted in the provided injection fluid; Roche) 3 times each week was initiated by day 11 (median) of age. Epoetin-beta was given as a bolus IV or by SC injection when IV access was no longer needed. Written parental consent for rhEpo treatment and urine sampling was obtained. The study was approved by the institutional review board (EA2-063-06). Enteral iron (3 mg/kg/day) was given from the beginning of rhEpo treatment and increased to as much as 9 mg/kg/day according to transferrin saturation. RhEpo treatment was interrupted when the transferrin saturation was ⬍30% or stopped when the development of hemangioma was observed, at 36 ⫹ 6 weeks of postconceptional age or earlier when infants were discharged from the hospital. Urine was collected in small plastic bags every 14 days before and 4 and 8 hours after rhEpo treatment. Epo concentrations were measured in duplicate by using the Quantikine Human Epo Immunoassay (R&D Systems; lower detection limit, 2.5 mU/mL). For statistical analysis, the Mann-Whitney U test was applied. A P value ⬍.05 (2-sided) was considered to be statistically significant.
RESULTS Before treatment, Epo was detected in 11 of 53 urine specimens (median, 0 mU/mL; range, 0-59 mU/mL). After 4 hours, Epo was detected in 32 of 54 samples (IV, 15/23; SC, 17/31). Urinary Epo concentrations were significantly higher after IV (median, 38.7 mU/mL; range, 0-376.5 mU/mL) than after SC rhEpo (median, 5.5 mU/mL; range, 0-200 mU/mL; P ⬍ .05). After 8 hours, Epo was detected in 31 of 52 specimens (IV, 17/23; SC, 14/29). Again, Epo concentrations were also significantly higher after IV (median, 57 mU/mL; range, 0-305 mU/mL) than after SC rhEpo (median, 0 mU/mL; range, 0-81 mU/mL; P ⬍ .05; Figure). If only Epo-positive specimens were considered, mean Epo concentrations were 3-fold higher after IV rhEpo (4 hours IV 138.6 mU/mL versus SC 46.6 mU/mL [P ⬍ .05]; 8 hours IV 121.9 versus SC 40.5 mU/mL [P ⬍ .05]). Epo EpoR IV
728
Erythropoietin Erythropoietin receptor Intravenous
rhEpo SC VLBW
Recombinant human erythropoietin Subcutaneous Very low birth weight
From Klinik für Neonatologie, Campus Virchow-Klinikum, Charité – Universitätsmedizin, Berlin, Germany. Submitted for publication Oct 16, 2007; last revision received Nov 19, 2007; accepted Jan 18, 2008. Reprint requests: Christof Dame, MD, Department of Neonatology, Campus Virchow-Klinikum, Charité – Universitätsmedizin Berlin, Augustenburger Platz 1, D13353 Berlin, Germany. E-mail: christof. [email protected]. 0022-3476/$ - see front matter Copyright © 2008 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2008.01.026
Table. Erythropoietin concentrations in 46 urine specimens from 19 infants that were taken 4 and 8 hours after intravenous treatment with 250 IU/kg recombinant human erythropoietin Week
Figure. Epo concentrations in urine specimens of VLBW infants taken before, 4 hours after, and 8 hours after IV (grey boxes) or SC treatment (light boxes) with 250 U/kg Epoetin-beta. Data are presented as box plots with the median and the 25th and 75th percentiles defining the box. Whiskers indicate the 10th and 90th percentile. Single data points (*) that lie outside the 10th and 90th percentile are also shown.
Epo concentrations were higher in urine specimens taken when the infants were at a gestational age ⬍31 ⫾ 0 weeks (Table) and did not correlate with serum creatinine or creatinine clearance (not shown). Considering the urine volume excreted within the first 8 hours after rhEpo, the maximal calculated amount of renal Epo loss was 4.5% after IV treatment versus 0.75% after SC treatment within that period.
DISCUSSION A significantly higher urinary Epo excretion was reported after IV treatment versus SC treatment of rhEpo in a preterm pair of twins.3 We systematically quantified the urinary loss of Epo for 8 hours after treatment. In contrast to Warwood et al,4 who analyzed urinary Epo concentrations after treatment with 4 g/kg darbepoietin (a higher glycosylated rhEpo derivate that equals 1600 IU rhEpo/kg), we detected significant amounts of Epo 4 and 8 hours after IV/SC treatment. When we tested cotton-containing materials for urine sampling as described,4 rhEpo could not be recovered (not shown), likely because of binding of highly glycosylated rhEpo to cotton-containing collecting material. Our study was not designed to prove whether IV or SC rhEpo treatment is more efficient. This question is still not answered, because, in most rhEpo trials, IV treatment is initiated and then continued by means of the SC route.1 In VLBW infants, as much as 5% of rhEpo is lost by urinary excretion within the first 8 hours after treatment, suggesting
Samples Median Range Quartiles (n) (mU/mL) (mU/mL) (mU/mL)
25 ⫹ 0 to 27 ⫹ 6
18
105.5
28 ⫹ 0 to 30 ⫹ 6
20
69.4
0-376.5
31 ⫹ 0 to 33 ⫹ 6
8
3.8
0-91.6
0-304
25th, 75th, 25th, 75th, 25th, 75th,
19.6 171.2 0.4 190.4 0 21
that SC treatment is more effective. However, urinary Epo concentrations may differ when rhEpo is given as bolus injection or short infusion. The question arises whether SC rhEpo is safer. Recently, a relation between Epo treatment and retinopathy stage ⱖ3 was identified in the Cochrane review,1 although the individual randomized, controlled trials did not identify that problem. In animals and human adults, circulating Epo rapidly peaks following IV treatment, but not after SC rhEpo treatment.5 This phenomenon may be reflected by higher urinary Epo concentrations in VLBW infants who undergo IV treatment. As postulated for other growth factors, peaks in circulating Epo may harm vascular development in the retina more than moderate but more constant elevations. Our data may be important for future concepts of the use of rhEpo as a neuroprotectant.6 To achieve significant neuroprotection after hypoxic-ischemic brain injury, rhEpo must be given in high doses, shortly after the onset of injury.2 Only higher doses of rhEpo result within a crucial time window in appropriate Epo concentrations in the cerebrospinal fluid.5 Even assuming that renal Epo loss may be low in near-term infants, confirmation is required because rhEpo dosages for neuroprotection after acute brain injury are 20fold higher. Neuroprotection with rhEpo in VLBW infants may require a different strategy. In preterm infants treated with rhEpo for preventing red blood cell transfusions, a post hoc analysis showed higher developmental index scores only in infants with circulating Epo concentrations ⱖ500 mU/mL.7,8 Thus, it is appealing to give rhEpo for neuroprotection in VLBW infants very early and repetitively, but in lower dosages than after asphyxia. Such a strategy may be important because neurotoxic effects of extremely high doses of rhEpo were identified in vitro.9 In general, subcutaneous administration might prevent peaks in circulating Epo and accumulation in tissues with high EpoR expression. We thank Peter Martus (Institut für Biometrie und klinische Biometrie, Charité – Universitätsmedizin Berlin) and Boris Metze (Klinik für Neonatologie, CVK, Charité – Universitätsmedizin Berlin) for supporting the analysis of the data.
Urinary Loss of Erythropoietin after Intravenous Versus Subcutatneous Epoetin-beta in Preterm Infants
729
REFERENCES 1. Ohlsson A, Aher S. Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants. Cochrane Database Syst Rev 2006;3: CD004863. 2. Juul S, Felderhoff-Mueser U. Epo and other hematopoietic factors. Semin Fetal Neonatal Med 2007;12:250-8. 3. Buhrer C, Obladen M, Maier R, Muller C. Urinary losses of recombinant erythropoietin in preterm infants. J Pediatr 2003;142:452-3. 4. Warwood TL, Ohls RK, Lambert DK, Leve EA, Veng-Pedersen P, Christensen RD. Urinary excretion of darbepoetin after intravenous versus subcutaneous administration to preterm neonates. J Perinatol 2006;26:636-9. 5. Xenocostas A, Cheung WK, Farrell F, Zakszewski C, Kelley M, Lutynski A, et al. The pharmacokinetics of erythropoietin in the cerebrospinal fluid after intravenous
730
Langer, Obladen, and Dame
administration of recombinant human erythropoietin. Eur J Clin Pharmacol 2005; 61:189-95. 6. Dame C, Fahnenstich H. Don’t give up on erythropoietin as a neuroprotective agent. Pediatrics 2005;116:521-2. 7. Ohls RK, Ehrenkranz RA, Das A, Dusick AM, Yolton K, Romano E, et al. Neurodevelopmental outcome and growth at 18 to 22 months’ corrected age in extremely low birth weight infants treated with early erythropoietin and iron. Pediatrics 2004;114:1287-91. 8. Bierer R, Peceny MC, Hartenberger CH, Ohls RK. Erythropoietin concentrations and neurodevelopmental outcome in preterm infants. Pediatrics 2006;118:e635-40. 9. Weber A, Dzietko M, Berns M, Felderhoff-Mueser U, Heinemann U, Maier RF, et al. Neuronal damage after moderate hypoxia and erythropoietin. Neurobiol Dis 2005;20:594-600.
The Journal of Pediatrics • May 2008
Urinary Loss of Erythropoietin after Intravenous Versus Subcutaneous Epoetin-beta in Preterm Infants JULIANE LANGER, BSC, MICHAEL OBLADEN, MD,
AND
CHRISTOF DAME, MD
Hematopoietic and non-hematopoietic effects of recombinant erythropoietin (Epo) given to preterm infants are controversially discussed. Because renal loss of Epo was significantly higher after intravenous versus subcutaneous Epoetin-beta administration, we suggest a reconsideration of whether subcutaneous recombinant Epo is more efficient and safer because of lower peaks of circulating Epo. (J Pediatr 2008;152:728-30)
ow erythropoietin (Epo) levels are the rationale for the treatment of anemia of prematurity with recombinant human Epo (rhEpo). Although early rhEpo treatment reduces the frequency and volume of blood transfusions, the recent Cochrane review indicates a very limited overall benefit, because most infants already underwent transfusion before entry into Epo trials.1 Furthermore, a significantly higher incidence of retinopathy stage ⱖ3 was observed in infants treated with rhEpo.1 Thus, the question arises whether peaks in circulating Epo or urinary loss of Epo contribute to these issues. Although some data on the pharmacokinetics of rhEpo in the circulation or cerebrospinal fluid have been reported,2 the question on its urinary loss after intravenous (IV) or subcutaneous (SC) treatment has not been addressed systematically.
L
METHODS This longitudinal study included 20 very low birth weight (VLBW) infants born at a gestational age ⬍31 ⫹ 0 weeks. Treatment with 250 IU/kg Epoetin-beta (NeoRecormon, diluted in the provided injection fluid; Roche) 3 times each week was initiated by day 11 (median) of age. Epoetin-beta was given as a bolus IV or by SC injection when IV access was no longer needed. Written parental consent for rhEpo treatment and urine sampling was obtained. The study was approved by the institutional review board (EA2-063-06). Enteral iron (3 mg/kg/day) was given from the beginning of rhEpo treatment and increased to as much as 9 mg/kg/day according to transferrin saturation. RhEpo treatment was interrupted when the transferrin saturation was ⬍30% or stopped when the development of hemangioma was observed, at 36 ⫹ 6 weeks of postconceptional age or earlier when infants were discharged from the hospital. Urine was collected in small plastic bags every 14 days before and 4 and 8 hours after rhEpo treatment. Epo concentrations were measured in duplicate by using the Quantikine Human Epo Immunoassay (R&D Systems; lower detection limit, 2.5 mU/mL). For statistical analysis, the Mann-Whitney U test was applied. A P value ⬍.05 (2-sided) was considered to be statistically significant.
RESULTS Before treatment, Epo was detected in 11 of 53 urine specimens (median, 0 mU/mL; range, 0-59 mU/mL). After 4 hours, Epo was detected in 32 of 54 samples (IV, 15/23; SC, 17/31). Urinary Epo concentrations were significantly higher after IV (median, 38.7 mU/mL; range, 0-376.5 mU/mL) than after SC rhEpo (median, 5.5 mU/mL; range, 0-200 mU/mL; P ⬍ .05). After 8 hours, Epo was detected in 31 of 52 specimens (IV, 17/23; SC, 14/29). Again, Epo concentrations were also significantly higher after IV (median, 57 mU/mL; range, 0-305 mU/mL) than after SC rhEpo (median, 0 mU/mL; range, 0-81 mU/mL; P ⬍ .05; Figure). If only Epo-positive specimens were considered, mean Epo concentrations were 3-fold higher after IV rhEpo (4 hours IV 138.6 mU/mL versus SC 46.6 mU/mL [P ⬍ .05]; 8 hours IV 121.9 versus SC 40.5 mU/mL [P ⬍ .05]). Epo EpoR IV
728
Erythropoietin Erythropoietin receptor Intravenous
rhEpo SC VLBW
Recombinant human erythropoietin Subcutaneous Very low birth weight
From Klinik für Neonatologie, Campus Virchow-Klinikum, Charité – Universitätsmedizin, Berlin, Germany. Submitted for publication Oct 16, 2007; last revision received Nov 19, 2007; accepted Jan 18, 2008. Reprint requests: Christof Dame, MD, Department of Neonatology, Campus Virchow-Klinikum, Charité – Universitätsmedizin Berlin, Augustenburger Platz 1, D13353 Berlin, Germany. E-mail: christof. [email protected]. 0022-3476/$ - see front matter Copyright © 2008 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2008.01.026
Table. Erythropoietin concentrations in 46 urine specimens from 19 infants that were taken 4 and 8 hours after intravenous treatment with 250 IU/kg recombinant human erythropoietin Week
Figure. Epo concentrations in urine specimens of VLBW infants taken before, 4 hours after, and 8 hours after IV (grey boxes) or SC treatment (light boxes) with 250 U/kg Epoetin-beta. Data are presented as box plots with the median and the 25th and 75th percentiles defining the box. Whiskers indicate the 10th and 90th percentile. Single data points (*) that lie outside the 10th and 90th percentile are also shown.
Epo concentrations were higher in urine specimens taken when the infants were at a gestational age ⬍31 ⫾ 0 weeks (Table) and did not correlate with serum creatinine or creatinine clearance (not shown). Considering the urine volume excreted within the first 8 hours after rhEpo, the maximal calculated amount of renal Epo loss was 4.5% after IV treatment versus 0.75% after SC treatment within that period.
DISCUSSION A significantly higher urinary Epo excretion was reported after IV treatment versus SC treatment of rhEpo in a preterm pair of twins.3 We systematically quantified the urinary loss of Epo for 8 hours after treatment. In contrast to Warwood et al,4 who analyzed urinary Epo concentrations after treatment with 4 g/kg darbepoietin (a higher glycosylated rhEpo derivate that equals 1600 IU rhEpo/kg), we detected significant amounts of Epo 4 and 8 hours after IV/SC treatment. When we tested cotton-containing materials for urine sampling as described,4 rhEpo could not be recovered (not shown), likely because of binding of highly glycosylated rhEpo to cotton-containing collecting material. Our study was not designed to prove whether IV or SC rhEpo treatment is more efficient. This question is still not answered, because, in most rhEpo trials, IV treatment is initiated and then continued by means of the SC route.1 In VLBW infants, as much as 5% of rhEpo is lost by urinary excretion within the first 8 hours after treatment, suggesting
Samples Median Range Quartiles (n) (mU/mL) (mU/mL) (mU/mL)
25 ⫹ 0 to 27 ⫹ 6
18
105.5
28 ⫹ 0 to 30 ⫹ 6
20
69.4
0-376.5
31 ⫹ 0 to 33 ⫹ 6
8
3.8
0-91.6
0-304
25th, 75th, 25th, 75th, 25th, 75th,
19.6 171.2 0.4 190.4 0 21
that SC treatment is more effective. However, urinary Epo concentrations may differ when rhEpo is given as bolus injection or short infusion. The question arises whether SC rhEpo is safer. Recently, a relation between Epo treatment and retinopathy stage ⱖ3 was identified in the Cochrane review,1 although the individual randomized, controlled trials did not identify that problem. In animals and human adults, circulating Epo rapidly peaks following IV treatment, but not after SC rhEpo treatment.5 This phenomenon may be reflected by higher urinary Epo concentrations in VLBW infants who undergo IV treatment. As postulated for other growth factors, peaks in circulating Epo may harm vascular development in the retina more than moderate but more constant elevations. Our data may be important for future concepts of the use of rhEpo as a neuroprotectant.6 To achieve significant neuroprotection after hypoxic-ischemic brain injury, rhEpo must be given in high doses, shortly after the onset of injury.2 Only higher doses of rhEpo result within a crucial time window in appropriate Epo concentrations in the cerebrospinal fluid.5 Even assuming that renal Epo loss may be low in near-term infants, confirmation is required because rhEpo dosages for neuroprotection after acute brain injury are 20fold higher. Neuroprotection with rhEpo in VLBW infants may require a different strategy. In preterm infants treated with rhEpo for preventing red blood cell transfusions, a post hoc analysis showed higher developmental index scores only in infants with circulating Epo concentrations ⱖ500 mU/mL.7,8 Thus, it is appealing to give rhEpo for neuroprotection in VLBW infants very early and repetitively, but in lower dosages than after asphyxia. Such a strategy may be important because neurotoxic effects of extremely high doses of rhEpo were identified in vitro.9 In general, subcutaneous administration might prevent peaks in circulating Epo and accumulation in tissues with high EpoR expression. We thank Peter Martus (Institut für Biometrie und klinische Biometrie, Charité – Universitätsmedizin Berlin) and Boris Metze (Klinik für Neonatologie, CVK, Charité – Universitätsmedizin Berlin) for supporting the analysis of the data.
Urinary Loss of Erythropoietin after Intravenous Versus Subcutatneous Epoetin-beta in Preterm Infants
729
REFERENCES 1. Ohlsson A, Aher S. Early erythropoietin for preventing red blood cell transfusion in preterm and/or low birth weight infants. Cochrane Database Syst Rev 2006;3: CD004863. 2. Juul S, Felderhoff-Mueser U. Epo and other hematopoietic factors. Semin Fetal Neonatal Med 2007;12:250-8. 3. Buhrer C, Obladen M, Maier R, Muller C. Urinary losses of recombinant erythropoietin in preterm infants. J Pediatr 2003;142:452-3. 4. Warwood TL, Ohls RK, Lambert DK, Leve EA, Veng-Pedersen P, Christensen RD. Urinary excretion of darbepoetin after intravenous versus subcutaneous administration to preterm neonates. J Perinatol 2006;26:636-9. 5. Xenocostas A, Cheung WK, Farrell F, Zakszewski C, Kelley M, Lutynski A, et al. The pharmacokinetics of erythropoietin in the cerebrospinal fluid after intravenous
730
Langer, Obladen, and Dame
administration of recombinant human erythropoietin. Eur J Clin Pharmacol 2005; 61:189-95. 6. Dame C, Fahnenstich H. Don’t give up on erythropoietin as a neuroprotective agent. Pediatrics 2005;116:521-2. 7. Ohls RK, Ehrenkranz RA, Das A, Dusick AM, Yolton K, Romano E, et al. Neurodevelopmental outcome and growth at 18 to 22 months’ corrected age in extremely low birth weight infants treated with early erythropoietin and iron. Pediatrics 2004;114:1287-91. 8. Bierer R, Peceny MC, Hartenberger CH, Ohls RK. Erythropoietin concentrations and neurodevelopmental outcome in preterm infants. Pediatrics 2006;118:e635-40. 9. Weber A, Dzietko M, Berns M, Felderhoff-Mueser U, Heinemann U, Maier RF, et al. Neuronal damage after moderate hypoxia and erythropoietin. Neurobiol Dis 2005;20:594-600.
The Journal of Pediatrics • May 2008