Single-Dose Darbepoetin Administration to Anemic Preterm Neonates
Journal of Perinatology(2005)
摘要
In 2004, Intermountain Health Care, which operates 18 hospitals in Utah and Idaho, suggested that its physicians consider switching their patients who require recombinant erythropoietin (rEpo) to darbepoetin, in order to improve cost effectiveness for the health system and its patients.1, 2 Darbepoetin was created by modifying five amino acids of the original rEpo protein in order to generate two additional carbohydrate binding sites, thereby significantly increasing the t½ and effectiveness.3 Information is available to guide physicians who wish to switch their adult patients from rEpo to darbepoetin,4 but no such guidelines and no studies on safety, efficacy, or pharmacokinetics have been published for neonates. Ohls and Dai5, 6 reported encouraging in vitro studies, where darbepoetin stimulated clonal maturation of erythroid progenitors from human umbilical cord blood, fetal bone marrow, and fetal liver. However, it is not clear whether the usual starting unit dose recommended for anemic adults (1 g/kg) and the usual dosing interval (once every 1 to 3 weeks) would be appropriate for neonates.4 In a recent meta-analysis, we reported that administering rEpo to preterm neonates after the first 3 weeks of life could reduce erythrocyte transfusions in a dose-dependent fashion.7 We also published a "consistent approach" to rEpo administration in the NICU, in which rEpo is given three times per week to targeted anemic neonates as a means of preventing "late" transfusions.8 Recent findings by Reiter et al.9, 10 support this approach. The potentially lower cost and fewer doses of darbepoetin needed might be the reasons neonatologists would consider for using darbepoetin rather than rEpo for selected anemic patients. However, because no clinical studies are available to guide the selection of a proper dose and dosing interval, we devised a small study to assess these issues. Specifically, as a step toward selecting a reasonable unit dose, we contrasted the erythropoietic effect, as judged by the immature reticulocyte fraction (IRF) and the absolute reticulocyte count (ARC), of two doses (1 vs 4 g/kg). As a step toward selecting a reasonable dosing interval, we measured serial plasma concentrations after dosing to estimate the t½. Before the dose, and at preset intervals after, blood was drawn for IRF, ARC, and plasma Epo concentration. For odd-numbered patients, the IRF and ARC were drawn 48 and 96 hours after the dose. For even-numbered patients, these were drawn 60 and 108 hours after the dose. This was performed to avoid excessive phlebotomy losses. Blood was obtained for plasma Epo concentrations before the dose of darbepoetin and at four preset intervals after. To avoid excessive phlebotomy losses, "population" pharmacokinetics was followed. Specifically, odd-numbered patients had phlebotomy before and 6, 48, 72, and 96 hours after the dose, and even-numbered patients before and 12, 60, 84, and 108 hours after the dose. The amount of each study phlebotomy was 0.2 ml for the IRF and ARC, and an additional 0.2 ml for the plasma Epo concentration. Institutional Review Board of Intermountain Health Care approved the study and the parents of each study participant signed informed consent. Between November 2004 and April 2005, the families of 15 NICU patients were contacted regarding participating in this study, on the basis that their neonate met the entry criteria. Consent was granted for 12 of the 15. All 12 for whom consent was granted were randomized, and all 12 received the dose of darbepoetin as stipulated by the study protocol. None of the 12 study subjects dropped out of the study. The demographics of the study infants are shown in Table 1. The 12 subjects had BWs averaging 1129245 g, they were 29.21.2 weeks gestation at delivery, and were 4312 days old with an Hgb concentration of 9.61.0 g/dl when the study drug was given. As designed, six subjects received 1 g/kg of darbepoetin and six received 4 g/kg once s.c. No adverse events were reported on any of the 12 recipients. In particular, no new rashes and no increase or decrease in blood pressure, heart rate, or need for supplemental oxygen were observed in relation to the dose. Table 2 lists the IRF and ARC before and twice following the dosing. Combining the low and high doses for comparison with baseline, both the IRF (p<0.05) and the ARC (p<0.05) increased after darbepoetin administration. As can be seen in Table 2, the IRF was somewhat more likely to increase in the 4 g/kg group than in the 1 g/kg group (p=0.06). The highest recorded drug concentrations in the 1 g/kg recipients were 185106 mU/ml (range 54 to 308 mU/ml), while in the 4 g/kg recipients they were 597238 mU/ml (range 268 to 980 mU/ml, p<0.002). The biovailability-normalized clearance for the 1 g/kg group was 17.1 ml/hour/kg and for the 4 g/kg group it was 20.7 ml/hour/kg. The terminal half-life was 29.6 and 21.5 hour for the two groups, respectively. The individual fits to the two groups did not show any significant dose dependency of the terminal half-life (p>0.05). This lack of significant dose dependency was also observed for the bioavailability-normalized clearance between the two dosing groups. The pharmacodynamic effect relative to the darbepoetin plasma level was quantified by the ratio between the AUC for the ARC vs time response and the darbepoetin AUC. This ratio was significantly (p<0.05) lower for the larger dose group (0.62 vs 2.61). Darbepoetin was developed to increase the erythropoietic activity of rEpo with less frequent dosing.1, 2, 3, 4 The darbepoetin molecule was created by biologically modifying five amino acids of the original protein in order to generate five carbohydrate binding sites, compared with three in rEpo. This modification resulted in a t½ more than twice that of rEpo. Egrie et al.3 compared the biological potency of darbepoetin and rEpo in CD-1 mice. In studies where darbepoetin was given once per week and rEpo three times per week, darbepoetin was 3.6 times more potent. In studies where both were administered once per week, darbepoetin was 13 to 14 times more potent.3 Darbepoetin has been studied extensively in two populations of anemic adults: those with anemia due to end-stage renal disease (ESRD) and those with anemia due to malignancy and chemotherapy. In patients with ESRD, doses were administered either i.v. or s.c., and ranged from 0.075 to 2.5 g/kg every 1 to 2 weeks. In oncology patients, doses ranged from 0.5 to 4.5 g/kg given every week,12 up to 15 g/kg given once every 3 weeks.13 Adult patients receiving darbepoetin have shown a significant erythropoietic response, and side effects were similar to those in patients receiving rEpo. No patients receiving darbepoetin developed antibodies to the drug. Only two reports have described darbepoetin administration to children.14, 15 Lerner et al.14 studied 12 children aged 3 to 16 years with chronic renal disease who were randomized to receive one dose of darbepoetin (0.5 g/kg) either i.v. or s.c. After 2 weeks the patients received a second dose by the alternate route. The t½ after s.c. dosing (43 hours) was similar to that in adult patients (49 hours). The t½ after i.v. dosing (22 hour) was also similar to that in adult patients (25 hours). De Palo et al.15enrolled seven children aged 7 to 15 years undergoing hemodialysis for over a year who had been treated with rEpo for at least 6 months. They used a conversion of 200 U rEpo=1 g darbepoetin. Six of the seven patients had a significant increase in Hgb within 1 month. Doses of darbepoetin were then adjusted as needed to maintain the Hgb level between 11 and 13 g/dl. In most cases, they decreased the dose, from an average of 1.6 g/kg once per week to 0.8 g/kg once per week, and eventually to 0.5 g/kg once per week. There are instances where administering rEpo to anemic preterm neonates is helpful in avoiding an erythrocyte transfusion.16, 17, 18, 19, 20, 21, 22, 23 For these infants, substituting darbepoetin for rEpo would have the advantages of fewer doses, and would likely result in lower costs. Based on the increases in IRF and ARC of our subjects, we predict that darbepoetin could indeed accelerate effective erythropoiesis in growing preterm neonates. The IRF quantifies the fraction of reticulocytes that have intermediate or high RNA staining intensity. On that basis, a higher IRF signifies release of earlier reticulocytes from the marrow. Studies by Davis,11 Brugnara,21 and Lesesve22 suggest that the IRF is more sensitive than the ARC in assessing an early erythropoietic response. The increase in IRF we observed following darbepoetin administration is consistent with stimulation of erythropoiesis. Our observation of an increase in ARC gives further support to that conclusion. Much pharmacokinetic information is available from adults treated with either rEpo or darbepoetin. Based on that data, a reasonable starting unit dose and dosing interval can be selected. For example, if the drug is to be given every three to four half-lives, to maintain a steady state, rEpo would be administered to adults about thrice weekly, and darbepoetin about once per week. However, a percentage of adult patients are successfully dosed with rEpo once per week. Similarly, some adult patients successfully receive darbepoetin only once every 2 to 3 weeks. Thus, the doses predicted by pharmacokinetics can probably be modified with equivalent clinical efficacy. In adults with ESRD, darbepoetin has a t½ of 49 hours when given s.c.24 but 25 hours when given i.v.25 The longer half-life after s.c. dosing indicates that the t½ is controlled by the release rate from the s.c. injection site. Thus, the shorter t½ we observed after s.c. dosing (26 hours) compared to that in adults (49 hours) suggests that the release of darbepoetin from the s.c. injection site is faster in NICU patients. We saw no significant difference in t½ between the two dosing groups, suggesting that the t½ is not dose-dependent. Similarly, we saw no significant difference in bioavailability-normalized clearance (Cl/F) between the two dosing groups, suggesting that the fraction of darbepoetin bioavailable from the s.c. dosing is not dose-dependent. The overall "integrated erythropoietic activity" from darbepoetin is expressed by the AUC of the ARC vs time response. In terms of pharmacodynamic activity, the AUCretic is appropriately related to the AUC for the darbepoetin+endogenous Epo response vs time, represented by AUCEpo. Owing to the nonselective nature of the assay, which does not differentiate between darbepoetin and endogenous Epo, it is not possible to accurately determine AUCdarbepoein. However, due to the comparatively low Epo baseline values relative to the total (Epo+darbepoetin) response resulting from the dosing, AUCdarbepoetin should be comparable to AUCEpo. Thus, the pharmacodynamic to pharmacokinetic relationship, given by the ratio AUCretic/AUCEpo in Table 3, indicates a nonlinear relationship between drug concentration in plasma and erythropoietic activity (Figure 1). The lower value of this ratio for the larger dose is due to a nonsignificant difference in AUCretic between the 4 and the 1g/kg dosing, while the AUCEpo largely increased in proportion to the dose, consistent with a dose-independent F and Cl for darbepoetin. Thus, increasing the dose from 1 to 4 g/kg does not give a four-fold increase in erythropoietic activity, but results in a much smaller gain. The relatively large variability we observed in half-life and clearance may be due to several factors. First, an s.c. administration produces a variable degree of bioavailability, which influences the clearance. Second, variable absorption from the s.c. injection site may create a so-called kinetic "flip-flop" situation that perturbs the terminal phase of the darbepoietin plasma level curve and thus the terminal half-life. Third, the plasma level data for the individual subjects introduced variability in the individual determinations for the clearance and the half-life. We did not find any significant correlation between half-life and clearance parameters and the demographic variables, such as gestational age at birth, BW, gender, or age or weight at the injection. Based on the limited data obtained in this pilot trial, it seems reasonable that an anemic preterm neonate who otherwise would be treated with rEpo could instead be treated with darbepoetin, but larger randomized trials are needed before darbepoetin can be recommended for routine clinical use in preterm neonates. The present results suggest that neonates will need a higher unit dose/kg and a shorter dosing interval than are generally used in adults. However, higher doses should not be expected to result in proportionately higher erythropoietic activity. We thank the NICU staff nurses at McKay-Dee Hospital and LDS Hospital for their valuable help with this study.
更多查看译文
关键词
study design,neonatology,birth weight
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络