 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
-L-Iduronidase (Laronidase)
a Willink Biochemical Genetics Unit, Royal Manchester Children's Hospital, Manchester, United Kingdom
b Department of Pediatrics, Children's Hospital, University of Mainz, Mainz, Germany
c Sleep and Respiratory Services, Great Ormond Street Hospital, London, United Kingdom
d Department of Pediatrics, Sophia Children's Hospital, Rotterdam, Netherlands
e Division of Pediatric Clinical Neuroscience, University of Minnesota, Minneapolis, Minnesota
f Genzyme Europe BV, Naarden, Netherlands
g BioMarin Pharmaceutical Inc, Novato, California
h Department of Pediatrics, Centre de Référence des Maladies Héréditaires du Métabolisme, Hôpital Edouard Herriot, Lyon, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
METHODS. This was a prospective, open-label, multinational study of 20 patients who had mucopolysaccharidosis I and were <5 years old (16 with Hurler syndrome, 4 with Hurler-Scheie syndrome) and were scheduled to receive intravenous laronidase at 100 U/kg (0.58 mg/kg) weekly for 52 weeks. Four patients underwent dosage increases to 200 U/kg for the last 26 weeks because of elevated urinary glycosaminoglycan levels at week 22.
RESULTS. Laronidase was well tolerated at both dosages. Investigators reported improved clinical status in 94% of patients at week 52. The mean urinary glycosaminoglycan level declined by 
50% at week 13 and was sustained thereafter. A more robust decrease in urinary glycosaminoglycan was observed in patients with low antibody levels and those who were receiving the 200 U/kg dosage. On examination, the liver edge was reduced by 69.5% in patients with a palpable liver at baseline and week 52 (n = 10). The proportion of patients with left ventricular hypertrophy decreased from 53% to 17%. Global assessment of sleep studies showed improvement or stabilization in 67% of patients, and the apnea/hypopnea index decreased by 5.8 events per hour (–8.5%) in those with abnormal baseline values. The younger patients with Hurler syndrome (<2.5 years) and all 4 patients with Hurler-Scheie syndrome showed normal mental development trajectories during the 1-year treatment period.
CONCLUSIONS. Laronidase seems to be well tolerated and to provide clinical benefit in patients who have severe mucopolysaccharidosis I and are <5 years old. Enzyme replacement therapy is not curative and may not improve all affected organs and systems in individuals when irreversible changes have developed. The long-term clinical outcome and effects of antibodies and laronidase dosing on glycosaminoglycan reduction warrant additional investigation.
Key Words: enzyme replacement therapy • Hurler syndrome • laronidase • MPS I
Abbreviations: MPS I—mucopolysaccharidosis I • HSCT—hematopoietic stem cell transplantation • ERT—enzyme replacement therapy • AE—adverse event • IgG—immunoglobulin G • AHI—apnea/hypopnea index • IAR—infusion-associated reaction
Mucopolysaccharidosis type I (MPS I) is caused by a deficiency of the lysosomal enzyme 
-L-iduronidase, which leads to the progressive accumulation of the glycosaminoglycans dermatan and heparan sulfate, ultimately interfering with cell functioning and compromising virtually all tissues and organs.1 Historically, patients with MPS I have been classified into 3 clinical syndromes on the basis of differences in disease progression: Hurler (severe), Hurler-Scheie (intermediate), and Scheie (mild). Patients with Hurler syndrome experience cognitive decline in early childhood, whereas patients with Hurler-Scheie and Scheie syndromes have relatively mild, if any, cognitive impairment. However, the 3 phenotypes are not always clearly delineated, because there can be substantial overlap in symptom presentation and considerable variability in both the severity and the progression of the disease.1,2 The disease is best described as a spectrum that ranges from severe (with cognitive decline) to attenuated (without cognitive decline).
Since 1980, hematopoietic stem cell transplantation (HSCT) has been used to treat patients with Hurler syndrome, first using bone marrow and more recently using umbilical cord blood.3 HSCT improves many of the somatic features of MPS I and can preserve cognitive function, but bone, heart valve, and eye disease seem to be recalcitrant to treatment. The procedure requires HLA-matched donor cells, is prone to engraftment failure, and is associated with considerable morbidity and mortality, which has limited its use to severely affected patients who are early in the course of disease. Enzyme replacement therapy (ERT) with laronidase (recombinant human -L-iduronidase) has been shown to be a safe and effective treatment in patients who have MPS I and are older than 5 years.4–6 In the pivotal study to confirm the drug's safety and efficacy, the use of functional assessments as co-primary end points (forced vital capacity and the 6-minute walk test) precluded the enrollment of young patients and those with severe cognitive impairment. To evaluate the safety, pharmacokinetics, and efficacy of laronidase in a young and severely affected patient population, this study investigated the effects of weekly laronidase infusions for 1 year in 20 patients who had MPS I and were younger than 5 years.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
All patients were naive to laronidase therapy, had to be younger than 5 years at initiation of treatment, and had to have a diagnosis of MPS I confirmed by fibroblast or leukocyte -L-iduronidase enzyme activity <10% of normal and by genotyping. Exclusion criteria included having undergone or being under consideration for HSCT, acute hydrocephalus, clinically significant organic disease not related to MPS I, administration of an investigational drug within 30 days before study enrollment, or known hypersensitivity to components of the laronidase solution. All parents or legal guardians provided written informed consent to participate in the study. The protocol was approved by each site's independent ethics committee. The study was designed and conducted in compliance with the principles of the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use guidelines for Good Clinical Practice.
Laronidase Treatment
All patients initially received weekly intravenous infusions of 100 U/kg (0.58 mg/kg) laronidase (recombinant human -L-iduronidase (Aldurazyme; BioMarin Pharmaceutical, Inc, Novato, CA; and Genzyme Corp, Cambridge, MA). Laronidase was diluted in 100 mL or 250 mL of 0.9% sodium chloride injection and infused over 4 hours. Unlike in previous studies, laronidase was administered without 0.1% human serum albumin, in accordance with the European summary of product characteristics. On the basis of the results of an interim analysis that was performed on the first 13 patients, the study protocol was amended to allow the final 7 patients to receive a dosage of 200 U/kg laronidase from week 26 onward if their urinary glycosaminoglycan level at week 22 was >200 µg/mg creatinine; 4 of these patients qualified and received the higher dosage for the second half of the study. A Port-a-Cath was placed in 8 patients. To minimize possible infusion-associated reactions (IARs), all patients received an antipyretic and an antihistamine before each infusion.
Evaluation of Safety
Safety monitoring included adverse event (AE) reporting, physical examination, clinical chemistries, hematology parameters, urinalysis, vital signs, and electrocardiograms. AEs were characterized by severity and by relationship to study drug. An infusion-associated reaction was defined as any AE that occurred from the start of an infusion to the end of the 30-minute postinfusion observation period (longer if deemed necessary by the investigator) and that was considered to be drug related by the investigator.
Antibody Testing
Antibody titers to laronidase (immunoglobulin G [IgG]) were measured every 4 weeks with the use of a modified enzyme-linked immunosorbent assay and confirmed by immunoprecipitation.7 IgE testing was to be performed after moderate to severe IARs.
Pharmacokinetic Assessment
Pharmacokinetic assessments were performed at weeks 1, 13, 26, and 52. Plasma laronidase activity was determined using 4-methylumbelliferyl iduronic acid as a substrate.8 Pharmacokinetic parameters for laronidase were calculated using standard noncompartmental methods.
Biochemical Evaluations and Efficacy Outcome Measures
Biochemical evaluations and clinical outcome measures that were relevant to this young severely affected MPS I patient population consisted of urinary glycosaminoglycan excretion, liver size, cardiac status, upper airway obstruction during sleep, growth velocity, the investigator's global assessment, and mental development.
Urinary glycosaminoglycan excretion was determined in the first morning void using an automated dimethylmethylene blue dye–binding procedure in a central laboratory (BioMarin Pharmaceutical Inc) and expressed as micrograms per milligram of creatinine.4
Liver size was evaluated by the distance of the liver edge below the right costal margin at the midclavicular line at physical examination. Quantitative volumetric measurements by MRI and/or computed tomography would have required anesthesia and were considered too high a risk for these patients.
Two-dimensional echocardiography was performed and interpreted by local cardiologists according to a centralized protocol. Left ventricular mass z scores were calculated using normative data from Children's Hospital (Boston, MA).9 In addition, local cardiologists assessed the degree of left ventricular hypertrophy and valvular appearance.
Polysomnograms with resulting apnea/hypopnea index (AHI) scores (the total number of episodes of apnea and hypopnea per hour of sleep) were interpreted centrally by a single, independent sleep study expert. Upper airway obstruction during sleep should be considered as a continuum that may not fully be represented by the AHI, especially in young children. Therefore, the severity of the upper airway obstruction during sleep was also assessed as mild, moderate, or severe by the same expert using a nonlinear clinical global assessment scale based on well-defined criteria that combined clinical observation and sleep laboratory testing results.10,11
Height and weight were measured at baseline and at weeks 13, 26, and 52 and expressed as height-for-age and weight-for-age z scores using the Centers for Disease Control and Prevention/National Center for Health Statistics clinical growth charts.12,13 z scores were also calculated for a historical control group that consisted of untreated patients with MPS I in the same age range (6–72 months) from the MPS I Registry (BioMarin and Genzyme, data on file).14
Investigators provided a global assessment of the patient's clinical status using a 7-point scale (marked, moderate, or slight decline; no change; mild, moderate, or marked improvement) before and after 13, 26, and 52 weeks of treatment with laronidase. Exploratory assessment on cognitive function was performed at baseline and weeks 26 and 52 using the Griffiths Mental Development Scales, which is validated for children from birth to 8 years and is widely used in Europe.15 Each patient's mental development trajectory was calculated by plotting the patient's mental age equivalent versus the patient's chronological age for each study time point. The normal trajectory for mental development has the mental age corresponding to the chronological age.
Statistics
No hypothesis testing was performed in this open-label study. Continuous data were summarized by using means, medians, and ranges. Categorical data were summarized by using frequencies and distributions. All analyses were performed with the use of SAS 8.0 software (SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Two patients died during the study as a consequence of events related to their underlying disease. A 13-month-old girl with Hurler syndrome died of cardiac failure after 25 weeks of laronidase therapy, but no autopsy was performed. She had an episode of cardiac failure before the study and became cyanotic 6 weeks before her death. A 3-year-old boy with Hurler syndrome underwent surgery for bilateral hip dysplasia and died of a postsurgical complication (accidental extubation with unsuccessful reintubation and tracheostomy) at week 48.
There were no clinically meaningful changes for any of the serum chemistry, hematologic, or urinary parameters assessed, and changes in vital signs or physical examination parameters during the study were unremarkable.
All patients in this study developed IgG antibodies to laronidase with a mean time to seroconversion of 25.8 days. Antibody titers generally rose during the first 20 weeks of the study; the highest measured titer during the study was 1:204800. After 1 year of laronidase treatment, 4 patients had titers 
1:1600 (1 patient was seronegative), and 12 patients had titers >1:1600. There was no apparent correlation between the time to seroconversion or antibody titer and the incidence of IARs. In the 4 patients (marked with a ‘‘b’’ in Table 1) who received the increased dosage of laronidase, the subsequent antibody titers did not change appreciably. Two patients tested negative for IgE antibodies after moderate IARs.
Pharmacokinetics
The range of individual patient values, although variable, was reasonably consistent across 52 weeks of treatment. The mean area under the curve increased from 0.58 ± 0.59 hours/U per mL at the first infusion to 0.94 ± 0.97 hours/U per mL at week 52, but the change was not significant. There was no apparent relationship between area under the curve and antibody titer over the 52 week period. The mean half-life ranged from 0.55 hours to 1.55 hours with large variability in individual values. The mean volume of distribution corrected for body weight decreased from 0.753 ± 0.497 L/kg to 0.246 ± 0.210 L/kg during treatment. There was a trend toward a lower volume of distribution as the antibody level increased, suggesting that the binding of antibodies to laronidase keeps more laronidase in the plasma, away from the renal epithelial cells that are believed to be the source of urinary glycosaminoglycans, resulting in a smaller distribution volume. Pharmacokinetic analyses were hampered by the relatively small sample size and large intragroup variations that did not allow for any meaningful correlation analyses.
Urinary Glycosaminoglycans and Efficacy Outcome Measures
After initiation of treatment with laronidase, urinary glycosaminoglycan levels showed a sharp decline within the first 13 weeks followed by a plateau. The mean reduction in urinary glycosaminoglycan level after 52 weeks of treatment was 61.3% for all patients, 59.1% for patients who were treated with 100 U/kg throughout the study, and 67.7% for the 4 patients who received 200 U/kg after week 26 (Table 2).
|
|
The number of patients who were categorized as having mild left ventricular hypertrophy by echocardiography decreased from 52.6% (10 of 19) of assessable patients at the start of the study to 16.7% (3 of 18) of assessable patients at the final study visit. At the start of study, the mean left ventricular mass was abnormally high as demonstrated by a mean z score of 3.8 (range: 0.6–8.4). At the end of study, the mean left ventricular mass z score had decreased by 0.9 (–11.3%) for the 17 patients with available data (Table 2). For the 14 patients with left ventricular mass z scores >2 at baseline, the mean reduction in z score was 1.3 (–28.7%), demonstrating that the largest decreases in left ventricular mass were observed in the patients with hypertrophy at baseline.
Six patients had normal AHI values at baseline (AHI <10 events per hour); in 4 of these patients, the AHI remained normal for the rest of the study, and in 2 patients, the AHI increased to 11.6 and 14.7 events per hour. AHI changes were assessed in the subgroup of patients (n = 9) who had clinically significant abnormal baseline values (AHI 
10) and available results at the end of study. In this subgroup, the mean AHI decreased from 45.3 to 39.6 events per hour, which corresponds to an 8.5% reduction. Predefined clinical significance (ie, a 25% reduction in events per hour) was reached by 5 patients (33% of the patients with available data). Of the 9 patients with abnormal baseline AHI values, 5 patients had tonsillectomies and/or adenoidectomies performed between baseline and week 11 that might have confounded the results. However, most of the AHI improvements occurred between weeks 26 and 52, suggesting that the benefit was attributable to laronidase rather than surgery.
In the 15 patients with available sleep study results at both baseline and week 52, global assessment of the sleep study data by the central expert revealed that 5 (33%) patients showed improvement (1 mild, 3 moderate, and 1 marked), 5 (33%) patients showed no change, and 5 (33%) patients showed worsening (1 marked, 1 moderate, and 3 slight).
Changes in the height and weight growth velocities of study patients were compared with cross-sectional data from an age-matched, untreated control group of patients (27 with Hurler syndrome and 17 with Hurler-Scheie syndrome) from the MPS I Registry (BioMarin and Genzyme, data on file). There was no apparent relationship between the gender of the patient and height-for-age or weight-for-age z score over the study period of 1 year. The regression line for untreated registry patients shows that patients with MPS I disease are initially taller than their age-matched normal peers during the first 2 years of life and that, as expected, the height-for-age z score declines with age (Fig 2). Seven study patients (4 with Hurler syndrome and 3 with Hurler-Scheie syndrome) showed a net increase in height-for-age z score after 52 weeks of treatment. Likewise, untreated patients showed a declining weight-for-age z score across the same age range, whereas 3 study patients (all with Hurler syndrome) showed a net increase in weight-for-age z score during the 1-year study duration (data not shown).
|
|
For determination of whether the presence of IgG antibodies to laronidase affected urinary glycosaminoglycan excretion, antibody titers at week 51 were plotted against the percentage of urinary glycosaminoglycan reduction at week 52 (Fig 4). A more robust decrease in urinary glycosaminoglycans (above the median of 63%) was observed in patients with low antibody levels (1:1600) and those who received the 200 U/kg dose.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Many of the patients with Hurler syndrome have 2 null mutations and are not expected to make cross-reacting immunologic enzyme.1 In theory, this could lead to a stronger humoral immune response to laronidase and an increased safety risk in these patients. All patients in this study developed IgG antibodies against laronidase. This is similar to the seroconversion rate (93%) that was seen in patients who had Hurler-Scheie and Scheie syndromes and received laronidase in previous studies.4,5 Although the overall IgG levels in this study were higher and the mean time to seroconversion was shorter than in previous studies, these findings were not associated with any noticeable differences between phenotypes in terms of IARs or mean percentage of glycosaminoglycan reduction. The type and the frequency of IARs were similar to those that were seen in older patients who were treated with laronidase in the Phase 3 double-blind study.5
In this study, patients with low or absent antibody levels had a more robust urinary glycosaminoglycan reduction than patients who had higher antibody levels and showed more variable urinary glycosaminoglycan reductions. The use of a higher dosage (200 U/kg) might overcome the effect of high antibody levels on the urinary glycosaminoglycan excretion, as shown by the more robust urinary glycosaminoglycan reductions in the patients who received the double dosage.
The pharmacokinetics of laronidase in patients who had MPS I and were 5 years of age were characterized in an earlier study in which patients received 100 U/kg once a week for 26 weeks rather than 52 weeks.5 Taking into account the small number of patients in these studies, there is reasonable agreement in the ranges for the key pharmacokinetic parameters, suggesting no major difference between patients who are <5 and 5 years of age.
The degree of reduction in urinary glycosaminoglycan level and hepatomegaly indicates effective clearance of accumulated glycosaminoglycan substrates in young and severely affected patients, consistent with the treatment effect that was observed previously in older patients with more attenuated disease.4,5 By comparison, bone marrow transplantation also leads to a reduction in urinary glycosaminoglycan levels to the upper limit of normal after several months.16 Weekly infusion of laronidase that was started 20 months after transplantation in 1 patient with mixed chimerism has normalized glycosaminoglycan excretion (N. Guffon, MD, verbal communication, 2006). Urinary glycosaminoglycan clearance was not proportionally increased by doubling the dosage of laronidase, but an increase in dosage may be beneficial in patients with persistently high urinary glycosaminoglycan levels after an initial period of treatment at the approved 100 U/kg (0.58 mg/kg) dosage. Early dosage-ranging studies that were conducted in dogs with MPS I showed a similar result with only a modest additional improvement in tissue glycosaminoglycan clearance at a dosage of 2.0 mg/kg versus 0.5 mg/kg laronidase (BioMarin, data on file).
Exploratory clinical efficacy end points also suggest that laronidase exerts positive effects on some organ systems in this young severe MPS I patient population. Cardiorespiratory dysfunction is an important cause of morbidity and mortality across the spectrum of patients with MPS I.17 Left-sided valvular heart disease that is caused by mitral and/or aortic valve dysplasia and primary myocardial involvement are well documented in MPS I. The left ventricular mass was significantly increased at baseline with a z score for the group of 3.8, but the ejection fraction and other echocardiogram parameters where within the limit of normal. No indirect signs of coronary occlusion could be detected, but 1 patient died from an undefined cardiac cause 1.5 months after the start of therapy. In this study, 70% of the patients who presented with left ventricular hypertrophy showed resolution after 1 year of treatment. This is consistent with other reports on a small number of patients that have shown regression of ventricular hypertrophy in the short term (5 years) after bone marrow transplantation.18,19 The largest decreases in left ventricular mass were observed in the patients with hypertrophy at baseline.
Resolution of cardiomyopathy after laronidase treatment also has been described in a young patient who had Hurler syndrome and initially was deemed ineligible for bone marrow transplantation.20 Once improved, the child was able to undergo successful transplantation. During this study, only small changes in valvular structure and function were observed. Longer term follow-up is needed to determine whether laronidase is able to halt or slow the progression of valvular disease and avoid the need for valve replacement surgery.
Exploratory mental development testing indicated that the patients with Hurler-Scheie syndrome had a normal to above-normal rate of cognitive growth during the 1-year study. Similarly, the younger (<2.5 years of age) patients with Hurler syndrome showed an increase in cognitive function at a rate similar to that of healthy children. In contrast, the older patients with Hurler syndrome did not show any significant gains or loss in cognition.
Because the intravenously infused enzyme is not expected to cross the blood-brain barrier in appreciable amounts at the administered dosage levels,21 it is possible that some of the developmental gains that were observed were indirectly related to improvement in overall health status. Long-term follow-up is needed to distinguish direct from indirect effects. The children were reported to benefit from the treatment as assessed by the investigator global assessment.
There are some limitations in the design and the methods that were used in this study. First, several factors precluded performing a randomized, controlled study, including the low incidence of MPS I, the rapidly progressive and fatal course of Hurler disease, the families' unwillingness to receive HSCT as an alternative therapy, and ethical concerns about withholding a medication that was anticipated to receive regulatory approval during the study. Historical comparison was largely impossible because of the paucity of data from untreated patients with Hurler syndrome. The relatively short follow-up period hampers definitive conclusion on the impact of laronidase on the long-term disease progression. Comparisons from the University of Minnesota database suggest little difference early in cognitive development from HSCT (E. Shapiro, PhD, written communication, 2006). A group of 15 children who had MPS I and underwent HSCT and were the same age and had the same gender distribution had the same median developmental quotient of 65 as observed in our study group after 1 year of follow-up. Both groups were better than untreated children from the same database whose median developmental quotient was 50 at a median age of 3.3 years. Longer term follow-up, particularly of the young patients with Hurler syndrome in this study, will clarify the cognitive development trajectory of patients who have Hurler syndrome and are treated with ERT.
The age range of the patients commanded choice of some nonquantitative explorative efficacy measures. Most, if not all, patients would not have been able to cooperate with sophisticated testing methods (eg, forced vital capacity, 6-minute walk test) as used in previous trials. In addition, the high risks that are related to anesthesia precluded the use of MRI. Because of the small sample size and variability, it was not possible to establish correlations between urinary glycosaminoglycan reductions and improvements in clinical end points in this trial.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The age at diagnosis, the severity of existing symptoms, the expected disease progression, and the availability of a donor in the case of transplantation are factors to take in account when discussing treatment options with parents. In this trial, all parents and guardians were fully informed of the risk/benefit profiles of HSCT and laronidase. A variety of reasons led parents to choose ERT for their children who were younger than 5 years: delayed diagnosis (after 2 years of age), significant developmental delay, lack of a compatible HSCT donor, non-Hurler phenotype (no expected neurocognitive regression), and concern over the safety risk of transplantation. MPS I is a rapidly progressive disorder; as for other lysosomal disorders, initiation of treatment as early as possible in the disease course is required to prevent and/or minimize irreversible damage.
|
ACKNOWLEDGMENTS |
|---|
We acknowledge the participation of study patients and their families and the expert assistance of all study-site coordinators and personnel.
|
FOOTNOTES |
|---|
Address correspondence to J. Edmond Wraith, MBChB, Royal Manchester Children's Hospital, Willink Biochemical Genetics Unit, Hospital Road, Pendlebury, Manchester M27 1HA, United Kingdom. E-mail: ed.wraith{at}cmmc.nhs.uk
Financial Disclosure: Dr Wraith provided paid and unpaid consultancy services for Genzyme, received travel expenses to attend meetings, and participated in clinical trials sponsored by Genzyme; Dr Beck has received unrestricted grants from BioMarin and Genzyme; Dr Lane received payment from Genzyme to analyze, interpret, and report the sleep studies that were performed in this study; Dr van der Ploeg has performed clinical trials for Genzyme and has received travel expenses and honoraria to attend and present at scientific meetings; Dr Shapiro is a consultant on neuropsychological methods and has received travel expenses to attend and present at scientific meetings; Dr Xue is employed by Genzyme; Dr Kakkis is employed by BioMarin and has financial conflicts related to laronidase; and Dr Guffon has performed clinical trials for Genzyme and has received travel expenses and honoraria to attend and present at scientific meetings.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2007 American Academy of Pediatrics. All rights reserved. |
||
|
||
