DEPARTMENT OF HEALTH & HUMAN SERVICES Public Health Service

Food and Drug Administration

Memorandum Center for Biologics Evaluation and Research

1401 Rockville Pike

Rockville, MD 20852

Division of Clinical Trial Design and Analysis

HFM-576

 

Date: December 13, 2002

Subject: Advisory Committee Briefing Document

 

BLA STN 125058/0

Original Submission

 

BioMarin Pharmaceuticals, Incorporated

 

ALDURAZYME™

Laronidase

For the Treatment of Mucopolysaccharidosis I

Clinical Review

Table of Contents

 

Table of Contents *

Introduction *

Proposed indication and dose *

Biochemical and Clinical Background *

Product Background *

Regulatory History *

Pre-clinical studies *

Toxicology studies *

Pharmacodynamic studies *

Safety *

Overview of Clinical Studies *

Study BIO7500 *

Overview *

Protocol *

Design *

Objectives *

Eligibility Criteria *

Treatment assignment / randomization *

Product information and administration *

Evaluations *

Safety Monitoring *

Endpoints *

Statistical analysis *

Study Conduct *

Protocol Violations *

FDA site inspection findings *

Results *

Subject disposition *

Baseline characteristics *

Study drug exposure *

Primary endpoints *

Hepatomegaly *

Splenomegaly *

Urinary Glycosaminoglycans *

Secondary endpoints *

Safety *

Summary *

Study ALID-003 *

Overview *

Protocol *

Design *

Objectives *

Eligibility Criteria *

Treatment assignment / Randomization *

Product information and administration *

Evaluations *

Safety Monitoring *

Endpoints *

Statistical Analysis *

Study Conduct *

Database Integrity *

Protocol violations *

FDA site inspection findings *

Results *

Subject disposition *

Baseline characteristics *

Study drug exposure *

Primary endpoints *

Percent Predicted Forced Vital Capacity (% FVC) *

Six minute walk distance (meters) *

Secondary endpoints *

Apnea / Hypopnea Index (AHI) of the sleep study *

Liver volume *

Disability Index from the CHAQ or HAQ *

Shoulder Flexion *

Tertiary Endpoints *

Pharmacokinetic Studies *

Safety *

Summary *

Study ALID-006 *

Overview *

Protocol *

Design *

Objectives *

Eligibility Criteria *

Treatment assignment / Randomization *

Product Information and Administration *

Evaluations *

Safety Monitoring *

Endpoints *

Statistical Analysis *

Study Conduct *

Database Integrity *

Protocol Violations *

FDA site inspection findings *

Results *

Subject disposition *

Study entry characteristics *

Study drug exposure *

Primary endpoints *

Percent Predicted Forced Vital Capacity (% FVC) *

6 Minute Walk Distance *

Secondary endpoints *

Tertiary endpoints *

Safety *

Summary *

Summary of Efficacy *

Summary of Safety *

Conclusions and Recommendations *

Appendix A: Tertiary Endpoints of Study ALID-003 *

Appendix B: Tertiary Endpoints of Study ALID-006 *

 

 

Introduction

This document is the medical officer’s review of the clinical data submitted with the Biologics License Application (BLA) Submission Tracking Number 125058. This application is for Laronidase (ALDURAZYME), a recombinant enzyme product which is proposed for usage as enzyme replacement therapy for patients with Mucopolysaccharidosis I.

Proposed indication and dose

Biomarin proposes that Laronidase is indicated as long term enzyme replacement therapy in patients with Mucopolysaccharidosis I (MPS I) to treat the non- central nervous system manifestations of the disease. The proposed dosage regimen of Laronidase is 100 U / kg (0.58 mg / kg) administered once weekly intravenously.

Reviewer’s comment: This disorder, as for all forms of lysosomal storage diseases, has complex and multiple organ and tissue manifestations. There are variable phenotypes and a broad spectrum of severity. Evidence-based determination of a treatment effect on specific endpoints should be required to support a claim of benefit for these endpoints. A claim of general benefit in the management of the disease as a whole may be inappropriate.

Biochemical and Clinical Background

Lysosomal storage disorders result from a genetic defect that causes deficient production or function of one or more of the lysosomal enzymes. The enzymatic deficiency results in an abnormal accumulation of metabolites within a lysosome and ultimately disruption of the normal cell function and cell death. Lysosomal storage disorders are usually classified according to the nature of the macromolecule that is abnormally catabolized and consequently accumulates within the lysosome. Sphyngolipidoses (including gangliosidoses) are associated with the accumulation of complex lipids, the basic structure of which is a sphingosine, a long chain amino-alcohol. Oligosaccharidoses or mucolipidoses are associated with the storage of complex glycoproteins. Mucopolysaccharidosis are caused by deficiencies of enzymes needed to degrade glycosaminoglycans (also known as mucopolysaccharides). Glycosaminoglycans themselves are lysosomal degradation products derived by proteolytic removal of the protein core of proteoglycans (macromolecules occurring in the cell membrane and extracellular matrix). Mucopolysaccharidosis I is the subject of this license application, and will be described here briefly.

Mucopolysaccharidosis I (MPS I) is characterized biochemically by the deficiency of the lysosomal enzyme a -L-iduronidase (IDU). This enzyme cleaves the terminal iduronic acid residues of dermatan sulfate and heparan sulfate (Figure 1).

Figure 1. Catabolism of glycosaminoglycans highlighting role of IDU

Source: www.genome.ad.jp/dbget-bin/get_pathway?org_name=hsa&mapno=00531 assessed on October 16, 2002

Experiments in cultured fibroblasts show that the enzyme is made in a precursor form, cleaved intracellularly to a 628 amino acid protein and incorporated with mannose 6-phosphate markers for targeting to lysosomes. The gene encoding IDU is located in the short arm of chromosome 4 (locus 4p16.3) and MPS I is inherited with autosomal recessive transmission. Homozygosity or compound heterozygosity for the W402X (stop codon) and Q70X mutations are the common causes of MPS I with a severe form in affected Caucasian individuals. Japanese patients with MPS-I have other predominant IDU gene mutations.

Historically, MPS I patients have been broadly categorized into three clinical syndromes: Hurler, Hurler-Scheie, and Scheie, representing severe, intermediate, and mild clinical phenotypes, respectively. These classifications are arbitrary categorizations of points on a continuous spectrum of patient phenotypes. Biochemically, patients with the milder disease form retain trace residual amounts of IDU that are sufficient to ameliorate the phenotype to varying degrees. However there is considerable heterogeneity in the severity and symptoms within each phenotype and substantial overlap of the three syndromes. The true incidence of MPS I is unknown, with estimates in the range of 1/76,000 for MPSIH, 1/280,000 for MPS IH-S, and less than 1 in 840,000 live births for MPS IS in Northern Ireland. Semi quantitative analysis of spot urinary glycosaminoglycans can be used for screening, but is subject to both false negative and false positive results. Definitive diagnosis is established by lysosomal enzymes assays in leukocytes, cultured skin fibroblasts or serum. Pre-natal diagnosis is established by IDU assays in cultured cells from amniotic fluid or chorionic villus biopsies. Carrier testing is currently performed with analysis of enzyme activity in family members, but may be superseded by molecular analysis of specific family mutations in the enzyme gene.

Hurler syndrome (MPS IH) is a progressive disorder that affects multiple organs and tissues and leads to death during childhood. The symptoms of MPS IH present between 6 months and 2 years of age. They include inguinal or umbilical hernias, hepatosplenomegaly, coarse facies, deafness, recurrent ear and sinus infections, macroglossia, noisy breathing, obstructive airway disease and sleep apnea, communicating hydrocephalus with increased intracranial pressure, prominent forehead, developmental delay, skeletal deformities (dysostosis multiplex), corneal clouding, joint stiffness, acute cardiomyopathy associated with endocardial fibroelastosis, valvular heart disease and pulmonary hypertension. The most frequent causes of death are related to obstructive airway disease, respiratory infections and cardiac complications.

Symptoms of Hurler-Scheie Syndrome (MPS IH-S) include dysostosis multiplex, corneal clouding, joint stiffness, short stature, deafness, and obstructive airway disease with little or no intellectual dysfunction. The onset of these symptoms is observed between ages 3 and 8 years of age, and death usually occurs in the second or third decade of life, usually from the same complications described in the Hurler patients.

Patients with Scheie Syndrome (MPS IS) have variable amounts of joint stiffness, aortic valve disease, mild hepatosplenomegaly, and corneal clouding, usually without neurologic involvement. Symptoms start in children older than 5 years and the diagnosis is usually made between 10 and 20 years of age. Patients achieve normal stature and normal lifespan.

Current management of patients with MPS I is restricted to supportive care and treatment of specific complications. Ventriculoperitoneal shunting in moderate to severe hydrocephalus, ventilating tubes and hearing aids, range of motion exercises, tracheostomy and high pressure nasal continuous positive airway pressure, mitral or aortic valve replacement can improve symptoms.

A treatment that can replace the defective enzyme by supplying either cells capable of normal IDU secretion or functional IDU can potentially improve the clinical manifestations of MPS I. Cell culture and animal model experiments have demonstrated feasibility and raised optimism. Despite numerous trials using different enzyme sources, the only significant advance has been transplantation of allogeneic bone marrow in patients with MPS IH, particularly if instituted before the age of 2 years. With stable bone marrow engraftment the biochemical and somatic features revert, and long-term survival is possible. Neuropsychological function and some skeletal abnormalities are not reversed in patients that undergo bone marrow transplantation after age 2. However, limitations in the appropriate donor pool, and the high risks of significant morbidity and mortality make this therapeutic option restricted to a few affected individuals.

With the cloning of complementary DNA for IDU, large scale production of the recombinant human a - L – IDU became possible, with the mannose 6 phosphate sites necessary for targeting lysosomes.

Product Background

Recombinant human α -L-iduronidase (rhIDU) is a 628 amino acid lysosomal hydrolase. The enzyme is a single polypeptide chain of molecular mass 70.1 kilodaltons (from translated cDNA sequence). rhIDU contains six N-linked glycosylation sites, two of which carry the bis mannose-6-phosphate oligomannose7 7 oligosaccharide that binds the target cell surface receptor. The apparent molecular mass of full-length rhIDU is approximately 83 kilodaltons, suggesting that posttranslational modifications of rhIDU contribute approximately 13 kilodaltons to the molecular mass.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- . The ------------- formulated drug product is supplied as a sterile solution at a concentration of 0.58 mg/mL (100 U/mL) in a sodium chloride and sodium phosphate buffer. Each vial of drug product contains 2.90 mg of Aldurazyme™ and is intended for single use.

The product used in the pre-clinical studies and the Phase 1 clinical study was originally defined with an activity of 125,000 U/mL and a protein concentration of 0.5 mg/mL. These characteristics of laronidase were subsequently redefined (due to changes in assay procedures) as 100 U /mL activity and 0.58 mg/mL protein concentration. These redefinitions do not represent any changes in actual enzyme activity or concentration, rather reflecting changes in assay methodology. However, terminology used in the Phase 1 study and in preclinical study reports are based upon the prior assay process.

Regulatory History

Development of laronidase was conducted under IND 7334, initially submitted September 19, 1997 proposing a Phase 1 clinical study. Subsequently, a single Phase 3 study was conducted.

Pre-clinical studies

Reviewer’s comment: It is important to note that the assay for a -L–iduronidase activity was changed during the drug development to make it more reproducible and robust. A change in the definition of activity unit to be more conventional occurred as a consequence. The dose used in the Phase 1 study was reported as 125,000 U, which was determined to be equivalent to 100 U under the revised activity definition. The method for determination of protein concentration was also changed, resulting in a small apparent mass increase, such that the 0.5 mg / ml formulation of rhIDU now measured 0.58 mg / ml in the new assay.

Dosing in all the pharmacodynamic studies was reported on an enzymatic activity basis (U / kg) using the old activity unit and protein assay. Dosing in the toxicity studies was reported only on a weight basis using the older protein assay. Most of the pre-clinical study reports present the dose used in a mg / kg basis, except for the pharmacodynamic studies, which also present data in the original activity units for comparison to the published literature.

Toxicology studies

Placebo and dose-controlled acute toxicity studies using single doses of laronidase were conducted in rats and dogs with a dose range of 0.29 to 5.8 mg kg in rats and 0.116 to 11.6 mg/kg in dogs. No treatment related toxicities were found, although dogs had low incidence of emesis and mucoid liquid stools following treatment with laronidase, unrelated to dose. Female rats developed small foci of hepatocellular necrosis of unclear relationship to the treatment and unclear significance.

An intravenous laronidase study in cynomolgus monkeys showed a slight increase in lymphocytes and eosinophils when the highest dose (16.6 mg/kg) was infused weekly for 26 weeks.

Pharmacodynamic studies

In vitro studies were carried out to explore the uptake of laronidase by MPS I patient fibroblasts, its effect on GAG storage and its cell half-life. These studies demonstrated effective laronidase endocytosis by the fibroblasts via a mannose-6-phosphate dependent receptor with a half-maximal uptake at approximately 0.7 nM enzyme. Laronidase reduced GAG storage in MPS I patient fibroblasts with half-maximal reduction at approximately 0.7 pM. The half-life of the enzyme in the fibroblasts was estimated at 5 days.

Reviewer’s comment: The sponsor claims this result supports a once weekly dosing regimen, but no in-vivo studies were performed to examine the tissue half-life of the enzyme.

Subsequent studies in the canine model using laronidase at 125,000 versus 500,000 U/kg/week (0.5 or 2.0 mg/kg/week) showed superiority of 9-hour weekly infusions over continuous infusions in raising tissue enzyme levels and decreasing tissue GAG accumulation. More importantly, this study also demonstrated that at the 2.0 mg/kg/week dose, the 9-hour infusion was more effective than continuous infusion at lowering GAG levels in cornea, kidney cortex, liver, myocardium, pancreas, synovium, cerebral cortex, and medulla. There were no significant differences in GAG accumulation within denser connective tissues (i.e. ligaments, cartilage, cornea, atrioventricular valve leaflets).

Preclinical Safety

Infusion related anaphylactoid reactions were observed in dogs and cats and managed by stopping the infusion, intravenous fluid administration and, if necessary, oxygen supplementation. The most likely cause of these reactions is IgG-mediated complement activation. Anti-histamines treatment prior to the infusions was effective in eliminating or decreasing these reactions. Lowering the infusion rate and the addition of canine serum albumin also mitigated these reactions in long term laronidase administration. BioMarin notes that the clinical grade product is more pure than the one used in the pre-clinical studies, and polysorbate 80 added as a detergent in the clinical grade product can potentially decrease microaggregation. Anaphylactoid reactions were not seen in monkeys or dogs treated with the clinical grade product, even without pre-treatment with antihistamines.

Overview of Clinical Studies

This application contains data from a total of 59 patients with various forms of MPS I. The initial open label study of laronidase has been extended as a single arm, uncontrolled trial. Study ALID-003 enrolled 45 subjects into the single randomized, placebo controlled, double blind trial reported with this application. At the end of the study all subjects were offered participation in an open label single arm laronidase protocol. Subjects previously randomized to placebo were converted to laronidase treatment.

There is also a small ongoing Open Label Study that provides treatment for patients with late-stage MPS I who are seriously ill and do not meet the selection criteria for participation in ongoing clinical studies. As of 29 April 2002, 4 patients had received laronidase treatment in this program.

The sponsor has conducted 3 clinical studies during the clinical development of laronidase (Table 1)

Table 1. Clinical Studies Included in this Submission

Protocol

Study Design

Completion Status / Study Dates

Location/ Number of Centers

Duration of treatment (weeks)

n

ALID-003

Phase 3 Double Blind, Placebo Controlled Randomized

Completed

12/28/00 to 9/6/01

USA:2

Canada: 1

UK: 1

Germany: 1

26

45

ALID-006

Phase 3 Open Label , non-randomized Extension

Ongoing

USA: 13

Canada: 2

UK: 1

Germany: 2

24

45

BIO7500

Phase 1 / 2 Open Label, non-randomized

Ongoing

Started 11/28/97

USA: 13

171 (safety)

10

Special Access & ALID-007

Uncontrolled administration

Ongoing

Australia: 3

35

3

Single patient

Completed (subject died)

USA: 1

28

1

A multicenter observational study (ALID-004) was also conducted in 2000 to collect normative data in patients with MPS I. The single session survey included 5 centers in the USA and Europe and enrolled 45 patients. The objective was to determine median values for the six minute walk distance test (6MWD) and the forced vital capacity (FVC) in patients with MPS I intended to participate in the Phase 3 clinical trial. Information from this study was used to partly guide design of Study ALID-003. A notable finding was the higher than expected distances on the six minute walk test. This relatively mild degree of impairment might have suggested that there would be difficulty in demonstrating a clinically meaningful difference in a randomized study.

 

Study BIO7500

"Phase 1 / 2, Open-Label Study of Recombinant Human a - L – Iduronidase as Enzyme Replacement Treatment for Mucopolysaccharidosis I (MPS I)".

Overview

This study was conducted under IND 7334 according to Protocol BIO7500-001. The study started on November 28, 1997 and is still ongoing.

The submitted study report contains safety and clinical outcome data derived from a 152 week period. The original protocol submitted proposed a 26 week study and was subsequently modified to extend the treatment duration. A change in product manufacture method and formulation occurred during the study. Crossover comparisons of pharmacokinetic and pharmacodynamic parameters were done.

The reported results demonstrate that the newer product had a comparable safety profile to the older product in the seven subjects that participated in the crossover study. An apparent increase in plasma enzyme activity 20 minutes after completion of infusion was observed with newer as compared to the older product.

Protocol

Title: "Phase 1 / 2, Open-Label Study of Recombinant Human a - L – Iduronidase as Enzyme Replacement Treatment for Mucopolysaccharidosis I (MPS I)". Protocol BIO7500-001.

The study was initiated on November 28, 1997 and the reporting period included in the corresponding study report included in this submission ended on March 30, 2001. Six protocol amendments occurred after initiation of the study. These served largely to clarify certain study criteria or procedures, or to extend the study duration, and are incorporated in the following description.

Design

Open label non-randomized, single dose, multicenter, phase 1 study in 10 subjects with MPS I. Effect of laronidase treatment was compared to pre-treatment levels. The study drug was to be administered intravenously on a weekly basis initially for a period of 26 weeks. Subsequent amendments to the protocol extended the study period to 152 weeks.

Reviewer’s comment: the sponsor classified this design as "patients serving as their own control" but this is in reality an uncontrolled study.

Reviewer’s comment: this study is classified as a multicenter study, but the design is such that all eligible subjects receive the first 6 weeks of product administration at one site (Harbor-UCLA Clinical Research Center) and would have to return to this site for efficacy assessments at 12, 26, 52 and 104 weeks of treatment.

Objectives

Eligibility Criteria

Inclusion criteria:

Exclusion Criteria:

Treatment assignment / randomization

In this non-randomized study all subjects were assigned to active treatment.

Product information and administration

Laronidase was diluted in normal saline with 0.1% (1.0 mg/mL) human serum albumin. The total laronidase dose was 125,000 U/kg (100 U/kg by redefined units), infused over a 3-4 hour period. Intravenous diphenidramine pre-medication was given 10-30 minutes prior to enzyme infusion. During the first hour, the enzyme was infused at a rate of 50 U/kg/min equivalent to 3,000 U/kg/hr. During the second and third hours, the rest of the dose was administered at a maximum rate of 61,000 U/kg/hr.

Evaluations

(except as noted, at baseline, weeks 6, 12, 26, 52, 104)

          1. Liver and spleen volume were assessed by MRI. Organ density assumed 1g/ml to enable calculation of organ weight as a percentage of body weight. An unblinded radiologist evaluated the MRI in real time. At study end a second radiologist blinded to subject identity and MRI timepoint analyzed organ volumes in a randomized fashion. These data were compared to healthy children and adolescents liver and spleen weights in proportion to the average weight for their age. Liver and spleen volume data is reported through week 104 in the unblinded extension study.
          2. Urinary GAG was collected at pre-treatment and weekly for the first 6 weeks, and at decreasing frequency thereafter. All determinations were made at Harbor-UCLA by a validated method using Alcyan blue dye and quantification by spectrophotometer. The values were corrected by urine creatinine concentration and expressed as micrograms of GAG per mg of creatinine.
          3. Joint range of motion (ROM) of shoulder, knee and elbow. Measurements obtained by a single physical therapist using a goniometer. 5 independent measurements for each joint and for each visit were obtained and averaged.
          4. Cardiac function was assessed in a prospectively defined but unvalidated "cardiac scoring system" which is a composite score of EKG (weight 3), echocardiogram (weight 5), history (weight 4), physical examination (weight 4), and radiological findings (weight 4). The subject was also assessed according to the New York Heart Association functional classification.
          5. Airway obstruction was assessed by a variety of parameters measured during polysomnography to investigate the frequency of apnea (cessation of airflow for 10 or more seconds), hypopnea (50% decrease in airflow per breath accompanied by arousal or desaturation), as well as minutes of hypoxia (oxygen saturation below 90%). Sleep studies were performed at pre-treatment, week 26, and if previously abnormal, at weeks 52 and 104. MRI assessments of the tongue and measurement of the airway index (ratio of mid-sagittal anterior-posterior width of the pharyngeal and tracheal airway to total width of the pharynx) were performed by an unblinded radiologist at pre-treatment and at weeks 6, 12, 26, 52 and 104.
          6. Eye disease was assessed by a combination of visual acuity testing, complete ophthalmologic exam, intraocular pressure measurements, corneal photographs.
          7. Central nervous system abnormalities were assessed by brain and cervical cord MRI, by lumbar puncture if there were no contraindications for the procedure, and by a pediatric neurologist examination. In addition the Wechsler Intelligence Scale was administered
          8. Genetic skeletal surveys .
          9. Height and weight were assessed from health records obtained from the 2 years prior to study to establish baseline growth rates, and every 4 weeks to week 52, and every 12 weeks to week 152 during treatment with rhIDU in pre-pubertal subjects (age 5 to 12). Only those assessments performed at Harbor-UCLA were used in the analysis of the results.
          10. Clinical Laboratory Safety assessments were performed periodically during the study. 24 hour creatinine clearance was performed at pre-treatment and weeks 12, 26, 52 and 104. Complement testing (CH50, C3 and C4) was done pre- and post-infusion, at pre-treatment and weeks 4, 6, 12, 26, 52, and 104. ELISA assessments for IgG serum antibodies against rhIDU with Western blot specificity confirmation were performed at pre-treatment and weeks 1, every other week to week 26, every 4 weeks to week 52, and every 12 weeks to week 152.
          11. Enzyme activity was assessed in buccal brushings. Enzyme activity was also assessed in leukocytes.
          12. Pharmacokinetic studies were performed at pre-treatment, and weeks 1, 2, 12 and 26. Pharmacokinetic studies were also performed on Week 6 for subjects 001 and 002, in addition to the other timepoints.
          13. Safety Monitoring

            A study monitor designated by BioMarin would monitor both safety and study conduct at the study sites.

            Endpoints

            Primary endpoints: The primary endpoints cited in the original protocol were:

            1. Proportion of subjects with 20 % reduction of the excess size of either liver or spleen or both. In order to calculate the excess organ size the sponsor proposed to subtract the normal organ size, which in the original protocol was estimated as 2.5 % of body weight for the liver and 0.2 % of body weight for the spleen.
            2. Reviewer’s comment: This endpoint has changed substantially in subsequent amendments of the protocol. Amendment 1 of 3/3/98 revised the definition of a significant reduction in organomegaly to a 20 % reduction in the total size of either the liver or the spleen or both. The 20% criterion was selected by reference to experience in Gaucher’s disease. However, it is unclear how to extrapolate the clinical meaning of hepatomegaly in one disorder to a different disorder.

            3. Percentage reduction in urinary GAG excretion as calculated by taking the average of the pretreatment samples compared with the average of samples from the last 6 weeks of therapy.

Reviewer’s comment: The definition of endpoint for analysis was very vague in the original protocol. Amendment 2 (9/13/98) revised the definition in the statistical analytical plan as "the four pre-treatment values will be averaged and compared with the last six specimens taken through week 26". As the study duration was extended, the timepoint for endpoint comparison has been modified.

A general comment related to the analysis of "efficacy" endpoints is that this study is uncontrolled, and variations in these endpoints would demonstrate, at best, bioactivity to guide more definitive studies.

Secondary endpoints:

  1. Cardiac ejection fraction, valvular regurgitation, pulmonary hypertension as assessed by the cardiac scoring system, and NYHA classification
  2. Corneal clouding, and visual acuity
  3. Joint stiffness / range of motion, based on goniometer, subjective reports of stiffness or pain, as well as video examinations of standard motions.

Statistical analysis

The revised analytic plan provided for specific analysis procedures which were not detailed in the original protocol.

Reviewer’s comments: In this uncontrolled study, pre- and post-treatment comparisons are not meaningful in support of laronidase efficacy.

Study Conduct

Study Conduct was monitored by 2 contract research organizations (CRO’s) during the study period: from 11/28/97 until 6/12/2000 Inveresk Research was charged with study monitoring and from 6/12/00 until the end of the study period Abt Associates Clinical Trials monitored study conduct.

This study report includes data collected through March 30, 2001. As of this date, the first patient enrolled had completed Week 171 and the last patient enrolled had completed Week 152.

Protocol Violations

The most common deviations involved variations from protocol-specified collection times for safety and efficacy assessments, particularly laboratory evaluations (urinalysis, urinalysis with microscopy, CBC, and blood chemistries) and urinary GAG measurements. Most variations resulted from missed study visits. Together the 10 subjects missed a mean of 19.8 infusions due to frequent unavailability of study drug from December 1998 through December 1999.

Reviewer’s comment: The sponsor does not explain the unavailability of laronidase during the study period. The sponsor states that 66% of missed infusions are due to laronidase unavailability, but this figure is inconsistent with the Appendix that lists protocol violations.

 

FDA site inspection findings

No site inspections have been performed for this study.

Results

Subject disposition

Ten subjects were enrolled into this study. The first subject signed the consent form and was enrolled on November 28, 1997, the first study dose was given on December 19, 1997, and the last subject completed the 152-week timepoint on March 29, 2001. All 10 subjects successfully completed the first 52 weeks of the study, and 8 subjects were active at Week 152. Subject 008 died between Weeks 103 and Week 104 of a viral illness. Subject 002 died 19 days after the patient’s last study drug infusion at Week 137 of complications following spinal fusion surgery for worsening of scoliosis.

Baseline characteristics

Table 2 shows the demographic and relevant baseline characteristics of the subjects enrolled in this study.

Table 2. Demographic and baseline characteristics

Subject

Age (yrs)

Height (cm)

Weight (kg)

Gender

Race

Clinical Status

001

17

132.5

38.1

M

White

MPS IH-S

002

10

121

22.6

F

White

MPS IH-S

003

9

122.5

27

M

White

MPS IH-S

004

8

125

34.6

M

White

MPS IH-S

005

12

127

24.4

M

White

MPS IH-S

006

22

152.7

64.5

M

White

MPS IH-S

007

17

160

57.2

F

White

MPS IS

008

5

87

14.8

F

White

MPS IH

009

9

118.5

24.2

F

White

MPS IH-S

010

14

160

54.6

M

White

MPS IH-S

There were 6 pre-pubertal subjects in the study. These subjects covered a broad spectrum of clinical presentation. The Principal Investigator / Medical Monitor used a protocol classification of MPS I phenotypes to describe the subjects disorder severity at the time of enrollment. Under this classification one subject had mild disease (MPS IS, Scheie), 8 subjects had moderate disease (MPS IH-S, Hurler-Scheie) and one subject had severe disease (MPS IH, Hurler). All subjects had in common the wide spectrum of MPS I impairments, with unknown and variable intensity in each affected organ, tissue or functional system. Most of the concomitant medications used by these subjects at the time of entry were analgesics or anti-inflammatory agents for relief of headaches or other pains.

Reviewer’s comment: It is important to note that the phenotype classification is based on a subjective impression of the overall degree of organ or tissue impairment seen at a specific age range.

Study drug exposure

Ten subjects participated in this uncontrolled, open label study and have received by the time of the study report weekly intravenous infusions of rhIDU for an average 151 weeks (range 103 – 171).

The mean (± SD) number of administered infusions per subject was 115.0 (± 17.5, range 82 to 135). As mentioned under Protocol Violations, the majority of missed infusions were due to unavailability of the study drug. Most of the other infusions were missed because of patient illnesses or surgeries, parents’ illnesses, logistical problems in getting to the clinic, vacations, or decisions on the part of the subject or parents to receive infusions less frequently than once a week.

Primary endpoints

Hepatomegaly

Results of the blinded review of liver volumes are reported in Table 3. Due to the wide variation in liver volumes at baseline, these results were normalized as percent reduction from baseline for each subject and averaged.

Table 3. Mean normalized liver volume as a percentage of pre-treatment

Time

Mean

SD

# Subjects with ≥ 20 % decrease

Pre-treatment

100

0

N/A

Week 6

79.9

6.5

5

Week 12

78.2

6.1

6

Week 26

76.6

10.2

8

Week 52

75.0

9.2

7

The 3 subjects that did not achieve the stated endpoint of 20 % reduction had liver volumes in the normal range for their ages by week 52, and achieved the 20% or greater reduction in liver volume in the unblinded readings at week 104.

Splenomegaly

Results of MRI blinded review of the spleen volumes paralleled those obtained for liver volumes. 7 of the 10 subjects demonstrated a ≥ 20% reduction at week 6. Subject 009 developed clarithromycin-induced hepatitis just prior to the Week 26 imaging timepoint, with a sharp increase in liver and particularly spleen size to twice her baseline volume. As her serum liver enzymes normalized by week 46, both the liver and spleen volumes returned to normal.

In conclusion, using blinded MRI readings, 7 of 10 subjects had a greater than 20% reduction in liver volumes at Week 52, and 5 of the 10 subjects had a 20% reduction in spleen volume at Weeks 26 and 52. Using unblinded MRI readings, 9 of the 10 subjects had normalized liver size by week 52 (8 / 9 by week 104) and 2 of the ten subjects had normalized spleen volumes by weeks 52 (1 / 9 by week 104).

BioMarin believes (----------------------------------------------------------) that the reduction in liver and spleen sizes caused improvements in comfort and edurance and the ability to eat and breathe.

Reviewer’s comment: It is important to note that even though the reduction in liver and spleen volumes was clearly demonstrated, its impact on general well being of the subjects may have been overstated by other factors, including the fact that this is an open label, uncontrolled study.

Urinary Glycosaminoglycans

All subjects had elevated urinary GAG levels prior to treatment that were generally proportional to the severity of disease in this limited sample of MPS I subjects. For example, the most severely affected subject (008) had urinary GAG exceeding 500 µg GAG/mg creatinine, whereas the subjects with the mildest disease (003 and 007) had levels of 63.25–112.8 µg GAG/mg creatinine. Table 4 shows the mean normalized urinary GAG level reported as percentage of baseline from pre-treatment to week 152. By week 152 mean GAG levels were within the normal range.

Table 4. Mean normalized urinary GAG as a percentage of pre-treatment

Time

Mean

SD

n

# Subjects with ≥ 50 % decrease

Pre-treatment

100

0

10

N/A

Week 6

31.8

5.9

10

10

Week 12

33.6

9.0

10

10

Week 26

31.3

8.2

10

10

Week 52

37.2

11.3

10

8

Week 104

26.4

6.7

9*

9

Week 152

21.5

9.3

7**

7

* Subject 008 died before week 104

** Subject 002 died before week 152

For comparison, mean (± SD) urinary GAG in 68 healthy subjects is as follows: 32.4 ± 6.9 μg GAG/ mg creatinine for ages 3 – 12 (n=42), 14.9 ± 3.4 μg GAG/mg creatinine for ages 13-18 (n=12), and 8.5 ± 1.8 μg GAG/mg creatinine for ages 19-52.

Reviewer’s comment: This study provides indication of laronidase activity through the substantial but incomplete reduction in urinary GAG levels. The sites of enzyme action and the extent of tissue and organ function recovery cannot be inferred from these data. Similarly, the clinical significance of this finding remains unclear.

Secondary endpoints

Joint ROM

Subjects were evaluated for shoulder flexion and extension, knee flexion and extension, and elbow extension. Results were analyzed for changes in ROM (the angle of a limb relative to the body at maximal flexion or extension) and changes in degrees of restriction of ROM (mean degrees of movement in an age-adjusted normal population minus the degrees of movement in the MPS I subjects).

By Week 26, the majority of subjects showed some improvement in ROM in one or more joints, although inter-subject variability was large. The movements that showed the most improvements at Week 52 were shoulder flexion, knee extension, and elbow extension (Table 5)

Table 5. Mean (± SD) Joint ROM and changes from pre-treatment

Joint

Time

Mean ± SD (angle)

Mean (± SD) change from pre-treatment

R Shoulder flexion

Pre-treatment

100.6 ± 17.5

28.1 ± 21.5

Week 52

128.7 ± 16.2

L Shoulder flexion

Pre-treatment

101.2 ± 18.5

26.1 ± 26.8

Week 52

127.4 ± 17.6

R Shoulder extension

Pre-treatment

32.1 ± 14.0

5.1 ± 19.1

Week 52

37.2 ± 7.1

L Shoulder extension

Pre-treatment

26.8 ± 6.9

7.6 ± 12.8

Week 52

34.4 ± 7.5

R Elbow extension

Pre-treatment

156.7 ± 13.4

7.0 ± 6.3

Week 52

163.7 ± 13.6

L Elbow extension

Pre-treatment

156.1 ± 19.1

7.1 ± 9.0

Week 52

163.2 ± 16.7

R Knee flexion

Pre-treatment

126.6 ± 14.1

4.7 ± 5.0

Week 52

131.3 ± 14.4

L Knee flexion

Pre-treatment

126.6 ± 19.7

3.6 ± 8.8

Week 52

130.1 ± 13.7

R Knee extension

Pre-treatment

172.0 ± 11.0

3.5 ± 10.4

Week 52

175.5 ± 5.6

L Knee extension

Pre-treatment

171.3 ± 11.0

2.8 ± 6.4

Week 52

174.1 ± 6.3

Reviewer’s comment: The increase in ROM for some joints was modest, although the mean changes seem to be favorable to the laronidase treatment. Caution should be exercised in the interpretation of this unblinded uncontrolled study in correlating the modest ROM changes to improvements seen in daily activities.

Cardiac Function

The subjects in this study exhibited typical MPS I cardiac disease, including valvular insufficiency, pulmonary hypertension, and congestive heart failure, although only Subject 006 was in serious heart failure at the start of the study.

At Week 26, 3 of 10 subjects had improvements in NYHA scores. After 52 and 104 weeks of treatment, NYHA scores improved in 10 of 10 and 9 of 9 subjects, respectively, with 6 subjects achieving normal scores of Class I. No subject was in Class I at pretreatment.

Tricuspid regurgitation also modestly improved some subjects.

Reviewer’s comment: The data suggest inconsistent echocardiographic improvement in a few subjects with underlying cardiac disorders in this small uncontrolled study. In contrast the improvement seen using the NYHA scores is more open to the bias inherent in open label studies. These findings may not be surprising in view of the lack of effect of laronidase on GAG accumulation in heart valve and myocardium in the canine model of MPS I.

Polysomnography and airway evaluations

Six subjects who had sleep apnea at pre-treatment demonstrated improvement in the number of apnea episodes by week 26. Five of the 9 subjects with pre-treatment hypopneas showed improvement in the number of hypopneas per night by week 26. The apnea hypopnea index (AHI) improved in 7 out of 10 subjects, and the mean AHI decreased from 2.08 to 0.97 (53% reduction) at week 26. However, some subjects had worsening of some parameters.

Other measures of airway function (airway index and tongue diameter) showed inconclusive results, in part due to technical difficulties

Eye disease

Three subjects had improvement in visual acuity by week 52, and with additional favorable changes noted at week 104. No changes were observed in intra-ocular pressure or in corneal clouding.

CNS abnormalities

MRI studies of the brain and cervical cord pretreatment demonstrated the substantial pathology expected in MPS I patients. This included perivascular storage, meningeal thickening and spinal cord compression or restriction, and miscellaneous other findings.

There were no substantial changes seen on MRI during the 104-week course of laronidase treatment.

Reviewer’s comment: These findings are consistent with the lack of enzyme activity in the central nervous system. They are also similar to those pre-clinical studies showing absence of CNS effects in dogs affected by MPS I.

Bone evaluations

No significant changes were observed in the genetic skeletal surveys performed during the study. All subjects had radiologic evidence of odontoid dysplasia and dysostosis multiplex, commonly seen in patients with MPS I.

Height and weight

In 6 pre-pubertal subjects, height increased by a mean of 6.0 cm at Week 52 and a mean of 10.0 cm at Week 104. Weight increased by a mean of 4.2 kg at Week 52 and a mean of 7.9 kg at Week 104 in these subjects.

Height growth rate normalized for 3 of 6 pre-pubertal subjects at Week 52. Weight growth rate normalized for 2 of the 6 pre-pubertal subjects at Week 52.

Pharmacokinetics

Intravenous administration of laronidase (125,000 U/kg) resulted in significant circulating plasma IDU activity levels of IDU for periods of 3 hours or more.

The circulating levels of IDU achieved peak at 100-200 U/ml for the majority of infusions, which was approximately 10 times the half-maximal uptake of the enzyme in vitro.

The mean circulating half-life (t˝) of IDU was approximately 1.8 to 1.9 hours at Weeks 1 and 2 and decreased to 1.2 to 1.4 hours at Weeks 12 and 26 in 7 of the 10 subjects.

Differences in pharmacokinetic parameters did not correspond to differences in urinary GAG excretion.

In examining the question of factors responsible for variations in pharmacokinetic parameters, the sponsor found that presence of antibodies to laronidase did not correlate with a decrease in terminal half-lives by linear regression analysis. The 3 subjects with consistent IDU-specific antibodies did not have consistent and clear changes—either increased or decreased half-life—compared to patients without IDU-specific antibodies.

In addition, no relationship was found between urinary GAG levels, enzyme activity in buccal mucosa brushings and pharmacokinetic parameters.

Enzyme uptake into buccal mucosa and leukocytes

Prior to treatment all subjects had very little or undetectable a - L –iduronidase in their buccal mucosa or peripheral leukocytes.

By the Week 2 pre-infusion assessment, enzyme was detectable in the buccal mucosa, reaching an average of about 1% of normal in all patients. Leukocyte enzyme levels have also shown persistently higher levels during treatment (mean of 18%, 12% and 35% of normal at weeks 26, 52 and 104 respectively.

Safety

Adverse events

All 10 subjects had at least 1 adverse event, and at least one deemed to be study drug-related. There were a total of 960 adverse events reported. Eight subjects experienced 32 serious adverse events, including 2 that resulted in death. Seven subjects had 33 adverse events reported as severe.

The most common adverse events reported during the study were: rhinitis (10 subjects), pain, asthenia and fever (9 subjects each), and increased cough, abdominal pain, vascular disorder, and rash (8 subjects each).

The most common of the 33 severe adverse events reported during the study were: bone disorder (including cervical cord compression and spinal deformities), apnea and vascular disorders, allergic reactions and headaches. Ten of these 33 severe events were judged to be definitely or possibly related to the study drug. Fourteen of these 33 severe adverse events were also serious adverse events.

Adverse events were also classified by having occurred on an infusion date after the start of the infusion or as having occurred in non-infusion days. The infusion day adverse events reported by most subjects were: rash, urticaria, allergic reactions and headaches and adverse events with the greatest occurrences were: urticaria (97 events), rash, and angioedema.

All subjects were pre-medicated with diphenidramine intravenously before the laronidase infusion, and some subjects also received corticosteroids and nonsteroidal anti-inflammatory drugs. If a reaction occurred during an infusion, the rate of infusion was slowed or temporarily stopped, the dosage could temporarily be reduced, additional doses of diphenidramine or corticosteroids could be administered or other interventions could be performed.

Reviewer’s comment: Approaches to prevent or treat an infusion reaction were so varied that guidelines for a label would be problematic.

Deaths

There were 2 deaths that occurred during the 152-week portion of the study.

Subject 008, age 7, died 4 days after the Week 103 infusion, of apnea (respiratory arrest). Respiratory distress occurred during a flight for the Week 104 assessment, followed by respiratory arrest, without response to intubation and full CPR efforts.. Autopsy findings indicated an apparent systemic viral illness with histologic evidence of an active lymphocytic myocarditis and patchy bronchiolitis. Subject 008 had a high antibody titer, including IgG specifically to laronidase, which persisted until the last assessment at week 100. Complement activation was seen at weeks 6 and 12, but not subsequently. Autopsy findings included mild immune complex deposits in glomerular basement membranes, and mild focal deposition of IgM in lung capillaries with minimal C3, but without any pathologic change by light or electron microscopy in either organ to suggest a functional effect. The subject’s residual underlying MPS I disease and/or viral illness may have contributed to death. This subject had received weekly infusion premedication with methylprednisolone starting at Week 8 and continuing throughout the study due to hypersensitivity-type reactions. Study drug did not appear to be directly implicated based on history and autopsy findings.

Reviewer’s comment: This death is likely related to acute complications of a viral illness in a subject with severe manifestations of the disorder. The use of systemic steroids as pretreatment for the enzyme infusions may have also contributed.

Subject 002, age 13, died 19 days after the subject’s last study infusion at Week 137, of cardiac and respiratory arrest. This subject had a significant medical history of musculoskeletal disease at study enrollment, including scoliosis. The subject was admitted to the hospital for posterior spinal fusion for worsening of her scoliosis from T5 to L4. Postoperatively, the subject was unable to move the lower extremities, and through additional testing and procedures, was diagnosed with spinal cord injury and ascending quadriparesis which progressed to respiratory failure. The subject and family requested discontinuation of the ventilation leading to the subjects death.

Laboratory Abnormalities

Serum Chemistries: Most subjects had elevated serum levels of alkaline phosphatase at screening and pre-treatment. These levels generally decreased by week 8, although for some subjects levels declined and remained above normal for the study period. One subject had an elevated serum LDH level at week 8, which returned to normal by week 10. One subject had an episode of presumed macrolide hepatotoxicity from week 26 until week 30.

Hematology: Several subjects had chronic anemia at pre-treatment and the indices related to red blood cell remained stable throughout the study period.

Antibody development

All subjects developed IgG antibodies to the laronidase product as detected by ELISA by week 6 or 12 of treatment. Serum antibody levels generally declined over time. Using a Western blot technique, the sponsor determined that the initial antibodies detected were directed against a 60 KD protein impurity. Specific immune responses to rhIDU were seen in 4 subjects by Western Blot which were either transient or declined with time.

IgG titer did not correlate with hypersensitivity-type reactions in these patients.

There was no apparent correlation between IgG antibody titer to laronidase product and reductions in excess urinary GAG levels, liver or spleen size. These antibodies did not neutralize enzyme activity.

Complement activation, as measured by the difference between the pre-and postdose levels of CH50 and either C3 or C4, was highest at Week 6 or 12 in 4 subjects, but was greatly reduced or resolved by Week 26 and absent at Week 52. There was no correlation between complement activation and need to receive glucocorticoids as pre-medications for infusion or to manage hypersensitivity-type reactions during infusions. There were no apparent effects of the immune responses on development of immune complex disease or on glomerulonephritis based on urinalysis and GFR results.

Summary

This uncontrolled Phase 1 study of laronidase in 10 subjects with MPS I was able to demonstrate bioactivity in clinical areas that correspond to large accumulation of GAG’s such as the reticulo-endothelial system and the kidneys, resulting in substantial reductions of hepatosplenomegaly and excretion of urinary GAG’s. An effect on other endpoints, particularly of clinical significance, was not demonstrated in this study. Some data suggested a favorable trend, but it is imperative to note that these data obtained from an open label uncontrolled study cannot provide support for laronidase efficacy in MPS I.

All subjects had infusion associated reactions, which were mild or moderate. All subjects had antibodies to laronidase as assessed by ELISA, but specific anti-IDU antibodies were confirmed in 4 subjects. 2 deaths occurred during this study, but these seem to be unrelated to the study drug, and consistent with the natural history of the severe presentation of MPS I.

Study ALID-003

Overview

This is the only Phase 3, double blind, placebo-controlled, randomized clinical trial conducted for this product to support a claim of efficacy and safety in the treatment of patients with MPS I. The study started on December 28, 2000 and was completed on September 6, 2001.

At the end of the 26-week study period, all subjects were offered participation in an open label extension study (ALID-006).

Protocol

Title: "A Randomized, Double-Blind, Placebo-Controlled, Multicenter, Multinational, Clinical Study of Recombinant Human Alpha-L-Iduronidase in Patients with Mucopolysaccharidosis Type I". Protocol ALID-003-99.

Design

This was a randomized, double-blind, placebo-controlled, multicenter, multinational study. The study was divided into 2 consecutive phases: a 2-week period to assess eligibility and obtain baseline parameters, and a 26-week treatment phase in which subjects are randomized to receive weekly laronidase or placebo.

Objectives

Eligibility Criteria

Inclusion criteria:

Exclusion criteria:

Treatment assignment / Randomization

Eligible subjects were randomized to the laronidase or placebo group according to a central randomization scheme blocked by site. All subjects, investigators, site personnel, and members of the sponsor’s staff were masked to the treatment assignment. The computer-generated randomization codes were held by Genzyme Corporation.

Product information and administration

Subjects randomized to the treatment group received laronidase at a dose of 100 U / kg intravenously weekly during the treatment phase for 26 weeks. Placebo subjects received a pH-adjusted, phosphate-buffered solution. Study solutions were infused weekly, with a minimum interval between consecutive infusions being 4 days.

For administration to subjects, the laronidase or placebo were diluted with between 100 mL to 250 mL (depending on the subject’s weight) of 0.1% human serum albumin in saline.

All subjects were pretreated 30 minutes to one hour prior to infusion with an antipyretic regimen (acetaminophen or ibuprofen) and an antihistamine regimen (diphenhydramine, hydroxyzine, chlorpheniramine, cetirizine, fexofenadine or loratadine).

The initial infusion rate of 0.01 mg/kg (2 U/kg) of body weight per hour for 15 minutes was incrementally increased if well tolerated to a maximum of 0.25 mg/kg (43 U/kg) of body weight per hour to deliver the total volume of the infusion over approximately 4 hours. The total infusion volume for subjects weighing between 5 and 20 kg was 100 mL and for subjects weighing between 21 and 100 kg was 250 mL

If a patient was unable to receive a scheduled treatment infusion within the 10-day maximum period allowed following the last scheduled infusion, that patient received a treatment infusion at the earliest possible date. After such an out-of-schedule treatment infusion, the next infusion was given within 7 ± 3 days. Subsequent treatment infusions were given according to the original weekly treatment infusion schedule.

 

Evaluations

(baseline and Weeks 4, 8, 12, 16, 20, 26 except as noted)

a. Forced Vital Capacity (FVC)

        1. Six-minute walk distance, reported in meters walked. Baseline exam repeated 3 times for training, and third trial used as baseline.
        2. Children’s Health Assessment Questionnaire (CHAQ) for subjects 5 –18 years of age or Health Assessment Questionnaire (HAQ) for subjects 19 years of age or older. (assessed at baseline, weeks 4, 12, and 26)
        3. Sleep study (to measure apnea/ hypopnea events and oxygen desaturation) are assessed at baseline, and weeks 4, 12, and 26
        4. Liver volume by MRI at baseline and week 26
        5. Urinary Glycosaminoglycans (GAG)
        6. Joint Range of Motion at baseline and weeks 12 and 26
        7. Child Health Questionnaire or SF-36 at baseline and weeks 12 and 26
        8. Resource utilization
        9. Standing heights for measurement of growth velocity
        10. Investigator global assessment at week 26
        11. Visual acuity by standard eye chart testing, ocular pressure, corneal clouding, and retina / optic nerve exam at baseline and week 26
        12. EKG and echocardiogram at baseline and week 26
        13. FEV1, total lung capacity and diffusing capacity at baseline and week 26
        14. Parent/caregiver quality of life at baseline and week 26
        15. Pharmacokinetic assessments at 2 selected sites for all enrolled subjects at those sites at week 1, 12 and 26, with samples collected at pre-dose, 15, 45, 90 minutes of infusion, 3 and 4 hours of infusion, 10, 30, 45, 60 minutes post-infusion, 2, 3, 4 and 6 hours post-infusion.
        16. Medical history, VS, and physical examination
        17. Brain / cranio-cervical junction MRI at baseline (all subjects) and at the investigator’s discretion at week 26
        18. Laboratory evaluations at baseline and weeks 4, 12 and 26, including clinical chemistry, hematology and urinalysis. Urine pregnancy test for females of childbearing potential at baseline and every 4 weeks.
        19. Antibody testing at baseline and every 4 weeks. Complement activation (CH100 or CH50 and C3 or C4 components) was measured if symptoms of a hypersensitivity reaction are noted.
        20. Adverse events

Safety Monitoring

An independent Allergic Reaction Review Board (ARRB) was created to review signs of moderate or severe hypersensitivity and provide guidance on management of these reactions. The ARRB interacted with the Genzyme Pharmacovigilance group and infrequently directly with investigators.

Endpoints