- Email: [email protected]
A randomized trial comparing the efficacy and safety of imiglucerase (Cerezyme) infusions every 4 weeks versus every 2 weeks in the maintenance therapy of adult patients with Gaucher disease type 1
Molecular Genetics and Metabolism 96 (2009) 164–170
Contents lists available at ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
A randomized trial comparing the efficacy and safety of imiglucerase (Cerezyme) infusions every 4 weeks versus every 2 weeks in the maintenance therapy of adult patients with Gaucher disease type 1 P.S. Kishnani a,*, M. DiRocco b, P. Kaplan c, A. Mehta d, G.M. Pastores e, S.E. Smith f, A.C. Puga f, R.M. Lemay f, N.J. Weinreb g a
Division of Medical Genetics, Department of Pediatrics, Duke University Medical Center, Box 103856 DUMC, 595 LaSalle St., GSRB1, 4th Floor, Durham, NC 27710, USA Unit of Rare Diseases, Department of Pediatrics, Gaslini Institute, Genoa, Italy Section of Metabolic Diseases, Department of Pediatrics, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA d Department of Haematology, Royal Free Hampstead NHS Trust, London, UK e Departments of Neurology and Pediatrics, New York University School of Medicine, New York, NY, USA f Genzyme Corporation, Cambridge, MA, USA g Gaucher Disease Treatment Center, University Research Foundation for Lysosomal Storage Disease, Inc., Coral Springs, FL, USA b c
a r t i c l e
i n f o
Article history: Received 13 October 2008 Received in revised form 22 December 2008 Accepted 23 December 2008 Available online 4 February 2009 Keywords: Enzyme replacement therapy Gaucher disease Glucocerebrosidase Glucosylceramide b-glucosidase Imiglucerase Infusion frequency Type 1
a b s t r a c t Imiglucerase (CerezymeÒ) has been the standard of care for treatment of Gaucher disease, a lysosomal storage disorder resulting from deficiency of glucocerebrosidase, since its approval in 1994. Infusions are typically given once every 2 weeks. However, many patients have expressed a desire for less frequent infusions as a matter of convenience. This clinical study assessed the safety and efficacy of intravenous imiglucerase infused once every 4 weeks (Q4) compared to once every 2 weeks (Q2) at the same total monthly dose in adult patients with clinically stable Gaucher disease type 1 (GD1). This was a 24-month, open-label, randomized, Phase 4, dose-frequency study conducted in 25 centers worldwide. Patients receiving imiglucerase were randomized to receive their monthly dose biweekly (n = 33) or every 4 weeks (n = 62). Changes from baseline in hemoglobin, platelets, liver and spleen volumes, bone crisis, and bone disease comprised a predefined composite endpoint; achievement or maintenance of established Gaucher disease therapeutic goals comprised a secondary endpoint. Sixty-three percent of Q4- and 81% of Q2-treated patients met the composite endpoint at Month 24; 89% of Q4- and 100% of Q2-treated patients met the therapeutic goals-based endpoint. The frequency of related adverse events was comparable between treatment groups. This study suggests that with comprehensive monitoring, a Q4 imiglucerase infusion regimen may be a safe and effective treatment option for the majority of clinically stable adult patients with GD1 but may not be appropriate for all GD1 patients. Continued monitoring in patients treated with Q4 dosing is required to assess long-term effectiveness. Ó 2008 Elsevier Inc. All rights reserved.
Gaucher disease is an inherited disorder characterized by insufficient activity of the enzyme acid b-glucosidase (glucocerebrosidase), resulting in the accumulation of glucosylceramide in the bone marrow compartment, liver, and spleen, and less often in the lungs [1]. The most common clinical manifestations of the disease include anemia, thrombocytopenia, hepatosplenomegaly and skeletal complications such as bone pain, bone crisis, bone lesions, cortico-medullary infarctions, medullary expansion, osteopenia, osteonecrosis, and pathological fractures [1–3]. The most common clinical variant is Gaucher disease type 1 (GD1) [1]. Treatment with enzyme replacement with imiglucerase (CerezymeÒ, Genzyme Corporation), a modified form of recombinant
* Corresponding author. Fax: +1 919 684 8944. E-mail address: [email protected] (P.S. Kishnani). 1096-7192/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2008.12.015
human b-glucocerebrosidase specifically targeted for uptake by macrophages, is currently the standard of care for symptomatic patients with GD1. Imiglucerase has been available since 1994, with over 5000 patients currently being treated [4]. Regular infusions with imiglucerase result in regression of the clinical manifestations of GD1 as evidenced by correction or amelioration of anemia and thrombocytopenia, reduction of hepatosplenomegaly, improvement in bone mineral density [5], diminished chronic bone pain, reduced frequency of bone crises [6], and improved responses on standardized patient health surveys [7–10]. Because of the highly variable manifestations and progression of Gaucher disease, disease treatment is individualized so that the imiglucerase dose and infusion regimen result in an optimal outcome for the patient [3]. The choice of starting imiglucerase dose and subsequent dosage adjustments are often influenced by an assessment
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
of age at initial presentation (generally, the earlier the symptoms, the more severe the disease), overall disease severity and progression, and the degree to which individualized therapeutic goals are achieved [3,11]. Regardless of dose, the infusion schedule is usually maintained at an interval of once every 2 weeks with an average infusion time of 1–2 h. Many patients, especially those who have milder or stable symptoms, have expressed a desire for less frequent infusions as a matter of convenience. Except for a single-institution, 11-patient study [12], reduced frequency infusions have never been prospectively evaluated as a maintenance regimen. We designed a randomized clinical trial to compare the safety and efficacy of the most commonly prescribed dosing regimen for imiglucerase (one infusion every 2 weeks) with a less frequent but equivalent overall dose infusion regimen (one infusion every 4 weeks). Materials and methods
165
at Month 24 (or discontinuation) based on a composite of hematologic, skeletal, and visceral measures (composite endpoint). To be classified as a success with respect to the composite endpoint, individual patients had to meet all five of the following criteria: Primary composite endpoint: Hemoglobin did not fall more than 1.25 g/dL for women or 1.5 g/ dL for men below the patient’s baseline value. Platelet count did not fall more than 25% below the patient’s baseline value, and did not fall below 80,000 mm3. Liver and spleen volumes were not greater than 20% above the patient’s baseline value. No new finding or progression of bone disease, including no new pathologic fractures, medullary infarction, lytic lesions or avascular necrosis. No bone crises during the study.
Study design This was a 24-month, open-label, multicenter, international, randomized, dose-frequency study conducted in adult patients (P18 years) with GD1. The study was initiated following the approval of the appropriate regulatory authorities and the approval of institutional ethics committees for each of the 25 trial sites worldwide. Eligible patients were randomized 2:1 to receive imiglucerase either once every 4 weeks (Q4) or once every 2 weeks (Q2) for 24 months. Patients in both treatment arms received approximately the same total monthly dose of imiglucerase as the average monthly dose the patient had received during the 6 months prior to study enrollment. After patients gave informed consent, medical history was obtained and a routine physical examination conducted, including clinical laboratory measures. Skeletal and visceral assessments and a general health survey (SF-36) were completed during the screening period. An independent data monitoring committee periodically reviewed source documents to advise the sponsor of any safety or efficacy concerns.
The original protocol was amended to account for fluctuations observed over time in hemoglobin levels, due in part to physiological variations or varying techniques at the site-based laboratories. The original criterion for hemoglobin was modified from an allowable decline of 1.0 g/dL at Month 24/discontinuation to a decline of not more than 1.25 g/dL for women and 1.5 g/dL for men. Intra-patient measurement variability was also observed in liver and spleen volume measurements early in the study. Therefore, the original criteria for liver and spleen volumes were modified from an allowable increase of no greater than 10% at Month 24 to an increase of no >20%. These modified criteria were used for the final analyses. Therapeutic goals endpoint In 2004, after the initiation of this study, therapeutic goals for the management of patients with GD1 were established [11]. The hematologic (hemoglobin and platelets) and visceral (liver and spleen volume) criteria of the primary composite endpoint were therefore re-defined in accordance with these particular therapeutic goals. To be classified a therapeutic success according to these goals, a patient had to meet all of the following criteria:
Eligibility criteria Therapeutic goals-based endpoint: Adult patients at least 18 years of age with a confirmed diagnosis of Gaucher disease (documented deficiency of acid b-glucosidase by enzyme assay) were eligible for this study if they met the following: at least 2 years of imiglucerase therapy, with a stable dose of 20–60 U/kg every 2 weeks for at least 6 months prior to enrollment, and evidence of stable GD1 at baseline as defined by: hemoglobin P11 g/dL (women) or P12 g/dL (men); platelet count >100,000/mm3; liver volume 61.8 multiples of normal (MN); spleen volume 610 MN (patients who had undergone partial or total splenectomy were also eligible); aspartate amino-transferase (AST) and alanine amino-transferase (ALT) < 2.0 the upper limit of normal (ULN); total bilirubin <2.0 ULN and serum creatinine <2.0 mg/dL; no evidence of new pathologic bone fractures, medullary infarctions, lytic lesions, or avascular necrosis secondary to Gaucher disease (confirmed by X-ray within 6 months of the study); and no evidence of bone crisis (defined as pain with acute onset requiring immobilization, narcotics, increased white blood cell (WBC) count, periosteal elevation, fever or debilitation >3 days) within 12 months of the study. Primary efficacy endpoint The primary efficacy endpoint was the proportion of patients who successfully maintained their baseline level of clinical response
Hemoglobin: >11 g/dL (women); >12 g/dL (men) Platelets: >100,000 mm3 Liver volume: <1.25 MN Spleen volume: <8 MN Skeletal assessments as described for the primary composite endpoint.
Additional assessments The key efficacy and safety assessments are provided in Table 1. Patient-reported health status was assessed using the SF-36 Health Survey (Standard Version) [13] at baseline and at 12 and 24 months. Laboratory parameters included biomarkers (ACE, TRAP and chitotriosidase) in addition to blood chemistry. Safety measures Safety was assessed by adverse event monitoring, clinical laboratory evaluations, and physical examinations. Patients could be withdrawn at the discretion of the investigator for non-compliance, intolerable adverse events, or failure on one or more of the composite endpoint criteria. Patients were free to withdraw consent and discontinue from the study at any time.
166
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
Table 1 Key study assessments.
Table 2 Patient disposition.
Assessment
Timepoint(s)
Laboratory and other assessments Hemoglobin and platelets Chemistry panela ACE, TRAP and chitotriosidaseb
Baseline; every 3 months Screening; month 24/discontinuation Baseline; every 3 months
Visceral assessments Liver and spleen volumec
Every 6 months
Skeletal assessments X-ray of femora/spined MRI of femorae Bone pain, bone crisis and mobilityf
Baseline; months 12, 24/discontinuation Baseline; months 12, 24/discontinuation Every 3 months
Patient questionnaire SF-36 general health survey
Baseline; every 6 months
Safety assessments Physical examination Adverse events Serum pregnancy testg
Baseline; every 6 months Every visit Baseline
Q2
Q4
Overall
Number of patients screened Number of patients randomized, N Number of patients in ITT Population, n Evaluable patients (ITT)a n (%)
37 33 26 (78.8)
65 62 57 (91.9)
130 102 95 83 (87.3)
Completion status (ITT) n (%) Completed study Discontinued study Adverse events Clinical failureb Otherc
26 (78.8) 7 (21.2) 1 (3.0) 2 (6.1) 4 (12.1)
40 (64.5) 22 (35.5) 5 (8.1) 13 (21.0) 4 (6.5)
66 (69.5) 29 (30.5) 6 (6.3) 15 (15.8) 8 (8.4)
a
a Included albumin, alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase, direct bilirubin, blood urea nitrogen, calcium, serum creatinine, lactate dehydrogenase, phosphorus, and total protein. b TRAP, tartrate resistant acid phosphatase; ACE, angiotensin converting enzyme; CHITO, chitotriosidase. c Volume per milliliter (mL) calculated from 0.2% of body weight assuming 1 g/ mL for volumes from computed tomography (CT) or volumetric magnetic resonance imaging (MRI) using a consistent modality. Multiples of normal (MN) calculated by dividing volume (mL) by [weight(kg) 25] for liver ([weight 2] for spleen). d Erlenmeyer flask deformity, lytic lesions, and fractures per anterior/posterior view of the entire femora and lateral view of the spine. e Marrow infiltration, medullary infarction and avascular necrosis assessed per coronal, T1 and T2-weighted MRI. f Bone pain, bone crisis and mobility per patient interview at study visits. g Serum b-human chorionic gonadotropin performed within two weeks of randomization for women of childbearing potential.
Statistical methods A sample size of 90 was chosen assuming a 95% success rate for the Q2 group and 80% for the Q4 group, using a 95% confidence interval (CI) for the difference between the percentage of success for the two treatment groups at Month 24 or last visit. All patients who enrolled in the study and received at least one infusion were included in the primary analysis population, the Intent-to-Treat (ITT). The safety population was equivalent to ITT. The primary efficacy endpoint was the proportion of ITT patients with a known outcome who were a success at Month 24 (or last visit if discontinued). A 95% two-sided CI was calculated for each treatment group using the exact method of Blyth-StillCastella [14,15]. A 95% two-sided CI was calculated for the difference between treatment groups using the exact method of Agresti and Min [16]. Patients who failed any of the criteria comprising the composite endpoint were classified as failures, even if they did not have complete testing data for all composite endpoint parameters. Patients who did not fail any of the composite endpoint criteria but who did not have complete data for all required tests were classified as unevaluable and were excluded from the primary analysis. The analysis based on the therapeutic goals-based endpoint was performed similarly. Formal statistical analysis was not performed for the SF-36 survey. Results Patient disposition, demographics, and disease history A total of 102 patients were randomized, and 95 patients received at least one on-study infusion and are included in the ITT population (Q2, n = 33; Q4, n = 62). Twenty-six of 33 (78.8%) Q2-
Based on the number of evaluable patients in the primary analysis. Withdrawn by investigator after failing on one or more criteria of the composite endpoint. c Other reasons for discontinuation: Q2: 1 pregnancy, 1 discontinuation at baseline, 1 noncompliant, 1 wishes to withdraw. Q4: 1 pregnancy, 1 return to Q2 regimen, 1 clinical baseline issue, 1 wishes to withdraw. b
treated and 40 of 62 (64.5%) Q4-treated patients completed the study to Month 24. Two Q2-treated patients and 13 Q4-treated patients were withdrawn by the treating physician after clinical deterioration led to failure to meet one or more of the criteria of the primary endpoint. A full summary of patient disposition is provided in Table 2. Baseline demographics of patients are detailed in Table 3. The mean age at randomization was 47 years (18–82 years). Fortyeight percent of patients were of Ashkenazi Jewish descent. Of note, patients were on imiglucerase for a median of 6.9 (Q2) and 5.2 (Q4) years at enrollment, at a mean imiglucerase dose of approximately 70 U/kg/4 wks in both groups. Baseline indicators of Gaucher disease were generally similar between treatment groups (Table 4). The majority (73%) of patients had an intact spleen and splenomegaly (overall median spleen volume 3.4 MN). One patient who did not meet entry criteria for organ volumes was inadvertently randomized and treated within the study, becoming part of the ITT population for the Q2 group. Baseline disease characteristics for the groups are presented in Table 4; organ volumes are summarized both with and without the inadvertently randomized patient. Overall, 40% of patients had evidence of medullary infarction or avascular necrosis (AVN) at baseline. Absent any historical, clinical, or radiologic evidence that these events occurred within 6 months prior to study entry, these patients were allowed to enroll based upon their current disease status, which was assessed as stable by the local principal investigator.
Table 3 Baseline demographics and disease chronology, ITT. Parameter
Statistic
Q2 (N = 33)
Q4 (N = 62)
Gender Male Female
n (%) n (%)
20 (60.6%) 13 (39.4%)
28 (45.2%) 34 (54.8%)
Ethnicity Caucasian: Ashkenazi Jewish Caucasian: non-Ashkenazi Jewish Othera
n (%) n (%) n (%)
17 (51.5%) 14 (42.4%) 2 (6.1%)
29 (46.8%) 29 (46.8%) 4 (6.4%)
Median (range) Median (range) Median (range)
23.7 (2, 68) 21.8 (1,68) 35.9 (10,74)
29.7 (4, 73) 23.1 (1,71) 41.9 (11,75)
Median (range)
6.9 (2, 12)
5.2 (2, 13)
70.4 (24.9)
69.7 (21.3)
Gaucher disease chronology Age at diagnosis, years Age at symptom onset, years Age at first imiglucerase infusion, years Years on imiglucerase prior to enrollment
Pre-enrollment imiglucerase dose (U/kg/4weeks) Dose Mean (SD) a
Hispanic, 3.2%; Other 3.2%.
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170 Table 4 Baseline Gaucher disease status, ITT. Parameter
Statistic
Q2 (N = 33)
Q4 (N = 62)
Hematologic Hemoglobin (g/dL) Platelets (103/mm3)
Median (range) Median (range)
14.1 (12–17) 182.0 (108–411)
14.1 (12–17) 187.5 (101–428)
Visceral Liver volume (MN) Liver volume (MN)a Spleen volume (MN) Spleen volume (MN)a
Median Median Median Median
1.0 1.0 4.6 4.2
1.0 (1–2)
Spleen status Intact spleen Partial splenectomy Total splenectomy
n (%) n (%) n (%)
23 (69.7%) 0 10 (30.3%)
46 (74.2%) 1 (1.6%) 15 (24.2%)
n n (%)
0 15 (45.5%)
0 23 (37.1%)
Skeletal Bone crisisb Medullary infarction or AVNc
(range) (range) (range) (range)
(0–15) (0–1.5) (2–61) (2–9)
3.2 (1–9)
MN, Multiples of normal; AVN, Avascular Necrosis. a Organ volumes recalculated with patient 260101 (liver = 14.6 MN; spleen = 60.9 MN) excluded. b Bone crisis within 12 months prior to study participation. c Determined by MRI at Baseline. Patients were to be excluded if there was evidence that medullary infarction and/or AVN occurred within 6 months prior to study start. However, the majority of patients had no available MRI prior to baseline to provide any historical comparison. Absent any historical or clinical evidence that these events occurred within 6 months prior to study entry, these patients were allowed to enroll based upon their current disease status, as assessed by the local principal investigator.
167
Logistic regression analyses were performed to identify potential factors (including genotype, disease duration, splenectomy status, years of treatment, current dose and baseline liver volume) associated with success as measured by the primary efficacy endpoint. No statistically significant effects were identified. Therapeutic goals-based endpoint At study entry, the majority of patients in both groups met the therapeutic goals-based endpoint (Q2 n = 27 [81.8%], Q4 n = 51 [82.2%]). A total of 46 of 52 (89%) Q4-treated patients met the criteria for the therapeutic goals-based endpoint at Month 24 compared with 24 of 24 (100%) of Q2-treated patients (Fig. 1). The 11% difference seen between the 2 treatment groups was not statistically significant (95% confidence interval for the difference: 0.231, 0.060). Using the therapeutic goals-based endpoint criteria, all evaluable Q2 and Q4 splenectomized patients met the hemoglobin, platelet, and liver volume parameters. Biomarkers by composite endpoint There was little change from baseline in median levels of TRAP, ACE, or chitotriosidase at Month 24 for either treatment group overall or according to the outcome of the composite endpoint. Chitotriosidase increased in Q4 patients but the extent of increase was not different between Q4 patients who met or did not meet the composite endpoint. Table 5 provides median biomarker values according to clinical outcome.
Primary composite endpoint SF-36 A total of 83 ITT patients (87%; Q2 n = 26 [78.8%], Q4 n = 57 [91.9%]) had the required assessments to determine the primary endpoint. Of these, a total of 36 (63.2%) Q4-treated patients met the criteria of the primary composite endpoint compared with 21 (80.8%) Q2-treated patients (Fig. 1). The difference is not statistically significant based on the 95% CI (Q4 minus Q2: 0.176 [ 0.357, 0.058]). Fig. 2A illustrates the proportion of ITT patients who met the success criteria for individual components at Month 24. The highest proportion of patients met the bone health criteria in both Q2 (100%) and Q4-treated patients (98.4%). Success rates for the individual endpoint components for patients with prior total or partial splenectomy were very similar and are shown in Fig. 2B.
100%
100%
Exposure The mean dose of imiglucerase was comparable between Q4 (69 U/kg/4 weeks) and Q2-treated patients (67 U/kg/4 weeks). The duration of infusion calculated in 4-week intervals was slightly longer for Q2-treated patients (median of approximately 2.9 h/4 weeks) compared to Q4-treated patients (median of approximately 2 h/4 weeks) indicating a slightly faster rate of infusion for Q4treated patients overall.
89%
-17.6% [-35.7, 5.8]
81%
-11.5% [-23.1, 6.0]
There were no significant differences in mean composite scores for the SF-36 health survey at baseline or at Month 24 (Table 6). The mean subscale and summary scores for both treatment groups approximated those of the general population at baseline and at Month 24.
Adverse events and other safety measures
80% 63% 60% Q2 40%
Q4
20% N=26
N=57
N=24
N=52
0% Composite Endpoint
Therapeutic Goals-Based Endpoint
Fig. 1. Patient outcomes: percentage of patients meeting the composite endpoint and the therapeutic-goals based endpoint at Month 24 or last visit. Ns indicate the number of evaluable patients at Month 24 or last visit. Differences are given with 95% confidence intervals.
A total of 83.9% of Q4 and 63.6% of Q2-treated patients experienced adverse events (AEs) during the study. The majority of AEs (Q2, 89.9%; Q4, 95.2%) were unrelated to imiglucerase therapy. Seven Q2 (21.2%) and 4 (6.5%) Q4-treated patients experienced serious adverse events (SAEs) during the study; none was related to study medication or resulted in discontinuation from the study. Infusion associated reactions (IARs) occurred at low frequency in both the Q2 and Q4 treatment groups (Table 7). None of the IARs was classified as serious or resulted in discontinuation from the study. Among the AEs occurring at different frequencies between groups were those associated with progression of Gaucher disease: fatigue (Q2, 0%; Q4, 9.7%), hemoglobin decreased or anemia (Q2, 0%; Q4, 9.7%), splenomegaly (Q2, 0%; Q4, 3.2%), and platelet count decreased or thrombocytopenia (Q2, 0%; Q4, 11.3%).
168
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
A
Q4 *N=29
Hemoglobin
Q2 94
87
N=53 N=31
Platelet Count
100
N=55
90
N=27
93
Liver Volume 90
N=54 N=26
Spleen Volume
90 85
N=50 N=26 Bone Disease Progression
100
N=51
98
N=32 Bone Crisis
100
N=60 0
10
98
20
30
40 50 60 Percentage of Patients
70
80
90
100
*Ns indicate the number of evaluable patients at Month 24 or last visit.
Q4
B
100
*N=8
Hemoglobin
88
N=14
Platelet Count
Liver Volume
N=8
100
N=16
100
N=7
88
N=14
88
N=8 Bone Disease Progression
100
N=15
94
N=8 Bone Crisis
100
N=15 0
Q2
10
94
20
30
40 50 60 Percentage of Patients
70
80
90
100
*Ns indicate the number of evaluable patients at Month 24 or last visit. Fig. 2. (A) Percentage of patients meeting individual primary composite endpoint components. (B) Percentage of patients meeting individual primary composite endpoint components: patients with prior total or partial splenectomy.
Discussion The study was conducted to determine whether for maintenance purposes, an infusion of imiglucerase given once every 4 weeks using the same total monthly dose was comparable in efficacy and safety to the most commonly prescribed regimen of once every 2 weeks in previously treated adult patients with stable GD1. The study found no clinically meaningful signs of Gaucher disease progression on the Q4 regimen in the majority of study patients followed up to 2 years. This finding broadens the treatment options for physicians treating adult patients with GD1 and may be welcomed by a sub-set of patients who find biweekly infusions to be burdensome. It also supports the clinical practice of temporarily adopting a Q4 regimen at a doubled dose to accommodate changes in patients’ schedules.
The study employed a highly conservative composite primary efficacy endpoint because of initial concerns about the safety and efficacy of a less frequent infusion schedule. Several of the endpoint parameters, such as hemoglobin and organ volumes, were modified during the study to account for physiological variations in hematological and visceral measures commonly observed over time in patients with GD1 and for the extent of test-to-test variability in imaging measurements for determining organ volume. This modification would not be expected to affect the success rates for the two treatment groups unequally, as the groups were randomized and were similar at baseline; additionally, because the original criteria were within the normal range of variability for the measurements, they would not have effectively measured the impact of treatment regimen on clinical status.
169
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170 Table 5 Median levels of Gaucher disease markers at baseline and Month 24 by outcome (Met/Not Met Criteria) for composite endpoint, ITT. Biomarker
Q2 (N = 33)
Q4 (N = 62)
Met criteria
Did not meet criteria
Met criteria
Did not meet criteria
ACE (IU/L), median N Baseline Month 24/discontinuation
16 101.0 88.5
3 70.0 62.0
33 79.0 67.0
14 80.0 88.5
TRAP (IU/L), median N Baseline Month 24/discontinuation
16 8.6 8.8
3 10.0 12.0
33 10.0 9.2
3 598.0 568.0
32 174.5 244.0
Chitotriosidase (nmol/mL/hr), median N 15 Baseline 396.0 Month 24/discontinuation 350.0
Table 7 Overview of adverse events. Event type
Number (%) of patients Q2 (N = 33)
Q4 (N = 62)
Adverse events (AEs)a Number (%) patients with AEs Number of events
21 (63.6) 127
52 (83.9) 252
Treatment-related AEsb Number (%) of patients Number of events
2 (6.1) 13
7 (11.3) 13
14 9.5 8.2
Infusion-associated (related) AEsc Number (%) of patients Number of events
1 (3.0) 11
3 (4.8) 5
13 126.0 241.0
Serious AEsd Number (%) of patients Number of events Related events
7 (21.2) 13 0
4 (6.5) 5 0
Reference ranges: ACE = 43–109 IU/L; TRAP = 4.3–9.1 IU/L; Chitotriosidase = 5– 95 nmol/mL/h.
The final results were also analyzed using an endpoint based on consensus therapeutic goals for GD1 that were published after ini-
Table 6 Mean change from baseline in SF-36 composite scores at Month 24/discontinuation, ITT. Composite scores
Statistic
Q2
Q4
Physical health (53.16) Baseline score Change from baseline
N Mean (SD) Mean (range)
33 49.8 (8.4) 1.5 ( 18, 11)
62 46.9 (10.3) 0.5 ( 27, 17)
Mental health (52.82) Baseline Change from baseline
N Mean (SD) Mean (range)
25 52.9 (7.6) 0.6 ( 13, 14)
53 53.0 (8.5) 0.5 ( 17, 25)
Values in parentheses indicate mean scores for the general population. Patients with both baseline and change from baseline scores are shown.
tiation of the study. This analysis showed that many patients who did not meet the primary composite endpoint actually had hematological and visceral measurements that remained within the range accepted as treatment goals for patients with GD1. Because inclusion in this study required that patients had been clinically stable on Q2 dosing for at least 2 years prior to study start, and because dose and schedule for the patients in the Q2 treatment group was essentially not changed for this study, we would not expect a significant number of Q2 patients to fail the primary composite endpoint. As 19% of Q2 patients failed the primary composite endpoint but no Q2 patients failed the therapeutic-goals based endpoint, the clinical criteria that comprise the therapeutic goals endpoint may be a better summary measure of clinical stability in GD1 patients than the primary composite endpoint that was prospectively defined for this study. The analysis demonstrated that the majority of patients in both the Q2 and Q4 treatment groups maintained clinical stability, as defined by the principal efficacy endpoints. There were no statistically significant differences in success rates between the two groups for either the primary composite endpoint or the therapeutic goals-based endpoint. Each individual endpoint was maintained by the large majority of patients in both the Q4 group (P85%) and Q2 group (P90%). There were several patients in both groups (7 Q2 and 5 Q4, or 12.6% of patients overall) who were unevaluable due to missing one or more individual efficacy assessments. However, many of these unevaluable patients may have actually been successes with respect to the composite efficacy endpoint if results
a Adverse events (AEs) are any unfavorable and unintended sign, symptom or disease temporally associated with the use of a medical treatment or procedure regardless of whether it is considered related to the medical treatment or procedure. b An AE is considered to be related if, in the opinion of the principal investigator, there is a reasonable likelihood that it was caused by the medical treatment or procedure. c An infusion-associated adverse event is an event related to and occurring on the same day as a study infusion. d Serious AEs are defined as an undesirable sign, symptom, or medical condition which is fatal, is life-threatening, requires or prolongs inpatient hospitalization, results in persistent or significant disability/incapacity, constitutes a congenital anomaly or birth defect, is medically significant and which the investigator regards as serious based on appropriate medical judgment.
from all required assessments were available. In fact, the conclusions of the study would not be substantially altered even if all unevaluable patients were imputed to be endpoint failures. The study design allowed physicians to withdraw their patients from the study if the patient’s condition deteriorated such that one or more of the primary composite endpoint parameters were not met; 15 patients were withdrawn for this reason, all but 2 of whom were in the Q4 dosing group. As the majority of the primary composite endpoint parameters could change over time (hematologic values and visceral volumes), there is some chance that patients who were withdrawn from the study might have re-met these criteria had they stayed in the study for the entire 24 months. Thus the number of patients failing may be overestimated, particularly in the Q4 treatment group; as a consequence the difference in the success rates between the two treatment groups may be overstated. This potential bias should not impact the main finding of this study that there is no significant difference in outcomes between the dosing groups. Safety analysis revealed that imiglucerase infusions are welltolerated when the same monthly dose is administered either every 2 or 4 weeks. Adverse events were generally non-serious, and few were considered related to therapy. Certain adverse events that relate to disease progression were only reported in the Q4 group; this finding could suggest more disease progression among Q4 patients but as this study was not blinded to treatment group, it is also possible that reporting biases affected adverse event reporting. The safety and efficacy results found in this study do not support a general recommendation applicable to all patients with GD1 for a less frequent infusion schedule as a maintenance regimen. The findings apply only to adults and do not apply to children. The patients selected for this study were required to be clinically stable with respect to manifestations of GD1 and on a stable dose of imiglucerase between 20 and 60 U/kg/2 wks for at least 6 months. Patients conforming to this clinical profile, i.e. those
170
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
with stable minimal residual disease, would seem to be the best candidates for a trial of the less frequent infusion regimen. This study also did not evaluate the effectiveness of maintaining the same infusion dose for a patient on Q2 dosing but moving to a Q4 schedule, reducing the total 4-week dose by half. Thus these study results do not support the simple elimination of every other dose for stable adult GD1 patients; such a dosing schedule has not been evaluated in a controlled clinical study. While no statistically significant difference between the two dosing groups was found, proportionally fewer patients in the Q4 group than in the Q2 group met the composite endpoints and each individual efficacy parameter. This highlights that treatment decisions, including dose and frequency, must be made on an individual basis. The clinical status of each individual patient with GD1 should be carefully assessed prior to any adjustments in infusion regimen. Although in our study, the outcomes in post-splenectomy patients on the Q4 regimen did not differ from those in patients with intact spleens, we suggest extra caution in applying a Q4 regimen to splenectomized patients in whom an enhanced risk of post-treatment osteonecrosis has been reported [17]. In addition, the level of clinical success observed with the Q4 infusion regimen cannot be extrapolated beyond a 2-year treatment duration based on the results of this study. The long-term effectiveness of the Q4 regimen will require years of additional observation. In conclusion, a Q4 infusion regimen at the same total monthly dose may be considered for adult patients who have achieved therapeutic goals and are clinically stable and who have been on imiglucerase therapy for P2 years at 20 to 60 U/kg every 2 weeks. Both dose and infusion frequency should be individualized based upon physician judgment and patient needs as determined by regular and comprehensive monitoring. Longer-term follow up of patients on a Q4 dosing regimen is necessary to determine other clinical outcomes, such as bone mineralization and marrow infiltration. Acknowledgments This study was sponsored by Genzyme Corporation, Cambridge, Massachusetts. The authors would like to thank Lyn Radke-Mitchell (LRM Medical & Scientific Writing, LLC) for assistance with drafting and revising the manuscript, Carolyn Sawyer for assistance with references, and Dr. Michael Yeh and Dr. Edward Kaye for critical review of the manuscript. The authors also gratefully acknowledge the contributions of the late Dr. Ainu Prakash-Cheng and all the efforts of the physicians participating in the study: Dominick Amato, MD; Generosa Andria, MD; John Barranger, MD; Bruno Bembi, MD; Zale Bernstein, MD; Manuel Callis, MD, PhD; Lopes de Castro, MD; Joel Charrow, MD; Debra-Lynn Day-Salvatore, MD, PhD; Harry Dunn, MD; Paul Fern-
hoff, MD; David Finegold, MD; Edwin Kolodny, MD; David Kuter, MD; Nancy Leslie, MD; Clarisse Lobo, MD; Nicola Longo, MD; K. Joseph Philip, MD; Beata Pieczara, MD; Robert Steiner, MD; Cynthia Tifft, MD; Anna Tylki-Szymanska, MD; and Melissa Wasserstein, MD. References [1] J. Charrow, H.C. Andersson, P. Kaplan, E.H. Kolodny, P. Mistry, G. Pastores, B.E. Rosenbloom, C.R. Scott, R.S. Wappner, N.J. Weinreb, A. Zimran, The Gaucher registry: demographics and disease characteristics of 1698 patients with GD1, Arch. Intern. Med. 9 (2000) 2835–2843. [2] P. Kaplan, H.C. Andersson, K.A. Kacena, J.D. Yee, The clinical and demographic characteristics of non-neuronopathic Gaucher disease in 887 children at diagnosis, Arch. Pediatr. Adolesc. Med. 6 (2006) 603–608. [3] H.C. Andersson, J. Charrow, P. Kaplan, P. Mistry, G. Pastores, A. Prakesh-Cheng, B.E. Rosenbloom, C.R. Scott, R.S. Wappner, N.J. Weinreb, Individualization of longterm enzyme replacement therapy for Gaucher disease, Genet. Med. 7 (2005) 105–110. [4] Data on file, Genzyme Corporation. [5] R.J. Wenstrup, K.A. Kacena, P. Kaplan, G. Pastores, A. Prakash-Cheng, A. Zimran, T. Hangartner, Effect of enzyme replacement therapy with imiglucerase on BMD in type Gaucher disease, J. Bone. Miner. Res. 22 (2007) 119–126. [6] J. Charrow, B. Dulisse, G.A. Grabowski, N.J. Weinreb, The effect of enzyme replacement therapy on bone crisis and bone pain in patients with type 1 Gaucher disease, Clin. Genet. 71 (2007) 205–211. [7] N. Weinreb, J. Barranger, S. Packman, A. Prakash-Cheng, B.E. Rosenbloom, K. Sims, J. Angell, A. Skrinar, G.M. Pastores, Imiglucerase (CerezymeÒ) improves quality of life in patients with skeletal manifestations of Gaucher disease, Clin. Genet. 71 (2007) 576–588. [8] A.M. Damiano, G.M. Pastores, J.E. Ware Jr., The health related quality of life of 20 adults with Gaucher disease receiving enzyme replacement therapy: results from a retrospective, Qual. Life Res. 7 (1998) 373–386. [9] B.J. Masek, K. Sims, C.M. Bove, M.S. Korson, P. Short, D.K. Norman, Quality of life assessments in adult patients with Gaucher disease, Qual. Life Res. 8 (1999) 263–268. [10] P. Giraldo, V. Solano, P. Juan-Ignacio, Quality of life related to adult patients with type 1 Gaucher disease: Spanish experience, Qual. Life Res. 14 (2005) 453–461. [11] G.M. Pastores, N.J. Weinreb, H. Aerts, G. Andria, T.M. Cox, M. Giralt, G. A Grabowski, P.K. Mistry, A. Tylki-Szymanska, Therapeutic goals in the treatment of Gaucher disease, Elsevier: Semin. Hematol. 41 (suppl 5) (2004) 4–14. [12] M. de Fost, J. Aerts, J. Groener, M. Maas, E.M. Akkerman, M.G. Wiersma, C.E. Hollak, Low frequency maintenance therapy with imiglucerase in adult type I Gaucher disease: a prospective randomized controlled trial, Haematologica 92 (2007) 215–221. [13] J.E. Ware Jr., SF-36 Health survey update, Spine 25 (2000) 3130–3139. [14] C. Blyth, H. Still, Binomial confidence intervals, J. Am. Stat. Assoc. 78 (1983) 108–116. [15] G. Casella, Refining binomial confidence intervals, Can. J. Stat. 14 (1986) 113– 129. [16] A. Agresti, Y. Min, On small-sample confidence intervals for parameters in discrete distributions, Biometrics 57 (2001) 963–971. [17] T.M. Cox, J.M. Aerts, N. Belmatoug, M.D. Cappellini, S. vom Dahl, J. Goldblatt, G.A. Grabowski, C.E. Hollak, P. Hwu, M. Maas, A.M. Martins, P.K. Mistry, G.M. Pastores, A. Tylki-Szymanska, J. Yee, N. Weinreb, Management of nonneuronopathic Gaucher disease with special reference to pregnancy, splenectomy, bisphosphate therapy, use of biomarkers and bone disease monitoring, J. Inherit. Metab. Dis. 31 (2008) 319–336.
Contents lists available at ScienceDirect
Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme
A randomized trial comparing the efficacy and safety of imiglucerase (Cerezyme) infusions every 4 weeks versus every 2 weeks in the maintenance therapy of adult patients with Gaucher disease type 1 P.S. Kishnani a,*, M. DiRocco b, P. Kaplan c, A. Mehta d, G.M. Pastores e, S.E. Smith f, A.C. Puga f, R.M. Lemay f, N.J. Weinreb g a
Division of Medical Genetics, Department of Pediatrics, Duke University Medical Center, Box 103856 DUMC, 595 LaSalle St., GSRB1, 4th Floor, Durham, NC 27710, USA Unit of Rare Diseases, Department of Pediatrics, Gaslini Institute, Genoa, Italy Section of Metabolic Diseases, Department of Pediatrics, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA d Department of Haematology, Royal Free Hampstead NHS Trust, London, UK e Departments of Neurology and Pediatrics, New York University School of Medicine, New York, NY, USA f Genzyme Corporation, Cambridge, MA, USA g Gaucher Disease Treatment Center, University Research Foundation for Lysosomal Storage Disease, Inc., Coral Springs, FL, USA b c
a r t i c l e
i n f o
Article history: Received 13 October 2008 Received in revised form 22 December 2008 Accepted 23 December 2008 Available online 4 February 2009 Keywords: Enzyme replacement therapy Gaucher disease Glucocerebrosidase Glucosylceramide b-glucosidase Imiglucerase Infusion frequency Type 1
a b s t r a c t Imiglucerase (CerezymeÒ) has been the standard of care for treatment of Gaucher disease, a lysosomal storage disorder resulting from deficiency of glucocerebrosidase, since its approval in 1994. Infusions are typically given once every 2 weeks. However, many patients have expressed a desire for less frequent infusions as a matter of convenience. This clinical study assessed the safety and efficacy of intravenous imiglucerase infused once every 4 weeks (Q4) compared to once every 2 weeks (Q2) at the same total monthly dose in adult patients with clinically stable Gaucher disease type 1 (GD1). This was a 24-month, open-label, randomized, Phase 4, dose-frequency study conducted in 25 centers worldwide. Patients receiving imiglucerase were randomized to receive their monthly dose biweekly (n = 33) or every 4 weeks (n = 62). Changes from baseline in hemoglobin, platelets, liver and spleen volumes, bone crisis, and bone disease comprised a predefined composite endpoint; achievement or maintenance of established Gaucher disease therapeutic goals comprised a secondary endpoint. Sixty-three percent of Q4- and 81% of Q2-treated patients met the composite endpoint at Month 24; 89% of Q4- and 100% of Q2-treated patients met the therapeutic goals-based endpoint. The frequency of related adverse events was comparable between treatment groups. This study suggests that with comprehensive monitoring, a Q4 imiglucerase infusion regimen may be a safe and effective treatment option for the majority of clinically stable adult patients with GD1 but may not be appropriate for all GD1 patients. Continued monitoring in patients treated with Q4 dosing is required to assess long-term effectiveness. Ó 2008 Elsevier Inc. All rights reserved.
Gaucher disease is an inherited disorder characterized by insufficient activity of the enzyme acid b-glucosidase (glucocerebrosidase), resulting in the accumulation of glucosylceramide in the bone marrow compartment, liver, and spleen, and less often in the lungs [1]. The most common clinical manifestations of the disease include anemia, thrombocytopenia, hepatosplenomegaly and skeletal complications such as bone pain, bone crisis, bone lesions, cortico-medullary infarctions, medullary expansion, osteopenia, osteonecrosis, and pathological fractures [1–3]. The most common clinical variant is Gaucher disease type 1 (GD1) [1]. Treatment with enzyme replacement with imiglucerase (CerezymeÒ, Genzyme Corporation), a modified form of recombinant
* Corresponding author. Fax: +1 919 684 8944. E-mail address: [email protected] (P.S. Kishnani). 1096-7192/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2008.12.015
human b-glucocerebrosidase specifically targeted for uptake by macrophages, is currently the standard of care for symptomatic patients with GD1. Imiglucerase has been available since 1994, with over 5000 patients currently being treated [4]. Regular infusions with imiglucerase result in regression of the clinical manifestations of GD1 as evidenced by correction or amelioration of anemia and thrombocytopenia, reduction of hepatosplenomegaly, improvement in bone mineral density [5], diminished chronic bone pain, reduced frequency of bone crises [6], and improved responses on standardized patient health surveys [7–10]. Because of the highly variable manifestations and progression of Gaucher disease, disease treatment is individualized so that the imiglucerase dose and infusion regimen result in an optimal outcome for the patient [3]. The choice of starting imiglucerase dose and subsequent dosage adjustments are often influenced by an assessment
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
of age at initial presentation (generally, the earlier the symptoms, the more severe the disease), overall disease severity and progression, and the degree to which individualized therapeutic goals are achieved [3,11]. Regardless of dose, the infusion schedule is usually maintained at an interval of once every 2 weeks with an average infusion time of 1–2 h. Many patients, especially those who have milder or stable symptoms, have expressed a desire for less frequent infusions as a matter of convenience. Except for a single-institution, 11-patient study [12], reduced frequency infusions have never been prospectively evaluated as a maintenance regimen. We designed a randomized clinical trial to compare the safety and efficacy of the most commonly prescribed dosing regimen for imiglucerase (one infusion every 2 weeks) with a less frequent but equivalent overall dose infusion regimen (one infusion every 4 weeks). Materials and methods
165
at Month 24 (or discontinuation) based on a composite of hematologic, skeletal, and visceral measures (composite endpoint). To be classified as a success with respect to the composite endpoint, individual patients had to meet all five of the following criteria: Primary composite endpoint: Hemoglobin did not fall more than 1.25 g/dL for women or 1.5 g/ dL for men below the patient’s baseline value. Platelet count did not fall more than 25% below the patient’s baseline value, and did not fall below 80,000 mm3. Liver and spleen volumes were not greater than 20% above the patient’s baseline value. No new finding or progression of bone disease, including no new pathologic fractures, medullary infarction, lytic lesions or avascular necrosis. No bone crises during the study.
Study design This was a 24-month, open-label, multicenter, international, randomized, dose-frequency study conducted in adult patients (P18 years) with GD1. The study was initiated following the approval of the appropriate regulatory authorities and the approval of institutional ethics committees for each of the 25 trial sites worldwide. Eligible patients were randomized 2:1 to receive imiglucerase either once every 4 weeks (Q4) or once every 2 weeks (Q2) for 24 months. Patients in both treatment arms received approximately the same total monthly dose of imiglucerase as the average monthly dose the patient had received during the 6 months prior to study enrollment. After patients gave informed consent, medical history was obtained and a routine physical examination conducted, including clinical laboratory measures. Skeletal and visceral assessments and a general health survey (SF-36) were completed during the screening period. An independent data monitoring committee periodically reviewed source documents to advise the sponsor of any safety or efficacy concerns.
The original protocol was amended to account for fluctuations observed over time in hemoglobin levels, due in part to physiological variations or varying techniques at the site-based laboratories. The original criterion for hemoglobin was modified from an allowable decline of 1.0 g/dL at Month 24/discontinuation to a decline of not more than 1.25 g/dL for women and 1.5 g/dL for men. Intra-patient measurement variability was also observed in liver and spleen volume measurements early in the study. Therefore, the original criteria for liver and spleen volumes were modified from an allowable increase of no greater than 10% at Month 24 to an increase of no >20%. These modified criteria were used for the final analyses. Therapeutic goals endpoint In 2004, after the initiation of this study, therapeutic goals for the management of patients with GD1 were established [11]. The hematologic (hemoglobin and platelets) and visceral (liver and spleen volume) criteria of the primary composite endpoint were therefore re-defined in accordance with these particular therapeutic goals. To be classified a therapeutic success according to these goals, a patient had to meet all of the following criteria:
Eligibility criteria Therapeutic goals-based endpoint: Adult patients at least 18 years of age with a confirmed diagnosis of Gaucher disease (documented deficiency of acid b-glucosidase by enzyme assay) were eligible for this study if they met the following: at least 2 years of imiglucerase therapy, with a stable dose of 20–60 U/kg every 2 weeks for at least 6 months prior to enrollment, and evidence of stable GD1 at baseline as defined by: hemoglobin P11 g/dL (women) or P12 g/dL (men); platelet count >100,000/mm3; liver volume 61.8 multiples of normal (MN); spleen volume 610 MN (patients who had undergone partial or total splenectomy were also eligible); aspartate amino-transferase (AST) and alanine amino-transferase (ALT) < 2.0 the upper limit of normal (ULN); total bilirubin <2.0 ULN and serum creatinine <2.0 mg/dL; no evidence of new pathologic bone fractures, medullary infarctions, lytic lesions, or avascular necrosis secondary to Gaucher disease (confirmed by X-ray within 6 months of the study); and no evidence of bone crisis (defined as pain with acute onset requiring immobilization, narcotics, increased white blood cell (WBC) count, periosteal elevation, fever or debilitation >3 days) within 12 months of the study. Primary efficacy endpoint The primary efficacy endpoint was the proportion of patients who successfully maintained their baseline level of clinical response
Hemoglobin: >11 g/dL (women); >12 g/dL (men) Platelets: >100,000 mm3 Liver volume: <1.25 MN Spleen volume: <8 MN Skeletal assessments as described for the primary composite endpoint.
Additional assessments The key efficacy and safety assessments are provided in Table 1. Patient-reported health status was assessed using the SF-36 Health Survey (Standard Version) [13] at baseline and at 12 and 24 months. Laboratory parameters included biomarkers (ACE, TRAP and chitotriosidase) in addition to blood chemistry. Safety measures Safety was assessed by adverse event monitoring, clinical laboratory evaluations, and physical examinations. Patients could be withdrawn at the discretion of the investigator for non-compliance, intolerable adverse events, or failure on one or more of the composite endpoint criteria. Patients were free to withdraw consent and discontinue from the study at any time.
166
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
Table 1 Key study assessments.
Table 2 Patient disposition.
Assessment
Timepoint(s)
Laboratory and other assessments Hemoglobin and platelets Chemistry panela ACE, TRAP and chitotriosidaseb
Baseline; every 3 months Screening; month 24/discontinuation Baseline; every 3 months
Visceral assessments Liver and spleen volumec
Every 6 months
Skeletal assessments X-ray of femora/spined MRI of femorae Bone pain, bone crisis and mobilityf
Baseline; months 12, 24/discontinuation Baseline; months 12, 24/discontinuation Every 3 months
Patient questionnaire SF-36 general health survey
Baseline; every 6 months
Safety assessments Physical examination Adverse events Serum pregnancy testg
Baseline; every 6 months Every visit Baseline
Q2
Q4
Overall
Number of patients screened Number of patients randomized, N Number of patients in ITT Population, n Evaluable patients (ITT)a n (%)
37 33 26 (78.8)
65 62 57 (91.9)
130 102 95 83 (87.3)
Completion status (ITT) n (%) Completed study Discontinued study Adverse events Clinical failureb Otherc
26 (78.8) 7 (21.2) 1 (3.0) 2 (6.1) 4 (12.1)
40 (64.5) 22 (35.5) 5 (8.1) 13 (21.0) 4 (6.5)
66 (69.5) 29 (30.5) 6 (6.3) 15 (15.8) 8 (8.4)
a
a Included albumin, alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase, direct bilirubin, blood urea nitrogen, calcium, serum creatinine, lactate dehydrogenase, phosphorus, and total protein. b TRAP, tartrate resistant acid phosphatase; ACE, angiotensin converting enzyme; CHITO, chitotriosidase. c Volume per milliliter (mL) calculated from 0.2% of body weight assuming 1 g/ mL for volumes from computed tomography (CT) or volumetric magnetic resonance imaging (MRI) using a consistent modality. Multiples of normal (MN) calculated by dividing volume (mL) by [weight(kg) 25] for liver ([weight 2] for spleen). d Erlenmeyer flask deformity, lytic lesions, and fractures per anterior/posterior view of the entire femora and lateral view of the spine. e Marrow infiltration, medullary infarction and avascular necrosis assessed per coronal, T1 and T2-weighted MRI. f Bone pain, bone crisis and mobility per patient interview at study visits. g Serum b-human chorionic gonadotropin performed within two weeks of randomization for women of childbearing potential.
Statistical methods A sample size of 90 was chosen assuming a 95% success rate for the Q2 group and 80% for the Q4 group, using a 95% confidence interval (CI) for the difference between the percentage of success for the two treatment groups at Month 24 or last visit. All patients who enrolled in the study and received at least one infusion were included in the primary analysis population, the Intent-to-Treat (ITT). The safety population was equivalent to ITT. The primary efficacy endpoint was the proportion of ITT patients with a known outcome who were a success at Month 24 (or last visit if discontinued). A 95% two-sided CI was calculated for each treatment group using the exact method of Blyth-StillCastella [14,15]. A 95% two-sided CI was calculated for the difference between treatment groups using the exact method of Agresti and Min [16]. Patients who failed any of the criteria comprising the composite endpoint were classified as failures, even if they did not have complete testing data for all composite endpoint parameters. Patients who did not fail any of the composite endpoint criteria but who did not have complete data for all required tests were classified as unevaluable and were excluded from the primary analysis. The analysis based on the therapeutic goals-based endpoint was performed similarly. Formal statistical analysis was not performed for the SF-36 survey. Results Patient disposition, demographics, and disease history A total of 102 patients were randomized, and 95 patients received at least one on-study infusion and are included in the ITT population (Q2, n = 33; Q4, n = 62). Twenty-six of 33 (78.8%) Q2-
Based on the number of evaluable patients in the primary analysis. Withdrawn by investigator after failing on one or more criteria of the composite endpoint. c Other reasons for discontinuation: Q2: 1 pregnancy, 1 discontinuation at baseline, 1 noncompliant, 1 wishes to withdraw. Q4: 1 pregnancy, 1 return to Q2 regimen, 1 clinical baseline issue, 1 wishes to withdraw. b
treated and 40 of 62 (64.5%) Q4-treated patients completed the study to Month 24. Two Q2-treated patients and 13 Q4-treated patients were withdrawn by the treating physician after clinical deterioration led to failure to meet one or more of the criteria of the primary endpoint. A full summary of patient disposition is provided in Table 2. Baseline demographics of patients are detailed in Table 3. The mean age at randomization was 47 years (18–82 years). Fortyeight percent of patients were of Ashkenazi Jewish descent. Of note, patients were on imiglucerase for a median of 6.9 (Q2) and 5.2 (Q4) years at enrollment, at a mean imiglucerase dose of approximately 70 U/kg/4 wks in both groups. Baseline indicators of Gaucher disease were generally similar between treatment groups (Table 4). The majority (73%) of patients had an intact spleen and splenomegaly (overall median spleen volume 3.4 MN). One patient who did not meet entry criteria for organ volumes was inadvertently randomized and treated within the study, becoming part of the ITT population for the Q2 group. Baseline disease characteristics for the groups are presented in Table 4; organ volumes are summarized both with and without the inadvertently randomized patient. Overall, 40% of patients had evidence of medullary infarction or avascular necrosis (AVN) at baseline. Absent any historical, clinical, or radiologic evidence that these events occurred within 6 months prior to study entry, these patients were allowed to enroll based upon their current disease status, which was assessed as stable by the local principal investigator.
Table 3 Baseline demographics and disease chronology, ITT. Parameter
Statistic
Q2 (N = 33)
Q4 (N = 62)
Gender Male Female
n (%) n (%)
20 (60.6%) 13 (39.4%)
28 (45.2%) 34 (54.8%)
Ethnicity Caucasian: Ashkenazi Jewish Caucasian: non-Ashkenazi Jewish Othera
n (%) n (%) n (%)
17 (51.5%) 14 (42.4%) 2 (6.1%)
29 (46.8%) 29 (46.8%) 4 (6.4%)
Median (range) Median (range) Median (range)
23.7 (2, 68) 21.8 (1,68) 35.9 (10,74)
29.7 (4, 73) 23.1 (1,71) 41.9 (11,75)
Median (range)
6.9 (2, 12)
5.2 (2, 13)
70.4 (24.9)
69.7 (21.3)
Gaucher disease chronology Age at diagnosis, years Age at symptom onset, years Age at first imiglucerase infusion, years Years on imiglucerase prior to enrollment
Pre-enrollment imiglucerase dose (U/kg/4weeks) Dose Mean (SD) a
Hispanic, 3.2%; Other 3.2%.
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170 Table 4 Baseline Gaucher disease status, ITT. Parameter
Statistic
Q2 (N = 33)
Q4 (N = 62)
Hematologic Hemoglobin (g/dL) Platelets (103/mm3)
Median (range) Median (range)
14.1 (12–17) 182.0 (108–411)
14.1 (12–17) 187.5 (101–428)
Visceral Liver volume (MN) Liver volume (MN)a Spleen volume (MN) Spleen volume (MN)a
Median Median Median Median
1.0 1.0 4.6 4.2
1.0 (1–2)
Spleen status Intact spleen Partial splenectomy Total splenectomy
n (%) n (%) n (%)
23 (69.7%) 0 10 (30.3%)
46 (74.2%) 1 (1.6%) 15 (24.2%)
n n (%)
0 15 (45.5%)
0 23 (37.1%)
Skeletal Bone crisisb Medullary infarction or AVNc
(range) (range) (range) (range)
(0–15) (0–1.5) (2–61) (2–9)
3.2 (1–9)
MN, Multiples of normal; AVN, Avascular Necrosis. a Organ volumes recalculated with patient 260101 (liver = 14.6 MN; spleen = 60.9 MN) excluded. b Bone crisis within 12 months prior to study participation. c Determined by MRI at Baseline. Patients were to be excluded if there was evidence that medullary infarction and/or AVN occurred within 6 months prior to study start. However, the majority of patients had no available MRI prior to baseline to provide any historical comparison. Absent any historical or clinical evidence that these events occurred within 6 months prior to study entry, these patients were allowed to enroll based upon their current disease status, as assessed by the local principal investigator.
167
Logistic regression analyses were performed to identify potential factors (including genotype, disease duration, splenectomy status, years of treatment, current dose and baseline liver volume) associated with success as measured by the primary efficacy endpoint. No statistically significant effects were identified. Therapeutic goals-based endpoint At study entry, the majority of patients in both groups met the therapeutic goals-based endpoint (Q2 n = 27 [81.8%], Q4 n = 51 [82.2%]). A total of 46 of 52 (89%) Q4-treated patients met the criteria for the therapeutic goals-based endpoint at Month 24 compared with 24 of 24 (100%) of Q2-treated patients (Fig. 1). The 11% difference seen between the 2 treatment groups was not statistically significant (95% confidence interval for the difference: 0.231, 0.060). Using the therapeutic goals-based endpoint criteria, all evaluable Q2 and Q4 splenectomized patients met the hemoglobin, platelet, and liver volume parameters. Biomarkers by composite endpoint There was little change from baseline in median levels of TRAP, ACE, or chitotriosidase at Month 24 for either treatment group overall or according to the outcome of the composite endpoint. Chitotriosidase increased in Q4 patients but the extent of increase was not different between Q4 patients who met or did not meet the composite endpoint. Table 5 provides median biomarker values according to clinical outcome.
Primary composite endpoint SF-36 A total of 83 ITT patients (87%; Q2 n = 26 [78.8%], Q4 n = 57 [91.9%]) had the required assessments to determine the primary endpoint. Of these, a total of 36 (63.2%) Q4-treated patients met the criteria of the primary composite endpoint compared with 21 (80.8%) Q2-treated patients (Fig. 1). The difference is not statistically significant based on the 95% CI (Q4 minus Q2: 0.176 [ 0.357, 0.058]). Fig. 2A illustrates the proportion of ITT patients who met the success criteria for individual components at Month 24. The highest proportion of patients met the bone health criteria in both Q2 (100%) and Q4-treated patients (98.4%). Success rates for the individual endpoint components for patients with prior total or partial splenectomy were very similar and are shown in Fig. 2B.
100%
100%
Exposure The mean dose of imiglucerase was comparable between Q4 (69 U/kg/4 weeks) and Q2-treated patients (67 U/kg/4 weeks). The duration of infusion calculated in 4-week intervals was slightly longer for Q2-treated patients (median of approximately 2.9 h/4 weeks) compared to Q4-treated patients (median of approximately 2 h/4 weeks) indicating a slightly faster rate of infusion for Q4treated patients overall.
89%
-17.6% [-35.7, 5.8]
81%
-11.5% [-23.1, 6.0]
There were no significant differences in mean composite scores for the SF-36 health survey at baseline or at Month 24 (Table 6). The mean subscale and summary scores for both treatment groups approximated those of the general population at baseline and at Month 24.
Adverse events and other safety measures
80% 63% 60% Q2 40%
Q4
20% N=26
N=57
N=24
N=52
0% Composite Endpoint
Therapeutic Goals-Based Endpoint
Fig. 1. Patient outcomes: percentage of patients meeting the composite endpoint and the therapeutic-goals based endpoint at Month 24 or last visit. Ns indicate the number of evaluable patients at Month 24 or last visit. Differences are given with 95% confidence intervals.
A total of 83.9% of Q4 and 63.6% of Q2-treated patients experienced adverse events (AEs) during the study. The majority of AEs (Q2, 89.9%; Q4, 95.2%) were unrelated to imiglucerase therapy. Seven Q2 (21.2%) and 4 (6.5%) Q4-treated patients experienced serious adverse events (SAEs) during the study; none was related to study medication or resulted in discontinuation from the study. Infusion associated reactions (IARs) occurred at low frequency in both the Q2 and Q4 treatment groups (Table 7). None of the IARs was classified as serious or resulted in discontinuation from the study. Among the AEs occurring at different frequencies between groups were those associated with progression of Gaucher disease: fatigue (Q2, 0%; Q4, 9.7%), hemoglobin decreased or anemia (Q2, 0%; Q4, 9.7%), splenomegaly (Q2, 0%; Q4, 3.2%), and platelet count decreased or thrombocytopenia (Q2, 0%; Q4, 11.3%).
168
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
A
Q4 *N=29
Hemoglobin
Q2 94
87
N=53 N=31
Platelet Count
100
N=55
90
N=27
93
Liver Volume 90
N=54 N=26
Spleen Volume
90 85
N=50 N=26 Bone Disease Progression
100
N=51
98
N=32 Bone Crisis
100
N=60 0
10
98
20
30
40 50 60 Percentage of Patients
70
80
90
100
*Ns indicate the number of evaluable patients at Month 24 or last visit.
Q4
B
100
*N=8
Hemoglobin
88
N=14
Platelet Count
Liver Volume
N=8
100
N=16
100
N=7
88
N=14
88
N=8 Bone Disease Progression
100
N=15
94
N=8 Bone Crisis
100
N=15 0
Q2
10
94
20
30
40 50 60 Percentage of Patients
70
80
90
100
*Ns indicate the number of evaluable patients at Month 24 or last visit. Fig. 2. (A) Percentage of patients meeting individual primary composite endpoint components. (B) Percentage of patients meeting individual primary composite endpoint components: patients with prior total or partial splenectomy.
Discussion The study was conducted to determine whether for maintenance purposes, an infusion of imiglucerase given once every 4 weeks using the same total monthly dose was comparable in efficacy and safety to the most commonly prescribed regimen of once every 2 weeks in previously treated adult patients with stable GD1. The study found no clinically meaningful signs of Gaucher disease progression on the Q4 regimen in the majority of study patients followed up to 2 years. This finding broadens the treatment options for physicians treating adult patients with GD1 and may be welcomed by a sub-set of patients who find biweekly infusions to be burdensome. It also supports the clinical practice of temporarily adopting a Q4 regimen at a doubled dose to accommodate changes in patients’ schedules.
The study employed a highly conservative composite primary efficacy endpoint because of initial concerns about the safety and efficacy of a less frequent infusion schedule. Several of the endpoint parameters, such as hemoglobin and organ volumes, were modified during the study to account for physiological variations in hematological and visceral measures commonly observed over time in patients with GD1 and for the extent of test-to-test variability in imaging measurements for determining organ volume. This modification would not be expected to affect the success rates for the two treatment groups unequally, as the groups were randomized and were similar at baseline; additionally, because the original criteria were within the normal range of variability for the measurements, they would not have effectively measured the impact of treatment regimen on clinical status.
169
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170 Table 5 Median levels of Gaucher disease markers at baseline and Month 24 by outcome (Met/Not Met Criteria) for composite endpoint, ITT. Biomarker
Q2 (N = 33)
Q4 (N = 62)
Met criteria
Did not meet criteria
Met criteria
Did not meet criteria
ACE (IU/L), median N Baseline Month 24/discontinuation
16 101.0 88.5
3 70.0 62.0
33 79.0 67.0
14 80.0 88.5
TRAP (IU/L), median N Baseline Month 24/discontinuation
16 8.6 8.8
3 10.0 12.0
33 10.0 9.2
3 598.0 568.0
32 174.5 244.0
Chitotriosidase (nmol/mL/hr), median N 15 Baseline 396.0 Month 24/discontinuation 350.0
Table 7 Overview of adverse events. Event type
Number (%) of patients Q2 (N = 33)
Q4 (N = 62)
Adverse events (AEs)a Number (%) patients with AEs Number of events
21 (63.6) 127
52 (83.9) 252
Treatment-related AEsb Number (%) of patients Number of events
2 (6.1) 13
7 (11.3) 13
14 9.5 8.2
Infusion-associated (related) AEsc Number (%) of patients Number of events
1 (3.0) 11
3 (4.8) 5
13 126.0 241.0
Serious AEsd Number (%) of patients Number of events Related events
7 (21.2) 13 0
4 (6.5) 5 0
Reference ranges: ACE = 43–109 IU/L; TRAP = 4.3–9.1 IU/L; Chitotriosidase = 5– 95 nmol/mL/h.
The final results were also analyzed using an endpoint based on consensus therapeutic goals for GD1 that were published after ini-
Table 6 Mean change from baseline in SF-36 composite scores at Month 24/discontinuation, ITT. Composite scores
Statistic
Q2
Q4
Physical health (53.16) Baseline score Change from baseline
N Mean (SD) Mean (range)
33 49.8 (8.4) 1.5 ( 18, 11)
62 46.9 (10.3) 0.5 ( 27, 17)
Mental health (52.82) Baseline Change from baseline
N Mean (SD) Mean (range)
25 52.9 (7.6) 0.6 ( 13, 14)
53 53.0 (8.5) 0.5 ( 17, 25)
Values in parentheses indicate mean scores for the general population. Patients with both baseline and change from baseline scores are shown.
tiation of the study. This analysis showed that many patients who did not meet the primary composite endpoint actually had hematological and visceral measurements that remained within the range accepted as treatment goals for patients with GD1. Because inclusion in this study required that patients had been clinically stable on Q2 dosing for at least 2 years prior to study start, and because dose and schedule for the patients in the Q2 treatment group was essentially not changed for this study, we would not expect a significant number of Q2 patients to fail the primary composite endpoint. As 19% of Q2 patients failed the primary composite endpoint but no Q2 patients failed the therapeutic-goals based endpoint, the clinical criteria that comprise the therapeutic goals endpoint may be a better summary measure of clinical stability in GD1 patients than the primary composite endpoint that was prospectively defined for this study. The analysis demonstrated that the majority of patients in both the Q2 and Q4 treatment groups maintained clinical stability, as defined by the principal efficacy endpoints. There were no statistically significant differences in success rates between the two groups for either the primary composite endpoint or the therapeutic goals-based endpoint. Each individual endpoint was maintained by the large majority of patients in both the Q4 group (P85%) and Q2 group (P90%). There were several patients in both groups (7 Q2 and 5 Q4, or 12.6% of patients overall) who were unevaluable due to missing one or more individual efficacy assessments. However, many of these unevaluable patients may have actually been successes with respect to the composite efficacy endpoint if results
a Adverse events (AEs) are any unfavorable and unintended sign, symptom or disease temporally associated with the use of a medical treatment or procedure regardless of whether it is considered related to the medical treatment or procedure. b An AE is considered to be related if, in the opinion of the principal investigator, there is a reasonable likelihood that it was caused by the medical treatment or procedure. c An infusion-associated adverse event is an event related to and occurring on the same day as a study infusion. d Serious AEs are defined as an undesirable sign, symptom, or medical condition which is fatal, is life-threatening, requires or prolongs inpatient hospitalization, results in persistent or significant disability/incapacity, constitutes a congenital anomaly or birth defect, is medically significant and which the investigator regards as serious based on appropriate medical judgment.
from all required assessments were available. In fact, the conclusions of the study would not be substantially altered even if all unevaluable patients were imputed to be endpoint failures. The study design allowed physicians to withdraw their patients from the study if the patient’s condition deteriorated such that one or more of the primary composite endpoint parameters were not met; 15 patients were withdrawn for this reason, all but 2 of whom were in the Q4 dosing group. As the majority of the primary composite endpoint parameters could change over time (hematologic values and visceral volumes), there is some chance that patients who were withdrawn from the study might have re-met these criteria had they stayed in the study for the entire 24 months. Thus the number of patients failing may be overestimated, particularly in the Q4 treatment group; as a consequence the difference in the success rates between the two treatment groups may be overstated. This potential bias should not impact the main finding of this study that there is no significant difference in outcomes between the dosing groups. Safety analysis revealed that imiglucerase infusions are welltolerated when the same monthly dose is administered either every 2 or 4 weeks. Adverse events were generally non-serious, and few were considered related to therapy. Certain adverse events that relate to disease progression were only reported in the Q4 group; this finding could suggest more disease progression among Q4 patients but as this study was not blinded to treatment group, it is also possible that reporting biases affected adverse event reporting. The safety and efficacy results found in this study do not support a general recommendation applicable to all patients with GD1 for a less frequent infusion schedule as a maintenance regimen. The findings apply only to adults and do not apply to children. The patients selected for this study were required to be clinically stable with respect to manifestations of GD1 and on a stable dose of imiglucerase between 20 and 60 U/kg/2 wks for at least 6 months. Patients conforming to this clinical profile, i.e. those
170
P.S. Kishnani et al. / Molecular Genetics and Metabolism 96 (2009) 164–170
with stable minimal residual disease, would seem to be the best candidates for a trial of the less frequent infusion regimen. This study also did not evaluate the effectiveness of maintaining the same infusion dose for a patient on Q2 dosing but moving to a Q4 schedule, reducing the total 4-week dose by half. Thus these study results do not support the simple elimination of every other dose for stable adult GD1 patients; such a dosing schedule has not been evaluated in a controlled clinical study. While no statistically significant difference between the two dosing groups was found, proportionally fewer patients in the Q4 group than in the Q2 group met the composite endpoints and each individual efficacy parameter. This highlights that treatment decisions, including dose and frequency, must be made on an individual basis. The clinical status of each individual patient with GD1 should be carefully assessed prior to any adjustments in infusion regimen. Although in our study, the outcomes in post-splenectomy patients on the Q4 regimen did not differ from those in patients with intact spleens, we suggest extra caution in applying a Q4 regimen to splenectomized patients in whom an enhanced risk of post-treatment osteonecrosis has been reported [17]. In addition, the level of clinical success observed with the Q4 infusion regimen cannot be extrapolated beyond a 2-year treatment duration based on the results of this study. The long-term effectiveness of the Q4 regimen will require years of additional observation. In conclusion, a Q4 infusion regimen at the same total monthly dose may be considered for adult patients who have achieved therapeutic goals and are clinically stable and who have been on imiglucerase therapy for P2 years at 20 to 60 U/kg every 2 weeks. Both dose and infusion frequency should be individualized based upon physician judgment and patient needs as determined by regular and comprehensive monitoring. Longer-term follow up of patients on a Q4 dosing regimen is necessary to determine other clinical outcomes, such as bone mineralization and marrow infiltration. Acknowledgments This study was sponsored by Genzyme Corporation, Cambridge, Massachusetts. The authors would like to thank Lyn Radke-Mitchell (LRM Medical & Scientific Writing, LLC) for assistance with drafting and revising the manuscript, Carolyn Sawyer for assistance with references, and Dr. Michael Yeh and Dr. Edward Kaye for critical review of the manuscript. The authors also gratefully acknowledge the contributions of the late Dr. Ainu Prakash-Cheng and all the efforts of the physicians participating in the study: Dominick Amato, MD; Generosa Andria, MD; John Barranger, MD; Bruno Bembi, MD; Zale Bernstein, MD; Manuel Callis, MD, PhD; Lopes de Castro, MD; Joel Charrow, MD; Debra-Lynn Day-Salvatore, MD, PhD; Harry Dunn, MD; Paul Fern-
hoff, MD; David Finegold, MD; Edwin Kolodny, MD; David Kuter, MD; Nancy Leslie, MD; Clarisse Lobo, MD; Nicola Longo, MD; K. Joseph Philip, MD; Beata Pieczara, MD; Robert Steiner, MD; Cynthia Tifft, MD; Anna Tylki-Szymanska, MD; and Melissa Wasserstein, MD. References [1] J. Charrow, H.C. Andersson, P. Kaplan, E.H. Kolodny, P. Mistry, G. Pastores, B.E. Rosenbloom, C.R. Scott, R.S. Wappner, N.J. Weinreb, A. Zimran, The Gaucher registry: demographics and disease characteristics of 1698 patients with GD1, Arch. Intern. Med. 9 (2000) 2835–2843. [2] P. Kaplan, H.C. Andersson, K.A. Kacena, J.D. Yee, The clinical and demographic characteristics of non-neuronopathic Gaucher disease in 887 children at diagnosis, Arch. Pediatr. Adolesc. Med. 6 (2006) 603–608. [3] H.C. Andersson, J. Charrow, P. Kaplan, P. Mistry, G. Pastores, A. Prakesh-Cheng, B.E. Rosenbloom, C.R. Scott, R.S. Wappner, N.J. Weinreb, Individualization of longterm enzyme replacement therapy for Gaucher disease, Genet. Med. 7 (2005) 105–110. [4] Data on file, Genzyme Corporation. [5] R.J. Wenstrup, K.A. Kacena, P. Kaplan, G. Pastores, A. Prakash-Cheng, A. Zimran, T. Hangartner, Effect of enzyme replacement therapy with imiglucerase on BMD in type Gaucher disease, J. Bone. Miner. Res. 22 (2007) 119–126. [6] J. Charrow, B. Dulisse, G.A. Grabowski, N.J. Weinreb, The effect of enzyme replacement therapy on bone crisis and bone pain in patients with type 1 Gaucher disease, Clin. Genet. 71 (2007) 205–211. [7] N. Weinreb, J. Barranger, S. Packman, A. Prakash-Cheng, B.E. Rosenbloom, K. Sims, J. Angell, A. Skrinar, G.M. Pastores, Imiglucerase (CerezymeÒ) improves quality of life in patients with skeletal manifestations of Gaucher disease, Clin. Genet. 71 (2007) 576–588. [8] A.M. Damiano, G.M. Pastores, J.E. Ware Jr., The health related quality of life of 20 adults with Gaucher disease receiving enzyme replacement therapy: results from a retrospective, Qual. Life Res. 7 (1998) 373–386. [9] B.J. Masek, K. Sims, C.M. Bove, M.S. Korson, P. Short, D.K. Norman, Quality of life assessments in adult patients with Gaucher disease, Qual. Life Res. 8 (1999) 263–268. [10] P. Giraldo, V. Solano, P. Juan-Ignacio, Quality of life related to adult patients with type 1 Gaucher disease: Spanish experience, Qual. Life Res. 14 (2005) 453–461. [11] G.M. Pastores, N.J. Weinreb, H. Aerts, G. Andria, T.M. Cox, M. Giralt, G. A Grabowski, P.K. Mistry, A. Tylki-Szymanska, Therapeutic goals in the treatment of Gaucher disease, Elsevier: Semin. Hematol. 41 (suppl 5) (2004) 4–14. [12] M. de Fost, J. Aerts, J. Groener, M. Maas, E.M. Akkerman, M.G. Wiersma, C.E. Hollak, Low frequency maintenance therapy with imiglucerase in adult type I Gaucher disease: a prospective randomized controlled trial, Haematologica 92 (2007) 215–221. [13] J.E. Ware Jr., SF-36 Health survey update, Spine 25 (2000) 3130–3139. [14] C. Blyth, H. Still, Binomial confidence intervals, J. Am. Stat. Assoc. 78 (1983) 108–116. [15] G. Casella, Refining binomial confidence intervals, Can. J. Stat. 14 (1986) 113– 129. [16] A. Agresti, Y. Min, On small-sample confidence intervals for parameters in discrete distributions, Biometrics 57 (2001) 963–971. [17] T.M. Cox, J.M. Aerts, N. Belmatoug, M.D. Cappellini, S. vom Dahl, J. Goldblatt, G.A. Grabowski, C.E. Hollak, P. Hwu, M. Maas, A.M. Martins, P.K. Mistry, G.M. Pastores, A. Tylki-Szymanska, J. Yee, N. Weinreb, Management of nonneuronopathic Gaucher disease with special reference to pregnancy, splenectomy, bisphosphate therapy, use of biomarkers and bone disease monitoring, J. Inherit. Metab. Dis. 31 (2008) 319–336.