Protocol AGAL01601: A Multicenter, Phase 2, Open-Label Study of Fabrazyme® (Recombinant Human α-Galactosidase A) Replacement Therapy in Pediatric Patients with Fabry Disease

Fabrazyme^® (agalsidase beta)

Drug Name	Generic Name	Studied Indications or Disease	Approved U.S. Drug Label
Fabrazyme®	agalsidase beta	Fabry disease	Prescribing Info

Investigators and Study Center(s)

This was a multicenter study conducted at 10 sites in Europe and 1 site in the United States.

Publication (Reference)

Publications

Studied Period

30 October 2002 (first patient enrolled) to
09 May 2005 (last patient completed)

Phase of Development

Phase 2

Objectives

The primary objective was to evaluate safety, efficacy and pharmacokinetics of 1.0 mg/kg Fabrazyme biweekly in pediatric patients (age ≥ 7 and ≤ 15 years) with Fabry disease.

Methodology

This was a multicenter, multinational, Phase 2, open-label study in pediatric patients with Fabry disease who had no prior treatment with Fabrazyme.
Screening, Baseline Visit and Baseline Observation Period: Pediatric patients who met the selection criteria provided written informed consent (and patient’s legal guardians) and following baseline assessments were observed for a period of 12 weeks; followed by a treatment period of 48 weeks.
Treatment Period: During the treatment period, subjects received 1.0 mg/kg Fabrazyme every 2 weeks for up to 48 weeks. Following the first infusion, clinical and/or laboratory evaluations were performed every 4 to 12 weeks, with a final safety assessment 2 weeks after the last infusion.

Number of Patients (Planned and Analyzed)

A maximum of 20 patients were to be enrolled with a minimum of 10 male patients qualifying for the pain inclusion criterion. A total of 16 patients (14 males, 2 females) were treated with Fabrazyme. One patient discontinued participation in the study due to an SAE.

Diagnosis and Main Criteria for Inclusion

The study enrolled pediatric (age ≥ 7 years and ≤ 15 years, and Tanner stage ≤ III) hemizygous male and heterozygous, non-pregnant female patients with a clinical diagnosis of Fabry disease and confirmed diagnosis of Fabry disease (males with plasma α–galactosidase A [αGAL] activity of < 1.5 nmol/hr/mL or a documented leukocyte αGAL activity of < 4 nmol/hr/mg and females with potentially higher levels of αGAL activity who should have been diagnosed by genotyping), whose active Fabry disease was confirmed by either a history of pain consisting of Fabry pain crisis and/or chronic pain not controlled with pain medication OR at least 1 of the following other signs or symptoms: Body Mass Index (BMI) ≤ 20; Electrocardiogram (ECG) PR interval <130 msec provided it was not caused by a disease condition other than Fabry disease; Proteinuria (>30 mg/dL albumin as measured by dipstick test with at least 2+) OR estimated glomerular filtration rate (GFR) < 80 mL/min (calculated using Schwartz formula) provided that serum creatinine was < 2 mg/dL (180 µmol/L); History of gastrointestinal (GI) symptoms (post-prandial abdominal pain, nausea, vomiting); or history of autonomic neuropathy (history of hypohidrosis, impaired pupillary constriction with pilocarpine, reduced production of tears). The patient’s legal guardian(s) provided written informed consent before any study-related procedures were performed.

Test Product, Dose, and Mode of Administration

Test Product Fabrazyme^® (agalsidase beta)
Dose: approximately 1 mg/kg every 2 weeks
Mode of Administration: intravenous (IV) infusion
During the first 8 treatment visits, patients received IV infusions at a rate of no more than 0.25 mg/min (15 mg/hr) over approximately 4.5 hours. The rate to start each of these infusions was advised not to exceed 0.05 mg/kg/h, which could then progressively be increased (not exceeding the limit of 0.25 mg/min). After the first 8 infusions, the infusion rate of 0.25 mg/min (15 mg/hr) could be increased. The infusion rate increases could be up to 0.083 mg/min (5 mg/hr), reached incrementally, as tolerated. The cumulative infusion rate could be increased at each infusion; however, the total infusion period could not be less than 2 hours for patients weighing ≥ 30 kg. For patients weighing < 30 kg, the maximum rate had to be equal to 15 mg/hr. The infusion duration for these patients was less than 2 hours.

Duration of Treatment

Patients were treated for a total of up to 48 weeks

Reference Therapy, Dose and Mode of Administration

None.

Criteria for Evaluation
Efficacy

Globotriasylceramide (GL-3) clearance (in plasma, measured at Baseline and every 4 weeks thereafter; in skin, measured at Baseline Observation Period, Week 24 and Week 48); Pain, recorded daily using an electronic patient diary (pain visual analogue scale [VAS] score, concomitant pain medication, triggers for pain [fever, exercise, ambient maximum temperature], body outline for recording pain localization and intensity); Quality of life (general health VAS score, school attendance, and difficulties in performing physical activities, recorded daily using an electronic patient diary); Autonomic neuropathy (hypohidrosis [daily VAS score using an electronic patient diary]); GI disease (number of stools [recorded daily using an electronic patient diary]; nausea, vomiting, and post-prandial pain [recorded via investigator questionnaire at Baseline and every 4 weeks thereafter]); Renal disease (GFR measured by inulin clearance [GFR_Inulin] in plasma [measured at Week 0, 24 and 48]; estimated GFR using the Schwartz formula [eGFR_Schwartz; measured at Baseline and at Week 0, 12, 24, 36, and 48]; serum creatinine [measured at Baseline and at Weeks 0, 12, 24, 36, and 48]; urinary protein [measured at Week 0, 24, and 48]; urinary microalbumin [measured at Week 0, 24, and 48]); Cardiac disease (ECG intervals [measured at Week -12, Week 0, 24 and 48]; Growth, weight and height [measured at Screening, Baseline, Week -8, -4, 0 and every 2 weeks thereafter]; BMI [measured at Screening, Baseline, Week -8, -4, 0 and every 2 weeks thereafter].

Pharmacokinetics

Pharmacokinetic (PK) profiles for Fabrazyme were determined at the first infusion, and after 24 and 48 weeks of treatment.

Safety

Safety was measured in terms of adverse events (AEs), and serial physical exams, measurements of vital signs, routine laboratory parameters, electrocardiography, and echocardiography. Serum was collected throughout the study to determine the development of immunoglobulin G (IgG) antibodies against Fabrazyme. The formation of circulating immune complexes was determined in seroconverted patients.

Statistical Methods
Efficacy

The efficacy analysis was conducted on the Intent-to-Treat population (all enrolled patients meeting all inclusion and exclusion criteria, who signed informed consent and received at least 1 infusion.). Testing whether the change was significantly different from zero was evaluated using a Wilcoxon signed-rank test on the following continuous variables: Plasma GL-3 (percentage change from Baseline to Week 24 and Week 48); Skin Biopsy Results (change from Baseline to Week 24 and Week 48 in majority scores); Renal Function (change from Baseline to Week 24 and Week 48 in serum creatinine and eGFR_Schwartz; change from Week 0 to Week 24 and Week 48 in, GFR_Inulin, total urinary protein, urinary micro-albuminuria, and Cardiac ECG Central reads (change from screening to Week 24 and Week 48 in PR interval). The following discrete variables were evaluated by way of a Binomial Matched Pairs Procedure: Skin Biopsy majority score collapsed into zero/nonzero (Baseline and Week 24); and GI Questionnaire (nausea, vomiting, and post-prandial pain [yes/no], Baseline compared to Week 24 and Week 48). Daily diary data were evaluated by way of a Repeated Measures model to test whether change scores were significantly different from zero.

Pharmacokinetics

PK parameters were summarized by infusion (Week 0, Week 24, and Week 48) using descriptive statistics.

Safety

All AEs were summarized. All analyses on safety parameters were conducted on the Intent-to-Treat population.

Summary – Conclusions
Efficacy

The 16 patients who were treated had a mean age of 12.1 ± 2.52 years and consisted of 14 (87.5%) males and 2 (12.5%) females. The age of onset of Fabry symptoms ranged from 5 to 11 years and the age of diagnosis with Fabry disease ranged from 0 to 14 years. At Baseline, a majority of the patients presented with typical symptomatology of Fabry disease including early renal and cardiac involvement. All 15 patients who completed the study received at least 80% of the planned 25 infusions. One patient discontinued his participation in the study after 15 infusions due to a serious adverse event (SAE).

At Baseline, 14/16 patients (88%; all males) had abnormal plasma GL-3 levels (i.e., >7.03 µg/mL), while at Week 20 the plasma GL-3 of all patients had normalized and remained normal. The 2 female patients had normal plasma GL-3 levels from the start of the study. Before the start of treatment, a majority of the male patients (12/14; 86%) already had significant accumulation of GL-3 in all skin cell types (even the youngest male aged 8.5 years had moderate GL-3 accumulation). Upon Fabrazyme treatment, GL-3 was cleared from the capillary endothelial cells in all patients, and in most patients (88% at Week 24) from the deep vessel endothelial cells. The effect in the perineurium was more modest. GL-3 was normalized or cleared from the plasma and skin by Week 24 and 48 for all patients, thus demonstrating sustained efficacy of Fabrazyme treatment.

An electronic diary (eD) was used daily by the patients to record a number of parameters; this tool is not validated for use in children or in Fabry disease, and only some of the parameters, such as the VAS score for pain and quality of life and the number of stools per day, are classically used in other pathologies. The pain VAS score used a scale from 0 to 100 (0 = no pain and 100 = severe pain). A modest statistically significant decrease in pain VAS score was observed between pre-treatment and the Week >12 to 24 and Week >36 to 48 periods during treatment (p<0.001), but this improvement was mainly due to the dramatic improvement of 1 of the only 4 patients with significant pain at Baseline. Given the unexpectedly low pain symptomatology at Baseline, the absence of a placebo arm and the unrestricted use of pain medication, the effect of Fabrazyme treatment on pain in this pediatric population could not be studied optimally. Still, given the natural course of pain in this disorder, it is noteworthy that none of the 16 patients experienced an important pain increase during the 1-year treatment period. Fabry pain can be triggered by fever, which is also a symptom of Fabry disease. A small trend for improvement was observed as the percentage of days with fever was slightly decreased at Week 48.

Prior to treatment, 8/16 patients (50%) reported using antidepressants or anti-epileptics, while at Week >36 to 48, 5/11 patients (45%) reported the use of these drugs. Opiates were used by 3/16 patients (19%) at pre-treatment and by 1/11 patients (9%) at Week >36 to 48. The percentage of days with no pain medications and the percentage of days with no pain medication and no pain did not show statistically significant differences between the pre-treatment period and the Week >12 to 24 and the Week >36 to 48 period (the diary only captured the frequency of administration and not the dose).

Quality of life, as measured by general health, physical activities, and school attendance, is often impaired in children with Fabry disease. These parameters were part of the Fabry Quality of Life Daily Questionnaire monitored using the eD. General health (VAS score on a 0 to 100 scale, where 0 = poor health and 100 = excellent health), which was already quite good at Baseline, showed a statistically significant improvement between pre-treatment and the Week >12 to 24 period (75.99% to 80.34%, p<0.001) and the Week >36 to 48 period (76.97% to 82.34%, p<0.001). This improvement in general health was consistent with the improvement of pain. The mean percentage of days with “good” general health (VAS score ≥ 75 and ≤ 100) increased from 65.8% to 75.4% between pre-treatment and Week >12 to 24, which is statistically significant (p=0.009). In the 11 patients with data in the Week >36 to 48 treatment period, an improvement from 69.2% in the pre-treatment period to 76.8% in the Week >36 to 48 period was observed, but these differences were not statistically significant. School attendance improved, demonstrated by a statistically significant decrease in the mean percentage of days that school was not attended due to sickness from 12.4% at Baseline to 7.4% at Week >36 to 48 (p=0.040). The percentage of days with difficulty performing low energy activity improved during treatment from 11.9% of days at Baseline to 3.2% of days at Week >36 to 48 (p<0.001). A trend in improvement was observed for moderate energy activity, but no significant effect was observed for high energy activity. However, the children did not necessarily try to perform high energy activities every day, so the lack of improvement may also reflect the lack of documentation of the effect.

Most children reported hypohidrosis at Baseline (subjective evaluation through the diary, with mean VAS score of 27.24, on a 0 to 100 scale, where higher scores were better), and a slight improvement upon treatment (between pre-treatment and Week >12 to 24 (+2.74; p=0.009) and Week >36 to 48 (+3.57; p=0.005). This must be interpreted as a qualitative change (improvement), as the diary was not adequate to evaluate the change quantitatively.

The number of stools per day was normal at Baseline and remained normal throughout the study. Other GI parameters, explored at every visit through the GI questionnaire, showed improvement upon treatment, consistent across the different components of the questionnaire. The number of patients with nausea decreased from 7 (50%) at Baseline to 2 (14%) at Week 48 (p=0.063), the number of patients with post-prandial pain decreased from 8 (57%) at Baseline to 4 (29%) at Week 48, and the number of patients with vomiting decreased from 4 (29%) at Baseline to 0 (0%) at Week 48.

Renal function was monitored by measuring GFR, 24-hour urinary protein, microalbuminuria, and serum creatinine. No clinically significant change in mean GFR as measured through inulin clearance or the Schwartz formula was observed, but 4 patients had abnormal values: 3 high and 1 low. High GFR (suggestive of hyperfiltration) is often predictive of future poor outcome of renal function (Sunder-Plassmann, 2004, Am J Kidney Dis); interestingly, 2 patients with high GFR values normalized upon treatment. Proteinuria was already abnormal in this young population, since 8 patients (7 males, 1 female) had levels > 150 mg/24 hr. A statistically significant decline (p=0.002) from 151.0 mg/24 hr at Week 0 to 90.0 mg/24 hr at Week 24 was observed. A more modest decrease compared to Week 0 was also seen at Week 48 (from 151.0 mg/24 hr to 126.0 mg/24 hr). Consistently, microalbuminuria was present in 2 patients at Week 0 and diminished during treatment in both of them. The trend for improvement in proteinuria and microalbuminuria and the normalisation of some of the patients with high GFR values are encouraging. These suggest that Fabrazyme may have beneficial effect on this early glomerular involvement; long-term follow-up is necessary to confirm this hypothesis.

ECG central reads did not show significant changes in mean PR interval or in any of the other measurements during treatment.

At the start of treatment, the weight, height and BMI for all patients were within the 95th percentiles for their age. However, a majority of the children (10/14 males and 2/2 females) were below the 50th percentile, suggesting a moderate growth impairment or delay. Mean weight increased by 3.0 kg during the 48 weeks of treatment, while mean height increased by 7.1 cm, but mean BMI decreased 0.4 kg/m2. Four of the male patients showed a clinically significant increase in growth percentile during the study. The evolution of BMI was variable between patients, but altogether, the mean decrease observed reflected a faster height growth during the treatment period, not associated with a similar weight gain

Pharmacokinetic Results

The PK analysis showed that Fabrazyme was largely distributed to the extravascular water compartment after dosing. IgG seroconversion from negative to positive resulted in a 71% decrease in clearance of Fabrazyme from 129 mL/min to 37 mL/min and a 60% decrease in steady state volume of distribution from 21.4 L to 8.6 L. Fabrazyme was rapidly removed from plasma, having a least square mean half-life of 84 minutes after the 1st dose, but increasing to 156 minutes after IgG seroconversion. The net result of this alteration was an increase in exposure. However, this change in exposure had no observable impact on the efficacy of Fabrazyme on the clearance of skin GL3. Therefore, the immunologic responses in pediatric patients may differ from those in adults, as IgG seroconversion in pediatric patients was associated with prolonged half-life concentrations of Fabrazyme, a phenomenon rarely observed in adult patients.

Safety Results

Overall, the safety profile of Fabrazyme remains favorable. A total of 377 Fabrazyme infusions were given to the 16 patients participating in this study. Fifteen patients received at least 80% of their infusions. The mean infusion time decreased gradually over time from 271.9 min (range 240-320 min) for the first infusion at Week 0 to 172.1 min (range 105-248 min) at Week 48.

All patients enrolled in this study experienced at least 1 AE. Fifteen of the 16 patients who participated in the trial experienced a total of 350 treatment-emergent AEs. The most frequently involved Body Systems were the Gastro-Intestinal System, Body as a Whole-General, the Central and Peripheral Nervous System, and the Respiratory System. The most frequently occurring treatment-emergent AEs in ≥30% of patients without regard to causality included headache (56%), abdominal pain (56%), pharyngitis (56%), fever (50%), nausea (50%), vomiting (44%), pain (38%), rhinitis (38%), and diarrhoea (31%). The majority of AEs were mild or moderate in intensity and not related to Fabrazyme.

There were no patient deaths during the study. One patient experienced 1 SAE, which was also an IAR, during his 15^th infusion in the study. This SAE was a reported event of anaphylactic shock which was severe in intensity and was considered by the investigator to be definitely related to Fabrazyme treatment. Treatment was discontinued and the patient received medication after which he recovered without sequelae the same day. A careful review of all available information suggests that this event did not represent anaphylactic shock. The event was associated with bradycardia which was responsive to atropine. This episode, in part, was more suggestive of a vasovagal response, especially given the likelihood of pre-existing autonomic dysfunction due to Fabry disease. At the time of onset of the reaction, he was receiving the infusion at a relatively rapid rate of 0.5 mg/min. There was no evidence of vascular collapse, epinephrine was not required, substantial respiratory symptoms were absent and the patient never became tachycardic. Two serum IgE tests against Fabrazyme were negative and serum tryptase remained within normal range (no mast cell degranulation). This patient discontinued his participation in the study after 15 infusions.

Eleven (69%) patients, all male, seroconverted in a mean time of 55.6 days since first infusion. Seroconversion did not have any discernible impact on efficacy, based on GL-3 clearance.

Six of the 16 patients (38%) experienced infusion-associated reactions (IARs), defined as related AEs that occurred on the same day as the infusion, excluding laboratory, ECG, and ECHO abnormalities. The most commonly occurring IARs were rigors (chills) (19%), headache (19%), nausea (19%), fever (13%) and temperature changed sensation (feeling hot and/or feeling cold) (13%). The majority of IARs were mild or moderate in intensity and managed by infusion rate adjustments and/or medications (e.g., anti-pyretics, anti-histamines and/or steroids). All patients recovered from the IARs.

Mean values for all laboratory and vital signs parameters were within normal ranges and there were no clinically meaningful changes overall or trends in changes over time.

No new safety concerns were identified based upon evaluation of AEs, vital signs, physical examination findings, laboratory values, ECGs, ECHOs and concomitant medication use.

Conclusion

Overall, the safety and efficacy profile of Fabrazyme treatment administered at 1 mg/kg every 2 weeks in pediatric patients is consistent with that seen in adults.