Drugs

, Volume 69, Issue 16, pp 2179–2205

Enzyme Replacement Therapy for Fabry Disease

A Systematic Review of Available Evidence

Authors

    • Department of MedicineUniversity of Muenster
  • Anna Tylki-Szymańska
    • Department of Metabolic Diseases, Endocrinology, and DiabetologyThe Children’s Memorial Health Institute
  • Max J. Hilz
    • Department of NeurologyUniversity of Erlangen-Nuremberg
    • Departments of Neurology, Medicine and PsychiatryNew York University
Review Article

DOI: 10.2165/11318300-000000000-00000

Cite this article as:
Schaefer, R.M., Tylki-Szymańska, A. & Hilz, M.J. Drugs (2009) 69: 2179. doi:10.2165/11318300-000000000-00000
  • 402 Views

Abstract

Fabry disease is a progressive and life-threatening glycolipid storage disorder affecting both males and females. The primary driver of the disease is the accumulation of glycolipids (globotriaosylceramide [GL-3]) in a variety of cell types, including vascular endothelial cells, a range of renal cell types, cardiomyocytes and neurons, which is caused by deficient activity of the lysosomal enzyme, α-galactosidase. The disease typically presents during childhood or adolescence. First manifestations reflect involvement of small nerve fibres of the peripheral and autonomic nervous systems. With age, severe complications involving the kidneys, heart and brain cause considerable morbidity and premature death.

Outside the US, enzyme replacement therapy (ERT) with agalsidase alfa 0.2 mg/kg every other week (EOW) and agalsidase beta 1.0mg/kg EOW is available for the treatment of patients with Fabry disease, while agalsidase beta 1.0mg/kg EOW is the only approved drug in the US.

To analyse the evidence for ERT, a systematic review of the literature was performed to identify prospectively designed randomized, controlled trials (RCTs) and open-label studies on the efficacy of agalsidase alfa and agalsidase beta. MEDLINE and EMBASE databases were searched; inclusion criteria for the systematic review were prospectively designed clinical studies evaluating ERT with quantifiable endpoints: double-blind and open-label studies were eligible. Exclusion criteria were review articles, case reports, case studies, letters to the editor and articles based on registry data (Fabry Outcome Survey or Fabry Registry). In addition, any studies with a retrospective design or data based on post hoc analyses were excluded. The evidence was reviewed with respect to the clinical benefits of ERT at the level of the end organ.

A total of 9 RCTs and 23 open-label studies were identified for inclusion. The efficacy of ERT in Fabry disease has been measured against a variety of endpoints, the majority of which were subclinical parameters rather than clinical outcomes. Plasma levels of GL-3 together with accumulation in the kidney, heart and skin were the most commonly studied endpoints, followed by renal endpoints of proteinuria and glomerular filtration rate, whereas cardiac and neurological endpoints were not commonly studied. To date, only one RCT with ERT defined hard clinical outcomes in the form of cardiac, renal or cerebrovascular events, or death as its primary endpoint.

The currently available data from prospective RCTs and open-label studies in patients with Fabry disease are more robust for ERT at a dose of 1 mg/kg EOW than a dose of 0.2mg/kg EOW, although the beneficial effects of ERT with either dose or preparation are variable.

Fabry disease is a glycolipid storage disorder associated with an early onset of an insidious cascade of processes leading to progressive tissue and organ damage, organ failure and premature mortality.[1] Although the inheritance of Fabry disease is X-linked, significant numbers of females may experience a similar constellation of symptoms, but less severe disease progression than that reported in classically affected males, possibly as a result of skewed X-chromosome inactivation. However, onset in females is often later in life.[25]

Clinical manifestations usually become apparent around the age of 10 years in affected males and several years later in females.[6,7] First signs and symptoms reflect damage to small nerve fibres of the peripheral and autonomic nervous systems.[810] They include chronic and/or acute episodic neuropathic pain in the hands and feet, angiokeratomas, cornea verticillata, ocular vessel tortuosity, hypo- or anhidrosis, tinnitus aurium, hearing loss, sensory losses (particularly of small nerve-fibre-mediated pain and temperature perception), gastrointestinal disturbances, and intolerance to heat, cold and exercise.[7,11,12]

First clinical signs of Fabry nephropathy include microalbuminuria, proteinuria and — later or subsequently — progressive decline in glomerular filtration rate (GFR) occurring if a critical number of nephrons have been damaged as a result of interstitial scarring, glomerulosclerosis and tubular atrophy.[5,1315] Most classically affected males will progress to end-stage renal disease by the fifth decade of life.[16,17] A significant proportion of females will develop moderate to severe kidney involvement.[5] Potential cardiovascular manifestations develop more often and earlier in males than in females and may include sinus node dysfunction, conduction system abnormalities, arrhythmia, left ventricular hypertrophy, valvular dysfunction, angina pectoris, myocardial infarction and heart failure.[18,19] Cerebrovascular complications in adulthood may include early stroke, transient ischaemic attacks, cerebral vasculopathy, thrombosis, ischaemic stroke, but also cerebral haemorrhages.[20,21] Quality of life and physical, social and job performance progressively decline and patients are at risk of chronic depression.[22,23]

Although the pathophysiology of Fabry disease is still not completely understood, the storage of globotriaosylceramide (GL-3) represents the initial insult and may start before birth.[24] Consistent with the progressive and multisystemic nature of Fabry disease, cellular dysfunction is followed by a cascade of disease processes involving functional impairment and structural damage to tissues and organs that may develop over a period of years or decades (figure 1).[1] These cascades are likely to differ depending on the organ systems affected and varying stages of disease progression. Certain patterns can be recognized and provide the basis for classifying disease processes according to the following: (i) storage of GL-3 (primary disease), which may result in structural changes, damage and abnormal cellular function that in turn can act as a stimulus to the surrounding tissue; (ii) tissue-level responses to pathological stimuli (secondary disease) such as ischaemia, inflammation, hypertrophy, fibrosis or proteinuria; (iii) and organ failure (tertiary disease).[1] Although the storage of GL-3 has been classified as the primary disease, no direct correlation exists between lipid storage and the clinical manifestations. A recent study by Aerts et al.[25] reported that globotriaosylsphingosine, a GL-3 metabolite (lysoGb3), is dramatically increased in the plasma of male Fabry patients, and plasma and tissues of Fabry mice, and might be the missing link in the pathogenesis of Fabry disease.
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Fig1.jpg
Fig. 1

Fabry disease model. Cascades involving different levels of disease processes (reproduced from Wanner,[1] with permission). GL-3 = globotriaosylceramide.

With the availability of disease-specific enzyme replacement therapy (ERT), the need for early diagnosis and comprehensive treatment of Fabry disease is paramount, as progression of the disease can be rapid. ERT is intended to address lysosomal storage of GL-3, the primary pathological event and driver of the various organ-specific disease cascades.[1] Following the US FDA evaluation of two different α-galactosidase A enzyme preparations, i.e. agalsidase alfa (Replagal®; Shire Human Genetic Therapies, Boston, MA, USA) 0.2 mg/kg every other week (EOW) and agalsidase beta (Fabrazyme®; Genzyme Corp., Cambridge, MA, USA) 1.0 mg/kg EOW; only agalsidase beta was granted marketing approval in the US. Both agalsidase alfa and agalsidase beta are marketed in the EU following approval in 2001. In February 2008, EU approval of agalsidase beta under exceptional circumstances was lifted to normal approval. Agalsidase alfa is produced in a human fibroblast cell line and agalsidase beta in a Chinese hamster ovary host cell line. These two products are similar, with comparable specific activities per milligram and glycosylation patterns.[2628] To date, one study — which was not prospectively designed — has directly compared the two licensed preparations at the approved doses, i.e. agalsidase alfa 0.2 mg/kg EOW and agalsidase beta 1 mg/kg EOW.[29]

The efficacy of ERT in Fabry disease has been measured against a variety of endpoints, with GL-3 accumulation being the most commonly employed in clinical studies. Although it is of great value to measure the efficacy of therapy against a marker representing the driver of the pathophysiological changes, it is important to recognize that GL-3 is a subclinical marker and should not be considered in isolation. Rather, subclinical markers need to be considered in a hierarchy that gives primary emphasis to clinically relevant endpoints, such as premature mortality, major clinical events and clinical symptoms (figure 2). Furthermore, data on the efficacy of ERT need to be assessed in terms of hierarchy of evidence, with evidence from randomized, controlled trials (RCTs) taking priority over prospective open-label studies and other study designs.
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Fig2.jpg
Fig. 2

Hierarchy of endpoints in Fabry disease. GFR = glomerular filtration rate; GL-3 = globotriaosylceramide; LV = left ventricular.

The objective of the present systematic review is to summarize the evidence for ERT so that physicians and others involved in the care of patients with Fabry disease have a firm basis to support treatment decisions. In particular, the evidence is reviewed with respect to the clinical benefits of ERT at the level of the end organs. We evaluate and compare the two current ERT options: agalsidase alfa 0.2 mg/kg EOW and agalsidase beta 1.0 mg/kg EOW.

1. Methods of Literature Review

English language medical literature was reviewed to identify abstracts and articles relating to recombinant α-galactosidase A in Fabry disease. MEDLINE and EMBASE databases were searched on 18 May 2009 using the search terms ‘recombinant alpha galactosidase’, ‘Fabry disease’, ‘Fabrazyme’, ‘Replagal’, ‘agalsidase alfa’, ‘agalsidase beta’ and combinations thereof. Each database was searched from the beginning of the database to the date of the search. Inclusion criteria for the systematic review were prospectively designed clinical studies evaluating ERT with quantifiable endpoints; double-blind and open-label studies were eligible. Exclusion criteria were review articles, case reports, case studies, letters to the editor and articles based on registry data (Fabry Outcome Survey or Fabry Registry). Registry data were excluded to avoid duplicate data from patients reported on in a clinical trial as well as in a registry, and because of potential confounding variables inherent to the nature of registry data. In addition, any studies with a retrospective design or data based on post hoc analyses were excluded.

1.1 Included Studies

The literature search returned 584 publications, including duplicates, for further screening. After exclusions, detailed in the QUality Of Reporting Of Meta-analyses (QUOROM) flow chart (figure 3), the search returned a total of nine randomized, placebo-controlled studies, including four studies of agalsidase beta and five studies of agalsidase alfa (table I).[3038] Six to 82 patients were randomized per study, the majority of whom were male. A total of 23 open-label studies were identified for inclusion, with a portion of these being open-label extension studies of randomized, placebo-controlled trials. Of the 23 open-label studies, 18 studies focused on male patients with Fabry disease, including 12 studies of agalsidase beta,[30,3949] five of agalsidase alfa[36,5053] and one that evaluated both agalsidase beta and agalsidase alfa[54] (table II). In addition to these 18 studies, one open-label study of agalsidase alfa focused on female patients (table II)[55] and four were open-label studies of children (one agalsidase beta,[56] three agalsidase alfa[5759]) that were predominantly designed as safety studies (table II). A further study[60] was excluded because agalsidase alfa was administered only once to each of ten patients, which was deemed to be unrepresentative of the extended dosing required in the real world for effective treatment of Fabry disease.
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Fig3.jpg
Fig. 3

QUOROM (QUality Of Reporting Of Meta-analyses) flow chart. Included were nine randomized, controlled trials and 23 open-label studies. Two of the open-label studies were extension studies of randomized, controlled trials, reported within the same publication, yielding a total study number of 30. ERT = enzyme replacement therapy.

https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab1.jpg
Table I

Randomized, controlled studies of enzyme replacement therapy in patients with Fabry disease. Trials are structured according to endpoint

https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab2.jpg
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab2a.jpg
Table II

Open-label studies of enzyme replacement therapy (ERT) in patients with Fabry disease. Trials are structured according to endpoint

2. Results

2.1 Overview

Clearance of GL-3 accumulation was widely adopted as a measure of efficacy in studies of agalsidase beta and agalsidase alfa in Fabry disease. However, it needs to be noted that although the storage of GL-3 has been classified as the primary disease, no direct correlation exists between lipid storage and the clinical manifestations.[25] Plasma levels of GL-3 together with accumulation in the kidney, heart and skin were the most commonly studied endpoints, followed by renal endpoints of proteinuria and GFR, whereas cardiac and neurological endpoints were studied less commonly. The findings of the studies included in this systematic review are summarized in table format and structured according to endpoint. The values reported in individual publications are given, together with an indication of increase or decrease in the investigated endpoint.

2.2 GL-3 Accumulation

2.2.1 Plasma GL-3

In all included RCTs, plasma GL-3 levels were assessed as secondary outcome measures (table III).[30,34,38] Both agalsidase beta and agalsidase alfa produced significant decreases in mean plasma GL-3 levels compared with baseline.[30,34,38] Rapid and complete clearance of plasma GL-3 to below detectable levels (<1.2 μg/mL) was achieved by week 14 of treatment with agalsidase beta (1.0 mg/kg EOW) in a single RCT.[30] The effect of treatment with agalsidase alfa (0.2 mg/kg EOW) for 24 weeks in two RCTs was a partial clearance at a magnitude of 54%[34] and 45%.[38]
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab3.jpg
Table III

Plasma globotriaosylceramide (GL-3) assessed in clinical trials of enzyme replacement therapy in patients with Fabry disease; data expressed as mean (SD)

Complete clearance of GL-3 with agalsidase beta was also reported in open-label studies, with four studies of agalsidase beta (1.0 mg/kg EOW; treatment durations of 5–54 months) showing rapid normalization of plasma GL-3 levels.[43,44,48,49] A fifth study[39] was a dose-finding study of 15 male patients with Fabry disease receiving five doses of agalsidase beta at one of five dosing regimens: 0.3 mg/kg EOW, 1.0 mg/kg EOW, 3.0 mg/kg EOW, 1.0 mg/kg every other day (EOD) or 3.0 mg/kg EOD. That study reported that of the three patients receiving the standard dose of 1.0 mg/kg EOW, plasma GL-3 was completely cleared in two patients following the first infusion. In patients receiving 3.0 mg/kg EOW, plasma GL-3 was completely cleared after the first infusion in all three patients. With 0.3 mg/kg EOW, plasma GL-3 reached the lowest value only following the fifth infusion. In patients receiving 1.0 or 3.0 mg/kg EOD, the decreases were less than those observed in patients with EOW dosing, with lowest plasma GL-3 levels following the fourth infusion.[39]

Mean plasma GL-3 levels were reduced by 50% in two open-label studies of agalsidase alfa.[51,52] In a randomized, open-label study comparing the standard regimen of agalsidase alfa (0.2 mg/kg EOW) with agalsidase beta administered at one-fifth of the approved dose (0.2 mg/kg EOW),[54] plasma GL-3 was elevated above the normal level in 8 of 13 patients in the agalsidase beta group and 10 of 16 patients in the agalsidase alfa group at baseline. Concomitant drug therapy at baseline included ACE inhibitors or angiotensin II receptor antagonists or blockers (ARBs), β-blockers, diuretics, calcium channel antagonists and aspirin. A significant decline was seen in both groups at 12 and 24 months, and no significant differences were observed between agalsidase alfa 0.2 mg/kg EOW and agalsidase beta 0.2 mg/kg EOW.[54]

To summarize the results of all the studies measuring plasma GL-3, both agalsidase beta and agalsidase alfa were associated with significant decreases in mean plasma GL-3 levels in RCTs and open-label studies. Reductions with agalsidase beta 1.0 mg/kg EOW have been reported to lead to rapid normalization of GL-3 levels in plasma, whereas partial clearance was reported with agalsidase alfa 0.2 mg/kg EOW.

2.2.2 Kidney GL-3

Studies that focused on histological measures of GL-3 accumulation in and around the endothelial vasculature of the kidney were predominantly performed with agalsidase beta (table IV). One RCT showed that the primary study endpoint of GL-3 deposits in the kidney decreased significantly after treatment with agalsidase beta 1.0 mg/kg EOW, with biopsies showing normalization in most (69%) treated patients.[30] Open-label studies using histological measures of GL-3 accumulation found that GL-3 clearance from kidney capillary endothelium and multiple other renal cell types had been achieved after only 5–6 months of ERT EOW, and was maintained during 54 months of agalsidase beta treatment.[40,49] By month 54, agalsidase beta was effective in improving or clearing GL-3 from podocytes, with a 1-point reduction in four of six patients from baseline podocyte GL-3 inclusions on an ordinal scale ranging from score 0 (no or trace amounts of GL-3) to score 3 (large GL-3 accumulations with specific clusters and bulging of the vessel lumens). Podocytes were not completely cleared of GL-3 deposits. However, all patients assessed had a 0 score for distal convoluted tubule/collecting duct cells at 54 months. In two other open-label studies reporting histological outcomes, GL-3 accumulation in kidney capillary endothelial cells was normalized after 20 weeks of treatment with agalsidase beta 1.0 mg/kg EOW in 12 of 13 patients (92%) in one study,[44] and mean GL-3 content decreased by 82% in four of five patients after treatment with five infusions in a dose-finding study.[39] The first study[44] also demonstrated complete clearance of GL-3 from glomerular endothelial cells and noncapillary (arteriolar) interstitial endothelial cells.
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab4.jpg
Table IV

Kidney globotriaosylceramide (GL-3) assessed in clinical trials of enzyme replacement therapy in patients (pts) with Fabry disease

In a single RCT of agalsidase alfa 0.2 mg/kg EOW, treatment for 6 months did not significantly alter the secondary outcome measure of kidney GL-3 accumulation with respect to baseline.[34]

In summary, normalization of GL-3 in multiple renal cell types (with the exception of podocytes) has been shown for agalsidase beta 1.0 mg/kg EOW in one RCT as a primary endpoint and multiple open-label studies, whereas no treatment effect of agalsidase alfa 0.2 mg/kg EOW was found in one RCT as a secondary outcome.

2.2.3 Urinary GL-3

In the RCTs, urinary GL-3 excretion was assessed as a secondary study outcome (table V). Agalsidase beta 1.0 mg/kg EOW and agalsidase alfa 0.2 mg/kg EOW both demonstrated a reduction from baseline in urinary GL-3 excretion, although it was not apparent whether these reductions represented a partial normalization associated with any clinically meaningful effect.[30,34,38] Comparison between treatments was difficult, as GL-3 excretion was not reported in the same units (nmol/g creatinine, nmol/24 hours or nmol/filter). Small, nonsignificant reductions in urinary GL-3 excretion were reported in open-label studies.[44,53,54]
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab5.jpg
Table V

Urinary globotriaosylceramide (GL-3) assessed in clinical trials of enzyme replacement therapy in patients with Fabry disease; data expressed as mean (SD)

2.2.4 Cardiac GL-3

GL-3 accumulation in the cardiovascular endothelium measured as a secondary efficacy endpoint decreased with agalsidase beta in one RCT[30] and in two open-label studies (tables VI and VII).[39,49] In the RCT with agalsidase beta 1.0 mg/kg EOW, active treatment for 20 weeks led to normalization of GL-3 level in myocardial endothelial cells, as assessed by histological examination following biopsy.[30] In an RCT with agalsidase alfa 0.2 mg/kg EOW that assessed cardiac GL-3 content as its primary endpoint, some patients (20%) showed decreases in myocardial GL-3 content (homogenized heart biopsies), but no agalsidase alfa-associated treatment effect could be demonstrated, as the difference between active treatment and placebo did not reach statistical significance (p = 0.42).[38] In one open-label dose-finding study, agalsidase beta treatment yielded a near normalization of GL-3 level in cardiac vascular endothelial cells assessed by ELISA, which was confirmed by histological examination.[39] In a 54-month open-label study, 80% of patients treated with agalsidase beta 1.0 mg/kg EOW had normalization of GL-3 in cardiac capillary endothelial cells by month 6.[49]
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab6.jpg
Table VI

Cardiac endpoints, including cardiac globotriaosylceramide (GL-3) level, cardiopulmonary function, imaging findings and electrocardiography assessed in clinical trials of enzyme replacement therapy in patients with Fabry disease; data expressed as mean (SD). More open-label studies with agalsidase beta are listed in table VII

https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab7.jpg
Table VII

Cardiac endpoints, including cardiac globotriaosylceramide (GL-3) level, cardiopulmonary function, imaging findings and electrocardiography assessed in open-label studies of enzyme replacement therapy with agalsidase beta in patients with Fabry disease; data expressed as mean (SD). More studies are listed in table VI

To summarize, agalsidase beta 1.0 mg/kg EOW has shown significant reductions in levels of GL-3 in cardiac vascular endothelial cells in RCTs. In open-label studies, these reductions have been reported to lead to normalization of GL-3 levels in capillary endothelium in cardiac tissue. Data for agalsidase alfa 0.2 mg/kg EOW did not reach statistical significance.

2.2.5 Skin GL-3

Accumulation of GL-3 in the skin was investigated in a total of five studies, all of agalsidase beta. In one RCT including skin GL-3 as a secondary endpoint, 20 weeks of treatment with agalsidase beta 1.0 mg/kg EOW resulted in 20 of 29 patients with score 0 (free of microvascular endothelial deposits of GL-3 in skin biopsy specimens).[30] After 6 months of extension as an open-label treatment, skin biopsies showed complete normalization in 96% of patients.[30] A second RCT with agalsidase beta treatment 1.0 mg/kg EOW for 5 months, which defined GL-3 clearance from dermal cell types as its primary efficacy endpoint, showed complete normalization of GL-3 level in superficial endothelial cells in 100% of patients, in deep endothelial cells (maintaining or achieving score 0) in 85% of patients, in smooth muscle cells in 33% of patients, and in the perineurium in 4% of patients.[31] In an open-label study of agalsidase beta, complete normalization of skin GL-3 level in superficial vascular endothelial cells was reported within 24 weeks in 100% of children, whereas moderate to severe deposits in the deep vascular endothelial cells were cleared or reduced to mild (based on a 4-category scoring system — none, mild, moderate, severe — for GL-3 accumulation).[56] Complete normalization of skin GL-3 level was reported in open-label studies of agalsidase beta after 30 months in 98% of adult patients[43] and after 54 months in 86% of adult patients.[49] No studies with agalsidase alfa addressed GL-3 levels in skin.

2.3 Neuropathy and Pain

Pain outcomes in Fabry disease were reported as a primary endpoint in one RCT with agalsidase alfa[34] and as a secondary measure in one RCT with agalsidase beta.[30] Seven open-label studies of agalsidase beta[39,4244,47,49,54] and two open-label studies of agalsidase alfa[50,54] also reported pain outcomes (table VIII).
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab8.jpg
Table VIII

Pain, including neuropathic pain outcomes assessed in clinical trials of enzyme replacement therapy in patients with Fabry disease; data expressed as mean (SD)

In the agalsidase beta RCT, Eng et al.[30] reported significant improvements in pain scores after 20 weeks of treatment in both the agalsidase beta 1.0 mg/kg EOW and placebo groups, despite low baseline pain scores. A possible placebo effect could not be ruled out. In the only RCT investigating pain outcomes as a primary endpoint, Schiffmann et al.[34] reported that 4 of 11 patients treated with agalsidase alfa 0.2 mg/kg EOW who were taking neuropathic pain medications at baseline discontinued these medications after 1–8 weeks of the study (mean 30.5 days) compared with none of 11 patients receiving placebo. Patients in the agalsidase alfa group were able to remain longer without pain medications compared with patients in the placebo group (74.5 vs 12.9 days; p = 0.02). A further RCT of agalsidase alfa 0.2 mg/kg EOW with a 12-month open-label extension investigated epidermal re-innervation. The investigators reported no significant difference in intraepidermal nerve fibre density between the treatment groups at 6 months and thermal thresholds remained unchanged, indicating the absence of epidermal nerve fibre regeneration.[37]

One open-label study of agalsidase alfa 0.2 mg/kg EOW showed significant reduction in pain after 6 months of treatment, but no further pain score reductions were observed thereafter.[50]

Agalsidase beta 1.0 mg/kg EOW has shown some benefit in open-label studies. In a cohort of patients with low pain scores at study entry, Wilcox et al.[43] reported that 5 of 58 (8.6%) patients stopped all pain medications and 6 of 58 (10.3%) had a reduction in dose and/or frequency of their pain medications after 30 months of open-label agalsidase beta treatment. Of the patients who reported pain at baseline, 54 months of agalsidase beta treatment was associated with a significant improvement in pain scores.[49] In 22 patients who underwent open-label treatment with agalsidase beta for 18–23 months, all patients reported a subjective reduction in frequency and intensity of pain.[42] Measures of neuropathy, including detection thresholds of vibration, heat-pain onset and intermediate heat-pain severity, were significantly improved following agalsidase beta treatment, with intermediate heat-pain in particular normalizing in 16 of 20 patients with initially abnormal thresholds.[42] Significant improvements from baseline in pain scores, including neuropathy, were also reported in a further open-label study of agalsidase beta,[39] while marked improvements were noted (but not statistically tested) in another.[47] The open-label study comparing agalsidase alfa 0.2 mg/kg EOW with agalsidase beta at one-fifth of the approved dose (i.e. 0.2 mg/kg EOW) reported no significant reduction in pain scores with either therapy after 24 months.[54]

To summarize, both agalsidase beta 1.0 mg/kg EOW and agalsidase alfa 0.2 mg/kg EOW have been reported to reduce patient pain scores in a single RCT each, although a possible placebo effect on this measure was suggested. In open-label studies, agalsidase beta 1.0 mg/kg EOW normalized heat-pain thresholds, whereas agalsidase alfa 0.2 mg/kg EOW had no effect on thermal thresholds.

2.4 Sweat Function

No RCTs addressed sweat function. In four open-label studies, both agalsidase beta and agalsidase alfa improved sweat function.[39,42,50,53]

In the two open-label studies of agalsidase beta, patients subjectively reported an increased ability to sweat.[39,42] Schiffmann et al.[50] measured pre- and post-ERT sweat excretion at a single timepoint in 17 patients and reported an improvement in sweat function 24–72 hours after treatment with agalsidase alfa 0.2 mg/kg EOW. Four patients normalized and seven remained nonresponders after treatment. The overall treatment effect of agalsidase alfa 0.2 mg/kg waned after 7 days and sweat function returned to pretreatment levels, leading the investigators to conclude that a more frequent administration or higher enzyme dose may be required.[50] In a second open-label study in which patients previously treated with agalsidase alfa 0.2 mg/kg EOW were switched to weekly dosing of agalsidase alfa 0.2 mg/kg, sweat function did not change significantly from baseline at any time during the study.[53]

2.5 Kidney Function

The effect of agalsidase beta treatment on measures of kidney function was assessed as a secondary study endpoint in two RCTs[30,33] and in five open-label studies.[43,44,47,49,54] Agalsidase alfa efficacy was investigated in one RCT[34] as a secondary endpoint and four reports of open-label studies.[37,51,53,54]

2.5.1 Glomerular Filtration Rate

In patients with normal kidney function or early-stage kidney disease, both agalsidase beta 1.0 mg/kg EOW and placebo yielded a stable mean GFR relative to baseline after 5 months of follow up (table IX).[30,33] In the phase IV study, both groups then received treatment with agalsidase beta 1 mg/kg and were observed for 35 months.[33] During this observation period, renal function remained stable for the entire study population. Of 58 patients enrolled in a 5-month RCT who subsequently received up to 54 months of agalsidase beta therapy in an open-label extension study, mean GFR remained in the normal range during the 5-year extension phase.[43,49] Six of ten patients with baseline proteinuria above 1 g/24 hours and four of six patients with baseline glomerulosclerosis of at least 50% received either ACE inhibitors or ARBs during the course of the study.[43,49] Stabilization of GFR was reported in three other open-label studies of agalsidase beta,[44,47,54] although patients with a GFR of <90 mL/min/1.73 m2 at baseline in one study experienced deterioration in renal function.[47]
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab9.jpg
Table IX

Estimated glomerular filtration rate assessed in clinical trials of enzyme replacement therapy in patients with Fabry disease; all values based on creatinine clearance; data expressed as mean (SD)

In an RCT with agalsidase alfa 0.2 mg/kg EOW, Schiffmann et al.[34] reported a mean improvement in creatinine clearance in patients treated with agalsidase alfa compared with declining creatinine clearance in patients receiving placebo after 24 weeks.

An open-label study reported stable GFR after 2 years of treatment with agalsidase alfa 0.2 mg/kg EOW.[54] In contrast, GFR deteriorated significantly after 48 months of double-blind and open-label agalsidase alfa treatment.[37,51] Although this decline was predominantly driven by four patients with stage III chronic kidney disease, functional decline was observed in patients of all disease stages.[51] Moreover, 11 of 24 patients in the sample were treated with ACE inhibitors, including all four patients with stage III chronic kidney disease, which might have prevented a more rapid decline in function.[51] Finally, in 11 patients who showed a progressive decline in GFR over an average of 3.4 years of therapy with agalsidase alfa 0.2 mg/kg EOW, switching to weekly agalsidase alfa slowed, but did not prevent, further progression.[53] Seven of 11 patients had to be started on ACE inhibitor or ARB treatment prior to entry, at entry or during the 2-year study period.[53]

2.5.2 Proteinuria

Proteinuria is an independent risk factor for progression of renal disease.[33,47,49,51] In the only RCT investigating the effects of ERT on proteinuria (as a secondary measure), Banikazemi et al.[33] reported stable proteinuria from baseline to the end of follow-up (35 months) in patients receiving either agalsidase beta 1.0 mg/kg EOW or placebo (table X). At or before baseline, 35% of patients were treated with ACE inhibitors or ARBs. Longitudinal analyses suggested a nonsignificant improvement in proteinuria over time for the agalsidase beta group compared with placebo.[33] In open-label studies, proteinuria was stable with agalsidase beta treatment for up to 54 months of follow up.[47,49,54]
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab10.jpg
Table X

Proteinuria assessed in clinical trials of enzyme replacement therapy in patients with Fabry disease

In two open-label studies of agalsidase alfa, proteinuria worsened in one 36-month study[51] and was stable after 24 months of follow-up in the other.[54] In another study in which 11 patients previously receiving agalsidase alfa 0.2 mg/kg EOW were switched to weekly dosing, agalsidase alfa was associated with improvement in proteinuria in six patients, no change in one and worsening in four patients. It is important to note that seven of the 11 patients were receiving ACE inhibitor or ARB treatment and this had remain the same for at least 6 months before agalsidase alfa therapy was initiated.[53]

2.6 Composite Endpoints, Including Renal, Cardiac or Cerebrovascular Events, or Death

In the only RCT that included a composite, hard, primary endpoint, agalsidase beta 1.0 mg/kg EOW compared with placebo slowed progression to a renal, cardiac or cerebrovascular event, or death in 82 patients with advanced Fabry disease and mild to moderate kidney disease.[33] The observed 53% risk reduction in the entire study group after adjustment for imbalance in baseline proteinuria did not reach statistical significance (p = 0.06 for agalsidase beta vs placebo), whereas secondary analyses of protocol-adherent patients showed a significant (61%) risk reduction in those who followed the protocol (p = 0.034 agalsidase beta vs placebo). Four patients in each group were excluded in the secondary, protocol-adherent analyses. The treatment effect was greater in the group of patients with baseline estimated GFR over 55 mL/min/1.73 m2. This group showed a significant (81%) risk reduction (p = 0.025 agalsidase beta vs placebo), highlighting the need for treatment in the early course of the disease.

2.7 Cardiac Function

In an RCT in which four patients were treated and available for follow up, the primary endpoint of exercise tolerance seemed to increase in patients receiving agalsidase beta 1.0 mg/kg EOW, and was associated with nonsignificant improvements in secondary outcomes of maximum oxygen consumption, mean oxygen pulse and estimated stroke volume not seen in patients receiving placebo (table VI).[32] Cardiac outcomes were also considered in six published reports of open-label studies of agalsidase beta. Weidemann et al.[41] demonstrated a reduction in left ventricular (LV) mass after 36 months of open-label treatment with agalsidase beta 1.0 mg/kg EOW, along with an improvement of regional myocardial function and a higher exercise capacity in patients without myocardial fibrosis at baseline. Minor or no improvements were seen in patients with mild or severe fibrosis. Beer et al.[46] reported significant reductions in LV mass during treatment with agalsidase beta 1.0 mg/kg EOW, as well as improvements in regional myocardial function in patients without late enhancement, whereas late enhancement was associated with impaired cardiovascular improvement.

Cardiac hypertrophy was investigated in three other open-label studies,[47,48,54] with agalsidase beta therapy associated with stable cardiac function in two studies,[48,54] and significant improvements in echocardiographical measures in patients with normal renal function in another.[47] In nine patients undergoing dialysis, Pisani et al.[45] showed a nonsignificant improvement in LV mass index after 2 years of treatment with agalsidase beta 1.0 mg/kg EOW.

Cardiac outcomes were assessed as secondary endpoints in two RCTs of agalsidase alfa 0.2 mg/kg EOW.[34,38] In one of these RCTs, the investigators reported a significant decrease in QRS-complex duration in agalsidase alfa-treated patients with normal cardiac function, although the result was contingent on a single patient with a right bundle branch block that resolved during the study.[34] In the second RCT, Hughes et al.[38] intended to enrol patients with increased LV mass, but four of the seven patients initially receiving placebo had normal LV mass as opposed to one patient in the agalsidase alfa group. That study also failed to demonstrate any effects associated with agalsidase alfa treatment on cardiac Fabry disease based on heart content of GL-3, cardiac histological pathology, echocardiographic and MRI outcomes, and ECG. The decrease in LV mass by week 13 in the agalsidase alfa group came almost entirely from one patient. Furthermore, a placebo patient with the largest baseline LV mass showed the largest decrease in LV mass and was eliminated from the prospectively planned analyses. No significant reduction in cardiac mass was observed in the open-label study of agalsidase alfa.[54]

2.8 Special Populations

2.8.1 Women

One open-label study of agalsidase alfa 0.2 mg/kg EOW was conducted in 15 women with Fabry disease with treatment durations varying from 3 to 12 months.[55] Baseline plasma GL-3 and urinary GL-3 values were variable and, 27 weeks after the start of ERT, the changes did not reach statistical significance. Renal function did not deteriorate and significant reductions from baseline were reported in both QRS interval (p = 0.003) and LV mass (p = 0.003) [table II].

2.8.2 Children

Two 6-month open-label studies of agalsidase alfa 0.2 mg/kg EOW were in paediatric populations with Fabry disease (table II, table XI).[5759] At baseline, mean plasma GL-3 level was elevated in boys but not girls in both studies. In the boys, treatment with agalsidase alfa led to a significant reduction in mean plasma GL-3 level (p < 0.001).[57] In contrast, plasma GL-3 level in the girls did not change from baseline values.[57] Sweat function (according to the Quantitative Sudomotor Axon Reflex Test) and pain (Brief Pain Inventory) improved during therapy, but differences were not significant. Estimated GFR, which was either normal or elevated at baseline, was stable during treatment.[57] In one study, 4 of 21 children with baseline microalbuminuria and a single patient with baseline proteinuria experienced a reduction in protein excretion with agalsidase alfa treatment 0.2 mg/kg EOW.[57]
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab11.jpg
https://static-content.springer.com/image/art%3A10.2165%2F11318300-000000000-00000/MediaObjects/40265_2012_69162179_Tab11a.jpg
Table XI

Efficacy outcomes in paediatric patients assessed in clinical trials of enzyme replacement therapy in patients with Fabry disease

One 48-week open-label study of agalsidase beta 1.0 mg/kg EOW was carried out in a paediatric population of 14 male and 2 female patients.[56] At baseline, plasma and skin GL-3 levels were elevated in boys, but not in girls. Treatment with agalsidase beta resulted in complete normalization of GL-3 levels in plasma and skin. In 8 of 15 patients with mild proteinuria (≥100 mg/m2/24 h) at baseline, the proteinuria resolved completely in five patients, remained unchanged in two and deteriorated in one. The percentage of patients who reported gastrointestinal symptoms of postprandial pain, nausea and vomiting decreased significantly (p = 0.031) following initiation of agalsidase beta treatment. No new Fabry disease symptoms developed during therapy.

2.9 Safety Profile

Mild to moderate infusion reactions are the most common adverse events with ERT and occur mostly in the form of rigor and fever. In RCTs of agalsidase beta 1.0 mg/kg EOW, rigor (48% vs 0%; p = 0.004) and fever (24% vs 3%; p = 0.024) occurred more commonly with agalsidase beta than with placebo in one trial,[30] and in another trial 55% of patients in the agalsidase beta group experienced infusion reactions compared with 23% in the placebo group.[33] In RCTs of agalsidase alfa 0.2 mg/kg EOW, the incidence of infusion reactions ranged from 4% to 57%.[34,3638] Immunoglobulin G (IgG) antibody formation has been reported in 68–88% of patients in RCTs with agalsidase beta 1.0 mg/kg EOW[30,33] and in 20–56% of patients in RCTs with agalsidase alfa 0.2 mg/kg EOW.[34,37,38] This difference in the rates of antibody formation is most likely due to the different dose regimens used for the agalsidase alfa and beta preparations.

3. Discussion

Fabry disease is associated with a complex array of signs and symptoms, many of which result from the initial pathological event, i.e. accumulation of GL-3 and its metabolite lysoGb3 in the vascular endothelium and a variety of other cell types. Classically, affected males — but also an important proportion of affected females[4] — develop anomalies of the kidneys, heart, brain, and the peripheral and autonomic nervous systems, with potential fatal consequences including renal failure, and cerebrovascular and cardiovascular events.[7]

The purpose of this systematic review was to critically examine data regarding efficacy of the two available regimens of ERT for treatment of patients with Fabry disease, i.e. agalsidase beta at a dose of 1.0 mg/kg EOW and agalsidase alfa at a dose of 0.2 mg/kg EOW. This evidence should be placed into a clinically useful context that considers not only the quality of the studies, but also the quality of the endpoints. Precedence has been given to RCTs and open-label extension studies of these RCTs. We have evaluated outcomes at different levels of the disease process, ranging from cellular GL-3 storage, to tissue involvement and clinical endpoints such as organ failure and death. A total of nine published reports of RCTs and 23 published reports of open-label studies were identified.

To date, ERT with agalsidase beta at a dose of 1.0 mg/kg has demonstrated improvement in hard clinical outcomes in the setting of one RCT, although this effect only reached significance in a secondary analysis of protocol-adherent patients. Specifically, agalsidase beta 1.0 mg/kg EOW has shown a benefit relative to placebo in delaying progression to a renal, cardiac or cerebrovascular event, or death in patients with mild to moderate kidney disease.[33] Agalsidase beta 1.0 mg/kg EOW has additionally demonstrated clearance of GL-3 from plasma, and kidney, heart and skin tissue. Improvement of clinical symptoms in the form of significant amelioration in pain scores and including neuropathy have been demonstrated for ERT with agalsidase alfa at a dose of 0.2 mg/kg EOW. Significant reductions of GL-3 by agalsidase alfa 0.2 mg/kg EOW have been demonstrated in plasma.

There were more studies of agalsidase alfa than agalsidase beta in special populations. In the only study specifically assessing ERT effects in female patients, a significant improvement in cardiac function was found after 3–12 months of treatment with agalsidase alfa 0.2 mg/kg EOW in 15 women.[55] With regard to paediatric populations, the single open-label study of agalsidase beta 1.0 mg/kg EOW showed that there was rapid normalization of plasma GL-3 levels and clearance of GL-3 from skin cells.[56] Treatment with agalsidase alfa 0.2 mg/kg EOW partially cleared GL-3 in two open-label studies of 24 and 13 children.[57,58] A nonsignificant improvement in pain and sweat function was observed, but requires further supporting evidence. Overall, studies of agalsidase beta and agalsidase alfa in women or paediatric populations were few and had low statistical power due to low patient numbers.

With regard to safety, the most common adverse events reported with ERT in Fabry disease are infusion reactions. Immunogenicity to therapeutic proteins is common, but data on the incidence of antibody formation in response to ERT in Fabry disease are nonconclusive at present. This is mainly due to the considerable variability that exists in the antibody assays used. The Fabry Antibody Standardization Initiative has been launched to establish a standard antibody assay and a common reporting methodology, which will allow better assessment of this safety aspect of ERT in the future. It is thought that development of antibodies is closely related to the residual α-galactosidase A enzyme protein.[61] Complete cross-reactivity of IgG antibodies to agalsidase alfa and agalsidase beta has been demonstrated.[62,63]

The only study that directly compared the efficacy of agalsidase alfa and agalsidase beta at their indicated doses of 0.2 mg/kg EOW and 1.0 mg/kg EOW, respectively, was not prospectively designed.[29] Therefore, it was not included in this review. However, it is important to mention that the results showed that the negative effects on efficacy associated with antibody formation may probably be overcome by administration of ERT at the dose of 1 mg/kg EOW. In this study, agalsidase beta 1 mg/kg EOW was compared with either agalsidase alfa 0.2 mg/kg EOW or agalsidase beta 0.2 mg/kg EOW, and LV mass significantly decreased in patients receiving 1 mg/kg EOW, whereas no decrease was seen in the groups receiving 0.2 mg/kg EOW. In addition, the dose of 1 mg/kg was more effective in reducing plasma and urinary GL-3 levels in the presence of neutralizing antibodies.[29]

One randomized, open-label study compared agalsidase alfa at the indicated dose of 0.2 mg/kg EOW with agalsidase beta administered at one-fifth of the approved dose, i.e. 0.2 mg/kg EOW.[54] The investigators reasoned that the comparison of efficacy of agalsidase beta and agalsidase alfa was best addressed at equal weight-dose, choosing the 0.2 mg/kg EOW dose on financial grounds. However, equimolar — as opposed to equal weight — preparations of agalsidase alfa and agalsidase beta have shown similar effects on GL-3 clearance in fibroblasts.[64] The comparative study of agalsidase alfa 0.2 mg/kg EOW and agalsidase beta 0.2 mg/kg EOW reported no differences in the primary endpoint of reduction in LV mass after 12 or 24 months, or of other outcomes such as renal failure, GFR, pain and decline in GL-3 levels.[54] At the selected doses, an undesirably high rate of treatment failure was reported in this study, which seems to support the fact that for both agalsidase preparations a dose of 0.2 mg/kg EOW was suboptimal. In this context, it is also of concern that investigators in studies of agalsidase alfa have reported limited effectiveness of 0.2 mg/kg EOW dosing, prompting an increase in frequency to weekly dosing of agalsidase alfa in several patients.[50,53]

Key considerations in the management of Fabry disease are the heterogeneity of patients’ symptoms and disease staging, as well as timely intervention. The greatest benefits of ERT have been reported in patients who had milder impairment of renal or cardiac function at the start of therapy,[33,41,46,49] and ERT is less likely to provide quick or effective control of disease in patients with more advanced disease. ERT cannot repair organ damage; instead delaying or preventing late complications are among the key therapeutic goals. Cardiac parameters such as the LV mass index have been reported to be better in Fabry patients taking ERT with renal transplants than in those remaining on maintenance dialysis.[65] Thus, patients with Fabry disease should be identified without undue delay and the patient’s family members should be screened for previously undiagnosed Fabry disease. Furthermore, in the situation of a lifelong, progressive disorder such as Fabry disease, the clinical results of ERT need long-term monitoring, which highlights the importance of Fabry disease registries.

4. Conclusions

We have critically examined the available data regarding efficacy of ERT at a dose of 1.0 mg/kg EOW with agalsidase beta and at a dose of 0.2 mg/kg EOW with agalsidase alfa, and placed the evidence into a clinically useful context that considers not only the quality of the studies, but also the quality of the endpoints. A large variety of reported endpoints and ways of measuring them complicates a rigorous comparison between the two ERT regimens. Further research is needed to address the role of ERT at different stages of the disease process and outcomes should be reported in a consistent way. Specific gaps in our knowledge relate to the clinical relevance of clearance of GL-3 accumulation, the role of neutralizing agalsidase antibody activity, when to start therapy, and whether early intervention enhances clinical benefits and prevents major organ damage. Based on the current analysis, ERT at a dose of 0.2 mg/kg with agalsidase alfa and at a dose of 1.0 mg/kg with agalsidase beta differ in their ability to provide effective reduction of the subclinical marker of GL-3 storage, which is most likely due to the different dosages. Complete normalization of GL-3 levels in plasma, skin, and renal and cardiac tissue has been demonstrated with agalsidase beta 1.0 mg/kg EOW, whereas agalsidase alfa 0.2 mg/kg EOW generally provides partial clearance in these tissues.

Although the beneficial effects of ERT with either dose or preparation are variable, the currently available data are more robust for ERT at a dose of 1 mg/kg EOW than a dose of 0.2 mg/kg EOW in Fabry disease based on clinical trial design and efficacy both in terms of normalizing tissue storage of GL-3 in major organs and in terms of ability to reduce the risk of renal, cardiac or cerebrovascular events, or death.

Acknowledgements

The authors received writing/editorial support in preparation of this article, funded by Genzyme Europe BV. Hester van Lier of Excerpta Medica provided writing support. The authors maintained full and independent responsibility for the content of this paper.

R.M. Schaefer certifies that there is no actual or potential conflict of interest in relation to this article, except that he received fees for lectures and consulting from both Genzyme and Shire. A. Tylki-Szymańska received fees for lectures from Genzyme and Shire. M.J. Hilz received funding support for studies, honoraria for scientific presentations, and compensation as a consultant from Genzyme Inc., Cambridge, MA, USA.

Copyright information

© Adis Data Information BV 2009