Anti-α-galactosidase A antibody response to agalsidase beta treatment: Data from the Fabry Registry

Molecular Genetics and Metabolism 105 (2012) 443–449

Contents lists available at SciVerse ScienceDirect

Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme

Anti-α-galactosidase A antibody response to agalsidase beta treatment: Data from the Fabry Registry William R. Wilcox a, b,⁎, Gabor E. Linthorst c, Dominique P. Germain d, Ulla Feldt-Rasmussen e, Stephen Waldek f, Susan M. Richards g, Dana Beitner-Johnson g, Marta Cizmarik g, J. Alexander Cole g, Wytske Kingma g, David G. Warnock h a

Medical Genetics Institute, Cedars-Sinai Medical Center, Los Angeles, CA, USA Department of Pediatrics, UCLA School of Medicine, Los Angeles, CA, USA c Academic Medical Center, Amsterdam, Netherlands d University of Versailles, Hôpital Raymond Poincaré, Garches, France e National University Hospital, Copenhagen, Denmark f Salford Royal NHS Foundation Trust, Manchester, UK g Genzyme, Cambridge, MA, USA h Division of Nephrology, Department of Medicine, University of Alabama at Birmingham, Birmingham, AL, USA b

a r t i c l e

i n f o

Article history: Received 1 October 2011 Received in revised form 8 December 2011 Accepted 8 December 2011 Available online 14 December 2011 Keywords: Fabry disease Lysosomal storage diseases Alpha-galactosidase Enzyme replacement therapy Glycosphingolipids Antibody formation

a b s t r a c t Agalsidase beta, a form of recombinant human α-galactosidase A (αGAL), is approved for use as enzyme replacement therapy (ERT) for Fabry disease. An immunogenic response against a therapeutic protein could potentially impact its efficacy or safety. The development of anti-αGAL IgG antibodies was evaluated in 571 men and 251 women from the Fabry Registry who were treated with agalsidase beta. Most men developed antibodies (416 of 571, 73%), whereas most women did not (31 of 251, 12%). Women were also significantly more likely to tolerize than men; whereas 18 of 31 women tolerized (58%, 95%CI: 52%–64%), only 47 of 416 men tolerized during the observation period (11%, 95% CI: 8%–15%). Patients who eventually tolerized had lower median peak anti-αGAL IgG antibody titers than patients who remained seropositive at their most recent assessment (400 versus 3200 in men, 200 versus 400 in women, respectively). Patients with nonsense mutations in the GLA gene were more likely to develop anti-αGAL IgG antibodies than patients with missense mutations. Approximately 26% of men (151 of 571) reported infusion-associated reactions (IARs), compared to 11% of women (27 of 251). Men who developed anti-αGAL IgG antibodies were more likely to experience IARs compared to those who remained seronegative. Nine percent of seronegative men and women (34 of 375) reported IARs. The majority of IARs occurred during the first 6 to 12 months of agalsidase beta treatment and decreased over time, in both seroconverted and seronegative patients. © 2012 Elsevier Inc. All rights reserved.

1. Introduction Fabry disease is a rare lysosomal storage disorder caused by a deficiency of the lysosomal enzyme α-galactosidase A (αGAL) and the resulting progressive accumulation of globotriaosylceramide (GL-3) and other glycosphingolipids within various cells and tissues [1,2]. Over time, this accumulation of GL-3 is associated with renal disease progression, cardiovascular disease and strokes [3–8]. Because Fabry disease is X-linked, hemizygous males typically experience the most severe manifestations of Fabry disease [1,2]. However, many

Abbreviations: CI, confidence interval; ERT, enzyme replacement therapy; αGAL, αgalactosidase A; GPS&RM, Genzyme Global Patient Safety & Risk Management; IAR, infusion-associated reaction. ⁎ Corresponding author at: Medical Genetics Institute, Cedars-Sinai Medical Center, Los Angeles, CA 90048, USA. E-mail address: [email protected] (W.R. Wilcox). 1096-7192/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2011.12.006

heterozygous females also develop substantial signs and symptoms of Fabry disease, including renal failure, cerebrovascular events, and cardiovascular events [4–6,8]. Agalsidase beta, a form of recombinant human αGAL, is approved for use as enzyme replacement therapy (ERT) for treatment of Fabry disease (Fabrazyme®, Genzyme, a Sanofi company, Cambridge, MA). In clinical studies, agalsidase beta was shown to clear microvascular endothelial GL-3 deposits from the kidney, heart, and skin [9,10], and to improve clinical outcomes when initiated before the onset of irreversible organ damage [10,11]. Agalsidase beta is administered via intravenous infusion at the recommended labeled dose of 1.0 mg/kg/2 weeks. Like other therapeutic proteins [12,13], agalsidase beta can cause immunogenicity [14,15], especially if patients have little or no endogenous enzyme. Anti-αGAL IgG antibodies could potentially impact the safety and efficacy of ERT in certain patients, possibly by altering its biodistribution or metabolic clearance [13]. Regular monitoring of serum

444

W.R. Wilcox et al. / Molecular Genetics and Metabolism 105 (2012) 443–449

samples for the presence of anti-αGAL IgG antibodies is recommended as part of the routine care for patients who receive ERT. The formation of anti-αGAL IgG antibodies was monitored in patients treated with agalsidase beta in placebo-controlled Phase 3 and Phase 4 clinical studies [9–11]. A more detailed retrospective analysis of data from these clinical trials and their corresponding open-label extension studies evaluated the impact of anti-αGAL IgG antibodies on efficacy [14]. However, randomized clinical trials of agalsidase beta included relatively limited numbers of subjects and most were men. In a prospective study by Vedder et al. [15], the immunogenicity and safety profile of agalsidase beta was found to be similar to that of agalsidase alfa (another form of recombinant human αGAL, Shire Human Genetic Therapies, Inc., Cambridge, MA), when administered at 0.2 mg/kg/2 weeks. The Fabry Registry is an observational database that compiles clinical and laboratory data from both treated and untreated patients; it was established to further investigate the long-term effects of ERT and the natural progression of Fabry disease in a larger and more diverse population. The objective of the present analysis was to evaluate the development of anti-αGAL IgG antibodies in Fabry Registry men and women who were treated with agalsidase beta. 2. Methods The Fabry Registry began enrolling treated and untreated patients in April 2001. As of 02 April 2010, it included 3552 patients (1760 men and 1792 women). All patients with Fabry disease are eligible to enroll in the Fabry Registry, regardless of age, gender, symptoms, or whether they are receiving ERT from any commercial source. Patient and physician participation is voluntary. Patients provide informed consent through local Institutional Review Boards/Ethics Committees and may decline to participate or withdraw consent at any time. To be included in the current analyses, patients must have received agalsidase beta as their only source of ERT and must have reported at least 1 post-baseline anti-αGAL IgG assessment. Anti-αGAL IgG antibody titers were assessed in diluted serum samples by enzyme-linked immunoabsorbent assay at the Genzyme Clinical Specialty Laboratory (Framingham, MA, US), which has been used in clinical trials with agalsidase beta, as described in detail by Benichou et al. [14]. Anti-αGAL IgG titers are reported as the reciprocal of the maximum dilution that yielded a positive result above the assay cut-point. Patients were designated as seronegative if they tested negative for anti-αGAL IgG

antibodies each time they were tested. Patients were designated as seroconverted, then tolerized if they had 1 or more negative anti-αGAL IgG tests after previously testing positive and whose most recent assessment was negative. Patients were designated as seroconverted, not tolerized if they had at least 1 positive anti-αGAL IgG test and their most recent assessment was positive. The Fabry Registry recommends that treating physicians collect serum samples for IgG antibody testing from all patients prior to the first agalsidase beta infusion (baseline), every 3 months thereafter for the first 18 months of treatment, then every 6 months until 2 consecutive negative results are confirmed (tolerization). However, treating physicians determine the actual frequency of assessments according to individual patients' needs. Plasma GL-3 levels were measured by tandem mass spectrometry [16] at the Genzyme Clinical Specialty Laboratory. The Fabry Registry Board of Advisors recommends that the treating physician collect plasma for GL-3 levels to be assessed prior to the first infusion, then every 3 months for the first 18 months of treatment, then every 6 months thereafter. The Genzyme Global Patient Safety & Risk Management (GPS&RM) pharmacovigilance database was searched for any infusion-associated reactions (IARs), defined as any related adverse event occurring during or after an agalsidase beta infusion on the day of the infusion among the patients in this cohort. Statistical analyses were performed using SAS statistical software version 9.1 (SAS Institute Inc., Cary, NC). The binomial proportion Fisher's exact test was used to determine the 95% lower and upper confidence intervals (CI) of the percentage of patients who tolerized. Student's two-sided tests were used to perform statistical significance testing of plasma GL-3 levels between seronegative and seroconverted patients (a p-value cut-point of an alpha-level of 0.05). The Chi-square test was used to perform statistical significance testing of the percentage of seronegative versus seroconverted men who reported IARs; the Chi-square exact test was used to perform statistical significance testing of the percentage of seronegative versus seroconverted women who reported IARs (a p-value cut-point of an alphalevel of 0.05). 3. Results As of 02 April 2010, a total of 822 Fabry Registry patients (571 men and 251 women) had reported post-baseline IgG titer data to the Fabry Registry. As shown in Fig. 1, 416 of 571 men (73%) reported Seroconverted, not tolerized (n=13)

Seroconverted, then tolerized (n=47)

Seroconverted, then tolerized (n=18)

Seronegative (n=155)

Seroconverted, not tolerized (n=369)

Seronegative (n=220)

Males

Females

(N=571)

(N=251)

Fig. 1. Most males, but few females develop anti-aGAL IgG antibodies after treatment with agalsidase beta. Seroconversion status is shown for Fabry Registry males and females. Patients designated as Seronegative tested negative for anti-αGAL IgG antibodies each time they were tested. Patients designated as Seroconverted, not Tolerized had at least 1 positive anti-αGAL IgG test, but did not tolerize. Patients designated as Seroconverted, then Tolerized had 1 or more negative anti-αGAL IgG tests after previously testing positive and whose most recent test was negative.

W.R. Wilcox et al. / Molecular Genetics and Metabolism 105 (2012) 443–449

445

Fig. 2. Females who seroconverted had lower peak anti-αGAL IgG titers than males and patients who tolerized had lower peak anti-αGAL IgG titers. Because peak anti-αGAL IgG titers varied widely across patients, the data are presented on a logarithmic scale. The y-axis on the right side of the graph shows the approximate corresponding peak titers. The boxes encompass the 25th–75th percentiles, and the bars indicate the minimum and maximum values. The horizontal black lines within each box represents the natural log of the median peak titer and “x” represents the natural log of the mean peak titer. The numbers of patients in each seroconversion category and gender are shown below the x-axis. Patients designated as Seronegative tested negative for anti-αGAL IgG antibodies each time they were tested. Patients designated as Seroconverted, not Tolerized had at least 1 positive anti-αGAL IgG test, but did not tolerize. Patients designated as Seroconverted, then Tolerized had 1 or more negative anti-αGAL IgG tests after previously testing positive and whose most recent test was negative.

a positive test for anti-αGAL IgG antibodies at some time during the treatment period, including 369 who had a positive test at the time of their most recent assessment and 47 who eventually tolerized (i.e., had a negative anti-αGAL IgG test at the most recent assessment) during the observation period. In contrast, most women remained seronegative; only 31 of 251 women (12%) reported the development of anti-αGAL IgG antibodies, including 13 who remained seropositive and 18 who eventually tolerized. Men were significantly less likely to tolerize than women; whereas 47 of 416 men tolerized (11%, 95% CI: 8%–15%), 18 of 31 women tolerized (58%, 95%CI: 52%–64%). Because the Fabry Registry is an observational database, IgG titer levels were not consistently reported at regular intervals, hence, these data cannot be used to determine when seroconversion occurred. Men who remained seropositive at their most recent assessment had median peak anti-αGAL IgG antibody titers of 3200 (min 100, max 819200), whereas men who eventually tolerized had median peak titers of 400 (min 100, max 12800). Peak titers were lower among the 31 women who seroconverted: median 400 (min 100, max 6400) among women who remained seropositive and median 200 (min 100, max 800) among women who tolerized. These data are displayed graphically (log scale) in Fig. 2. Plasma GL-3 data were available for 379 of 571 men (66%) and 179 of 251 women (71%) in this cohort (Table 1). Mean plasma GL-3 levels varied widely among individual patients, but were within the normal range (≤7.0 μg/mL) in all seroconversion categories. Mean plasma GL3 levels were slightly, but statistically significantly higher in men with anti-αGAL IgG antibodies (4.9 ± 1.92 μg/mL), compared to men who remained seronegative (4.1 ± 1.45 μg/mL, p b 0.001 by t-test) and men who eventually tolerized (4.1 ± 1.29 μg/mL, p = 0.029 by t-test). There were no significant differences in plasma GL-3 levels among women with anti-αGAL IgG antibodies, compared to women who remained seronegative or tolerized and very few women had plasma GL-3 levels outside of the normal range. Over 500 mutations in the α-galactosidase A gene have been identified and most genotypes are specific to individual families (Fabry Registry database). The 20 most common genotypes in the overall Fabry Registry are shown in Table 2. Among this cohort, the most common genotype was the R227X nonsense mutation, which was

reported by 23 men and 16 women. In general, men with nonsense mutations were more likely to develop anti-αGAL IgG antibodies (34 of 37) than were men with missense mutations (29 of 52). Certain missense mutations, such as R342Q, were associated with high rates of seroconversion in men; however, this was based on a small number of patients (8 of 9 men with the R342Q mutation developed antiαGAL IgG antibodies). The Fabry Registry does not track adverse events in patients exposed to agalsidase beta, but such events are captured in the global safety database and should be reported to Genzyme GPS&RM. A review of the GPS&RM database identified 178 Fabry Registry patients in this cohort

Table 1 Seroconversion status and plasma GL-3 levels.

Number of patients tested after first infusion date, N Seronegative, n Plasma GL-3 (μg/mL), n Mean (SD) Median Min, Max Seroconverted, but not tolerized, n Plasma GL-3 (μg/mL), n Mean (SD) Median Min, Max Seroconverted, then tolerized, n Plasma GL-3 (μg/mL), n Mean (SD) Median Min, Max

Males

Females

571 155 119 4.1 (1.45) 4.0 2.1, 11.3 369 229 4.9 (1.92)⁎,† 4.6 0.0, 17.0 47 31 4.1 (1.29) 3.8 2.0, 8.5

251 220 160 3.7 (1.21) 3.6 2.0, 10.7 13 8 3.8 (0.80) 4.2 2.3, 4.3 18 11 3.2 (0.74) 3.1 2.2, 4.4

Plasma GL-3 data are expressed as μg/mL. Normal plasma GL-3 range is defined as ≤7.0 μg/mL. For Seronegative patients, plasma GL-3 values represent the most recent data available. For Seroconverted, but Not Tolerized patients, plasma GL-3 values represent the value that was obtained most closely to the most recent IgG test (within ± 6 months). For Seroconverted, then Tolerized patients, plasma GL-3 values represent the most recent value that was obtained after tolerization. ⁎ p b .0001 by t-test, compared to seronegative males. † p = 0.029 by t-test, compared to tolerized males.

446

W.R. Wilcox et al. / Molecular Genetics and Metabolism 105 (2012) 443–449

Table 2 Relationship between seroconversion status and genotype for the 20 most common genotypes in the Fabry Registry. Males Genotype and category Nonsense R227X R220X R301X R342X Missense R227Q N215S R342Q R112C G85N P40S L415P R301Q A143P I317T A156T N34S P259R R112H Frameshift c.718_719 deletion c.1235_1236 deletion

Females

Seronegative n, (%) Seroconverted, but Seroconverted, then Total, N Seronegative n, (%) Seroconverted, but Seroconverted, then Total, N not tolerized n, (%) tolerized n, (%) not tolerized n, (%) tolerized n, (%)

5 (100%) 5 (83%) 1 (100%) 4 (100%) 1 (100%) 4 (100%) – 2 (100%) 5 (100%) – – – 1 (100%) 3 (100%)

– – – – – – 1 (17%) – – – – – – – – – – – –

– – – – – – – – – – – – – –

5 6 1 4 1 4 0 2 5 0 0 0 1 3

4 (100%) –

– –

– –

4 0

3 (13%) – – –

18 (78%) 9 (100%) 1 (100%) 3 (75%)

2 (9%) – – 1 (25%)

23 9 1 4

14 (88%) 3 (100%) 3 (100%) 2 (100%)

10 (67%) 3 (100%) 1 (11%) 3 (38%) 1 (100%) – – – 1 (25%) 1 (50%) 1 (100%) – 1 (100%) 1 (100%)

3 (20%) – 8 (89%) 5 (63%) – 2 (100%) 2 (100%) – 2 (50%) 1 (50%) – 3 (100%) – –

2 (13%) – – – – – – – 1 (25%) – – – – –

15 3 9 8 1 2 2 0 4 2 1 3 1 1

1 (20%) –

4 (80%) –

– –

5 0

2 (13%) – – –

16 3 3 2

Percentages are based on the total number of patients within each genotype. These 20 genotypes represent the most common genotypes in the overall population of the Fabry Registry. Seronegative patients tested negative for anti-αGAL IgG antibodies each time they were tested. Seroconverted, but not Tolerized patients had at least 1 positive anti-αGAL IgG test, but did not tolerize. Seroconverted, then Tolerized patients had 1 or more negative anti-αGAL IgG tests after previously testing positive and whose most recent test was negative.

who reported IARs, as defined in Methods. Because anti-αGAL IgG antibody titers were not measured at regular intervals, antibody titers and/ or seroconversion status at the time an IAR occurred could not always be determined. As an approximation, patients were placed in 2 general categories: those who remained seronegative at all assessments and those who ever tested positive for anti-αGAL IgG antibodies. Men who seroconverted were significantly more likely to report IARs than those who remained seronegative: 139 of 416 men (33%) who seroconverted had IARs, whereas only 12 of 155 men (8%) who remained seronegative had IARs (pb 0.001 by Chi-square test). Women were less likely to report IARs than men: 27 of 251 women reported IARs (17%), versus 151 of 571 men (26%). There was not a statistically significant difference in the percentage of seroconverted women who reported IARs (5 of 31, 16%), compared to the percentage of seronegative women who reported IARs (22 of 220, 10%) (p = 0.348 by exact Chi-square test). Overall, 9% of seronegative patients reported IARs. As shown in Fig. 3, most IARs were reported during the first 6 months of agalsidase beta treatment, with some occurring after 6 to 12 months of treatment. The number of patients reporting IARs decreased over time with continued treatment, among both seroconverted and seronegative patients. Table 3 summarizes the most frequently reported IARs, stratified by gender and by seroconversion status. Chills, nausea, pyrexia and/or increased body temperature were among the most frequently reported IARs in both seroconversion categories and in both genders. There were no obvious differences in the types of IARs reported by seroconversion status or by gender. However, relatively few females and seronegative men reported IARs. 4. Discussion The purpose of this investigation was to evaluate the development of anti-αGAL IgG antibodies in Fabry Registry men and women who were treated with agalsidase beta. Most men in this cohort (73%) developed anti-αGAL IgG antibodies at some point during the

treatment period, which is consistent with the fact that most men with Fabry disease have little or no endogenous αGAL enzyme activity [1,2]. In contrast, heterozygous women with Fabry disease usually have residual αGAL enzyme activity [1,2] and only a small percentage of women treated with agalsidase beta developed anti-αGAL IgG antibodies (12%). The women who developed an immune response had low peak titers and most tolerized. These findings are consistent with results from clinical studies of agalsidase beta, which found that most men seroconverted [9–11,14,17]. Similarly, a comparative study by Vedder et al. [15] reported that men treated with either agalsidase beta or agalsidase alfa frequently develop anti-αGAL IgG antibodies. The present investigation represents the largest analysis of anti-αGAL IgG antibodies in women with Fabry disease to date; the findings clearly show that the vast majority of women do not develop an immunological response against agalsidase beta. Patients with mutations that result in truncated, non-functional versions of the αGAL protein would be expected to develop a stronger immune response than patients with mutations that encode a full-length αGAL protein with some level of residual enzyme activity. Indeed, men with nonsense mutations were generally more likely to develop antiαGAL IgG antibodies than were men with missense mutations. However, certain missense mutations were also associated with seroconversion in men (R342Q, P40S, L415P and N34S), and it possible that these mutations may be associated with a “classical” early-onset phenotype [18,19]. Vedder et al. did not observe a relationship between the presence of nonsense versus missense mutations and seroconversion status among patients treated with agalsidase beta or agalsidase alfa, although that investigation included a substantially smaller number of patients [15]. One concern about the development of antibodies against therapeutic proteins is that such antibodies may impair the efficacy of the infused product, through various mechanisms [13]. If an appropriate biological marker is available, reductions in efficacy can be detected by monitoring such biomarkers in patients who develop an immune response. For example, in the case of alglucerase therapy for Gaucher disease, analysis

W.R. Wilcox et al. / Molecular Genetics and Metabolism 105 (2012) 443–449

447

Number of Seroconverted Patients with IARs

A. Ever Seropositive 50

40

Men Women

30

20

10

0 0

1

2

3

4

5

6

7

Years after First Agalsidase Beta Infusion

B. Always Seronegative

Number of seronegative patients with IARs

50

40

Men Women

30

20

10

0 0

1

2

3

4

5

6

7

Years after First Agalsidase Beta Infusion Fig. 3. Most IARs were reported by during the first 12 months of Fabrazyme treatment. Panel A shows the number of patients who were ever seropositive reporting IARs during each 3-month interval over the course of agalsidase beta treatment. Panel B shows the number of seronegative patients reporting IARs during each 3-month interval over the course of agalsidase beta treatment.

of hematologic and organ volume data from clinical trials [20] and the Gaucher Registry [21] demonstrated that the presence of antiglucocerebrosidase IgG antibodies did not correlate with a reduction in the clinical efficacy of alglucerase therapy in the vast majority of patients. For Fabry disease, it is more difficult to detect and quantify possible effects of treatment-induced antibodies on clinical outcome, given the lack of meaningful biomarkers and the slowly progressive

and highly variable natural course of the disease. However, there is some evidence that the formation of anti-αGAL IgG antibodies may have negative consequences, particularly when agalsidase beta is administered at doses lower than 1 mg/kg/2 weeks [17]. Anti-αGAL IgG antibodies have been associated with increased accumulation of GL-3 in skin [14,17] and elevated urinary excretion of GL-3 [15,17,22]. A prospective, open-label clinical trial evaluated whether transitioning from

448

W.R. Wilcox et al. / Molecular Genetics and Metabolism 105 (2012) 443–449

Table 3 Summary of most frequently reported infusion-associated reactions. Seroconverted males (ever seropositive, n = 416)

Seronegative males (always seronegative, n = 155)

Adverse event (preferred term)

Number of patients with IARs (n = 139)

Number of events (n = 753)

Adverse event (preferred term)

Number of patients with IARs (n = 12)

Number of events (n = 40)

Chills Nausea Vomiting Pyrexia Dyspnoea Body temperature increased Chest discomfort Flushing Feeling cold Blood pressure increased Headache Paraesthesia Pruritis Urticaria

78 25 25 25 24 20 18 17 16 16 15 15 15 12

119 37 35 31 31 22 23 23 21 19 21 18 19 18

Nausea Urticaria Chest discomfort Chills Pyrexia Paraesthesia Pruritis

3 3 2 2 2 2 2

3 5 2 2 2 2 2

Seroconverted females (ever seropositive, n = 31)

Seronegative females (always seronegative, n = 220)

Adverse event (preferred term)

Number of patients with IARs (n = 5)

Number of events (n = 38)

Adverse event (preferred term)

Number of patients with IARs (n = 22)

Number of events (n = 98)

Chills Body temperature increased Nausea Headache

3 2 2 2

3 2 2 2

Chills Headache Fatigue Nausea Vomiting Pyrexia Pruritus Urticaria Abdominal discomfort Pain in extremity Dyspnoea

5 5 4 4 4 3 3 3 2 2 2

9 6 6 5 4 5 7 5 3 3 2

Preferred terms are from the Medical Dictionary for Regulatory Activities Preferred Terms (version 11.1). Events that were reported by 12 or more patients are summarized for seroconverted men. For seronegative men and for women, events that were reported by 2 or more patients are summarized. Patients may have had more than 1 type of event.

the standard dose of agalsidase beta (1 mg/kg/2 weeks) to a lower dose (0.3 mg/kg/2 weeks) affected clearance of GL-3 from renal capillary endothelial cells and other cell types [17]. Lubanda et al. found that while a lower dose of agalsidase beta may be sufficient to maintain GL-3 clearance in patients who remained seronegative or developed only low anti-αGAL IgG antibody titers, some patients with higher titers did not maintain GL-3 clearance when transitioned to a lowered dose [17]. Similarly, a more robust reduction in urinary GL-3 levels was observed when agalsidase beta was administered at 1.0 mg/kg/2 weeks compared to a lower dose of 0.2 mg/kg/2 weeks [15]. However, the clinical significance of GL-3 in skin or urine is not known. Plasma GL-3 levels are reported in the present analysis because this parameter has been most frequently reported to the Fabry Registry. Although mean plasma GL-3 levels were slightly higher in men who seroconverted compared to those who remained seronegative, mean plasma GL-3 levels were within the normal range in both groups. A validated biomarker that can be routinely evaluated is needed to further investigate possible effects of anti-αGAL IgG antibodies on agalsidase beta treatment. Globotriaosylsphingosine is an interesting candidate biomarker [23,24]; however, further study is needed to establish whether it can be used to monitor Fabry disease progression and ERT outcomes. A second concern about patients who develop an immune response to ERT is the impact of anti-αGAL IgG antibodies on drug safety. Safety data are not reported to the Fabry Registry, but are collected in the Genzyme GPS&RM database. Among patients in this cohort who reported adverse events to Genzyme GPS&RM, men were more likely to experience IARs than women, and men who seroconverted were more likely to report IARs than those who remained seronegative (33% versus 8%). Although relatively few women seroconverted, a slightly higher percentage of these women reported IARs, compared to women who remained seronegative (16% versus 10%). In all patients,

most IARs occurred during the first 12 months of treatment, and decreased over time with continued administration of agalsidase beta. The most frequently reported IARs included chills, nausea, vomiting, pyrexia, and dyspnoea, which are consistent with what has been observed in clinical trials of agalsidase beta [25]. There were no obvious differences in the types of IARs reported between seroconversion categories or between genders. Overall, 9% percent of seronegative patients reported IARs. The basis for the occurrence of IARs in seronegative patients is not clear. However, similar findings have been reported for patients with Gaucher disease who experienced IARs after infusion of alglucerase, but who had not developed IgG antibodies against alglucerase [21]. In those cases, patients were found to be sensitive to compounds in the drug formulation, such as the buffer solution, and did not develop an antibody response to alglucerase itself. It is not known whether a similar mechanism may cause IARs in Fabry Registry patients who are seronegative. There are important limitations to consider when interpreting data from observational registries. Unlike subjects in a clinical trial, many Fabry Registry patients did not have anti-αGAL IgG titers regularly assessed at 3-month intervals. Thus, the data reported here may not accurately reflect patients' current seroconversion status or anti-αGAL IgG titer levels. Similarly, adverse events are spontaneously reported to GPS&RM in the post-marketing setting; it cannot be ascertained whether other Fabry Registry patients experienced unreported adverse events. In addition, the timing of adverse events usually cannot be accurately correlated with the timing of seroconversion or with antibody titers, because many patients did not have anti-αGAL IgG titer data reported at or near the time of such events. The development of anti-αGAL IgG antibodies has been reported in patients treated with agalsidase alfa [15,26,27], as well as in those treated with agalsidase beta. Because the manufacturers of these drugs have

W.R. Wilcox et al. / Molecular Genetics and Metabolism 105 (2012) 443–449

used different assays to measure anti-αGAL IgG antibodies, it has not been possible for physicians to directly compare seroconversion rates and antibody titer levels between these different ERT products. Currently, there is no standardized anti-αGAL IgG antibody assay that allows the treatment community to interpret antibody data from different labs equally. Given the potential impact of these antibodies, it is vital to have such harmonization in the public domain, to facilitate comparisons of different treatment strategies and the impact on long-term outcomes. In summary, most women treated with agalsidase beta do not develop an immune response, whereas most men do. Men who developed anti-αGAL IgG antibodies were more likely to experience IARs than those who remained seronegative, and most IARs occurred during the first 6 to 12 months of agalsidase beta treatment and decreased over time with continued administration of ERT. Disclosures Genzyme sponsors the Fabry Registry. WRW, DGW, GEL, DPG, and UF-R are members of the Fabry Registry Board of Advisors. Authors who have received research funds or travel support from Genzyme include WRW, DPG, GEL, UF-R, and DGW. Authors who have received speaking fees from Genzyme include DPG and WRW. WK, SMR, JAC, MC, and DBJ are Genzyme employees. All honoraria received by GEL are donated to the Gaucher Stichting, a national foundation that supports research in the field of lysosomal storage disorders. WRW has served as a paid consultant to Amicus and Shire. DGW has served as a paid consultant to Genzyme and also has consultancies and has received travel funds from Abbot, Amgen, Amicus, Gilead, Parion, Relypsa, and Shire. Acknowledgments The authors would like to thank the many patients who have agreed to participate in the Fabry Registry as well as the physicians and research coordinators that have entered clinical data on these patients. We also acknowledge our colleagues at Genzyme (Cambridge, MA): Rob Pomponio, for genotype analysis; Tim Foley, for analysis of plasma GL-3, Karen Welch, for analysis of anti-αGAL IgG antibody titers, and Badari Gudivada for statistical programming support. References [1] R.J. Desnick, Y.A. Ioannou, C.M. Eng, Alpha-Galactosidase A deficiency: Fabry disease, in: C. Scriver, A. Beaudet, W. Sly, D. Valle (Eds.), The Metabolic Bases of Inherited Disease, McGraw-Hill, New York, 2001, pp. 3733–3774. [2] D.P. Germain, Fabry disease, Orphanet J. Rare Dis. 5 (2010) 30. [3] M.H. Branton, R. Schiffmann, S.G. Sabnis, G.J. Murray, J.M. Quirk, G. Altarescu, L. Goldfarb, R.O. Brady, J.E. Balow, H.A. Austin III, J.B. Kopp, Natural history of Fabry renal disease: influence of alpha-galactosidase A activity and genetic mutations on clinical course, Medicine (Baltimore) 81 (2002) 122–138. [4] A. Ortiz, B. Cianciaruso, M. Cizmarik, D.P. Germain, R. Mignani, J.P. Oliveira, J. Villalobos, B. Vujkovac, S. Waldek, C. Wanner, D.G. Warnock, End-stage renal disease in patients with Fabry disease: natural history data from the Fabry Registry, Nephrol. Dial. Transplant. 25 (2010) 769–775. [5] R. Schiffmann, D.G. Warnock, M. Banikazemi, J. Bultas, G.E. Linthorst, S. Packman, S.A. Sorensen, W.R. Wilcox, R.J. Desnick, Fabry disease: progression of nephropathy, and prevalence of cardiac and cerebrovascular events before enzyme replacement therapy, Nephrol. Dial. Transplant. 24 (2009) 2102–2111. [6] K. Sims, J. Politei, M. Banikazemi, P. Lee, Stroke in Fabry disease frequently occurs before diagnosis and in the absence of other clinical events: natural history data from the Fabry Registry, Stroke 40 (2009) 788–794.

449

[7] C. Wanner, J.P. Oliveira, A. Ortiz, M. Mauer, D.P. Germain, G.E. Linthorst, A.L. Serra, L. Marodi, R. Mignani, B. Cianciaruso, B. Vujkovac, R. Lemay, D. Beitner-Johnson, S. Waldek, D.G. Warnock, Prognostic indicators of renal disease progression in adults with Fabry disease: natural history data from the Fabry Registry, Clin. J. Am. Soc. Nephrol. 5 (2010) 2220–2228. [8] W.R. Wilcox, J.P. Oliveira, R.J. Hopkin, A. Ortiz, M. Banikazemi, U. Feldt-Rasmussen, K. Sims, S. Waldek, G.M. Pastores, P. Lee, C.M. Eng, L. Marodi, K.E. Stanford, F. Breunig, C. Wanner, D.G. Warnock, R.M. Lemay, D.P. Germain, Females with Fabry disease frequently have major organ involvement: lessons from the Fabry Registry, Mol. Genet. Metab. 93 (2008) 112–128. [9] C.M. Eng, N. Guffon, W.R. Wilcox, D.P. Germain, P. Lee, S. Waldek, L. Caplan, G.E. Linthorst, R.J. Desnick, Safety and efficacy of recombinant human alphagalactosidase A—replacement therapy in Fabry's disease, N. Engl. J. Med. 345 (2001) 9–16. [10] D. Germain, S. Waldek, M. Banikazemi, D. Bushinsky, J. Charrow, P. Lee, T. Loew, A.C. Vedder, R. Abichandani, W.R. Wilcox, N. Guffon, Sustained, long-term renal stabilization after 54 months of agalsidase beta therapy in patients with Fabry disease, J. Am. Soc. Nephrol. 18 (2007) 1547–1557. [11] M. Banikazemi, J. Bultas, S. Waldek, W.R. Wilcox, C.B. Whitley, M. McDonald, R. Finkel, S. Packman, D.G. Bichet, D.G. Warnock, R.J. Desnick, Agalsidase-beta therapy for advanced Fabry disease: a randomized trial, Ann. Intern. Med. 146 (2007) 77–86. [12] S. Porter, Human immune response to recombinant human proteins, J. Pharm. Sci. 90 (2001) 1–11. [13] J. Wang, J. Lozier, G. Johnson, S. Kirshner, D. Verthelyi, A. Pariser, E. Shores, A. Rosenberg, Neutralizing antibodies to therapeutic enzymes: considerations for testing, prevention and treatment, Nat. Biotechnol. 26 (2008) 901–908. [14] B. Benichou, S. Goyal, C. Sung, A.M. Norfleet, F. O'Brien, A retrospective analysis of the potential impact of IgG antibodies to agalsidase beta on efficacy during enzyme replacement therapy for Fabry disease, Mol. Genet. Metab. 96 (2009) 4–12. [15] A.C. Vedder, F. Breunig, W.E. Donker-Koopman, K. Mills, E. Young, B. Winchester, I.J. Ten Berge, J.E. Groener, J.M. Aerts, C. Wanner, C.E. Hollak, Treatment of Fabry disease with different dosing regimens of agalsidase: effects on antibody formation and GL-3, Mol. Genet. Metab. 94 (2008) 319–325. [16] T.P. Roddy, B.C. Nelson, C.C. Sung, S. Araghi, D. Wilkens, X.K. Zhang, J.J. Thomas, S.M. Richards, Liquid chromatography-tandem mass spectrometry quantification of globotriaosylceramide in plasma for long-term monitoring of Fabry patients treated with enzyme replacement therapy, Clin. Chem. 51 (2005) 237–240. [17] J.C. Lubanda, E. Anijalg, V. Bzduch, B.L. Thurberg, B. Benichou, A. Tylki-Szymanska, Evaluation of a low dose, after a standard therapeutic dose, of agalsidase beta during enzyme replacement therapy in patients with Fabry disease, Genet. Med. 11 (2009) 256–264. [18] C. Eng, L.A. Resnick-Silverman, D.J. Niehaus, K.H. Astrin, R.J. Desnick, Nature and frequency of mutations in the alpha-galactosidase A gene that cause Fabry disease, Am. J. Hum. Genet. 53 (1993) 1186–1197. [19] D.P. Germain, J. Shabbeer, S. Cotigny, R.J. Desnick, Fabry disease: twenty novel αgalactosidase a mutations and genotype–phenotype correlations in classical and variant phenotypes, Mol. Med. 8 (2002) 306–312. [20] S.M. Richards, T.A. Olson, J.M. McPherson, Antibody response in patients with Gaucher disease after repeated infusion with macrophage-targeted glucocerebrosidase, Blood 82 (1993) 1402–1409. [21] M. Rosenberg, W. Kingma, M.A. Fitzpatrick, S.M. Richards, Immunosurveillance of alglucerase enzyme therapy for Gaucher patients: induction of humoral tolerance in seroconverted patients after repeat administration, Blood 93 (1999) 2081–2088. [22] T. Ohashi, M. Sakuma, T. Kitagawa, K. Suzuki, N. Ishige, Y. Eto, Influence of antibody formation on reduction of globotriaosylceramide (GL-3) in urine from Fabry patients during agalsidase beta therapy, Mol. Genet. Metab. 92 (2007) 271–273. [23] J.M. Aerts, W.W. Kallemeijn, W. Wegdam, M. Joao Ferraz, M.J. van Breemen, N. Dekker, G. Kramer, B.J. Poorthuis, J.E. Groener, J. Cox-Brinkman, S.M. Rombach, C.E. Hollak, G.E. Linthorst, M.D. Witte, H. Gold, G.A. van der Marel, H.S. Overkleeft, R.G. Boot, Biomarkers in the diagnosis of lysosomal storage disorders: proteins, lipids, and inhibodies, J. Inherit. Metab. Dis. 34 (2011) 605–619. [24] M.J. van Breemen, S.M. Rombach, N. Dekker, B.J. Poorthuis, G.E. Linthorst, A.H. Zwinderman, F. Breunig, C. Wanner, J.M. Aerts, C.E. Hollak, Reduction of elevated plasma globotriaosylsphingosine in patients with classic Fabry disease following enzyme replacement therapy, Biochim. Biophys. Acta 1812 (2011) 70–76. [25] Fabrazyme [package insert], Genzyme, Cambridge, MA, 2010. [26] R. Schiffmann, J.B. Kopp, H.A. Austin III, S. Sabnis, D.F. Moore, T. Weibel, J.E. Balow, R.O. Brady, Enzyme replacement therapy in Fabry disease: a randomized controlled trial, JAMA 285 (2001) 2743–2749. [27] G.E. Linthorst, C.E. Hollak, W.E. Donker-Koopman, A. Strijland, J.M. Aerts, Enzyme therapy for Fabry disease: neutralizing antibodies toward agalsidase alpha and beta, Kidney Int. 66 (2004) 1589–1595.