Fabry Nephropathy: Indications for Screening and Guidance for Diagnosis and Treatment by the European Renal Best Practice

Wim Terryn; Pierre Cochat; Roseline Froissart; Alberto Ortiz; Yves Pirson; Bruce Poppe; Andreas Serra; Wim Van Biesen; Raymond Vanholder; Christoph Wanner


Nephrol Dial Transplant. 2013;28(3):505-517. 

In This Article

Treatment of Fabry Nephropathy

4.1 We do not recommend starting ERT in patients with proteinuria [protein-to-creatinine ratio >1 g/g (>0.1 = gram/mol) creatinine] or eGFR <60 mL/min/1.73 m2, except for non-renal indications. (1D)

4.2 We recommend that when ERT is deemed indicated, it should be started as part of a well-designed clinical trial, either observational or interventional. (Ungraded statement)

4.3 In a patient on haemodialysis, and when ERT is deemed indicated, we recommend administering the ERT during a haemodialysis session. (1A)

4.4 We recommend kidney transplantation as a valuable option in patients who are eligible for this intervention. (Ungraded statement)

4.5 After renal transplantation, we do not suggest ERT for renal indications, but it can be continued for non-renal indications. (Ungraded statement)

As discussed above, proteinuria is an important risk factor for the progression of renal FD. The use of ACE-i and ARB has been shown to be nephro-protective in other proteinuric renal diseases, and could thus be important in FD as well. As such, the use of ACE-i or sartane would be acceptable in FD. In a recent paper,[47] it has been demonstrated that ERT interacts with ACE and inhibits its activity, possibly by removing the galactose residues from the enzyme. The clinical relevance of this observation is unclear, and should not be seen as a reason to prohibit the use of ACE-i.

Kidney Disease Improving Global Outcomes (KDIGO) guidelines suggest that in patients with CKD Stages 3–5, vitamin D deficiency be corrected.[48] Emerging evidence in patients with CKD show that vitamin D can reduce proteinuria or albuminuria even in the presence of angiotensin-converting enzyme inhibition.[49] Selective activation of the vitamin D receptor with paricalcitol lowered urinary albumin excretion, as was demonstrated in patients with Type 2 diabetes in a recent randomized controlled trial.[50] In cultured human podocytes, vitamin D receptor activation prevented lyso-Gb-3-induced, TGFβ1-mediated, up-regulation of extracellular matrix proteins.[51] Even lacking more definitive evidence of a beneficial effect of vitamin D on Fabry nephropathy, it seems advisable to place particular emphasis in following guidelines on vitamin D management in CKD patients in patients with FD.

Two forms of recombinant α-Gal A have been approved in Europe: agalsidase alpha (Replagal®; Shire Human Genetic Therapies, Boston, MA) and agalsidase beta (Fabrazyme®; Genzyme, Cambridge, MA). Agalsidase alpha is produced in a continuous human cell line and is administered as an intravenous infusion over 40 min at a dose of 0.2 mg/kg body weight every 2 weeks. Agalsidase beta is produced in Chinese hamster ovary (CHO) cells and is given as an intravenous infusion over a 4-h period at a dose of 1.0 mg/kg body weight every 2 weeks.

According to a recent Cochrane review, the evidence base in favour of ERT is weak. Only five (total n = 187) poor quality randomized controlled trials are available. They all concern surrogate end points, such as decrease in plasma Gb-3 levels in plasma and tissues and evolution of renal function. According to the Cochrane review, these studies show no evidence for a clinical benefit of the use of agalsidase alpha or beta to treat Fabry nephropathy.[52] As there are at present no hard data that ERT alters the natural course of Fabry nephropathy (Table 2), we recommend starting ERT only in the context of a clinical trial, interventional or observational. All data from observational trials should be entered in a central registry.

Besides randomized controlled trials open-label studies and retrospective analyses have been performed. It is of interest to compare the evolution of renal disease in the historical untreated and treated cohorts of an international industry sponsored registry on FD.[43,53] It is difficult to compare the data presented in both publications, as the design of the analyses and the presentation of data were different, and there was a substantial risk for selection bias, as only a minor proportion of all those enrolled could be evaluated because of missing data. Nevertheless, in both studies, patients were stratified into quartiles according to severity indices of renal involvement. The slope of change in GFR was similar in comparable quartiles of the treated and untreated cohorts, especially in men. Hence, one cannot deny the reflection that ERT might have no marked impact on the decline of kidney function. From this comparison, it is also clear that, irrespective of ERT, proteinuria was the strongest predictor of outcome. In patients without proteinuria, renal function remained stable, equally in males as in females. In those with proteinuria, the slope of deterioration of eGFR appeared to be similar with or without ERT. It is unclear what the implications of these observations are with regard to ERT: either it implies that ERT should be given before proteinuria develops (but these subjects have no deterioration of kidney function anyway) or that it should not be given for renal protection in those with already existing heavy proteinuria. It would be interesting to include complete data sets in a registry of patients developing proteinuria at early stages to see how the evolution of renal function is in this cohort. Remarkably, in the Fabry Registry, data on proteinuria were available in only 462 of 2850 (historical cohort) and 213 of 2887 (ERT cohort) patients.[43]

Other observational studies in male FD patients showed that renal function remained stable under ERT during a follow-up period up to 54 months in the case of normal or near normal baseline function (CKD 1–2) and low proteinuria (<1 g/g creatinine) in the majority of patients.[54] However, as only treated patients were observed, it cannot be excluded that these patients would have had no progression even without therapy, as it is clear from registry data that proteinuria <0.3 g/g creatinine is a favourable prognostic marker. Other publications demonstrate that in FD patients with CKD Stage 4, or with glomerulosclerosis >50% or proteinuria >1 g/g creatinine, renal function continues to deteriorate despite ERT (decline in renal function varying from 6.4 to 8.9 mL/min/1.73 m2/year.[54,55] In the case of CKD Stage 3, the decline in eGFR seems to be attenuated by ERT in comparison with historical data [−3.0 (male) and −1.0 versus −6.8 mL/min/1.73 m2/year).[56] Again, these data are small-scaled and use historic data as controls.

Few studies report on the effect of ERT on renal function in females. In a recent retrospective study of the Fabry Outcome Survey (FOS), the rate of decline in eGFR in females under ERT was similar to the normal expected age-related rate over a 4-year follow-up period, whereas the rate in men was approximately double the expected age-related rate of decline.[57] Another study reported on a stable renal function in female patients treated with ERT.[58]

In summary, these studies suggest that, for the renal aspect of FD, treatment is at best only effective in CKD Stage 1 or 2, before the deterioration of renal function or onset of overt proteinuria, as it does not reduce proteinuria per se. Once proteinuria (>1 g/day) or CKD Stage 3 (eGFR <60 mL/min/1.73 m2) develops, there are no data supporting a potential protective effect of ERT. Taking this and the very high cost (>200 000 Euro/year) into account, we do not recommend treatment in these cases.

ERT has few side effects, except for mild infusion-related reactions consisting primarily of chills that can be treated with paracetamol, antihistamines or steroids. It has been shown that the infusions can be safely performed in a home setting.[59,60]

The administration of ERT leads to the formation of antibodies in the majority of patients, and this is for both brands. These antibodies, especially the IgG, have inhibitory effects on the enzyme activity in vitro.[5,6,61,62] Although both agalsidase alpha and agalsidase beta have been associated with IgG formation, the reported incidence of antibodies has generally been higher for agalsidase beta.[62] In a study in 134 males and females, there was no correlation between anti α-Gal A IgG titres and the onset of clinical events or the rate in change in estimated GFR during treatment. However, a statistically significant association was found between anti-α-Gal A IgG titers and Gb-3 deposition in the dermal capillary endothelial cells during treatment, suggesting that Gb-3 clearance could be impaired.[63] In another study, there was less normalization of urinary Gb-3 in the seropositive patients compared with the seronegative ones.[64,65]

Analysing the consequences of antibodies is challenging because the assays are not uniform and there are no international antibody standards. Currently, numerous laboratories are performing α-Gal A-antibody testing. Potential differences between antibody assays and their respective sensitivities make comparison of titre values across the Fabry community difficult. The objective of the Fabry Antibody Standardization Initiative is to identify differences in analytical methods and to standardize α-Gal A antibody assays across the industry to allow the medical community involved in treatment to interpret antibody data equally.[66]

We have very few data on the efficiency of higher doses than the ones registered for agalsidase alpha (0.2 mg/kg EOW) and agalsidase beta (1 mg/kg EOW). One open-label trial studied 11 adult male patients with FD who demonstrated a continuing decline in renal function despite 2–4 years of conventionally dosed agalsidase alpha therapy (0.2 mg/kg EOW).[67] After switching to weekly dosing, three patients demonstrated an improvement in eGFR and six patients demonstrated a slow down in the rate of eGFR decline. Two patients failed to improve their eGFR slope. A multiple regression model confirmed that the weekly infusion regimen was the strongest explanatory variable for the change in eGFR, with a weaker contribution from the concomitant use of angiotensin-converting enzyme inhibitors/ARB, but the patient number was too low to allow meaningful conclusions.

We also have very few data comparing the two formulas. In a study by Vedder et al.,[65] the low number of patients and the dose of agalsidase beta that was used (0.2 mg/kg instead of the licensed 1.0 mg/kg) precluded firm conclusions. In a larger group of patients (n = 146), there was no difference in a composite outcome of renal, cardiac and neurological events after 30 months of treatment (West, Molecular Genetics and Metabolism, 2011, abstract).

Tahir et al. found stabilization of renal function in a small open-label observational study in patients with CKD Stage 1–2 (n = 4) and CKD Stage 3–4 (n = 6) treated with a combination of agalsidase beta 1 mg/kg EOW and ACEi or ARB. The surprisingly favourable response in patients with GFR <60 mL/1.73 m2/min and proteinuria >1 g/day was unexpected and should be confirmed in a larger study.[68] It is unclear in how far the positive effect, when confirmed, should be attributed to the ACE-i or the ERT. There is an on-going open-label, prospective, multi-centre study [The Fabrazyme® and ARB's and ACE Inhibitor Treatment (FAACET) Study, registered at ClinicalTrials.gov NCT00446862], with as primary hypothesis that titration of ACEi and ARBs to reduce urine protein excretion to <500 mg/day in Fabry patients receiving agalsidase beta therapy at 1 mg/kg every 2 weeks will slow the progression rate of decline of GFR compared with case–controls drawn from a Genzyme-sponsored Phase III extension study (GFR 60–125 mL/min/1.73 m2, urine protein >1 g/day) or the Phase IV study (GFR 20 to 60 mL/min/1.73 m2, urine protein >0.5 g/day).

Survival of Fabry patients on RRT is poor, with a reported 3-year survival of 60–63%, which is lower than that of non-diabetic-matched controls.[69] There is no proof of an improved survival in RRT patients on ERT.

In patients with CKD Stage 5, where ERT is deemed to be an appropriate option, ERT can be performed during the haemodialysis sessions, which do not alter pharmacokinetics.[70]

ERT diminished extra renal symptoms, and improved quality of life and in CKD Stage 5 patients on dialysis in a small (n = 9), non-placebo controlled cross-sectional study.[71] In another observational cross-sectional study (n = 16) on dialysis patients, with a mean follow-up of 45 months of ERT, mortality was very high (7/11), when patients were not transplanted.[72] These limited data suggest that, although typical Fabry symptoms such as pain crises can be controlled with ERT, we have no proof of improvement of cerebrovascular or cardiac morbidity or mortality in CKD Stage 5. Instead, mortality remains high if these patients are not transplanted. Transplantation without ERT has shown acceptable results. In a retrospective study, patient and graft survival was good for the first 10 years, although this study was probably undertaken in a selected patient group with little co-morbidity. After 10 years, mortality increases very quickly, probably due to progression of FD.[73] Data from the organ procurement Transplant Network/United Network for Organ Sharing (n = 197) were compared with a matched cohort of non-Fabry and non-diabetic CKD Stage 5 patients; although 5-year graft survival was similar, Fabry patients had a higher risk of death [RR 2.15 (1.52–3.02)].[74] All these data seem to indicate that transplantation can be successful in patients with Fabry nephropathy, and that transplanted patients have a stable kidney function without ERT.