This study suggests that enzyme replacement therapy with either agalsidase alfa or agalsidase beta has limited effectiveness on the course of renal, cardiac and cerebral manifestations. Complications occurred in both treated and untreated patients with Fabry disease. However, increased treatment duration reduced the odds of developing a first and second complication in another organ. This means that in general, prolonged treatment delays the occurrence of complications. This is in line with a placebo-controlled trial showing that the hazard ratio for developing complications was lower in patients treated with agalsidase beta compared to placebo [
9]. While patients in that study had advanced Fabry disease, our study indicates that complications can also not be prevented in milder affected patients.
Of interest is whether males and females respond differently to ERT. In our study females had a stable renal function resembling the healthy population [
29]. Also, LVmass remained stable in females, while LVmass increased in males. This is in line with previous studies, showing a more favourable course in females [
30,
31]. Kidney failure influences cardiac outcome as CKD contributes to increase of LVmass, independent of blood pressure [
32,
33]. Here we made a similar observation: progression of LVmass was primarily seen in males with advanced renal disease. Presence of cardiac fibrosis may explain unresponsiveness to ERT [
13,
34,
35]. In our cohort, insufficient cardiac (MRI) data were available for this analysis. Antibody formation in males is associated with a less robust decline of plasma lysoGb3, Gb3 and urinary Gb3, which may reflect worse treatment outcome, but needs to be studied in larger patient groups [
36].
Our study has several limitations. The natural history cohort consisted of patients with an indication for ERT who developed complications before ERT became available or who remained untreated, thus creating an untreated cohort with comparable disease severity. It is possible however, that disease progression in this NH cohort is underestimated, as not all complications may have been recorded as rigorously in the pre-ERT era. Alternatively, the inclusion of more atypical patients in the NH study cohort may underestimate the effectiveness of ERT. Excluding atypical patients from the analyses demonstrated a more beneficial effect of ERT duration. This illustrates that ERT is less effective in patients with a more attenuated disease course compared to patients with a classic phenotype. In contrast, the effect of ERT may have been overestimated, as patients in the ERT cohort received improved supportive care.
In this study we combined outcome data for agalsidase alfa and agalsidase beta. The controversy whether the drugs are equally effective remains unresolved. Both drugs appear to have an identical rate of complications at equal dose [
16] and another ongoing study suggests no difference in outcome at registered dose [
17]. Another important issue is the use of ACE/ARB medication that may have influenced treatment outcome. In the ERT group more patients had proteinuria, and during follow-up more patients on ERT started ACE-ARB medication. As the number of patients with proteinuria in the NH and ERT group was not equal, further stratification by ACE/ARB was not performed. Despite these limitations, the most important conclusion that can be drawn from the current study is that long term outcome of ERT is of limited effectiveness. Earlier reports frequently emphasize lack of progression or delay in progression as important benefits [
12,
37]. ERT with agalsidase alfa was reported to lead to substantial and sustained benefits. The data indicated, however, that there was a decline in renal function despite therapy and no change in left ventricular mass [
37]. Similarly, in 58 patients, mainly males, treated with agalsidase beta for 4.5 years, stabilization of renal function was reported but patients with a more rapid decline were left out of the analysis [
12]. Smaller, uncontrolled cohort studies have reported on progression of disease despite treatment in patients with more advanced renal or cardiac disease, especially in patients with decreased renal function and cardiac fibrosis. A better outcome in less affected patients has been shown [
13]. In a recent study, concerning a large cohort of Fabry disease patients from the UK, increasing duration of ERT had a small but beneficial effect on left ventricular mass and on renal function (in females) [
14]. The risk of stroke/TIA was not influenced by ERT. Interestingly quality of life scores declined over time, but no correction for disease severity was made, which makes it difficult to interpret these data. Also, the authors commented that the study was weakened by substantial amounts of missing data [
14]. If ERT cannot prevent disease progression in severely affected patients and if longer treatment duration is associated with decrease of complications, it is tempting to assume that early therapy is most effective. However, the results of early treatment in minimally affected boys, is still awaited (trial number NCT00701415,
http://www.clinicaltrials.gov). In the studied adolescents here, eGFR declined, LVmass remained stable but WMLs were not prevented. This suggests that even in patients with early disease manifestations, progression occurs. In conclusion, long term ERT combined with optimal supportive care (additional interventions and medication) does not prevent disease progression, but longer treatment duration may lower the risk of developing additional complications. The risk of developing a first or second complication declines with increasing treatment duration, but as no short term beneficial effect of ERT is expected, ERT in advanced disease seems to be of little benefit.