Introduction
Fabry disease (FD) (OMIM#301500) is a rare X-linked lysosomal storage disorder. The disease is characterized by deficiency of the lysosomal enzyme α-galactosidase A (α-Gal A, E.C. 3.2.1.22). This results in a systemic accumulation of globotriaosylceramide (Gb3) and related glycosphingolipids in lysosomes in cells throughout the body. The prevalence of FD is estimated at 1:40.000–170.000 live births [
1‐
3] although recent newborn and high-risk group screening studies suggested that the prevalence of non-classical FD may be much higher than previously thought [
4,
5]. Phenotypically, FD can be distinguished in the more severe classical form of FD, predominantly affecting males, and a non-classical form, more prominent in males with residual enzyme activity. Although females can be as severely affected as male patients with classical FD, most of them have a more variable and attenuated phenotype and are therefore better characterised as non-classical patients [
6].
Early symptoms in classically affected male and female patients include angiokeratoma, anhydrosis, neuropathic pain, gastrointestinal symptoms and microalbuminuria. Later in life, progressive renal failure, heart failure and stroke generally occur. In non-classically affected male patients and most females, the disease presents with a more attenuated and variable disease course [
7‐
11]. The shortened life expectancy and the morbidity of Fabry patients are strongly related to the degree of end-organ damage.
Currently, two enzyme preparations are available for the treatment of FD (agalsidase alfa, Shire HGT, Boston MA, USA, and agalsidase beta, Genzyme Inc, Boston MA, USA). The initial clinical trials showed beneficial effects on neuropathic pain, cardiac mass and kidney function. However, it has been shown that despite enzyme replacement therapy (ERT), disease complications may still occur [
12‐
14].
Patients who suffer from FD have a lower quality of life (QoL) compared to healthy individuals. Neuropathic pain and anhidrosis are predictors of decreased QoL, presumably as a marker of more severe disease [
3]. It has been postulated that ERT has a positive effect on QoL [
15,
16]. However, these studies used different measures of QoL and were only reported for small cohorts of patients. Interest in QoL measurements has increased over the past decades, because it is well recognized that, in addition to physical disabilities, emotional and psychological factors play an important role in the lives of patients with FD. Additionally, patient involvement with decision making and assessment of quality of care is increasing. Lastly, QoL measurements are needed for cost-effectiveness analyses, nowadays a requirement for reimbursement of therapy for some governments in the EU [
17]. It is therefore important to gain a good understanding of the information available to us now.
This systematic review provides an overview of the current literature with the aim to improve our understanding of the QoL amongst patients with FD and to enhance the appropriate use of QoL instruments in clinical practice. We specifically focus on which QoL measures have been used to determine if these different measures reveal similar results. Furthermore we review the literature on the potential effect of ERT on QoL.
Methods
Search strategy and study selection
The following electronic databases have been searched via OvidSP: Medline (1946 till December 10, 2014), Embase (1947 till December 10, 2014) and PsycInfo (1806 till December week 1, 2014). The Cochrane Central Register of Controlled Trials (CENTRAL, accessed December 10, 2014) has been searched as well.
The search terms used were: Fabry disease, quality of life, questionnaires, SF-36, EQ5D, pain measurement, BPI, peds QL, and their synonyms, Mesh terms (Medline) and headings (Embase). No limits were used. Detailed search strategies can be found in Additional file
1.
The title and abstract of all articles obtained by the search were screened to identify studies where quality of life in patients with FD was studied. Reference lists of identified papers were hand searched for additional relevant citations. Original articles published in English, French and German were included. Case reports, case series on less than 5 patients, and review articles were excluded.
Data were recorded on the type of study (clinical trial, cohort study, before-after study, case series or registry study), number of subjects, gender and age groups (children and/or adults), together with the type of questionnaire used to assess QoL, disease severity and therapy status at the time of QoL assessment.
Statistical analyses
A meta-analysis was performed on studies reporting SF-36 or RAND-36 results using a fixed effect inverse variance weighting. Meta-analysis of other QoL measurements was not feasible because data were either not given in sufficient detail or QoL instruments were only used in single studies. Articles were included in the meta-analysis when mean domain scores with standard deviations or confidence intervals were provided. Pooled analysis for all studies combined, as well as for subgroups of studies, were performed. Subgroups were defined as: (1) studies performed in the period before ERT was available (untreated, mostly classically affected patients), (2) studies on the effect of ERT that report baseline measurements (untreated patients but with a treatment indication) and (3) studies in which only ERT treated patients were included. Results from the Bodily Pain and General Health subdomains from the RAND-36 were excluded because different scoring algorithm are used for these subdomains.
Discussion
This systematic review of quality of life in Fabry patients from 54 articles and abstracts has led to two major findings. Firstly, a consistent finding from all studies is that Fabry patients suffer from a considerably worse quality of life as compared with the general population. This was found for all domains in the SF-36 and in the EQ-5D questionnaires. Secondly, the studies on the effect of ERT on QoL are inconclusive.
Both the SF-36 and EQ-5D revealed that patients with FD clearly have lower QoL scores in comparison with the healthy population. However, in the interpretation of these results, some of the study characteristics need to be taken into consideration; firstly, disease severity is rarely comprehensively reported and if reported, the data varies between studies. Rombach et al. showed that disease severity plays an important role for measuring QoL and should therefore be taken into account when measuring QoL scores [
59]. This is further supported by the finding of a correlation between the MSSI (a measure for disease severity) and QoL [
39,
55]. In addition, more severe kidney disease has been shown to lead to reduced QoL, in particular after initiation of renal replacement therapy. [
58]. Also, no studies on the difference in QoL between patients with either classical or non-classical FD have been performed, although this would have been interesting considering their different disease courses. The influence of phenotype is illustrated by Gold et al., who measured SF-36 scores for untreated male patients before the introduction of ERT. In these severely, mostly classically affected males, domain scores ranged between 24 and 61, which is worse compared to QoL scores found after the introduction of ERT, even if only baseline scores (prior to start of ERT) are considered. This difference can be partly explained by the inclusion of non-classical patients in the more recent studies. In addition, most studies investigated a Fabry population consisting of both males and females. As noted by Wilcox et al. differences are observed between males and females and at what age the quality of life starts to decline for either gender [
9]. Both are factors that need to be taken into consideration. Thirdly, many of the cohorts studied consisted of treated and untreated patients together. More studies in subgroups of patients are needed to gain a better insight into the influence of phenotype, gender and treatment on QoL.
If specifically looking at the effect of ERT on QoL, only a limited number of studies reported baseline and follow-up data in detail, showing different results. One study of only women, with a small sample size and without a control arm discovered a very minor change after 27 weeks. Another study reported a minimal improvement in BPI interference score after 24 weeks, although it should be noted that baseline BPI scores were different between both treatment arms and decreased in both groups [
15]. One might argue that 6 months is too short to detect any effect from ERT on QoL scores. However, a third study showed no change in subdomains of the SF-36 after 4 and 7 years of therapy except for Social Functioning, which worsened after 7 years of therapy [
61]. Another explanation could be that the questionnaires are not sensitive enough to show a clear effect in patients with FD. Baumstarck et al. demonstrated that generic questionnaires often are more suitable for universal applications where QoL is compared in different populations, while disease specific instruments focus on particular health problems and are more sensitive for detecting and quantifying small changes [
75,
76]. This would suggest that a Fabry specific QoL questionnaire would provide a more sensitive tool to investigate the effects of ERT on the QoL. At this point no validated FD specific QoL questionnaire exists and it would be worthwhile to develop such a questionnaire for this patient group. The studies based on data from the Fabry Registry or Fabry Outcome Survey all showed an improvement [
16,
53,
54,
67]. Despite being large, these registries have their shortcomings as has been published by Hollak et al. [
77]. Follow-up data on QoL of only a very small percentage of patients enrolled in these registries were available for the analysis making the results susceptible to selection bias. Furthermore, the limited information on genotype, phenotype and disease severity of the patients in these registry studies makes comparison between cohorts impossible. Finally, differences before and after therapy are small, especially when comparing to the 3–5 point (SF-36) or 0.074 (EQ-5D) minimally clinically important difference (MCID). However, whether the standard MCID’s are applicable to FD can be questioned. For example, Wyrwich et al. demonstrated that three different expert panels provided three different Clinically Important Differences for three different diseases; chronic obstructive pulmonary disease (COPD), asthma, and heart disease, respectively [
78]. This would imply that a FD specific MCID needs to be defined for an optimal interpretation of the results. Altogether, no clear answer can be given whether ERT has a positive or negative effect on standard QoL scores in patients with FD.
Apart from the need for more sensitive questionnaires and disease specific MCID’s, another important caveat is the lack of QoL data from untreated patients with similar disease severity as those treated with ERT. Only two short-term placebo-controlled trials of ERT in FD with QoL as a secondary outcome measure have been performed. One of those studies revealed significant improvement in both placebo and treated arms [
62], while the other showed a small improvement in BPI interference score compared to the placebo group [
15]. The 3 studies published on the effects of the shortage did provide us with an opportunity to see how ERT affected the QoL once the preparation or dose was changed. However, these studies were of relatively short duration and no clear conclusion can be drawn from the results. Two studies reported stable QoL scores, while one study established a decline in two subdomains of the SF-36, only in female patients [
42]. Whether anxiety of patients due to the situation of shortage played a role is unknown as well.
Finally, several national governments currently ask for cost effectiveness analyses to aid in reimbursement decisions. SF-36 and EQ-5D utility scores play a central role in these analyses. Based on these scores, QALY’s are calculated and subsequently used to obtain costs per QALY. This development stresses the importance for the collection of high quality QoL data both before and during treatment.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
MA has no competing interests do declare. MB and CH have received research support, educational grants, travel support, honoraria for consultancies and speakers fees from Genzyme, Shire HGT, Amicus, Protalix and Actelion. All fees are donated to the Gaucher Stichting or the AMC Medical Research for research support.
Authors’ contributions
MA: study design, data acquisition, data analyses, data interpretation, first draft of manuscript. MB and CH: study design, data interpretation, revision of manuscript. All authors read and approved the final manuscript.