The French National Registry of patients with Facioscapulohumeral muscular dystrophy

With an estimated prevalence between 1/20,000 and 1/8,000 [1, 2], autosomal dominant Facioscapulohumeral muscular dystrophy (FSHD) is globally the third most common inherited myopathy, while, in adult, it is the second most frequent myopathy after the Steinert Myotonic Dystrophy (DM1).

FSHD is characterized by an important clinical variability including age of onset, disease progression, and presence or absence of multisystemic involvement [3]. FSHD patients are typically characterized by a progressive and asymmetric involvement of facial, shoulder girdle, anterior forelegs and abdominal muscles. Pelvic and axial muscles can also be affected. The disease progression is usually very slow and only 10–20% of patients with FSHD will become wheelchair-bound after age 50 [4]. This portion increases to more than 40% by age 18 in the early form of the disease [5]. Respiratory involvement is infrequent and, in the majority of cases, has been reported in wheelchair-bound patients or as secondary to the skeletal deformation (cyphosis, hyperlordosis) that can be seen in severe forms of the disease [6]. Cardiac involvement, manifesting as a predilection to atrial arrhythmias, is seen in about 5% of patients, few of whom require treatment [7]. Sensorineural deafness has been described in a variable percentage of FSHD patients, ranging from 25 to 65% in different studies [8‐11]. Retinal vascular involvement is rare and is frequently presented as asymptomatic retinal telangiectasia. Nevertheless, in severely affected FSHD patients, retinal abnormalities can have dramatic consequences including complete loss of vision due to retinal exudation leading to retinal detachment (Coat’s syndrome) [12‐14]. Dysphagia has been reported as affecting 10% of adult FSHD patients and has been interpreted as secondary to the involvement of orofacial muscles [15]. In childhood-onset FSHD patients, tongue atrophy, epilepsy and mental retardation have been described as an uncommon feature [16]. Other symptoms frequently reported by FSHD patients are pain and fatigue [17, 18]. Differences in expression of this disease related to gender have been described, women tending to be less symptomatic than men (late disease onset and slower progression) [19, 20]. Life expectancy is not modified except for patients with a childhood-onset FSHD.

FSHD is a genetically heterogeneous disease with two distinct forms. FSHD type 1 (FSHD1) is the most common form with more than 90% of FSHD patients, while the FSHD type 2 (FSHD2) is the rarest form of the disease, affecting less than 5% of FSHD patients. FSHD1 is caused by the contraction to 10 repeats or less of the 3.3 Kb D4Z4 repeat units on chromosome 4. In contrast, FSHD2 is caused by mutations in the chromatin modifier SMCHD1 gene located on chromosome 18. Alternatively, pathogenic mutations in the DNMT3B gene have been described in association with FSHD2 [21].

To date, in France, molecular diagnosis is available only for the FSHD1 type; indeed, clinical FSHD patients with no D4Z4 contraction might be explored in a research context, and positive results are not fully integrated in the diagnosis process yet.

With the recent interest from pharmaceutical industries to develop novel therapeutic solutions for FSHD patients, it is now mandatory to create national registries to facilitate fully characterized patients’ recruitment in clinical trials and better understand the natural history of the disease to identify key parameters to efficiently evaluate treatment benefits. Today various national registries have been or are being set up as reported in the TREAT-NMD website (http://​www.​treat-nmd.​eu). They are either disease-centric in the United-Kingdom [22], Italy [23], Czech Republic, Netherlands, or part of larger registries including multiple neuromuscular diseases in the United States [24], Canada [25], Egypt, New-Zealand, Spain and Russia. A recent ENMC international workshop was dedicated to FSHD registries to ultimately harmonize data content to allow data sharing and match-making between registries [26]. Here we report the creation of the French national FSHD registry and highlight its original design allowing a strong involvement of both patients and physicians, and its evolution since 2013.

From an international perspective, the French National Registry of patients with FSHD is one of the 13 FSHD registries set up by the end of 2016 [26]. The French registry was designed to ensure the active participation of both patients and professionals (physicians and geneticists). To this end, we use a data-collection mixed model combining self-report and clinical evaluation questionnaires, filled by patients and professionals respectively. While this approach has already been successfully adopted for other Neuromuscular registries [29], such a dual approach is rare among existing FSHD registries at the international level as, to our knowledge, only the UK registry has adopted this approach [22]. Most other registries collect data via forms filled by medical practitioners, while the US registry has a self-report component, but its scope is essentially restricted to patient recruitment and expanding knowledge on the burden of the disease [30]. Besides, some FSHD registries are part of larger projects aiming at collecting data from major neuromuscular pathologies [24, 25]. In the French FSHD registry, most items in the clinical evaluation form are mirrored in the self-report questionnaire to assess data quality and identify, for each item, who is best suited to provide the answer (either the patient or the physician).

The French registry includes a strong patient-centered component. Indeed, the self-administered questionnaire has been designed by patients from the “groupe FSH”, a satellite association from the AFM. The questionnaire design was part of a study sponsored by the Nice University Hospital (10-AOI-07).

The French FSHD registry includes various tools to visualize and analyze data. For instance, it is possible to display the number of concomitant affections among patients suffering or not from a given affection as illustrated in Fig. 5 for a respiratory affection. Data from the French FSHD registry will thus allow assessing the nature and prevalence of comorbidities.

Because this registry contains sensitive data, it was approved by the appropriate French review boards (CNIL and CCTIRS) and is hosted within a secured environment. The registry is accessible through a website (http://​www.​fshd.​fr) containing the secured database itself with a freely accessible part. It contains general information on the French National Registry, the disease, the diagnosis, the associated French network centers, the state of research, reference publications and clinical trials. Links to other national registries are provided. Video and various documents are available and can be downloaded and used to join the registry. Newsletters also ensure regular information to patients about the registry evolution and advertise about forthcoming regional meetings with experts.

To help achieve a high data quality level required for a fruitful exploitation of the registry, we developed computer assisted data input. Various data consistency and data pre-filling rules have thus been defined to assist the curator (Additional file 4: Table S1, Additional file 5: Table S2).

After 5 years, the registry now contains about 21% of French FSHD patients. The proportion of registered FSHD patients is expected to be further increased through the strong support provided by the French association dedicated to Muscular Dystrophies (AFM-Telethon) and the active participation of experts from the French national network for rare neuromuscular diseases (FILNEMUS), which is one of the 23 rare disease healthcare networks built through the second French national plan on rare diseases.

To ensure interoperability at the global level, we use genetics standards coupled to clinical description. We therefore mapped these terms to ontologies, dictionaries and terminology systems to evaluate if one of the tested systems can be used for match-making [31]. We observed that no system can be used to fully describe patients clinical features: the match rates of SNOMED-CT, CTV3 and NCIt are respectively of 75, 61.7 and 53.3%. Other ontologies and dictionaries, including HPO, contain less than 50% of terms used in the database. This underlines that interoperability between databases and registries cannot be ensured by a single approach and that systems must be able to handle heterogeneous queries to ensure efficient interoperability. While the rare disease field is trying to develop the Human Phenotype Ontology (HPO) and the Orphanet Rare Disease Ontology (ORDO) to simplify interoperability, our experience demonstrates that HPO still needs improvements to allow direct mapping of used clinical descriptions. In fact, the systems that showed the highest matching rates in the mapping process are not ontologies but a clinical health terminology system (SNOMED-CT), a clinical term system (CTV3) and a reference terminology system (NCIt). Future developments made through international initiatives as RD-Connect [31], the ELIXIR [32], the IRDiRC [33] and others, should ensure the inclusion of these terms and concepts in HPO. Meanwhile, match-making and other initiatives to link rare disease resources should consider using multiple ontologies, dictionaries and terminology systems.

The French National Registry of FSHD patients is part of a national effort to develop national databases [22‐25]. These initiatives should now interact with other initiatives to build a European and/or an international FSHD virtual registry. The French registry is applying the FAIR principles as data are Findable, Accessible, Interoperable and Reusable. While individual data are protected, collaborators can have access to aggregated data after their requests have been granted by the registry oversight committee. This data sharing policy will evolve in the future with the emergence of the European Reference Networks (ERN) and the European Joint Program (EJP) on rare diseases and the associated actions to promote data standardization and interoperability.

FSHD patients will benefit of the development and improvement of the French National FSHD Registry. Thanks to collected data, the registry will facilitate and accelerate clinical and translational research on the disease. Furthermore, specific guidelines on FSHD medical care will be regularly updated and disseminated, contributing to the national standardization of medical care for FSHD patients. Moreover, thanks to the registry, the assessment of feasibility and recruitment in clinical trials and studies will be facilitated. Despite strong efforts from physicians to ensure the collection of longitudinal data, this aspect is always a weakness in patients’ registries. In order to make it more efficient, we are currently modifying the system to allow the direct data submission from patients. The curators will then validate these new information to ensure their quality.

The regular update of the website will contribute to the improvement of the knowledge of both patients and professionals on FSHD, as well as facilitate interaction with academic and industrial partners. In addition, through the website, users already have access to educational links and other relevant resources (for example FSHD meetings and congresses).