Introduction

Facioscapulohumeral dystrophy (FSHD) is a genetically acquired condition that is characterised by gradually progressive asymmetrical muscle weakness of the face, scapular region, upper limbs (humeral) and distal lower limbs (peroneal) [1]. It is the third most common adult-onset muscular dystrophy, and the estimated prevalence of FSHD is approximately 4 to 12 cases per 100,000 individuals [2,3,4]. At present, there is no disease-modifying treatment for FSHD. However, ongoing research in epigenetics has led to a deeper understanding of the underlying pathogenesis of FSHD, spurring the identification of potential therapeutic targets. In this article, we discuss the pathogenesis, clinical features and diagnosis of FSHD and review current management strategies as well as potential therapeutics for patients with FSHD.

Pathogenesis

FSHD can be classified into 2 subtypes: FSHD1 and FSHD2. FSHD1 accounts for 95% of cases, whereas FSHD2 makes up the remaining 5% [2]. Both subtypes are clinically indistinguishable and arise due to inappropriate expression of the double homeobox protein 4 (DUX4) gene in the skeletal muscles [5••]. DUX4 encodes for a transcription factor that is involved in the regulation of genes for pre- and post-implantation embryogenesis [6•]. It is typically epigenetically suppressed in most somatic cells, except in the thymus and testis [7, 8]. When expressed in skeletal muscles, it can induce downstream effects like cell death, oxidative stress, inflammation and disrupted myogenesis, leading to the development of FSHD [6•, 9, 10].

The DUX4 gene lies within a macrosatellite repeat array that comprises 3.3 kb D4Z4 repeat units, in the subtelomeric region of chromosome 4 at 4q35 [8, 11, 12]. In healthy individuals, this array is made up of 11–100 D4Z4 repeat units, which are normally highly methylated and exist as euchromatin in most cells. In FSHD1, this array is contracted to 1–10 repeat units [8, 13,14,15]. Contraction of D4Z4 repeat arrays leads to hypomethylation and chromatin relaxation, facilitating inappropriate DUX4 expression [5••, 16]. Additionally, expression of DUX4 requires polyadenylation of the DUX4 transcript, which only occurs with the 4qA but not the 4qB haplotype. Hence, FSHD is manifested in individuals with D4Z4 repeat contractions of 1–10 units on the permissive 4qA haplotype.

FSHD1 is inherited via an autosomal dominant pattern [17], although 10 to 30% of FSHD1 cases exist due to sporadic occurrences, from de novo pathogenic contraction of the D4Z4 locus [18]. In FSHD1, there is an inverse correlation between the size of the D4Z4 repeat and the severity of the disease. Patients with 1–3 repeat units are most severely affected and have an earlier disease onset, compared to those with 8–10 repeat units who appear to have a milder disease which is later in onset [19, 20].

Patients with FSHD2 exhibit contraction independent, DNA hypomethylation on both copies of D4Z4, due to pathogenic variants in chromatin-modifying genes [16, 21]. The inheritance of FSHD2 is digenic, requiring the inheritance of dysfunctional chromatin-modifying genes and a moderate repeat contraction of D4Z4 repeat number between 8 and 30 on the 4qA permissive haplotype [21]. Eighty-five percent of patients with FSHD2 carry a variant in the SMCHD1 (structural maintenance of chromosomes flexible hinge domain containing 1) gene on chromosome 18 [17, 21]. SMCHD1 serves as an epigenetic repressor that binds to the D4Z4 repeat to maintain a repressed chromatin state in somatic cells via methylation, and its reduced activity in FSHD2 leads to hypomethylation of the D4Z4 array, enabling the aberrant expression of the DUX4 protein [22,23,24]. Variants in DNMT3B (de novo DNA methyltransferase gene) and LRIF1 genes similarly lead to chromatin relaxation and inappropriate DUX4 expression [15, 25, 26], manifesting as FSHD.

Clinical characteristics

Muscle weakness

FSHD is characterised by progressive muscle weakness that develops in a rostro-caudal pattern, involving the face, scapular stabilisers, upper arm, abdomen, lower leg (peroneal muscles) and hip girdle [5••]. In contrast to other dystrophies, FSHD often has asymmetric muscle involvement [27]. The disease onset varies from infancy to middle age, although most affected patients develop symptoms by the second decade [28]. The clinical progression is usually slow, and patients typically have a normal or near-normal lifespan. Disease severity is highly variable amongst individuals, and in general, patients who develop symptoms at an earlier onset have more severe disease [1, 27, 29]. In the long run, approximately 20% of the patients become wheelchair-dependent [30].

Weakness of the facial muscles, especially the orbicularis oculi and orbicularis oris, develops in the initial stages [27, 31]. This results in difficulties with closing the eyes tightly, smiling, pursing the lips and whistling [1]. In FSHD, facial weakness can be absent or mild early in the course of the disease, and may remain mild for many years [1].

Scapular winging is commonly noted early in the course of the disease. During abduction of the arms, there is characteristic upward and lateral riding of the scapula, due to preferential weakness of the lower trapezius muscles [28]. The deltoid muscles typically remain largely unaffected until the later stages of the disease. In contrast, the pectoral muscles, biceps and triceps are often affected early on, resulting in marked weakness and atrophy of the upper arm [32]. The forearm muscles are commonly spared, giving rise to the appearance of a “Popeye-arm” appearance [33]. In individuals with more severe disease, distal upper extremity weakness can be present, affecting the wrist and finger extensors as well [33].

In the abdomen, the lower abdominal muscles are selectively involved, resulting in a protuberant abdomen, exaggerated lumbar lordosis and a positive Beevor’s sign [34, 35]. A positive Beevor’s sign is characterised by upward movement of the umbilicus upon flexion of the neck in a supine position, and it occurs due to lower abdominal muscle weakness [34]. It has been extensively described in patients with FSHD and has a sensitivity and specificity of approximately 90% [36].

Lower limb weakness manifests with peroneal muscle weakness predominantly, leading to foot drop [37]. In some patients, weakness of the hip girdle muscles may be present as well.

Other systemic manifestations

Beyond skeletal muscle manifestations, FSHD can also lead to the involvement of other systems, causing respiratory dysfunction, retinal vasculopathy, hearing loss and pain.

Respiratory insufficiency in FSHD is predominantly related to weakness of the expiratory abdominal muscles, diaphragmatic dysfunction and chest wall deformities [38, 39]. In 10 to 39% of the FSHD population, a restrictive ventilatory pattern can be seen on spirometry testing [39, 40]. However, only approximately 1 to 3% of patients require respiratory support with chronic non-invasive ventilation [38, 41].

Retinal vasculopathy can occur in up to 50 to 75% of patients with FSHD, resulting in increased vascular tortuosity, telangiectatic blood vessels and microaneurysms [42]. The changes are usually bilateral and subtle and can only be demonstrated via fluorescein angiography [43]. While vision is generally unaffected in FSHD, a small percentage of patients may experience a Coats-like syndrome [43]. This syndrome occurs due to retinal telangiectasia and exudative retinopathy that can progress to retinal detachment, causing visual loss [44]. Sensorineural hearing loss may also be present in individuals with FSHD and is usually gradual and progressive [45]. The risks of hearing loss and/or exudative retinopathy are postulated to be higher in patients with larger D4Z4 repeat contraction sizes and those with early-onset disease [44, 46, 47].

Chronic pain is a significant, troubling and under-recognised symptom in patients with FSHD and has been reported to be present in up to 82% of patients [48,49,50]. It commonly affects the shoulders and lower back [51]. The pain is likely multifactorial, stemming from factors such as hyperlordosis of the lumbar spine, and muscle weakness and atrophy resulting in a restricted range of motion and discomfort [51].

FSHD does not typically result in cardiomyopathy. However, cardiac arrhythmias have been reported in patients with FSHD, though the majority of patients are asymptomatic. An incomplete right bundle branch block is most commonly described and was shown to be present in approximately 23 to 33% of patients [52], followed by supraventricular tachycardia in approximately 10% of patients [53].

Diagnosis

Genetic testing confirms the diagnosis of FSHD and should be obtained in patients with typical presentations and no first-degree relatives with genetic confirmation of the disease, or in patients with atypical presentations. First-degree relatives of a genetically confirmed proband who present with a classical FSHD phenotype may be diagnosed without further genetic testing [54].

The Southern blot method is typically used for the diagnosis of FSHD1. This procedure involves cleaving genomic DNA into specific fragments using restriction enzymes, separation of fragments by size using gel electrophoresis and, subsequently, hybridization with a p13E-11 probe [55, 56]. A reduction in fragment size of less than 10 D4Z4 repeats on the 4q35 chromosome, on a permissive 4qA allele, is consistent with a diagnosis of FSHD1.

Despite Southern blotting being the standard diagnostic tool for FSHD1, it has its limitations. It requires large amounts of high-quality molecular weight DNA, is labour-intensive and time-consuming and may require the use of radioactive material. It estimates the number of D4Z4 repeats based on the size of detected bands, which can lead to inaccuracies [57, 58]. To distinguish between 4qA, 4qB and 10q haplotypes, multiple restriction enzymes and probes are required [56, 59, 60]. Cases with somatic mosaicism or rearrangements may be undetected with standard gel electrophoresis, although this can be mitigated by using pulsed-field gel electrophoresis (PFGE) [60].

Optical genomic mapping (OGM), which maps locations of restriction enzymes in DNA molecules, is emerging as a valuable tool in genetic testing of FSHD as it addresses certain limitations associated with Southern blotting. Studies have shown that OGM can measure the number of D4Z4 repeats with higher precision, distinguish between DNA segments from 4q35 and 10q26 and accurately identify cases with mosaicism [58, 61,62,63]. In addition, it is more cost-effective and has a shorter turnaround time [62]. However, it is unable to detect rearrangements as it cannot differentiate the 4q35 and 10q26 D4Z4 repeats and telomere ends [61].

In individuals who display the classical phenotype of FSHD but do not have the D4Z4 repeat contraction typically seen in FSHD1, FSHD2 should be considered. The diagnosis of FSHD2 requires the identification of a pathogenic variant in chromatin modifier genes (SMCHD1, DNMT3B, LRIF1) with the identification of decreased 4q35 methylation on the permissive 4qA haplotype [64]. It is advisable to first evaluate for mutations in the SMCHD1 gene, as they account for approximately 80–85% of all FSHD2 cases [21, 65]. If available, whole-exome sequencing (WES) should be offered, as it can evaluate SMCHD1, DNMT3B and LRIF1 concurrently [66, 67].

It has been proposed that OGM in conjunction with WES can help provide a comprehensive approach to the detection of both FSHD1 and FSHD2. However, it is important for healthcare providers to interpret genetic testing outcomes with caution due to inherent test limitations. In addition, the length of D4Z4 repeat does not reliably predict the disease course or severity, due to phenotypic variability and incomplete penetrance [68, 69].

Current management

The present approach to managing FSHD is primarily supportive in nature, since disease-modifying therapy has not yet progressed beyond clinical trials. This includes exercise and rehabilitation, optimization of pain control and conducting longitudinal surveillance for extra-skeletal systemic manifestations. Certain patients also benefit from orthopaedic interventions, such as scapular fixation surgery.

Clinical trials have shown that aerobic exercises may help improve the patient’s exercise performance and cardiovascular fitness, without damaging muscle tissue [70,71,72]. The physiotherapist can tailor exercises based on the individual’s physical status, with the aim to enhance range of motion and alleviate pain [5••, 54]. Orthotic devices, such as ankle-foot orthoses and lumbar corsets, are commonly recommended. It is notable that approximately 20% of patients may require a wheelchair for mobility after reaching the age of 50 [30, 73]. As upper limb weakness may restrict the use of a manual wheelchair, a motorised wheelchair is the preferred option for patients with FSHD. Along with exercises, nonsteroidal anti-inflammatory drugs can be used for managing acute pain, while chronic pain can be addressed with anti-depressants or anti-seizure medications [51, 74••].

All individuals with FSHD should undergo a baseline pulmonary function test, and those with kyphoscoliosis, lumbar hyperlordosis, chest wall deformities, co-existing chronic lung or cardiac conditions, severe disease leading to wheelchair dependence or severe proximal weakness should have annual testing [54, 75]. Approximately 1% of FSHD patients require nocturnal non-invasive ventilatory support, and this usually occurs only decades after the onset of the disease [41]. Sleep-disordered breathing such as obstructive sleep apnea and nocturnal hypoventilation can also be present in patients with FSHD [39, 76, 77]. As such, clinicians should also screen patients for symptoms such as early-morning headaches and non-restorative sleep and, if present, consider polysomnography for further evaluation [76, 77]. It is recommended to initiate nocturnal non-invasive ventilation in FSHD patients with a forced vital capacity of less than 60% on lung function tests or those with sleep-disordered breathing disorders [54]. Routine cardiac screen is not required unless the patient is symptomatic [54].

In terms of surveillance for ophthalmic manifestations, all patients should undergo a baseline fundoscopy and dilated retinal examination [54]. There is a higher risk of retinal complications in patients with early-onset FSHD or those with D4Z4 repeat array fragments that are less than 15 kb in size, suggesting a need for closer monitoring in these patients [44]. If signs of retinal vasculopathy are detected, prompt intervention with photocoagulation can help to prevent further retinal damage [78]. Additionally, some patients with FSHD have weakness of the orbicularis oculi, resulting in difficulties with eyelid closure and lagophthalmos. Topical lubricants, ointments and eye patches can be used at night to prevent exposure keratopathy from developing as a consequence of this.

Individuals with FSHD have an increased risk of developing sensorineural hearing loss. Much like the risk associated with retinal complications, the risk of developing hearing loss is greater in those with shorter D4Z4 repeat arrays and those with FSHD characterised by an earlier onset (e.g. infantile or adolescent-onset) [47]. Regular evaluations are recommended for these specific groups [46]. In patients with adult-onset FSHD, routine hearing assessments are not necessary unless symptoms are present [79].

Scapular fixation surgery involves surgical fixation of the scapula to the posterior thorax. Apart from cosmetic improvements, scapulothoracic arthrodesis has resulted in functional improvements in shoulder flexion and abduction, for patients with severe scapular winging and preserved deltoid strength [80, 81]. The Horwitz manoeuvre, a bedside manual scapular fixation test which imitates the post-surgical mechanics, can help predict post-surgical improvement [80, 81]. However, physicians should carefully evaluate the potential complications of surgery in comparison to its benefits, taking into consideration the patient’s disease progression rate and the need for prolonged post-surgical bracing [82].

Future therapies

Previously, clinical trials involving albuterol, salbutamol, diltiazem, corticosteroids and certain myostatin inhibitors (MYO-029 and ACE-083) did not demonstrate clinical benefit for individuals with FSHD [74••, 83,84,85,86,87,88,89,90,91]. However, recent advancements in research have provided further insight into the fundamental pathophysiology of FSHD, particularly the aberrant expression of the DUX4 protein in skeletal muscles. This understanding has paved the way for the development of targeted therapies directed at suppressing DUX4 expression or mitigating its downstream effects. This is done via the following: (i) epigenetic silencing of the D4Z4 repeats, (ii) blocking DUX4 mRNA production and (iii) targeting downstream pathways triggered by DUX4 expression [92, 93••].

In this segment, we discuss targeted therapies that are currently being investigated in clinical and/or pre-clinical studies. The main avenues for targeting DUX4 include small molecules, oligonucleotide therapeutics and CRISPR-based approaches. At present, all potential therapies are in the pre-clinical stage, with the exception of losmapimod, which is currently undergoing a phase 3 clinical trial [94••]. Although none of these novel therapies have been approved yet, they represent a pivotal change in the treatment landscape of FSHD, moving beyond traditional supportive and symptomatic therapies to treatments that directly target the fundamental root cause.

Small molecules

Small molecule drugs are developed through chemical synthesis that bind to cellular targets to affect disease processes [95]. In contrast to biologics, they are non-immunogenic and have low molecular weight, allowing for oral administration with favourable cellular uptake, and are generally more cost-effective [95].

Losmapimod is an oral, selective, small molecular inhibitor of the P38 mitogen-activated protein kinase (MAPK) pathway, which is a modulator of DUX4 expression and a mediator of inflammation [96]. Previous pre-clinical studies done with mice models showed a significant reduction of DUX4 levels of approximately 80% with losmapimod [9, 96]. It has been shown to be well tolerated with no serious adverse events in a phase 1 study [97•].

A randomised, double-blind, placebo-controlled multicentre phase 2b clinical trial on losmapimod (ReDUX4) involving 80 patients with FSHD1 was recently completed [98]. The patients were randomised 1:1 to receive either losmapimod 15mg, twice daily or a placebo for 48 weeks. The primary endpoint, which was a reduction in DUX4-driven gene expression in skeletal muscle, was not achieved. However, there was a statistically significant benefit in the secondary endpoints in terms of structural, functional and patient-reported outcomes [98,99,100]. After 48 weeks of treatment, patients who received losmapimod showed reduced progression of muscle fat infiltration (MFI) on MRI (0.03% vs. 0.52%; disparity, − 0.49; 95% CI, − 0.86 to − 0.12; p = 0.01) compared to those who received placebo [101]. Reachable workspace (RWS), used as a performance measure of the shoulder and proximal arm function, also showed that patients who received losmapimod performed better than those who received a placebo in the RWS measure with weights. Analysis of reachable surface area (RSA) showed that the annualised rate of change (%/year) in total RSA for losmapimod versus placebo in the dominant arm was − 0.44 versus − 8.42, p = 0.07; in the non-dominant arm, this was 4.88 versus − 4.02, p = 0.01 [98]. In the losmapimod group, assessment of maximum voluntary isometric contraction (MVICT) via hand-held dynamometry also showed stabilisation across various parameters [94••]. Additionally, relative to placebo, these patients also reported significant improvement in the Patients’ Global Impression of Change (PGIC) assessment as well (difference, − 0.58; p = 0.02) [98].

An open-label extension of the above trial was conducted, and preliminary data was presented at the 2023 AAN Annual Meeting [102]. Participants who were on losmapimod continued to receive the drug (LOS/LOS) and, at week 96, were assessed for durability of treatment response, via assessment of RWS. Participants who received a placebo were converted to losmapimod at week 48 (PBO/LOS) and received the drug for another 48 weeks. Annualised total RSA showed stability in the LOS/LOS group in the 2nd year (0.18%/year) compared to the 1st (− 0.77%/year) [102]. In the PBO/LOS group, participants exhibited trends of slowing or halting of disease progression based on RWS, as shown by improvement in annualised total RSA in the 2nd versus the 1st year (4.07%/year versus − 9.96%/year, respectively) [102]. Throughout the extended duration, no drug-related serious adverse events or discontinuation due to adverse events were reported.

At present, Fulcrum Therapeutics has just completed the enrolment of 260 patients in a double-blind, multi-national, placebo-controlled phase 3 trial (REACH) (ClinicalTrials Identifiers: NCT05397470), to further evaluate the efficacy and safety of losmapimod for the treatment of FSHD [103, 104]. The primary endpoint involves evaluation of change from baseline RWS, along with secondary endpoints such as analysis of MFI using whole-body MRI, quality of life in the neurological disorders upper extremity scale (Neuro-QoL UE) and PGIC [104]. Preliminary data is anticipated to be reported in the fourth quarter of 2024.

Antisense oligonucleotides

Antisense oligonucleotides (ASOs) are modified single-stranded DNA or RNA sequences that bind to complementary targeted mRNA sequences, thereby preventing or altering the translation of protein. DUX4 expression can be targeted by ASOs that bind to specific DUX4 mRNA sequences [105]. Pre-clinical studies with ASOs have proven efficacy of reduction of DUX4 and DUX4 target genes in cultured FSHD myocytes and FSHD mouse models [105,106,107,108,109]. The drawback with ASOs, however, is limited bioavailability and poor cellular uptake, which limits the effectiveness of delivery to the target muscle tissue [110].

AOC 1020, developed by Avidity Biosciences, comprises a unique monoclonal antibody that binds to the transferrin receptor 1 (TfR1) combined with a siRNA designed to specifically target DUX4 mRNA [111, 112]. FORTITUDE, a randomised, placebo-controlled, double-blind phase 1/2 clinical trial (ClinicalTrials.gov Identifiers: NCT05747924), is currently ongoing, to evaluate AOC 1020 in 72 participants with FSHD, with the aim to evaluate the safety and tolerability of the drug when administered intravenously [112, 113]. The primary objective of this study is to evaluate the safety and tolerability of AOC 1020, whereas secondary objectives include analysing the pharmacokinetics and pharmacodynamics of the drug. There are three parts to this study—part A consists of dose titration to evaluate the safety of the drug at two low doses, whereas part B involves ascending doses of the drug to study two presumably effective doses [112]. Finally, part C aims to evaluate clinical outcomes. The study will assess measures of mobility and muscle strength, including the use of MRI to measure muscle volume and composition [112]. There will also be an open-label extension study, whereby eligible participants will be given the option to enrol in. Avidity intends to share data from a preliminary assessment of approximately half of the study participants in the first half of 2024.

Myostatin inhibitors

Myostatin is a growth differentiation factor that plays an essential role in regulating skeletal muscle growth [114]. As such, myostatin inhibition has been postulated to help increase muscle mass and, in turn, muscle strength [115]. GYM329 (RO7204239) is an investigational monoclonal anti-myostatin antibody which targets inactive latent myostatin, preventing its conversion to active myostatin, thus reducing the levels of myostatin in muscle and blood [116].

MANOEUVRE (ClinicalTrials.gov Identifiers: NCT05548556) is a multi-centre, randomised, placebo-controlled, double-blind phase 2 trial that aims to evaluate the safety, tolerability, efficacy, pharmacokinetics and pharmacodynamics of GYM329 (RO7204239) in adult patients with genetically confirmed FSHD1 or FSHD2 [116, 117]. It is currently in the recruitment phase, with the aim of enrolling 48 participants. The trial involves participants receiving subcutaneous RO7204239 or placebo injection every 4 weeks, over a treatment period of 52 weeks. Primary outcome measures include assessing percentage change from baseline in contractile muscle volume (CMV) of quadriceps femoralis via MRI bilaterally as well as analysis of adverse effects experienced by participants, whereas secondary outcome measures include assessment of motor function and strength and change from baseline in CMV in other muscle groups as assessed by MRI as well as changes in serum myostatin levels [116]. After completion, there will be an open-label extension, and participants will be given the option to participate and receive RO7204239 for another 52 weeks.

Gene therapy (CRISPR)

CRISPR/CAS9 (clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9) gene editing techniques are in development for various genetic diseases, including FSHD. The technique involves combining a guide RNA sequence complementary to the target DNA, with the CAS9 enzyme, forming the CRISPR/CAS9 complex. The complex can target DNA sequences complementary to the guide RNA, allowing the CAS9 enzyme to make targeted double-stranded DNA breaks. This can be used to disrupt the cut genes or utilise DNA repair to insert new DNA template sequences.

Applications of CRISPR/CAS9 techniques for FSHD are in various pre-clinical stages of development. One application involves the use of an inactivated form of the CAS9 enzyme (dCAS9-KRAB system) to induce epigenetic silencing of the DUX4 gene (instead of creating double-stranded DNA cuts), resulting in decreased production of the DUX4 transcripts and downregulation of target genes [118, 119]. Other studies have targeted the DUX4 polyadenylation signal, required to stabilise the DUX4 transcript [120, 121••]. In FSHD2, a study targeted the intronic variant of the methylation regulation gene SMCHD1 with CRISPR/CAS9 gene editing, which restored SMCHD1 expression and suppression of DUX4 [122]. There is potential progress made towards human studies, with an announcement of plans in 2024 for a first-in-human trial for a CRISPR/CAS9 treatment targeting epigenetic silencing of DUX4 expression, delivered by an adeno-associated virus vector [123].

Conclusion

FSHD is one of the most common muscular dystrophies in the adult population that manifests with disabling skeletal muscle weakness and multi-systemic complications. At present, the mainstay of management is limited to supportive management to preserve and optimise functional independence, with the aim to improve quality of life. In recent years, further understanding of the underlying molecular pathophysiology of FSHD has led to advances in pre-clinical and clinical trials for targeted therapy. This holds a promising potential for disease-modifying management in the foreseeable future, which may alter the disease trajectory for this condition.