Fragile X Syndrome and FMR1-Associated Disorders

Literature and regulatory information updated: 6 October 2026; integration of the French national protocol (PNDS of the HAS): 2 November 2026.

Fragile X syndrome (full mutation), fragile X-associated tremor/ataxia syndrome (FXTAS) and fragile X-associated primary ovarian insufficiency (FXPOI) in premutation carriers, intermediate alleles, mosaicism, AGG interruptions and risk of maternal expansion; premature ovarian insufficiency of other genetic causes in the differential diagnosis

0. Initial Reference Table

1. Introduction

FMR1-associated disorders are the paradigmatic example of a single gene giving rise to distinct clinical entities according to the size of a trinucleotide repeat and its methylation status. The full mutation, with more than 200 CGG repeats and promoter hypermethylation, silences the gene and causes fragile X syndrome (FXS), the most frequent heritable monogenic cause of intellectual disability and the most frequent known monogenic cause of autism spectrum disorder (ASD), accounting for around 2% to 3% of all ASD cases [2–5]. The premutation, with 55 to 200 unmethylated repeats, does not cause FXS, but is associated with two adult-onset entities — fragile X-associated tremor/ataxia syndrome (FXTAS) and fragile X-associated primary ovarian insufficiency (FXPOI) — and a set of neuropsychiatric manifestations grouped under the name FXAND; intermediate alleles, between 45 and 54 repeats, do not cause disease, but may be unstable and expand to a premutation in the next generation [2, 6–8]. This article addresses the complete FMR1 spectrum and includes, in relation to FXPOI, the remaining genetic causes of premature ovarian insufficiency that the geneticist must consider when assessing a woman with hypergonadotrophic amenorrhoea before 40 years [9, 10].

The history of FXS follows that of medical genetics itself. In 1943, James Purdon Martin and Julia Bell, in London, published the pedigree of a family with intellectual disability transmitted through the maternal side and restricted to males, the first demonstration of a form of X-linked intellectual disability, which gave rise to the eponym Martin–Bell syndrome [4, 11]. In 1969, Herbert Lubs identified, in lymphocytes from affected men and some carrier women in a family, an X chromosome with a constriction on the long arm, which he called “marker X” [12]. Macro-orchidism without endocrine abnormalities was described by Turner and colleagues (1975) and Cantú and colleagues (1976) [4], and in 1977 Grant Sutherland demonstrated that expression of the fragile site depended on the folate-deficient culture medium, enabling cytogenetic diagnosis for around fifteen years [4]. This method overestimated prevalence — the initial estimates of 80 per 100,000 men included other fragile sites near FRAXA — and was abandoned with molecular identification of the gene [2].

In 1991, Verkerk and colleagues, in a consortium including the groups of Ben Oostra, Stephen Warren and David Nelson, cloned the FMR-1 gene and demonstrated that it contained a CGG repeat coinciding with the fragile site's breakpoint region and varying in length in patients [13]. In the same year, Fu and colleagues showed that repeat instability explained the so-called “Sherman paradox” — the observation that the risk of intellectual disability in a family increases from generation to generation, with unaffected transmitting men — establishing the concept of dynamic mutation [14]. The premutation was considered clinically silent for a decade, until the description in 2001 by Randi Hagerman and colleagues of five grandfathers of children with FXS with progressive intention tremor, executive deficits and generalised cerebral atrophy, associated with elevated FMR1 messenger RNA (mRNA) levels [15]; Greco and colleagues described in 2002 the neuronal and astrocytic intranuclear inclusions constituting the neuropathological marker of FXTAS [16], and Jacquemont and colleagues proposed in 2003 the clinical–radiological diagnostic criteria still in use [17]. The association between the premutation and early menopause was systematised by Stephanie Sherman in 2000, who estimated that around 20% of women with a premutation have ovarian failure before 40 years [18].

2. Molecular Genetics and Pathophysiology

2.1 The gene and protein

FMR1 is located at Xq27.3 and encodes FMRP (fragile X messenger ribonucleoprotein), an RNA-binding protein associated with polyribosomes, expressed in the cytoplasm of many cell types, with particular abundance in neurons and gonads [2]. FMRP contains two RNA-binding KH domains and interacts with a subset of mRNAs with G-quadruplex motifs; it regulates alternative splicing, stability, dendritic transport and local postsynaptic translation of a broad set of mRNAs, and, through liquid–liquid phase separation, forms cytoplasmic ribonucleoprotein granules that maintain mRNAs in translational repression until synaptic activity releases them [2, 19]. FMRP thus acts as a brake on local protein synthesis at the synapse, essential for dendritic spine maturation and synaptic plasticity [2, 4].

2.2 The CGG repeat and allelic categories

The CGG repeat is located in the 5′ UTR of exon 1. In the population, allele distribution peaks at 29 to 31 repeats, with smaller clusters around 20 and 40 [2]. In stable alleles, the CGG sequence is interrupted by an AGG trinucleotide every nine to ten CGGs, usually at positions 10 and 20; premutation alleles are distinguished by the absence of AGGs or the presence of a single interruption at the 5′ end, with long stretches of pure CGGs at the 3′ end [2, 3]. Boundaries between categories are not absolute — they depend on family history and observed instability — and vary between publications, as the following table shows [2–4, 20].

The laboratory usually indicates an imprecision of ± 2 to 3 repeats in measuring intermediate alleles, so a result of 55 repeats should be treated as a potential premutation; when precision is not stated in the report, the laboratory should be contacted [2]. The upper limit of 54 repeats for an intermediate allele is conservative: to date, no maternal transmission of an allele with 55 or fewer repeats has produced an affected individual [2, 21]. The full mutation almost always shows extensive somatic variation in repeat number in peripheral blood, so laboratories report it as a range of several hundred repeats [2].

2.3 Full mutation mechanism: loss of function

When the repeat exceeds around 200 CGGs, the promoter CpG island and the repeat itself become hypermethylated, chromatin acquires repressive histone marks and FMR1 transcription is silenced; FMRP is no longer produced or is produced in residual amounts [2, 4, 22]. The association of FMR1 mRNA with polysomes, high in normal cells, decreases progressively as repeat size increases, and the FMRP level correlates with the degree of cognitive impairment in both sexes [3]. In the absence of FMRP, dendritic translation dependent on the group I metabotropic glutamate receptor (mGluR5) becomes dysregulated and exaggerated, with increased long-term depression and persistence of long, immature dendritic spines — the so-called fragile X “mGluR theory” [2, 23]. Neuroimaging documents changes in total white matter volume and regional grey matter volumes, which correlate with FMRP levels and intelligence quotient (IQ) [4]. The mechanism is therefore loss of function, explaining why partial or complete gene deletions and rare sequence variants (for example, missense variants in the KH2 domain) cause the same phenotype [2].

2.4 Premutation mechanism: RNA toxicity and non-canonical translation

The premutation is unmethylated and, paradoxically, is associated with increased FMR1 transcription, with elevated mRNA levels, while FMRP decreases slightly only in the upper premutation range [2, 24]. The dominant pathogenic model is a toxic RNA gain of function: mRNA with expanded CGG repeats sequesters RNA-binding proteins and compromises their normal function [2, 24]. A second mechanism, described by Todd and colleagues in 2013, is repeat-associated non-AUG translation (RAN), which produces a polyglycine peptide (FMRpolyG) present in intranuclear inclusions and toxic in cellular and animal models [25]. Eosinophilic, ubiquitin-positive intranuclear inclusions in neurons and astrocytes constitute the neuropathological substrate of FXTAS [3, 16]. In FXPOI, elevated FMR1 mRNA levels in granulosa cells correlate with lower oocyte yield, suggesting the same toxic mechanism at follicular level [2]. Because the full mutation silences the gene, manifestations dependent on toxic RNA do not occur in the same way in full mutation carriers: women with a full mutation have no increased risk of FXPOI or signs of diminished ovarian reserve [2].

2.5 Repeat instability, AGG interruptions and maternal expansion

Repeat expansion is a dynamic mutation: expansion risk increases with allele size, in a continuum from intermediate allele to premutation, without a defined “step”; classification into categories is primarily a practical risk management tool [3]. After the initial expansion period, during which one or more allele sizes may be generated, the expansion maintains remarkable somatic stability [3]. Three factors determine premutation allele instability: the sex of the transmitting parent, repeat number and the number and position of AGG interruptions [4]. Expansion to a full mutation occurs only in female meiosis; in paternal transmission, the premutation remains a premutation, with small expansions or contractions, and daughters of a transmitting man frequently receive a slightly shorter allele [2, 3]. Several studies have shown that men with a full mutation in somatic tissues have only premutation-sized repeats in sperm and gonadal tissue [3].

The stability of alleles with fewer than 90 repeats depends strongly on the number of AGG interruptions, both for variation in the size of intermediate alleles and small premutations and for expansion to a full mutation of premutations above around 60 repeats [2, 21, 26]. In 375 mothers and 48 fathers with alleles of 45 to 69 repeats, Nolin and colleagues found that the nine expansions to a full mutation arose from maternal alleles without any AGG and that the magnitude of expansion was greater in alleles without interruptions [26]. In a subsequent study of 1,040 alleles from 705 families with 45 to 90 repeats, alleles without AGGs had the highest risk of instability in all size categories, and 89% (24 of 27) of maternal alleles with fewer than 75 repeats that expanded to a full mutation had no AGGs; a possible effect of maternal age was also observed, without statistical significance (p = 0.06) [27]. AGG genotyping is clinically available and may refine the risk for premutations with fewer than 100 repeats, although it is not performed routinely and its clinical utility is not completely established; the impact of AGGs on clinical premutation manifestations (FXTAS, FXPOI) is unknown and studies on FXPOI are contradictory [2].

The smallest allele described with expansion to a full mutation in a single generation had 56 repeats [3]; in Nolin and colleagues' collaborative series of 13 laboratories, the smallest premutation alleles expanding to a full mutation had 59 repeats, without AGG interruptions [28]. The same series showed that most transmissions of intermediate alleles in families without a history of fragile X are stable (72.8%), with increasing instability with size (19% for 49–54 repeats, 30.9% for 55–59 and 80% for 60–65), and that the proportion of full mutations does not differ between male and female fetuses [28]. Among 2,932 mother–child transmissions in the normal and intermediate range, Youings and colleagues observed only five changes; instabilities were around 90 times more frequent in alleles of 40 to 59 repeats than in alleles with fewer than 40 [3].

2.6 Mosaicism

Mosaicism is frequent in FXS, occurring in around 20% of cases, and takes two forms [2, 4]. In size mosaicism (full mutation/premutation mosaicism), cell lines with an unmethylated premutation and a methylated full mutation coexist; in methylation mosaicism, the full mutation shows variable degrees of methylation [2]. Individuals with mosaicism tend to have higher intellectual performance than carriers of a fully methylated full mutation, but most have intellectual disability [2]. Individuals with methylation mosaicism or a completely unmethylated full mutation and normal intelligence have rarely been described, reflecting FMRP production from unmethylated alleles and showing that expansion and methylation are not absolutely coupled [2, 4]. Han and colleagues described a man with complex mosaicism (full mutation, premutation and deletion) who had only learning difficulties and 22% of normal FMRP levels [2]. Like premutations, unmethylated (active) full mutations frequently express elevated mRNA levels, considered toxic [4, 22].

3. Clinical Spectrum and Manifestations by System

3.1 Fragile X syndrome in males (full mutation)

The phenotype of boys with FXS varies according to puberty [2]. In childhood, medical problems include hypotonia, gastro-oesophageal reflux and feeding difficulties, strabismus, seizures, sleep disturbances, joint laxity, flat feet, scoliosis and recurrent otitis media; the skin is frequently soft and smooth [2, 4]. Growth is usually normal, with head circumference above the 50th centile and relative macrocephaly — generally close to the 75th centile — and linear growth accelerates in childhood and decelerates in adolescence; the combination of rapid growth in the early years and motor delay may suggest Sotos syndrome [4, 29]. Motor and language milestones are delayed: sitting unsupported at 10 months, walking at 20.6 months and first clear words at 20 months, on average [2].

Intellectual disability is almost universal in boys with a methylated full mutation. Mean IQ has been reported between 40 and 45, with a range from below 10 to normality [2]; Smith's describes usually moderate to profound disability, with IQ between 30 and 55, sometimes in the mild or borderline range [4]. The Oxford Desk Reference relates IQ to molecular status: mean IQ of 41 in men with a methylated full mutation, 60 in men with premutation/full mutation mosaicism and 88 in men with a full mutation unmethylated in more than 50% of cells [3]. IQ tends to decrease with age, reflecting a slower rate of acquisition than peers rather than regression [4].

Language ranges from absent speech to mild communication difficulties; speech lacks fluency, with incomplete sentences, repetition, echolalia, perseverative speech in less affected boys and short bursts of repetitive speech in those more affected [3, 4]. Behaviour is the domain with the greatest impact. It includes attention deficit hyperactivity disorder (ADHD) symptoms — hyperactivity, impulsivity, distractibility — stereotypies such as hand-flapping or hand-biting (≈ 60%), tactile defensiveness, sensory hyperarousal, anxiety, shyness, gaze avoidance (≈ 90%), tantrums, irritability, aggression and self-injurious behaviours [2, 4]. In the differential diagnosis of self-injury in Lesch–Nyhan disease, the Atlas of Inherited Metabolic Diseases reports self-injurious behaviour in up to 58% of boys with FXS (Symons and colleagues, 2003), mainly biting of the fingers or the back of the hand and head-banging, without information on severity (tissue loss or amputations) [30]. ASD is present in 50% to 70% of boys and is associated with more severe behavioural problems and more frequent seizures [2, 4]. Most children are affectionate and interested in social contact, but have difficulty interacting, appearing shy and anxious in groups; social avoidance emerges in the first year of life and persists into adulthood [3, 4]. The behavioural course is characteristic: lower reactivity and activity before two years, followed by increasing hyperactivity and the emergence of anxiety and irritability during childhood [2].

The French national protocol (PNDS of the HAS, 2021) adds quantitative data on the cognitive and communicative profile, drawn mainly from large North American family surveys [1]. The North American national parent survey by Bailey and colleagues (2009) analysed 981 families with at least one child with a full mutation, a figure that the PNDS reproduces correctly; the methods section reports 1,298 children with a full mutation (1,015 male and 283 female), the abstract 1,105 males, and the figures sum to 1,089 and 299 by age group, an internal discrepancy that the article does not explain [1, 31]. Among males aged 20 years or older (n = 281), according to their parents, 91% used single words or gestures, 86% two- or three-word phrases and 62% complex sentences, 92% understood others and only 59% could sustain a conversation; the PNDS figures of 62%, 24% and 14% (complex sentences, only two- to three-word phrases, only single words or no expressive language) are derived from the differences between these proportions and refer only to adult males, not to boys in general, and the 14% comprises about 5% who used only single words and 9% with neither words nor gestures [1, 31]. In reading, 78% of adult males recognised letters, 65% recognised words by sight, 59% knew letter sounds, 44% read picture books or simple stories and 19% read books with new words or concepts; the 19%, 25%, 34% and 22% that the PNDS attributes to "fragile X adults" are likewise derived from these proportions and describe males only, because among adult females (n = 97) 99% recognised letters, 91% words, 91% read simple stories and 76% books with new content [1, 31]. Articulation remains limited in adult males: 75% pronounced words clearly, 71% spoke at an appropriate volume, 62% at a normal rate and 55% intelligibly overall [31]. Speech, initially highly echolalic, becomes rapid and choppy, sometimes with stuttering, and is associated with phonological, syntactic and pragmatic disorders, with perseveration on words or topics [1]. Mathematics is the most impaired school domain, because it depends on sequential processing, working memory, visuospatial skills, attention and abstraction; ADHD is described as almost constant in boys, an expression of broader executive dysfunction (planning, inhibition, flexibility), with hyperactivity that attenuates in adolescence and may disappear in adulthood [1]. Short-term memory, especially verbal, is weak, whereas long-term episodic memory and implicit procedural learning are relatively preserved, which explains why a child remembers a place visited a year earlier but not an instruction given minutes before [1]. Perception of basic emotions is not globally impaired but, compared with age-matched controls, boys identify anger, fear, sadness and disgust less well, and theory of mind is frequently impaired; motor or vocal tics (about 6%) do not appear more frequent than in the general population [1]. The PNDS states that 10% to 13% of boys with a full mutation have no intellectual disability (Hagerman, 1994) and that the severity of disability should now be graded by adaptive function, not by IQ alone, according to DSM-5 [1]. Boys with mosaicism or incomplete methylation may have preserved reasoning abilities, while retaining anxiety, attention deficit, executive dysfunction and visuospatial difficulties [1].

Regarding ASD, the PNDS figures are lower than those of GeneReviews: FXS would account for about 5% of autistic presentations (Hagerman, 2009), and 8% to 30% of children with FXS (up to 50% in some studies) would meet criteria for autism (Hagerman, 2002; Kaufmann and colleagues, 2017) [1]. The divergence from the 50% to 70% of boys with ASD and the 2% to 3% of ASD cases attributed to FXS cited above reflects the difficulty of diagnosing ASD in a population in which most individuals have symptoms that mimic it (gaze avoidance, stereotypies, social anxiety), differences between instruments and the fact that series do not distinguish the sexes in the same way [1, 2]. Boys with FXS and ASD had more epilepsy (20.7% versus 7.6%), persistent sleep disorders, aggressive behaviours and neuroleptic prescription than boys with FXS without ASD [1]. The PNDS concludes that therapeutic approaches developed for ASD are useful in all people with FXS, with or without that formal diagnosis [1].

The face — long and narrow, with a prominent forehead, large prominent ears with soft cartilage and concavity of the upper helix, prognathism and thickening of the nasal bridge — becomes more evident with age and is rarely marked before school age; many young children have a normal appearance, and only a subgroup has the typical features, which are not reliable for diagnosis [2–4]. Smith's adds pale blue irides, epicanthic folds, a high-arched palate and dental crowding [4]. After puberty, macro-orchidism is identified in virtually all men, and anxiety (social phobia, specific phobias, anticipatory, performance, separation and generalised anxiety) and irritable or aggressive behaviour may increase; caregivers identify anxiety as the most disabling problem [2]. A subgroup has obesity with a Prader–Willi syndrome-like phenotype (hyperphagia and hypogonadism) [2, 4].

Connective tissue involvement — attributed to dysregulation of extracellular matrix proteins — includes joint hyperextensibility, particularly of the finger joints, soft skin or mild cutis laxa, flat feet, pectus excavatum, scoliosis, mitral valve prolapse, tricuspid and pulmonary regurgitation and aortic root dilatation; dilated ureters and vesicoureteric reflux may occur [2, 4]. Mitral valve prolapse is rare in childhood, but may occur in around 50% of adults [3]. Recurrent otitis media occurs in 60% to 80% of children and causes conductive hearing loss; seizures occur in around 20% and usually resolve in adolescence [3]. Occasional findings include nystagmus, hypermetropia, myopia or astigmatism, torticollis, cleft palate and abnormal ossification patterns [4]. Life expectancy is normal [4].

The French PNDS quantifies the somatic manifestations with data that partly diverge from those above [1]. In the newborn, sucking difficulties are rare, cleft palate is uncommon (8% of cases, according to the PNDS, which describes it as exceptional without indicating the source series) and cases of Pierre Robin sequence have been described; in infancy, gastro-oesophageal reflux is frequent, sometimes with oesophagitis and, very rarely, oesophageal dyskinesia with megaoesophagus [1]. Chronic constipation may delay toilet training, with persistent encopresis in about 10%, whereas other patients have rapid transit linked to sympathetic hyperactivity and anxiety; these digestive complaints persist into adulthood in about one third (Utari and colleagues, 2010), and swallowing without chewing is common [1]. In the ear, nose and throat sphere, 85% of children have had at least one episode of acute otitis media and 23% sinusitis; 45% to 63% have recurrent otitis, compared with 15% of siblings and 38% of children with intellectual disability without FXS, and one third to one half require tympanostomy tubes, often with adenoidectomy [1]. In a child without language, otitis may manifest only as irritability, head-banging, sleep disturbance, anorexia or vomiting, which justifies examining the tympanic membranes even with mild fever [1]. Refractive errors occur in 13% to 17% (mainly hypermetropia, with or without astigmatism), nystagmus in 5% to 13% and strabismus, usually divergent, in 17.5% (8% to 36% depending on the series) [1]. Pes planovalgus affects about 50% of boys; recurrent patellar dislocations may occur owing to joint hyperlaxity, kyphoscoliosis is rare but should be monitored in adolescence, and there may be a subtle pyramidal syndrome without spasticity [1].

Epilepsy (excluding febrile seizures, which are also more frequent) affects 9% to 44% depending on the series, 12% to 15% of boys and 3% to 5% of girls in the large North American series, and 13% in a series of 113 boys and men aged 2 to 51 years (Berry-Kravis, 2002) — values lower than the approximately 20% cited by the Oxford Desk Reference [1, 3]. It begins on average at around 5 years, with focal motor or autonomic seizures, related to sleep and sometimes prolonged; status epilepticus, almost always at onset, is exceptional [1]. The electroencephalogram (EEG) shows, in 25% to 50% of children, broad triphasic centrotemporal spikes activated by sleep, similar to those of benign epilepsy with centrotemporal spikes; the epilepsy is usually mildly active, controlled with monotherapy (valproate, lamotrigine or carbamazepine), and seizures cease in adolescence in 75% of cases [1]. Sleep disorders affect about half of children under 5 years and a quarter of adults over 20 years in a parental questionnaire covering more than 1,000 boys (Kronk and colleagues, 2010), in the form of difficulty falling asleep (especially before 5 years), multiple awakenings or early morning waking; obstructive sleep apnoea occurs in 7% to 34% [1]. As regards the heart, the PNDS stresses that the only series with systematic echocardiography (Sreeram and colleagues, 1989; 23 asymptomatic adults) found mitral valve prolapse in 22% and dilatation of the aortic arch in 52%, without worsening with age, whereas in a cohort of 198 clinically assessed men, without systematic echocardiography, prolapse was recorded in only 0.5% (Kidd and colleagues, 2014); echocardiographic criteria for prolapse have become more restrictive since the 1980s, which qualifies the estimate of about 50% of adults [1, 3].

In adulthood, men with a full mutation show relatively preserved daily living skills compared with communication and socialisation abilities; nevertheless, most require some degree of residential support [3].

3.2 Fragile X syndrome in females (full mutation)

Women with a full mutation are less affected because the normal X chromosome, active in a proportion of cells, produces variable amounts of FMRP; variability arises from the proportion in the brain of active X chromosomes carrying the full mutation relative to inactive ones with the normal allele [2, 3]. Around 50% have intellectual disability, generally less severe than in men, and around 50% have normal intelligence [2, 3]. Smith's gives a mean IQ of 82 in women with a full mutation, IQ below 70 in 30% to 50% and below 85 in 50% to 70% [4]. Even with normal IQ, learning difficulties are frequent — with better verbal than performance abilities and specific needs in arithmetic, visuospatial skills, abstract concepts and visual and auditory memory — as are ADHD, anxiety and socioemotional dysfunction; behavioural problems are expressed more through social withdrawal than hyperactivity [2, 3, 32]. Because of their milder phenotype, women may achieve greater independence and quality of life, albeit with cognitive, behavioural and, particularly, socioemotional challenges [4]. In the survey by Bailey and colleagues (2009), girls with a full mutation had, according to their parents, higher mean scores than boys in all seven domains assessed (eating, dressing, toileting, bathing and personal hygiene, communication, articulation and reading; p < 0.001 in all) and acquired skills earlier: at 6–10 years (n = 55), 96% undressed themselves, and at 11–15 years (n = 55) 96% dressed independently, 95% buttoned their clothes, 91% chose appropriate clothing and 87% tied their shoes [31]. Among women over 20 years (n = 97), 100% used the toilet independently, 93% wiped themselves, 94% managed bathing without a reminder, 96% used a towel without help, 86% a toothbrush and 91% a comb, 91% sustained a conversation, 89% used complex sentences and 91% spoke clearly; the acquisition curve was curvilinear, with most gains by 6–10 years and subsequent stabilisation of mean scores, except for reading books with new content, which continued to progress into adulthood (76%) [31].

The French PNDS gives a lower estimate of intellectual disability: one quarter to one third of girls with a full mutation, with a mean IQ between 74 and 90 in most and sometimes strictly normal (Cronister and colleagues, 1991; Bennetto and Pennington, 2001), compared with about 50% in GeneReviews and 30% to 50% with IQ below 70 in Smith's; the differences arise mainly from the origin of the samples (families ascertained through an affected boy or clinical series) [1, 2, 4]. The typical profile combines specific learning difficulties in arithmetic, attention deficit, weak working memory and difficulties with planning and visuospatial reasoning with generally adequate written language; marked anxiety, social withdrawal, extreme shyness, obsessive-compulsive disorders and depression may compromise schooling and social integration, even though emotion perception is preserved [1]. Diagnosis in girls is still often late, and the PNDS stresses its value for genetic counselling and for the girl's and her family's understanding of the biological vulnerability underlying her difficulties [1].

3.3 Genotype–phenotype correlation

The male phenotype depends almost entirely on the nature of the variant; the female phenotype depends on the variant and random X-inactivation [2]. The table summarises the correlation described in GeneReviews [2].

Within the full-mutation range (about 200 to 1,000 repeats), the number of CGG repeats does not correlate with clinical severity; what modulates the phenotype is residual FMRP production, through mosaicism or partial methylation [1]. The statement that 100% of men with a methylated full mutation have intellectual disability (GeneReviews) and the estimate of 10% to 13% of boys with a full mutation without intellectual disability (PNDS, based on Hagerman, 1994) become compatible if the latter includes carriers of mosaicism and incomplete methylation, which the table separates; the PNDS does not specify the methylation status of the original sample [1, 2].

3.4 Fragile X-associated tremor/ataxia syndrome (FXTAS)

FXTAS is a late-onset multisystemic neurodegenerative disease affecting premutation carriers, characterised by progressive cerebellar gait ataxia and intention tremor, followed by cognitive decline [2, 33]. Onset is typically between 60 and 65 years and rarely before 50; age at onset and rate of progression vary considerably and are related to repeat size, sex and other factors [2, 33]. The first sign is usually intention tremor, followed by ataxia; most patients require a walking aid within ten years of diagnosis [2]. Cognitive impairment begins with executive functions and extends to working memory and processing speed; almost half of patients meet dementia criteria [2]. Associated findings include short-term memory impairment, parkinsonism, peripheral neuropathy with neuropathic pain, proximal lower-limb muscle weakness and dysautonomia [2, 24]. Psychiatric disturbances are frequent — anxiety, irritability, agitation, hostility, obsessive–compulsive disorder, apathy and depression — and worsen with disease stage, with differences between sexes [2, 34].

Penetrance depends on age and sex. In men with a premutation, risk is 17% between 50 and 59 years, 38% between 60 and 69, 47% between 70 and 79 and 75% from 80 years [2, 3]. Overall prevalence in men with a premutation above 50 years is estimated at around 40% to 45%, and in women at 16% to 20%; in a study of Spanish families with FXS, penetrance above 50 years was 45.5% in men and 16.5% in women [2, 33, 35]. Smith's indicates that around 38% of men and 17% of women with a premutation develop some sign of FXTAS after 60 years [4]. Most premutations found in patients with FXTAS have more than 70 repeats [3]. Larger repeats are associated with greater motor impairment (tremor, ataxia and parkinsonism), more severe neuropathy, more inclusions, greater magnetic resonance imaging (MRI) abnormalities and earlier onset [2].

The French PNDS places FXTAS in carriers of 60 to 200 repeats, with cases mainly above 70, and cites penetrance in men as 17% at 50–59 years, 38% at 60–69, 45% at 70–79 and 75% above 80, attributing it to Jacquemont and colleagues (2003); these values in fact come from the same group's penetrance study published in 2004 (192 carriers and controls from families recruited through Californian associations, survey and neurological examination in 2002–2003), which gives 47%, not 45%, at 70–79 years, as in the following table, and presents them as minimum estimates of the combination of intention tremor and gait ataxia [36]; in women it gives 8% to 16%, below the 16% to 20% of GeneReviews [1, 2]. Action tremor of the upper limbs is usually the first sign, but is absent in 10% to 20% of patients; it is present in 80% to 90% and may be a fine distal tremor of about 6 Hz, similar to essential tremor (but never isolated), a more proximal or axial, wide and slow tremor of about 3 Hz, of cerebellar type, or a parkinsonian rest tremor [1]. Cerebellar ataxia occurs in almost all patients, and the first falls occur on average after six years of disease; parkinsonism, of variable intensity and sometimes levodopa-responsive, occurs in 30% to 60%, and peripheral neuropathy in 50% to 80%, sometimes as the presenting feature, either as a non-length-dependent axonal sensory ganglionopathy or as a length-dependent axonal sensory neuropathy [1]. Cognitive decline of subcortical–frontal dysexecutive type affects about 50% of patients and later extends to working memory and visuospatial abilities, sparing episodic memory, language and praxis for a long time, unlike Alzheimer disease; cognitive symptoms may be the presenting feature, in the form of mild cognitive impairment, and anxiety affects about 50% [1]. In women with FXTAS, myalgia and hypothyroidism may be added [1].

Source: GeneReviews, adapted from Grigsby et al., 2005 [2].

3.5 Fragile X-associated primary ovarian insufficiency (FXPOI)

FXPOI is defined as hypergonadotrophic hypogonadism before 40 years in women with a premutation and occurs in around 20% of these women, compared with 1% in the general population [2, 9, 18]. The relationship with premutation size is non-linear: risk is highest between 80 and 99 repeats and lower in larger premutations [2, 3]. It may arise as early as 11 years, with primary amenorrhoea and pubertal delay, and affected women have a high rate of infertility and menopausal-type symptoms — vasomotor symptoms, mood changes and vaginal dryness [2]. Unlike menopause, ovarian function in primary ovarian insufficiency is erratic and unpredictable: some women maintain irregular ovulations and menstruation for years after diagnosis, whereas others become completely amenorrhoeic; it is estimated that up to 12.6% of women conceive after diagnosis of FXPOI [2, 37]. Hypo-oestrogenism leads to osteoporosis and cardiovascular disease, and primary ovarian insufficiency is associated with a higher risk of thyroid disease [2, 38]. Smoking advances the age at onset of ovarian insufficiency by an average of five years [2, 39]. Initial studies suggesting increased risk in high-normal and intermediate alleles were not confirmed by more robust studies [2].

The French PNDS describes the relationship between premutation size and FXPOI risk somewhat differently: maximum risk between 59 and 99 repeats, decreasing above 100 (Ennis and colleagues, 2006; Allen and colleagues, 2007; Wittenberger and colleagues, 2007), whereas GeneReviews and the following table place the peak between 80 and 99 [1, 2]. Both formulations describe the same inverted U-shaped curve: the odds ratio is already elevated between 59 and 79 repeats and reaches its maximum between 80 and 99. In addition to FXPOI itself, carriers, especially in the 80 to 99 repeat range, have diminished ovarian reserve, with FSH higher than that of control women in all cycles, particularly after 30 years, without reaching the values of ovarian insufficiency, and lower AMH; on average, menopause occurs about five years earlier than in women without a premutation (Murray and colleagues, 2014) [1]. In established ovarian insufficiency, the probability of spontaneous pregnancy is low [1].

Source: GeneReviews, from Sherman, 2005 [2].

3.6 Other premutation-associated manifestations (FXAND and medical findings)

The designation FXAND was proposed to promote characterisation of premutation-associated neuropsychiatric manifestations — anxiety, depression, ADHD, social phobia and social anxiety — affecting around 50% of carriers according to some series [4, 6]. Results are, however, heterogeneous: many studies focus on mothers of children with FXS, in whom maternal stress is a confounder, although others suggest that the premutation confers greater susceptibility to stress and dysregulation of the hypothalamic–pituitary axis [2]. Studies of cognitive difficulties, ASD and neurodevelopmental diagnoses in children with a premutation are contradictory and subject to selection bias; in the great majority of cases, premutation carriers have normal IQ, and when a child with learning difficulties has a premutation, it should not be assumed to be the cause without investigating other aetiologies [2, 3]. Increased rates of arterial hypertension, hypothyroidism, fibromyalgia, migraine, insomnia, sleep apnoea, restless legs syndrome, central sensitisation syndrome and autoimmune diseases have been reported, particularly in women with a premutation, but on the basis of questionnaires and record reviews, with discordant results between groups [2]. A study of 110 daughters of men with FXTAS showed a higher incidence of neurological and psychiatric symptoms than in controls, and there are reports of tremor and ataxia in carriers who do not fulfil FXTAS criteria [2].

3.7 Intermediate alleles

Intermediate alleles do not cause FXS, and more robust studies have not confirmed the association with FXPOI [2]. Their clinical significance lies in the potential for transgenerational instability: around 14% expand into the premutation range during maternal transmission, with risk for subsequent generations — grandchildren being the nearest generation potentially at risk of FXS — and around 16% of maternal transmissions undergo minimal changes (one or two repeats) [2, 3].

4. Clinical Recognition and Warning Signs

4.1 When to consider fragile X syndrome

Signs of FXS in early childhood are non-specific and developmental delay is the almost universal manifestation; FXS should therefore be considered in any child of either sex with speech, language or motor delay of unknown cause, particularly when there is a family history of intellectual disability, a compatible physical and behavioural phenotype and no structural brain malformations or other congenital defects [2]. GeneReviews recommends testing in men and women with intellectual disability or developmental delay of unknown cause, in men with unexplained ASD and in women with unexplained ASD accompanied by an additional indicator — a compatible phenotype, a family history of X-linked neurodevelopmental disorders, or premature ovarian insufficiency, ataxia or tremor in close relatives [2]. The Oxford Desk Reference considers molecular FMR1 testing indicated in all individuals with intellectual disability, even without dysmorphism, because of the non-specific presentation in early childhood and high recurrence risk [40, 41]. Between 1% and 3% of individuals with developmental delay, intellectual disability or ASD have a full mutation [4].

Warning signs in the history and examination are: language delay with repetitive or echolalic speech; gaze avoidance; hand-flapping or hand-biting; hyperarousal and sensory overload in noisy environments (for example, supermarkets), with tantrums, hyperactivity, withdrawal or repetitive behaviours; relative macrocephaly; a long face and prominent ears with concavity of the upper helix; joint hyperlaxity, particularly of the metacarpophalangeal joints; flat feet; soft skin; postpubertal macro-orchidism; a midsystolic click or mitral prolapse murmur; recurrent otitis [2–4]. The pedigree is the most sensitive tool: it should cover at least three generations, extending along the maternal line — maternal aunts and their children, maternal grandmother — and should look for intellectual disability in boys, milder difficulties in girls, menopause before 40 years and late-onset tremor or ataxia in grandparents [3, 40, 41]. The presence of premutations produces pedigrees with apparent skipping of generations or apparently sporadic cases [2].

The French PNDS highlights, as early signs, ligamentous hyperlaxity with valgus feet and subluxation of the thumbs, present before the facial features, and notes that the newborn is of normal size, sometimes with mild macrocephaly, and that the infant may be hypotonic without marked postural delay; macrocephaly may be associated with advanced stature [1]. In the family history, early menopause in female relatives and degenerative neurological disease in men over 60 years strengthen the suspicion [1]. Diagnostic delay persists: in France, the mean age at diagnosis of boys fell from 9.2 to 5.8 years between 1994 and 2001 (Biancalana and colleagues, 2004), with diagnoses still late, especially in girls [1].

The postal survey by Bailey, Skinner and Sparkman (2003) describes the path to diagnosis in 274 North American families with at least one child with FXS, one response per family, selected from the 299 families that returned the questionnaire (out of about 500 contacted through the FRAXA foundation and a longitudinal study) and restricted to biological parents who were unaware of their carrier status before the birth of the first affected child; the mother was the carrier in all cases [42]. In the sample as a whole, the first concern arose on average at 15.6 months, professional confirmation of delay at 25.9 months, entry into early intervention or special education at 32 months, the test request at 56.2 months and the diagnosis at 60 months, with diagnoses between 6 months and 30 years; variability depended mainly on year of birth, before or after identification of the gene in 1991 [42]. In boys born in 1990 or later (n = 117), the means were 13.3 months for the first concern, 20.7 months (about 21 months) for confirmation of delay, 24.5 months for eligibility for special services, 29.8 months for the test request and 31.5 months (standard deviation 16.6; close to 32 months) for the diagnosis, the latest at 8.4 years, with a mean interval of more than 18 months between first concern and diagnosis; in boys born before 1990 (n = 98), diagnosis occurred on average at 86.6 months, and in girls born in 1990 or later (n = 10) at 37.4 months, about six months later than boys, in a group that probably includes only the most affected [42]. Parents were the first to be concerned in 71.2% of cases, another relative in 7.7%, a health professional in 10.2% and a teacher in 4.4%; the main reason was developmental delay (84.3%), followed by behavioural problems (7.1%), excessive activity (4.3%), feeding difficulties (4.3%) and aversion to being held or to physical contact (1.4%) [42]. When parents took their concern to a professional, 28.2% were told to wait because development might improve and 18.1% that the child was normal, similar proportions before and after 1990, and 58.7% described it as difficult to get professionals to recognise the problem [42]. Before fragile X testing was requested, 17.2% of families had had one or two consultations, 28.1% three to five, 9.5% six to ten and 37.6% ten or more (7.6% did not answer); the test was proposed by the family doctor or attending paediatrician in 18.2% of cases, by a neurologist in 24.8%, by a geneticist in 17.9%, by a developmental paediatrician in 9.1% and, in 19.3%, by someone outside the medical pathway (the parents themselves, another relative, a teacher or a therapist) [42]. In those born in 1990 or later, no child factor (sex, year of birth, presenting sign), family factor (income, education, ethnicity) or health-system factor was associated with age at diagnosis, the interval to diagnosis or the number of consultations [42]. The convenience sample was 94.2% white, with education and income above the US average, so these times probably represent the most favourable scenario [42]. The PNDS cites this survey as showing that 24% of families had consulted a professional ten times before the diagnosis was confirmed; the article reports 37.6% with ten or more consultations before the test was requested [1, 42].

4.2 When to consider FXTAS

FXTAS should be considered in men and women with late-onset intention tremor and cerebellar ataxia without a defined cause, in patients with dementia accompanied by ataxia, parkinsonism or tremor, and in patients with multiple system atrophy of cerebellar type, especially of prolonged course [2]. About 2% to 4% of men with adult-onset cerebellar ataxia and no family history have a premutation; there are no equivalent studies in women [2]. FXTAS is the most frequent known cause of ataxia after 50 years according to Smith's, and may be present in 3% to 5% of adults over 50 years with essential tremor [4]. In a series of 56 patients, 98 different diagnoses had been assigned before the diagnosis of FXTAS, mainly parkinsonism, tremor, ataxia, dementia, dysautonomia and stroke [2]. The diagnosis of FXS in a grandson immediately identifies the maternal grandfather as an obligate premutation carrier at risk of FXTAS [2]. The French PNDS, following Berry-Kravis and colleagues (2007), recommends testing for the premutation in the presence of cerebellar ataxia with onset after 50 years and undetermined cause; action tremor with onset after 50 years, of undetermined cause, with parkinsonism and cognitive decline; a patient over 50 years with T2/FLAIR hyperintensity of the middle cerebellar peduncles or of the splenium of the corpus callosum and compatible symptoms; a family history of an FMR1 alteration in an individual at risk of being a carrier and with compatible symptoms; or a family history of infertility or early menopause in a patient with suggestive symptoms [1, 43]. Testing is justified even without a family history of intellectual disability, without tremor, without the middle cerebellar peduncle sign and in women [1].

4.3 When to consider FXPOI

FXPOI should be considered in any woman with unexplained primary ovarian insufficiency or ovarian failure (hypergonadotrophic hypogonadism) before 40 years [2]. Among women with sporadic idiopathic premature ovarian failure, around 2% have a premutation, a proportion rising to around 14% in familial forms [9]; GeneReviews gives 3.2% of isolated cases and 11% of familial cases [2]. The 2024 European and North American guideline strongly recommends FMR1 premutation testing in all women with non-iatrogenic premature ovarian insufficiency, emphasising that multigene panels and next-generation sequencing do not detect the premutation [10]. In the history, age at menopause in female relatives, family history of intellectual disability and history of chemotherapy or radiotherapy are key points [9].

5. Diagnosis: Criteria, Tests and Algorithms

5.1 Molecular tests

Diagnosis of FMR1-associated disorders requires molecular testing targeting the CGG repeat; FXS typically requires a full mutation (> 200 CGG), whereas diagnosis of FXTAS or FXPOI is associated with a premutation (55–200 CGG) [2, 20]. The most important practical point is that a normal exome, genome or multigene panel does not exclude FXS: these methods do not size the repeat and are useful only when no expansion is detected and clinical suspicion persists [2].

Source: GeneReviews, Table 1, and Oxford Desk Reference [2, 3].

The classic laboratory algorithm begins with PCR: if two distinct normal alleles are detected, additional analysis is usually unnecessary; if no normal allele is detected in a man, or only one allele in a woman (who may be homozygous for a normal allele), repeat-primed PCR is used to determine whether an expanded allele is present [3]. Methylation status is essential in full mutations, because hypermethylation silences the gene [2]. PCR, targeted analysis and, in some cases, Southern blot do not detect exon or whole-gene deletions on an X chromosome in a heterozygous woman [2]. If the phenotype is strongly suggestive and leucocyte testing is normal, testing of another tissue (for example, fibroblasts) should be considered, because cryptic mosaicism has been described [2]. In women with intermediate alleles or small premutations, AGG genotyping may help estimate expansion risk [2]. The 2021 technical standard of the American College of Medical Genetics and Genomics (ACMG) and the 2015 best practice of the European Molecular Genetics Quality Network (EMQN) establish requirements for sizing, methylation and report writing, including stating measurement precision [20, 44]. Cytogenetic demonstration of the fragile site in folate-poor medium is historical and should not be used [2].

In the context of intellectual disability, the yield of FMR1 testing in unselected boys is low (around 3% to 6% in early studies and 1.2% in a more recent study), and chromosomal abnormalities and copy number variations are detected equally or more frequently; chromosomal microarray should therefore form part of laboratory assessment [2]. In Rauch and colleagues' series, FXS (1.2%) was the third most frequent diagnosis after Down syndrome, 22q11.2 deletion (2.4%) and Williams–Beuren syndrome (1.3%) [40]. In mild intellectual disability, a classic series of 439 patients identified fragile X in 4.6%, and in severe intellectual disability the proportion varies between 2% and 6% [40].

5.2 Diagnostic criteria for FXTAS

The criteria of Jacquemont and colleagues, revised by Berry-Kravis and colleagues and the National Fragile X Foundation consensus, apply to premutation carriers and classify the diagnosis into three levels of confidence [2, 17, 43].

Source: GeneReviews [2].

The French PNDS adopts a revised version of these criteria, in which hyperintensity of the splenium of the corpus callosum (Apartis and colleagues, 2012) becomes a major radiological criterion, alongside the middle cerebellar peduncle sign; the minor clinical criteria are parkinsonism, short-term memory or executive function deficit and neuropathy, and the minor radiological criteria are cerebral white matter hyperintensities on T2 or FLAIR and moderate to severe generalised atrophy [1]. The combination rules coincide (definite: one major clinical criterion with one major radiological criterion or typical inclusions; probable: two major clinical criteria, or one major radiological criterion with one minor clinical criterion; possible: one major clinical criterion with one minor radiological criterion), but, by classifying the splenium sign as major, a patient with intention tremor and that sign alone has a definite diagnosis under the PNDS and only a possible one under the GeneReviews table [1, 2]. MRI is almost always abnormal in FXTAS, but the middle cerebellar peduncle sign is absent in about 40% of patients (Leehey, 2009), and the splenium sign is useful precisely in women and in patients without the MCP sign; hyperintensities of the brainstem and dentate nuclei and atrophy of the corpus callosum or cerebellum are also frequent [1]. The MCP sign is highly suggestive but not specific in the context of cerebellar ataxia: the radiological differential diagnosis includes multiple system atrophy, cerebrotendinous xanthomatosis, POLG variants, LMNB1 duplication, polyglucosan diseases, sarcoidosis and paraneoplastic syndromes [1].

White matter lesions on MRI in the middle cerebellar peduncles and/or brainstem are characteristic but not specific, and white matter changes may also occur in metabolic and acquired diseases, such as multiple sclerosis [3, 33]. In adults with ataxia, the Oxford Desk Reference includes FMR1 in the DNA testing panel alongside the most frequent spinocerebellar ataxias (SCA1, 2, 3, 6, 7, 12 and 17), dentatorubral–pallidoluysian atrophy, Friedreich ataxia and ataxia with vitamin E deficiency, and notes that trinucleotide expansions may not be detected by next-generation sequencing [33].

5.3 Diagnostic criteria for FXPOI

GeneReviews defines primary ovarian insufficiency by the presence of four to six months of amenorrhoea and two follicle-stimulating hormone (FSH) measurements in the menopausal range at least one month apart, in a woman under 40 years [2]. The 2024 ESHRE/ASRM/CREWHIRL/IMS guideline adopts as criteria disordered menstrual cycles (spontaneous amenorrhoea or irregular cycles) for at least four months and FSH above 25 IU/l, to be repeated after four to six weeks in case of doubt, with no need for timing within the cycle; it recommends karyotype and FMR1 premutation testing in all women with non-iatrogenic premature ovarian insufficiency [10]. The Oxford Desk Reference states that raised FSH (> 20 IU/l) with low oestradiol characterises ovarian failure, whereas normal FSH with raised oestrogen indicates diminished ovarian reserve, and recommends karyotype with screening for mosaicism in 30 cells and FRAXA analysis [9]. Anti-Müllerian hormone (AMH) assesses ovarian reserve [2]. The French PNDS, drawn up before the 2024 guideline, follows the 2016 ESHRE guideline (Webber and colleagues): cycle disturbance (oligomenorrhoea or amenorrhoea) for at least four months and FSH above 25 IU/l on two occasions at least four weeks apart; the 2024 guideline retained the 25 IU/l threshold, but now requires repetition only in case of doubt [1, 10]. The PNDS recommends premutation testing in all women with premature ovarian insufficiency and considers that systematic testing in women with merely diminished ovarian reserve is not consensual and should not be recommended [1].

5.4 Practical algorithm

In a child with developmental delay, intellectual disability or ASD, chromosomal microarray and targeted FMR1 testing (PCR with methylation) are requested in parallel, followed by exome or genome sequencing if both are normal; if suspicion of FXS persists after normal PCR, FMR1 sequencing and deletion testing and investigation of a second tissue are considered [2, 40]. In an adult with late-onset ataxia and tremor, MRI looking for the MCP sign and FMR1 testing form part of the initial investigation alongside spinocerebellar ataxia tests [2, 33]. In a woman with premature ovarian insufficiency, karyotyping and premutation testing are requested together and, if both are normal, next-generation sequencing may be offered after genetic counselling [10]. Once an expansion has been identified, genetic counselling and cascade testing along the maternal line are integral to diagnosis [2, 45].

The pathway proposed by the French PNDS has three entry points — psychomotor delay, intellectual disability or ASD; premature ovarian insufficiency; a neurological presentation suggestive of FXTAS with MRI — which converge on taking the family history and targeted testing for the expansion (commercial PCR or Southern blot) [1]. The test is indicated in any patient, of either sex, with confirmed psychomotor delay, intellectual disability or ASD, the family history and morphology being supplementary arguments and not prerequisites; chromosomal microarray is requested as a first-line test in parallel and may detect deletions or duplications of FMR1 exons [1]. When the patient has dysmorphic or malformative features that do not fit FXS, microarray followed by high-throughput sequencing (panel or exome) as first-line testing is a discussed alternative, bearing in mind that these techniques do not identify repeat expansions [1]. A positive result entails mandatory genetic counselling; a negative result leads to testing for other FMR1 variants or another genetic cause (microarray, sequencing), or to investigation of other causes of ovarian insufficiency or neurodegenerative disease [1]. In France, genetic diagnosis is the responsibility of the geneticist or of a physician (paediatrician, paediatric neurologist, psychiatrist, family doctor) trained in requesting and communicating molecular genetic results, and must always be followed by a genetics clinic appointment with a geneticist or genetic counsellor [1].

5.5 Limitations of genetic testing and report content

Each method covers only part of the mutational spectrum. Repeat-targeted PCR and methylation analysis identify the cause in more than 99% of FXS cases, but do not detect sequence variants or partial or complete FMR1 deletions, which require sequencing and quantitative methods (MLPA, quantitative PCR, microarray); conversely, short-read exome or genome sequencing and multigene panels do not size the CGG repeat or assess methylation, so a normal result from these methods does not exclude FXS, FXTAS or FXPOI [2, 10, 20]. In women, detection of a single allele by PCR may represent homozygosity for a normal allele or an unamplified expanded allele, and a deletion on one X chromosome may go unnoticed [2, 3]. Size and methylation mosaicism are assessed in the tissue studied, usually leucocytes, which may not reflect the brain; with a convincing phenotype and a normal test, analysis of another tissue is considered [2]. Long-read sequencing, which sizes the repeat and reads interruptions on the same molecule, is being used in research, as in the Nigerian CHAMP-FX protocol, but does not replace methods validated by laboratory standards [20, 46].

The report should state the repeat number of each allele with measurement precision (usually ± 2 to 3 repeats for intermediate alleles), allelic category, methylation status of full mutations, presence of size or methylation mosaicism, number of AGG interruptions when genotyped, tissue analysed and method limitations, as established by the 2021 ACMG standard and EMQN best practice [20, 44]. Nomenclature uses the MANE Select transcript NM_002024.6 [47]. In ClinVar, the premutation is recorded as NM_002024.6(FMR1):c.-128GGC[55_200] and the full mutation as c.-128GGC[200_?]: the 3′ alignment rule of Human Genome Variation Society (HGVS) nomenclature shifts the repeat unit to GGC, whereas clinical literature and reports continue to refer to “CGG repeats” — this is the same alteration with two notations, and both should be stated [48]. Sequence variants are normalised on the same transcript: the historical KH2 domain missense variant I304N is written NM_002024.6:c.911T>A, p.(Ile304Asn) and is classified as pathogenic, whereas other missense variants, such as c.413G>A, p.(Arg138Gln), remain of uncertain significance [48]. Classification of sequence variants follows ACMG/AMP criteria and that of copy number variations the joint ACMG/ClinGen standard [49, 50]. A variant of uncertain significance does not become pathogenic through a therapeutic response or partial segregation within a family, and segregation analysis should be described as additional investigation rather than confirmation; zygosity (hemizygosity in men, heterozygosity in women) and chromosomal sex should be included in the report, because they determine interpretation.

6. Differential Diagnosis

6.1 Fragile X syndrome

Sources: GeneReviews, Table 5; Oxford Desk Reference, p. 194–196 and 224–225; Smith's, p. 198 [2, 4, 29, 41].

The French PNDS stresses that some forms of FXS mimic other syndromes: marked macrocephaly with advanced stature suggests Sotos syndrome, and a "Prader–Willi phenotype" (hyperphagia with obesity, hypogenitalism and acromicria) suggests Prader–Willi syndrome, from which FXS is distinguished by the absence of overt neonatal hypotonia and early sucking difficulties [1].

FRAXA and FRAXE are distinct fragile sites, although close together on the X chromosome; the genes spanning them (FMR1 and AFF2) have no detectable similarity at DNA level and the clinical entities are distinct [2, 3]. The approach to a family in which only males have intellectual disability, without male-to-male transmission, includes confirmation of the X-linked pattern, separation of syndromic and non-syndromic forms and exclusion of unbalanced chromosomal rearrangements mimicking X-linked inheritance; FMR1 testing is indicated in all cases [41].

6.2 FXTAS

The differential diagnosis of FXTAS is broad and includes Parkinson disease and other parkinsonian disorders, essential tremor, cerebellar-type multiple system atrophy, autosomal dominant spinocerebellar ataxias (with more marked anticipation in SCA7), dentatorubral–pallidoluysian atrophy, Huntington disease, Gerstmann–Sträussler–Scheinker disease, late-onset Friedreich ataxia, ataxia with vitamin E deficiency (treatable), autosomal recessive cerebellar ataxia due to SYNE1 or ANO10, cerebrotendinous xanthomatosis, mitochondrial ataxias (POLG, MERRF, NARP), dementia with Lewy bodies, multiple sclerosis and cerebrovascular disease [2, 33, 51]. Neuronal intranuclear inclusion disease, associated with a CGG expansion in NOTCH2NLC, resembles FXTAS in morphology and clinical presentation, including white matter hyperintensities, and should be considered in patients with a compatible presentation and normal FMR1 testing, particularly of East Asian ancestry [2, 52].

6.3 FXPOI and other genetic causes of premature ovarian insufficiency

Before establishing primary ovarian insufficiency, other causes of menstrual irregularity, such as polycystic ovary syndrome, thyroid disease and hyperprolactinaemia, should be excluded [2]. Once ovarian insufficiency is established, around one-third to one-half of cases remain idiopathic, and the probability of a genetic, autoimmune or idiopathic cause depends on age at diagnosis [9]. The entities to consider are summarised in the following table.

Sources: Oxford Desk Reference, p. 700–701 and 778–779; Smith's, p. 326–327; Síndromes Genéticas: Reconhecer para Tratar, p. 38; GeneReviews [2, 9, 53–55].

In BPES, FOXL2 mutations are found in 67% to 72% of patients, larger genomic rearrangements in 10% to 15%, deletions outside the transcription unit in 5% and cytogenetically visible translocations or deletions at 3q22 in around 2%, the latter frequently with intellectual disability, microcephaly and growth retardation; ovarian involvement is variable and cannot be predicted from genotype, so division into types I and II has historical value, and oocyte cryopreservation may be considered in women with type I [9, 54]. In balanced X;autosome translocations, the normal X is usually the one inactivated, and the phenotype depends on whether the breakpoint interrupts a gene, a mutation is present on the active X or there is tissue variation in inactivation [53].

7. Treatments and Risk Management Strategies

7.1 General principles and regulatory status

There is no disease-modifying treatment or drug approved by the Food and Drug Administration (FDA) or European Medicines Agency (EMA) with an indication for FXS, FXTAS or FXPOI [2, 56]. Treatment is symptomatic and supportive and, in FXS, combines psychopharmacological treatment of target symptoms with therapeutic interventions — behavioural intervention, speech and language therapy, occupational therapy and individualised educational support — and routine treatment of associated medical problems [2]. In the US, the Fragile X Clinical and Research Consortium (FXCRC), established by the National Fragile X Foundation, brings together 32 specialist clinics and maintains consensus documents by age and domain [2]. Individuals with FXS are more sensitive to psychotropic adverse effects, so these drugs should be started at low doses and titrated slowly [2]. The drugs used are the same as in the general population and, in the context of FXS, are prescribed off-label for the genetic diagnosis, although they may be approved for the symptom (for example, ADHD or ASD-associated irritability).

7.2 Assessments after diagnosis of FXS

After diagnosis, GeneReviews recommends: comprehensive developmental and educational assessment (motor, adaptive, cognitive and language) and occupational therapy assessment; comprehensive behavioural and neuropsychiatric assessment (attention, anxiety, obsessive–compulsive disorder, aggression, depression and ASD); orthopaedic and physiotherapy assessment (hyperextensibility, flat feet, scoliosis, hypotonia); neurological assessment if there are episodes suggestive of seizures; gastroenterological and nutritional assessment if there are feeding difficulties or reflux; ophthalmological assessment (strabismus); otorhinolaryngological and audiological assessment; cardiac assessment with echocardiography (mitral prolapse, aortic root dilatation); and a sleep study if there are apnoea symptoms; medical genetics consultation is recommended in all phenotypes [2, 57].

The French PNDS defines the objectives of the initial assessment as recognising the signs, confirming the diagnosis with a genetic counselling appointment, looking for and treating associated somatic conditions, assessing the cognitive and affective profile, informing about the need for multidisciplinary follow-up and applying for recognition as a long-term condition (ALD) and for the support of the Maison départementale des personnes handicapées (MDPH) [1]. The systematic examinations are hearing assessment by an otorhinolaryngologist and ophthalmological examination; the physicians usually involved are the paediatrician or family doctor, the paediatric neurologist, the otorhinolaryngologist, the ophthalmologist and the psychiatrist and, as needed, the gastroenterologist, the physical medicine and rehabilitation physician, the orthopaedic surgeon, the cardiologist, the sleep specialist and the neurologist [1]. The PNDS diverges from GeneReviews and the American Academy of Pediatrics on two examinations: it does not recommend systematic echocardiography, but careful cardiac auscultation from adolescence, with echocardiography in case of any doubt (midsystolic click, murmur), and discussion of endocarditis prophylaxis for dental or surgical care if mitral valve prolapse is confirmed; and it does not indicate systematic EEG, only in case of clinical suspicion of seizures [1, 2, 57].

7.3 Non-pharmacological interventions in FXS

Early intervention (0–3 years) should include occupational therapy, physiotherapy, speech and feeding therapy and infant mental health support; from three years, inclusion in preschool education with an individual educational plan and, later, transition plans for adult life [2]. Children with FXS learn better when auditory and visual distractions are minimised and work is organised in short blocks (no more than 15 minutes), presented side by side rather than face to face; most can attend mainstream education with support, although behavioural difficulties may make appropriate special education necessary earlier than for other children with a similar cognitive level [3]. Small classes, individualised attention and avoidance of sudden changes are recommended by parents and teachers [2]. Augmentative and alternative communication does not impair speech development and should be considered when verbal language is severely delayed [2]. Applied behaviour analysis may benefit children who meet ASD criteria, and occupational therapy guides self-regulation of hyperarousal and sensory adaptations [2]. Stimulating family and school environments are associated with better performance [4].

The French PNDS details the educational principles, which should be defined by a multidisciplinary team (teachers, speech and language therapist, psychomotor therapist, occupational therapist, neuropsychologist, physicians, educators) in partnership with the child and parents, regularly reassessed and not limited to academic learning [1]. Order and regularity reduce anxiety, and unanticipated changes should be avoided or prepared with verbal explanation, pictures, pictograms or timelines [1]. Short work sessions of 5 to 10 minutes are recommended, with achievable goals presented in advance and a visual timer — a shorter duration than the 15 minutes indicated by the Oxford Desk Reference —, seating in the front row, away from distractions and next to a calm child, breaks to release restlessness and headphones that attenuate distracting sounds [1, 3]. Eye contact should not be forced (looking away reflects distress, not inattention), an activity once started should be allowed to finish, helping to complete it more quickly, and, when the child cannot explain something, it helps to ask questions with binary answers or to be completed [1]. Information should be learned in context, repeated and linked to areas of interest (for example, counting plates when laying the table); since visual memory is better than auditory memory, verbal instructions should be short and broken down and supported by simple visual cues [1]. Fine motor dyspraxia may justify ergonomic pencils and the use of a computer, with training by an occupational therapist; reading benefits from multisensory methods (visual, auditory, tactile) that reinforce grapheme–phoneme association [1].

The same protocol recommends multidisciplinary assessments at the points of educational decision — entry into preschool and primary school, adolescence and transition to adult life —, with adaptive behaviour scales (Vineland, VABS-2) and behaviour scales (Nisonger), while IQ tests remain useful mainly in boys with mosaicism and in girls [1]. Allied health support combines speech and language therapy, psychomotor therapy, occupational therapy, physiotherapy (static disorders or significant postural delay), psychotherapy (preferably group or family) and orthoptics (neurovisual disorders) [1]. In France, this support is provided in private practice, at home or in dedicated facilities: early medico-social action centres (CAMSP, 0–6 years), special education and home care services (SESSAD, 6–16 years), medico-psycho-pedagogical centres (CMPP, 3–18 years), medico-psychological centres (CMP) and, when ASD coexists, the Centres Ressources Autisme; schooling may take place in a mainstream school with a support assistant, in inclusion units (ULIS) or in medico-educational institutes (IME), depending on age, cognitive level, associated disorders and family resources [1]. A secure family environment, with a firm and affectionate educational attitude that values progress, and access to artistic and sporting activities strengthen self-esteem, including in girls with preserved intellectual abilities [1].

Coordination rests, according to the PNDS, on a coordinating physician identified from diagnosis (paediatrician, paediatric neurologist, psychiatrist, geneticist, family doctor or neurologist, depending on the context), with knowledge of the syndrome and of health, medico-social and educational resources, and on a non-medical care pathway coordinator (a specialised educator or nurse, for example), a role that the protocol itself acknowledges is still rare [1]. The gynaecologist (adolescents and adult women) and the dentist trained in caring for people with intellectual disability are part of the follow-up team [1]. Transition to adult care should be anticipated at the end of adolescence: the paediatrician hands over follow-up to a family doctor familiar with intellectual disability and, when necessary, to neurology (persistent epilepsy), psychiatry (ongoing psychotropic drugs) or orthopaedics (advanced scoliosis); the optimal age of transition has not been determined [1].

7.4 Symptomatic pharmacological treatment of FXS

Sources: GeneReviews; PNDS of the HAS (2021); Erickson et al., 2011; Greiss Hess et al., 2016; posologies from the official labels cited in each row [1, 2, 66, 70]. In FXS, these drugs are used symptomatically and largely outside approved indications.

Aripiprazole was associated with a clinical response in 10 of 12 patients (87%) with FXS in a 12-week open-label study, with discontinuation in two because of akathisia, sialorrhoea and tiredness [70]. In the randomised trial of low-dose sertraline in 52 children aged 2 to 6 years, there was no difference from placebo in the primary endpoints (expressive language and clinical global impression), but there was improvement in secondary motor, visuoperceptual and social participation endpoints, without a difference in adverse effects [66]. Medical complications — strabismus, otitis media (with ventilation tubes or prophylactic antibiotics if necessary), gastro-oesophageal reflux, cardiac and musculoskeletal problems and seizures — are treated conventionally [2, 3]. For hypertonia or dystonia, specialists may be involved for the use of baclofen, tizanidine, botulinum toxin (Botox®) or antiparkinsonian drugs [2].

The French PNDS considers that treatment of common conditions (infections, arterial hypertension, anaesthetic induction) requires no specific precautions in FXS and that psychotropic drugs are reserved for behaviours with self- or other-directed aggression, hyperactivity with risk and superimposed psychiatric disorders (anxiety, depression, obsessive-compulsive disorder), always as an adjunct to regular psychiatric follow-up in partnership with the family and the educational team [1]. Benzodiazepines, although indicated for anxiety, are often poorly tolerated in FXS and should be limited to episodes of acute anxiety, never as chronic treatment [1]. Prolonged or exaggerated reactions to some drugs may be observed; neuroleptics require monitoring of weight, screening for dyslipidaemia and assessment for extrapyramidal effects (dyskinesias), and all psychotropic drugs, except methylphenidate, should be withdrawn gradually, with periodic reassessment of the benefit–risk balance to maintain the minimum effective dose, as monotherapy whenever possible [1]. In infants with severe gastro-oesophageal reflux compromising sleep and growth, the PNDS proposes milk thickeners, antacids or alginates and, if symptoms persist, proton pump inhibitors [1]. As regards sleep, melatonin was considered relatively or very effective by 60% of parents in the survey by Kronk and colleagues, but controlled evidence is limited to a crossover trial in 12 patients with FXS and/or ASD (Wirojanan and colleagues, 2009), 10 of whom improved more with melatonin than with placebo; when obstructive apnoea is suspected (snoring, sweating, pauses, difficult waking, daytime sleepiness), otorhinolaryngological assessment is indicated and, if suspicion is strong, polygraphy, and tonsillectomy or nocturnal positive pressure ventilation may be discussed [1].

7.5 Targeted therapies: clinical trials and status in 2026

Understanding the role of FMRP in mGluR5-dependent synaptic protein synthesis prompted trials of negative mGluR5 modulators. Two randomised controlled phase 2b trials of mavoglurant (AFQ056), in 175 adults and 139 adolescents, did not meet the primary endpoint of behavioural improvement on the Aberrant Behavior Checklist (FXS-specific version) after 12 weeks, nor did they confirm methylation status as a predictor of response [3, 77]. Mavoglurant was subsequently studied in combination with intensive language intervention in young children, and other targets (GABA-B receptors, among others) were explored in clinical trials [2, 4].

Zatolmilast (BPN14770), an allosteric phosphodiesterase 4D (PDE4D) inhibitor increasing cyclic adenosine monophosphate (cAMP) signalling, which is deficient in the absence of FMRP, showed good tolerability and improvement in cognitive language measures (oral reading, picture vocabulary and the NIH Toolbox crystallised cognition index) and caregiver scales of language and daily functioning in a phase 2 crossover trial in 30 adult men (25 mg twice daily) [78]. The phase 3 EXPERIENCE trials in adults (NCT05358886, 171 participants, completed in July 2025) and adolescents aged 9 to 17 years (NCT05163808, 163 participants, completed in September 2025) did not meet the primary endpoint (cognition measured by the NIH Toolbox), according to preliminary results released in May 2026; the adult trial showed statistically significant improvement on a caregiver numerical scale, which was a secondary endpoint [56, 79]. In October 2026, the sponsor suspended the open-label extension for adolescents because of the absence of an efficacy signal in this group, without new safety findings, while maintaining the adult extension and dialogue with the FDA [80].

Transdermal cannabidiol (ZYN002), which in a phase 2/3 trial (CONNECT-FX) had not met the primary endpoint but had shown benefit in an ad hoc analysis in patients with full methylation [81], was tested in the phase 3 RECONNECT trial, which did not meet the primary endpoint of improvement in social avoidance in participants with full methylation (≥ 90%), owing to a higher than expected placebo response (September 2025) [82]. Metformin, which corrects deficits in murine and Drosophila models, was studied in a six-month randomised double-blind trial in 34 Chinese children aged 2 to 16 years (250–1,000 mg/day adjusted for weight, trial doses outside the terms of the label, because metformin has no indication in FXS): it improved hyperactivity and sleep, without a difference in total Aberrant Behavior Checklist score or cognitive, social or repetitive behaviour measures; adverse effects (transient appetite reduction, mild self-limiting lactic acidosis, IGF-1 reduction) were comparable to placebo [2, 83]. In 2026, trials are ongoing or in preparation for a second PDE4D inhibitor (MRM-3379, phase 2), BK channel modulators (KER-0193, SPG601), a vasopressin V1a receptor antagonist (SRX246) targeting aggression, repurposed drugs (ethosuximide, tacrine) and genetic strategies using antisense oligonucleotides to restore FMRP production; transcriptional reactivation of FMR1 remains a research avenue [4, 84]. None of these agents is approved.

The ClinicalTrials.gov registry, consulted in October 2026, shows the status of these programmes: the phase 2 MRM-3379 study in male participants with FXS (BLOOM, NCT07209462; 60 planned participants) is recruiting; SPG601, a BK channel activator, has a registered phase 2/3 study (NCT07439510; 248 planned participants) not yet recruiting; SRX246 has a phase 2 study targeting aggression (NCT07328529) yet to start; and a University of Alberta metformin trial (NCT03862950; 125 participants) ended in December 2025 without results published in the registry [85]. The only published clinical data for SPG601 come from a single-dose (800 mg) crossover study in 10 adult men, funded by the sponsor, whose primary endpoint was safety: the drug was well tolerated and modified exploratory resting electroencephalographic measures and an inhibitory control test, constituting a pharmacodynamic signal rather than a demonstration of clinical efficacy [86]. The zatolmilast open-label extension (NCT05367960) remains active, not recruiting [85].

The French PNDS places these failures within a longer list of strategies that corrected the phenotype of the Fmr1 knockout mouse without demonstrated benefit in humans: GABA-B agonists (baclofen, arbaclofen) and acamprosate (with GABA-A and GABA-B action), minocycline (normalisation of matrix metalloproteinase 9), lithium (GSK3β inhibition), Ras-ERK pathway inhibitors (lovastatin, metformin), the CB1 cannabinoid receptor antagonist rimonabant, modulators of calcium-activated potassium channels (BKCa) and pioglitazone (dysregulation of diacylglycerol kinase kappa) [1]. From reviews of these programmes, the PNDS draws methodological lessons: the knockout mouse has a variable phenotype and mild cognitive deficit, so preclinical studies should be replicated in at least two independent laboratories and, if possible, in more than one model; outcomes should be transposable between animal and patient and adapted to cognitive level, covering the three priorities of families (anxiety, behaviour and learning); it is important to define windows of plasticity, control for the placebo effect, which is greater in open-label trials, and ensure sufficient statistical power [1].

Gene therapy for FXS remains at the preclinical stage, as the PNDS describes [1]; a search of ClinicalTrials.gov on 2 November 2026 identified no registered human trial of gene transfer or genome editing, which does not exclude unregistered programmes. Three strategies can be distinguished. The first is gene replacement with recombinant adeno-associated virus (rAAV) vectors carrying the FMR1 coding sequence, administered intracerebrally: in an Fmr1 knockout mouse, hippocampal expression of the main central nervous system isoform of FMRP normalised the exaggerated long-term depression at CA3–CA1 synapses (Zeier and colleagues, 2009), and a review of murine AAV studies shows results ranging from complete correction to partial or no effect, depending on the behavioural or electrophysiological test [87, 88]. The unresolved problems are sufficient brain distribution from local injections, the percentage of transduced neurons required, the risk of FMRP overexpression (a regulatory protein whose dosage is critical), vector immunogenicity and the developmental window in which replacement is still effective [88]. The second strategy is epigenetic reactivation of the endogenous gene: targeted demethylation of the CGG repeat with dCas9-Tet1 and a guide RNA converted the FMR1 promoter from heterochromatin to active chromatin in patient-derived induced pluripotent stem cells, persistently restored gene expression, corrected the electrophysiological abnormalities of derived neurons, maintained expression after engraftment into the mouse brain and also reactivated the gene in post-mitotic neurons (Liu and colleagues, 2018) [89]. This approach preserves physiological regulation of the gene, but requires delivery of a large system to the brain and has not been demonstrated in vivo in animals carrying the human mutation. The third is editing of downstream modifier genes: intracranial injection of CRISPR-Cas9 in gold nanoparticles (CRISPR-Gold), targeting the mGluR5 receptor gene, reduced local receptor levels in the striatum and corrected the exaggerated repetitive behaviours of the Fmr1 knockout mouse, without detectable toxicity at the doses used (Lee and colleagues, 2018) [90]. Risks common to editing strategies include off-target effects, editing mosaicism, immune responses to the nuclease and irreversibility; none of the three has, to date, efficacy or safety data in humans.

7.6 Treatment of FXTAS

Treatment is symptomatic, multidisciplinary and based on anecdotal evidence and extrapolation from similar diseases; it includes neurology with a movement disorders specialist, psychiatry, psychological counselling, speech therapy, occupational therapy, physiotherapy and gait training [2].

Source: GeneReviews, Table 9 [2].

Typical and atypical antipsychotics with a significant antidopaminergic effect and metoclopramide (worsen parkinsonism), anticholinergics (worsen cognitive complaints), excessive alcohol consumption (worsens cerebellar dysfunction and postural instability) and, with caution, agents with known cerebellar toxicity should be avoided [2]. Allopregnanolone, a natural neurosteroid, improved neuronal abnormalities in animal models of FXTAS and is in early human studies [2].

The French PNDS, as in most hereditary ataxias, acknowledges that no trial has demonstrated the efficacy of any drug in FXTAS and proposes symptomatic and empirical measures guided by the dominant signs [1]. For tremor it lists propranolol, levetiracetam, primidone, topiramate and pregabalin (the latter also useful for associated neuropathic pain); for ataxia, physiotherapy and speech and language therapy; for parkinsonism, levodopa and, for akinesia, amantadine; for subcortical–frontal dementia, a low-dose acetylcholinesterase inhibitor as a therapeutic trial; for depression and generalised anxiety, antidepressants (mainly selective serotonin reuptake inhibitors) and psychological or psychiatric support [1]. Memantine, the most studied drug in FXTAS, was not superior to placebo in a randomised double-blind trial, with benefit only in a subgroup [1]. In severe psychiatric disorders, the PNDS allows a low-dose atypical neuroleptic, preferring clozapine when parkinsonism is present, which agrees with the recommendation to avoid antipsychotics with significant antidopaminergic effect [1, 2]. Two cases of lasting benefit on tremor with deep brain stimulation have been described, one in the ventral intermediate nucleus of the thalamus (unilateral, high frequency) and the other in the VoP/zona incerta region, evidence at case-report level [1]. The PNDS does not give doses for these options; those in the table were taken from the labels and reviews cited in each row, and the others follow the posology for the analogous condition (essential tremor, Parkinson disease, dementia) in the respective label.

7.7 Treatment of FXPOI

There is no specific treatment; gynaecological or reproductive medicine assessment guides hormonal therapy and reproductive counselling [2]. After diagnosis, gynaecological assessment with hormonal markers of ovarian reserve and transvaginal ultrasound, psychological referral, bone densitometry by dual-energy X-ray absorptiometry (DXA) and thyroid function assessment are recommended [2].

Hormone replacement therapy should be started at diagnosis and continued at least until the usual age of menopause (around 50 to 51 years), to prevent osteoporosis, vasomotor symptoms and premature cardiovascular disease [2, 9, 10]. The 2024 guideline suggests a daily dose of at least 2 mg oral oestradiol or 100 µg transdermal oestradiol, or equivalent, to optimise bone mineral density, and recommends adding a progestogen in all women with a uterus, increasing the progestogen dose when higher oestrogen doses are used; delayed initiation and non-adherence should be avoided [10]. The US oestradiol labels include hypoestrogenism due to primary ovarian failure: orally, starting with 1–2 mg/day adjusted to clinical response, aiming for the lowest effective maintenance dose [97]; transdermally (weekly system), starting with 0.025 mg/day adjusted to clinical response [98]; neither label defines a maximum dose for this indication. In the Portuguese SmPC of oral oestradiol (Zumenon® 2 mg), the indication is postmenopausal hormone replacement therapy, with 2 mg/day continuously and a progestogen for 12–14 days per 28-day cycle in women with a uterus, at the lowest effective dose [99]. The guideline's oral dose of 2 mg falls within the US labelled range; starting with 100 µg/day transdermally exceeds the labelled starting dose, and in the EU use in FXPOI is an extrapolation from the postmenopausal indication, both outside the terms of the label. Transdermal oestradiol (for example, Estradot®, Climara®, Oestrogel®) with cyclic progesterone or a levonorgestrel-releasing intrauterine system (Mirena®) reproduces physiological levels, is superior to combined contraception for bone density in small trials and minimises thromboembolic risk [2]. Hormonal therapy is preferable to the contraceptive pill, which provides an artificial cycle but does not allow ovulation; pregnancy may occur during hormonal therapy [9]. Adverse effects include mastalgia, irregular bleeding, headache and, with oral formulations, increased thromboembolic risk; migraine is not a contraindication in premature ovarian insufficiency [10]. Adequate calcium and vitamin D intake and weight-bearing exercise are recommended [2]. Surveillance includes annual gynaecological consultation and DXA every two years if initial bone mineral density is low or hormonal therapy inadequate [2]. The French PNDS indicates bone densitometry at diagnosis of ovarian insufficiency, repeated according to the initial result, and annual cardiovascular risk assessment (blood pressure, smoking, body mass index, lipid profile and fasting blood glucose); to women with established ovarian insufficiency who wish to have children it proposes oocyte donation [1].

Regarding fertility, some women have sporadic ovulation and the possibility of spontaneous conception, without any known treatment increasing it (ethinylestradiol showed preliminary results); in vitro fertilisation with donated oocytes or embryos is the option with the highest probability of success, whereas in vitro fertilisation with their own oocytes has very low success rates [2]. Women who do not wish to conceive need effective contraception — long-acting devices, tubal ligation or partner vasectomy — because combined contraception may have a higher failure rate in this population [2]. Pregnancies in women with FXPOI do not have more complications, but pregnancies through oocyte donation have a higher risk of hypertensive disease, prematurity and growth restriction [2].

7.8 Risk management in premutation carriers

Women with a premutation should avoid smoking, which diminishes ovarian reserve and advances FXPOI onset [2, 39]. The French PNDS recommends, in all premutation carriers, a systematic assessment of ovarian function at diagnosis — menstrual cyclicity, oestradiol, FSH and AMH, and ovarian ultrasound with antral follicle count — and its follow-up over time, in order to counsel on fertility and discuss fertility preservation in a specialised clinic; ovarian reserve also determines access to preimplantation genetic testing, which requires stimulation and in vitro fertilisation [1]. In older carriers, aggressive control of cardiovascular risk factors (blood pressure and cholesterol) is recommended, given diffuse white matter damage in FXTAS, as is annual assessment of cognition (Montreal Cognitive Assessment, MoCA) and mood (Beck Depression Inventory or equivalent) in patients with FXTAS [2]. Premutation carriers should be informed about the risks of FXTAS and FXPOI at cascade diagnosis [3, 45]. A 2026 multidisciplinary review proposes that care for FXTAS, in which a frontosubcortical-type major neurocognitive disorder leads to marked functional decline (particularly in men), be coordinated by the primary care clinician, with multispecialty and multidisciplinary collaboration, caregiver support and palliative and end-of-life care planning; this is an expert review rather than a guideline with graded evidence [100].

7.9 Pivotal and negative trials in FXS

These trials do not allow interventions to be ranked or efficacy to be inferred from secondary outcomes. First, positive phase 2 trials (zatolmilast, the ad hoc CONNECT-FX analysis) were not confirmed by phase 3 trials with a prespecified primary outcome, and a subgroup analysis defined after the results — such as that of methylation ≥ 90% — generates hypotheses rather than proof; RECONNECT, designed precisely to confirm it, was negative [81, 82]. Second, primary outcomes differ between programmes (caregiver-completed behavioural scales, computerised cognitive batteries, language measures), and caregiver scales show high placebo responses, making indirect comparison of differences between trials misleading. Third, a statistically significant effect on a secondary endpoint in a trial whose primary endpoint failed does not have the confirmatory value of a primary result. A 2026 scoping review bringing together 23 pharmacological FXS trials (14 randomised and placebo-controlled, 9 open-label; 1,469 participants) found a small median effect size in controlled trials (|d| = 0.22) and a moderate one in open-label trials (|d| = 0.70), with substantial heterogeneity in design, samples and instruments — a difference illustrating the weight of the placebo effect and lack of blinding [102]. As of October 2026, phase 3 zatolmilast and RECONNECT results are known only through sponsor communications disseminated by patient foundations, without peer-reviewed publication, and should be read as preliminary results [56, 82].

7.10 Comparative regulatory status and evidence hierarchy

The evidence hierarchy used in this article distinguishes peer-reviewed publication (for example, the mavoglurant, arbaclofen, CONNECT-FX and metformin trials), a sponsor announcement or preliminary result (phase 3 zatolmilast and RECONNECT results), an agency opinion and an authorisation decision; no drug has a known authorisation application for FXS, FXTAS or FXPOI in the sources consulted. A marketing authorisation does not guarantee marketing, reimbursement or access in each country. Symptomatic drugs used in FXS are authorised for symptoms or disorders (ADHD, irritability associated with autistic disorder, insomnia), not for the syndrome, and indications differ between jurisdictions.

Sources: Drugs@FDA, the cited US labels and EMA medicinal product data file, consulted in October 2026.

Two concepts should be distinguished. Off-label use depends on label wording, age and jurisdiction: aripiprazole for irritability in an 8-year-old boy with FXS and autistic disorder is within the US label but outside the European label; guanfacine in a European child with ADHD who has not tried stimulants, or atomoxetine in a 4-year-old child, are off-label uses even for the correct symptom. Extrapolation of evidence is different: it consists of applying results obtained in other populations (idiopathic ADHD, autistic disorder, essential tremor, premature ovarian insufficiency of other causes) to FXS, because syndrome-specific trials are scarce and mostly negative or open-label [66, 70, 102]. In FXTAS, all symptomatic drugs — primidone, propranolol, gabapentinoids, duloxetine — are used by extrapolation from essential tremor and neuropathic pain, and are within the indication only when the label covers the symptom treated [2, 91].

7.11 Emergency situations and hospital admission

The French PNDS is the only one of the guidelines consulted that addresses emergency situations in FXS in a structured way [1]. Faced with a first epileptic seizure, the approach is the usual one: look for a provoking cause independent of the syndrome (hypoglycaemia, electrolyte disturbance, toxic, traumatic or infectious cause) through the history, examination and, if necessary, laboratory tests, and systematically perform an EEG, ideally including sleep; most seizures stop spontaneously and require only observation — rest in a calm environment, lateral decubitus position, airway clearance, recording of duration, characteristics, tolerance and recovery time [1]. Rescue treatment with rectal diazepam or buccal midazolam is reserved for seizures lasting more than 5 minutes or with cyanosis and for repeated seizures in close succession (for example, three in one hour), a rare situation in FXS; therefore, advance prescription for administration by carers is limited to patients who have already had a prolonged seizure or to an explicit indication from the reference physician [1]. The PNDS does not give doses; the official labels define them. Buccal midazolam (Buccolam®, authorised in the EU for prolonged acute convulsive seizures from 3 months of age to adulthood, in patients with diagnosed epilepsy) is administered as a single age-based dose — 2.5 mg from 3 months to the first year (between 3 and 6 months only in a hospital setting), 5 mg from 1 to under 5 years, 7.5 mg from 5 to under 10 years and 10 mg from 10 years —, and medical assistance should be sought if the seizure does not stop within 10 minutes, with no second dose without medical advice [105]. Diazepam rectal gel (US label, seizure clusters from 2 years) uses 0.5 mg/kg from 2 to 5 years, 0.3 mg/kg from 6 to 11 years and 0.2 mg/kg from 12 years, rounded up to the next available unit, with a possible second dose 4 to 12 hours later and a maximum of five episodes per month and one episode every five days [106]. Hospital admission is justified for seizures that are prolonged, poorly tolerated from a respiratory standpoint or repeated, unresponsive to rescue treatment, or followed by unusual recovery [1]. At school, seizure management may be framed within an individualised reception plan (PAI) drawn up by the school doctor with the educational team, which is unnecessary when seizures occur only during sleep and there is no indication for emergency treatment [1].

An acute behavioural crisis — violent anger in the face of frustration or unrecognised pain, more often with self-injury (biting, head-banging, slapping) than with aggression towards others — is managed first without drugs: settle the person in a calm place until they calm down, supervising to prevent injury, and then, with the carer, listen and speak in simple words, without trying to argue; calming techniques such as a weighted blanket may be used [1]. Drugs are used only in case of absolute necessity; in adults, the PNDS proposes hydroxyzine or a sedative and anxiolytic neuroleptic [1]. Hydroxyzine has no dose in the PNDS; in the EU, since the 2015 re-evaluation of the risk of QT interval prolongation and torsade de pointes, the maximum daily dose is 100 mg in adults and 50 mg in older people (if use cannot be avoided), and 2 mg/kg/day in children up to 40 kg, the lowest effective dose should be used for the shortest possible time, and it is contraindicated in congenital or acquired QT prolongation and in the presence of known risk factors [107]. Once the episode has passed, the cause is sought with the patient and carer: superimposed somatic disease (especially painful or infectious, or epilepsy) or unrecognised psychiatric disease (anxiety, depression); communication difficulties, in which the frustration of not being understood is the most frequent origin; and environmental factors (changes, interruption of support, conflicts, transitions, inadequate care setting) [1]. The first-line response is to understand the trigger, adjust the environment (spaces, tasks, preparation for changes) and apply educational and behavioural approaches (positive reinforcement, social skills, alternative communication); when a drug is necessary, it complements these measures, at low dose, as monotherapy and with regular reassessment [1].

Emergency admission should be avoided through regular follow-up and early detection of common illnesses [1]. When it occurs, the PNDS recommends first contacting the attending physician or the specialist who knows the patient (or the medical emergency number), presenting the DéfiScience network's fragile X emergency card, keeping an up-to-date medical liaison file for patients with chronic conditions or in institutional care, allowing a carer to be present, including for adults — a right framed by French disability compensation legislation and by the 2017 HAS recommendation on access of people with disabilities to healthcare facilities — and assessing pain with a validated behavioural scale, such as the GED-DI modified for ASD, whose baseline scoring should be done outside a crisis and reviewed annually [1]. The emergency card records the people to contact, the attending and specialist physicians and the reference centre, associated conditions (recurrent otitis, gastro-oesophageal reflux, significant constipation, epilepsy, enuresis, encopresis, heart disease, arterial hypertension, sleep apnoea, hearing loss, low vision, difficulty in standing), current treatment with posology, the mode of communication (oral expression, understanding of simple instructions, reading, writing, use of pictures, sign language or pointing) and the warning that the person has difficulty expressing and describing pain [1].

Source: PNDS of the HAS, section 4.3.3 and annex 5 [1].

7.12 Divergences between the French PNDS and the other guidelines

8. Reproductive Options and Family Counselling

8.1 Mode of transmission and risk to relatives

FMR1-associated disorders have X-linked transmission, and the full mutation can only be inherited from the mother [2, 3]. All mothers of individuals with a full mutation are heterozygous for a premutation or full mutation; if the proband has a premutation, the mother may have an intermediate allele or a premutation [2]. The father of a male proband does not have an expansion and does not need testing; in a female proband, the expansion may come from the mother or a father with a premutation, so testing of both parents is recommended [2]. If the mother has a premutation, the maternal grandfather may be a “normal transmitting man” at risk of FXTAS, or the maternal grandmother may have an intermediate or premutation allele [2].

A woman with a premutation or full mutation has a 50% probability in each pregnancy of transmitting the expanded allele; the Oxford Desk Reference formulates this as four equally probable outcomes in each pregnancy — a non-carrier boy, a non-carrier girl, a boy with a premutation or full mutation and a girl with a premutation or full mutation — emphasising that expansion does not necessarily occur in every pregnancy, but that massive expansion is possible from any premutation [2, 3]. For alleles above 100 repeats, expansion to a full mutation in female meiosis is virtually constant, so half of the children will have FXS; on the basis of random X-inactivation, the risk of a daughter of a woman with a premutation being clinically affected is lower (around 15% to 30%, depending on allele size) [4]. A man with a premutation transmits it to all daughters and no sons; the repeat remains relatively stable (small expansions and contractions are possible), daughters will not have FXS, but are at risk of FXTAS, FXPOI and FXAND and of having children with FXS [2, 3]. Men with a full mutation rarely reproduce; a woman with a full mutation, with or without manifestations, transmits it to 50% of children [2].

Source: GeneReviews, Table 14 [2].

8.2 Risk of maternal expansion

The risk of a maternal premutation expanding to a full mutation depends on repeat number and, below 100 repeats, the number of AGG interruptions [2, 27]. The table reproduces values presented in the Oxford Desk Reference, which relate to women with a family history of fragile X; risks are lower in women without this history [3].

Source: Oxford Desk Reference, Table 3.16 [3].

The influence of family history was quantified by Nolin and colleagues in the analysis of 1,112 prenatal samples: maternal premutations of 70 to 79 repeats had an expansion risk of 54% when there was a family history of fragile X and 18% in its absence [2, 108]. Contractions are rare and more frequent in paternal transmissions, particularly in the 70 to 90 repeat range [2]. In the 45 to 90 repeat range, absence of AGGs is the main marker of instability, and AGG genotyping provides more precise estimates of full mutation expansion risk in women with 45 to 90 repeats [26, 27]. A maternal premutation may originate in the carrier's mother or father, whereas the full mutation is always maternal in origin [3].

8.3 Cascade testing, testing of minors and screening

Cascade testing of adult relatives is indicated [3, 45]. The proband's maternal aunts and their children may be carriers or affected, and apparently healthy brothers of a carrier mother may be transmitting men, who should be offered testing because of the risk to grandchildren through their daughters and the risk of FXTAS [2, 3]. Testing of asymptomatic sisters of affected boys, without school difficulties, is usually deferred until 16 years or later, when they may wish to know their status before family planning; if they have mild school difficulties, vision and hearing assessment and developmental or psychoeducational assessment should be offered, and diagnostic testing may be useful if a learning difficulty without another evident cause is confirmed [3]. Genetic counselling should be offered to young adult carriers or those at risk, ideally before pregnancy [2].

The French PNDS organises the communication of the diagnosis in two stages [1]. The disclosure takes place in a dedicated consultation, which explains that this is a permanent neurodevelopmental disorder and not a recoverable delay, acknowledges the uncertainty about future expression, the guilt associated with the mode of transmission and the difficulty of retaining information at that moment, invites the family to repeat their questions as often as necessary, liaises with the attending physician and medico-social teams for local support in the following days and provides information about patient associations [1]. The genetics clinic appointment, with a geneticist or genetic counsellor and, if necessary, a psychologist, should be separate from the disclosure: it revisits the mode of transmission, reconstructs the pedigree in the light of the diagnosis, and informs the parents about the risk of recurrence and the options of prenatal or preimplantation diagnosis and about the risk of FXTAS and FXPOI in relatives [1]. In France, the patient (or the parents, or the guardian) has a legal obligation to inform potentially at-risk relatives, under Decree No. 2013-527 of 20 June 2013; if they do not wish to do so personally, they may ask the physician to inform the relatives of the existence of information that concerns them [1]. A genetics clinic appointment may be useful for the patients' siblings, especially in adolescence, to inform them about the risk of transmission and, in girls, about the risk of premature ovarian insufficiency; it should not be offered systematically, but the team should be available [1]. Psychological support, for the patient, parents, couple, siblings and grandparents, and the support of the social services complete the care [1].

Diagnosis of the first case in a family is frequently late: in a survey, in around half of families diagnosis was made more than a year after the first developmental concerns, and half of families had subsequent pregnancies before diagnosis of the first affected child [2]. In the survey by Bailey and colleagues (2003), 55.5% of the 274 families had another child before the diagnosis of the first (61.2% when the first had been born before 1990 and 50% when born in 1990 or later); in the 121 families with information, 191 children were born in that interval, of whom 109 (57%) had a full mutation, 7 a premutation, 67 no FMR1 alteration and 8 were not tested [42]. After the diagnosis, 59% of parents said that it influenced their decision to have more children, almost always towards not having them; 21 families subsequently had 31 more children (5 with a full mutation, 1 with a premutation, 23 without an alteration and 2 with no known result) [42]. Genetic counselling was received by 85% of families (33.9% at the time of diagnosis, 14.6% in the following week, 25.8% between one week and one month and 21.5% more than one month later), and was considered very useful by 39.5%, somewhat useful by 45.1% and not useful by 15.5%; the sources from which they learned most about the condition were written materials (41.6%), genetic counsellors (29.6%), the Internet (17.9%), other parents (6.9%) and the family doctor (3.7%) [42]. In 93.8% of families the extended family had to be informed of the risk of being a carrier; in 91% of these cases the parents themselves did so and in less than 7% a professional, the experience was very stressful for 34.8% and, in 70.8%, at least one relative was tested, whereas denial and refusal of testing were the reactions most often described in the others [42]. The most frequently cited advantages of the diagnosis were understanding the child and knowing how to help (63.9%) and knowing the cause (52.9%), followed by informing the family's reproductive decisions (13.9%) and finding support (10.9%); the most frequently cited challenges were readjusting expectations for the child and family (37.6%), explaining the condition to others (17.9%), accessing services and professionals knowledgeable about FXS (15.7%) and the guilt of being a carrier (10.2%) [42]. In the US context, 11% reported changes in health insurance after the diagnosis (refusal, cancellation or exclusion of therapies) [42]. Newborn screening for FXS has been tested in pilot projects, but is not current practice [2]. The 2021 ACMG practice resource on carrier screening during pregnancy and preconception recommends a consistent and equitable approach for all individuals, regardless of ancestry; many gamete banks include FMR1 in expanded donor screening, after transmission of a premutation through a sperm donor was documented [2, 109].

8.4 Prenatal diagnosis and preimplantation genetic testing

Once an expanded or altered allele has been identified in the family, prenatal diagnosis and preimplantation genetic testing (PGT) are possible [2]. Chorionic villus sampling detects the presence of a premutation or full mutation, requiring an adequate sample and exclusion of maternal cell contamination; because FMR1 methylation is established in placental tissue only between 10 and 12 weeks, amniocentesis may be necessary to confirm the methylation status of a full mutation [2, 3]. Confirmation by linkage analysis, available rapidly, may be used, given potential technical difficulties with Southern blot [3]. The French PNDS considers prenatal diagnosis reliable, but stresses that, in a female fetus with a full mutation, it does not allow prediction of the degree of cognitive impairment or of the severity of associated disorders, especially psychiatric ones; preimplantation diagnosis may be offered after the opinion of a multidisciplinary prenatal diagnosis centre and is performed only in a few accredited centres [1]. Information about fetal sex is useful, because female fetuses with a full mutation usually have a less severe phenotype; however, the phenotype of a female fetus with a full mutation, or of a fetus with a premutation (because of variable FXPOI and FXTAS expressivity), cannot be predicted, so prior counselling about these uncertainties is essential [2, 3].

PGT requires in vitro fertilisation with ovarian stimulation, which is difficult in women with FXPOI, who have very low or absent ovarian reserve: in a published case, a woman with FSH of 33 IU/l had an unaffected child after 20 oocyte retrievals to accumulate 15 embryos for biopsy [2]. Even without FXPOI, women with a premutation need higher gonadotrophin doses and obtain fewer oocytes, particularly in the intermediate repeat range (70–100), with fewer embryos available and higher cancellation rates [2]. PGT does not currently determine whether a premutation has expanded to a full mutation and is considered a screening test, given the risk of mosaicism and discordance between trophectoderm and inner cell mass, so confirmatory prenatal diagnosis should be offered [2]. Oocyte donation eliminates transmission risk and is the option with the highest probability of success in women with FXPOI [2, 9].

8.5 Specific counselling aspects

A diagnosis of FXPOI does not eliminate the possibility of conception, and there are reports of women with a premutation who had children with FXS after FXPOI diagnosis; those not wishing to conceive should be informed of the need for effective contraception and those who may conceive should be referred for genetic counselling [2]. Information about FXPOI risk, higher in the 80 to 100 repeat range, may be useful for reproductive planning, including the decision about PGT [3, 9]. Relatives of women with premature ovarian insufficiency due to a premutation or another identified genetic cause should receive genetic counselling and testing, and sisters and daughters of women with non-iatrogenic ovarian insufficiency should be informed that they are at increased risk and that there are no established methods to predict or prevent it [10]. The 2012 NSGC guideline provides detailed guidance on counselling and cascade testing [45].

When counselling the parents of a child newly diagnosed with a full mutation, the functional data from the survey by Bailey and colleagues (2009) allow concrete expectations to be given: according to the parents of 281 men and 97 women over 20 years, most men eat, dress, use the toilet and bathe independently and 91% communicate at least through single words or gestures, whereas conversation, intelligible speech and reading of new text remain limited in a large proportion of them (59%, 55% and 19%), and women reach proportions of 85% to 100% in most skills [31]. These proportions describe a North American convenience sample with high education and income, based on cross-sectional parental report, and do not predict the course of an individual child, which depends mainly on learning ability, adaptability, attention and the number of coexisting conditions [31].

In Europe, referral of patients with FXS and other neurodevelopmental disorders of genetic cause is through the European Reference Network for congenital anomalies and rare intellectual disability (ERN ITHACA); FXTAS falls within the remit of the rare neurological disease network (ERN-RND) and premature ovarian insufficiency within that of the rare endocrine disease network (Endo-ERN) [110]. Patient organisations include the National Fragile X Foundation and FRAXA Research Foundation (US), which maintain information on FXS, the premutation, FXTAS and FXPOI, the Fragile X Society (UK) and, in Brazil, the Instituto Buko Kaesemodel “Eu Digo X” programme [111].

In France, the 2021 PNDS was drawn up by the reference centre for intellectual disabilities of rare causes (CRDI), coordinated at the Pitié-Salpêtrière Hospital (Paris), whose constituent site in Lyon (Hôpital Femme Mère Enfant, Bron) is the reference for FXS and the other X-linked intellectual disabilities, within the DéfiScience rare neurodevelopmental disease health network [1]. At the time of the PNDS, the national network included ten constituent reference centres and thirteen competence centres for intellectual disability of rare cause, in addition to the reference and competence centres for rare diseases with psychiatric expression, which may be contacted by the patient, the family or any professional, at any stage, for an opinion on complex situations, an initial synthesis consultation or a multidisciplinary assessment [1]. FXS is recognised as a long-term condition (ALD), with exemption from patient co-payment under a care protocol established by the attending physician with the specialist; the list of procedures and services annexed to the PNDS includes specialist consultations, EEG, sleep polygraphy, speech and language therapy, physiotherapy and orthoptics, whereas psychomotor therapy, occupational therapy and neuropsychological assessment depend on specific conditions (healthcare facilities such as CAMSP, CMP or CMPP, coordination and orientation platforms for neurodevelopmental disorders, MDPH notification or compensation through the education allowance for children with disabilities, AEEH) [1]. The French patient associations are the Association Fragile X France and Mosaïques — Association des «X fragile»; at European level, the PNDS refers to the associative networks Fragile X networks and X fragile — Europe and, as information resources, the Alliance Maladies Rares, Orphanet and the Maladies Rares Info Services helpline [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1 Global epidemiology

FXS prevalence estimates have been revised downwards since identification of FMR1: cytogenetic estimates of 80 per 100,000 men were replaced by 16 to 25 per 100,000 (using intellectual disability as the cardinal finding), an incidence of 19 per 100,000 male newborns in methylation screening of 36,124 boys in the US and a prevalence of 14 per 100,000 in a meta-analysis [2, 112, 113]. The Oxford Desk Reference gives around 1 in 5,500 men with a full mutation and Smith's 1 in 5,000, calculated from neonatal blood samples [3, 4]. Female prevalence is estimated at half the male prevalence because of reduced penetrance [2]. FRAXE is rarer, at around 1 in 23,000 men [3]. The French PNDS uses older and higher estimates — 1 in 4,000 boys and 1 in 8,000 girls (Crawford and colleagues, 2001) — and estimates the total number of carriers of the anomaly in France, with or without manifestations, at about 10,000, without indicating the method of calculation; it also considers underdiagnosis in women to be likely [1]. For the premutation it gives a prevalence of 1 in 260 in the general population, without distinction by sex, a value intermediate between the sex-specific estimates described below [1].

The premutation is much more frequent than the full mutation: around 0.5% of the population has a premutation allele [3]. Hunter and colleagues' meta-analysis estimated 117 per 100,000 men (≈ 1/855) and 344 per 100,000 women (≈ 1/291), and a population sample of 19,996 adults in Wisconsin (US) estimated 345 per 100,000 men (≈ 1/290) and 677 per 100,000 women (≈ 1/148); in the same sample, intermediate alleles (45–54 repeats) were found in 1 in 33 women and 1 in 62 men [2, 114]. From premutation prevalence and penetrance, FXTAS prevalence in US men was estimated at 1/4,848 and FXPOI prevalence in women at 1/3,560 [2, 115]. Thus, premutation manifestations are collectively more frequent in the population than FXS itself [2].

9.2 Population variation and founder effect

The distribution of repeat sizes and AGG interruption patterns differs between populations. In a study of 3,065 normal alleles from nine populations (Australia, Chile, United Arab Emirates, Guatemala, Indonesia, Italy, Mexico, Spain and the US), Yrigollen and colleagues found significant differences in the frequency of long alleles, long uninterrupted CGG stretches and alleles with more than two AGGs, factors that may modulate the prevalence of FMR1-associated disorders in each population [116]. Peprah's review synthesises the worldwide CGG repeat distribution and haplotypes associated with instability [117].

The most extreme example of a founder effect is Ricaurte, a small locality in southwestern Colombia (Valle del Cauca), with an FXS prevalence of 1 in 38 men and 1 in 100 women — around a hundred times the worldwide prevalence — the largest known FXS cluster; a common haplotype of 33 single nucleotide polymorphisms including FMR1 was found in all male expansion carriers, demonstrating a founder effect [118].

9.3 Regional data: Portugal, Latin America, Asia and Africa

In Portugal, a retrospective study of 94 women with premature ovarian insufficiency in a tertiary centre found chromosomal abnormalities in 16.5% (78.6% involving the X chromosome) and an FMR1 premutation in 6.7% (6/90), without a significant difference between familial and sporadic cases [119]. In a Portuguese public gamete bank, one of 63 female donor candidates carried a premutation [120]. In Latin America, a scoping review of 25 studies (5,531 participants) identified 38 FXTAS cases and 386 premutation carriers without FXTAS, diagnosed between 53 and 85 years, with generalised cerebral atrophy more frequent than the MCP sign and atypical presentations (spastic paraparesis, progressive supranuclear palsy-like presentations) delaying diagnosis; the authors conclude that FXTAS is probably underdiagnosed in the region and that prospective prevalence and penetrance studies are lacking [121]. A systematic review of 51 Asian studies found that the physical phenotype resembles that described in other populations — large prominent ears in 72.6%, an elongated face in 57.5% and macro-orchidism in 45.2% — with intellectual disability in 99%, hyperactivity in 77% and social withdrawal in 55%, and that physical features are not reliable for early recognition [122].

No national population-based prevalence studies of the full mutation or premutation in Portugal or Brazil were identified in the sources consulted; available data come from selected clinical series, measuring the proportion of carriers in a referred population rather than prevalence in the general population, and this gap in the sources does not demonstrate that such studies do not exist. For sub-Saharan Africa there are no national data: a Nigerian multicentre screening study in children with neurodevelopmental disorders (CHAMP-FX), with long-read sequencing, was published as a protocol in 2026 precisely because of the absence of data [46]. Data on FXTAS prevalence in women are also scarce, and there are no studies on the proportion of premutations in women with adult-onset ataxia [2]. These estimates should not be extrapolated without local validation.

9.4 Special populations

In gamete donors, FMR1 screening is increasingly included in expanded carrier screening [2]. In women with premature ovarian insufficiency, premutation prevalence is around 3% in isolated forms and 11% to 14% in familial forms, justifying universal testing [2, 9, 10]. In men with late-onset ataxia without a family history, 2% to 4% have a premutation [2]. In individuals with ASD, FXS represents 2% to 3% of cases [2]. In childhood cancer survivors treated with chemotherapy or radiotherapy, premature ovarian failure (one in six) should be distinguished from FXPOI [9].

10. Illustrative Clinical Case

10.1 Published case: fragile X-associated primary ovarian insufficiency with a diagnostic odyssey

Fink and colleagues described in 2018 a woman followed at the Clinical Center of the National Institutes of Health (NIH) [38]. Cascade testing, performed at 4 years after FXS diagnosis in her older brother (identified at 9 years), had shown a premutation of 100 to 110 CGG repeats; her mother and an aunt were carriers and both had experienced “early menopause” [38]. Menarche occurred at 11 years, without regular cycles; she took an oral contraceptive from 13 to 16 years for dysmenorrhoea and menorrhagia, and subsequently had only scant bleeding every 3–4 months, with night sweats from 17–18 years [38]. At 18 years she developed severe depression, prompting interruption of her first university year and several psychotropic regimens; levothyroxine was prescribed despite normal thyroid function and, at a reference centre, she was even investigated for Cushing syndrome because of elevated morning and urinary cortisol and an abnormal dexamethasone suppression test [38]. On admission to the NIH at 20 years (2006), daytime cortisol measurement was normal; transvaginal ultrasound showed an endometrium of 2 mm and a small left ovary without visible follicles, FSH was 46 IU/l, LH 26 IU/l and oestradiol 23.3 pg/ml, anti-21-hydroxylase and antithyroid antibodies were negative, and prolactin and pituitary MRI were normal, establishing the diagnosis of FXPOI [38]. She started transdermal oestradiol 100 µg/day with medroxyprogesterone acetate 10 mg/day for the first 12 days of each month, psychotropics and levothyroxine were reduced and discontinued, cycles became regular and anxiety and depression resolved; the patient described around three years without anyone understanding her presentation [38]. Already on hormonal therapy and without fertility treatment, she conceived in 2012; prenatal diagnosis showed a full mutation in the fetus and the pregnancy was terminated. She subsequently had two spontaneous pregnancies, in 2013 and 2016 — before the latter, an in vitro fertilisation clinic had estimated a conception probability below 0.1% and suggested oocyte donation — resulting in the birth of two healthy girls [38].

The authors accompany the case with a review in which Figure 2 shows changes in bone mineral density at the femoral neck and lumbar spine over three years of physiological hormonal therapy in women with spontaneous 46,XX primary ovarian insufficiency (https://pmc.ncbi.nlm.nih.gov/articles/PMC6278244/figure/F2/), Figure 1 the follicle count in premutation mice (https://pmc.ncbi.nlm.nih.gov/articles/PMC6278244/figure/F1/) and Figure 4 the proposed design for an international FXPOI natural history study (https://pmc.ncbi.nlm.nih.gov/articles/PMC6278244/figure/F4/) [38]. The case illustrates four practical points: psychiatric and vasomotor manifestations of hypo-oestrogenism may be mistaken for primary psychiatric disease or endocrine disease of another nature, even when carrier status is known; FXPOI does not equate to sterility, and spontaneous pregnancies on hormonal therapy require counselling on contraception and transmission risk; the risk of expansion of a maternal allele of around 100 repeats materialised in the first pregnancy; and the hormonal regimen used in this report coincides with the minimum doses suggested by the 2024 guideline for bone density, but the choice of progestogen should follow that guideline rather than the isolated report [2, 10, 38].

10.2 Integrated teaching case: one family, three generations

This is a composite case, constructed for teaching from data in the cited sources, and does not represent a real family. A 4-year-old boy is seen in the genetics clinic for language delay (first words at 26 months, two-word phrases at 4 years), hyperactivity, gaze avoidance and hand-flapping when excited. His head circumference is at the 85th centile, with an elongated face, prominent soft-cartilage ears, metacarpophalangeal hyperlaxity and flat feet; he has had recurrent otitis media with placement of ventilation tubes. Chromosomal microarray is normal. FMR1 PCR with repeat-primed PCR shows an allele with more than 200 CGGs, and methylation analysis confirms full methylation, establishing a diagnosis of FXS due to a full mutation [2, 3].

His 32-year-old mother has 92 CGG repeats without AGG interruptions; she reports irregular cycles since 29 years and FSH is 38 IU/l on two measurements a month apart, with undetectable AMH, fulfilling FXPOI criteria [2, 10]. She is referred to reproductive medicine and starts transdermal oestradiol 100 µg/day with a cyclic progestogen, calcium and vitamin D supplementation and DXA; she stops smoking [2, 10]. She receives counselling on the 50% risk of transmitting the expanded allele in each pregnancy, with a risk of full mutation expansion above 80% for an allele of this size, and on the options of prenatal diagnosis, PGT (with limitations imposed by ovarian reserve) and oocyte donation [2, 3].

His 68-year-old maternal grandfather, followed in neurology for “essential tremor” since 63 years, now has gait instability and planning difficulties. He carries a premutation of 88 repeats; MRI shows T2 hyperintensities in the middle cerebellar peduncles and moderate cerebral atrophy. With one major radiological sign (MCP sign) and two major clinical signs (intention tremor and gait ataxia), he meets criteria for definite FXTAS [2, 17]. He starts low-dose primidone, physiotherapy and gait training; medication is reviewed, with discontinuation of metoclopramide prescribed for dyspepsia, and vitamin B12 and vitamin D deficiencies are corrected [2]. The 28-year-old maternal aunt, asymptomatic, opts for cascade testing before planning a pregnancy [3, 45].

11. Prognosis and Long-Term Follow-Up

11.1 Fragile X syndrome

Life expectancy in FXS is normal [4]. Functional prognosis depends on the degree of intellectual disability, presence of ASD, severity of anxiety and behavioural disorders, and educational and family support; stimulating family and school environments are associated with better performance [2, 4]. Adults show relatively preserved daily living skills, but most men need some residential support, and women, with a milder phenotype, may achieve greater independence [3, 4]. Seizures usually resolve in adolescence, but mitral prolapse may arise in around half of adults [3].

The survey by Bailey and colleagues (2009) is the largest cross-sectional description of functional skills across the lifespan in the full mutation: 981 North American families recruited through the National Fragile X Foundation, the FRAXA Research Foundation, the Conquer Fragile X Foundation, researchers and clinicians (1,250 enrolled, 1,075 with a completed questionnaire, 86%), who rated 37 concrete skills of each child on a four-level scale; the authors considered a skill acquired when performed "fairly well" or "very well", and the age groups 0–5, 6–10, 11–15, 16–20 and over 20 years comprised, according to the figures, 158, 260, 222, 168 and 281 males and 53, 55, 55, 39 and 97 females [31]. In males over 20 years, basic activities of daily living were present in most: 97% ate with a spoon or fork and 77% used a knife, 98% undressed and 95% dressed themselves, 83% chose appropriate clothing and 79% buttoned it, 93% used the toilet independently, 89% asked to go to the toilet and 83% managed toileting hygiene without a reminder, 82% used a towel without help and 75% managed bathing without a reminder [31]. Below about two thirds were eating at a normal pace (59%), tying shoes (66%), wiping themselves (66%), brushing their teeth (66%) and hair (61%), sustaining a conversation (59%), speaking clearly (55%) and at a normal rate (62%) and reading books with new words or concepts (19%), skills that the authors propose as explicit intervention goals, noting that about one third of adult males performed them [31]. In males, mean scores increased linearly across all age groups in eating, dressing, toileting, bathing and personal hygiene, communication and articulation (p < 0.01), with late gains in some skills — managing bathing without a reminder increased by about 25 percentage points between 16–20 years and adulthood, and using a towel without help rose from 49% at 6–10 years to 67% at 16–20 years —, whereas reading plateaued after 6–10 years; reading from the figures, complex sentences rose from about 20% at 0–5 years to 41% at 6–10, 49% at 11–15, 55% at 16–20 and 62% after 20 years, conversation from about 15% to 59%, and reading of simple stories stood at around 40% to 50% from 6–10 years onwards, without progression in adulthood [31]. In females, acquisition was earlier and adult proportions exceeded 85% in most skills, except reading books with new content (76%) and amount of food eaten (76%) [31].

In children aged 11 years or older, an age at which all the skills assessed would be acquired in the absence of intellectual disability, a higher total functional score (0 to 21, sum of the means of the seven domains) was associated in males, adjusted for age, with better reasoning and learning ability, better mood, better adaptability, better attention and fewer than five coexisting conditions (attention problems, hyperactivity, aggression, self-injury, autism, seizures, anxiety, depression; p < 0.001 in each model), but not with the respondent's education; in females the associations with reasoning, adaptability, attention and coexisting conditions were repeated, and there was a single significant interaction between the respondent's education and attention (p < 0.05) [31]. These figures have limitations: the design is cross-sectional and differences between age groups may reflect different educational and therapeutic experiences between generations rather than individual trajectories; the data are parent-reported on a non-validated scale; the convenience sample was 90% white, with 58% of respondents holding a university degree and 56% with an annual income of at least 75,000 dollars, and 89% of respondents were mothers; better-functioning women, often undiagnosed, are probably under-represented; and grouping all adults over 20 years prevents description of the course in adulthood [31]. The survey did not assess employment, independent living or type of residential support, so it provides no estimates for these outcomes, and it describes a North American generation preceding the targeted therapies under trial, serving as a comparator for intervention studies [31].

Sources: GeneReviews, Table 11; American Academy of Pediatrics health supervision; PNDS of the HAS, section 4.3.1 [1, 2, 57].

11.2 FXTAS

FXTAS is progressive: most patients need a walking aid within ten years of diagnosis, and almost half progress to dementia [2]. Severity, number of inclusions and MRI changes increase with repeat size [2]. Psychiatric manifestations worsen with motor stage and have prognostic value, justifying early intervention even in initial stages [34]. Follow-up includes neurological assessment with a movement disorders specialist, physiotherapy and occupational therapy, annual cardiovascular risk factor control, annual MoCA and depression scale, and medication review to avoid drugs worsening parkinsonism, cognition or ataxia [2]. According to the French PNDS, citing Berry-Kravis and colleagues (2007), the natural history, still poorly known, appears to be marked by loss of walking at around 75 years and death at around 80, with the patient bedridden, dysarthric and incontinent; this is an estimate from clinical series, not a population survival curve [1].

11.3 FXPOI

Premature ovarian insufficiency without hormonal therapy is associated with reduced life expectancy, particularly from cardiovascular disease, supporting hormonal therapy until the usual age of menopause [10]. Follow-up includes annual gynaecological or reproductive medicine consultation (maintenance of hormonal therapy and fertility options), DXA every two years if bone mineral density is low or therapy inadequate, thyroid function and psychological support [2]. The possibility of spontaneous conception, estimated at up to 12.6%, should be recalled at every visit [2, 37]. The PNDS adds densitometry at diagnosis, repeated according to the result, and annual cardiovascular risk assessment [1].

11.4 Premutation and intermediate allele carriers

Asymptomatic premutation carriers benefit from information about early FXTAS and FXPOI signs, avoidance of smoking and excessive alcohol, vascular risk factor control and surveillance of mood and anxiety [2, 6]. Intermediate allele carriers have no established personal risk; follow-up is limited to counselling about transgenerational risk [2].

12. Summary and Final Considerations

The FMR1 CGG repeat gives rise to a spectrum of disorders determined by repeat number, methylation and sex. The methylated full mutation (> 200 CGG) silences the gene and causes FXS through loss of FMRP, a regulator of synaptic translation; it is the most frequent heritable monogenic cause of intellectual disability and ASD, with an incidence of around 1 in 5,000 boys, intellectual disability in all men with full methylation and around half of women, and a behavioural phenotype marked by hyperarousal, anxiety, gaze avoidance and ASD in 50% to 70% [2–4]. The premutation (55–200 CGG), unmethylated and with elevated mRNA, acts through RNA toxicity and RAN translation and causes FXTAS in around 40% to 45% of men and 16% to 20% of women after 50 years, FXPOI in around 20% of female carriers, and a spectrum of neuropsychiatric manifestations (FXAND) [2, 24, 25]. Intermediate alleles (45–54 CGG) do not cause disease, but may expand to a premutation [2].

Diagnosis requires repeat-targeted testing with methylation analysis, which no exome or genome replaces; size and methylation mosaicism explain phenotypic variability and may require testing of a second tissue [2, 20]. The risk of maternal expansion to a full mutation increases with premutation size — from around 4% below 60 repeats to 100% above 140 — and is modulated by AGG interruptions below 100 repeats; expansion to a full mutation never occurs in paternal transmission [2, 3, 27]. There is no approved therapy: phase 3 zatolmilast and transdermal cannabidiol trials, completed between 2025 and 2026, did not meet their primary endpoints, and treatment remains symptomatic, multidisciplinary and educational in FXS, symptomatic in FXTAS, with avoidance of neuroleptics, metoclopramide, anticholinergics and alcohol, and hormonal in FXPOI [2, 10, 56, 82]. Diagnosis of a single family member identifies three generations at risk and makes genetic counselling and cascade testing an essential part of treatment [2, 45]. Programmes ongoing in October 2026 — MRM-3379, SPG601, SRX246 — are in phase 2 or 2/3, and the only known phase 3 results are negative; symptomatic drugs used in FXS are authorised for the symptom rather than the syndrome, with relevant differences between US and European labels [85, 102]. The 2021 French national protocol adds to this synthesis the organisation of care in France (DéfiScience reference centres, ALD, coordination and transition), an age-based surveillance scheme, management of epileptic and behavioural emergencies with an emergency card, psychotropic doses lower than those of the labels and explicit divergences from the North American sources, particularly regarding systematic echocardiography, the frequency of ASD and the radiological criteria for FXTAS [1].

13. Bibliographic References

1. Haute Autorité de Santé; Centre de Référence Déficiences Intellectuelles de causes rares; Filière DéfiScience. Protocole National de Diagnostic et de Soins (PNDS): Syndrome de l'X fragile. Saint-Denis La Plaine: HAS; 2021 Aug 17 (text of December 2020, version of July 2021). Available from: https://www.has-sante.fr/jcms/p_3264435
2. Hunter JE, Berry-Kravis E, Hipp H, Todd PK. FMR1 Disorders. 1998 Jun 16 [updated 2024 May 16]. In: Adam MP, Bick S, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. PMID: 20301558. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1384/
3. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Fragile X syndrome (FRAX); p. 424–6.
4. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Fragile X syndrome; p. 198–201.
5. Orphanet. Fragile X syndrome (ORPHA:908) [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/908
6. Hagerman RJ, Protic D, Rajaratnam A, Salcedo-Arellano MJ, Aydin EY, Schneider A. Fragile X-associated neuropsychiatric disorders (FXAND). Front Psychiatry. 2018;9:564. doi:10.3389/fpsyt.2018.00564
7. Orphanet. Fragile X-associated tremor/ataxia syndrome (ORPHA:93256) [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/93256
8. Orphanet. Fragile X-associated primary ovarian insufficiency (ORPHA:642691) [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/642691
9. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Premature ovarian failure (POF); p. 778–9.
10. Panay N, Anderson RA, Bennie A, Cedars M, Davies M, Ee C, et al.; ESHRE, ASRM, CREWHIRL and IMS Guideline Group on POI. Evidence-based guideline: premature ovarian insufficiency. Hum Reprod Open. 2024;2024(4):hoae065. doi:10.1093/hropen/hoae065
11. Martin JP, Bell J. A pedigree of mental defect showing sex-linkage. J Neurol Psychiatry. 1943;6(3-4):154-7. doi:10.1136/jnnp.6.3-4.154
12. Lubs HA. A marker X chromosome. Am J Hum Genet. 1969;21(3):231-44. PMID: 5794013. Available from: https://pubmed.ncbi.nlm.nih.gov/5794013/
13. Verkerk AJ, Pieretti M, Sutcliffe JS, Fu YH, Kuhl DP, Pizzuti A, et al. Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome. Cell. 1991;65(5):905-14. doi:10.1016/0092-8674(91)90397-h
14. Fu YH, Kuhl DP, Pizzuti A, Pieretti M, Sutcliffe JS, Richards S, et al. Variation of the CGG repeat at the fragile X site results in genetic instability: resolution of the Sherman paradox. Cell. 1991;67(6):1047-58. doi:10.1016/0092-8674(91)90283-5
15. Hagerman RJ, Leehey M, Heinrichs W, Tassone F, Wilson R, Hills J, et al. Intention tremor, parkinsonism, and generalized brain atrophy in male carriers of fragile X. Neurology. 2001;57(1):127-30. doi:10.1212/wnl.57.1.127
16. Greco CM, Hagerman RJ, Tassone F, Chudley AE, Del Bigio MR, Jacquemont S, et al. Neuronal intranuclear inclusions in a new cerebellar tremor/ataxia syndrome among fragile X carriers. Brain. 2002;125(Pt 8):1760-71. doi:10.1093/brain/awf184
17. Jacquemont S, Hagerman RJ, Leehey M, Grigsby J, Zhang L, Brunberg JA, et al. Fragile X premutation tremor/ataxia syndrome: molecular, clinical, and neuroimaging correlates. Am J Hum Genet. 2003;72(4):869-78. doi:10.1086/374321
18. Sherman SL. Premature ovarian failure in the fragile X syndrome. Am J Med Genet. 2000;97(3):189-94. doi:10.1002/1096-8628(200023)97:3<189::AID-AJMG1036>3.0.CO;2-J. PMID: 11449487.
19. UniProt Consortium. Q06787 · FMR1_HUMAN — Fragile X messenger ribonucleoprotein 1 [Internet]. UniProtKB; 2026. Available from: https://www.uniprot.org/uniprotkb/Q06787/entry
20. Spector E, Behlmann A, Kronquist K, Rose NC, Lyon E, Reddi HV; ACMG Laboratory Quality Assurance Committee. Laboratory testing for fragile X, 2021 revision: a technical standard of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2021;23(5):799-812. doi:10.1038/s41436-021-01115-y
21. Nolin SL, Glicksman A, Tortora N, Allen E, Macpherson J, Mila M, et al. Expansions and contractions of the FMR1 CGG repeat in 5,508 transmissions of normal, intermediate, and premutation alleles. Am J Med Genet A. 2019;179(7):1148-56. doi:10.1002/ajmg.a.61165
22. Kraan CM, Godler DE, Amor DJ. Epigenetics of fragile X syndrome and fragile X-related disorders. Dev Med Child Neurol. 2019;61(2):121-7. doi:10.1111/dmcn.13985
23. Bear MF, Huber KM, Warren ST. The mGluR theory of fragile X mental retardation. Trends Neurosci. 2004;27(7):370-7. doi:10.1016/j.tins.2004.04.009
24. Hagerman RJ, Hagerman P. Fragile X-associated tremor/ataxia syndrome — features, mechanisms and management. Nat Rev Neurol. 2016;12(7):403-12. doi:10.1038/nrneurol.2016.82
25. Todd PK, Oh SY, Krans A, He F, Sellier C, Frazer M, et al. CGG repeat-associated translation mediates neurodegeneration in fragile X tremor ataxia syndrome. Neuron. 2013;78(3):440-55. doi:10.1016/j.neuron.2013.03.026
26. Nolin SL, Sah S, Glicksman A, Sherman SL, Allen E, Berry-Kravis E, et al. Fragile X AGG analysis provides new risk predictions for 45-69 repeat alleles. Am J Med Genet A. 2013;161A(4):771-8. doi:10.1002/ajmg.a.35833
27. Nolin SL, Glicksman A, Ersalesi N, Dobkin C, Brown WT, Cao R, et al. Fragile X full mutation expansions are inhibited by one or more AGG interruptions in premutation carriers. Genet Med. 2015;17(5):358-64. doi:10.1038/gim.2014.106
28. Nolin SL, Brown WT, Glicksman A, Houck GE Jr, Gargano AD, Sullivan A, et al. Expansion of the fragile X CGG repeat in females with premutation or intermediate alleles. Am J Hum Genet. 2003;72(2):454-64. doi:10.1086/367713
29. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Macrocephaly; p. 224–5.
30. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. Lesch-Nyhan disease and variants; p. 498.
31. Bailey DB Jr, Raspa M, Holiday D, Bishop E, Olmsted M. Functional skills of individuals with fragile X syndrome: a lifespan cross-sectional analysis. Am J Intellect Dev Disabil. 2009;114(4):289-303. doi:10.1352/1944-7558-114.4.289-303. PMID: 19642710.
32. Bartholomay KL, Lee CH, Bruno JL, Lightbody AA, Reiss AL. Closing the gender gap in fragile X syndrome: review of females with FXS and preliminary research findings. Brain Sci. 2019;9(1):11. doi:10.3390/brainsci9010011
33. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Ataxic adult; p. 70–2.
34. Chi MH, Bourgeois JA, Santos E, Kim K, Ponzini MD, Mendoza G, et al. Psychiatric manifestations in early to middle stages of fragile X-associated tremor-ataxia syndrome (FXTAS). J Neuropsychiatry Clin Neurosci. 2025;37(1):20-8. doi:10.1176/appi.neuropsych.20230215
35. Rodriguez-Revenga L, Madrigal I, Pagonabarraga J, Xunclà M, Badenas C, Kulisevsky J, et al. Penetrance of FMR1 premutation associated pathologies in fragile X syndrome families. Eur J Hum Genet. 2009;17(10):1359-62. doi:10.1038/ejhg.2009.51
36. Jacquemont S, Hagerman RJ, Leehey MA, Hall DA, Levine RA, Brunberg JA, et al. Penetrance of the fragile X-associated tremor/ataxia syndrome in a premutation carrier population. JAMA. 2004;291(4):460-9. doi:10.1001/jama.291.4.460
37. Hipp HS, Charen KH, Spencer JB, Allen EG, Sherman SL. Reproductive and gynecologic care of women with fragile X primary ovarian insufficiency (FXPOI). Menopause. 2016;23(9):993-9. doi:10.1097/GME.0000000000000658
38. Fink DA, Nelson LM, Pyeritz R, Johnson J, Sherman SL, Cohen Y, et al. Fragile X associated primary ovarian insufficiency (FXPOI): case report and literature review. Front Genet. 2018;9:529. doi:10.3389/fgene.2018.00529
39. Allen EG, Sullivan AK, Marcus M, Small C, Dominguez C, Epstein MP, et al. Examination of reproductive aging milestones among women who carry the FMR1 premutation. Hum Reprod. 2007;22(8):2142-52. doi:10.1093/humrep/dem148
40. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Intellectual disability; p. 188–92.
41. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Intellectual disability with apparent X-linked inheritance; p. 194–7.
42. Bailey DB Jr, Skinner D, Sparkman KL. Discovering fragile X syndrome: family experiences and perceptions. Pediatrics. 2003;111(2):407-16. doi:10.1542/peds.111.2.407. PMID: 12563071.
43. Berry-Kravis E, Abrams L, Coffey SM, Hall DA, Greco C, Gane LW, et al. Fragile X-associated tremor/ataxia syndrome: clinical features, genetics, and testing guidelines. Mov Disord. 2007;22(14):2018-30. doi:10.1002/mds.21493
44. Biancalana V, Glaeser D, McQuaid S, Steinbach P. EMQN best practice guidelines for the molecular genetic testing and reporting of fragile X syndrome and other fragile X-associated disorders. Eur J Hum Genet. 2015;23(4):417-25. doi:10.1038/ejhg.2014.185
45. Finucane B, Abrams L, Cronister A, Archibald AD, Bennett RL, McConkie-Rosell A. Genetic counseling and testing for FMR1 gene mutations: practice guidelines of the National Society of Genetic Counselors. J Genet Couns. 2012;21(6):752-60. doi:10.1007/s10897-012-9524-8
46. Mbachu CNP, Eseigbe EE, Mbachu II, Eleje GU, Udigwe IB, et al. Advancing mental health and well-being of Nigerian children through public health screening for Fragile X disorders (CHAMP-FX): protocol of a prospective multicentre screening study with longitudinal follow-up. PLoS One. 2026;21(8):e0355384. doi:10.1371/journal.pone.0355384
47. Morales J, Pujar S, Loveland JE, Astashyn A, Bennett R, Berry A, et al. A joint NCBI and EMBL-EBI transcript set for clinical genomics and research. Nature. 2022;604(7905):310-5. doi:10.1038/s41586-022-04558-8. PMID: 35388217. MANE GRCh38 v1.5 file: https://ftp.ncbi.nlm.nih.gov/refseq/MANE/
48. National Center for Biotechnology Information. ClinVar: VCV000623467 (NM_002024.6(FMR1):c.-128GGC[55_200]), VCV000623468 (c.-128GGC[200_?]), VCV000009969 (c.911T>A, p.Ile304Asn) and VCV001701686 (c.413G>A, p.Arg138Gln) [Internet]. Bethesda (MD): National Library of Medicine; [accessed 2026 Oct 6]. Available from: https://www.ncbi.nlm.nih.gov/clinvar/?term=FMR1%5Bgene%5D
49. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-24. doi:10.1038/gim.2015.30. PMID: 25741868.
50. Riggs ER, Andersen EF, Cherry AM, Kantarci S, Kearney H, Patel A, et al. Technical standards for the interpretation and reporting of constitutional copy-number variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics (ACMG) and the Clinical Genome Resource (ClinGen). Genet Med. 2020;22(2):245-57. doi:10.1038/s41436-019-0686-8. PMID: 31690835.
51. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Parkinson's disease; p. 516.
52. Ishiura H, Shibata S, Yoshimura J, Suzuki Y, Qu W, Doi K, et al. Noncoding CGG repeat expansions in neuronal intranuclear inclusion disease, oculopharyngodistal myopathy and an overlapping disease. Nat Genet. 2019;51(8):1222-32. doi:10.1038/s41588-019-0458-z
53. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. X-autosome translocations; p. 700–1.
54. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Blepharophimosis-ptosis-epicanthus inversus syndrome; p. 326–7.
55. Ribeiro EM, editor. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. Síndrome do triplo X; p. 37–9.
56. FRAXA Research Foundation. Shionogi update on zatolmilast Fragile X clinical trial findings [Internet]. Newburyport (MA): FRAXA; 2026 May 19 [cited 2026 Oct 3]. Available from: https://www.fraxa.org/shionogi-update-on-zatolmilast-fragile-x-clinical-trial-findings/
57. Hersh JH, Saul RA; Committee on Genetics. Health supervision for children with fragile X syndrome. Pediatrics. 2011;127(5):994-1006. doi:10.1542/peds.2010-3500
58. Sandoz Inc. RITALIN (methylphenidate hydrochloride) tablets: prescribing information [revised 2025 Feb]. Silver Spring (MD): U.S. Food and Drug Administration; 2025. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=d6fb2750-cdab-4749-ba0d-7534840a5892
59. Janssen Pharmaceuticals. CONCERTA (methylphenidate hydrochloride) extended-release tablets: prescribing information [revised 2026 Apr]. Silver Spring (MD): U.S. Food and Drug Administration; 2026. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=1a88218c-5b18-4220-8f56-526de1a276cd
60. INFARMED. Concerta (metilfenidato) comprimidos de libertação prolongada: resumo das características do medicamento. Lisboa: INFARMED; [accessed 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=34285&tipo_doc=rcm
61. Eli Lilly and Company. STRATTERA (atomoxetine) capsules: prescribing information [revised 2026 Jun]. Silver Spring (MD): U.S. Food and Drug Administration; 2026. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=309de576-c318-404a-bc15-660c2b1876fb
62. Generis Farmacêutica. Atomoxetina Generis (atomoxetina) cápsulas: resumo das características do medicamento [revised 2024 Nov]. Lisboa: INFARMED; 2024 [accessed 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=619907&tipo_doc=rcm
63. Takeda Pharmaceuticals America. INTUNIV (guanfacine) extended-release tablets: prescribing information [revised 2025 Jun]. Silver Spring (MD): U.S. Food and Drug Administration; 2025. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=b972af81-3a37-40be-9fe1-3ddf59852528
64. European Medicines Agency. Intuniv (guanfacine) prolonged-release tablets: EPAR — product information (summary of product characteristics). Amsterdam: EMA; [accessed 2026 Oct 26]. Available from: https://www.ema.europa.eu/en/documents/product-information/intuniv-epar-product-information_en.pdf
65. Actavis Pharma. Clonidine hydrochloride extended-release tablets (ADHD): prescribing information [revised 2023 Oct]. Silver Spring (MD): U.S. Food and Drug Administration; 2023. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0100c70d-7fde-46a1-8374-940550a27e43
66. Greiss Hess L, Fitzpatrick SE, Nguyen DV, Chen Y, Gaul KN, Schneider A, et al. A randomized, double-blind, placebo-controlled trial of low-dose sertraline in young children with fragile X syndrome. J Dev Behav Pediatr. 2016;37(8):619-28. doi:10.1097/DBP.0000000000000334
67. Viatris Specialty. ZOLOFT (sertraline hydrochloride) tablets and oral solution: prescribing information [revised 2023 Aug]. Silver Spring (MD): U.S. Food and Drug Administration; 2023. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=fda754f6-d0f3-4dce-a17a-927d64f912f7
68. Viatris. Zoloft (sertralina) comprimidos revestidos por película: resumo das características do medicamento. Lisboa: INFARMED; [accessed 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=9485&tipo_doc=rcm
69. Otsuka America Pharmaceutical. ABILIFY (aripiprazole) tablets and oral solution: prescribing information [revised 2025 Jan]. Silver Spring (MD): U.S. Food and Drug Administration; 2025. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=c040bd1d-45b7-49f2-93ea-aed7220b30ac
70. Erickson CA, Stigler KA, Wink LK, Mullett JE, Kohn A, Posey DJ, et al. A prospective open-label study of aripiprazole in fragile X syndrome. Psychopharmacology (Berl). 2011;216(1):85-90. doi:10.1007/s00213-011-2194-7
71. Janssen Pharmaceuticals. RISPERDAL (risperidone) tablets, oral solution and orally disintegrating tablets: prescribing information [revised 2026 May]. Silver Spring (MD): U.S. Food and Drug Administration; 2026. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7e117c7e-02fc-4343-92a1-230061dfc5e0
72. European Medicines Agency. Risperdal and associated names: Article 30 referral (CHMP opinion of 24 July 2008; European Commission decision of 7 October 2008) and harmonised risperidone SmPC, section 4.1. Amsterdam: EMA; 2008. Available from: https://www.ema.europa.eu/en/medicines/human/referrals/risperdal
73. Janssen Farmacêutica Portugal. Risperdal (risperidona) comprimidos revestidos por película e solução oral: resumo das características do medicamento. Lisboa: INFARMED; [accessed 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=7607&tipo_doc=rcm
74. European Medicines Agency. Slenyto (melatonin) prolonged-release tablets: EPAR — product information (summary of product characteristics), sections 4.1 and 4.2. EMEA/H/C/004425; first authorisation 20 Sep 2018; extension to neurogenetic diseases (II/0025) 26 Aug 2024; ADHD indication (II/0028) 17 Mar 2025; product information updated on 10 Jun 2026 [accessed 2026 Nov 7]. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/slenyto
75. European Medicines Agency. Valproate and related substances: Article 31 referral — new measures to avoid valproate exposure in pregnancy (CMDh position of 21 March 2018; European Commission decision) [Internet]. Amsterdam: EMA; 2018 [accessed 2026 Nov 2]. Available from: https://www.ema.europa.eu/en/medicines/human/referrals/valproate-related-substances-0
76. Consoli A, Çabal Berthoumieu S, Raffin M, Thuilleaux D, Poitou C, Coupaye M, et al. Effect of topiramate on eating behaviours in Prader-Willi syndrome: TOPRADER double-blind randomised placebo-controlled study. Transl Psychiatry. 2019;9(1):274. doi:10.1038/s41398-019-0597-0. PMID: 31685813
77. Berry-Kravis E, Des Portes V, Hagerman R, Jacquemont S, Charles P, Visootsak J, et al. Mavoglurant in fragile X syndrome: results of two randomized, double-blind, placebo-controlled trials. Sci Transl Med. 2016;8(321):321ra5. doi:10.1126/scitranslmed.aab4109
78. Berry-Kravis EM, Harnett MD, Reines SA, Reese MA, Ethridge LE, Outterson AH, et al. Inhibition of phosphodiesterase-4D in adults with fragile X syndrome: a randomized, placebo-controlled, phase 2 clinical trial. Nat Med. 2021;27(5):862-70. doi:10.1038/s41591-021-01321-w
79. ClinicalTrials.gov. A study of BPN14770 in male adults (aged 18 to 45) with fragile X syndrome (NCT05358886); A randomized study of BPN14770 in male adolescents (aged 9 to < 18 years) with fragile X syndrome (NCT05163808) [Internet]. Bethesda (MD): National Library of Medicine; 2026 [cited 2026 Oct 3]. Available from: https://clinicaltrials.gov/study/NCT05358886
80. FRAXA Research Foundation. Shionogi to end zatolmilast open-label extension for adolescents [Internet]. Newburyport (MA): FRAXA; 2026 Oct 1 [cited 2026 Oct 3]. Available from: https://www.fraxa.org/shionogi-to-end-zatolmilast-open-label-extension-for-adolescents/
81. Berry-Kravis E, Hagerman R, Budimirovic D, Erickson C, Heussler H, Tartaglia N, et al. A randomized, controlled trial of ZYN002 cannabidiol transdermal gel in children and adolescents with fragile X syndrome (CONNECT-FX). J Neurodev Disord. 2022;14(1):56. doi:10.1186/s11689-022-09466-6
82. FRAXA Research Foundation. RECONNECT trial results: ZYN002 does not meet primary endpoint in Fragile X [Internet]. Newburyport (MA): FRAXA; 2025 Sep 24 [cited 2026 Oct 3]. Available from: https://www.fraxa.org/reconnect-trial-results-zyn002-does-not-meet-primary-endpoint-in-fragile-x/
83. Zhu Y, Li D, Hu C, Tian Y, Lu P, Hagerman RJ, et al. Effects of metformin on children with fragile X syndrome: a randomized, double-blind, placebo-controlled trial. Mol Autism. 2025;16(1):57. doi:10.1186/s13229-025-00691-z
84. FRAXA Research Foundation. Fragile X treatment pipeline: current and emerging clinical trials [Internet]. Newburyport (MA): FRAXA; 2026 Jun 16 [cited 2026 Oct 3]. Available from: https://www.fraxa.org/fragile-x-treatment-pipeline-current-and-emerging-clinical-trials/
85. ClinicalTrials.gov. Records NCT07209462 (MRM-3379, BLOOM), NCT07439510 (SPG601, phases 2/3), NCT07328529 (SRX246), NCT03862950 (metformin, University of Alberta), NCT05367960 (open-label extension of BPN14770) and NCT04977986 (ZYN002, RECONNECT) [Internet]. Bethesda (MD): National Library of Medicine; [accessed 2026 Oct 6]. Available from: https://clinicaltrials.gov/search?cond=Fragile%20X%20Syndrome
86. Pedapati EV, Vanderklish PW, Sarraf ST, Gargosky S, Nelson M, Richter S, et al. SPG601-associated modulation of resting-state EEG and improvement in executive function in a fragile X syndrome randomized controlled crossover study. Sci Rep. 2026;16(1):11705. doi:10.1038/s41598-026-46928-6. PMID: 41946887.
87. Zeier Z, Kumar A, Bodhinathan K, Feller JA, Foster TC, Bloom DC. Fragile X mental retardation protein replacement restores hippocampal synaptic function in a mouse model of fragile X syndrome. Gene Ther. 2009;16(9):1122-9. doi:10.1038/gt.2009.83. PMID: 19571888
88. Hampson DR, Hooper AWM, Niibori Y. The application of adeno-associated viral vector gene therapy to the treatment of fragile X syndrome. Brain Sci. 2019;9(2):32. doi:10.3390/brainsci9020032. PMID: 30717399
89. Liu XS, Wu H, Krzisch M, Wu X, Graef J, Muffat J, et al. Rescue of fragile X syndrome neurons by DNA methylation editing of the FMR1 gene. Cell. 2018;172(5):979-92.e6. doi:10.1016/j.cell.2018.01.012. PMID: 29456084
90. Lee B, Lee K, Panda S, Gonzales-Rojas R, Chong A, Bugay V, et al. Nanoparticle delivery of CRISPR into the brain rescues a mouse model of fragile X syndrome from exaggerated repetitive behaviours. Nat Biomed Eng. 2018;2(7):497-507. doi:10.1038/s41551-018-0252-8. PMID: 30948824
91. Dash D, Bruno V, Schwingenschuh P, Lyons KE, Tan EK, Testa CM, et al. Update on medical treatments for essential tremor: an International Parkinson and Movement Disorder Society evidence-based medicine review. Mov Disord. 2026;41(4):815-25. doi:10.1002/mds.70184. PMID: 41556478.
92. Mysoline (primidona) 250 mg comprimidos: resumo das características do medicamento [text revised 2026 Apr]. Lisboa: INFARMED; 2026 [accessed 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/
93. NCS HealthCare of KY (Vangard Labs). Propranolol hydrochloride tablets: prescribing information [revised 2024 Aug]. Silver Spring (MD): U.S. Food and Drug Administration; 2024. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=013a485a-4227-46ae-b19f-c550302fcb96
94. Viatris Specialty. NEURONTIN (gabapentin) capsules, tablets and oral solution: prescribing information [revised 2026 May]. Silver Spring (MD): U.S. Food and Drug Administration; 2026. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=97935fd9-1d4a-43b6-a5d9-de994591187b
95. Parke-Davis (Pfizer). LYRICA (pregabalin) capsules and oral solution: prescribing information [revised 2020 Jun]. Silver Spring (MD): U.S. Food and Drug Administration; 2020. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=60185c88-ecfd-46f9-adb9-b97c6b00a553
96. Golden State Medical Supply. Duloxetine delayed-release capsules: prescribing information [revised 2026 Aug]. Silver Spring (MD): U.S. Food and Drug Administration; 2026. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=00628e5e-4c5b-2573-e063-6294a90a0e3b
97. Direct_Rx. Estradiol tablets, USP (ANDA040275): prescribing information [revised 2025 Jan]. Silver Spring (MD): U.S. Food and Drug Administration; 2025 [accessed 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=01789bad-4be4-ae22-e063-6294a90af392
98. A-S Medication Solutions. Estradiol transdermal system, continuous delivery (once-weekly) (ANDA075182): prescribing information [revised 2026 Jun]. Silver Spring (MD): U.S. Food and Drug Administration; 2026 [accessed 2026 Oct 26]. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=1c38f0d0-a360-4b66-bc0e-7d54e92465a1
99. Zumenon (estradiol) 2 mg comprimido revestido por película: resumo das características do medicamento. Lisboa: INFARMED; [accessed 2026 Oct 26]. Available from: https://extranet.infarmed.pt/INFOMED-fo/
100. Bourgeois JA, Schneider A, Klusek J, Christensen TR, Levin E, Gardner K, et al. Comprehensive, multidisciplinary care for fragile X-associated tremor/ataxia syndrome. Front Neurol. 2026;17:1746002. doi:10.3389/fneur.2026.1746002. PMID: 41940304.
101. Berry-Kravis E, Hagerman R, Visootsak J, Budimirovic D, Kaufmann WE, Cherubini M, et al. Arbaclofen in fragile X syndrome: results of phase 3 trials. J Neurodev Disord. 2017;9:3. doi:10.1186/s11689-016-9181-6. PMID: 28616094.
102. Van Steenbergen A, Kaur M, Rubanarayana K, Bolduc F. Clinical trial designs for rare disorders: a scoping review of the effectiveness of pharmacologic interventions in fragile X syndrome. Neurol Genet. 2026;12(2):e200373. doi:10.1212/NXG.0000000000200373. PMID: 41890791.
103. U.S. Food and Drug Administration. Drugs@FDA: NDA 010187 (Ritalin, 1955-12-05), NDA 021121 (Concerta, 2000-08-01), NDA 021411 (Strattera, 2002-11-26), NDA 022037 (Intuniv, 2009-09-02), NDA 019839 (Zoloft, 1991-12-30), NDA 021436 (Abilify, 2002-11-15; efficacy supplement dated 2009-11-19) and NDA 020272 (Risperdal, 1993-12-29; efficacy supplement dated 2006-10-06) [Internet]. Silver Spring (MD): FDA; [accessed 2026 Oct 6]. Available from: https://www.accessdata.fda.gov/scripts/cder/daf/
104. European Medicines Agency. Medicines data file (EPAR): Intuniv (authorisation 2015-09-17), Abilify (2004-06-04), Slenyto (2018-09-20) and Tuzulby (2025-02-28) — current indications [Internet]. Amsterdam: EMA; [accessed 2026 Oct 6]. Available from: https://www.ema.europa.eu/en/medicines/download-medicine-data
105. European Medicines Agency. Buccolam (midazolam) oromucosal solution: EPAR — product information (summary of product characteristics; first authorisation on 5 September 2011). Amsterdam: EMA; [accessed 2026 Nov 2]. Available from: https://www.ema.europa.eu/en/documents/product-information/buccolam-epar-product-information_en.pdf
106. Zydus Pharmaceuticals USA. Diazepam rectal gel (ANDA217468): prescribing information [revised 2026 Apr]. Silver Spring (MD): U.S. Food and Drug Administration; 2026 [accessed 2026 Nov 2]. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=652714c7-25c4-4d05-979d-8eee3b6f2f4d
107. European Medicines Agency. Hydroxyzine-containing medicinal products: Article 31 referral — CMDh final position (27 March 2015) [Internet]. Amsterdam: EMA; 2015 [accessed 2026 Nov 2]. Available from: https://www.ema.europa.eu/en/medicines/human/referrals/hydroxyzine
108. Nolin SL, Glicksman A, Ding X, Ersalesi N, Brown WT, Sherman SL, et al. Fragile X analysis of 1112 prenatal samples from 1991 to 2010. Prenat Diagn. 2011;31(10):925-31. doi:10.1002/pd.2815
109. Gregg AR, Aarabi M, Klugman S, Leach NT, Bashford MT, Goldwaser T, et al. Screening for autosomal recessive and X-linked conditions during pregnancy and preconception: a practice resource of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2021;23(10):1793-806. doi:10.1038/s41436-021-01203-z
110. European Reference Networks: ERN ITHACA (https://ern-ithaca.eu), ERN-RND (https://www.ern-rnd.eu) and Endo-ERN (https://endo-ern.eu) [Internet]. [accessed 2026 Oct 6].
111. National Fragile X Foundation (https://fragilex.org); FRAXA Research Foundation (https://www.fraxa.org); Fragile X Society (https://www.fragilex.org.uk); Instituto Buko Kaesemodel — Programa Eu Digo X (https://www.xfragil.org.br) [Internet]. [accessed 2026 Oct 6].
112. Coffee B, Keith K, Albizua I, Malone T, Mowrey J, Sherman SL, et al. Incidence of fragile X syndrome by newborn screening for methylated FMR1 DNA. Am J Hum Genet. 2009;85(4):503-14. doi:10.1016/j.ajhg.2009.09.007
113. Hunter J, Rivero-Arias O, Angelov A, Kim E, Fotheringham I, Leal J. Epidemiology of fragile X syndrome: a systematic review and meta-analysis. Am J Med Genet A. 2014;164A(7):1648-58. doi:10.1002/ajmg.a.36511
114. Maenner MJ, Baker MW, Broman KW, Tian J, Barnes JK, Atkins A, et al. FMR1 CGG expansions: prevalence and sex ratios. Am J Med Genet B Neuropsychiatr Genet. 2013;162B(5):466-73. doi:10.1002/ajmg.b.32176. PMID: 23740716.
115. Hantash FM, Goos DM, Crossley B, Anderson B, Zhang K, Sun W, et al. FMR1 premutation carrier frequency in patients undergoing routine population-based carrier screening: insights into the prevalence of fragile X syndrome, fragile X-associated tremor/ataxia syndrome, and fragile X-associated primary ovarian insufficiency in the United States. Genet Med. 2011;13(1):39-45. doi:10.1097/GIM.0b013e3181fa9fad
116. Yrigollen CM, Sweha S, Durbin-Johnson B, Zhou L, Berry-Kravis E, Fernandez-Carvajal I, et al. Distribution of AGG interruption patterns within nine world populations. Intractable Rare Dis Res. 2014;3(4):153-61. doi:10.5582/irdr.2014.01028
117. Peprah E. Fragile X syndrome: the FMR1 CGG repeat distribution among world populations. Ann Hum Genet. 2012;76(2):178-91. doi:10.1111/j.1469-1809.2011.00694.x
118. Ramírez-Cheyne J, López D, Payán-Gómez C, Arcos-Burgos M, Saldarriaga W. Fragile X syndrome in the largest world clustering. I. Genetic epidemiology and founder effect outline. Am J Med Genet A. 2024;194(5):e63523. doi:10.1002/ajmg.a.63523
119. Neves AR, Pais AS, Ferreira SI, Ramos V, Carvalho MJ, Estevinho A, et al. Prevalence of cytogenetic abnormalities and FMR1 gene premutation in a Portuguese population with premature ovarian insufficiency. Acta Med Port. 2021;34(9):580-5. doi:10.20344/amp.13490
120. Soares CA, Tkachenko N, Vale-Fernandes E, Barreiro M, Abreu M, Reis CF. Genetic counseling and carrier screening in candidates for gamete donation at a Portuguese center. JBRA Assist Reprod. 2023;27(2):180-4. doi:10.5935/1518-0557.20220012
121. Schmidmajer A, Páez-García S, Martínez S, Ramírez-Triana J, O'Hara-Veintimilla K, Ricaurte-Fajardo A, et al. FXTAS and the spectrum of FMR1 premutation-associated phenotypes in Latin America: a scoping review. Mov Disord Clin Pract. 2026;13(9):2045-58. doi:10.1002/mdc3.70617
122. Napitupulu CLM, Tengkawan J, Utari A, Winarni TI, Sihombing NRB. Phenotypic variability of fragile-X syndrome in Asian population: a systematic review. Intractable Rare Dis Res. 2026;15(3):232-46. doi:10.5582/irdr.2026.01037