Ataxia-Telangiectasia

Louis-Bar syndrome; biallelic ATM gene disease — classic and variant forms, and cancer predisposition in heterozygotes.

Literature and regulatory information updated: 6 October 2026.

0. Initial Reference Table

1. Introduction

Ataxia-telangiectasia is an autosomal recessive multisystem disorder combining, in the same patient, childhood-onset cerebellar neurodegeneration, oculocutaneous telangiectasias, variably expressed combined immunodeficiency, progressive lung disease, predisposition to malignancies — particularly lymphoid malignancies — and hypersensitivity to ionising radiation [1, 2]. It belongs to the group of DNA damage response disorders, in which failure to maintain genomic integrity accelerates the accumulation of genetic alterations and favours neoplastic transformation, and in which defective repair of double-strand breaks also interferes with V(D)J recombination of immunoglobulin and antigen receptor genes, explaining the coexistence of immunodeficiency and chromosomal instability [5].

The history of the disease spans almost a century. The first clinical descriptions are attributed to Syllaba and Henner (1926) and Denise Louis-Bar (1941), whose name the syndrome retained as an eponym [7]. The modern founding paper is that of Elena Boder and Robert P. Sedgwick, published in 1958 in Pediatrics, which described a “familial syndrome of progressive cerebellar ataxia, oculocutaneous telangiectasia and frequent pulmonary infection” and gave it its current name [7, 21]. Subsequent laboratory milestones remain useful in practice: Waldmann and McIntire demonstrated elevated serum AFP in these patients in 1972 [22]; Taylor and colleagues showed in 1975 that cells from patients with A-T are abnormally sensitive to ionising radiation, an observation reinforced by the severe reactions described after radiotherapy [23]; Gatti and colleagues mapped the locus to 11q22-23 by linkage analysis (1988) [24]; and Savitsky and colleagues cloned a single causative gene in 1995, whose product resembles phosphatidylinositol-3 kinases (PI3K) and which was named ATM [25].

Until the gene was identified, A-T was defined by clinical criteria and cellular complementation groups; molecular genetics demonstrated that it is a single genetic entity with a continuous phenotypic spectrum. Classic A-T, without ATM kinase activity, is now distinguished from variant A-T, in which some residual protein activity persists; this distinction does not define different diseases, but has prognostic value because it determines the rate of neurological deterioration, severity of immunodeficiency, respiratory risk, age at first malignancy and life expectancy [1, 26]. Variant A-T has its own Orphanet code (ORPHA:370109) and is described as a persistent, later-onset combined dystonia, often without telangiectasias or immunodeficiency [3].

Heterozygotes for pathogenic ATM variants, who do not develop A-T, have a moderately increased risk of breast, pancreatic, prostate and stomach cancer; this dominant condition has had its own GeneReviews entry since June 2026 [8]. In this article, predisposition in heterozygotes is addressed only insofar as it concerns counselling families of patients with A-T.

The year 2026 changed the therapeutic landscape. The phase 3 NEAT trial, which evaluated dexamethasone encapsulated in autologous erythrocytes, did not meet its primary endpoint and the sponsor abandoned the programme; by contrast, levacetylleucine demonstrated benefit in a placebo-controlled crossover phase 3 trial and received the first authorisation of a medicine for A-T ataxia from the US Food and Drug Administration (FDA) in September 2026 [27–30].

2. Molecular Genetics and Pathophysiology

2.1 The ATM gene and ATM protein

The ATM gene is located at 11q22.3 and the current reference transcript (MANE Select) is NM_000051.4, corresponding to Ensembl transcript ENST00000675843 and comprising 63 exons; the encoded protein, NP_000042.3, contains 3056 amino acids and weighs approximately 350 kDa [4]. Some classic sources describe the gene as having 66 exons (four non-coding and 62 coding) across approximately 150 kb of genomic DNA; the difference reflects annotation of the 5′ non-coding exons and does not alter the coding sequence numbering used in HGVS nomenclature [5, 7].

The ATM protein is a serine/threonine kinase of the PI3K-related kinase (PIKK) family that acts as the principal sensor and coordinator of the response to DNA double-strand breaks. Activated by the MRE11-RAD50-NBN (MRN) complex at the break site, it phosphorylates hundreds of substrates at serine/threonine-glutamine motifs, including p53, CHK2, BRCA1, NBN (nibrin), MRE11 and histone H2AX at serine 139, triggering cell cycle checkpoint arrest, DNA repair or apoptosis [1, 4, 8]. ATM is also required for repair of breaks within heterochromatin and participates in allelic exclusion of immunoglobulin heavy-chain genes during recombination mediated by RAG enzymes [4, 5].

Beyond the nucleus, ATM participates in RNA metabolism, transcriptional regulation, redox and calcium homeostasis, and mitochondrial and peroxisomal function; absence of the protein therefore disrupts multiple cellular pathways, helping to explain the multisystem nature of the disease [1]. Activation of ATM by oxidative stress, independently of DNA breaks, and its links with the mTORC1 pathway and insulin metabolism offer plausible explanations for the insulin resistance, dyslipidaemia and hepatic steatosis observed in older patients [1, 7].

2.2 Tissue-specific pathophysiology

In the nervous system, abnormalities predominate in the cerebellum, with Purkinje cell loss and atrophy of the vermis and hemispheres, but other regions and the peripheral nervous system become progressively involved; the exact mechanisms of neurodegeneration remain uncertain, with proposed mechanisms including defective DNA repair in postmitotic neurons, oxidative stress and mitochondrial dysfunction with nicotinamide adenine dinucleotide (NAD+) deficiency [1, 7, 31]. This last hypothesis supported trials of NAD+ precursors, described in section 7 [32, 33].

In the immune system, defective V(D)J recombination and class-switch recombination compromise production of a diverse B- and T-lymphocyte repertoire; thymic hypoplasia causes T-cell lymphopenia, preferentially reducing naïve CD4+ cells [1]. The same repair failure during recombination explains the non-random translocations and inversions involving T-cell receptor and immunoglobulin loci on chromosomes 7 and 14, visible on T-lymphocyte karyotyping, and provides the basis for lymphomagenesis [5, 6]. Radiosensitivity results from the inability to signal and repair breaks induced by ionising radiation and radiomimetic drugs; in patients' cells, chromosomal breaks induced in vitro may be up to ten times more numerous than in controls [1, 5].

2.3 Allelic spectrum and disease mechanism

The disease mechanism is loss of function. Pathogenic variants are distributed throughout the gene, without hotspots, and are usually truncating (nonsense, frameshift, splice-altering or initiation codon variants), leading to nonsense-mediated mRNA decay and absence of protein; some missense variants or in-frame deletions produce a stable but inactive protein [1, 5]. In variant A-T, variants are predominantly missense with residual activity or “leaky” splicing variants, in which some of the transcript is processed normally [1, 8].

Sequencing of exons and junctions detects most variants, but a relevant proportion lies in deep intronic regions that panels and exome sequencing do not interrogate: GeneReviews identifies, among others, c.5763-1050A>G, c.2639-384A>G, c.2839-579_2839-576delAAGT, c.3994-159A>G and c.1803-270T>G [8]. In the A-T Children's Project cohort, comprising predominantly North American patients, whole-genome sequencing of 235 individuals from 209 families enabled a complete molecular diagnosis in almost all; 9% had variants “probably” and 6% “possibly” amenable to correction with splice-modulating antisense oligonucleotides, mostly in deep intronic regions [34].

2.4 Genotype–phenotype correlations and nomenclature

As a rule, the combination of two null alleles produces the classic phenotype, whereas missense and leaky splicing variants are associated with variant A-T; within each family the phenotype tends to be similar [1, 26]. The presence of ATM protein with residual kinase activity correlates with a milder phenotype and, in a series of 296 patients from the British Isles and the Netherlands, provided marked protection against childhood lymphoid tumours (incidence rate ratio of 0.15 for tumours diagnosed by age 16 years) [26, 35]. The following table brings together variants with clinical or population significance described in the sources, with nomenclature normalised to the MANE transcript (confirmed in VariantValidator) and ClinVar classification where consulted.

Variant c.7865C>T illustrates an important interpretative point: the in silico prediction of the protein effect (alanine to valine, considered benign) is misleading, because the pathogenic effect results from creation of a splice donor site that truncates the exon by 64 bp [34]. Conversely, identification of two variants of uncertain significance, or one pathogenic variant and one variant of uncertain significance, neither establishes nor excludes the diagnosis; a clinical response to treatment does not convert a variant of uncertain significance into a pathogenic variant [1].

2.5 Allelic disorders and somatic alterations

Heterozygous germline pathogenic variants are associated with an increased risk of female breast, exocrine pancreatic, prostate and gastric cancer, with modest, unconfirmed associations for ovarian and colon cancer and melanoma; variant c.7271T>G confers a higher risk than other variants and warrants individualised management [1, 8]. Somatic ATM variants occur in sporadic tumours — B-cell chronic lymphocytic leukaemia, mantle cell lymphoma and T-cell prolymphocytic leukaemia, among others — and do not imply hereditary predisposition [1, 43]. In single-allele carriers, detection of low-level mosaicism in leucocytes suggests clonal haematopoiesis rather than a constitutional variant [8].

3. Clinical Spectrum and Manifestations by System

3.1 Overview: classic A-T and variant A-T

GeneReviews describes A-T as a continuum and systematises the differences between its two extremes; the following table reproduces that comparison and adds data from recent cohorts [1].

Adapted from GeneReviews, Table 2, and variant A-T cohorts [1, 44, 45]. “Always present”, “usual”, “occasional” and “rare” translate the original ++, +, ±, − scale.

3.2 Nervous system in classic A-T

In most children, cerebellar ataxia becomes apparent when they begin to sit and walk: the child appears normal until walking begins, at the usual age, but walking does not improve and remains unstable, with a broad base, swaying when standing, and backward positioning of the arms and hands reflecting dystonia [1, 5, 21]. In the early years, natural acquisition of motor skills may mask progression, giving a false impression of a static disorder [1]. The Oxford Desk Reference describes ataxia as evident from approximately age 1 year [46]; the GeneReviews overview of hereditary ataxias includes A-T among ataxias with onset before age five years [9].

Most children develop extrapyramidal involvement — dystonia, choreoathetosis, myoclonus and tremor — and parkinsonism may emerge in the second or third decades [1]. Involuntary hand movements hinder fine motor tasks in late childhood [5]. From approximately age 8 years, signs of progressive sensorimotor peripheral neuropathy appear, with hyporeflexia or areflexia, distal weakness and atrophy, and reduced vibration sensation; by approximately age 10 years, most require a wheelchair [1]. Dysarthria accompanies progression in all patients, although oral communication generally remains possible [1, 47]. Dysphagia, frequent in adolescence and young adulthood, makes meals tiring and prolonged, contributing to reduced intake and chronic aspiration [1, 48].

Oculomotor abnormalities include gaze-evoked nystagmus, hypometric saccades, saccadic intrusions, impaired convergence and accommodation, and vestibulo-ocular reflex abnormalities, with fixation difficulties that compromise reading; oculomotor apraxia, with compensatory head movements, and strabismus generally appear several years after onset [1, 5]. Moderate to severe intellectual disability is not part of the phenotype, and its presence should prompt investigation of another cause; however, cerebellar and extracerebellar network dysfunction causes deficits in intellectual functioning, non-verbal memory, abstract verbal reasoning, calculation and executive function, resembling cerebellar cognitive affective syndrome, from the end of the first decade [1, 49].

3.3 Nervous system in variant A-T

In variant A-T, the first signs emerge from childhood to adulthood, but 81% of the 57 patients in the largest published cohort had symptoms before age 10 years; diagnostic delay exceeded 10 years in 68% and 20 years in one third of probands, and 43% were still walking 20 years after onset [44]. Ataxia may be absent, extrapyramidal movement disorders — particularly dystonia and dystonic tremor — dominate, and patients with exclusively extrapyramidal presentations have a more benign course; chorea and parkinsonism are rare, and most have axonal sensorimotor polyneuropathy [1, 44]. An analysis of 105 patients with mild neurological phenotypes (14 Dutch patients and 91 from the literature) identified six trajectories: childhood extrapyramidal signs with later ataxia (18 patients), childhood ataxia with later extrapyramidal signs (35), dystonia without ataxia (23), isolated ataxia (22), prominent muscle weakness (2) and adult extrapyramidal signs with subsequent anterior horn disease (5) [45]. A patient presenting with cancer at a young age and an apparently static motor disorder interpreted as cerebral palsy should raise suspicion of variant A-T [1].

3.4 Immunodeficiency and infections

Immunodeficiency is highly variable and may be humoral, cellular or combined; it generally remains stable over time [1]. The Oxford Desk Reference reports immunodeficiency in 80% of patients, typically manifesting as sinusitis, bronchitis or pneumonia [5]. The most common humoral abnormalities are selective IgA deficiency, hypogammaglobulinaemia and IgG subclass deficiency (particularly IgG2 and IgG4), with reduced responses to polysaccharide antigens, especially when IgA and IgG2 deficiencies coexist [1, 50]. Approximately 10% of patients with IgA and IgG deficiency have normal or elevated IgM — the hyper-IgM phenotype — which is associated with severe immunodeficiency, respiratory failure and shorter survival [1, 51, 52]. Lymphopenia, with reduced CD4+ and CD8+ cells, is the most frequent cellular abnormality [1].

Despite these deficiencies, severe bacterial, viral and opportunistic infections are uncommon; mild recurrent sinopulmonary infections predominate and aggravate lung disease. Until age 15 years, the most common bacterial organisms are Haemophilus influenzae, Streptococcus pneumoniae and Staphylococcus aureus, and Pseudomonas aeruginosa becomes more frequent in older patients [1, 50]. Cutaneous granulomas, occurring in fewer than 10% of patients, have been linked to persistence of vaccine-strain rubella virus [7, 53, 54]. A minority develop gammopathies or marked immunoglobulin elevations, sometimes with hyperviscosity [7].

3.5 Lung disease

Lung disease is present in most patients and, alongside cancer, is the main cause of death [1, 13]. The European Respiratory Society distinguishes three overlapping phenotypes: recurrent infections due to immunodeficiency, with bronchiectasis and pleural abnormalities; disease associated with dysphagia and ineffective cough due to neuromuscular impairment, with chronic aspiration; and interstitial lung disease or fibrosis [13]. Interstitial disease usually presents in adolescence with dry cough, tachypnoea, hypoxaemia and dyspnoea without infection; in one series, 19 of 25 patients died within 24 months of onset, and only those treated early with systemic corticosteroids showed clinical and radiological improvement [1, 13]. Restrictive disease, due to respiratory muscle weakness and incoordination, scoliosis or fibrosis, reduces forced vital capacity; in 12 wheelchair-using adolescents (median forced vital capacity 44% predicted), polysomnography showed infrequent partial obstructions and minimal nocturnal hypoxaemia, but reduced sleep efficiency [13].

3.6 Malignancies

The risk of malignancy in classic A-T was 24.5% (69 of 279 patients) in the French primary immunodeficiency registry: 38 non-Hodgkin lymphomas (predominantly aggressive B-cell), 12 Hodgkin lymphomas, 8 acute leukaemias (4 T-cell acute lymphoblastic leukaemias), 3 T-cell prolymphocytic leukaemias and 8 carcinomas, at median ages of 9.7; 10.6; 8.3; 24.2 and 31.4 years, respectively; Epstein–Barr virus was associated with all Hodgkin lymphomas and half of non-Hodgkin lymphomas [55]. GeneReviews summarises this risk as 25%, with a median age at diagnosis of 12.5 years [1]. The Oxford Desk Reference reports 10%–30% of patients, including 15% of those younger than 16 years, and a 2017 North American review reports “up to 40%” [5, 14]. At one German centre, 19 cancers occurred in 17 of 69 patients (25%), with higher risk in boys and a median survival of 2.11 years after malignancy diagnosis [56]. These proportions describe referral-centre cohorts or registries, with different follow-up periods and eras, and are not absolute lifetime risks.

In adults, solid tumours occur in addition to haematological malignancies: breast, ovary, stomach, liver, oesophagus, melanoma, leiomyomas and sarcomas [1, 7]. Breast cancer is part of the phenotype: eight cases in a series of 296 patients corresponded to a standardised mortality ratio of 31.1 and an estimated cumulative risk of 45% by age 50 years (95% confidence interval 24%–76%) [35]. In variant A-T, the overall risk of malignancy was 29.5% in the analysis of 105 patients with mild neurological phenotypes and 25% in the largest published variant A-T cohort (57 patients), with particular risk of premenopausal breast cancer and haematological malignancies; missense variants were associated with milder neurological disease but higher cancer risk than splicing variants [44, 45].

3.7 Endocrinology, growth and metabolism

Linear growth impairment is common and multifactorial — nutritional problems, recurrent infections and suboptimal growth hormone levels — but also appears to be a primary disease feature, more pronounced in females [1, 57]. In girls, elevated luteinising hormone (LH) and follicle-stimulating hormone (FSH) levels are frequent, indicating ovarian insufficiency, with a spectrum from absent pubertal development to normal menstruation; in boys, FSH, LH and testosterone levels are generally normal, although gonadal failure and abnormal spermatogenesis have been described [1, 57]. Insulin resistance, with hyperglycaemia and hyperinsulinaemia, develops in some adults, rarely with glycosuria or ketosis; thyroid and adrenal function are usually normal [1]. In adulthood, elevated transaminases, hepatic steatosis, non-alcoholic cirrhosis, hypercholesterolaemia and hypertriglyceridaemia, osteopenia and vitamin D deficiency are frequent [1, 7]. In a Brazilian cross-sectional study (13 patients, mean age 14.6 years), 30.8% were malnourished and 23.1% had short stature, with an atherogenic lipid profile [58]; in a Latin American registry of 218 patients from nine countries, the proportion with severe thinness increased with age, 23.6% had elevated triglycerides, 31.7% elevated total cholesterol and 34.0% abnormal LDL cholesterol, and aspartate aminotransferase rose progressively in paired samples [59].

3.8 Skin, telangiectasias and other findings

Telangiectasias of the bulbar conjunctiva and sun-exposed areas (pinnae, face, neck, hands) are generally evident by approximately age 6 years — between 4 and 8 years according to the Oxford Desk Reference — and are absent at disease onset; they do not affect visual acuity [1, 7, 46]. They may develop in other organs: the bladder, particularly after intravenous cyclophosphamide, with potentially severe haematuria; the gastrointestinal tract; and the brain, where they usually require no intervention [1]. Premature ageing manifests as greying hair, hypopigmentation, café-au-lait macules and adolescent onset of diseases associated with older age, such as diabetes and liver disease [1, 5]. With increased survival, postural scoliosis, progressive foot deformities, gastro-oesophageal reflux, early menopause, depression and incontinence related to transfer difficulties have also been described [7].

4. Clinical Recognition and Warning Signs

Classic A-T is often diagnosed late because, in the early years, ataxia may be interpreted as motor immaturity and the most characteristic signs — telangiectasias and oculomotor apraxia — appear only years after onset [1, 7]. For the same reason, A-T should be considered in any child with ataxia without an established diagnosis [7]. The following findings should trigger targeted investigation.

The history should include a three-generation pedigree assessing consanguinity, pregnancy history and birth weight, motor developmental milestones (particularly age at walking), growth, photosensitivity, types of infection and vaccine responses [5, 46]. Examination should assess gait and truncal ataxia, oculomotor abnormalities, conjunctival telangiectasias, extrapyramidal signs, neuropathy and growth parameters [5, 46].

5. Diagnosis: Criteria, Tests and Algorithms

5.1 Criteria and scenarios raising suspicion

No consensus clinical diagnostic criteria have been published; the diagnosis is established in a proband with suggestive findings and biallelic pathogenic (or likely pathogenic) ATM variants [1]. GeneReviews describes two scenarios: abnormal newborn screening for severe combined immunodeficiency (SCID) and a symptomatic proband [1]. As an alternative to molecular confirmation, absent or deficient ATM protein or kinase activity in cultured cell lines also confirms the diagnosis, according to the 2016 North American review [7].

5.2 TREC newborn screening

Newborn SCID screening quantifies T-cell receptor excision circles (TREC) in dried blood spots. Newborns with classic A-T, still asymptomatic, may have TREC levels as low as children with SCID; screening is estimated to identify approximately 50% of children with classic A-T, but does not detect variant A-T, which is not accompanied by immunodeficiency [1]. In the California programme, exome sequencing of newborns with unexplained T-cell lymphopenia led to two unexpected A-T diagnoses and, among archived samples from 13 older patients, 7 would have screened positive by the programme's criteria [61]. A newborn with a positive screen requires immediate immunological assessment at a centre experienced in SCID, including its genetic causes and haematopoietic cell transplantation [1]. Neonatal identification allows avoidance of infectious complications and unnecessary radiation, and informs the family about recurrence risk and cancer risk in carriers, although no curative treatment for the neurological disease yet exists [61].

5.3 Biomarkers and additional investigations

The Oxford Desk Reference describes the traditional laboratory work-up: full blood count, 3–5 mL of EDTA blood for immunoglobulins, subclasses and lymphocyte subsets, and 5 mL in lithium heparin for radiosensitivity testing and establishment of a lymphoblastoid cell line, in which ATM protein can be studied by western blot and residual activity assessed, followed by gene sequencing [5]. In current approaches, genomic analysis (panel, exome or genome) usually precedes cellular assays, which are reserved for resolving variants of uncertain significance [1, 5].

5.4 Molecular tests and their limitations

These percentages describe the proportion of known pathogenic variants detected by each method and do not equal the clinical sensitivity of a test in an individual patient: a negative result or identification of a single variant in the presence of a convincing phenotype (elevated AFP, immunodeficiency, telangiectasias) does not exclude the diagnosis and should prompt CNV analysis, genome or RNA analysis, and functional studies [1, 34]. If the proband appears homozygous, a deletion of one or more exons on the other allele (pseudohomozygosity) and uniparental isodisomy must be excluded; if only one parent is a carrier, with paternity and maternity confirmed, a de novo variant in the proband or parental mosaicism should be considered [1].

The report should specify the gene, versioned transcript (NM_000051.4, MANE Select), DNA and protein HGVS nomenclature, zygosity and phase (confirmed by parental testing), classification according to ACMG/AMP criteria with the ATM expert panel specifications, and technical limitations (intronic coverage, CNV, mosaicism) [1, 10]. When the literature uses older nomenclature — such as 5762ins137 or IVS40-1050A>G for c.5763-1050A>G — the report should provide both to allow correspondence [1, 35]. Functional assays and family segregation may support reclassification, but partial segregation constitutes research rather than proof of pathogenicity [1].

5.5 Proposed diagnostic algorithm

6. Differential Diagnosis

The differential diagnosis depends on presentation. In early-onset ataxia with oculomotor abnormalities, ataxias with oculomotor apraxia and other autosomal recessive ataxias should be considered; in immunodeficiency with neurological signs or radiosensitivity, other chromosomal instability syndromes; and in isolated dystonia, genetic dystonias [1, 5, 9]. None of the ataxias in the GeneReviews table is associated with immunological abnormalities, providing a means of distinction [1]. Tables on the approach to inborn errors of metabolism also include A-T among causes of basal ganglia and brainstem hyperintensities and oculomotor apraxia [67].

To distinguish A-T from ATLD and NBS, the combination of AFP, immunoglobulins, head circumference and telangiectasias is generally sufficient, but definitive diagnosis is molecular [5, 73]. MRE11 heterozygotes, relatives of patients with ATLD, should not be managed by analogy with ATM carriers: in a meta-analysis of 53 studies, pathogenic MRE11 variants did not increase breast cancer risk (OR 0.87; 95% CI 0.66–1.13), whereas NBN variants were associated with melanoma, pancreatic, haematological and prostate malignancy risks; risks for individual MRN complex genes cannot be transferred to the others [76].

7. Treatments and Risk Management Strategies

7.1 General principles

Until September 2026 there was no approved treatment for A-T, and the authorisation subsequently granted in the USA concerns symptomatic treatment of ataxia, rather than a curative or proven disease-modifying therapy [1, 16]. Supportive care aims to improve quality of life, maximise function and reduce complications, ideally through a multidisciplinary team including (paediatric) neurology, respiratory medicine, immunology, paediatrics and internal medicine, physical and rehabilitation medicine, physiotherapy, speech and language therapy, occupational therapy and nutrition, and, according to the problems, oncology, clinical genetics, endocrinology, orthopaedics, dermatology, mental health and social work [1, 12]. Most recommendations rely on expert opinion and observational series, owing to a lack of clinical trials [1, 13]. Distinguishing classic from variant phenotypes is not always possible, so plans should be guided by each patient's problems [1].

7.2 Neurological manifestations and rehabilitation

For focal dystonia, botulinum toxin A is the first choice, as in torticollis; for generalised dystonia, anticholinergics and GABA-mimetic drugs may help; in dopa-responsive generalised dystonia, a therapeutic trial of levodopa may be useful [1, 39]. Deep brain stimulation of the internal globus pallidus, described in only two patients, produced conflicting results [1]. Overall efficacy of pharmacotherapy for movement disorders in A-T is low to moderate and rests on non-randomised studies [1]. In a small series, 4-aminopyridine improved oculomotor and vestibular function, and acetyl-DL-leucine affected downbeat nystagmus in a study of six patients; nystagmus and oculomotor apraxia generally require no pharmacological treatment [1]. Drugs and doses for these uses are not included in A-T labels, so these are off-label uses and doses should follow study or specialty protocols.

Rehabilitation is the basis of treatment for motor disability: physiotherapy to maintain strength, fitness and activity and prevent contractures, with foot orthoses, a wheelchair providing correct sitting posture (reducing scoliosis and choking), and tilt tables for standing; occupational therapy for activities of daily living; speech and language therapy for dysarthria, with augmentative and alternative communication when needed [1, 47]. Ophthalmologists can advise on visual aids for reading [1].

7.3 Levacetylleucine (Aqneursa®)

Levacetylleucine (N-acetyl-L-leucine) is the L-enantiomer of an acetylated leucine derivative; its molecular mechanism in A-T is unknown, and the label describes metabolism to acetate and L-leucine without cytochrome P450 involvement [16]. The FDA approved it in September 2024 for neurological manifestations of Niemann-Pick disease type C and, through an efficacy supplement submitted on 19 March 2026 and approved on 18 September 2026 with priority review and orphan designation, for treatment of ataxia in adults and children with A-T weighing ≥15 kg [16, 30].

7.4 Pivotal trials and interpretative limitations

These results require careful interpretation. Both eDSP programmes failed their primary endpoint, with point estimates of the same order (approximately −1.3 to −1.4 points) and confidence intervals including zero; the hypothesis of benefit at ages 6–9 years, generated by ATTeST subgroup analysis, was not confirmed in NEAT [27, 77]. The sponsor announced NEAT topline results on 29 January 2026 and discontinued development; the open-label extension was terminated and European orphan designation was withdrawn [28, 79, 80]. In the levacetylleucine trial, the −1.88-point SARA effect must be interpreted within a crossover design with 12 weeks per period, measuring symptomatic improvement without demonstrating modification of progression; the FDA based approval on a shortened functional scale (fSARA), with an absolute difference of 0.6 out of 16 points, supported by the original SARA [16, 29]. A 2026 systematic review (13 studies, 314 paediatric patients) concluded that the recurring pattern of promising open-label results followed by negative phase 3 trials reflects small samples, heterogeneous populations, variable endpoints and non-linear progression with age-dependent responses [81]. These interventions cannot legitimately be ranked through indirect comparison across trials with different scales, ages and durations.

7.5 Comparative regulatory status

Orphan designation is not a marketing authorisation, and authorisation does not guarantee marketing, reimbursement or access in individual countries; no information on levacetylleucine availability in Portugal (INFARMED) or Brazil (ANVISA/CONITEC) was found in the sources consulted.

7.6 Immunodeficiency, infections and vaccination

Prophylactic antibiotics should be considered early in patients with recurrent or severe infections and those with IgG deficiency or specific antipolysaccharide antibody deficiency [1]. Immunoglobulin replacement therapy, intravenously or subcutaneously, is indicated in severe IgG deficiency and the hyper-IgM phenotype, and may be considered for recurrent infections, mild humoral defects or poor responses to vaccine boosters [1, 7]. For intravenous immunoglobulins, labels warn of thrombosis, renal dysfunction and acute renal failure, haemolysis and aseptic meningitis, and contraindicate use in patients with IgA deficiency and anti-IgA antibodies with a history of hypersensitivity, and, for Privigen®, in hyperprolinaemia; the most frequent adverse effects are headache, fatigue, nausea, chills and fever, minimised by progressively increasing infusion rates and hydration [84, 85]. In patients with absent IgA, washed red cell concentrates are recommended for elective transfusions [7].

Inactivated vaccines are safe in all patients; influenza vaccine (annually, also for household members), pneumococcal conjugate vaccine and the 23-serotype polysaccharide vaccine should be administered [1, 7]. In current practice, the 13-valent conjugate vaccine referred to in the sources should be interpreted in accordance with current national schedules. Sources differ regarding live vaccines: the 2016 North American review recommends live measles, mumps, rubella and varicella vaccines in patients not requiring gammaglobulin, whereas GeneReviews 2023 recommends avoiding rubella vaccine in severe immunodeficiency because of the risk of vaccine-strain virus-associated cutaneous granulomas [1, 7, 53, 54]. Decisions should be based on individual immunological assessment. Cutaneous granulomas have been treated with potent topical corticosteroids or ciclosporin, intralesional injections, immunoglobulin or TNF-α inhibitors for extensive lesions [7].

7.7 Respiratory disease and nutrition

The European Respiratory Society statement recommends proactive multidisciplinary monitoring of immune function, infections, lung function, swallowing, nutrition and scoliosis; it did not use the GRADE method and therefore represents a consensus statement rather than a formal set of recommendations [13]. Spirometry is feasible with adaptations (head stabilisation, cheek support, mouthpiece adaptation) and allows decline to be tracked; when interstitial disease is suspected, high-resolution computed tomography is considered justified, and early systemic corticosteroid treatment was associated with a better course [13]. Airway clearance techniques, regular exercise and early antibiotic treatment of exacerbations should be encouraged [13].

Good nutritional status is necessary for other treatments to be effective; malnutrition results from mechanical eating difficulties, fatigue and endocrine abnormalities [1]. Correct sitting posture, straws, thickened fluids and easy-to-chew foods are useful; early gastrostomy placement improves outcomes and is indicated when a child cannot eat enough to grow, when any patient cannot maintain weight, when aspiration is problematic or when meals become excessively long or stressful [1, 7].

7.8 Malignancies: treatment and radiation

There are no consensus protocols for treating malignancies in A-T; treatment is individualised according to patient factors (mobility, lung function) and tumour factors, and chemotherapy may begin with dose-adjusted protocols, increasing doses if tolerated [1]. In T-cell acute lymphoblastic leukaemia, many patients had not yet been diagnosed with A-T when leukaemia was diagnosed; complete remission rates were high with both modified and unmodified strategies, and treatment is feasible and should be undertaken through shared decision-making [63]. In the French registry, survival improved in patients with a major treatment response (3.46 vs. 0.87 years) [55]. Radiotherapy is contraindicated in both classic and variant forms because of radiosensitivity, which can cause very severe complications; a North American review allows it only in rare circumstances and at reduced doses [1, 7]. This contraindication applies only to biallelic patients. In heterozygous relatives who develop cancer, the 2025 American College of Medical Genetics and Genomics (ACMG) clinical practice resource recommends offering radiotherapy when indicated, without avoiding it and without modifying the dose because of the genetic result: in heterozygous women with breast cancer, adjuvant radiotherapy reduced local recurrence, did not increase acute or late toxicity and was not associated with more second primary tumours [8, 11].

Allogeneic haematopoietic cell transplantation has heterogeneous outcomes: in a systematic review with individual analysis of 16 patients (2000–2025), the two who received myeloablative conditioning died, mortality with reduced-intensity conditioning was 33.3%, half had relevant toxicity, half had graft-versus-host disease and half survived; an adolescent with T-cell leukaemia achieved leukaemia-free survival with treosulfan dosed using pharmacokinetic monitoring [87]. The effect of transplantation on neurodegeneration is uncertain and its use outside oncology remains experimental [1].

Exposure to diagnostic ionising radiation should be limited to situations in which it changes management, favouring magnetic resonance imaging and ultrasound [1, 7]. The European Respiratory Society adopts a pragmatic position: a single diagnostic radiograph or a computed tomography scan limited to the area of interest, at the lowest possible dose, is justified when omission could delay a relevant diagnosis, since there are no data on adverse effects of diagnostic doses [13].

7.9 Endocrinology, metabolism and perioperative care

Insulin-resistant diabetes and hypercholesterolaemia in young adults are treated according to general population protocols, although the limited life expectancy of the classic form may justify a decision not to treat hypercholesterolaemia [1]. In women with gonadal insufficiency (primary or secondary amenorrhoea), oestrogen supplementation prevents osteoporosis; growth hormone may be an option in children with severe growth impairment [1, 88]. Before anaesthesia, the increased risk of complications from lung disease and malnutrition should be considered; patients should undergo surgery in hospitals experienced in A-T, and expedited ventilator weaning with early extubation may improve outcomes in severe restrictive disease [1, 89].

7.10 Investigational therapies

The antisense oligonucleotide atipeksen (AT008), targeting variant c.7865C>T, restored splicing and ATM signalling in fibroblasts and was administered intrathecally to a child diagnosed after low TREC on newborn screening, beginning at age 2 years with escalation and maintenance at 42 mg every three months, with good tolerability over three years; this regimen is experimental and does not constitute a dose recommendation [34]. A phase 1/2 trial (NCT07215416), open to patients up to age 17 years with classic A-T and at least one copy of c.7865C>T, began recruitment in 2026 [90]. Other studies include sirolimus in a pilot study of 6 patients at Mexico's Instituto Nacional de Pediatría (1 mg/m²/day, 6 months; completed in 2026 without published results), triheptanoin and nicotinamide riboside, and gene therapy using viral vectors is limited by the size of the gene [31, 91, 92]. Oral corticosteroids (betamethasone) produced transient neurological improvements in small series, but prolonged use exposes patients to severe adverse effects and is not recommended outside a trial [1, 81].

8. Reproductive Options and Family Counselling

8.1 Inheritance and risk to relatives

A-T results from biallelic pathogenic ATM variants and is inherited in an autosomal recessive manner, with complete penetrance of the biallelic phenotype and variable expressivity [1, 5]. The parents of a patient are generally heterozygotes; molecular testing is recommended to confirm carrier status, estimate recurrence risk accurately and clarify their own cancer risk [1]. When both parents are heterozygotes, each sibling of the proband has, at conception, a 25% probability of having A-T, a 50% probability of being heterozygous and a 25% probability of inheriting neither familial variant; the phenotype tends to be similar between affected siblings, although presentation and immunological abnormalities may differ [1]. Each sibling of the proband's parents has a 50% probability of being heterozygous [1].

With a carrier frequency of approximately 1 in 200 in the general population, risk to the extended family is low in the absence of consanguinity. The Oxford Desk Reference provides examples: the risk of a healthy sibling of a patient having an affected child with an unrelated partner is 2/3 × 1/200 × 1/4 = 1/1200, and that for an uncle or aunt is 1/2 × 1/200 × 1/4 = 1/1600 [5]. Carrier testing in partners should be considered, particularly when they share the same ancestry and when this corresponds to a population with a founder variant (Druze from northern Israel, Moroccan and Tunisian Jews, Spanish Roma) [1, 8]. For a confirmed heterozygote (for example, a parent or a carrier sibling of a patient) whose unrelated partner has not been tested, the ACMG estimates a risk of about 1 in 800 of having a child with A-T (1 × 1/200 × 1/4, assuming a carrier frequency of 1 in 200, which varies between populations) [11]. Partner testing should be offered with counselling, particularly when there is A-T in the family or consanguinity, either as single-gene testing or as part of an expanded carrier screening panel; in the ACMG tiered classification for preconception and prenatal carrier screening, ATM belongs to tier 4, comprising genes that are not offered routinely but only in specific circumstances, such as consanguinity or a family history that warrants testing [11]. Funding varies between health systems: in England, the National Health Service funds partner testing only when the carrier frequency in the relevant population exceeds 1 in 70 [11]. A negative result in the partner reduces the risk but does not eliminate it, because no test detects all pathogenic variants (see section 5.4).

Most patients with classic A-T do not have children, but exceptions exist, and reproduction has been described in several women with variant A-T; all children of a patient are obligate heterozygotes [1, 44]. Pregnancy in a woman with A-T requires consideration of cancer risk, respiratory function and drug safety; the levacetylleucine label requires exclusion of pregnancy before initiation and effective contraception during treatment and for the following 7 days [16].

8.2 Heterozygotes: cancer risk and management

Heterozygotes do not develop A-T but have an increased cancer risk, particularly women for breast cancer before age 50 years [1]. Detailed estimates and surveillance include a breast cancer risk of 21%–24% by age 80 years, approximately 43% for c.7271T>G, pancreatic, prostate and stomach cancer risks, mammography and breast magnetic resonance imaging protocols, and considerations on risk-reducing surgery, based on the 2025 American College of Medical Genetics and Genomics (ACMG) clinical practice resource and GeneReviews 2026 [8, 11]. Two points have immediate relevance when counselling families of patients with A-T. First, the 2017 Oxford Desk Reference recommendation to avoid regular mammography in female carriers because of radiation concerns has been superseded: current recommendations provide for annual mammography from age 40 years and annual magnetic resonance imaging from age 30–35 years, depending on the country [5, 8]. Second, unlike biallelic patients, heterozygous women with breast cancer did not experience greater acute or late radiotherapy toxicity, so it may be used when indicated [8]. A 2016 meta-analysis by the Nijmegen group also estimated increased mortality from cancer and ischaemic heart disease in carriers (relative risk 1.7) [93].

8.2.1 Risk estimates and the divergence over c.7271T>G

The ACMG treats ATM as a moderate-penetrance gene but recommends regarding risk as a continuous variable, modified by family history, hormonal and lifestyle factors and a polygenic score [11]. For truncating variants, the two large 2021 case-control studies in women of European ancestry gave breast cancer odds ratios of 2.10 (95% CI 1.71–2.57; BCAC/BRIDGES consortium, 294 case and 150 control carriers) and 1.82 (95% CI 1.46–2.27; CARRIERS, all pathogenic variants), with risk concentrated in oestrogen receptor-positive tumours and no demonstrated excess of contralateral cancer [8, 11]. For the pancreas, in a cohort of 2227 members of 130 families with pancreatic cancer from US and Canadian registries, including 155 heterozygotes, the cumulative risk in carriers was 1.1% at age 50 years, 6.3% at 70 and 9.5% (95% CI 5.0%–14.0%) at 80 years; this estimate derives from selected families and may be inflated by ascertainment bias. For the prostate, the association is mainly with aggressive disease (odds ratio between 1.88 and 4.4 depending on the study) [11].

The missense variant NM_000051.4:c.7271T>G p.(Val2424Gly), in the kinase domain, is the main exception (nomenclature confirmed in VariantValidator; classified as pathogenic by an expert panel in ClinVar, VCV000003023). Sources give different estimates because they use different populations and outcomes:

The 52% and 60% estimates come from 2006 and 2011 studies, predating the large consortia; the ACMG itself notes that a recent large case-control study did not reproduce such a high risk, although other studies confirm a risk higher than that of truncating variants [11]. The 43% estimate by age 80 years rests on the most recent and largest dataset, which also showed an increased risk of prostate cancer in male carriers of this variant (RR 4.84 in UK Biobank) [8]. All sources agree that c.7271T>G confers a breast cancer risk above that of truncating variants, approaching the high-risk threshold and warranting individualised management. This variant is directly relevant to A-T families: in combination with another pathogenic allele it is associated with the variant form, because the mutant protein retains some kinase activity, and parents and siblings heterozygous for it carry the risk described in this row rather than that of truncating variants [8, 11].

8.2.2 ACMG recommendations for heterozygotes

The 2025 ACMG recommendations apply to parents (obligate heterozygotes), carrier siblings and other adult relatives of a patient with A-T [11]:

• Breast: surveillance is based on an individual risk assessment with a model such as CanRisk (BOADICEA), which incorporates the risk of truncating variants; most heterozygous women meet the criteria for enhanced surveillance and some those for breast MRI, according to each country's thresholds. As a rule, annual mammography is offered from age 40 years, or earlier, and annual MRI from age 30–35 years in countries that provide for it. CanRisk does not incorporate non-truncating variants such as c.7271T>G and underestimates their risk, so it should not be used for them [8, 11].
• Risk-reducing mastectomy: bilateral mastectomy in unaffected heterozygous women and contralateral mastectomy in those who have already had breast cancer are not offered routinely, because the risk does not usually reach the threshold; they may be considered after individual risk estimation and shared decision-making. For very high-risk variants such as c.7271T>G, additional preventive measures may be considered according to national thresholds [11].
• Ovary: risk-reducing salpingo-oophorectomy is not recommended routinely; it may be discussed in the presence of a family history of ovarian cancer or gynaecological surgery planned for another reason [11].
• Prostate: it is reasonable to consider annual PSA from age 40 years (NCCN position adopted by the ACMG), and digital rectal examination may help interpret the PSA [11].
• Pancreas: surveillance should be considered, preferably within a clinical trial. North American guidelines (NCCN and the American Gastroenterological Association) include it even without a family history, from age 50 years or 10 years before the earliest family case, with annual endoscopic ultrasound or MRI/magnetic resonance cholangiopancreatography in an experienced centre; the international CAPS consortium reserves it for those with an affected first- or second-degree relative on the side from which the variant was inherited; in the United Kingdom it is not recommended outside research. The quality of evidence is low: in a series of 1461 high-risk individuals under surveillance, none of the 9 cancers occurred in the 93 ATM heterozygotes, and 5 of the 8 resections for cystic lesions showed only low-grade dysplasia [11].
• Radiotherapy: when indicated, it is offered without avoidance and without dose modification (see section 7.8) [11].
• PARP inhibitors: olaparib and the combination of talazoparib with enzalutamide are FDA-approved for metastatic castration-resistant prostate cancer with homologous recombination repair gene variants, including ATM, but the specific benefit in ATM tumours is weak or absent in the trials; in other tumours there is no evidence of benefit [11].

An example from the ACMG itself shows the effect of an A-T diagnosis in a child on the mother's management: in a 3-year-old child with ataxia and oculocutaneous telangiectasias, with two truncating variants in trans (c.2483del p.(Lys828SerfsTer8) and c.3802del p.(Val1268Ter)), the breast cancer risk by age 80 years of the 40-year-old mother, calculated in CanRisk, rose from 11% (family history alone) to 21% after she was confirmed as heterozygous; enhanced breast surveillance was offered to her, and testing was offered to her siblings and to her partner's sister [11]. The surveillance schedule for adult heterozygous relatives is summarised in section 11.3.

8.3 Testing relatives and minors

Testing for familial variants should be offered to apparently asymptomatic siblings, older or younger, to identify affected individuals early — who benefit from care, cancer surveillance and radiation avoidance — and heterozygotes [1]. As a rule, carrier status testing is not recommended before age 18 years, because cancer surveillance for heterozygotes begins only in adulthood; the exception is minors whose parents are both carriers, who are at risk of A-T and should be tested because the result changes childhood management [8]. The ACMG frames this exception in the same way: predictive testing for a familial variant is not recommended routinely in childhood, but siblings of a patient with A-T may be tested, with appropriate counselling, when there is doubt as to whether they are affected [11]. Parents are obligate heterozygotes, and other adult relatives should receive counselling and testing to know their risk, with counselling before and after predictive testing [11].

Relatives who test negative for the familial variant are not left with zero risk: the ACMG recommends giving them an estimate of residual risk, particularly of breast cancer, which continues to depend on family history and other factors, together with surveillance advice appropriate to that risk [11]. A variant of uncertain significance in ATM should not guide clinical management or be used for predictive testing of relatives; because ATM variants may be reclassified, the possibility of recontact should be discussed at the time of testing [11].

Heterozygotes may be identified independently of A-T: on multigene panels requested for a personal or family history of cancer, in preconception or prenatal carrier screening, or through a family history of A-T. ATM is not on the ACMG secondary findings list (version 3.2), so a heterozygote is generally not reported as a secondary finding in exome or genome sequencing requested for another reason; the absence of such a finding does not exclude carrier status [11].

8.4 Reproductive options

The ideal time to determine genetic risk and discuss prenatal diagnosis and preimplantation genetic testing is before pregnancy; genetic counselling should be offered to young adults who are affected, heterozygous or at risk of being heterozygous [1]. Once familial variants have been identified, prenatal diagnosis (by chorionic villus sampling, usually at 11 weeks, or amniocentesis) and preimplantation genetic diagnosis are possible using standard molecular techniques and should be coordinated with an experienced laboratory [1, 5]. When variants are unknown but the affected child's diagnosis is unequivocal, haplotype analysis has been used [7]. The ACMG recommends discussing preimplantation genetic testing and prenatal diagnosis by chorionic villus sampling or amniocentesis with couples in which both partners are carriers, with the aim of detecting the biallelic genotype [11]. By contrast, prenatal diagnosis aimed at detecting a single heterozygous variant in the fetus is controversial, because it confers a moderate-penetrance, adult-onset cancer predisposition; for the same reason, preimplantation testing to avoid transmitting heterozygous status is controversial, unlike in high-penetrance or childhood-onset syndromes, with few exceptions [11]. Other options include gamete donation and adoption. Perspectives on these techniques may differ among professionals and within families, and the decision is personal [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1 Incidence and prevalence

Reported incidence ranges from 1:300,000 to 1:40,000 births, and the 2016 North American review places birth prevalence between 1:40,000 and 1:100,000 [1, 7]. GeneReviews also reports an estimated prevalence of 1–9 per 100,000; this figure is difficult to reconcile with a birth incidence of 0.33–2.5 per 100,000 in a life-shortening disorder, so the incidence estimate is preferable for counselling [1]. The Oxford Desk Reference reports a prevalence of approximately 1:300,000 [5]. In the USA, approximately 350 children with A-T are known to the A-T Children's Project, and heterozygote frequency was estimated at approximately 1 in 200 from North American incidence [1]. In control populations of recent meta-analyses (BRIDGES, CARRIERS and UK Biobank), 0.35% of unaffected individuals carried a pathogenic ATM variant, with gnomAD v3.1 frequencies ranging from 1/142 (Latino/admixed Amerindian individuals) to 1/1204 (South Asians), 1/277 in non-Finnish Europeans and 1/357 in African Americans [8].

Orphanet records country-specific estimates with different measures and sources, which should not be interchanged: birth prevalence of 1.7 per 100,000 in the USA; point prevalence of 0.4 per 100,000 in Norway (southeast), 0.25 in France and the United Kingdom (the latter not yet validated), 1.19 in Italy, 0.15 in Portugal and 0.49 per 100,000 for Europe overall [2, 94]. These point prevalences, obtained mainly from hereditary ataxia surveys, underestimate incidence because of early mortality and underdiagnosis, and do not measure carrier frequency. An industry estimate reports approximately 4600 diagnosed patients in the USA and 5000 in the United Kingdom and the four largest EU countries, based on medical claims data; this is a sponsor press release rather than a peer-reviewed epidemiological study [28].

9.2 Founder variants

The Oxford Desk Reference describes the disease in all ethnic groups and the consulted sources do not describe ATM founder variants in Portugal or Brazil [5]. In Portugal, the population survey of hereditary ataxias conducted since 1993 documented the importance of ataxias with oculomotor apraxia (estimated prevalence 0.41 per 100,000; AOA1 with shared variant p.Trp279Ter; AOA4 as the most frequent AOA form), relevant to the differential diagnosis of A-T [20, 69]. In Brazil, available data derive from clinical series: a study of 13 patients from São Paulo on nutritional status and atherogenic risk and the Latin American immunodeficiency (LASID) registry, with 218 patients from nine countries and participation by Brazilian centres [58, 59]. These series describe patients followed at referral centres and cannot estimate prevalence; absence of Brazilian estimates in the consulted sources does not prove that none exist.

9.3 Sex, age and access

Risk of growth impairment appears higher in females, and women have a particular risk of ovarian insufficiency and early breast cancer [1, 35]. At one German centre, cancer risk was higher in male patients [56]. Patients who survive into adulthood face diseases associated with ageing and transition to adult services, which are often less familiar with the disease [7, 98]. The availability of multidisciplinary centres and registries — European (ESID), Latin American (LASID) and the North American association's registry — influences access to early diagnosis and clinical trials [1, 34, 59].

9.4 European networks and patient associations

In Europe, A-T lies at the intersection of two European Reference Networks: ERN-RND, which includes the cerebellar ataxia and hereditary spastic paraplegia group, and ERN RITA, dedicated to primary immunodeficiencies and other rare immunological diseases; no A-T-specific page or guideline published by an ERN was identified in the sources consulted [18, 19]. The European Society for Immunodeficiencies (ESID) registry collects data on patients with A-T [1]. Patient and family associations include the A-T Children's Project (USA), A-T Society (United Kingdom), Twan Foundation (Netherlands), National Ataxia Foundation (USA), Ataxia UK, the euro-ATAXIA federation and, in Portugal, the Associação Portuguesa de Ataxias Hereditárias (APAHE) [1, 9, 20].

10. Illustrative Clinical Case

10.1 Severe radiotherapy reaction as the presentation of variant A-T (Byrd et al., 2012)

Byrd and colleagues, from the University of Birmingham, described in the British Journal of Cancer a 44-year-old woman, a mother and professionally active, with right breast carcinoma (T2 G2, two positive lymph nodes, oestrogen receptor-positive), treated with lumpectomy, adjuvant FEC chemotherapy and conventional radiotherapy of 40 Gy in 15 fractions over 3 weeks [64]. After treatment she developed an acute reaction at the severe end of the usual spectrum, followed by a late reaction with breast retraction and distortion, photographically documented in Figure 1 of the original article [64]. BRCA1 and BRCA2 sequencing and MLPA were negative [64].

The severity of the reaction prompted investigation. The patient walked into the clinic unaided, had no ataxia, had normal tandem gait, feeding and swallowing, and no oculomotor apraxia, hypometric saccades, pursuit abnormalities or involuntary movements; findings were limited to mild right esotropia, terminal dysmetria on finger-to-nose testing, reduced vibration and position sense up to the knees, absent ankle and knee reflexes, and mild axonal polyneuropathy on electromyography [64]. Skin fibroblasts showed increased radiosensitivity in the clonogenic assay, below that of a patient with classic A-T (Figure 2), and lymphocytes had approximately three times more induced chromosomal damage than the control [64].

Sequencing identified two variants in trans, confirmed in cDNA and the parents: c.8672G>A, p.(Gly2891Asp), paternally inherited, in the kinase domain, and c.1A>G, p.(Met1?), maternally inherited, in the initiation codon (Figure 3) [64]. The original nomenclature was described on NM_000051.3; VariantValidator conversion confirmed the same numbering on the MANE transcript NM_000051.4 [4, 64]. Cell lines expressed reduced quantities of ATM protein, with residual kinase activity of approximately 12% of normal towards the p53 substrate (Figures 4 and 5), and modelling showed that p.Gly2891Asp protein retains activity, whereas the c.1A>G allele produces a truncated protein without measurable activity [64].

The authors concluded that this was a very mild form of A-T, which might have contributed to breast cancer and caused the radiotherapy reaction, and proposed investigating A-T in patients with unexpected severe acute and late reactions to breast radiotherapy [64]. The case illustrates several points in this article: variant A-T may remain unrecognised until adulthood; absence of ataxia and oculomotor abnormalities does not exclude the diagnosis; early breast cancer is part of the biallelic phenotype; radiosensitivity also exists in the variant form; and functional studies help interpret atypical variants [1, 35, 44]. The oncological management described (conventional radiotherapy) preceded diagnosis and should not be read as a recommendation; with a known diagnosis, current guidelines contraindicate radiotherapy in both classic and variant forms [1].

10.2 Neonatal diagnosis and individualised therapy case (Kim et al., 2023)

In the A-T Children's Project programme, a child was diagnosed in infancy after low TREC on newborn screening; exome sequencing revealed compound heterozygosity for c.7865C>T and the 27 bp deletion c.8585-13_8598del at the intron–exon junction, confirmed by trio sequencing [34]. At age 2 years, intrathecal treatment with the oligonucleotide atipeksen began under an individual protocol, and at age 6 years tolerability remained good without serious adverse events after three years, with “encouraging signs” described in a supplementary note; clinical efficacy cannot be inferred from a single patient [34]. The case shows how newborn screening and comprehensive sequencing can bring diagnosis forward to before neurological deterioration, creating a window for genotype-directed interventions [34, 61].

11. Prognosis and Long-Term Follow-Up

11.1 Survival and prognostic determinants

Life expectancy in classic A-T is greatly reduced by cancer, lung disease and infections; most patients do not live beyond age 30 years [1]. In two North American cohorts, one prospective and one retrospective, median survival was 25 and 19 years, with wide variation, and did not correlate well with neurological severity [99]. In the French registry of 240 patients born between 1954 and 2005, survival at age 20 years was 53.4%, without improvement over the period; cancer (hazard ratio 2.7) and respiratory infections (2.3) were independently associated with mortality, cancer predominantly in patients with null variants and respiratory infections in those with hypomorphic variants [100]. In a systematic literature review, median age at death (294 patients) was 14 years (interquartile range 10–23 years; extremes 1 to 76 years), and malignancy was the most frequently reported cause of death [60]. At a German centre with 69 patients followed between 2002 and 2019, median overall survival was 26.9 years and survival at age 15 years was 82.5%, falling to a median of 2.11 years after malignancy diagnosis [56]. These estimates derive from different populations, eras and centres and are not directly comparable.

In the Dutch cohort of 61 patients, classic A-T had poorer survival than the variant form (hazard ratio 5.9), worsening after malignancy diagnosis (2.5); the hyper-IgM phenotype with hypogammaglobulinaemia (9.2) and IgG2 deficiency (7.8) were associated with poorer survival, and most patients with hyper-IgM died before age 15 years from respiratory failure [1, 52]. Patients with c.3576G>A have longer survival and less cancer; compound heterozygotes for c.8147T>C have a variant phenotype with survival that exceptionally approaches normal [36]. Long-term survivors with classic A-T accumulate neurological disability, lung disease, metabolic disorders and solid malignancies [7, 98]. In variant A-T, longer survival results from slower neurological decline, the usual absence of respiratory and immunological complications, and later age at malignancy [1].

11.2 Assessments after diagnosis

11.3 Surveillance

Recommended surveillance is the same for classic and variant forms, with the pulmonary exception noted above [1]. Sources differ on the extent of cancer screening. The 2017 American Association for Cancer Research (AACR) workshop review suggests, for children, annual physical examination, full blood count, metabolic profile and LDH, immunoglobulin monitoring, annual skin examination, annual scoliosis and diabetes assessment, and limiting X-rays, without proposing systematic imaging [14]. A 2021 North American expert review suggests that adults may benefit from annual whole-body MRI [101], and a British study evaluated this technique in A-T (NCT05252819). The 2023 Delphi consensus, involving 35 experts from 13 countries, concluded that evidence-based guidelines are needed, with different recommendations for adults and children; that a full blood count should be included, with a blood film for abnormal white cells or cytopenias; that surveillance should be at least annual and have no stopping age; and that starting age, optimal tests and differentiation between classic and variant forms require research [15]. Biomarkers of progression and treatment response have not yet been validated, limiting monitoring and trials [102].

Parents and adult heterozygous relatives do not have A-T and follow a different schedule, centred on the moderate cancer risk (see section 8.2); the table summarises the 2025 ACMG and 2026 GeneReviews recommendations [8, 11].

11.4 Transition, quality of life and palliative care

Transition to adult medicine should be planned, because adults accumulate steatosis and liver disease, dyslipidaemia, diabetes, osteopenia, scoliosis, reflux, early menopause, depression and solid tumours [7, 98]. Patients' emotional well-being and ageing caregivers require explicit attention; for difficult life-prolonging decisions, GeneReviews suggests consultation with independent clinical teams and timely involvement of palliative care [1, 7].

12. Summary and Final Considerations

Ataxia-telangiectasia is an autosomal recessive disorder caused by biallelic variants in ATM, the kinase coordinating the response to DNA double-strand breaks, and manifests as a continuous spectrum between the classic form, without residual kinase activity, and the variant form, with residual activity [1, 26]. In the classic form, ataxia emerges with walking, followed by extrapyramidal movement disorder, neuropathy, dysarthria, dysphagia and oculomotor abnormalities; telangiectasias appear around age 6 years; immunodeficiency is variable and lung disease and lymphoid malignancies determine survival, with a median between 19 and 27 years in the described cohorts [1, 56, 99]. The variant form, often dystonic and diagnosed after delays of decades, retains cancer risk and radiosensitivity [44, 64].

Diagnosis rests on clinical suspicion, supported by elevated AFP after age 2 years and immunological abnormalities, and is confirmed by identification of two pathogenic variants interpreted on transcript NM_000051.4; technical limitations — deletions, deep intronic variants and apparently missense variants affecting splicing — justify genome sequencing and functional studies when results are incomplete [1, 34]. TREC newborn screening identifies approximately half of children with the classic form [1, 61]. Differentiation from AOA1, AOA2, AOA4, ATLD and Nijmegen breakage syndrome uses the combination of AFP, immunoglobulins, telangiectasias, head circumference and molecular genetics [1, 5, 73].

Management is multidisciplinary and preventive: rehabilitation, nutrition with early gastrostomy, antibiotic prophylaxis and immunoglobulin when indicated, vaccination adjusted to immunity, at least annual cancer surveillance, individualised cancer treatment, and avoidance of unnecessary ionising radiation and of radiotherapy [1, 13, 15]. In 2026, erythrocyte-encapsulated dexamethasone failed its second phase 3 trial and development ceased, whereas levacetylleucine obtained the first US authorisation for A-T ataxia, with a symptomatic effect demonstrated over 12 weeks and no equivalent authorisation in the European Union [16, 17, 27, 29]. Antisense oligonucleotides targeting splicing variants open a path to individualised treatment for a minority of patients [34, 90]. Genetic counselling should cover the 25% recurrence risk, prenatal and preimplantation diagnosis, and cancer risk in heterozygotes [1, 8]. For parents and heterozygous relatives, the ACMG recommends individualised breast surveillance with CanRisk, no routine risk-reducing mastectomy, annual PSA from age 40 years and considered pancreatic surveillance, preferably within a trial; radiotherapy is offered without dose modification; the risk with an untested partner is about 1 in 800; and those who test negative receive a residual risk estimate [11].

13. Bibliographic References

1. Veenhuis S, van Os N, Weemaes C, Kamsteeg EJ, Willemsen M. Ataxia-Telangiectasia. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. First published 19 Mar 1999; full update 5 Oct 2023; revisions 4 Dec 2025 and 22 Sep 2026. PMID:20301790. NBK26468.
2. Orphanet. Ataxia-telangiectasia. ORPHA:100 [Internet]. Definition and epidemiological data via Orphadata. Disease entry. Accessed 6 Oct 2026.
3. Orphanet. Ataxia-telangiectasia variant. ORPHA:370109 [Internet]. Disease entry. Accessed 6 Oct 2026.
4. Reference records for the ATM gene: VariantValidator (GRCh38, MANE selection), Ensembl (ENST00000675843, MANE Select, 63 exons) and UniProt (Q13315, 3056 amino acids; functional annotation: activation by double-strand breaks, substrates with a serine/threonine-glutamine motif, including TP53, CHK2, BRCA1, NBN, MRE11 and H2AX, and a role in immunoglobulin heavy-chain allelic exclusion). VariantValidator; Ensembl; UniProt Q13315. Accessed 6 Oct 2026.
5. Firth HV, Hurst JA. DNA repair disorders. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 398–401. Local chapter.
6. Poli MC, Aksentijevich I, Bousfiha AA, Cunningham-Rundles C, Hambleton S, Klein C, et al. Human inborn errors of immunity: 2024 update on the classification from the International Union of Immunological Societies Expert Committee. J Hum Immun. 2025;1(1):e20250003. doi:10.70962/jhi.20250003. PMID:41608114.
7. Rothblum-Oviatt C, Wright J, Lefton-Greif MA, McGrath-Morrow SA, Crawford TO, Lederman HM. Ataxia telangiectasia: a review. Orphanet J Rare Dis. 2016;11(1):159. doi:10.1186/s13023-016-0543-7. PMID:27884168.
8. Schon KR, Pal T, Stewart DR, Tischkowitz M, Hanson H. ATM-Related Cancer Predisposition. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Published 4 Jun 2026. NBK623002.
9. Perlman S. Hereditary Ataxia Overview. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Full update 16 Jun 2022; revision 9 Jul 2026. PMID:20301317. NBK1138.
10. Richardson ME, Holdren M, Brannan T, de la Hoya M, Spurdle AB, Tavtigian SV, et al. Specifications of the ACMG/AMP variant curation guidelines for the analysis of germline ATM sequence variants. Am J Hum Genet. 2024;111(11):2411–2426. doi:10.1016/j.ajhg.2024.08.022. PMID:39317201.
11. Pal T, Schon KR, Astiazaran-Symonds E, Balmaña J, Foulkes WD, James P, et al. Management of individuals with heterozygous germline pathogenic variants in ATM: a clinical practice resource of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2025;27(1):101243. doi:10.1016/j.gim.2024.101243. PMID:39636577.
12. van Os NJH, Haaxma CA, van der Flier M, Merkus PJFM, van Deuren M, de Groot IJM, et al. Ataxia-telangiectasia: recommendations for multidisciplinary treatment. Dev Med Child Neurol. 2017;59(7):680–689. doi:10.1111/dmcn.13424. PMID:28318010.
13. Bhatt JM, Bush A, van Gerven M, Nissenkorn A, Renke M, Yarlett L, et al. ERS statement on the multidisciplinary respiratory management of ataxia telangiectasia. Eur Respir Rev. 2015;24(138):565–581. doi:10.1183/16000617.0066-2015. PMID:26621971.
14. Walsh MF, Chang VY, Kohlmann WK, Scott HS, Cunniff C, Bourdeaut F, et al. Recommendations for childhood cancer screening and surveillance in DNA repair disorders. Clin Cancer Res. 2017;23(11):e23–e31. doi:10.1158/1078-0432.CCR-17-0465. PMID:28572264.
15. Neves R, De Dios Perez B, Panek R, Jagani S, Wilne S, Bhatt JM, et al. Development of cancer surveillance guidelines in ataxia telangiectasia: a Delphi-based consensus survey of international experts. Cancer Med. 2023;12(13):14663–14673. doi:10.1002/cam4.6075. PMID:37264737.
16. IntraBio Inc. AQNEURSA (levacetylleucine) for oral suspension: prescribing information. Revision 9/2026 (NDA 219132/S-002). U.S. Food and Drug Administration. Label. Accessed 6 Oct 2026.
17. European Medicines Agency. Aqneursa (L-acetylleucine): European public assessment report (EMEA/H/C/006327); authorisation on 19 Jan 2026. EPAR. Accessed 6 Oct 2026.
18. European Reference Network on Rare Neurological Diseases (ERN-RND). Disease groups: cerebellar ataxias & hereditary spastic paraplegias. Institutional website. Accessed 6 Oct 2026.
19. European Reference Network on Rare Immunodeficiency, Autoinflammatory and Autoimmune Diseases (ERN RITA). Institutional website. Accessed 6 Oct 2026.
20. Coutinho P, Barbot C. Ataxia with Oculomotor Apraxia Type 1 — retired chapter, for historical reference only. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Last updated 19 Mar 2015. PMID:20301629. NBK1456.
21. Boder E, Sedgwick RP. Ataxia-telangiectasia; a familial syndrome of progressive cerebellar ataxia, oculocutaneous telangiectasia and frequent pulmonary infection. Pediatrics. 1958;21(4):526–554. PMID:13542097.
22. Waldmann TA, McIntire KR. Serum-alpha-fetoprotein levels in patients with ataxia-telangiectasia. Lancet. 1972;2(7787):1112–1115. doi:10.1016/s0140-6736(72)92717-1. PMID:4117204.
23. Taylor AM, Harnden DG, Arlett CF, Harcourt SA, Lehmann AR, Stevens S, et al. Ataxia telangiectasia: a human mutation with abnormal radiation sensitivity. Nature. 1975;258(5534):427–429. doi:10.1038/258427a0. PMID:1196376.
24. Gatti RA, Berkel I, Boder E, Braedt G, Charmley P, Concannon P, et al. Localization of an ataxia-telangiectasia gene to chromosome 11q22-23. Nature. 1988;336(6199):577–580. doi:10.1038/336577a0. PMID:3200306.
25. Savitsky K, Bar-Shira A, Gilad S, Rotman G, Ziv Y, Vanagaite L, et al. A single ataxia telangiectasia gene with a product similar to PI-3 kinase. Science. 1995;268(5218):1749–1753. doi:10.1126/science.7792600. PMID:7792600.
26. Verhagen MM, Last JI, Hogervorst FB, Smeets DF, Roeleveld N, Verheijen F, et al. Presence of ATM protein and residual kinase activity correlates with the phenotype in ataxia-telangiectasia: a genotype-phenotype study. Hum Mutat. 2012;33(3):561–571. doi:10.1002/humu.22016. PMID:22213089.
27. Russo S, Zielen S, Pietrucha B, Trischler J, Perlman S, Zielonka D, et al. Neurological effects of encapsulated dexamethasone sodium phosphate in children aged 6–9 years with ataxia telangiectasia (NEAT): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Child Adolesc Health. 2026;10(10):719–730. doi:10.1016/S2352-4642(26)00125-2. PMID:42600624.
28. Quince Therapeutics. Quince Therapeutics announces topline results from pivotal phase 3 NEAT clinical trial of eDSP in ataxia-telangiectasia (press release of 29 Jan 2026, exhibit 99.1 to form 8-K) and update of 12 Feb 2026. U.S. Securities and Exchange Commission. 8-K. Accessed 6 Oct 2026.
29. Martakis K, Bremova-Ertl T, Bolton C, Foltan T, Del Mar Garcia Romero M, Gautschi M, et al. Safety and efficacy of levacetylleucine in ataxia-telangiectasia: a phase 3, randomised, double-blind, placebo-controlled crossover trial. Lancet Neurol. 2026;25(7):633–644. doi:10.1016/S1474-4422(26)00158-4. PMID:42309084.
30. U.S. Food and Drug Administration. NDA 219132/S-002 supplement approval letter: Aqneursa (levacetylleucine) for the treatment of ataxia in adults and pediatric patients with ataxia-telangiectasia weighing ≥15 kg. Sep 2026. Approval letter; Drugs@FDA. Accessed 6 Oct 2026.
31. Nayler S, Foster S, Lavin M, Coman D. Recent advances in therapy for the neurodegenerative disorder ataxia-telangiectasia. Int J Mol Sci. 2026;27(16):7347. doi:10.3390/ijms27167347.
32. Veenhuis SJG, van Os NJH, Janssen AJWM, van Gerven MHJC, Coene KLM, Engelke UFH, et al. Nicotinamide riboside improves ataxia scores and immunoglobulin levels in ataxia telangiectasia. Mov Disord. 2021;36(12):2951–2957. doi:10.1002/mds.28788. PMID:34515380.
33. Presterud R, Deng WH, Wennerström AB, Burgers T, Gajera B, Mattsson K, et al. Long-term nicotinamide riboside use improves coordination and eye movements in ataxia telangiectasia. Mov Disord. 2024;39(2):360–369. doi:10.1002/mds.29645. PMID:37899683.
34. Kim J, Woo S, de Gusmao CM, Zhao B, Chin DH, DiDonato RL, et al. A framework for individualized splice-switching oligonucleotide therapy. Nature. 2023;619(7971):828–836. doi:10.1038/s41586-023-06277-0. PMID:37438524.
35. Reiman A, Srinivasan V, Barone G, Last JI, Wootton LL, Davies EG, et al. Lymphoid tumours and breast cancer in ataxia telangiectasia; substantial protective effect of residual ATM kinase activity against childhood tumours. Br J Cancer. 2011;105(4):586–591. doi:10.1038/bjc.2011.266. PMID:21792198.
36. van Os NJH, Chessa L, Weemaes CMR, van Deuren M, Fiévet A, van Gaalen J, et al. Genotype-phenotype correlations in ataxia telangiectasia patients with ATM c.3576G>A and c.8147T>C mutations. J Med Genet. 2019;56(5):308–316. doi:10.1136/jmedgenet-2018-105635. PMID:30819809.
37. National Center for Biotechnology Information. ClinVar: ATM variants on transcript NM_000051.4 — c.3576G>A (VCV000003035), c.5763-1050A>G (VCV000003021), c.7271T>G (VCV000003023) and c.7865C>T (VCV000449346). Accessed 6 Oct 2026.
38. McConville CM, Stankovic T, Byrd PJ, McGuire GM, Yao QY, Lennox GG, et al. Mutations associated with variant phenotypes in ataxia-telangiectasia. Am J Hum Genet. 1996;59(2):320–330. PMID:8755918.
39. Charlesworth G, Mohire MD, Schneider SA, Stamelou M, Wood NW, Bhatia KP. Ataxia telangiectasia presenting as dopa-responsive cervical dystonia. Neurology. 2013;81(13):1148–1151. doi:10.1212/WNL.0b013e3182a55fa2. PMID:23946315.
40. Saunders-Pullman R, Raymond D, Stoessl AJ, Hobson D, Nakamura K, Nakamura T, et al. Variant ataxia-telangiectasia presenting as primary-appearing dystonia in Canadian Mennonites. Neurology. 2012;78(9):649–657. doi:10.1212/WNL.0b013e3182494d51. PMID:22345219.
41. Stankovic T, Kidd AM, Sutcliffe A, McGuire GM, Robinson P, Weber P, et al. ATM mutations and phenotypes in ataxia-telangiectasia families in the British Isles: expression of mutant ATM and the risk of leukemia, lymphoma, and breast cancer. Am J Hum Genet. 1998;62(2):334–345. doi:10.1086/301706. PMID:9463314.
42. Telatar M, Teraoka S, Wang Z, Chun HH, Liang T, Castellvi-Bel S, et al. Ataxia-telangiectasia: identification and detection of founder-effect mutations in the ATM gene in ethnic populations. Am J Hum Genet. 1998;62(1):86–97. doi:10.1086/301673. PMID:9443866.
43. ESMO Handbook — Cancer Genomics. Haematological malignancies: genomics. Lugano: ESMO Press. p. 113. Local chapter.
44. Schon K, van Os NJH, Oscroft N, Baxendale H, Scoffings D, Ray J, et al. Genotype, extrapyramidal features, and severity of variant ataxia-telangiectasia. Ann Neurol. 2019;85(2):170–180. doi:10.1002/ana.25394. PMID:30549301.
45. van Os NJH, Hensiek A, van Gaalen J, Taylor AMR, van Deuren M, Weemaes CMR, et al. Trajectories of motor abnormalities in milder phenotypes of ataxia telangiectasia. Neurology. 2019;92(1):e19–e29. doi:10.1212/WNL.0000000000006700. PMID:30504431.
46. Firth HV, Hurst JA. Ataxic child. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 74–76. Local chapter.
47. Veenhuis SJG, van Os NJH, van Gerven MHJC, van Haaften L, Mulder EH, Weemaes CMR, et al. Dysarthria in children and adults with ataxia telangiectasia. Dev Med Child Neurol. 2021;63(4):450–456. doi:10.1111/dmcn.14811. PMID:33521952.
48. Lefton-Greif MA, Crawford TO, Winkelstein JA, Loughlin GM, Koerner CB, Zahurak M, et al. Oropharyngeal dysphagia and aspiration in patients with ataxia-telangiectasia. J Pediatr. 2000;136(2):225–231. doi:10.1016/s0022-3476(00)70106-5. PMID:10657830.
49. Hoche F, Daly MP, Chutake YK, Valera E, Sherman JC, Schmahmann JD. The cerebellar cognitive affective syndrome in ataxia-telangiectasia. Cerebellum. 2019;18(2):225–244. doi:10.1007/s12311-018-0983-9. PMID:30338439.
50. Nowak-Wegrzyn A, Crawford TO, Winkelstein JA, Carson KA, Lederman HM. Immunodeficiency and infections in ataxia-telangiectasia. J Pediatr. 2004;144(4):505–511. doi:10.1016/j.jpeds.2003.12.046. PMID:15069401.
51. Noordzij JG, Wulffraat NM, Haraldsson A, Meyts I, van't Veer LJ, Hogervorst FB, et al. Ataxia-telangiectasia patients presenting with hyper-IgM syndrome. Arch Dis Child. 2009;94(6):448–449. doi:10.1136/adc.2008.149351. PMID:19224889.
52. van Os NJH, Jansen AFM, van Deuren M, Haraldsson A, van Driel NTM, Etzioni A, et al. Ataxia-telangiectasia: immunodeficiency and survival. Clin Immunol. 2017;178:45–55. doi:10.1016/j.clim.2017.01.009. PMID:28126470.
53. Bodemer C, Sauvage V, Mahlaoui N, Cheval J, Couderc T, Leclerc-Mercier S, et al. Live rubella virus vaccine long-term persistence as an antigenic trigger of cutaneous granulomas in patients with primary immunodeficiency. Clin Microbiol Infect. 2014;20(10):O656–O663. doi:10.1111/1469-0691.12573. PMID:24476349.
54. Buchbinder D, Hauck F, Albert MH, Rack A, Bakhtiar S, Shcherbina A, et al. Rubella virus-associated cutaneous granulomatous disease: a unique complication in immune-deficient patients, not limited to DNA repair disorders. J Clin Immunol. 2019;39(1):81–89. doi:10.1007/s10875-018-0581-0. PMID:30607663.
55. Suarez F, Mahlaoui N, Canioni D, Andriamanga C, Dubois d'Enghien C, Brousse N, et al. Incidence, presentation, and prognosis of malignancies in ataxia-telangiectasia: a report from the French national registry of primary immune deficiencies. J Clin Oncol. 2015;33(2):202–208. doi:10.1200/JCO.2014.56.5101. PMID:25488969.
56. Bakhtiar S, Salzmann-Manrique E, Donath H, Woelke S, Duecker RP, Fritzemeyer S, et al. The incidence and type of cancer in patients with ataxia-telangiectasia via a retrospective single-centre study. Br J Haematol. 2021;194(5):879–887. doi:10.1111/bjh.17736. PMID:34337741.
57. Nissenkorn A, Levy-Shraga Y, Banet-Levi Y, Lahad A, Sarouk I, Modan-Moses D. Endocrine abnormalities in ataxia telangiectasia: findings from a national cohort. Pediatr Res. 2016;79(6):889–894. doi:10.1038/pr.2016.19. PMID:26891003.
58. Andrade IGA, Costa-Carvalho BT, da Silva R, Hix S, Kochi C, Suano-Souza FI, et al. Risk of atherosclerosis in patients with ataxia telangiectasia. Ann Nutr Metab. 2015;66(4):196–201. doi:10.1159/000430790. PMID:26045095.
59. de Leon CDA, Amantea SL, Pereira RA, Dantas EO, Loekmanwidjaja J, Costa-Carvalho BT, et al. Nutritional status and metabolic alterations in patients with ataxia-telangiectasia. Orphanet J Rare Dis. 2025;20(1):330. doi:10.1186/s13023-025-03785-2. PMID:40597142.
60. Petley E, Yule A, Alexander S, Ojha S, Whitehouse WP. The natural history of ataxia-telangiectasia (A-T): a systematic review. PLoS One. 2022;17(3):e0264177. doi:10.1371/journal.pone.0264177. PMID:35290391.
61. Mallott J, Kwan A, Church J, Gonzalez-Espinosa D, Lorey F, Tang LF, et al. Newborn screening for SCID identifies patients with ataxia telangiectasia. J Clin Immunol. 2013;33(3):540–549. doi:10.1007/s10875-012-9846-1. PMID:23264026.
62. Firth HV, Hurst JA. Immunodeficiency and recurrent infection. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 464–471 (table, p. 468). Local chapter.
63. Schoenaker MH, Suarez F, Szczepanski T, Mahlaoui N, Loeffen JL. Treatment of acute leukemia in children with ataxia telangiectasia (A-T). Eur J Med Genet. 2016;59(12):641–646. doi:10.1016/j.ejmg.2016.05.012. PMID:27238889.
64. Byrd PJ, Srinivasan V, Last JI, Smith A, Biggs P, Carney EF, et al. Severe reaction to radiotherapy for breast cancer as the presenting feature of ataxia telangiectasia. Br J Cancer. 2012;106(2):262–268. doi:10.1038/bjc.2011.534. PMID:22146522. Full text.
65. Stray-Pedersen A, Borresen-Dale AL, Paus E, Lindman CR, Burgers T, Abrahamsen TG. Alpha fetoprotein is increasing with age in ataxia-telangiectasia. Eur J Paediatr Neurol. 2007;11(6):375–380. doi:10.1016/j.ejpn.2007.04.001. PMID:17540590.
66. Moreira MC, Koenig M. Ataxia with Oculomotor Apraxia Type 2. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Updated 12 Jul 2018. PMID:20301333. NBK1154.
67. Saudubray JM, García-Cazorla Á. Clinical approach to inborn errors of metabolism in paediatrics. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, section 1.5 (Tables 1.23 and 1.29). doi:10.1007/978-3-662-63123-2_1.
68. Moreira MC, Barbot C, Tachi N, Kozuka N, Uchida E, Gibson T, et al. The gene mutated in ataxia-ocular apraxia 1 encodes the new HIT/Zn-finger protein aprataxin. Nat Genet. 2001;29(2):189–193. doi:10.1038/ng1001-189. PMID:11586300.
69. Bras J, Alonso I, Barbot C, Costa MM, Darwent L, Orme T, et al. Mutations in PNKP cause recessive ataxia with oculomotor apraxia type 4. Am J Hum Genet. 2015;96(3):474–479. doi:10.1016/j.ajhg.2015.01.005. PMID:25728773.
70. Stewart GS, Maser RS, Stankovic T, Bressan DA, Kaplan MI, Jaspers NG, et al. The DNA double-strand break repair gene hMRE11 is mutated in individuals with an ataxia-telangiectasia-like disorder. Cell. 1999;99(6):577–587. doi:10.1016/s0092-8674(00)81547-0. PMID:10612394.
71. Orphanet. Ataxia-telangiectasia-like disorder. ORPHA:251347 [Internet]. Disease entry. Accessed 6 Oct 2026.
72. Bajek A, Przewodowska D, Koziorowski D, Jędrzejowska M, Szlufik S. Cervical dystonia and no oculomotor apraxia as new manifestation of ataxia-telangiectasia-like disorder 1 — case report and review of the literature. Front Neurol. 2023;14:1243535. doi:10.3389/fneur.2023.1243535. PMID:37808486.
73. Varon R, Demuth I, Chrzanowska KH. Nijmegen Breakage Syndrome. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle: University of Washington; 1993–2026. Updated 18 Aug 2022; revision 30 Nov 2023. PMID:20301355. NBK1176.
74. Varon R, Vissinga C, Platzer M, Cerosaletti KM, Chrzanowska KH, Saar K, et al. Nibrin, a novel DNA double-strand break repair protein, is mutated in Nijmegen breakage syndrome. Cell. 1998;93(3):467–476. doi:10.1016/s0092-8674(00)81174-5. PMID:9590180.
75. Firth HV, Hurst JA. Ataxic adult. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 70–71. Local chapter.
76. Stastna B, Dolezalova T, Matejkova K, Nemcova B, Zemankova P, Janatova M, et al. Germline pathogenic variants in the MRE11, RAD50, and NBN (MRN) genes in cancer predisposition: a systematic review and meta-analysis. Int J Cancer. 2024;155(9):1604–1615. doi:10.1002/ijc.35066. PMID:38924040.
77. Zielen S, Crawford T, Benatti L, Magnani M, Kieslich M, Ryan M, et al. Safety and efficacy of intra-erythrocyte dexamethasone sodium phosphate in children with ataxia telangiectasia (ATTeST): a multicentre, randomised, double-blind, placebo-controlled phase 3 trial. Lancet Neurol. 2024;23(9):871–882. doi:10.1016/S1474-4422(24)00220-5. PMID:39152028.
78. Beyraghi-Tousi M, Sahebkar A, Houra M, Sarvghadi P, Jamialahmadi T, Bagheri R, et al. Efficacy and safety of N-acetyl-L-leucine in patients with ataxia telangiectasia: a randomized, double-blind, placebo-controlled, crossover clinical trial. Eur J Paediatr Neurol. 2024;50:57–63. doi:10.1016/j.ejpn.2024.04.009. PMID:38669738.
79. ClinicalTrials.gov. NCT06193200 (NEAT) and NCT06664853 (open-label extension, terminated: “eDSP development discontinued”). NCT06193200; NCT06664853. Accessed 6 Oct 2026.
80. European Medicines Agency. Orphan designation EU/3/13/1158: dexamethasone sodium phosphate encapsulated in human autologous erythrocytes for the treatment of ataxia telangiectasia (status: withdrawn). Designation. Accessed 6 Oct 2026.
81. Panvino F, Paparella R, Urciuolo G, Galosi S, Spagnoli C, Leuzzi V, et al. Pharmacological treatment of cerebellar ataxia in pediatric ataxia-telangiectasia: a systematic review. Eur J Neurol. 2026;33(7):e70683. doi:10.1111/ene.70683. PMID:42387837.
82. European Medicines Agency. Orphan designation EU/3/18/2124: acetylleucine for the treatment of ataxia telangiectasia (11 Jan 2019). Designation. Accessed 6 Oct 2026.
83. ClinicalTrials.gov. NCT07380165 — Expanded access program of levacetylleucine for ataxia-telangiectasia (IntraBio). Registry entry. Accessed 6 Oct 2026.
84. CSL Behring AG. PRIVIGEN (immune globulin intravenous [human], 10% liquid): prescribing information. DailyMed, version published on 6 May 2026. Prescribing information. Accessed 6 Oct 2026.
85. European Medicines Agency. Privigen: summary of product characteristics (EPAR, product information). SmPC. Accessed 6 Oct 2026.
86. European Medicines Agency. Orphan designation EU/3/26/3254: 3-(aminocarbonyl)-1-(2,3,5-tri-O-acetyl-beta-D-ribofuranosyl)-pyridinium chloride for the treatment of ataxia telangiectasia (22 Jun 2026; sponsor AdRes EU B.V.). Designation. Accessed 6 Oct 2026.
87. Bekis Bozkurt H, Spaccamela VM, Hassan M, Felber M, Zeilhofer U, Zeckanovic A, et al. Systematic review of hematopoietic stem cell transplantation in patients with ataxia telangiectasia: a case report and an individual patient-level analysis. Front Immunol. 2026;17:1754093. doi:10.3389/fimmu.2026.1754093. PMID:42051534.
88. Woelke S, Pommerening H, Kieslich M, Schubert R, Zielen S. Growth hormone treatment in patients with ataxia telangiectasia. Growth Factors. 2017;35(2–3):125–130. doi:10.1080/08977194.2017.1367681. PMID:28948852.
89. Lockman JL, Iskander AJ, Bembea M, Crawford TO, Lederman HM, McGrath-Morrow S, et al. Anesthetic and perioperative risk in the patient with ataxia-telangiectasia. Paediatr Anaesth. 2012;22(3):256–262. doi:10.1111/j.1460-9592.2011.03739.x. PMID:22098343.
90. ClinicalTrials.gov. NCT07215416 — A phase 1/2 study of antisense oligonucleotide therapy for treatment of ataxia-telangiectasia (atipeksen; ATM c.7865C>T). Registry entry. Accessed 6 Oct 2026.
91. ClinicalTrials.gov. NCT07820579 — Immunological, dermatological, and neurological effects of sirolimus in ataxia-telangiectasia. Registry entry. Accessed 6 Oct 2026.
92. Coman D, Jeggo P, Lavin M. Lets talk about ataxia-telangiectasia: meeting report of the AT clinical research conference June 2025. DNA Repair (Amst). 2025;155:103907. doi:10.1016/j.dnarep.2025.103907. PMID:41151249.
93. van Os NJ, Roeleveld N, Weemaes CM, Jongmans MC, Janssens GO, Taylor AM, et al. Health risks for ataxia-telangiectasia mutated heterozygotes: a systematic review, meta-analysis and evidence-based guideline. Clin Genet. 2016;90(2):105–117. doi:10.1111/cge.12710. PMID:26662178.
94. Erichsen AK, Koht J, Stray-Pedersen A, Abdelnoor M, Tallaksen CM. Prevalence of hereditary ataxia and spastic paraplegia in southeast Norway: a population-based study. Brain. 2009;132(Pt 6):1577–1588. doi:10.1093/brain/awp056. PMID:19339254.
95. Fares F, Axelord Ran S, David M, Zelnik N, Hecht Y, Khairaldeen H, et al. Identification of two mutations for ataxia telangiectasia among the Druze community. Prenat Diagn. 2004;24(5):358–362. doi:10.1002/pd.874. PMID:15164409.
96. Gilad S, Bar-Shira A, Harnik R, Shkedy D, Ziv Y, Khosravi R, et al. Ataxia-telangiectasia: founder effect among north African Jews. Hum Mol Genet. 1996;5(12):2033–2037. doi:10.1093/hmg/5.12.2033. PMID:8968760.
97. Mancebo E, Bernardo I, Castro MJ, Fernández-Martinez FJ, Barreiro E, De-Pablos P, et al. Rapid molecular prenatal diagnosis of ataxia-telangiectasia by direct mutational analysis. Prenat Diagn. 2007;27(9):861–864. doi:10.1002/pd.1787. PMID:17600866.
98. van Os NJH, van Deuren M, Weemaes CMR, van Gaalen J, Hijdra H, Taylor AMR, et al. Classic ataxia-telangiectasia: the phenotype of long-term survivors. J Neurol. 2020;267(3):830–837. doi:10.1007/s00415-019-09641-1. PMID:31776720.
99. Crawford TO, Skolasky RL, Fernandez R, Rosquist KJ, Lederman HM. Survival probability in ataxia telangiectasia. Arch Dis Child. 2006;91(7):610–611. doi:10.1136/adc.2006.094268. PMID:16790721.
100. Micol R, Ben Slama L, Suarez F, Le Mignot L, Beauté J, Mahlaoui N, et al. Morbidity and mortality from ataxia-telangiectasia are associated with ATM genotype. J Allergy Clin Immunol. 2011;128(2):382–389.e1. doi:10.1016/j.jaci.2011.03.052. PMID:21665257.
101. McGrath-Morrow SA, Rothblum-Oviatt CC, Wright J, Schlechter H, Lefton-Greif MA, Natale VA, et al. Multidisciplinary management of ataxia telangiectasia: current perspectives. J Multidiscip Healthc. 2021;14:1637–1644. doi:10.2147/JMDH.S295486. PMID:34234451.
102. Tiet MY, Guțu BI, Springall-Jeggo P, Coman D, Willemsen M, Van Os N, et al. Biomarkers in ataxia-telangiectasia: a systematic review. J Neurol. 2025;272(2):110. doi:10.1007/s00415-024-12766-7. PMID:39812834.