Alagille Syndrome

An autosomal dominant developmental disorder caused predominantly by JAG1 haploinsufficiency and, in a minority, NOTCH2 variants, combining intrahepatic bile duct paucity and chronic cholestasis with congenital heart disease affecting the pulmonary circulation and vertebral, ocular, facial, renal and vascular abnormalities, with highly variable expressivity within the same family.

Literature and regulatory information updated: 10 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Clinical description and historical delineation

Alagille syndrome (ALGS) was first described in 1969 by the French paediatric hepatologist Daniel Alagille and colleagues, in a communication on intrahepatic biliary atresia with patent extrahepatic bile ducts [7]. In 1973, Watson and Miller reported five families with 21 affected individuals in the United Kingdom, combining familial pulmonary artery stenosis with neonatal liver disease, and proposed the name arteriohepatic dysplasia, still used as a synonym [7, 17]. In 1975, Alagille, Odièvre, Gautier and Dommergues analysed 30 children with hepatic ductular hypoplasia and normal extrahepatic bile ducts and identified 15 forming a homogeneous group: chronic cholestasis, characteristic facies, a midsystolic murmur, vertebral arch defects, growth and developmental delay and hypogonadism; biochemical and histological findings distinguished this group from other childhood biliary diseases [18].

In 1987, the same team reviewed 80 cases of syndromic paucity of interlobular bile ducts, comparing them with 31 cases of non-syndromic paucity, and established the five major criteria that still structure clinical diagnosis: peculiar facies (95%), chronic cholestasis (91%), posterior embryotoxon (88%), butterfly vertebrae (87%) and peripheral pulmonary artery hypoplasia or stenosis, isolated or associated with complex heart disease (85%) [19]. In that series, 26 patients had the “complete” syndrome, with all five criteria, 42 had four and 12 had three; growth retardation occurred in approximately half, and 21 of the 80 patients (26%) died. Treatment at the time was limited to medium-chain triglycerides (MCTs), fat-soluble vitamins, colestyramine or phenobarbital, making these figures a picture of the era before paediatric liver transplantation and IBAT inhibitors [19]. The authors already considered autosomal dominant inheritance with variable expressivity.

Subsequent large North American and European series refined frequencies and natural history: Emerick and colleagues described 92 patients followed in Philadelphia in 1999, and Subramaniam and colleagues reviewed 25 years of experience at King's College Hospital, London, in 2011 [7, 20]. The international Global ALagille Alliance (GALA) cohort, published in 2023, with 1,433 children from 67 centres in 29 countries, became the main reference for hepatic natural history [9].

1.2. Discovery of the molecular basis

The association of the syndrome with deletions of the short arm of chromosome 20 was noted in the 1980s, when Byrne and colleagues reported patients with del(20p) and manifestations of arteriohepatic dysplasia [7]. In July 1997, two groups published identification of the gene in the same issue of Nature Genetics: Li and colleagues showed that JAG1, encoding a Notch receptor ligand, lay in the critical 20p12 region, that cytogenetically visible deletions were present in fewer than 7% of patients and that four families had frameshift variants, proposing haploinsufficiency as the mechanism [21]; Oda and colleagues narrowed the critical region to 250 kb by fluorescence in situ hybridisation (FISH) and found three frameshift variants, two splice donor site variants and one abolishing expression of the allele, reaching the same conclusion [22]. In 2006, McDaniell and colleagues, after identifying JAG1 variants in 94% of patients, investigated 11 probands without a variant in this gene and found NOTCH2 variants segregating with disease in two families, with five affected individuals, all with renal involvement; ALGS thus came to be considered a genetically heterogeneous disorder of the Notch pathway, with a second type, ALGS2 [23].

1.3. Nomenclature, classification and scope

Orphanet classifies Alagille syndrome (ORPHA:52) as a malformation syndrome and defines it as a rare developmental disorder characterised by the variable association of chronic cholestasis due to intrahepatic bile duct paucity, congenital heart disease including pulmonary artery stenosis, butterfly vertebrae, posterior embryotoxon, characteristic facies, frequent growth retardation, glomerular or tubular renal disease and diffuse arterial abnormalities [2]. Three subtypes are grouped under this entry: Alagille syndrome due to a JAG1 point variant (ORPHA:261619; OMIM 118450), Alagille syndrome due to a NOTCH2 point variant (ORPHA:261629; OMIM 610205) and Alagille syndrome due to a 20p12 microdeletion (ORPHA:261600; OMIM 118450), the last when the deletion encompasses JAG1 and sometimes contiguous genes [2]. In the GeneGuidance database, the ALGS2 entry is associated with code ORPHA:955 and Hajdu–Cheney syndrome with code ORPHA:261629; the codes are transposed, and the correct correspondence, confirmed in Orphanet, is ORPHA:261629 for NOTCH2-related ALGS and ORPHA:955 for Hajdu–Cheney syndrome [2, 5, 24].

This article treats the three entities as a single clinical disease, distinguishing them where the gene or mechanism changes the phenotype, testing or counselling. Hajdu–Cheney syndrome, caused by gain-of-function variants in the last exon of NOTCH2, is addressed only as an allelic disease and differential diagnosis. Non-syndromic tetralogy of Fallot with a JAG1 variant and the general approach to congenital heart disease are not developed; only aspects specific to heart disease in ALGS are described here.

2. Molecular Genetics and Pathophysiology

2.1. The Notch pathway and JAG1 and NOTCH2 proteins

JAG1 (Jagged canonical Notch ligand 1) lies at 20p12.2 and encodes jagged-1, a single-pass transmembrane ligand of Notch receptors; NOTCH2 (Notch receptor 2) lies at 1p12 and encodes one of the four mammalian Notch receptors, recognising JAG1, JAG2 and DLL1 [1, 24]. Signalling is juxtacrine: when JAG1, exposed on one cell's membrane, binds the extracellular domain of NOTCH2 on a neighbouring cell, the receptor undergoes successive proteolytic cleavages releasing the Notch intracellular domain (NICD); this moves to the nucleus, associates with the transcription factor RBPJ and activates target genes of the HES and HEY families [1, 24]. The name “Notch” comes from the notched wing of Drosophila flies with only one functional copy of the gene; in flies, absence of both copies is lethal, with nervous system hypertrophy [1].

The Notch pathway determines cell fates in several developing tissues, which explains the multisystem distribution of ALGS: it patterns the intrahepatic bile ducts, the pulmonary arterial tree, the somites giving rise to the vertebrae, the anterior chamber of the eye and the kidney, and also participates in haematopoiesis and angiogenesis [24]. In the liver, periportal mesenchymal cells express JAG1 and adjacent hepatoblasts express NOTCH2; NOTCH2 activation in these hepatoblasts directs them towards cholangiocytes and the formation of the ductal plate and interlobular ducts, and murine models demonstrate that this signalling is indispensable for biliary development [1]. JAG1 is also a ligand of NOTCH1 [21], a receptor involved in cardiac and valvular development; the relationship between JAG1–Notch signalling and right-sided heart defects extends to non-syndromic tetralogy of Fallot.

2.2. Pathogenic mechanism: haploinsufficiency

The predominant mechanism of JAG1-related ALGS is haploinsufficiency. Approximately 87% of disease-associated JAG1 variants are loss-of-function variants (nonsense, frameshift, splice, intragenic or whole-gene deletions), and many missense variants, particularly those removing or introducing cysteine residues in extracellular domains, produce a misfolded protein retained in the endoplasmic reticulum and absent from the membrane, or unable to activate NOTCH2 [1]. ClinGen assigned JAG1 a haploinsufficiency score of 3 (sufficient evidence of disease due to dosage loss), meaning that a whole-gene deletion has the same significance as a truncating variant [24]. Deletion of one entire JAG1 copy, initially recognised by cytogenetics, is the most direct evidence that half the gene dosage suffices to cause disease [21, 22].

In NOTCH2-related ALGS, variants are predominantly missense and the pathogenicity of each is less well characterised; the accepted mechanism is also loss of function, in contrast to the gain of function causing Hajdu–Cheney syndrome [1, 24]. The resulting tissue effects are deficient formation and progressive involution of interlobular bile ducts, with cholestasis, and diffuse arterial vasculopathy, with stenoses, hypoplasias and aneurysms; the Jag1Ndr/Ndr mouse model shows sporadic cerebral haemorrhages, a thin skull, tortuous vessels and deficient arterial smooth muscle coverage, worsened by hypertension [25].

2.3. Genetic architecture and proportion attributed to each gene

Table 2.1 reproduces the molecular distribution published in GeneReviews, based on the series by Gilbert and colleagues (2019). Percentages refer to proportions of clinically diagnosed probands in whom each variant type was found; they are not test sensitivity values.

Table 2.1. Molecular distribution of Alagille syndrome [1]

Reference sources differ in this proportion. Smith's textbook reports JAG1 in 89% of patients, whole-gene deletion in fewer than 7% and NOTCH2 in fewer than 1%; the Oxford Desk Reference reports 95% for JAG1, with up to 7% of 20p12 deletions detectable by FISH; and the Brazilian textbook Síndromes Genéticas: Reconhecer para Tratar reports 98% for JAG1 and 2% for NOTCH2 [6–8]. This article gives precedence to the GeneReviews distribution (94.3%, 2.5% and 3.2%), because it is the most recent and specifies the case series and methods; differences result from the era of the series, inclusion criteria (strict clinical diagnosis or incomplete phenotype) and technical sensitivity in each period. In the GALA genetic cohort, with 952 molecularly diagnosed individuals, 902 had a JAG1 variant and 34 a NOTCH2 variant [26]; in ASSERT, 92.3% of participants had a JAG1 variant and 7.7% a NOTCH2 variant [14]. When no variant is found in either gene, possibilities include an as yet undetectable variant in these genes (regulatory region, introns, copy-number-neutral rearrangement), a variant in another Notch pathway gene or mosaicism, although none of these hypotheses has been demonstrated [1].

2.4. Variant spectrum and nomenclature

Variants should be described relative to MANE Select transcripts, NM_000214.3 for JAG1 and NM_024408.4 for NOTCH2 [3]. Many older reports and articles use earlier transcript versions (for example, NM_000214.2) or protein numbering without the transcript prefix; before comparing a variant with databases, it must be normalised to the current version. In the published case described in section 10, the variant cited in the article as NM_000214.2:c.2122_2125del corresponds, with the same coding description, to NM_000214.3:c.2122_2125del p.(Gln708ValfsTer34), NC_000020.11:g.10645249_10645252del (GRCh38), classified as pathogenic in ClinVar by several submitters without conflict [3, 27, 28].

The JAG1 spectrum includes whole-gene or exon deletions, nonsense and frameshift variants distributed throughout the gene, splice variants and missense variants concentrated in extracellular receptor-binding domains [1, 8]. The Korean Seoul cohort, with 41 patients with a JAG1 variant, showed 34.1% frameshift, 26.8% nonsense, 24.4% missense, 9.8% splice and 4.9% in-frame deletions, without structural variants, differing from the GALA distribution (p=0.043) [29].

Interpreting missense variants is the main difficulty. Gilbert and colleagues constructed a library of 2,832 nucleotide variants in JAG1 exons 1 to 7 and measured expression of each protein on the cell membrane, calibrating the assay with 175 variants of known classification: 486 variants had abnormal or probably abnormal expression, 90.3% of them missense; up to 85% of JAG1 missense variants described in patients were classified as uncertain or conflicting, and of 144 patient variants of uncertain significance (VUS), 27 had abnormal expression and 26 could be reclassified as likely pathogenic [30]. For NOTCH2, GALA reviewed 61 published variants and reclassified 19 (31.1%) from pathogenic or likely pathogenic to VUS, retaining the original classification in only 21 (34.4%); the authors proposed American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) rules adapted to this gene [26]. The variant in the first case in section 10, NM_000214.3(JAG1):c.820G>T p.(Gly274Cys), NC_000020.11:g.10652534C>A (GRCh38), illustrates the problem: the article classifies it as likely pathogenic, but ClinVar records it as of uncertain significance, with a single submitter and criteria provided, evaluated in July 2021 [27, 31]. The substitution introduces a cysteine into an extracellular domain, the type of alteration that functional assays frequently show to disrupt JAG1 expression [30], but without such data for this specific variant the discrepancy remains and should be communicated.

2.5. Contiguous 20p12 deletions and mosaicism

Approximately 7% of patients have a 20p12 deletion including the entire JAG1 gene; when the deletion is larger and encompasses neighbouring genes, ALGS may be associated with developmental delay, deafness and autism spectrum disorder, which are not part of the usual presentation [1, 8]. When a whole-gene deletion is found, a full cytogenetic study of the proband is recommended, because the deletion rarely results from a translocation or inversion, with implications for recurrence risk [1].

Parental mosaicism is a relevant feature of ALGS. In a screen of 61 families with a JAG1 variant, Giannakudis and colleagues identified five cases of parental mosaicism, more than 8.2%, and observed that mosaic parents could have a very mild phenotype [4]. Asymptomatic parents mosaic for the 20p deletion have also been described [1]. Counselling implications are addressed in section 8.

2.6. Penetrance and expressivity

ALGS expressivity is among the most variable of autosomal dominant disorders. Kamath and colleagues assessed 53 relatives carrying the JAG1 variant of 34 probands: 11 (21%) had a presentation that would itself lead to diagnosis; 17 (32%) had mild signs detected only after targeted assessment; 25 (47%) did not meet clinical criteria; and only 2 had no signs, corresponding to a penetrance of 96% (51/53) [1, 32]. Characteristic facies was the sign with the highest penetrance in this series, and cardiac and hepatic disease frequencies were lower in secondarily identified relatives than in probands [8, 32]. In NOTCH2-related ALGS, penetrance appears complete in the individuals described, although numbers are small [1]. The Brazilian textbook of genetic syndromes describes inheritance as having “incomplete penetrance” [6]; Kamath's data show that penetrance, understood as the presence of any sign, is high, and that severity is chiefly what varies.

2.7. Genotype–phenotype correlations and differences between genes

There is no correlation between the type or location of the JAG1 variant and heart defect type, hepatic severity or renal involvement; identical variants produce discordant presentations within the same family [1, 33, 34]. In the Korean cohort, 4 of the 5 deaths occurred in patients with frameshift variants (p=0.035), a single-centre association that has not been replicated [29].

Differences between genes are more consistent. In GALA, the 34 patients with NOTCH2 had lower frequencies of characteristic facies, posterior embryotoxon, cardiac involvement and butterfly vertebrae than the 902 with JAG1 (p<0.001), without a difference in native liver survival or overall survival [26]. The practical implication is that the classical definition, based mainly on the JAG1 phenotype, may miss patients with a NOTCH2 variant, and testing should include both genes even when extrahepatic findings are sparse [26].

2.8. Allelic diseases, disputed phenotypes and somatic alterations

Table 2.2. Allelic phenotypes of JAG1 and NOTCH2 [1, 5, 24]

Hajdu–Cheney syndrome illustrates how the same gene can cause opposite diseases through the mechanism. Truncating variants in exon 34 escape nonsense-mediated messenger RNA decay (NMD) and produce a NOTCH2 intracellular domain without the PEST domain, which is no longer degraded by the SCF/FBW7 ubiquitin ligase complex; signalling becomes excessive and sustained, with increased osteoclastic bone resorption [24]. Clinically, it is a progressive bone disease: acro-osteolysis of distal phalanges, osteoporosis with fractures and vertebral collapse, short stature, wormian bones, platybasia and basilar invagination with sometimes fatal neurological complications, delayed suture closure, early loss of permanent teeth, cleft palate, serpentine fibulae, joint laxity, heart defects and polycystic kidneys [1, 24]. There is no curative treatment; bisphosphonates, including zoledronic acid, have been reported to increase bone mineral density without established clinical benefit, and an antisense oligonucleotide targeting Notch2 has been studied only in mice [24].

Somatic NOTCH2 variants in sporadic splenic marginal zone lymphoma do not confer hereditary predisposition and are unrelated to germline ALGS [1, 24].

3. Clinical Spectrum and Manifestations by System

3.1. Overview and frequency of manifestations

The ALGS spectrum extends from potentially fatal hepatic or cardiac disease in the first year of life to subclinical manifestations identifiable only through targeted assessment, such as butterfly vertebrae, posterior embryotoxon or facial features; some carriers have an isolated finding, such as a heart defect or aneurysmal disease [1]. Published frequencies depend greatly on recruitment: paediatric hepatology centre series select patients with liver disease, and series of relatives or carriers identified by genetic testing show much lower rates. In one series not recruited by organ, only 61% of individuals had liver disease (da Palma et al. 2021, according to GeneReviews) [1]. Table 3.1 compares sources and their respective populations.

Table 3.1. Frequency of manifestations in different series

Smith's textbook presents similar frequencies, derived mainly from the Emerick and Subramaniam series, without an explicit denominator: typical facies 95%, embryotoxon 88%, pulmonary branch stenosis 67%, butterfly vertebrae 87%, duct paucity 85%, chronic cholestasis 96%, structural renal abnormalities 39% and functional abnormalities 74%, some degree of pancreatic insufficiency 41% and neurovascular events 15% [7].

3.2. Liver and bile ducts

Most patients with hepatic expression have neonatal jaundice with conjugated hyperbilirubinaemia; cholestasis develops in the first three months in 44% and between 4 months and 3 years in the remainder, and manifests as pruritus, acholic or hypocholic stools, xanthomas and hepatomegaly [7]. In GALA, 85% of children had a history of neonatal cholestasis; pruritus developed in 74%, with a median onset age of 12 months, and xanthomas in 24%, with a median age of 25 months [1, 9]. Laboratory findings include increased serum bile acids, alkaline phosphatase, gamma-glutamyltransferase (GGT), cholesterol and triglycerides, elevated aminotransferases, fat-soluble vitamin deficiency and malnutrition [1]. Hypercholesterolaemia is often marked and is mainly due to lipoprotein X, which is not considered atherogenic [10]. Pruritus is frequently the most disabling manifestation, with excoriations, sleep disturbance, fatigue and impact on the family, and is among the most frequent indications for liver transplantation [1, 36].

Interlobular bile duct paucity is progressive and may be absent in neonates: before six months, biopsy may show ductular proliferation or a neonatal hepatitis pattern, potentially leading to misdiagnosis as biliary atresia [1, 7]. In Emerick's series, duct paucity and fibrosis increased with age [20]. Extrahepatic bile duct involvement, with hypoplasia, occurs in approximately 20% [7].

The hepatic natural history of disease untreated with IBAT inhibitors is described mainly by GALA, a retrospective study of 1,433 children (57% boys) born between January 1997 and August 2019, followed at 67 centres in 29 countries [9]. Native liver survival was 54.4% at 10 years and 40.3% at 18 years, and by 10 and 18 years, 51.5% and 66.0% had experienced at least one adverse hepatic event (clinically evident portal hypertension, transplantation or death) [9]. Among children with neonatal cholestasis, 50.4% underwent transplantation by 18 years, with a median age of 2.8 years and 72% of transplants in the first five years; cholestasis was the primary indication in 72% and portal hypertension in 30%, the latter more frequent in older children, and more than half had more than one indication [1]. This is a cohort from paediatric reference centres, which does not measure liver disease frequency in the carrier population.

Bilirubin levels in the second half of the first year of life distinguish different trajectories. In GALA, compared with median total bilirubin below 5.0 mg/dL between 6 and 12 months, bilirubin between 5.0 and 10.0 mg/dL was associated with a 4.1-fold higher risk of clinically evident portal hypertension (95% confidence interval, 95% CI, 1.6–10.8) and a 4.8-fold higher risk of transplantation (2.4–9.7), and bilirubin of 10.0 mg/dL or more with risks 8.0-fold (3.4–18.4) and 15.6-fold (8.7–28.2) higher; 79% of children with bilirubin below 5.0 mg/dL reached adulthood with their native liver [9]. In an analysis of 570 GALA patients, serum bile acids correlated moderately with bilirubin (r=0.47), and values below 102 µmol/L, used as a time-dependent covariate, predicted better native liver survival (hazard ratio, HR, 3.78; 95% CI 2.39–5.99) and event-free survival (HR 3.44; 2.35–5.04); when median bile acids in the first year exceeded 102 µmol/L, native liver survival at 7 years was 67.2% versus 83.5% (p=0.05) [37]. These data concern patients untreated with IBAT inhibitors; they suggest, without proving, that lowering bile acids could improve outcomes.

Cirrhosis and portal hypertension develop in some patients, with splenomegaly, thrombocytopenia and varices, and were the most frequent transplantation indication at older ages [1, 20]. There is an increased risk of hepatocellular carcinoma (HCC) in children and adults with a JAG1 or NOTCH2 variant, even without cirrhosis and even in carriers without apparent liver disease [1, 38]. Schindler and colleagues described HCC in a 58-year-old woman with a pathogenic JAG1 variant and no evident hepatic manifestations and reviewed published cases [38]; according to the review by Ayoub and colleagues, 21 paediatric and 14 adult cases had been described, without cirrhosis in 11% of children and 67% of adults, and with normal alpha-fetoprotein (AFP) in 38%, making AFP alone unreliable; large regenerative nodules, over 5 cm, may mimic HCC [39].

3.3. Heart and pulmonary circulation

Cardiovascular involvement is almost constant: in a series of 200 individuals with a JAG1 variant and/or ALGS, McElhinney and colleagues found cardiovascular involvement in 94%, 75% with an abnormality documented by imaging and 19% with only a peripheral pulmonary stenosis murmur; lesions were right-sided in 123, left-sided in 22 and bilateral in 12, with pulmonary artery branch stenosis or hypoplasia in 76% and tetralogy of Fallot in 23 patients, 8 of whom had pulmonary atresia [33]. Involvement of the pulmonary vasculature (valve, trunk and branches) reaches up to 94%, peripheral or branch pulmonary stenosis 67%, and tetralogy of Fallot, in 7–16%, is the most common complex defect; ventricular septal defect (VSD), atrial septal defect (ASD), aortic stenosis and coarctation of the aorta follow in descending order [1]. The Oxford Desk Reference reports approximately 90% with one or several peripheral pulmonary artery stenoses and one-third with other malformations, mainly tetralogy of Fallot [40]. In Emerick's series, intracardiac disease was associated with higher mortality (p<0.001) [20].

3.4. Systemic and intracranial vascular abnormalities

Vasculopathy is the component of ALGS with the greatest impact on mortality outside the liver and heart. Among 268 individuals, Kamath and colleagues found 25 (9%) with non-cardiac vascular abnormalities or vascular events: 16 with structural abnormalities (2 basilar artery aneurysms, 7 internal carotid abnormalities, 1 middle cerebral artery aneurysm, 1 moyamoya syndrome, 3 aortic aneurysms, 2 coarctations and 1 renal artery stenosis) and 9 with intracranial events without a documented abnormality; vascular events accounted for 34% of mortality [41]. Neurovascular events affect up to 15% of patients, and renovascular abnormalities, mid-aortic syndrome and visceral artery aneurysms and stenoses have also been described [1, 7]. Smith's textbook states that moyamoya syndrome may be present in one-third of patients [7]; this proportion has no counterpart in the largest series of vascular abnormalities, where moyamoya occurred in one patient among 268 [41], and should not be used as a frequency estimate. Moyamoya disease and monogenic moyamoya angiopathy are part of the differential diagnosis.

Two unrelated families with a JAG1 variant had only aneurysmal disease, without the other syndrome signs [1]. According to Ayoub's review, cerebrovascular abnormalities were found in 50% of screened asymptomatic patients and almost all symptomatic patients, and vascular abnormalities, particularly with intracranial haemorrhage, account for most non-cardiac deaths in 25–50% of paediatric cohorts; in adults, five cases of subarachnoid haemorrhage between 21 and 30 years had been described, with death in two, three of them due to vertebrobasilar aneurysm rupture [39]. A systematic review found more girls than boys with spontaneous intracranial haemorrhage and described increased vascular tortuosity on patient retinal photographs, leading to the proposal of retinal photography as a non-invasive vascular assessment, still under investigation [25].

3.5. Eye

Posterior embryotoxon, a prominent and anteriorly displaced Schwalbe line, is an anterior chamber defect visible on slit-lamp examination, present in 78–89% of patients; it does not affect visual acuity and occurs in 8–15% of the general population, so it is supportive rather than diagnostic [1]. Axenfeld and Rieger anomalies, optic disc drusen (90% of 20 children examined by ocular ultrasound), retinal pigmentary changes (32% in one study), ectopic pupils, strabismus, choroidal folds and refractive errors have been described [1, 7]. The Brazilian textbook also reports congenital glaucoma, retinal atrophy and high myopia [6]. Visual prognosis is generally good, but some patients have peripheral chorioretinal changes and visual field loss, macular atrophy and progressive visual loss [1].

3.6. Skeleton and bone

Butterfly vertebrae result from failure of fusion of the halves of the vertebral body, mainly in the thoracic spine, and are generally asymptomatic; associated findings may include hemivertebrae, spina bifida occulta, rib abnormalities and absence of the normal increase in interpedicular distance from L1 to L5 [1, 7]. Short distal phalanges, fifth-finger clinodactyly, a short ulna, radioulnar synostosis and clubfoot occasionally occur [7]. Craniosynostosis, usually unilateral coronal, occurs in 0.9% of patients versus 0.03% of the general population [1]. Fracture risk is high, particularly in the long bones of the lower limbs, with significant bone mineral deficiency on dual-energy X-ray absorptiometry (DXA), related to cholestasis, vitamin D malabsorption and malnutrition [1, 7].

3.7. Face and oral cavity

The characteristic face, described as an “inverted triangle”, has a broad forehead, deep-set eyes with moderate hypertelorism, a straight or concave nasal bridge with a bulbous tip and a pointed chin; it is almost universal in patients with JAG1, less frequent with NOTCH2, and becomes more evident with age, with prognathism in adults [1, 26, 40]. Figures in Smith's textbook show an eighteen-month-old child with a broad forehead and prominent chin and auricular xanthomas [7]. Facial recognition depends on population: North American dysmorphologists had difficulty recognising the face in a cohort of Vietnamese children [1]. Oral findings include altered mandibular morphology, hypodontia, abnormal tooth structure, greenish discolouration of deciduous teeth attributed to exposure to elevated bilirubin during calcification, gingival inflammation and palatal and gingival xanthomas [1, 6, 7].

3.8. Kidney

In an analysis of 466 patients with a JAG1 variant, 187 had evaluable renal information, and 73 (39%) had renal involvement; renal dysplasia was the most frequent abnormality, without an association with variant type, and the authors proposed renal involvement as a sixth defining criterion [34]. Structural abnormalities include small hyperechogenic kidneys, cysts and pelviureteric junction obstruction; functional abnormalities mainly include renal tubular acidosis (RTA), reduced creatinine clearance and mesangiolipidosis [1, 7]. Hypertension and renal artery stenosis have been described in adults [1]. Before liver transplantation, glomerular filtration rate (GFR) was below 90 mL/min/1.73 m² in 18% of children with ALGS versus 5% of children with biliary atresia [42], and renal function tends to deteriorate in the long term after transplantation [43]. In patients with NOTCH2, renal involvement was the common feature of the first families described [23].

3.9. Growth, nutrition and pancreas

Growth impairment affects 50–90% of patients and is attributed to malabsorption of fat and vitamins, cholestasis and insufficient intake [1]. The GALA-specific growth charts, constructed from 1,204 term children with native livers for weight and 1,106 for height, show median birth weights of 2.8 kg in boys and 2.6 kg in girls and 50th-percentile height at 18 years of 171.5 cm and 156.5 cm, versus 176 cm and 163 cm on Centers for Disease Control and Prevention (CDC) charts [44]. Some degree of exocrine pancreatic insufficiency has been described in 41% [7], and the Brazilian textbook includes type 1 diabetes among manifestations [6].

3.10. Neurodevelopment, endocrinology and other features

In Emerick's series, 16% had mild gross motor delay and 2% mild intellectual disability [1]; more marked delay or disability suggests a contiguous 20p12 deletion [1]. Attention deficits and executive dysfunction are more frequent, and early screening is recommended [1]. Special education was required in 50% of transplanted children with ALGS versus 30% transplanted for biliary atresia [42]. Delayed puberty, hypogonadism, a high-pitched voice, growth hormone deficiency, hypothyroidism, splenomegaly, cleft palate, inner ear abnormalities and thyroid cancer have been described [1, 7]. The Brazilian textbook also reports dystonia and tremor, linked by Deveny and colleagues to basal ganglia hyperintensity and elevated blood manganese, and intracranial hypertension [6].

3.11. Disease in adulthood

Adults with ALGS fall into two groups: those with childhood disease, transplanted or with a native liver, and mildly affected carriers recognised after a child's diagnosis. In the latter, liver disease may be minimal, but risks of HCC, intracranial vascular events, renal progression and pregnancy complications persist [38, 39]. Adult data are scarce and come from small series and expert reviews.

4. Clinical Recognition and Warning Signs

4.1. In neonates and infants

The most frequent presentation is prolonged neonatal jaundice with conjugated hyperbilirubinaemia, sometimes with hypocholic or acholic stools, hepatomegaly and poor weight gain. In neonatal cholestasis, ALGS should be considered when accompanied by a cardiac murmur or known heart disease (particularly pulmonary branch stenosis or tetralogy of Fallot), suggestive facial features, elevated cholesterol, butterfly vertebrae on chest radiography, posterior embryotoxon or renal abnormalities [1, 8]. GGT is usually elevated in ALGS, unlike progressive familial intrahepatic cholestasis (PFIC) types 1 and 2, arthrogryposis, renal dysfunction and cholestasis (ARC) syndrome and bile acid synthesis defects, in which it is low or normal [8]. The most important warning sign is suspected biliary atresia: a hypoplastic gallbladder, absent excretion on scintigraphy and non-visualisation of bile ducts on cholangiography are frequent in ALGS and may lead to a Kasai portoenterostomy that worsens prognosis (sections 5.3 and 7.8) [45, 46].

4.2. In older children

In children, suspicion arises with intense pruritus with excoriations and sleep disturbance, xanthomas on extensor surfaces, palmar creases and auricles, growth impairment, fractures after minimal trauma, peripheral pulmonary stenosis discovered through a murmur, or renal dysplasia or RTA without an apparent cause [1, 6, 7]. The face becomes more recognisable with age. A child with duct paucity and developmental delay, deafness or autism spectrum disorder should prompt consideration of a contiguous 20p12 deletion [1].

4.3. In adolescents and adults

In adults, ALGS is recognised mainly in three settings: the parent of a diagnosed child, with mild signs (slightly abnormal liver tests, butterfly vertebrae, embryotoxon, facies); detection of unexplained peripheral pulmonary stenosis or intracranial or aortic aneurysmal disease; and HCC in a non-cirrhotic liver [1, 32, 38]. Assessment of a potential related living donor for liver transplantation is another context in which carriers are identified [1, 43].

4.4. Emergency signs

Emergency signs are sudden severe headache, focal neurological deficits, seizures or altered consciousness, requiring exclusion of intracranial haemorrhage or ischaemic stroke by magnetic resonance imaging (MRI), MR angiography (MRA) and, if necessary, conventional angiography [1]. Variceal gastrointestinal bleeding, ascites and encephalopathy indicate hepatic decompensation, contraindicate IBAT inhibitors according to the American label and warrant referral to a transplant centre [10, 11, 13]. Spontaneous haemorrhages are an important cause of death in paediatric ALGS [25]. In patients with heart disease, right heart failure or cyanosis also warrant urgent assessment.

4.5. Neonatal screening

There is no specific neonatal screening for ALGS. Early detection depends on systematic assessment of prolonged neonatal jaundice, with conjugated bilirubin measurement, the common entry point for all neonatal cholestases [8].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Clinical criteria

The classical criteria, derived from Alagille's 1987 series, require interlobular bile duct paucity on liver biopsy and at least three of the five major criteria: cholestasis, heart disease (mainly peripheral pulmonary artery stenosis), skeletal abnormalities (mainly butterfly vertebrae), ocular abnormalities (mainly posterior embryotoxon) and characteristic facies [7, 19]. GeneReviews, based on criteria published by Mitchell and colleagues (2018) and Kohut and colleagues (2021), accepts a clinical diagnosis when there is duct paucity and any three of the five major criteria, or one of the five criteria in an individual with a family history of ALGS; the molecular diagnosis is established with suggestive findings and a heterozygous pathogenic or likely pathogenic variant in JAG1 or NOTCH2 [1]. Absence of a family history does not exclude the diagnosis, given the frequency of de novo variants and mildly affected parents [1]. The French national biliary atresia protocol (PNDS (French National Diagnosis and Care Protocol) of the Haute Autorité de Santé, 2024) uses a different formulation: it defines paucity as ductopenia in more than half of the portal tracts, describes the facies by a prominent forehead, a small pointed chin and hypertelorism, and accepts a clinical diagnosis with cholestasis and at least two of the other four major criteria, treating elevated GGT and duct paucity on histology as supporting elements rather than as a mandatory condition [16]. In practice, this threshold coincides with the three-of-five criteria of GeneReviews, but requires one of them to be cholestasis and dispenses with biopsy, which brings it close to the EASL position on histology (section 5.2). Renal involvement has been proposed as a sixth criterion [34]. The clinical criteria have known limitations: JAG1 variants have been found in one-third of patients with very incomplete presentations, with one or two features, mainly cardiac or hepatic [7], and patients with NOTCH2 variants have fewer extrahepatic manifestations [26].

Table 5.1. Major criteria and assessment methods [1, 19, 20, 33]

5.2. Liver biopsy

Duct paucity is the most consistent finding, but is late: it is present in only 65% of biopsies in the first three months of life, and is eventually found in approximately 95% of patients, possibly an overestimate due to ascertainment bias [1]. The 2024 European Association for the Study of the Liver (EASL) guideline recommends against routine liver biopsy to determine prognosis or guide liver disease management in ALGS (level of evidence 3, strong recommendation, strong consensus), reserving it for limited situations, such as VUS interpretation [10]. The rationale rests on the low sensitivity of duct paucity in infants and the value of non-invasive markers: according to EASL, in a series of 144 children (Mouzaki et al.), fibrosis on biopsy before five years tripled the risk of a severe hepatic outcome, and in a series of 40 children (Shiau et al.), the aspartate aminotransferase-to-platelet ratio index (APRI) was higher at fibrosis stages F3–F4 than F0–F2 (3 versus 1.7; p=0.029) [10]. GeneReviews continues to describe duct paucity as the central criterion for clinical diagnosis [1]; the two positions are compatible if biopsy is understood as unnecessary when genetic testing confirms diagnosis and useful when it is inconclusive.

5.3. Investigation of neonatal cholestasis and distinction from biliary atresia

Neonatal cholestasis has more than a hundred causes; treatable causes (sepsis, hypothyroidism, classical galactosaemia) should be excluded first, and biliary atresia, the most frequent identifiable cause, must be diagnosed in time for early portoenterostomy [1]. ALGS may be confused with biliary atresia because of biochemical, histological and imaging overlap. Radiotracer excretion into the duodenum on hepatobiliary scintigraphy excludes biliary atresia, but absent excretion is frequent in ALGS (in Emerick's series, more than 60% of 36 patients had no excretion at 24 hours) [20, 45]. On cholangiography, 37% of ALGS patients do not show proximal extrahepatic ducts and another 37% have hypoplastic ducts, potentially leading to Kasai portoenterostomy, which, according to the authors, probably worsens hepatic prognosis [45]. Ayoub and Kamath recommend, when time permits and examinations are inconclusive, rapid simultaneous molecular analysis of JAG1 and NOTCH2 (3 to 4 weeks) and state that serum matrix metalloproteinase 7 (MMP-7) is superior to other markers for diagnosing biliary atresia [45].

The French biliary atresia PNDS (French National Diagnosis and Care Protocol), drafted by the reference centre for biliary atresia and genetic cholestasis, attributes 10 to 15% of neonatal cholestasis to ALGS, compared with 25 to 40% for biliary atresia, 5 to 10% for biallelic alpha-1 antitrypsin deficiency and approximately 10% for transient neonatal cholestasis [16]. These proportions come from the experience of specialised centres, which receive already selected cases, and the protocol gives neither denominator nor period; they are not population frequency estimates. The proportion attributed to ALGS is much higher than in consecutive series: in a systematic review of 17 series with 1,692 infants with conjugated hyperbilirubinaemia, ALGS accounted for 1.0% (16 cases) and interlobular duct paucity recorded without a syndromic diagnosis for 2.5%, compared with 25.9% for extrahepatic biliary atresia and 4.1% for alpha-1 antitrypsin deficiency, with heterogeneity of investigation protocols and populations [47]. The PNDS estimate of 10–15% should therefore not be used as a pre-test probability [16, 47]. In the newborn, the protocol organises the work-up into two groups to be explored first: causes requiring urgent treatment (biliary atresia, urinary tract infection or sepsis, cortisol deficiency, obstruction of the common bile duct) and easily detected causes, among which it places ALGS, alongside choledochal cyst, cystic fibrosis and alpha-1 antitrypsin deficiency [16]. On history and examination, the combination of facial dysmorphism, a heart murmur and posterior embryotoxon points towards ALGS, and the presence of other affected relatives towards an autosomal dominant disease [16]. The proposed initial laboratory work-up includes a full blood count with platelets, prothrombin time before and after vitamin K and factor V, aminotransferases, GGT, total and conjugated bilirubin, total bile acids, total cholesterol, blood glucose, vitamins A, D and E, alpha-1 antitrypsin level and verification of the newborn screening immunoreactive trypsin or a sweat test; vitamin K should be given by injection as soon as cholestasis is recognised, to prevent haemorrhage [16]. The cholestasis threshold is a conjugated bilirubin above 17 µmol/L (1 mg/dL); the relative fraction appears in the protocol text sometimes as above 10% and sometimes as above 20% of total bilirubin, so the absolute threshold is the operative criterion [16].

Regarding imaging, the PNDS considers hepatobiliary scintigraphy and magnetic resonance cholangiopancreatography useless for confirming or excluding biliary atresia and advises against them because they delay portoenterostomy; the test of patency is cholangiography, performed surgically when suspicion is strong, by percutaneous transhepatic cholecystography when the gallbladder is patent and of sufficient volume, or, more rarely, by the endoscopic retrograde route when suspicion is low [16]. A gallbladder visible on ultrasound does not exclude biliary atresia, and, in ALGS, an atrophic gallbladder may wrongly point towards it [16]. Histological signs of obstruction (bile plugs, ductular proliferation, portal fibrosis) are not specific to biliary atresia, and immunohistochemistry with antibodies against the proteins involved in genetic cholestasis helps guide the diagnosis once atresia has been excluded [16].

In practice, in neonatal cholestasis with elevated GGT, targeted ALGS assessment before any surgery includes echocardiography, chest and spine radiographs, slit-lamp ophthalmological examination, renal ultrasound and facial assessment [8, 46]. The Oxford Desk Reference recommends genomic analysis (panel, exome or genome) when the presentation suggests a syndromic diagnosis [8].

5.4. Systemic assessment after diagnosis

Table 5.2. Recommended assessments after diagnosis [1]

In adults with suspected ALGS, including relatives undergoing screening, Ayoub and colleagues propose alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase, GGT, total cholesterol and bile acids; hepatobiliary ultrasound with Doppler; biopsy only in atypical cases; transthoracic echocardiography; slit-lamp examination; spine radiography; facial assessment; BP, electrolytes, creatinine, urinary protein/creatinine ratio and renal Doppler ultrasound; brain MRI and MRA; and JAG1 and NOTCH2 analysis [39]. No biochemical test is specific to ALGS, and no non-molecular test distinguishes JAG1-related from NOTCH2-related disease [24].

5.5. Genetic testing

Strategy. GeneReviews describes two approaches. When the phenotype is typical, sequential testing is possible: JAG1 sequencing (missense, nonsense, splice and small insertion or deletion variants); if negative, JAG1 deletion and duplication analysis (exons and whole gene); if negative and suspicion is strong, NOTCH2 analysis [1]. Alternatively, a multigene panel including JAG1, NOTCH2 and differential diagnosis genes; if the patient does not meet clinical criteria, exome sequencing is the most commonly used option, with genome sequencing an alternative [1]. The EASL guideline recommends early genetic testing in investigating unexplained cholestasis in infants and children and, in adults, after common causes have been excluded, by next-generation sequencing (NGS) panel or exome; it discourages Sanger sequencing of individual genes, recommends considering genome sequencing if the exome is negative and reanalysing data at least every three years when no diagnosis has been made, with genetic counselling before testing [10]. The two sources differ on single-gene testing: the sequential GeneReviews model remains acceptable when the phenotype is unequivocal, but in as yet undifferentiated neonatal cholestasis a panel or exome offers greater yield.

Dosage analysis and cytogenetics. The ClinGen haploinsufficiency score of 3 makes JAG1 deletion and duplication analysis mandatory, by multiplex ligation-dependent probe amplification (MLPA), copy-number variant calling from NGS or chromosomal microarray (CMA) [24]. The Brazilian textbook also describes FISH with a 20p12 probe, showing absence of the 20p12.2 signal with the 20q control signal present [6]. If there is a whole-gene deletion, a full cytogenetic study should be performed, and if developmental delay, deafness or autism is present, CMA to delineate a contiguous deletion [1].

Test limitations. A negative result does not exclude ALGS. Some variants escape routine sequencing, such as copy-number-neutral inversions and promoter variants; when the clinical diagnosis is firm and no variant is found, genome sequencing focused on JAG1 and NOTCH2 or RNA sequencing may identify it, as shown by Rajagopalan and colleagues in 2021 [1]. NOTCH2 analysis has a specific technical difficulty: the NOTCH2NLR pseudogene, whose first four exons correspond to NOTCH2 exons 1 to 4, may generate false positives or false negatives in this region if the laboratory does not use a specific alignment or amplification strategy [1]. Low-level parental mosaicism may not be detected in peripheral blood DNA, and exclusively germline mosaicism never is [1]. The proportions in Table 2.1 (94.3%, 2.5%) are proportions of probands with each variant type and should not be presented as the clinical sensitivity of a test; approximately 3% of clinically diagnosed patients remain without an identified variant [1]. Consistently, the French biliary atresia PNDS states that JAG1 and NOTCH2 sequencing establishes the molecular diagnosis in more than 95% of cases [16].

Reporting and interpretation. The report should state the gene, versioned transcript (NM_000214.3 or NM_024408.4), coding and protein description according to Human Genome Variation Society (HGVS) nomenclature, genomic coordinate with genome build, zygosity (heterozygous), ACMG/AMP classification with the criteria applied and, for deletions, boundaries and included genes [1, 3]. A VUS in JAG1 or NOTCH2 neither confirms nor excludes diagnosis; family segregation analysis, targeted clinical assessment of carriers, functional data on JAG1 membrane expression and NOTCH2-specific rules are routes to reclassification [1, 26, 30]. Pathogenic and likely pathogenic classes are equivalent for clinical decision-making [1]. When there is no molecular diagnosis, storage of proband DNA is recommended [1].

5.6. Prenatal diagnosis

Prenatal diagnosis and preimplantation genetic testing (PGT) are possible when the familial variant is identified, but do not predict severity [1, 6]. Fetal echocardiography in a fetus at 50% risk may detect significant structural heart disease, but a normal echocardiogram excludes neither ALGS nor heart disease [1]. In eight reports of prenatal diagnosis, ultrasound findings included intrauterine growth restriction (IUGR) in 75%, heart disease (two tetralogies of Fallot and one severe pulmonary stenosis), hemivertebrae and a single umbilical artery [39]. Smith's textbook also reports prenatal ductus arteriosus stenosis [7].

5.7. Practical algorithm

Table 5.3. Diagnostic algorithm by scenario [1, 8, 10, 45]

6. Differential Diagnosis

The differential diagnosis is organised around the finding that prompted suspicion: duct paucity and neonatal cholestasis, posterior embryotoxon, pulmonary circulation abnormalities and butterfly vertebrae. Each finding is non-specific in isolation, and their combination suggests ALGS.

Table 6.1. Differential diagnosis of Alagille syndrome [1, 6, 8, 24, 40]

6.1. Neonatal cholestasis and biliary atresia

The most important distinction is from biliary atresia, because portoenterostomy is beneficial in biliary atresia when performed early and harmful in ALGS [1, 46]. Elevated GGT, absent excretion on scintigraphy and non-visualisation of the bile ducts are common to both diseases; pulmonary heart defects, butterfly vertebrae, embryotoxon or suggestive facies, elevated cholesterol and rapid genetic testing point towards ALGS [8, 45]. The Oxford Desk Reference notes that, once genetic causes of atresia are excluded, the recurrence risk of biliary atresia is low, in contrast with the 50% transmission risk from a parent with ALGS [8]. PFIC 1 and 2 and bile acid synthesis defects are distinguished by low or normal GGT [8]. Heart disease alone does not separate the two diseases: 10 to 20% of biliary atresias are syndromic, mainly with polysplenia syndrome (polysplenia, midline liver, preduodenal portal vein, absent inferior vena cava, situs inversus) or other laterality anomalies, and are associated with venous, digestive, urinary and cardiac malformations; laterality anomalies point towards syndromic biliary atresia, and branch pulmonary stenosis, embryotoxon and butterfly vertebrae towards ALGS [16]. Because expressivity is variable, many newborns with ALGS do not yet show the full set of signs, and the French PNDS recommends that, in such cases, the work-up and therapeutic decision be made in a reference or competence centre [16].

6.2. Posterior embryotoxon and anterior chamber abnormalities

Posterior embryotoxon is frequent in Axenfeld–Rieger syndrome and occurs in 8–15% of the general population, so it has value only when associated with other signs [1]. In Axenfeld–Rieger syndrome, anterior chamber abnormalities are more extensive, with glaucoma risk, and there is no liver disease [1].

6.3. Pulmonary circulation abnormalities

Peripheral pulmonary artery stenoses also occur in Williams and Noonan syndromes, NF1 and Down syndrome, and may be isolated [1, 40]. In 22q11.2 deletion, the coexistence of VSD, tetralogy of Fallot, butterfly vertebrae and short stature mimics ALGS, but there is no liver disease, and genetic testing distinguishes them [1]. In apparently isolated tetralogy of Fallot or peripheral pulmonary stenosis, a JAG1 variant is possible, and hepatic, ocular and vertebral assessment of the patient may reveal mild ALGS [7, 33].

6.4. Hajdu–Cheney syndrome

Hajdu–Cheney syndrome shares the NOTCH2 gene, dominant inheritance, renal cysts and some heart defects, but has neither duct paucity nor cholestasis; the predominant presentation is skeletal and progressive [1, 24]. The molecular distinction depends on variant location and type: exon 34 truncating variants, with gain of function, in Hajdu–Cheney syndrome; predominantly missense variants, with loss of function, in ALGS [1]. A report identifying a NOTCH2 variant should therefore interpret the mechanism in light of the phenotype.

7. Treatments and Risk Management Strategies

7.1. General principles and organisation of care

There is no curative treatment for ALGS [1]. Management is multidisciplinary, coordinated by paediatric hepatology and involving genetics, nutrition, cardiology, ophthalmology, nephrology, transplant hepatology and child development [1]. Until 2024 there were no ALGS-specific clinical practice guidelines [1]; the 2024 EASL guideline on genetic cholestatic diseases became the only formal guideline consulted with graded ALGS recommendations, and the expert review by Ayoub and colleagues (2023) is the reference for adult management, without formal evidence grading [10, 39]. China's National Health Commission published an ALGS diagnosis and treatment guideline in 2025, whose Chinese text was not consulted [48]. Objectives are to control pruritus and cholestasis consequences, ensure growth, prevent and treat fat-soluble vitamin deficiencies, detect and treat cardiac, vascular and renal complications, and refer in time for liver transplantation.

7.2. Nutrition and fat-soluble vitamins

Cholestasis reduces the absorption of fat and of vitamins A, D, E and K; treatment of growth impairment rests on nutritional optimisation and fat-soluble vitamin replacement, and may require nasogastric tube (NGT) or gastrostomy feeding [1]. MCT formulas have been used since Alagille's series and remain in use in reported practice [19, 27]. The sources consulted do not provide ALGS-specific fat-soluble vitamin doses; supplementation is adjusted to serum levels and the international normalised ratio (INR), and IBAT inhibitors may worsen deficiencies (section 7.4) [11, 12].

In the absence of an ALGS-specific regimen, the French biliary atresia protocol, drafted by the reference centre that also follows genetic cholestasis, provides a regimen for chronic cholestasis in infants and children that can be extrapolated to ALGS, with the caveat that it was defined for biliary atresia [16]. As soon as cholestasis is recognised, it recommends a 10 mg injection of vitamin K, intravenously or intramuscularly, whatever the weight, to prevent intracranial haemorrhage; correction of the prothrombin time after the injection, with a normal factor V, also distinguishes cholestasis from hepatocellular failure. Subsequent supplementation depends on total bilirubin (Table 7.0) [16]. In any regimen, vitamin A, D and E concentrations and the prothrombin time are monitored regularly, and vitamin E is adjusted to the ratio of serum vitamin E to the sum of cholesterol and triglycerides [16]; in ALGS, the ratio is particularly necessary, because hyperlipidaemia raises total serum vitamin E and may mask a deficiency. The same protocol proposes an MCT-enriched diet and, in France, mentions oral multivitamin supplements classified as foods for special medical purposes (DEKAs Essential®, DEKAs AquaE®), whose use in this indication is under evaluation [16].

Table 7.0. French fat-soluble vitamin supplementation regimen in chronic cholestasis (defined for biliary atresia; extrapolation to ALGS) [16]

ALGS-specific growth charts help interpret growth and inform decisions on transplant listing [44].

7.3. Symptomatic treatment of pruritus and xanthomas

Cholestatic pruritus in ALGS responds poorly to conventional drugs, and combinations are frequent [1]. Table 7.1 summarises the drugs described for pruritus, all used off-label in ALGS except IBAT inhibitors, with efficacy as assessed by EASL. Doses are given only when present in the sources consulted; those of the French biliary atresia protocol are extrapolated from another paediatric cholestasis treated with the same drugs.

Table 7.1. Drugs for cholestatic pruritus in ALGS [1, 10, 50]

When pruritus is severe and refractory, GeneReviews allows surgical biliary diversion (section 7.8) and, in refractory cholestasis, liver transplantation [1].

Hypercholesterolaemia and xanthomas in ALGS mainly reflect lipoprotein X, which is not considered atherogenic, and EASL recommends against statins for hyperlipidaemia or xanthomas (level of evidence 4, strong recommendation, strong consensus), based on the absence of evidence beyond one case report [10]. A phase 4 trial of atorvastatin for xanthomas in ALGS, with 15 patients, was completed in China in August 2024, without published registry results at the access date [52]. Xanthomas decrease with reduced cholestasis on IBAT inhibitors (section 7.6) [53].

7.4. Ileal bile acid transporter inhibitors

Mechanism. Maralixibat and odevixibat are selective, reversible, minimally absorbed IBAT inhibitors acting locally in the distal ileum, reducing bile acid reabsorption, increasing faecal elimination and decreasing serum bile acid concentration, a “medical interruption of enterohepatic circulation” [1, 12]. Bile acid reduction does not correlate with systemic pharmacokinetics [14]. Efficacy requires preserved enterohepatic circulation and canalicular bile salt transport; conditions, drugs or surgery compromising intestinal motility or enterohepatic circulation, such as surgical biliary diversion, may reduce efficacy [12, 14]. EASL recommends offering an IBAT inhibitor, when available, to patients with ALGS and cholestatic pruritus (level of evidence 2, strong recommendation, strong consensus) [10].

Maralixibat (Livmarli®, Mirum Pharmaceuticals). In the USA, the 9.5 mg/mL oral solution is indicated for cholestatic pruritus in ALGS from 3 months; the 19 mg/mL solution is intended for PFIC, and 10, 15, 20 and 30 mg tablets may be used in ALGS and PFIC in patients weighing 25 kg or more who can swallow tablets [11]. The recommended ALGS dose is 380 mcg/kg once daily, 30 minutes before a morning meal, starting with 190 mcg/kg once daily and increasing to 380 mcg/kg after one week, as tolerated; the maximum daily dose is 28.5 mg (3 mL) with oral solution and 30 mg with tablets [11]. In the EU, the 9.5 mg/mL solution is indicated for cholestatic pruritus in ALGS from 2 months, and tablets only in adults and adolescents aged 12 years or more; the target dose and titration are the same, administration is before (up to 30 minutes) or with a morning meal, and if poorly tolerated the dose may be reduced from 380 to 190 mcg/kg/day or interrupted, with a further escalation attempt [12]. The European summary of product characteristics (SmPC) adds that alternative treatment should be considered if there is no benefit after 3 months of continuous daily treatment [12].

Table 7.2. Maralixibat 9.5 mg/mL oral solution in ALGS: volume per dose by weight (identical in USA and EU labels) [11, 12]

Tablet labels differ. In the USA, oral solution is used below 25 kg; from day 8, the dose is 10 mg at 25 to 32 kg, 15 mg at 33 to 43 kg, 20 mg at 44 to 65 kg (10 mg on days 1 to 7) and 30 mg from 66 kg (15 mg on days 1 to 7); for 25 to 43 kg the label specifies no tablet for the first week [11]. In the EU, titration uses oral solution, and from day 8 the dose is 15 mg at 33 to 43 kg, 20 mg at 44 to 55 kg, 25 mg at 56 to 65 kg and 30 mg from 66 kg, the maximum dose [12].

Odevixibat (Bylvay® in the USA; Kayfanda® in the EU; Ipsen). In the USA, Bylvay® is indicated for cholestatic pruritus in ALGS from 12 months of age, at 120 mcg/kg once daily, in the morning with a meal; oral pellets (200 and 600 mcg) are used below 19.5 kg and capsules (400 and 1,200 mcg) from 19.5 kg; the US label does not define a maximum dose for ALGS, and the weight-band table ends at 7,200 mcg/day from 55.5 kg [13]. If tolerability problems occur without another cause, the dose may be reduced to 40 mcg/kg/day and returned to 120 mcg/kg/day once they stabilise [13]. Pellets may be mixed with up to two tablespoons (30 mL) of soft food at or below room temperature, and capsules should not be crushed or chewed [13]. In the EU, the ALGS indication was authorised under the brand Kayfanda®, from 6 months of age, with the same dose of 120 mcg/kg once daily, in the morning, with or without food, and a maximum daily dose of 7,200 mcg; reduction to 40 mcg/kg/day is provided for diarrhoea lasting 3 or more days, severe diarrhoea or diarrhoea requiring intravenous hydration, and alternative treatment should be considered if there is no benefit after 6 months [14]. The 200 and 600 mcg capsules are intended to be opened and sprinkled, and the 400 and 1,200 mcg capsules to be swallowed, although they may also be opened; the contents may be administered in breast milk, formula or 5 mL of water, by syringe [14].

Table 7.3. Odevixibat in ALGS: total daily dose by weight (120 mcg/kg/day) [13, 14]

Contraindications and hepatotoxicity. Since 2025, the American label for both drugs contraindicates IBAT inhibitors, as a class effect, in patients with previous or current hepatic decompensation events (variceal bleeding, ascites, hepatic encephalopathy) [11, 13]. European SmPCs do not have this contraindication: Kayfanda® contraindicates only hypersensitivity, and Livmarli® hypersensitivity and, in PFIC, severe hepatic or renal impairment with the 9.5 mg/mL solution; both recommend close monitoring in end-stage or decompensated liver disease [12, 14]. The American label recommends baseline liver tests and frequent monitoring during the first 6 to 8 months and thereafter as clinically indicated; reduce the dose or interrupt for new abnormalities or clinical hepatitis without another cause, and permanently discontinue if these persist, recur on reintroduction or decompensation occurs [11, 13]. With odevixibat, establishing the baseline pattern of liver test variability, monitoring ALT, AST, total and direct bilirubin and INR and restarting when values return to baseline are recommended [13]. In ALGS, aminotransferase fluctuations are intrinsic to the disease, complicating causal attribution; in the odevixibat programme (PFIC and ALGS), two of the six patients with drug-induced liver injury underwent transplantation, and in the maralixibat PFIC programme, of four attributed cases, one led to transplantation and another to death [11, 13].

Adverse effects and management. In the maralixibat ALGS programme (5 studies, 86 patients, doses up to 760 mcg/kg/day, median exposure 32.3 months, mostly uncontrolled), adverse effects were diarrhoea (55.8%), abdominal pain (53.5%), vomiting (40.7%), nausea (8.1%), fat-soluble vitamin deficiency (25.6%), increased aminotransferases (18.6%) and fractures (9.3%); ALT rose to more than 3 times baseline in 26% and more than 5 times in 3%, 7 patients (8.1%) discontinued because of increased ALT, and reductions or interruptions for diarrhoea, abdominal pain or vomiting occurred in 5 (6%) [11]. In the European SmPC, diarrhoea was always mild to moderate, began in the first month and lasted a median of less than one week [12]. In ASSERT, with odevixibat, diarrhoea occurred in 29% versus 6% with placebo, abdominal pain in 14% versus 6%, haematoma in 9% versus 0 and weight loss in 6% versus 0, without discontinuations due to adverse effects; in the 72-week open-label extension (50 patients), vitamin D deficiency occurred in 14%, vitamin E in 10% and vitamin K in 6%, fractures in 4%, and the most frequent cause of discontinuation was increased bilirubin [13]. In two open-label post-marketing studies, fracture frequency was 4% in ALGS and 5% in PFIC [13]. Diarrhoea is managed with dose reduction or temporary interruption and reintroduction at the last tolerated dose; treatment is interrupted for bloody diarrhoea, vomiting, dehydration requiring treatment or fever [11]. Before initiation, vitamins A, D and E and INR should be measured, and deficiencies supplemented during treatment; if deficiency persists despite supplementation, or fracture or haemorrhage occurs, interruption is considered [11, 12].

Excipients, interactions and overdose. Maralixibat solution contains propylene glycol (364.5 mg/mL); at 380 mcg/kg/day, exposure is 14.6 mg/kg/day according to the American label and up to 17 mg/kg/day according to the European SmPC, with greater risk below 5 years [11, 12]. Bile acid sequestrants (colestyramine, colesevelam, colestipol) should be taken at least 4 hours before or after the IBAT inhibitor [11, 13]. Maralixibat inhibits OATP2B1 in vitro, potentially reducing absorption of substrates such as statins (fluvastatin, rosuvastatin) [11, 12]. Odevixibat is a P-glycoprotein substrate (itraconazole increases exposure by 50–60%, without clinical relevance) and reduces ethinylestradiol area under the curve by 17% [14]. Interaction with UDCA and rifampicin has not been studied [12, 14]. In maralixibat overdose, single doses up to 500 mg were tolerated in healthy adults, and the relevant toxicity is that of propylene glycol, removable by dialysis [11, 12]; with odevixibat, single doses of 10,000 mcg had no consequences [14].

7.5. Comparative regulatory status

Maralixibat was the first drug approved for ALGS, by the USA Food and Drug Administration (FDA), in September 2021. In the EU, both drugs were authorised by the European Commission, following a European Medicines Agency (EMA) opinion, under “exceptional circumstances”, with annual reassessment of new information and additional monitoring [12, 14].

Table 7.4. Regulatory status of IBAT inhibitors in ALGS (accessed on 10 October 2026) [11–14, 54–57]

Labels also differ in emphasis regarding pregnancy: the American maralixibat label does not expect measurable fetal exposure, whereas the European SmPC prefers avoiding the drug in pregnancy and breastfeeding; odevixibat caused cardiac and vascular malformations in rabbits at all tested doses, and the European SmPC requires effective contraception (section 7.11) [11–14].

Access depends on each country's health technology assessment. In Ireland, the National Centre for Pharmacoeconomics (NCPE) did not recommend reimbursement of maralixibat for cholestatic pruritus in ALGS from 2 months unless cost-effectiveness improved (30 March 2026) [58]. In England, National Institute for Health and Care Excellence (NICE) assessment was suspended in March 2024 because of an incremental cost-effectiveness ratio above the acceptable level, resumed following commercial negotiation, and the final draft document was published on 9 October 2026, without its content having been consulted [59]. A Takeda database study of liver injury with maralixibat in Japanese patients, initiated in January 2026, indicates commercial use in Japan [60], and odevixibat was the subject of a therapeutic note in Australia in 2026 [61]. Sources consulted do not include authorisation or reimbursement data for Portugal (INFARMED) or Brazil (ANVISA, CONITEC); centralised EU authorisation applies to Portugal, but effective access depends on national assessment.

7.6. Clinical trials

Table 7.5. IBAT inhibitor trials relevant to ALGS

The pruritus scales used (ItchRO(Obs), observed scratching scale, from 0 to 4) are caregiver-reported; a 1-point change was considered clinically significant [12]. The phase 2 odevixibat study by Baumann and colleagues (2021), cited in GeneReviews, preceded ASSERT [1].

Evidence limitations. Trials are small, with short controlled duration, and manufacturer-funded (Mirum in ICONIC, Albireo/Ipsen in ASSERT) [62, 64]. ICONIC used a 4-week randomised withdrawal in patients who had already responded, a design enriching the population for responders and measuring loss of effect rather than gain; ASSERT is the only parallel placebo-controlled phase 3 trial in ALGS. ITCH did not meet its primary outcome in the full cohort [10], and the Japanese trial, with seven patients, did not demonstrate bile acid reduction [68]; these negative or null results should be read alongside the positive ones. Primary outcomes are caregiver-reported pruritus and serum bile acids, markers of symptoms and cholestasis; no trial measured native liver survival as a primary outcome. There is no direct comparison between maralixibat and odevixibat, and numerical differences between trials should not be interpreted as efficacy differences, because design, relative dose, duration and population differ. The unanchored indirect comparison published in 2026, with ASSERT individual data and ICONIC aggregate data, found numerically greater improvements with odevixibat in pruritus and bile acids, without significance, a total PedsQL difference of 14.64 points (95% CI 5.70–23.58) and fewer gastrointestinal adverse effects (odds ratio, OR, 0.22; 0.06–0.87); this is low-certainty evidence, because of differing designs and authors' links to the odevixibat manufacturer [69]. Efficacy in infants under 12 months rests on small open-label series with heterogeneous results [12, 67], and adult data are minimal: 11 patients who started maralixibat before 16 years and 3 who started afterwards [70].

7.7. Long-term outcomes and experience outside trials

Long-term data suggest that treatment response is associated with a better course, without demonstrating causality. In an analysis of 76 patients treated with maralixibat for at least 48 weeks in three trials, with a median duration of 4.7 years, 16 events occurred (10 transplants, 3 decompensations, 2 deaths and 1 surgical biliary diversion); event-free survival at 6 years was 88% in patients with an ItchRO(Obs) reduction greater than 1 point versus 57% in the remainder (p=0.005), 90% versus 43% with bilirubin below 6.5 mg/dL at week 48 (p<0.0001) and 85% versus 49% with bile acids below 200 µmol/L (p=0.001) [71]. In the external comparison of 84 maralixibat-treated patients and 469 GALA patients meeting trial eligibility criteria, aligned by an estimated index time and balanced for age, bilirubin, GGT and ALT, event-free survival (manifest portal hypertension, biliary diversion, transplantation or death) was better with maralixibat (HR 0.305; 95% CI 0.189–0.491; p<0.0001) [72]. This is a non-randomised comparison with a historical control, in which bile acids were not used for matching and which was partly manufacturer-sponsored [72]. EASL summarises that initial data link bile acid reduction to better transplant-free survival and that, in responders, IBAT inhibitors appear to delay transplantation for pruritus, but whether they alter progression to end-stage liver disease or HCC risk is unestablished [10].

Response is also associated with better quality of life: in ICONIC, 20 of 27 patients (74%) responded on ItchRO(Obs) at week 48, and responders had improved fatigue (+25.8 versus −3.1 points; p=0.03) and family impact, without a difference in generic PedsQL [36]. In an integrated post hoc analysis of ICONIC and ITCH (63 children, 43% with baseline xanthomas), the clinical xanthoma scale improved at 96 weeks (p=0.004), the proportion without xanthomas rose from 60% to 86% and cholesterol fell by 55 mg/dL [53]. With odevixibat, ASSERT described improvements in sleep, cholesterol and xanthomas, maintained in the extension [14].

Experience outside trial criteria is descriptive. Eight patients who would have been excluded (previous biliary diversion, administration by gastrostomy or NGT, transplant candidates) had reduced pruritus and good tolerability [73]; two patients improved with maralixibat before and after reversal of surgical biliary diversion [74]; and a 16-year-old boy with severe pruritus 13 years after liver transplantation, due to early chronic graft dysfunction, had resolution of pruritus in two months, an off-label use described in a single report [75]. Two Spanish cases described fibrosis regression and improved portal hypertension on maralixibat (section 10); these uncontrolled observations generate hypotheses and do not demonstrate disease modification [27].

7.8. Biliary surgery: biliary diversion and Kasai portoenterostomy

Partial biliary diversion, external or internal, is surgical interruption of enterohepatic circulation used for severe refractory pruritus [1, 7]. With the arrival of IBAT inhibitors, its position changed: previous diversion may reduce their efficacy, was an exclusion criterion in ICONIC, and reports describe patients whose diversion was reversed and pruritus controlled with maralixibat [12, 73, 74].

Kasai portoenterostomy is a different case. Smith's textbook states that biliary diversion procedures, “such as Kasai operations”, may improve cholestasis and prevent cirrhosis and liver failure [7]. Available evidence does not support this statement regarding Kasai. At the Children's Hospital of Philadelphia, among 430 patients with ALGS, the 19 who underwent portoenterostomy were compared with 36 ALGS patients matched for age, heart disease and neonatal jaundice, with equivalent bilirubin at 6–10 weeks (9.6 and 8.7 mg/dL): transplantation occurred in 47.3% versus 13.9% (p=0.01) and mortality in 31.6% versus 2.8% (p=0.005) [46]. The authors concluded that Kasai offers no benefit and appears to worsen outcomes, and did not appear merely to mark more severe disease, given similar initial bilirubin [46]. A meta-analysis of six retrospective studies, with 394 patients, associated Kasai with more transplantation (OR 6.46; 95% CI 3.23–12.89), lower native liver survival (OR 25.88; 2.83–236.84) and higher mortality (OR 15.05; 2.70–83.93) [76]. The authors themselves caution that whether surgery causes deterioration is unclear, because operated patients may have had more severe disease initially; all studies are retrospective, and confounding by severity cannot be excluded [76]. In an Asian and Turkish living donor transplant series, previous Kasai had no significant impact on transplant outcomes [43]. The practical conclusion is that Kasai portoenterostomy is not ALGS treatment and that ALGS should be excluded clinically and genetically before proposing Kasai in an infant with high-GGT cholestasis [45, 46].

7.9. Liver transplantation

Transplant indications in ALGS are refractory cholestasis, with disabling pruritus, disfiguring xanthomas, severe growth impairment or fractures, and end-stage liver disease or decompensated portal hypertension; transplantation remains standard treatment for these situations, and EASL strongly recommends referral to a transplant centre for patients with substantial hepatic involvement [1, 10]. In GALA, post-transplant survival was 88% at 20 years [1].

ALGS carries specific transplant risks. In the North American and Canadian Studies of Pediatric Liver Transplantation (SPLIT) registry, 91 children with ALGS, compared with 236 age-matched children with biliary atresia, more frequently had GFR below 90 mL/min/1.73 m² before transplantation (18% versus 5%) and height deficit (66% versus 22%), and had lower one-year survival (87% versus 96%; p=0.002), with deaths concentrated in the first 30 days and associated with biliary, vascular, central nervous system and renal complications [42]. In the Vanguard living donor transplant study in 49 children with ALGS (2001–2020; Japan, Korea and Turkey), median transplant age was 1.1 years, main indications were severe growth retardation (91.8%) and advanced liver disease, patient and graft survival at 5 years were 93.7% and 91.5%, transplantation before 2 years was associated with greater height and weight recovery, and renal function deteriorated in the long term, particularly more than 15 years after transplantation and when performed at an older age [43].

Pre-transplant assessment, according to Ayoub and colleagues, includes stress echocardiography in all patients, with catheterisation and pharmacological testing in selected cases (reserve considered sufficient when cardiac output increases by 40% and the right ventricular-to-aortic pressure ratio is 0.5 or less, a criterion not predictive in another study), brain MRI and MRA and triphasic abdominal computed tomography (CT) for visceral vasculopathy; they recommend renal-sparing immunosuppression [39]. Related living donors should be assessed for ALGS: testing for the familial variant when known and, if unknown, extensive clinical assessment with abdominal vascular imaging and liver biopsy [1, 39]. In Vanguard, 11% of donor candidates were excluded for genetic or anatomical reasons [43].

7.10. Heart, vessels and kidney

Heart defects should be treated by cardiologists and cardiothoracic surgeons experienced in ALGS. Repeated oxygenator failure during cardiopulmonary bypass (CPB) has been described in patients with ALGS (Moore et al. 2022), and surgical repair of peripheral pulmonary artery stenoses was reported in 145 patients over two decades at a single centre (Felmly et al. 2023), according to GeneReviews [1]. The efficacy of presymptomatic screening for vascular abnormalities has not been formally assessed; any patient with neurological symptoms should be investigated aggressively with MRI, MRA and/or angiography [1]. Moyamoya syndrome has been treated with surgical revascularisation (Baird et al. 2015, according to the Brazilian textbook) [6]. Renal abnormalities are followed by nephrology, with BP control [1].

7.11. Agents and circumstances to avoid, pregnancy, perioperative care and transition

To avoid. Contact sports should be avoided, particularly with chronic liver disease, splenomegaly or vascular involvement, and alcohol should be avoided with liver disease [1]. IBAT inhibitors are contraindicated in the USA after hepatic decompensation [11, 13], and statins are not recommended for xanthomas [10].

Pregnancy. Pregnancy requires hepatic and cardiac surveillance, because portal hypertension and cardiac dysfunction may worsen; successful pregnancies have been described [1]. Ayoub and colleagues describe risks of variceal bleeding, prematurity and miscarriage due to worsening portal hypertension, cardiac deterioration particularly with pulmonary hypertension due to peripheral pulmonary stenoses, and aneurysm rupture with the Valsalva manoeuvre; they suggest caesarean section or assisted vaginal delivery with intracranial vasculopathy, clinically significant portal hypertension or severe heart disease, and unassisted vaginal delivery otherwise [39]. They recommend preconception counselling, offering prenatal testing, referral to high-risk obstetrics and obstetric anaesthesia, endoscopic variceal screening before conception (or in the second trimester if not performed in the previous year) in pregnant women with cirrhosis, strict adherence to fat-soluble vitamins if cholestasis is present and detailed fetal ultrasound when indicated [39]. Maternal cholestasis is associated with fetal distress, meconium-stained fluid, preterm delivery and fetal death near term; maternal fat-soluble vitamin deficiency may cause neonatal vitamin K deficiency bleeding, maxillary hypoplasia, fetal hypocalcaemia and IUGR, and coagulopathy may contraindicate neuraxial anaesthesia [39].

Drugs during pregnancy and lactation. The American maralixibat label does not expect measurable fetal exposure, because of low systemic absorption and absent effects in animal studies, and notes that vitamin supplementation may need to increase during pregnancy [11]; the European SmPC prefers avoiding it during pregnancy, as a precaution, and breastfeeding, because of propylene glycol [12]. Odevixibat caused cardiac, great vessel and other vascular malformations in rabbits at all doses, with maternal exposure 2.1 times the maximum human dose at the lowest dose, and may cause cardiac malformations if the fetus is exposed [13]; in the EU it is not recommended in pregnancy, women of childbearing potential should use effective contraception and, during breastfeeding, risk cannot be excluded [14]. The 17% reduction in ethinylestradiol exposure should be considered in contraceptive choice [14].

Perioperative care. Before major surgery, Ayoub and colleagues recommend brain MRI and MRA if not already performed [39]; in cardiac surgery the possibility of oxygenator failure should be anticipated [1], and vitamin K deficiency coagulopathy should be corrected.

Transition to adult care. At transition, brain MRI and MRA if not previously performed, continued HCC, renal and cardiac surveillance and reproductive counselling for young adults are recommended [1, 39].

7.12. Developments in the last 12 months

Developments between October 2025 and October 2026 are mainly regulatory and observational evidence, without newly published randomised trials in ALGS.

Regarding authorised label changes, on 19 December 2025 the FDA approved inclusion in the Bylvay® label of long-term open-label extension data and revision of the fracture section (4% in ALGS in open-label post-marketing studies) [13, 55]. In the EU, there were new European Commission decisions on Livmarli® (18 February 2026, revision 10, with tablets for adults and adolescents) and Bylvay® (26 March 2026), without a new ALGS indication [12, 56].

Regarding technology assessment opinions, there was a negative reimbursement opinion from the Irish NCPE for maralixibat (30 March 2026) and a NICE final draft document (9 October 2026), whose content was not consulted [58, 59].

Published studies include a low-certainty odevixibat–maralixibat indirect comparison [69], an integrated analysis of xanthomas with maralixibat [53], ALGS-specific growth charts [44], a prognostic bile acid threshold of 102 µmol/L in GALA [37], national Brazilian and Korean cohorts [29, 35], long-term living donor transplant outcomes [43] and cases of fibrosis regression on maralixibat [27].

As for registered trials and post-authorisation studies, RISE infant trial results were published in the registry (June 2026) [67]; the phase 4, low-intervention LEAP-EU study, with approximately 230 Livmarli®-treated patients in the EU (September 2025; primary completion scheduled for 2030) [77], an observational odevixibat safety study in France and Italy (July 2026) [78] and a Japanese study of liver injury with maralixibat (January 2026) [60] began. The LEAP-US observational registry [79], the American observational odevixibat study on biliary diversion, transplantation and survival [80] and ASSERT-EXT [65] are ongoing.

No Notch pathway-targeted therapy is in clinical trials in ALGS in the sources consulted; a negative search does not demonstrate non-existence.

7.13. Guideline recommendations and divergences

Table 7.6. Recommendations of guidance sources and their divergences

The 2024 French biliary atresia PNDS (French National Diagnosis and Care Protocol) adds four divergences in the neonatal phase [16]. The first is epidemiological: the PNDS attributes 10–15% of neonatal cholestasis to ALGS, whereas the systematic review of consecutive series found approximately 1% (with an additional 2.5% of duct paucity without a syndromic diagnosis); the systematic review serves as the reference [47]. The second concerns hepatobiliary scintigraphy: Ayoub and Kamath use excretion of the radiotracer to exclude biliary atresia [45], whereas the PNDS considers it useless and a cause of delay to portoenterostomy, as is magnetic resonance cholangiopancreatography, and reserves the decision for cholangiography; the PNDS also does not mention MMP-7. The third concerns the clinical criteria: the PNDS accepts cholestasis with two other major criteria, without requiring biopsy, whereas the classical Alagille criteria require duct paucity and three criteria, and GeneReviews accepts three of five with duct paucity [1, 19]. The fourth is internal to the protocol, which gives two conjugated fraction thresholds (10% and 20%) while keeping the absolute threshold of 17 µmol/L; the 2017 NASPGHAN/ESPGHAN guideline uses only the absolute threshold of 1.0 mg/dL (17 µmol/L), whatever the total bilirubin [81]. On portoenterostomy in ALGS, the PNDS takes no explicit position and refers incomplete cases for decision to a reference centre. As for doses, the PNDS calculates UDCA by body surface area (600 mg/m²/day), which, using the Mosteller formula, corresponds to approximately 34 mg/kg/day in a 5 kg, 57 cm infant and approximately 23 mg/kg/day in a 20 kg, 110 cm child, above the 10–20 mg/kg/day reported by Cheng and colleagues [50]; as neither source is an ALGS-specific guideline and EASL sets no dose, the two dosing regimens are presented side by side.

8. Reproductive Options and Family Counselling

8.1. Inheritance and recurrence risks

ALGS is inherited in an autosomal dominant manner. Approximately 40% of probands have an affected parent and approximately 60% have a de novo variant (unpublished GALA data cited in GeneReviews) [1]; the Oxford Desk Reference reports a new variant rate of approximately 60% and the Brazilian textbook 50–70% [6, 8]. These proportions depend on parental assessment: many carrier parents are recognised only after targeted examination.

Table 8.1. Recurrence risks by scenario [1, 4, 32]

8.2. Assessment of parents and family

In an apparently sporadic case, parents should be assessed carefully. If the proband's variant is known, parents are tested for it; if unknown, clinical assessment includes liver tests and cholesterol, echocardiography, spine radiography, slit-lamp examination and facial assessment by a geneticist [1, 8]. Great intrafamilial variability warrants this assessment: in Kamath's series, only 21% of carrier relatives would have been diagnosed clinically, and 47% did not meet criteria [32]. At-risk relatives should have their status clarified by testing for the familial variant or, in its absence, the same clinical examinations [1]. Identifying mildly affected carriers has specific consequences: a 50% reproductive risk, HCC surveillance even without apparent liver disease and vascular assessment [1, 39]. Relatives being considered as living liver or kidney donors should be tested (section 7.9) [1, 39].

8.3. Mosaicism

Parental mosaicism was found in more than 8.2% of families in a screen of 61 JAG1 variants, and mosaic parents may be mildly or minimally affected [1, 4]. Leucocyte DNA may not detect low-level somatic mosaicism and never detects exclusively germline mosaicism; a parent with a negative blood result therefore does not have a zero recurrence risk [1]. A small number of parents with mild phenotypes may have gonosomal mosaicism [8].

8.4. Prenatal diagnosis and preimplantation genetic testing

When the familial variant is identified, prenatal diagnosis by chorionic villus sampling or amniocentesis and PGT are possible; neither predicts severity, making the decision particularly difficult in a disease whose expression varies from subclinical to fatal within the same family [1, 6]. Fetal echocardiography may detect structural heart disease, but a normal result does not exclude ALGS [1]. Perspectives on prenatal diagnosis differ between families and professionals, and the appropriate time to discuss it is before pregnancy [1]. In pregnant women with ALGS, Ayoub and colleagues recommend offering prenatal genetic testing as part of preconception counselling [39].

8.5. Counselling minors and young adults

Genetic counselling should be offered to affected or at-risk young adults, ideally before reproductive age [1]. In at-risk minors, testing has direct clinical utility, because a carrier needs cardiac, hepatic, renal and vascular assessment and HCC surveillance; testing an asymptomatic child is therefore justified by medical benefit, beyond reproductive planning alone [1, 39]. Counselling should explain variable expressivity, unpredictable severity in a child, mosaicism risk and the meaning of a VUS, and should be accompanied by DNA banking when there is no molecular diagnosis [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each figure measures

Table 9.1. ALGS frequency estimates

All these estimates measure recognised disease frequency, mainly the paediatric hepatic form; mildly affected adults, often identified only after a child's diagnosis, are uncounted, and GeneReviews itself considers the disease underdiagnosed [1, 24]. Orphanet has no validated worldwide prevalence [2].

9.2. Regional cohorts and differences between populations

Most natural history data come from North America and Europe: the Philadelphia series, the USA and Canadian SPLIT cohort, French and British series and the international GALA cohort, with 1,433 children from 29 countries [9, 19, 20, 42]. Recent years have brought cohorts from other regions, whose numbers reflect reference centres rather than the population:

Table 9.2. Recent regional cohorts [9, 29, 35, 84]

Native liver survival in the Korean cohort was higher than in GALA, which may reflect differences in severity, referral criteria or transplant practice, and not necessarily biological differences [29]. The Brazilian cohort is the main source on Brazil to date; the Brazilian textbook of genetic syndromes also mentions a national case report (Torres et al. 2006) [6, 35]. Sources consulted do not include Portuguese data. No JAG1 or NOTCH2 founder variants are described in the sources consulted, consistent with the high proportion of de novo variants.

Facial recognition depends on population: features considered typical were described mainly in patients of European ancestry, and North American dysmorphologists had difficulty recognising them in Vietnamese children [1]. This reinforces genetic testing's role outside the populations in which criteria were developed.

9.3. Sex

Paediatric cohorts have a slight male predominance (57% in GALA, 52.5% in Brazil, 66% of those randomised in ICONIC) [9, 35, 62]. The Brazilian textbook states that disease is more frequent in males [6], but autosomal dominant inheritance and other sources do not support a true frequency difference; cohort predominance may reflect referral patterns. A systematic review found more girls than boys with spontaneous intracranial haemorrhage, suggesting sex differences in bleeding risk that remain unexplained [25].

9.4. Special populations

Adults, pregnant women, transplanted patients and infants under 12 months are the populations with the least evidence. In adults, HCC, vascular and renal surveillance rests on expert recommendations [39]; in pregnancy, on case series [1, 39]; in transplanted patients, renal function and vascular complications dominate follow-up [42, 43]; in infants, IBAT inhibitor efficacy rests on small open-label studies [12, 67]. Children with a contiguous 20p12 deletion have additional neurodevelopmental and hearing needs [1].

9.5. Reference networks and patient associations

In Europe, the reference network for rare liver diseases, including paediatric diseases, is ERN RARE-LIVER (https://rare-liver.eu/) [15]. The Alagille Syndrome Alliance (ALGSA; https://alagille.org/), based in the USA with international reach, is the main disease-specific patient association [85]. GeneReviews also lists the American Liver Foundation, the Childhood Liver Disease Research Network (ChiLDReN), Liver Canada and Liver UK [1]. In the United Kingdom, the Children's Liver Disease Foundation and Gene People took part as patient organisations in the NICE appraisal [59]. International collaborative research is organised around the Global ALagille Alliance (GALA) [9]. In France, ALGS is followed at the reference centre for biliary atresia and genetic cholestasis (AVB-CG, coordinated within the Paris-Saclay hospital network) and in the network of competence centres of the rare liver diseases in children and adults health network (FILFOIE, https://www.filfoie.com/); the patient association is the Association Maladies du Foie depuis l'Enfance (AMFE, http://www.amfe.fr/), chronic liver disease gives access to the long-term condition scheme ALD 6, and transplantation to ALD 28 [16].

10. Illustrative Clinical Case

Quintero-Bernabeu and colleagues, from Barcelona, published two ALGS cases treated with maralixibat in 2025, with serial liver biopsies [27].

The first case concerns a term neonate weighing 2.6 kg, the first child of non-consanguineous parents, admitted at 10 days of life for jaundice and pale stools, with hepatomegaly without splenomegaly. UDCA (15 mg/kg/day), fat-soluble vitamins and MCT formula were started. A hypoplastic gallbladder and hypocholic stools led to suspected biliary atresia, but percutaneous transhepatic cholangiography at 30 days showed contrast passage into the bowel and excluded it. Systemic assessment revealed mild pulmonary branch stenosis, posterior embryotoxon and renal dysplasia. Exome sequencing at 42 days identified the heterozygous variant NM_000214.3(JAG1):c.820G>T p.(Gly274Cys), classified by the authors as likely pathogenic [27]. In ClinVar, the same variant (VCV001057247) is classified as of uncertain significance by a single submitter, evaluated in July 2021 [31]; diagnosis therefore rested on combining the variant with a phenotype meeting clinical criteria (cholestasis, heart disease, embryotoxon), rather than the variant alone. At 2.5 months he had moderate to severe pruritus (ItchRO 2.5 in the morning and 3.0 at night); he required an NGT at 2 months and gastrostomy at 7 months. Liver biopsy showed intrahepatic cholestasis, aberrant ductular proliferation and cirrhosis (Ishak 4–5); the spleen measured 16 cm, platelets were 73,000/µL, with INR 0.9, albumin 3.8 g/dL and ammonia 18 µmol/L. He was referred for liver transplantation and enrolled in a maralixibat early access programme, at 190 µg/kg/day for one week and then 380 µg/kg/day. Bilirubin, cholesterol, bile acids and pruritus improved, with a transient aminotransferase increase not requiring dose adjustment. A second biopsy, taken 13 months later during cardiac catheterisation, showed F2 fibrosis, with reduced liver stiffness and splenomegaly and increased platelets [27]. Figure 1 of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC12855487/figure/F1/) shows initial haematoxylin-eosin histology (A), cytokeratin 7 immunostaining (B), Sirius Red staining with Ishak 4–5 fibrosis (C) and the same stain with F2 fibrosis after one year of treatment (D).

The second case concerns a 15-year-old girl diagnosed by exome sequencing at 6 months (NM_000214.3(JAG1):c.2122_2125del p.(Gln708ValfsTer34), classified as pathogenic in ClinVar), with typical facies, pulmonary branch stenosis, moderate growth retardation, embryotoxon and bilateral renal dysplasia without dysfunction [27, 28]. She had splenomegaly of 20 cm, mild excoriations (ItchRO 1.5), platelets 105,000/µL, ALT 61 and AST 47 U/L, normal bilirubin, bile acids 76 µmol/L, a nodular liver with a possible splenorenal shunt, elastography 13.8 kPa and grade II oesophagogastric varices, while receiving rifampicin (5 mg/kg every 12 hours) and UDCA (15 mg/kg/day). After 24 months of maralixibat, with the same titration schedule, ItchRO was 0, the spleen measured 15 cm, platelets were 152,000/µL and liver stiffness 9.2 kPa; at 12 months varices were grade I–II [27].

The case illustrates several points of this article: how easily ALGS is confused with biliary atresia and the value of excluding the latter by cholangiography before any surgery; the importance of cardiac, ocular and renal assessment in neonatal cholestasis; the need to normalise the variant and compare the article's classification with databases; and the severity of pruritus and malnutrition in the first year. Several management choices in the report do not correspond to current guideline practice or labels. Liver biopsy for staging runs counter to EASL's recommendation against routine use for prognosis [10]. Maralixibat was started in an infant with cirrhosis and portal hypertension being assessed for transplantation, a population excluded from trials, and the European label recommends close monitoring in end-stage disease or decompensation, whereas the American label contraindicates the drug after decompensation [11, 12]. UDCA and rifampicin doses are those used by the authors and do not constitute a recommendation. Finally, fibrosis regression in two patients without a control group is an observation generating a disease modification hypothesis, rather than evidence that maralixibat modifies the disease [10, 27].

11. Prognosis and Long-Term Follow-Up

11.1. Survival and causes of death

Prognosis depends on severity and duration of early cholestasis, heart disease severity, hepatic course (failure or portal hypertension) and vascular events [7]. Table 11.1 brings together mortality and survival data from different eras and populations that are not directly comparable.

Table 11.1. Prognostic data by cohort

Smith's textbook reports mortality of approximately 10%, mainly vascular, cardiac and hepatic, and transplantation in 15% [7]; the Brazilian textbook states that disease tends to stabilise between 10 and 14 years and that the main causes of death are congenital heart disease, haemorrhagic stroke and liver disease [6]. Life expectancy may be reduced by transplant complications, heart disease, severe liver disease and intracranial haemorrhage [1].

11.2. Predictors of hepatic course

Total bilirubin between 6 and 12 months is the most robust predictor: below 5.0 mg/dL, 79% of children reach adulthood with their native liver, and higher values multiply portal hypertension and transplant risk (section 3.2) [9]. Bile acids below 102 µmol/L are associated with better native liver and event-free survival [37]. In the Brazilian cohort, direct bilirubin above 6 mg/dL, cholesterol above 300 mg/dL, AST above 5 times normal and APRI above 1 were associated with transplantation or death [35]. According to EASL, fibrosis on biopsy before 5 years, xanthomas and bilirubin above 65 µmol/L predicted a severe hepatic outcome in a series of 144 children, and APRI tracks fibrosis stage [10]. In Emerick's series, absent excretion on scintigraphy had no predictive value [20].

11.3. Prognosis with treatment

The prognostic data above concern patients untreated with IBAT inhibitors. In maralixibat-treated patients, pruritus response and reductions in bilirubin and bile acids at week 48 were associated with better event-free survival at 6 years, and the external GALA comparison suggested lower event risk (HR 0.305) [71, 72]. Neither study is randomised, and whether IBAT inhibitors alter progression to end-stage liver disease or HCC risk is unestablished [10].

11.4. Surveillance

There are no formally validated surveillance guidelines for ALGS. Table 11.2 combines GeneReviews' proposal, based on the authors' experience, Ayoub's expert review for adults and IBAT inhibitor label requirements.

Table 11.2. Recommended surveillance

The paediatric renal interval of 6 to 12 months and annual adult interval reflect different sources, without studies comparing them. The 6-month HCC screening interval is expert opinion, not a validated interval, but also applies to apparently non-penetrant carrier relatives [1, 39].

11.5. Transition and adult follow-up

Transition to adult care should ensure continued hepatic, HCC, renal, cardiac and vascular surveillance, with brain MRI and MRA if not already performed, and include reproductive counselling and information about pregnancy risks [1, 39]. In transplanted patients, follow-up centres on renal function, which deteriorates in the long term, and vascular complications, which arise mainly after adolescence [43].

12. Summary and Final Considerations

Alagille syndrome is an autosomal dominant Notch pathway disorder, caused in 94.3% of probands by JAG1 variants acting through haploinsufficiency, in 2.5% by NOTCH2 variants and in approximately 3% by an unidentified cause. Disease combines cholestasis with bile duct paucity, pulmonary territory heart disease, butterfly vertebrae, posterior embryotoxon, characteristic facies, renal involvement and vasculopathy responsible for a substantial share of mortality. Expressivity varies so greatly that relatives with the same variant may range from fatal childhood disease to signs detectable only by targeted examination, and approximately 60% of cases are de novo, with parental mosaicism in more than 8% of families studied.

Diagnosis rests on clinical criteria and genetic testing, which should include JAG1 sequencing and dosage analysis and NOTCH2 analysis, preferably by panel or exome in as yet undifferentiated neonatal cholestasis. A negative test does not exclude disease, missense variant interpretation requires functional and segregation data, and a NOTCH2 variant should be interpreted in light of mechanism, because gain of function in exon 34 causes Hajdu–Cheney syndrome. Distinction from biliary atresia is the diagnostic decision with the greatest immediate consequences: Kasai portoenterostomy does not benefit ALGS patients and is associated in retrospective studies with more transplantation and death, contrary to wording in a reference textbook.

The natural history of untreated liver disease is severe: fewer than half the children in GALA reached 18 years with their native liver, and bilirubin between 6 and 12 months separates trajectories. The IBAT inhibitors maralixibat and odevixibat are the first drugs authorised for cholestatic pruritus in ALGS, with different minimum ages and brands in the USA and EU; they reduce pruritus and bile acids in small, short trials, and response is associated with better event-free survival in non-randomised analyses, without demonstration that they alter disease progression or hepatocellular carcinoma risk. Liver transplantation remains treatment for refractory cholestasis and end-stage disease, with specific renal, cardiac and vascular risks.

Follow-up is multidisciplinary and extends into adulthood, with hepatocellular carcinoma surveillance even in carriers without liver disease, attention to neurological symptoms, renal monitoring and reproductive counselling. The most relevant gaps are absent prevalence data outside Europe, North America and Australia, scarce adult and pregnancy data, lack of validated doses for classical antipruritic agents and absence of trials measuring hard hepatic outcomes with IBAT inhibitors.

13. References

1. Spinner NB, Loomes KM, Krantz ID, Gilbert MA. Alagille Syndrome. 2000 May 19 [Updated 2024 Jan 4]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1273/. PMID: 20301450.
2. Orphanet. Alagille syndrome (ORPHA:52) [Internet]. Paris: INSERM; [updated 2026; accessed 2026-10-10]. Available from: https://www.orpha.net/en/disease/detail/52
3. VariantValidator (v4.0.0). gene2transcripts_v2 (JAG1, NOTCH2, MANE Select) and validation of NM_000214.3:c.820G>T and c.2122_2125del. Available from: https://rest.variantvalidator.org/
4. Giannakudis J, Röpke A, Kujat A, Krajewska-Walasek M, Hughes H, Fryns JP, et al. Parental mosaicism of JAG1 mutations in families with Alagille syndrome. Eur J Hum Genet. 2001;9(3):209-16. doi:10.1038/sj.ejhg.5200613. PMID: 11313761.
5. Orphanet. Hajdu-Cheney syndrome (ORPHA:955) [Internet]. Paris: INSERM; [accessed 2026-10-10]. Available from: https://www.orpha.net/en/disease/detail/955
6. Paula ECO, Almeida LT, Ribeiro EM. Síndrome de Alagille. In: Ribeiro EM, editor. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 59–61.
7. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 822–825 (Alagille syndrome).
8. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 288–290 (Prolonged neonatal jaundice and jaundice in infants below 6 months).
9. Vandriel SM, Li LT, She H, Wang JS, Gilbert MA, Jankowska I, et al. Natural history of liver disease in a large international cohort of children with Alagille syndrome: Results from the GALA study. Hepatology. 2023;77(2):512-529. doi:10.1002/hep.32761. PMID: 36036223; PMC9869940.
10. European Association for the Study of the Liver. EASL Clinical Practice Guidelines on genetic cholestatic liver diseases. J Hepatol. 2024;81(2):303-325. doi:10.1016/j.jhep.2024.04.006. PMID: 38851996.
11. Mirum Pharmaceuticals. LIVMARLI (maralixibat) oral solution; tablets. Prescribing information. NDA 214662 / NDA 219485. Revised 4/2025. DailyMed setid 64000394-1ef6-4e76-8ba8-11f25ba1b167 (version 11). Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=64000394-1ef6-4e76-8ba8-11f25ba1b167
12. European Medicines Agency. Livmarli (maralixibat): EPAR — Product information (SmPC). EMEA/H/C/005857; authorisation 09/12/2022; latest EC decision 18/02/2026 (revision 10). Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/livmarli
13. Ipsen Biopharmaceuticals. BYLVAY (odevixibat) capsules; oral pellets. Prescribing information. NDA 215498. Revised 12/2025. DailyMed setid 0935cd31-3251-4a0a-ac23-25ca4be12be9 (version 4). Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0935cd31-3251-4a0a-ac23-25ca4be12be9
14. European Medicines Agency. Kayfanda (odevixibat): EPAR — Product information (SmPC). EMEA/H/C/006462; authorisation 19/09/2024; latest EC decision 07/05/2025. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/kayfanda
15. ERN RARE-LIVER — European Reference Network on Rare Hepatological Diseases [Internet]. Available from: https://rare-liver.eu/
16. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Atrésie des voies biliaires. Centre de référence de l'atrésie des voies biliaires et cholestases génétiques. Saint-Denis La Plaine: HAS; 2024 Dec 2 [text dated November 2024; cited 2026 Nov 6]. Available from: https://www.has-sante.fr/jcms/p_3563486
17. Watson GH, Miller V. Arteriohepatic dysplasia: familial pulmonary arterial stenosis with neonatal liver disease. Arch Dis Child. 1973;48(6):459-66. doi:10.1136/adc.48.6.459. PMID: 4712776; PMC1648450.
18. Alagille D, Odièvre M, Gautier M, Dommergues JP. Hepatic ductular hypoplasia associated with characteristic facies, vertebral malformations, retarded physical, mental, and sexual development, and cardiac murmur. J Pediatr. 1975;86(1):63-71. doi:10.1016/s0022-3476(75)80706-2. PMID: 803282.
19. Alagille D, Estrada A, Hadchouel M, Gautier M, Odièvre M, Dommergues JP. Syndromic paucity of interlobular bile ducts (Alagille syndrome or arteriohepatic dysplasia): review of 80 cases. J Pediatr. 1987;110(2):195-200. doi:10.1016/s0022-3476(87)80153-1. PMID: 3806290.
20. Emerick KM, Rand EB, Goldmuntz E, Krantz ID, Spinner NB, Piccoli DA. Features of Alagille syndrome in 92 patients: frequency and relation to prognosis. Hepatology. 1999;29(3):822-9. doi:10.1002/hep.510290331. PMID: 10051485.
21. Li L, Krantz ID, Deng Y, Genin A, Banta AB, Collins CC, et al. Alagille syndrome is caused by mutations in human Jagged1, which encodes a ligand for Notch1. Nat Genet. 1997;16(3):243-51. doi:10.1038/ng0797-243. PMID: 9207788.
22. Oda T, Elkahloun AG, Pike BL, Okajima K, Krantz ID, Genin A, et al. Mutations in the human Jagged1 gene are responsible for Alagille syndrome. Nat Genet. 1997;16(3):235-42. doi:10.1038/ng0797-235. PMID: 9207787.
23. McDaniell R, Warthen DM, Sanchez-Lara PA, Pai A, Krantz ID, Piccoli DA, et al. NOTCH2 mutations cause Alagille syndrome, a heterogeneous disorder of the notch signaling pathway. Am J Hum Genet. 2006;79(1):169-73. doi:10.1086/505332. PMID: 16773578; PMC1474136.
24. GeneGuidance. JAG1 and NOTCH2 — gene records (clinical curation; version of 26 September 2026).
25. Hankeova S, Van Hul N, Laznovsky J, Verboven E, Mangold K, Hensens N, et al. Sex differences and risk factors for bleeding in Alagille syndrome. EMBO Mol Med. 2022;14(12):e15809. doi:10.15252/emmm.202215809. PMID: 36345711; PMC9728057.
26. Vandriel SM, Li LT, She H, Wang JS, Loomes KM, Piccoli DA, et al. Phenotypic Divergence of JAG1- and NOTCH2-Associated Alagille Syndrome & Disease-Specific NOTCH2 Variant Classification Guidelines. Liver Int. 2025;45(9):e70251. doi:10.1111/liv.70251. PMID: 40742203; PMC12312628.
27. Quintero-Bernabeu J, Padrós-Fornieles C, Mercadal-Hally M, Larrarte-King M, Mameli S, Miserachs-Barba M, et al. Beyond pruritus in Alagille syndrome: potential effects of maralixibat on fibrosis and portal hypertension-insights from two case studies. Front Med (Lausanne). 2025;12:1707258. doi:10.3389/fmed.2025.1707258. PMID: 41625768; PMC12855487.
28. ClinVar. VCV000177941 — NM_000214.3(JAG1):c.2122_2125del (p.Gln708fs). Available from: https://www.ncbi.nlm.nih.gov/clinvar/variation/177941/
29. Choi S, Kim DU, Kim Y, Kim L, Shim JO, Moon JS, et al. Native liver survival and genetic associations in Korean patients with Alagille syndrome. Eur J Pediatr. 2026;185(5). doi:10.1007/s00431-026-06917-3. PMID: 41961327; PMC13068675.
30. Gilbert MA, Keefer-Jacques E, Jadhav T, Antfolk D, Ming Q, Valente N, et al. Functional characterization of 2,832 JAG1 variants supports reclassification for Alagille syndrome and improves guidance for clinical variant interpretation. Am J Hum Genet. 2024;111(8):1656-1672. doi:10.1016/j.ajhg.2024.06.011. PMID: 39043182; PMC11339624.
31. ClinVar. VCV001057247 — NM_000214.3(JAG1):c.820G>T (p.Gly274Cys). Available from: https://www.ncbi.nlm.nih.gov/clinvar/variation/1057247/
32. Kamath BM, Bason L, Piccoli DA, Krantz ID, Spinner NB. Consequences of JAG1 mutations. J Med Genet. 2003;40(12):891-5. doi:10.1136/jmg.40.12.891. PMID: 14684686; PMC1735339.
33. McElhinney DB, Krantz ID, Bason L, Piccoli DA, Emerick KM, Spinner NB, et al. Analysis of cardiovascular phenotype and genotype-phenotype correlation in individuals with a JAG1 mutation and/or Alagille syndrome. Circulation. 2002;106(20):2567-74. doi:10.1161/01.cir.0000037221.45902.69. PMID: 12427653.
34. Kamath BM, Podkameni G, Hutchinson AL, Leonard LD, Gerfen J, Krantz ID, et al. Renal anomalies in Alagille syndrome: a disease-defining feature. Am J Med Genet A. 2012;158A(1):85-9. doi:10.1002/ajmg.a.34369. PMID: 22105858; PMC4511708.
35. Carvalho E, Porta G, Miura IK, Ferreira CHT, Silva LR, Sandy NS, et al. The Brazilian Alagille syndrome study: New insights from a multicenter national cohort. JPGN Rep. 2026 [Epub 2026 Jul 31]. doi:10.1002/jpr3.70224. PMID: 42540447; PMC13425789.
36. Kamath BM, Goldstein A, Howard R, Garner W, Vig P, Marden JR, et al. Maralixibat Treatment Response in Alagille Syndrome is Associated with Improved Health-Related Quality of Life. J Pediatr. 2023;252:68-75.e5. doi:10.1016/j.jpeds.2022.09.001. PMID: 36096175.
37. Perez CFM, Vandriel SM, Gonzales EM, Wang JS, Li LT, She H, et al. Elevated Serum Bile Acids Predict Poor Liver Outcomes in Children With Alagille Syndrome: Results From the GALA Study Group. Liver Int. 2025;45(12):e70423. doi:10.1111/liv.70423. PMID: 41250932; PMC12625800.
38. Schindler EA, Gilbert MA, Piccoli DA, Spinner NB, Krantz ID, Loomes KM. Alagille syndrome and risk for hepatocellular carcinoma: Need for increased surveillance in adults with mild liver phenotypes. Am J Med Genet A. 2021;185(3):719-731. doi:10.1002/ajmg.a.62028. PMID: 33369123; PMC7898517.
39. Ayoub MD, Bakhsh AA, Vandriel SM, Keitel V, Kamath BM. Management of adults with Alagille syndrome. Hepatol Int. 2023;17(5):1098-1112. doi:10.1007/s12072-023-10578-x. PMID: 37584849; PMC10522532.
40. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 112–115 (Congenital heart disease).
41. Kamath BM, Spinner NB, Emerick KM, Chudley AE, Booth C, Piccoli DA, et al. Vascular anomalies in Alagille syndrome: a significant cause of morbidity and mortality. Circulation. 2004;109(11):1354-8. doi:10.1161/01.CIR.0000121361.01862.A4. PMID: 14993126.
42. Kamath BM, Yin W, Miller H, Anand R, Rand EB, Alonso E, et al. Outcomes of liver transplantation for patients with Alagille syndrome: the studies of pediatric liver transplantation experience. Liver Transpl. 2012;18(8):940-8. doi:10.1002/lt.23437. PMID: 22454296.
43. Yanagi Y, Okumura S, Wakiya T, Hirata Y, Kim J, Honda M, et al. Long-Term Outcomes of Living Donor Liver Transplantation in Children With Alagille Syndrome: Results From Vanguard Multicenter Study of International Living Donor Liver Transplantation Group. Pediatr Transplant. 2026;30(8):e70426. doi:10.1111/petr.70426. PMID: 42583707; PMC13463210.
44. Huysentruyt K, Vandriel SM, Roelants M, Piccoli DA, Loomes KM, Rand EB, et al. Condition-Specific Growth Charts for Children With Alagille Syndrome. JAMA Netw Open. 2025;8(11):e2545294. doi:10.1001/jamanetworkopen.2025.45294. PMID: 41284294; PMC12645329.
45. Ayoub MD, Kamath BM. Alagille Syndrome: Diagnostic Challenges and Advances in Management. Diagnostics (Basel). 2020;10(11). doi:10.3390/diagnostics10110907. PMID: 33172025; PMC7694636.
46. Kaye AJ, Rand EB, Munoz PS, Spinner NB, Flake AW, Kamath BM. Effect of Kasai procedure on hepatic outcome in Alagille syndrome. J Pediatr Gastroenterol Nutr. 2010;51(3):319-21. doi:10.1097/MPG.0b013e3181df5fd8. PMID: 20601899.
47. Gottesman LE, Del Vecchio MT, Aronoff SC. Etiologies of conjugated hyperbilirubinemia in infancy: a systematic review of 1692 subjects. BMC Pediatr. 2015;15:192. doi:10.1186/s12887-015-0506-5. PMID:26589959.
48. 中华人民共和国国家卫生健康委员会 [National Health Commission of the PRC]. 阿拉杰里综合征诊疗指南（2025版）[Guideline for diagnosis and treatment of Alagille syndrome (2025 edition)]. J Clin Hepatol. 2025;41(9):1756-1758. doi:10.12449/JCH250907.
49. European Medicines Agency. Vedrop (tocofersolan) 50 mg/mL oral solution: EPAR — Product information (SmPC). EU/1/09/533; first authorisation 24/07/2009; renewal 23/04/2014. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/vedrop
50. Cheng K, Rosenthal P. Diagnosis and management of Alagille and progressive familial intrahepatic cholestasis. Hepatol Commun. 2023;7(12). doi:10.1097/HC9.0000000000000314. PMID: 38055640; PMC10984671.
51. U.S. Food and Drug Administration. RIFAMPIN prescribing information. Epic Pharma; revised 2023-12-20 (set id 0391f1e1-70d4-4b15-8b58-055bb6a385b8). Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0391f1e1-70d4-4b15-8b58-055bb6a385b8
52. ClinicalTrials.gov. NCT05488067: atorvastatin for xanthomas in ALGS. Available from: https://clinicaltrials.gov/study/NCT05488067
53. Hoskins BJ, Mogul DB, Chen C, Karnsakul W. Maralixibat Improves Xanthomas and Hypercholesterolemia in Children with Alagille Syndrome: A Post Hoc Integrated Analysis From Two Clinical Trials. J Pediatr. 2026;294:115108. doi:10.1016/j.jpeds.2026.115108. PMID: 42002216.
54. U.S. Food and Drug Administration. Approval letters NDA 214662 (S-000 corrected 21/10/2021; S-004 14/03/2023; S-005/S-008; S-009/S-010) and NDA 219485 (21/04/2025). Drugs@FDA. Available from: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=214662
55. U.S. Food and Drug Administration. Approval letters NDA 215498 (S-000 20/07/2021; S-003 14/06/2023; S-005 23/12/2025). Drugs@FDA. Available from: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=215498
56. European Medicines Agency. Bylvay (odevixibat): EPAR — Product information (SmPC). EMEA/H/C/004691; authorisation 16/07/2021; latest EC decision 26/03/2026 (revision 12). Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/bylvay
57. European Medicines Agency. Medicines data. Available from: https://www.ema.europa.eu/en/medicines/download-medicine-data
58. National Centre for Pharmacoeconomics (Ireland). Maralixibat (Livmarli®) for cholestatic pruritus in Alagille syndrome — HTA ID 25001. Summary, 30 March 2026. Available from: https://www.ncpe.ie/
59. National Institute for Health and Care Excellence. Maralixibat for treating cholestatic pruritus in Alagille syndrome [ID3941], GID-TA10832 — project page (in development). Available from: https://www.nice.org.uk/guidance/indevelopment/gid-ta10832
60. ClinicalTrials.gov. NCT07293897: database study of maralixibat (TAK-625) in Japan. Available from: https://clinicaltrials.gov/study/NCT07293897
61. Odevixibat for progressive familial intrahepatic cholestasis and Alagille syndrome-associated cholestatic pruritus. Aust Prescr. 2026;49(4):156-157. doi:10.18773/austprescr.2026.031. PMID: 42621451; PMC13487663.
62. Gonzales E, Hardikar W, Stormon M, Baker A, Hierro L, Gliwicz D, et al. Efficacy and safety of maralixibat treatment in patients with Alagille syndrome and cholestatic pruritus (ICONIC): a randomised phase 2 study. Lancet. 2021;398(10311):1581-1592. doi:10.1016/S0140-6736(21)01256-3. PMID: 34755627.
63. Shneider BL, Spino CA, Kamath BM, Magee JC, Ignacio RV, Huang S, et al. Impact of long-term administration of maralixibat on children with cholestasis secondary to Alagille syndrome. Hepatol Commun. 2022;6(8):1922-1933. doi:10.1002/hep4.1992. PMID: 35672955; PMC9315125.
64. Ovchinsky N, Aumar M, Baker A, Baumann U, Bufler P, Cananzi M, et al. Efficacy and safety of odevixibat in patients with Alagille syndrome (ASSERT): a phase 3, double-blind, randomised, placebo-controlled trial. Lancet Gastroenterol Hepatol. 2024;9(7):632-645. doi:10.1016/S2468-1253(24)00074-8. PMID: 38670135.
65. ClinicalTrials.gov. NCT05035030 (ASSERT-EXT): long-term odevixibat in ALGS. Available from: https://clinicaltrials.gov/study/NCT05035030
66. Miethke AG, Moukarzel A, Porta G, Covarrubias Esquer J, Czubkowski P, Ordonez F, et al. Maralixibat in progressive familial intrahepatic cholestasis (MARCH-PFIC): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Gastroenterol Hepatol. 2024;9(7):620-631. doi:10.1016/S2468-1253(24)00080-3. PMID: 38723644; PMC13409310.
67. ClinicalTrials.gov. NCT04729751 (RISE): maralixibat in infants with PFIC and ALGS. Available from: https://clinicaltrials.gov/study/NCT04729751
68. ClinicalTrials.gov. NCT05543174: TAK-625 (maralixibat) in ALGS (Japan). Available from: https://clinicaltrials.gov/study/NCT05543174
69. Kaltenbach E, Guiraud-Diawara A, Schwarzbard J, Artaud C. Indirect treatment comparison of odevixibat and maralixibat for the treatment of cholestatic pruritus in patients with Alagille syndrome. J Comp Eff Res. 2026;15(10):e260138. doi:10.57264/cer-2026-0138. PMID: 42755381; PMC13648047.
70. Hirschfield G, Vandriel SM, Mogul DB, Baek M, Vig P, Kamath BM. Maralixibat Reduces Serum Bile Acids and Improves Cholestatic Pruritus in Adolescents With Alagille Syndrome. Liver Int. 2025;45(2):e16201. doi:10.1111/liv.16201. PMID: 39823161; PMC11740003.
71. Sokol RJ, Gonzales EM, Kamath BM, Baker A, Vig P, Mogul DB, et al. Predictors of 6-year event-free survival in Alagille syndrome patients treated with maralixibat, an ileal bile acid transporter inhibitor. Hepatology. 2023;78(6):1698-1710. doi:10.1097/HEP.0000000000000502. PMID: 37278241; PMC10653287.
72. Hansen BE, Vandriel SM, Vig P, Garner W, Mogul DB, Loomes KM, et al. Event-free survival of maralixibat-treated patients with Alagille syndrome compared to a real-world cohort from GALA. Hepatology. 2024;79(6):1279-1292. doi:10.1097/HEP.0000000000000727. PMID: 38146932; PMC11095900.
73. Himes R, Rosenthal P, Dilwali N, Smith K, Venick R, Gonzalez-Peralta RP. Real-world experience of maralixibat in Alagille syndrome: Novel findings outside of clinical trials. J Pediatr Gastroenterol Nutr. 2024;78(3):506-513. doi:10.1002/jpn3.12101. PMID: 38334237.
74. Nwachukwu C, LeVine D, Himes R, Seal J, Domenick B, Rand EB, et al. Treatment of intractable pruritus with maralixibat in patients with Alagille syndrome before and after reversal of biliary diversion. JPGN Rep. 2025;6(4):508-514. doi:10.1002/jpr3.70092. PMID: 41245040; PMC12611582.
75. Ganschow R, Weigert A, Katzer D. Treatment of Pruritus With Maralixibat in Early-Stage Chronic Graft Dysfunction of a Child With Alagille Syndrome. Cureus. 2025;17(7):e88529. doi:10.7759/cureus.88529. PMID: 40851733; PMC12369989.
76. Fujishiro J, Suzuki K, Watanabe M, Uotani C, Takezoe T, Takamoto N, et al. Outcomes of Alagille syndrome following the Kasai operation: a systematic review and meta-analysis. Pediatr Surg Int. 2018;34(10):1073-1077. doi:10.1007/s00383-018-4316-3. PMID: 30073479.
77. ClinicalTrials.gov. NCT07290257 (LEAP-EU). Available from: https://clinicaltrials.gov/study/NCT07290257
78. ClinicalTrials.gov. NCT07585097: long-term odevixibat safety in ALGS (France, Italy). Available from: https://clinicaltrials.gov/study/NCT07585097
79. ClinicalTrials.gov. NCT06193928 (LEAP-US). Available from: https://clinicaltrials.gov/study/NCT06193928
80. ClinicalTrials.gov. NCT06850038: observational odevixibat (Bylvay) in ALGS (USA). Available from: https://clinicaltrials.gov/study/NCT06850038
81. Fawaz R, Baumann U, Ekong U, Fischler B, Hadzic N, Mack CL, et al. Guideline for the evaluation of cholestatic jaundice in infants: joint recommendations of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition and the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition. J Pediatr Gastroenterol Nutr. 2017;64(1):154–68. doi:10.1097/MPG.0000000000001334. PMID:27429428.
82. Saleh M, Kamath BM, Chitayat D. Alagille syndrome: clinical perspectives. Appl Clin Genet. 2016;9:75-82. doi:10.2147/TACG.S86420. PMID: 27418850; PMC4935120.
83. Leonard LD, Chao G, Baker A, Loomes K, Spinner NB. Clinical utility gene card for: Alagille Syndrome (ALGS). Eur J Hum Genet. 2014;22(3). doi:10.1038/ejhg.2013.140. PMID: 23881058; PMC3925268.
84. Vázquez-Frias R, Varela-Fascinetto G, Acosta-Rodríguez-Bueno CP, Consuelo A, Carrillo A, Reyes-Apodaca M, et al. A Retrospective Study of Clinical and Genetic Features in a Long-Term Cohort of Mexican Children with Alagille Syndrome. Int J Mol Sci. 2025;26(15). doi:10.3390/ijms26157626. PMID: 40806755; PMC12347795.
85. Alagille Syndrome Alliance (ALGSA) [Internet]. Available from: https://alagille.org/