Introduction
Alagille syndrome (AGS) is an autosomal dominant multisystem disorder with variable phenotypic penetrance. It is caused by mutations in genes active in the intercellular NOTCH signaling pathway, including JAG1 encoding the JAGGED1 ligand1,2 and NOTCH2 encoding the NOTCH2 receptor.3 Of the two genes, JAG1 mutations are responsible for about 95% of AGS cases.4 The JAG1 gene contains 26 exons, is located on chromosome 20p12.2, and encodes the JAGGED1 ligand. The ligand binds to the NOTCH transmembrane receptor encoded by NOTCH2 gene. The NOTCH intercellular signaling pathway has been shown to be related to cell fate during development in both invertebrates and vertebrates. The NOTCH proteins are involved in the regulation of development of various mammalian organs such as liver, heart, skeleton, kidney, and others (Fig. 1).5,6 The classical diagnosis of AGS7 is based on the presence of bile duct paucity in liver biopsies and at least three of the following five features, chronic cholestasis, cardiac disease (most commonly peripheral pulmonary stenosis), skeletal anomalies (typical butterfly vertebrae), ocular abnormalities (primarily posterior embryotoxon), and characteristic facial features (broad forehead, deep-set eyes, straight nose, pointed chin). AGS is almost exclusively diagnosed as cholestasis in infancy or early childhood. The prevalence is estimated as one case per 70,000 live births,8 but the availability of molecular genetics for diagnosis has increased the estimate to 1 in 30,000 live births.9 That may still be an underestimate, as some patients do not present with liver disease in infancy. In China, AGS is considered a rare disease, no reports clearly show the disease burden, and is often misdiagnosed in adults for reasons that include lack of awareness and the variable clinical findings. We conducted this retrospective study of AGS in adult patients to better understand the disease and establish a better diagnostic algorithm.
Methods
Patients
Eight adult patients with AGS from the Department of Infectious Diseases of the Third Affiliated Hospital of Sun Yat-sen University between June 2016 and June 2019 were included in the study. The diagnosis of AGS was based on the revised criteria described by Turnpenny et al,10 which include presence of a variant of JAG1 or NOTCH2 combined with a positive family history or one of the following, chronic cholestasis, skeletal abnormalities, peculiar facies, ocular abnormalities, renal involvement, or cardiovascular disorders. Informed consent to use the test results for medical research was obtained from the patients. The study conformed to the ethical principles of the Declaration of Helsinki and Good Clinical Practice, as well as Chinese regulatory requirements.
Clinical data were obtained from the patients’ medical records, and included demographic characteristics, symptoms, physical examination results, complications, and laboratory data. The patient characteristics included in the physical examination results were facial features, and the presence of heart murmurs and symptoms typical of chronic hepatitis (e.g., jaundice, liver palms, spider angiomas). Routine evaluation included ultrasonography of the liver and kidneys, radiography of the spine, echocardiography, and spiral computed tomography of the chest. Routine blood and urine routines were performed to rule out common liver diseases and to assess liver and renal function. Cholestasis was defined as serum direct/conjugated bilirubin >1.0 mg/dL (17 μmol/L)11 or serum alkaline phosphatase (ALP) >1.5 times the upper limit of normal (ULN) and glutamyl transpeptidase (GGT) >3 times the ULN.12
Percutaneous liver biopsy
Percutaneous liver biopsies were performed in four cases and the fixed, paraffin-embedded tissue was sectioned and stained for hematoxylin and eosin, Masson trichrome, cytokeratin-7, and cytokeratin-19. The histology was reviewed by liver pathologists. Bile duct paucity was defined as the absence of interlobular bile ducts in at least 50% of the portal tracts in liver specimens containing ten or more portal tracts.13
Genetic variation analysis
Genomic DNA was extracted from 3 mL of peripheral blood leucocytes with QIAamp Blood Midi Kits (Qiagen, Hilden, Germany) following the manufacturer’s instructions. Genes related to the target regions were selected for the panel. Biotinylated 60mer probes (P039-Exome; MyGenostics Inc., Beijing, China) were designed to align with the exons of the genes. Samples were organized into Illumina sequencing libraries, which were enriched for target region-associated genes with the MyGenostics Target Region Enrichment protocol. The captured libraries were sequenced with an Illumina HiSeq 2000 Sequencer (FC-404-2004; Illumina, San Diego, CA, USA).
Statistical analysis
Continuous variables were reported as means and SD or medians. Categorical variables (e.g., age and sex) were reported as numbers and percentages (%). The statistical analysis was performed with SPSS Statistics version 20 (IBM Corp., Armonk, NY, USA).
Discussion
AGS is a genetic disorder caused by mutations in JAG1 and NOTCH2, which are components of the NOTCH signaling pathway associated with the development of multiple organs,18,19 including bile duct and cardiovascular system20,21 Because the disease is most common in childhood and highly variable across patients,22 AGS is underdiagnosed in adults, especially those without a family history. The patients in this study were from 16 to 56 years of age. Most previously published studies describe pediatric patients. With the exception of some case reports, few summaries of the clinical features of AGS in adults are available. To the best of our knowledge, our study is the largest diagnostic series of adult AGS in China.
According to the traditional diagnostic algorithm, AGS should manifest as bile duct paucity and at least three other features, cholestasis, characteristic facies, vertebral anomalies, ocular abnormalities, and cardiovascular malformations. In our cohort, only case 2 met these criteria. The diagnostic criteria for AGS have changed considerably with the development of molecular testing. Kamath et al.23 and Turnpenny et al.10 revised the diagnostic criteria for AGS to include one or more characteristic clinical features and mutations in JAG1 or NOTCH2. Those with a family history of AGS and defects, even in the absence of some clinical manifestations, should be considered to have AGS and evaluated regularly.
As reported in other large cohorts,22,24 butterfly vertebrae were common in our patients, occurring in five (62.5%). However, in contrast to previously reported high cholestasis rates, we found that only 50% of the cases had cholestasis. More importantly, the percentages of congenital cardiac disease (12.5%), vascular anomalies (0%), posterior embryotoxon (0%), and typical facial features (12.5%) were also low in our cohort. It is possible that the lesions in these patients were mild and difficult to identify. Another possibility is that the manifestations of AGS are not the same in adults as in children. Previous studies22,25 have reported renal anomalies in 40–70% of clinically diagnosed AGS, including renovascular disease, tubulointerstitial nephritis, and renal dysplasia or agenesis. Some studies26 proposed that renal abnormalities should be considered as a disease-defining criterion for AGS. However, in this study, renal function (blood urea nitrogen and creatinine) and routine urine tests were normal, and renal ultrasonography found no structural abnormalities of the kidneys.
Children with AGS present with growth and mental retardation.27 Chronic disease and malabsorption of fat because of cholestasis may be causative factors. Other abnormalities, such as pancreatic insufficiency, hypothyroidism, small bowel stenosis, atresia, and tracheal stenosis, have also been reported in AGS.28,29 However, there were no similar findings in our study. Two cases had thalassemia, which we believe may be explained by the high incidence of the disease in Guangdong Province.30 The clinical features of AGS vary widely.31 Some patients have only mild elevations of serum bilirubin, bile acids, or liver transferases, but others have severe cholestasis that eventually leads to liver failure. In our study, all patients had hepatic involvement, but cholestasis occurred in only 50%, consistent with the guidelines for the evaluation of cholestatic jaundice in infants.11 However, in adults, the European Association for the Study of the Liver recommends levels of ALP >1.5 ULN and γGGT >3 ULN as biochemical thresholds for diagnosing cholestatic liver diseases.12 From that perspective, none of our cases met the requirements for the diagnosis of cholestasis.
Bile duct paucity is recognized as a necessary feature of AGS, but some patients are reluctant to undergo a liver biopsy because of its invasiveness, and not all patients who undergo liver biopsy show the absence of intrahepatic bile ducts. In our study, one of the four patients with a liver biopsy did not have bile duct loss, which may have been due to sampling error (Fig. 3). Therefore, liver biopsy is not the gold standard for diagnosis. In addition, the clinical manifestations of some patients, especially adults, are not typical, leading to misdiagnosis. Currently, genetic testing is an important diagnostic tool and perhaps should be considered the gold standard for diagnosis. AGS is a multisystem disorder caused by heterozygous mutations in the JAG1 and NOTCH2 genes.32 Our study found greater involvement of NOTCH2 mutations than it’s been previously reported. In our study mutant NOTCH2 accounted for half of the cases, but in previous reports, JAG1 accounted for a larger proportion, 70–94% of the cases.33–35 Perhaps the difference resulted from the older age of onset in our patients or racial differences. As few reports of specific genetic mutations and variations in adults with AGS have been published,36,37 more studies including patients from different regions and nationalities are needed to verify the results.
The genotype-phenotype correlations in the literature34 are not well documented. We found wide variation in phenotypic expression in our cohort. It was interesting to note the trend in increased bilirubin levels associated with NOTCH2, and that all patients with JAG1 mutations had butterfly vertebrae, and that only one case with a NOTCH2 mutation had butterfly vertebrae. In addition, one patient with a JAG1 mutation had abnormal cardiac sounds and characteristic facial features, but the other patients did not. However, because of the small sample size, no conclusions can be drawn. Studies of links between genotypes and phenotypes in AGS are ongoing, but no clear connections have been found. Factors that may contribute to the variability of genotypes known to be involved, probably involve a second gene that modifies the effects of a JAG1 or NOTCH2 mutation.32 The search for those genes is underway, and there is evidence that the post-translational addition of sugar moieties (glycosylation and fucosylation) to JAG1 protein has an impact on JAG1-mediated NOTCH signaling. The genes may be involved include Lunatic Fringe (LFNG), Radical Fringe (RFNG), and Manic Fringe (MFNG).38 Ryan et al. found that those genes affected the postnatal growth of bile ducts in mice that were haploinsufficient for both JAG1 and one of the other three genes.
Currently there is no novel treatments for AGS, not to mention genetic therapy. The best option for improving quality of life is to diagnose AGS as early as possible and initiate symptomatic treatment to slow the disease process. Lack of branching and elongation of bile ducts has been reported during postnatal liver growth in a patient with AGS.39 Theoretically, novel treatments targeting the NOTCH signaling pathway might promote formation of new bile ducts, thus improving the symptoms of cholestatic liver disease caused by AGS. The aim of treatment would be to increase the expression of JAGGED1 ligand or the NOTCH2 receptor. Such studies are underway to address the urgent need for those with early stages of liver cholestasis. For those who develop into end stage liver disease, liver transplant will be required. For younger patients, the predicted probability of survival to 20 years of age in all patients is 75%. The probability of survival to age 20 is 80% in patients who did not require liver transplantation.22 However, similar survival estimates of adult patients is not available, more studies are required, and antenatal screening of parents for AGS-related mutations is necessary.