JPGN reports · 7 authors, 3 centres
This summary was generated by AI from a single paper. It has not been reviewed by a clinician and is not clinical advice. Verify against the source before acting on it.
This study evaluated a 61-gene panel for diagnosing neonatal/infantile intrahepatic cholestasis (NIIC) in 315 Japanese patients. The panel achieved a 26.6% diagnostic rate in a prospective cohort, with Alagille syndrome, Dubin-Johnson syndrome, and citrin deficiency accounting for 78.8% of genetic diagnoses. The findings support expanded genetic testing and suggest that Crigler-Najjar syndrome type 2 patients may present with cholestasis and should be considered in NIIC evaluations.
**Background:** Genetic liver diseases such as Alagille syndrome (ALGS), progressive familial intrahepatic cholestasis (PFIC), and cystic fibrosis are important causes of neonatal/infantile intrahepatic cholestasis (NIIC), which can progress to fibrosis and cirrhosis. Clinical diagnosis is difficult due to overlapping features. The authors previously used an 18-gene panel achieving a 26% diagnostic rate, leaving ~70% of patients undiagnosed. This study aimed to evaluate a new 61-gene panel (advanced NIIC gene panel) in two cohorts.
**Methods:** Patients were recruited from 150 hospitals in Japan between May 2013 and October 2019. Cholestasis was defined as serum direct bilirubin (D.Bil) >1.0 mg/dL. Inclusion criteria: cholestasis, onset <12 months, no prior definitive molecular diagnosis. Exclusion: extrahepatic cholestasis, chromosomal abnormalities, birth before January 2010. The retrospective reanalysis group included 191 patients previously tested with an 18-gene panel (May 2013–September 2017) without a diagnosis. The prospective analysis group included 124 newly recruited patients (October 2017–October 2019) with no prior genetic testing. The 61-gene panel covered 998 exons, 1,387 amplicons, and 155,387 total targeted bases with 99.0% theoretical coverage. Sequencing used Ion PGM or Ion GeneStudio S5 systems. Variants were classified per ACMG guidelines. For UGT1A1, biallelic null variants were classified as Crigler-Najjar syndrome (CNS) type 1, biallelic severe reduction variants as CNS type 2, and biallelic mild reduction as Gilbert syndrome (GS).
**Key Results:** In the retrospective reanalysis group, 10 of 191 patients (5.2%) received a definitive molecular diagnosis: 3 mitochondrial DNA depletion syndrome (MPV17), 2 neonatal Dubin-Johnson syndrome (nDJS; ABCC2), 2 PFIC type 5 (NR1H4), 1 Niemann-Pick disease type C (NPC1), 1 cerebrotendinous xanthomatosis (CYP27A1), and 1 cystic fibrosis (CFTR). Eight of these 10 had causative variants in genes not included in the old 18-gene panel. In the prospective analysis group, 33 of 124 patients (26.6%) received a diagnosis: 10 ALGS (JAG1/NOTCH2; 8.1%), 9 nDJS (ABCC2; 7.3%), 7 NICCD (SLC25A13; 5.6%), 2 PFIC2 (ABCB11; 1.6%), 1 mitochondrial DNA depletion syndrome (POLG; 0.8%), 1 NPC (NPC1; 0.8%), 1 cystic fibrosis (CFTR; 0.8%), 1 PFIC1 (ATP8B1; 0.8%), and 1 PFIC3 (ABCB4; 0.8%). ALGS, nDJS, and NICCD together accounted for 78.8% of genetic diagnoses. Twenty of the causative variants were novel. In 37 patients (20 retrospective, 17 prospective), only a single heterozygous pathogenic/likely pathogenic variant was identified in autosomal recessive disease genes, most commonly ABCB11 (12 patients), SLC25A13 (5), and ABCC2 (5). Three patients in the retrospective group had genotypes associated with CNS type 2, showing severe indirect hyperbilirubinemia and mild cholestasis (T.Bil/D.Bil: 24.0/1.3, 19.7/1.4, and 24.3/2.4 mg/dL). GS genotypes were found in 21 of 191 (11.0%) retrospective and 11 of 124 (8.9%) prospective patients. Median age at NGS was 46 months (IQR 33–64) in the retrospective group and 4 months (IQR 3–6) in the prospective group (P<0.01).
**Clinical Implications:** The advanced 61-gene panel improved diagnostic yield over the 18-gene panel by including rare but clinically important genes (MPV17, NR1H4, NPC1, CYP27A1, CFTR, POLG). The 26.6% diagnostic rate in the prospective cohort is similar to the prior 25.7% rate, suggesting that panel size alone does not dramatically increase yield in Japan where three conditions (ALGS, nDJS, NICCD) predominate. However, the panel enabled diagnosis of rare diseases (e.g., cystic fibrosis, cerebrotendinous xanthomatosis) that would otherwise be missed, supporting use of a broad, universally applicable panel regardless of ethnicity. The identification of CNS type 2 patients presenting with direct hyperbilirubinemia indicates that clinicians should consider UGT1A1-related disorders in NIIC cohorts. The 37 patients with single heterozygous variants highlight the need for whole genome sequencing or RNA analysis to identify cryptic mutations. The study is limited by potential selection bias from different recruitment periods and the lack of functional validation for many variants.