Motivation Diagnosis of rare liver diseases with genetic origin in children remains a challenging
task, often due to unspecific symptoms and limited accessibility to comprehensive
genetic testing. Previous work has shown efficacy of diagnosis rates for targeted
gene-panel-analysis, using preselected patient cohorts based on phenotypes, mainly
in cholestatic diseases. The aim of this work was to evaluate this efficacy for a
heterogenous patient group with different phenotypes throughout all age subgroups.
Methods 56 patients (24 female, mean age: 47,5±69,3 months) were evaluated for unexplained
liver disease in 11 pediatric hepatology centers in Germany. All underwent high-throughput-sequencing
(targeted exome-panel-analysis performed at CeGaT Tuebingen between 2014-2020) with
clinical interpretation of up to 128 genes. We subdivided this cohort according to
the degree of phenotypic characterization before sequencing: I: phenotype associated
with specific suspected clinical diagnosis; II: phenotype likely associated with specific
group of diseases (e.g., PFIC); III: no association.
Results The diagnostic yield was 53,6%. Upon identification of a specific genotype in this
group, a therapeutic concept could be established. The most common diagnosis was PFIC
(n=10). The diagnostic yields in subgroups were I: 64%; II: 42%; III: 57%.
Conclusion The diagnostic yield was comparable to previous studies even though a heterogenous
group without preselection was evaluated. We show that exome-panel-analysis can be
effective even for patients in which phenotyping does not allow association to a specific
suspected disease pre-genotyping. In the future (trio-)whole-genome-sequencing and/or
transcriptome-sequencing may lead to further improve the diagnostic yield and potentially
identify new disease-causing genes.
