MicroRNA Sequencing Identifies a Serum MicroRNA Panel, Which Combined With Aspartate Aminotransferase to Platelet Ratio Index Can Detect and Monitor Liver Disease in Pediatric Cystic Fibrosis

Abstract

Cystic fibrosis (CF)-associated liver disease (CFLD) is a hepatobiliary complication of CF. Current diagnostic modalities rely on nonspecific assessments, whereas liver biopsy is the gold standard to assess severity of fibrosis. MicroRNAs (miRNAs) regulate liver disease pathogenesis and are proposed as diagnostic biomarkers. We investigated the combined use of serum miRNAs and aspartate aminotransferase (AST) to platelet ratio (APRI) to diagnose and assess CFLD severity. This was a cross-sectional cohort study of the circulatory miRNA signature of 124 children grouped by clinical, biochemical, and imaging assessments as follows: CFLD (n = 44), CF patients with no evidence of liver disease (CFnoLD; n = 40), and healthy controls (n = 40). Serum miRNAs were analyzed using miRNA sequencing (miRNA-Seq). Selected differentially expressed serum miRNA candidates were further validated by qRT-PCR and statistical analysis performed to evaluate utility to predict CFLD and fibrosis severity validated by liver biopsy, alone or in combination with APRI. Serum miR-122-5p, miR-365a-3p, and miR-34a-5p levels were elevated in CFLD compared to CFnoLD, whereas miR-142-3p and let-7g-5p were down-regulated in CFLD compared to CFnoLD. Logistic regression analysis combining miR-365a-3p, miR-142-3p, and let-7g-5p with APRI showed 21 times greater odds of accurately predicting liver disease in CF with an area under the receiver operating characteristics curve (AUROC) = 0.91 (sensitivity = 83%, specificity = 92%; P < 0.0001). Expression levels of serum miR-18a-5p were correlated with increasing hepatic fibrosis (HF) stage in CFLD (r_s = 0.56; P < 0.0001), showing good diagnostic accuracy for distinguishing severe (F3-F4) from mild/moderate fibrosis (F0-F2). A unit increase of miR-18a-5p showed a 7-fold increased odds of having severe fibrosis with an AUROC = 0.82 (sensitivity = 93%, specificity = 73%; P = 0.004), indicating its potential to predict fibrosis severity. Conclusion: We identified a distinct circulatory miRNA profile in pediatric CFLD with potential to accurately discriminate liver disease and fibrosis severity in children with CF.