Gamma-glutamyl transpeptidase-to-platelet ratio as a biomarker of liver disease and hepatic fibrosis severity in paediatric Cystic Fibrosis.

ABSTRACT

BACKGROUND: Cystic fibrosis (CF)-associated liver disease (CFLD) causes significant morbidity and mortality in children with CF. Diagnosis of liver disease prior to development of cirrhosis or portal hypertension (PHT) is challenging. While imaging modalities using Elastography show great promise they are still not widely available to all clinicians. This study investigated gamma-glutamyl transpeptidase-to-platelet ratio (GPR) as a non-invasive biomarker to detect liver disease and stage fibrosis severity in children with CF. METHODS: 237 children were enroled including 76 with CFLD and 161 with CF and no detectable liver disease (CFnoLD). CFLD was diagnosed using standard clinical, biochemical and imaging practice guidelines. Hepatic fibrosis was staged on liver biopsies available from 54 children with CFLD. Serum liver biochemistry was used to calculate GPR (median, [IQR]) and receiver operating characteristics (ROC) analysis assessed utility to detect liver disease and stage fibrosis severity. RESULTS: GPR was significantly increased in CFLD versus CFnoLD (0.33 [0.19-0.96] vs. 0.15 [0.11-0.21], P<0.0001). GPR demonstrated good diagnostic utility for detecting CFLD with an area under the curve (AUC) of 0.81 (95% confidence Interval [CI] [0.75-0.87]; P<0.0001), with sensitivity of 74% and specificity of 73%, using a cut-off of 0.20. GPR increased with increasing hepatic fibrosis stage. GPR discriminated both moderate-advanced (F2-F4) fibrosis vs. F0-F1 (AUC=0.82; 95%CI [0.71-0.94]; P<0.0001) and advanced (F3-F4) fibrosis vs. F0-F2 (AUC=0.77; 95%CI [0.64-0.90]; P = 0.004), with a cut-off 0.32 and 0.61, respectively. An elevated GPR of >0.84 was predictive of PHT at diagnosis of CFLD (AUC=0.81; 95%CI [0.67-0.95]; P = 0.0003). CONCLUSIONS: GPR demonstrates good diagnostic utility for assessing the presence of liver disease, PHT and hepatic fibrosis severity in children with CF. These findings will aid in better identification of patients at risk for CF-related liver involvement and the potential for more targeted and timely follow-up and treatment.