TBA (16S125)

Developing a pragmatic community-based algorithm for prioritization of HCV treatment.

Author(s)

D Storan, S Rutledge, C Kiat, S Stewart

Department(s)/Institutions

Centre for Liver Disease, Mater Misericordiae University Hospital, 55 Eccles St, Dublin 7

Introduction

HCV is a leading cause of cirrhosis and liver related mortality in Ireland. The advent of direct acting antivirals has removed many of the contraindications to treatment associated with interferon. High cost, however, has led to the rationing of therapy and we have largely limited treatment to those with established cirrhosis based on imaging, biopsy or FibroScan® (FS) score > 12kPa.

Aims/Background

As we look to expand access to treatment, we sought to assess more accessible modalities for staging, namely APRI and FIB-4, both of which can be calculated using readily available blood tests. The aim of this study was to determine if FIB-4 or APRI score could be used in primary care to detect patients with a FS >12kPa.

Method

All FS tests performed from 26/7/2011 to 16/9/2015 were analysed and duplicates and inaccurate FS scores (SR <60%, IQR <0.3) were removed. 223 patients were included in the study. Laboratory values closest to the date of FS were used in the formulae with a limit of 1 year either side of the FS date. Data were analysed using Microsoft Excel. APRI score was calculated as (AST/ULN)/(Platelet count x 100) and FIB-4 was calculated as age x AST/platelet count x √ALT.

Results

55/223 (25%) had a FS score of ≥12kPa and qualified for treatment. The performance of APRI and FIB-4 in detecting these patients at various cut-offs is as follows: APRI >0.7 (110/223); 80% sensitivity, 61% specificity, 40% PPV, 90% NPV APRI >1 (69/223); 62% sensitivity, 79% specificity, 49% PPV, 86% NPV APRI >1.5 (40/223); 47% sensitivity, 92% specificity, 65% PPV, 84% NPV FIB-4 >1.45 (120/223); 91% sensitivity, 58% specificity, 42% PPV, 95% NPV FIB-4 >3.25 (46/223); 60% sensitivity, 92% specificity, 72% PPV, 88% NPV

Conclusions

The FIB-4 score is superior to the APRI when used to detect HCV patients with a Fibroscan score of >12 kPa. A cut-off of 3.25 performs best overall, but a cut-off of 1.45 would be the most practical for use in the community. This could be used to reduce those requiring referral for Fibroscan by 46% while detecting 91% of those that will be approved for treatment.