Article Text

PDF

Clinical hepatology
P15 Non-invasive assessment of hepatic fibrosis in prior non-responders to hepatitis C virus treatment—a comparison of eight marker panels of liver fibrosis
  1. S Tanwar1,
  2. E Ellis2,
  3. C Herold3,
  4. W Rosenberg1
  1. 1Centre for Hepatology, University College London, London, UK
  2. 2University of Southampton, Southampton, UK
  3. 3University of Erlangen, Erlangen, Germany

Abstract

Introduction The detection of advancing fibrosis in patients who have previously failed standard therapy for hepatitis C virus (HCV) is important both for ascertaining prognosis and stratifying patients for further treatment with emerging therapies. Whilst liver biopsy remains the reference standard, non-invasive markers of liver fibrosis may be able to reduce the need for liver biopsy in this group of patients.

Method 80 previous non-responders to pegylated interferon and ribavirin (46 male, 34 female, age 24–98 years, mean 48.9) were recruited from five centres. Serum was taken at the time of liver biopsy. Seven tests of liver fibrosis and simple biochemical markers were compared. These were: Hyaluronic acid (HA); Indirect tests: APRI, Forn's, Fib-4; Tests including HA: SHASTA, Hepascore, Fibrometer and ELF test. Area under receiver operating characteristic curves (AUROC) were plotted for minimal fibrosis (F0–1 vs F2–6), mild fibrosis (F0–2 vs F3–6), moderate fibrosis (F0–3 vs F4–6), and severe fibrosis/cirrhosis (F0–4 vs F5–6).

Results AUROCs (and asymptotic 95% confidence intervals) are presented for each test for minimal, mild and severe fibrosis.

TestF0–1 vs F2–6 n=26 n=54F0–2 vs F3–6 n=35 n=45F0–4 vs F5–6 n=57 n=23
ELF0.802
0.705–0.900
0.851
0.770–0.932
0.859
0.765–0.953
Hepascore0.777
0.674–0.880
0.757
0.653–0.861
0.859
0.766–0.925
Fibrometer0.775
0.669–0.876
0.740
0.632–0.849
0.808
0.698–0.917
SHASTA0.652
0.532–0.772
0.654
0.535–0.773
0.806
0.691–0.921
HA0.668
0.551–0.785
0.710
0.597–0.822
0.769
0.637–0.902
FIB-40.714
0.597–0.832
0.768
0.666–0.869
0.814
0.704–0.924
Forns0.717
0.597–0.837
0.763
0.660–0.866
0.813
0.701–0.925
APRI0.654
0.529–0.780
0.674
0.557–0.791
0.764
0.650–0.878
Abstract P15 Table 1

Results

ELF was best at detecting lesser degrees of fibrosis and was better than the indirect marker panels either with or without HA. ELF and Hepascore were best at detecting advanced fibrosis, but other panels performed adequately with similar AUCs.

Conclusion In this cohort of previous non-responders to HCV treatment, the ability to discriminate advanced fibrosis appears to be similar amongst most of the markers tested. In contrast, of the eight markers tested in this study, the ELF panel appears to have the most consistent diagnostic performance across all stages of fibrosis and performs well in the detection of minimal and mild fibrosis. ELF testing would permit stratification of previous non-responders for further anti-HCV and anti-fibrotic therapy and for screening for complications of cirrhosis.

Statistics from Altmetric.com

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.