Article Text


PWE-140 Comparison Of 4 Serum Markers Panels of Fibrosis in Chc: Variants of the Hyaluronic Acid (HA) Assay Significantly Affect Their Diagnostic Performance
  1. S Tanwar1,
  2. P M Trembling1,
  3. B Hogan1,
  4. E L Ellis2,
  5. J Parkes2,
  6. P Grant3,
  7. E Nastouli3,
  8. D Schuppan5,
  9. W M Rosenberg1
  1. 1UCL Institute for Liver & Digestive Health, University College London, London
  2. 2University of Southampton, Southampton
  3. 3Department of Clinical Microbiology and Virology, University College London Hospitals NHS Foundation Trust, London, UK
  4. 4University of Erlangen, Erlangen
  5. 5Molecular and Translational Medicine, Department of Medicine, University of Mainz, Mainz, Germany


Introduction The detection of advancing fibrosis in patients with CHC and prior treatment failure is important for ascertaining prognosis. HA has been used alone and as a constituent component of fibrosis marker panels. The aim of this study was to compare the performance of 4 marker panels in the detection of moderate-to-severe fibrosis (Metavir F2–4) and to assess the influence on diagnostic performance of using 2 different validated assays for HA.

Methods 80 patients with CHC, all non-responders or relapsers to IFN-based treatment, were included in this study. Sera obtained within 6 months of liver biopsy were used to measure 4 biomarker panels incorporating HA (ELF, Fibrospect-II, Hepascore, Fibrometer-2G) using 2 validated assays for HA (ELISA-Siemens, radiometric-Pharmacia). Diagnostic performance for the detection of moderate-to-severe fibrosis was assessed by AUROC and by evaluating biomarker performance at published thresholds.

Results The prevalence of moderate-to-severe fibrosis was 63% (F0–8%, F1–29%, F2–24%, F3–30%, F4–9%). The AUROC of the Siemens HA assay was higher than the Pharmacia assay (0.80 Vs. 0.69, P = 0.005). Incorporating the Siemens assay for HA, the performance of the panels were not statistically significantly different but Fibrometer 2G generated the highest AUROC. Using the Siemens assay, ELF and Fibrometer 2G had the highest PPV and NPV respectively (88% and 100% using published thresholds). The use of the Pharmacia assay for HA to calculate the biomarkers did not reduce the discriminatory ability of the 4 panels (p = NS). However whereas the other panels were largely unaffected, the use of the Pharmacia assay resulted in a dramatic reduction in the performance of published thresholds of the ELF test with the a prioriand a posteriori probability of fibrosis being equal.

Abstract PWE-140 Table

Conclusion In this study the performance of the 4 biomarker panels to detect moderate-to-severe fibrosis was comparable. The diagnostic performance of biomarker panels may be significantly effected by the selection of the individual component assays as demonstrated by comparison of the results obtained with different HA assays.

Disclosure of Interest None Declared.

Statistics from

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.