Elsevier

Journal of Hepatology

Volume 41, Issue 6, December 2004, Pages 935-942
Journal of Hepatology

Evaluation of a panel of non-invasive serum markers to differentiate mild from moderate-to-advanced liver fibrosis in chronic hepatitis C patients

https://doi.org/10.1016/j.jhep.2004.08.008Get rights and content

Background/Aims

In chronic hepatitis C (CHC) infection, a liver biopsy provides important information that guides treatment decisions, but is invasive, expensive and associated with possible complications. Extracellular matrix remodeling proteins may be useful non-invasive markers of fibrosis. The aim of this study was to evaluate the diagnostic accuracy of a panel of these markers in CHC patients, develop a predictive algorithm that differentiates no/mild (METAVIR F0–F1) from moderate/severe (F2–F4) fibrosis, and validate the model in external cohorts.

Methods

A combination of matrix markers were initially evaluated and optimized in 294 CHC patients from a single center, and validated in an external cohort of 402 patients.

Results

Hyaluronic acid, TIMP-1 and alpha2-macroglobulin were selected as having the best predictive accuracy for F2–F4 fibrosis (combined AUROC=0.831). At an index cut-off >0.36 and prevalence for F2–F4 of 52%, results in all 696 patients indicated positive and negative predictive values of 74.3 and 75.8% with an accuracy of 75%.

Conclusions

The three-marker panel may reliably differentiate CHC patients with moderate/severe fibrosis from those with no/mild fibrosis, although accurate delineation between stages was not possible. Prospective studies are required to determine the potential utility of the marker panel in guiding treatment decisions and following disease progression.

Introduction

An accurate assessment of disease severity by liver biopsy is important in providing information on prognosis and guiding treatment decisions in patients with chronic hepatitis C infection (CHC). However, liver biopsies are invasive, costly, and associated with possible complications [1]. Other limitations include sampling variability, low reproducibility and modest inter and intra-observer concordance on disease staging [2], [3]. Thus, accurate and cost-effective non-invasive methods to assess disease severity and monitor CHC patients would be a useful additional clinical tool. Unfortunately, clinical parameters alone are inaccurate in this respect [4].

Various simple serum biochemical markers have been proposed as potential markers of disease severity [5], [6], [7], [8]. Serum markers of fibrogenesis that have been evaluated in chronic liver disease include products of collagen synthesis or degradation [9], [10], [11], enzymes involved in matrix synthesis or degradation [12], [13], extracellular matrix glycoproteins [14] and glycosaminoglycans [15], [16], [17]. Single marker studies using hyaluronic acid, PIIINP or YKL-40 [18], [19] have noted only a moderate correlation with histological fibrosis, with their main utility being in excluding patients with cirrhosis [15], [17]. Studies evaluating a combination of serum markers have produced better results in terms of their predictive accuracy for significant fibrosis [20]. Indices based on five and six biochemical markers [21] and other algorithms with high predictive values for moderate-severe fibrosis have been proposed [22], [23], and may reduce the need for a liver biopsy in a substantial number of CHC patients. However, the clinical utility of some of these algorithms has been limited by indeterminate values for a significant proportion of patients.

The aims of this study were to initially evaluate the diagnostic accuracy of a panel of extracellular matrix and connective tissue protein markers of fibrogenesis in CHC patients, develop an accurate predictive model for the detection of moderate-to-severe stages of fibrosis, and to validate this model in patient cohorts from other centers.

Section snippets

Patient population

The initial study cohort included 194 CHC patients that had previously been evaluated for treatment at a single center (Scripps Clinic, La Jolla, CA) between 1992 and 2000. Selection of this initial cohort was based on histology, in order to provide an approximately equivalent number of patients for each of the fibrosis stages F0–F4. Seven fibrogenesis markers were then evaluated blindly in stored serum samples from this initial cohort in order to select the best combination of markers that

Patient characteristics

The demographic and baseline characteristics of the study population are summarized in Table 1. The patients were predominantly male, with a mean age of 45±8.7 years at the time of biopsy. Overall, patients had 2–3-fold elevations in ALT. Liver biopsy results for the estimation study cohort indicated a mean (±SEM) length of 13.1±0.22 mm. Biopsies were obtained within a mean (±SD) of 6.51±7.72 and 2.01±3.31 months of the serum sample for the estimation (n=294) and validation (n=402) cohorts,

Discussion

The present study outlines the design and validation of a serodiagnostic test panel for the detection of moderate-to-severe stages of fibrosis in a large cohort of CHC patients from the US and Europe. Our results indicate that a regression index model, derived from the combination of three non-invasive serum markers of fibrosis (hyaluronic acid, TIMP-1 and alpha2-macroglobulin), has good diagnostic accuracy and reasonable predictive values for the detection of moderate-to-severe stages of

Acknowledgements

The authors thank Arfana Kishan, MD of William Beaumont Hospital for assistance with this study. Financial support: Supported in part by research grants from Prometheus Laboratories Inc., San Diego, CA, The Duke Clinical Research Institute, Durham, NC, and a clinical research center grant from Scripps Clinic, La Jolla, CA (MO1-RR00833). Disclosure: KP, JMcH performed this work at Scripps Clinic, La Jolla, CA and Duke University, Durham, NC; EO and KMS are employees of Prometheus Laboratories

References (34)

  • X. Forns et al.

    Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model

    Hepatology

    (2002)
  • C.T. Wai et al.

    A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C

    Hepatology

    (2003)
  • P. Bedossa et al.

    An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group

    Hepatology

    (1996)
  • P. Bedossa et al.

    Sampling variability of liver fibrosis in chronic hepatitis C

    Hepatology

    (2003)
  • P. Bedossa

    Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. The French METAVIR Cooperative Study Group

    Hepatology

    (1994)
  • J. Westin et al.

    Interobserver study of liver histopathology using the Ishak score in patients with chronic hepatitis C virus infection

    Liver

    (1999)
  • R.J. Fontana et al.

    Noninvasive monitoring of patients with chronic hepatitis C

    Hepatology

    (2002)
  • Cited by (0)

    View full text