933. Hepascore: A Non-invasive Marker for Advanced Fibrosis and Response to Antiviral Therapies in Chronic Hepatitis B Infection

[Guest guest]

http://www.hbvadvocate.org/news/reports/HBV_AASLD_2008/Abstracts/Nov2%20Disease%\

20Progression.htm#Nov2a933

933. Hepascore: A Non-invasive Marker for Advanced Fibrosis and Response to

Antiviral Therapies in Chronic Hepatitis B Infection

S. C. Raftopoulos; E. Rossi; B. De Boer; G. P. ; D. J. Speers; G. C.

MacQuillan; L. A.

Staging liver fibrosis in chronic hepatitis B (CHB) infection is important for

prognostication and management, however few validated non-invasive markers of

liver fibrosis are currently available, and it is unknown whether these are

responsive to changes in fibrosis over time. We examined the accuracy of a

non-invasive model developed for hepatitis C infection (Hepascore) in

determining liver fibrosis in CHB infection, and secondly, the longitudinal

response of Hepascore over time among patients with or without anti-viral

therapy.

Methods:

Patients with CHB infection had serum taken within 6 months of a liver biopsy

analysed for hyaluronic acid, alpha 2 macroglobulin, bilirubin, gamma

glutamyltransferase and combined with age and gender to produce a Hepascore

ranging from 0-1. A subset of patients with serum stored over time had serial

Hepascore’s performed. Liver biopsies were interpreted by a single liver

pathologist and scored according to METAVIR. Fibrosis was quantified as

significant (F2-4), advanced (F3-4) or cirrhosis (F4). Accuracy of Hepascore was

calculated by receiver operator characteristic (ROC) curves.

Results: 70 patients (73% male) with a mean ±SD age of 42.5 ±12.6 years

underwent biopsy. The majority were Asian (59%) and eAg negative (60%). The

median time between serum collection and biopsy was 0.7 months (range 0-6).

Thirty two (46%) had significant fibrosis, 21 (30%) had advanced fibrosis and 12

(17%) were cirrhotic. Hepascore had an area under the ROC curve for significant

fibrosis, advanced fibrosis and cirrhosis of 0.91 (95% CI 0.84-0.98), 0.94 (95%

CI 0.89-0.99) and 0.92 (95% CI 0.85-0.99) respectively. A cutoff of 0.62

provided sensitivity, specificity, positive and negative predictive values (NPV)

of 85%, 84%, 82% and 87% respectively for significant fibrosis. The same cutoff

provided a 100% NPV for advanced fibrosis and cirrhosis. Among 18 patients who

were monitored over a median of 1.7 (range 0.4-6.8) years without anti-viral

treatment, mean Hepascore tended to increase over time (0.59 ±0.30 to 0.72

±0.29, p=0.09). Among 21 patients who had anti-viral therapy over a median of

2.9 (0.8-7.0) years, mean Hepascore did not change significantly (0.65 ±0.29 to

0.61 ±0.35, p=0.56). The annual rate of change of Hepascore was significantly

higher in those that did not receive anti-viral therapy compared to those who

did (0.07 ±0.15 vs -0.04 ±0.13, p=0.04).

Conclusion:

Hepascore is an accurate predictor of liver fibrosis in patients with CHB

infection with potential to avoid liver biopsy. The rate of change for hepascore

varies with anti-viral therapy suggesting it is responsive to changes in

fibrosis over time.