120
Can Routine Non-invasive Tests Predict Liver Histology in HIV/HCV Co-infection? Analysis of Patients Entering the AIDS PEGASYS Ribavirin International Co-infection Trial (APRICOT)
Richard Sterling*1, E Lissen2, N Clumeck3, R Sola4, M Correa5, J Montaner6, M Sulkowski7, F Torriani8, D Dieterich9, D Thomas7, D Messinger10, and M Nelson11
1Virginia Commonwealth Univ Hlth System, Richmond, USA; 2Virgen del Rocío Univ Hosp, Seville, Spain; 3Ctr Hosp Univ St-Pierre, Brussels, Belgium; 4Hosp del Mar, Univ Autónoma de Barcelona, Spain; 5HCFMUSP, Casa da AIDS, Sao Paulo, Brazil; 6Univ of British Columbia, Vancouver, Canada; 7Johns Hopkins Univ Sch of Med, Baltimore, MD, USA; 8Univ of California, San Diego, Treatment Ctr, USA; 9Mt Sinai Sch of Med, New York, NY, USA; 10IST GmbH, Mannheim, Germany; and 11Chelsea and Westminster Hosp, London, UK
Background: Liver biopsy
remains the gold standard in the assessment of severity of liver disease in
patients with chronic hepatitis C virus (HCV). However, liver biopsy is not
without risk. Use of non-invasive tests has gained popularity
to predict liver histology in patients with HCV. However, many models include
tests not readily available and there are limited data from pts with HIV/HCV co-infection.
We aimed to develop a model using routine laboratory
tests to predict liver histology in pts with HIV/HCV.
Methods: Retrospective analysis of liver histology in 832
of 868 patients entering APRICOT. The design and
results of APRICOT have been published. Liver histology was assessed by Ishak score (missing in 36 patients) and patients were
categorized as none-mild (0 to 1, 36%), moderate (2 to 3, 43%), or advanced (4
to 6, 21%) fibrosis scores. Patients were randomly assigned to training (n = 555)
or validation (n = 277) sets. Variables possibly associated with liver
histology included demographics, laboratory, virologic,
immunologic, and HAART use. Variables associated with fibrosis severity on univariate logistic regression analyses were entered into a
stepwise multiple logistic regression for 3-level
response.
Results: Demographic, laboratory, and histology were
similar between the training and validation sets. Univariate
analysis found age, platelet count, INR, PTT, albumin, AST, bilirubin,
and alkaline phosphatase to be significantly
different between fibrosis groups (p <
0.05). Multivariate logistic regression analysis found that PLT, age, AST, and
INR remained significant. Additional analysis revealed PLT, age, AST, and ALT
as an alternative model. Based on this, a simple index (FIB-4) was developed: age [years] AST [U/L]/(PLT [109/L]/(ALT
[U/L])½.
The values vary from about 0.2 to 10 and the ROC AUC of the index is 0.76. At a
cut-off of ≤ 1.45 in the validation set, the negative predictive value to
exclude advanced fibrosis (stage 4 to 6) was 90% with sensitivity of 70%. A
cut-off of ≥ 3.25 had a positive predictive value of 65% and specificity
of 97%. Using these cut-offs, 87% of the 198 patients with FIB-4 values outside
1.45 to 3.25 would be correctly classified and could avoid liver biopsy in 198
of 277 patients (71%) of the validation group. FIB-4 was less predictive
between stage 0 to 1 and stage 2 to 6.
Conclusions: Noninvasive tests can accurately predict hepatic
fibrosis and may reduce the need for liver biopsy in some HIV/HCV-co-infected patients.
The utility of a novel index (FIB-4) using routine laboratory tests requires
further study.
Keywords: liver histology; hepatic fibrosis; hepatitis C virus
