Non-invasive screening for liver fibrosis in HIV/HCV coinfected patients

A research team from the University of Barcelona has found that serum levels of hyaluronic acid (HA) and a specific tissue inhibitor of metalloproteinase (TIMP-1) are predictive of higher stages of hepatic fibrosis in patients coinfected with HIV and hepatitis C virus.

These markers may serve as non-invasive alternatives to liver biopsy for determining the stage of liver fibrosis in coinfected patients. The results were published in the November 1^st edition of the Journal of Acquired Immune Deficiency Syndromes.

Liver disease such as hepatitis C virus (HCV) infection generally leads, over the long term, to liver tissue damage including fibrosis and cirrhosis. Liver biopsy, an invasive procedure requiring hospitalisation, is still the standard means of assessing the extent of hepatic (liver) tissue damage. Research efforts have been investigating other, non-invasive means of quantifying hepatic tissue damage without resorting to biopsy. The two main approaches are newer scanning technologies such as Fibroscan and the development of indices based on various metabolic markers in the blood serum. (See, for instance, the reports on aidsmap.com here, here and here.

Several such markers are already being studied in people with chronic hepatitis C infection: procollagen type III–N-terminal peptide (PIIINP) and hyaluronic acid (HA) have been extensively studied as markers of fibrosis. The metalloproteinase's and tissue inhibitors of metalloproteinases - markers of cellular changes in connective tissue – have shown a correlation with cirrhosis in people with chronic hepatitis C but have been little studied in people coinfected with HIV.

This study chose to assess the predictive value of five of these potential markers: metalloproteinases MMP-1 and MMP-2, specific tissue inhibitor of metalloproteinase TIMP-1, procollagen type III–N-terminal peptide (PIIINP), and hyaluronic acid (HA). (Other markers have also been investigated in the coinfected population – see this aidsmap.com report here).

The study population consisted of 119 HIV/HCV-coinfected patients at the clinic who consented to a liver biopsy as part of the study. All were on stable antiretroviral treatment for HIV, with HIV viral loads below 200 copies/ml and CD4 cell counts ≥ 250 cells/mm³; none had received any treatment for HCV. Patients were excluded for liver disease other than hepatitis C, autoimmune disease, or any other symptomatic infection, decompensated cirrhosis, cardiovascular disease, or pregnancy.

Median age was 39 years; 66% of the participants were male, and 77% had a history of injection drug use. Median length of HIV infection was twelve years and, of HCV infection, 19 years. Median CD4 count was 515 cells/mm³. Fibrosis stage, as determined by biopsy, was F0 in 22 participants (19%), F1 in 36 (30%), F2 in 24 (20%), F3 (bridging fibrosis) in 19 (16%), and F4 (compensated cirrhosis) in 18 (15%).

Various statistical models were used to evaluate the predictive value of the serum markers. Individually, all but MMP-1 were found to correlate positively and significantly with degree of fibrosis. Correlations were as follows, as measured by r-value (Sperman's correlation): TIMP-1, r = 0.6 (p < 0.001); MMP-2, r = 0.2 (p = 0.044); PIIINP, r = 0.4 (p < 0.001); HA, r = 0.5 (p < 0.001). Although MMP-1 levels were not themselves significantly correlated with fibrosis, the TIMP-1/MMP-1 ratio (r = 0.5, p< 0.001) was, as was TIMP-1/MMP-2 (r = 0.3, p < 0.001).

Investigators then analysed the ability of each marker to discriminate fibrosis stages of moderate or greater (F2 to F4) – the point at which therapy for HCV is generally recommended – from the lowest stages, F0 and F1.

TIMP-1 and HA levels proved moderately predictive of stage F2 or higher fibrosis. TIMP-1 was the best marker: TIMP-1 levels of 908.5 ng/ml or higher predicted moderate-or-higher fibrosis with a sensitivity of 65% and a specificity of 85%. HA levels were the next most useful predictor: a cutoff level of 39 µg/ml yielded a sensitivity of 60% and specificity of 88%. By multivariate analysis, both markers remained independently predictive of stage F2 or higher fibrosis – TIMP-1 with an odds ratio (OR) of 1.004 (95% confidence interval [CI] 1.002 to 1.006; p = 0.001), and HA>95 µg/mL, with an OR of 6.041 (95% CI, 1.184 to 30.816; p = 0.031).

A combined TIMP-1 and HA score was able to identify cases of F2 or higher fibrosis with a sensitivity of 72.9% and specificity of 83.1%. The predictive ability of the combined score compared favourably to that of other non-invasive predictors studied in the literature.

In summary, the researchers suggest that "an index of serum markers, including TIMP-1 [tissue inhibitor of matrix metalloproteinase-1] and HA [hyaluronic acid], may be clinically useful for detecting fibrosis in HIV/HCV-coinfected individuals. The combination serum levels of TIMP-1 and HA results were well correlated with the stages of liver fibrosis [and] could be used for establishing priorities in the management of HCV."