Abstract

Background: In patients with NASH, progression of liver fibrosis is the central determinant of poor liver related outcomes. Therefore, there is a critical need for treatment that improves or attenuates fibrosis in NASH. Recently, a RCT of pentoxifylline (PTX) in 55 patients with NASH showed greater decrease in fibrosis score among patients on PTX compared to placebo [Zein CO et al, Hepatology 2011]. In a smaller RCT of PTX in NASH hepatic expression of fibrosis genes was reduced by PTX [Van Wagner LB et al, Ann Hepatol 2011]. Liver biopsy is invasive, risky, and limited by discordance of fibrosis staging related to both sampling and inter-observer variability, thus accurate prediction of fibrosis using non-invasive markers is attractive. Thus far, studies of serum fibrosis markers in NASH have been few and limited to cross-sectional design. Aims: To compare levels of serum markers of fibrosis: TIMP-1, hyaluronic acid (HA) and amino-terminal propeptide of type 3 collagen (P3NP) in patients treated with PTX or placebo for one year. And, to correlate changes in these markers with the changes in fibrosis on liver biopsy before and after treatment. Methods: 80 patients treated with PTX (n=44) or placebo (n=36) who completed one year of therapy as part of two recently published RCTs were included. Biopsies of subjects in both trials were reviewed and scored by a single blinded pathologist. TIMP-1, HA, and P3NP were measured and a European Liver Fibrosis (ELF) panel score calculated using stored blood samples collected at entry and end of study. Changes from baseline were compared between treatment arms and correlated with histologic fibrosis stage. Results: After one year, patients treated with PTX had a greater decrease in all measured markers of fibrosis compared to those on placebo: TIMP-1 [mean change -14 (+/-41) vs +12.5 (+/-34), p=0.005]; HA [-15.7 (+/-50.7) vs +8.9 (+/-42.5), p=0.03]; P3NP [-1.3 (+/-2.3) vs +0.29 (+/-2), p=0.005]; and calculated ELF score [-0.18 (+/-0.4) vs +0.12 (+/-0.45), p=0.005]. In addition, subjects with improved or stable fibrosis on biopsy had significant decreases from baseline in levels of TIMP-1 (p=0.045), P3NP (p=0.033) and ELF calculation (p=0.041) compared to subjects with progression of fibrosis on histology. Conclusions: TIMP-1, HA, P3NP, and calculated ELF score decreased in patients with NASH treated with PTX compared to placebo. Furthermore, changes from baseline on TIMP-1, P3NP, and calculated ELF score correlated with non-progression or progression of fibrosis on follow up biopsy. These results not only further substantiate that PTX appears to have a favorable effect on fibrosis in patients with NASH, but also provide novel information on the performance of serum markers of fibrosis in relation to histological changes over time in the setting of a therapeutic intervention in NASH.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call