1033
Immunopathogenesis of Hepatic Flares after Initiation of ART in HIV/HBV-co-infected Individuals
Megan Crane*1, B Oliver2, G Matthews3, A Avihingsanon4, P Price2,5, K Visvanathan1, M French2,5, G Dore3, K Ruxrungtham4, and S Lewin1,6
1Monash Univ, Melbourne, Australia; 2Univ of Western Australia, Perth; 3Natl Ctr for HIV Epidemiology and Clin Res, Univ of New South Wales, Australia; 4HIV Netherlands Australia Thailand Res Collaboration, Bangkok; 5Royal Perth Hosp, Australia; and 6Alfred Hosp, Melbourne, Australia
Background: Following the initiation of hepatitis B
virus (HBV) -active HAART in HIV/HBV-co-infected individuals, hepatic flare or
worsening alanine aminotransferase (ALT) is a major clinical problem. We
hypothesized that in this setting, hepatic flare is a form of immune
restoration disease (IRD) and is a consequence of enhanced activated T cell recruitment
to the liver in the setting of high HBV viral load.
Methods: TICO is a prospective randomized (1:1:1)
trial of tenofovir (TDF) vs lamivudine (LMV) vs TDF/LMV within an efavirenz (EFV)
-based HAART regimen initiated in 36 HIV/HBV-co-infected antiretroviral-naïve
patients in Thailand. Hepatic flare was defined as an alanine aminotransferase
(ALT) >5 X upper limit of normal or >200 IU/mL above baseline, within 12
weeks of HAART initiation. Quantification of immune mediators (interleukin [IL]
-18, -2, -6, -8, -10; soluble[s] CD26, sCD30, IP-10, macrophage chemotactic
protein [MCP] -1, tumor necrosis factor [TNF] -a,
interferon [IFN] -g and -a) in baseline, week 4, 8, and 12 sera was
performed by enzyme-linked immunosorbent assay (ELISA), antigen capture
bioassay, or cytometric bead array. We compared cases (hepatic flare, n =
8) with controls (n = 28), using either a Kruskal-Wallis with Dunn's
multiple comparison post test (non-parametric) or repeated measures ANOVA with
post-hoc analysis (parametric) based on the distributions of the log
transformed data. In addition, immune mediator levels were correlated with ALT,
HBV viral load, HIV viral load, and CD4 count at each time-point (Spearman's
test).
Results: Cases had significantly higher HBV viral
loads (p = 0.011) and ALT (p = 0.008) than controls prior to
initiation of HAART. Following initiation of HAART, IP-10 significantly
decreased by weeks 8 and 12 in controls (p <0.05, Kruskal-Wallis),
while in cases there was no significant decrease in IP-10 over time. In cases,
sCD30 and ALT levels peaked at week 8 (p <0.05, ANOVA) while there
was no significant change in controls. Significant positive correlations were
found between ALT and IP-10 (p = 0.005, r = 0.46), sCD30 (p
= 0.006, r = 0.46), MCP-1 (p = 0.032, r = 0.36), and IL-18
(p = 0.009, r = 0.44) at week 8. There was a significant increase
in sCD26 over time in both cases and controls (p <0.05, ANOVA). IL-2,-6,
-8, and -10 and TNF-a and IFN-g and -a were
not detectable in the majority of case and control sera.
Conclusions: IP-10 (an activated T cell and natural
killer cell chemokine) and sCD30 (a T cell activation marker) play an important
role in the pathogenesis of hepatic flare following initiation of HBV-active
HAART in HIV/HBV-co-infected patients. Also implicated are markers of IFN-g induction (IL-18) and activity (MCP-1). These
data support our hypothesis that hepatic flare is a consequence of IRD.
|