Antiretroviral Treatment Interruption May Affect HCV Viral Load, HBV Rebound, and Liver Fibrosis Progression in Coinfected Patients

By Liz Highleyman

Over the past few years, evidence has accumulated showing that antiretroviral treatment interruption is a potentially risky strategy, and that ongoing HIV replication is associated with a variety of non-AIDS conditions even in people with relatively well-preserved immune function.

In the large SMART (Strategic Management of Antiretroviral Therapy) study, participants who interrupted treatment had a higher rate of liver disease and other conditions not traditionally considered AIDS-related. Over time, chronic viral hepatitis can lead to advanced liver disease including cirrhosis and hepatocellular carcinoma. Furthermore, research indicates that HIV positive people coinfected with hepatitis B or C virus (HBV or HCV) are more likely to experience rapid liver disease progression.

A set of studies presented at the 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009) this month in Montreal looked at outcomes among HIV-HCV and HIV-HBV coinfected SMART participants.

Briefly, SMART included 5472 mostly treatment-experienced HIV patients with a baseline CD4 count above 350 cells/mm3. Of these, 930 (17.0%) participants had chronic viral hepatitis: 120 (2.2%) HIV-HBV coinfected, 796 (14.5%) HIV-HCV confected, and 14 (0.3%) HIV-HBV-HCV triple-infected.

Participants were randomly assigned to either start or remain on continuous antiretroviral therapy ("viral suppression" arm) or to interrupt treatment while their CD4 count was above 350 cells/mm3 and resume when it fell to 250 cells/mm3 ("drug conservation" arm).

The study was halted in January 2006 after it became apparent that participants in the treatment interruption arm not only had a higher rate of AIDS-related opportunistic illness or death due to any cause, but were also more likely to develop serious cardiovascular, kidney, and liver disease.

HCV Viral Load

Jurgen Rockstroh presented findings from an analysis looking at changes in HCV viral load among HIV-HCV coinfected SMART participants. The investigators randomly selected 100 people with both HIV and HCV, but not HBV, from each treatment arm.

Participants in the treatment interruption and continuous therapy groups were well matched with regard to age (median 48 years) and percentage with HIV RNA < 400 copies/mL (about 55%), but the median baseline CD4 count was higher in the former group (roughly 650 vs 550 cells/mm3). Overall, 86% had HCV genotype 1, and mean baseline HCV viral load was higher in the treatment interruption group.

HCV RNA levels were measured in stored blood samples at baseline and at months 2, 4, and 6 of follow-up. CD4 count and HIV viral load assessment had been done at the same time points. Analysis of covariance was used to compare changes from baseline in HCV RNA according to antiretroviral treatment strategy.

Results

At month 2, HCV RNA decreased by 0.19 log10 IU/mL in the treatment interruption group and increased by 0.05 log10 IU/mL in the continuous therapy group, a difference that barely reached statistical significance (P = 0.048).

At month 4, HCV RNA fell by 0.13 log10 IU/mL in the treatment interruption group and rose by 0.08 log10 IU/mL in the continuous therapy group, but this was not significant (P = 0.091).

At month 6, HCV RNA declined by just 0.04 log10 IU/mL in the treatment interruption group and again rose by 0.08 log10 IU/mL in the continuous therapy group, also not significant (P = 0.691).

However, a significant decline in HCV RNA was noted at all time points in the treatment interruption group when considering only genotype 1 patients.

"In patients interrupting ART, a transient decrease in HCV RNA levels occurred contrasting with a slight but non-significant increase in HCV RNA levels over time in the viral suppression arm," the investigators concluded.

To explain this finding, they suggested, "Renewed HIV replication following cessation of ART may induce changes in cytokine patterns, such as rises in endogenous interferon levels, which then might impair HCV replication."

HBV Rebound

HIV-HBV coinfected individuals face an added level of complexity because several antiretroviral agents -- including tenofovir (Viread, also in the Truvada and Atripla combination pills), lamivudine (3TC, Epivir), and emtricitabine (Emtriva) -- are also active against HBV. When coinfected patients taking these drugs interrupt treatment, they also risk HBV reactivation.

Vincent Soriano presented an analysis of changes in HBV viral load at months 1, 2, 4, 6, 8, 10, and 12 among HIV-HBV coinfected SMART participants. Time to ART re-initiation was assessed using Kaplan Meier analysis.

Results

In the treatment interruption arm, 31%-33% of coinfected participants experienced HBV DNA rebound of at least 1 log10 from baseline during months 1-4.

In the treatment interruption arm, HBV viral load rebound of this magnitude was much less fequent, at 3%-4%.

13 HBV positive patients experienced HBV DNA rebound of > 3 log10, all but 1 of whom were in the treatment interruption arm.

8 of the patients with > 3 log10 increases were using tenofovir-containing regimens at the time of interruption.

Factors independently associated with HBV DNA rebound were:

Assignment to the treatment interruption arm (P < 0.0001);

Undetectable HBV DNA at baseline (P = 0.008);

Use of a tenofovir-containing ART regimen (P = 0.04);

Black race (P= 0.03)

HBV positive patients in the treatment interruption arm restarted ART sooner than either HCV positive participants or those without viral hepatitis (roughly 8, 16, and 18 months, respectively; P < 0.0001).

Likewise, more HBV positive participants had to restart treatment due to adverse outcomes than HCV positive people or those without viral hepatitis (63%, 47%, and 40%; P = 0.0002).

Over the firts 4 months follow-up in the treatment interruption arm, the median slope of CD4 decline was about 70 cells/mm3 per month for HBV positive patients and 50 cells/mm3 for both HCV positive participants and people without viral hepatitis.

However, median CD4 counts at the time of ART re-initiation did not differ significantly across the 3 arms (230-240 cells/mm3), and reasons for restarting therapy were similar.

HBV DNA rebound was correlated with both HIV viral load increase and CD4 cell decline.

Based on these findings, the researchers concluded, "Frequent HBV DNA rebound following ART interruption may be associated with accelerated immune deficiency."

Therefore, they advised, "maintenance of HBV viral control should be recommended in the setting of ART interruption."

Liver Fibrosis

In contrast with HIV, HBV and HCV viral load levels do not have a strong association with disease progression. But individuals who sustained HBV and/or HCV clearance -- either spontaneously or with antiviral treatment -- are less likely to develop advanced liver disease.

Lars Peters presented results from an analysis assessing whether ART interruption was associated with increased levels of hyaluronic acid, a biomarker for liver fibrosis. As background, the researchers noted that in some prior observational studies, fibrosis progression appeared to be accelerated in coinfected people with lower CD4 counts.

This analysis included 110 HIV-HBV coinfected, 553 HIV-HCV coinfected, and 12 HIV-HBV-HCV triple-infected SMART participants with available plasma samples, as well as a control group of HIV monoinfected participants matched for sex, age, randomization date, treatment arm, and alcohol use. Coinfected participants were more likely to be black and had a longer median time since ART initiation, but were less likely to be on treatment at study entry.

Stored plasma samples were tested for hyaluronic acid at baseline and at months 6, 12 (cases only), and 24 (normal range 0 to 75 ng/mL). Follow-up continued for median 33 months. Other biomarkers, including the inflammatory marker interleukin 6 (IL-6) and the coagulation marker D-dimer, were also measured.

Results

At baseline, coinfected participants in the treatment interruption arm had a median CD4 count of about 600 cells/mm3 compared with about 400 cells/mm3 in the continuous therapy arm.

After 12 months, however, patients in both treatment arms had a similar CD4 cell level (about 550 cells/mm3).

Hepatitis coinfected participants had significantly higher hyaluronic acid levels than HIV monoinfected individuals at baseline (about 30 vs 20 ng/mL, respectively).

During follow-up, hyaluronic acid increased significantly in the hepatitis coinfected group compared with the HIV monoinfected group.

At 6 months, about one-quarter of coinfected patients had hyaluronic acid > 75 ng/mL, but levels then reached a plateau and did not rise further.

Coinfected patients with elevated baseline hyaluronic acid also had significantly higher levels of IL-6 and D-dimer compared to people with normal hyaluronic acids levels.

21 coinfected patients developed advanced liver disease, including 17 with cirrhosis.

Within the coinfected group, 31 patients in the treatment interruption arm and 21 in the continuous therapy arm died during follow-up; 4 deaths were liver-related.

Among coinfected participants with a high baseline hyaluronic acid, the risk of death was nearly 4 times greater in the treatment interruption arm compared with the continuous therapy arm (hazard ratio 3.8).

Baseline hyaluronic acid was an independent predictor of non-AIDS-related death -- but not AIDS-related death -- in coinfected SMART participants.

"Interruption of ART led to higher levels of hyaluronic acid in HIV-hepatitis coinfected but not in HIV monoinfected participants," the investigators concluded."

"The higher levels were observed 6 months from study entry but not subsequently, possibly reflecting an acute inflammatory response or liver injury associated with interrupting ART," they hypothesized.

2/27/09

References

J Rockstroh and the INSIGHT SMART Study Group. Does Treatment Interruption within the Strategies for the Management of ART Trial Lead to Changes in HCV Load in HIV/HCV-co-infected Patients? 16th Conference on Retroviruses and Opportunistic Infections (CROI 2009). Montreal, Canada. February 8-11, 2009. Abstract 857.

V Soriano and the SMART/INSIGHT Study Group. ART Reinitiation and HBV Rebound among HIV/HBV-co-infected Patients following ART Interruption in the Strategies for the Management of ART Study. CROI 2009. Abstract 816.

L Peters and the INSIGHT SMART Study Group. Interruption of ART and Changes in Hyaluronic Acid as Marker of Liver Fibrosis Progression in Strategies for Management of ART Viral Hepatitis-co-infected Participants and Matched Controls. CROI 2009. Abstract 858.