HCV treatment in the setting of HIV infection remains a difficult and challenging management issue. The combination of faster rates of hepatic fibrosis combined with lower rates of response to standard of care therapy, when compared to HCV monoinfection, underscores the need to explore variations of management paradigms to reduce disease impact in this population. The SLAM-C study focused on the role of maintenance therapy to modulate fibrotic progression, but also included elements that permitted exploratory examination of the role of weight-based ribavirin, the effect of race on interferon-based therapy response and for responders, the tolerability and effect of prolonged treatment regimens.
The concept of maintenance therapy for HCV arose in the 1990s from several distinct lines of reasoning and evidence. First, analysis of cohort data suggested an altered natural history for patients who had undergone interferon-based therapy but failed to achieve SVR. Reports suggested that both hepatic fibrosis and the risk of hepatocellular carcinoma (HCC) were reduced.10–12
This concept was supported in examination of data from randomized clinical trials. Shiffman et al. selected and randomized subjects with significant (>50% decline from baseline) histologic response to continue to maintenance therapy. Notably, the group reported that a one log drop in HCV viral load was associated with decreased hepatic fibrosis in a treatment group. 13
These observations were congruent with a growing body of data from the HIV field that demonstrated that degree of viral suppression was closely linked to clinical outcomes.14
Finally, the absence of a therapeutic intervention for HCV that was effective and tolerable in the majority of treated patients led to a growing interest in defining the role of interferon maintenance therapy for HCV in non-responder patients.15
This interest culminated in the development of several large multicenter trials for HCV including HALT-C, EPIC and COPILOT. Similarly, the NIH-funded AIDS Clinical Trials Group (ACTG) supported the development and implementation of the SLAM-C maintenance therapy trial in HCV/HIV coinfected patients.
In the last year, it has become clear that interferon maintenance therapy in HCV monoinfection may not represent a viable modality for management of HCV non-responder patients. The HALT-C trial randomized 1050 subjects with advanced fibrosis to receive either half-dose pegylated interferon alfa 2a or no therapy for 3.5 years. There was no difference between groups in any primary outcome (death, HCC, hepatic decompensation or increased fibrosis score by ≥ 2 points in those with bridging fibrosis).16
Though our sample was considerably smaller than that analyzed in HALT-C, the projected rates of progression were significantly greater, permitting the smaller sample size that was utilized in our design. Data regarding long-term outcomes of patients in EPIC-3 and COPILOT remain unpublished, though interim results of COPILOT were presented previously, and these did suggest that certain clinical outcomes may be modulated by maintenance therapy.17
Another randomized trial found a numerical, but not statistically meaningful advantage to prolonged therapy of cirrhotic patients when clinical outcomes were the endpoint of interest.18
Though the sample size was small in SLAM-C compared to HALT-C, the study population and the intervention design were selected to reflect previously reported differences between mono- and coinfection. However, a recent meta-analysis pooled 17 reports of fibrosis progression, and reported a lower rate of fibrotic progression than our study was powered to evaluate (0.11–0.12 Metavir units/year).19
These data were not available at the time SLAM-C was designed however. Extrapolating the metanalysis findings to our study, we would have expected at least a 0.11 worsening in the Metavir score, but in fact there was no worsening of fibrosis in the control arm. The reason for failure to see progression in the control arm is unclear, but several hypotheses may be considered. It is possible that the initial twelve weeks of therapy halted progression long enough to prevent fibrotic progression during the subsequent 72 weeks. Additionally, it is possible that the high rate of HIV viral suppression was associated with decreased rates of fibrotic progression compared to early reports derived from untreated/inadequately treated HIV populations. In our study, 74% of subjects had undetectable HIV RNA. In contrast, Benhamou et al
. reported that while 67.2% of subjects were receiving anti-HIV therapy, few were on HAART (10.9%). The median CD4 count was 305 cell/mm3
In addition, heavy alcohol use was an important cofactor in fibrosis progression in prior literature, but active alcohol abusers were excluded from the SLAM-C study.2
Subsequent analysis of this cohort suggested that both protease inhibitor use and undetectable HIV viremia were associated with decreased fibrotic progression, though viral undetectability fell out of a multivariate model.20
However, Brau et al. has reported slower fibrosis progression in coinfected patients with HAART-associated HIV suppression.21
This is further supported by a recent paper which used transient elastography to suggest that fibrosis rates in HCV/HIV coinfected subjects was now similar that seen in HCV monoinfection.22
Finally, we suspect that there has been a significant evolution in antiretroviral therapy which is reflected in decreased rates of drug-associated hepatoxicity. For example, SLAM-C discouraged use of didanosine during HCV treatment, an agent which was associated with significant hepatotoxicity in prior clinical trials and few patients used this agent. 23
. Subsequently this guidance has been reflected in the product insert which indicates that co-administration of ribavirin and didanosine is not recommended. (from product insert, Genentech dated June 2010).
This study has several limitations. The underlying assumptions regarding rates of fibrotic progression may have overestimated the actual rates that occur in subjects using current and less hepatotoxic antiretroviral regimens. Another limitation of this study is our inability to characterize the effects of initial histologic response (during the first 12 weeks of therapy) on subsequent response during the maintenance phase due to the lack of access to the slides for centralized review of pretreatment biopsies. Since the study did not permit enrollment of patients with very high HIV viral load (>50,000 copies/ml) and very low CD4 (<200 cells/mm3) any conclusions drawn may not be generalized to these subpopulations. Finally, the proportion of woman enrolled did not permit detailed sub-analysis of gender associations with histologic response.
This clinical trial did provide valuable information regarding treatment response to pegylated interferon with weight-based ribavirin. Prior studies of treatment of HCV in the setting of HIV utilized either fixed dosing with 800 mg daily of ribavirin (APRICOT, RIBAVIC) or a unique dose-escalation regimen (600–1000 mg/day) in ACTG 5071.24–26
The dose approved by the U.S. FDA is 800 mg/day for use in coinfected patients. The SLAM-C study utilized weight based dosing of ribavirin in the entry Step 1 of the protocol. However, this study was performed in a highly comparable population to that utilized in ACTG 5071.25
The EVR rate in SLAM-C was 56% which is somewhat higher than a historic comparison of 47% (95% CI: 34–60%) in the PEG/Ribavirin arm of ACTG A5071 (personal communication from RTC). In fact, EVR rate was so high that early planned interim analysis required resizing of the sample population from Step 1, since there was a concern that a lack of non-responders would jeopardize Step 2 enrollment. Similarly, the PRESCO trial also utilized weight-based ribavirin also reported higher response rates than previously noted in coinfection trials which they attributed to the increased ribavirin dosing.27
However, data from both trials are uncontrolled, and should be used with caution when selecting the ribavirin dose in HCV/HIV coinfected patients. A controlled trial of ribavirin dosing in coinfected patients is in progress which may provide additional insights to this issue (NCT00353418).
SLAM-C does provide a unique window into racial disparities in treatment response in coinfected patients. Prior large multicenter trials in HCV/HIV coinfection enrolled relatively few minority subjects, while this study is much more reflective of the U.S. population of HCV/HIV coinfected patients. In SLAM-C more than 50% of subjects were not Caucasian, permitting a unique comparison of the effect of race on the biological response to pegylated interferon/ribavirin on viral decline. EVR was significantly less common in African-Americans and Hispanic subjects, consistent with reports in HCV monoinfected patients.28–30
Some experts had previously suggested that the overall decrement in interferon-based treatment response in those with HIV would blur the race effects, but our data clearly shows that race remains a significant factor in treatment outcome. However, the confounding role of potential response confounders (e.g. BMI, gender) do not permit detailed attribution of the race effect size. The proportional differences between EVR and cEVR that are illustrated in are also worth noting. These data support that concept of decreased rates of viral decline in minority populations. In 2009, several reports describing the relationship between IL-28b polymorphisms, viral clearance and treatment outcome and race were published.31, 32
These studies clearly demonstrate that viral clearance is associated with specific polymorphism near and within the coding domain for the IL-28b gene, and explain a significant portion of observed racial disparities. The SLAM-C cohort has not undergone analysis of the distribution of these polymorphisms, but such studies are planned.
In summary, the SLAM-C trial failed to find evidence of hepatic fibrosis progression among an untreated control group followed for 72 weeks after a short course of pegylated interferon/ribavirin therapy. Further investigation of the mechanism(s) associated with a halt in fibrotic progression are warranted. Finally, race appears to be an important factor in response to interferon-based therapy in HCV/HIV coinfected subjects. The association of this finding with IL28b polymorphisms is under active investigation.