Hepatitis C Viral Kinetics During Treatment With Peg IFN-alpha-2b in HIV/HCV Coinfected Patients

Hepatitis C Viral Kinetics During Treatment With Peg IFN-alpha-2b in HIV/HCV Coinfected Patients as a Function of Baseline CD4+ T-Cell Counts

JAIDS Journal of Acquired Immune Deficiency Syndromes:

December 2009

Avidan, Neumann U PhD; Goldstein, Deborah MD; Rozenberg, Lynn MSc; McLaughlin, Mary RN; Ferenci, Peter MD; Masur, Henry MD; Buti, Maria MD; Fauci, Anthony S MD; Polis, Michael A MD, MPH; Kottilil, Shyam MD, PhD

From the *Bar-Ilan University, Ramat-Gan, Israel; †LIR, NIAID, NIH, DHHS, Bethesda, MD; ‡LBM, NIDDK, NIH, DHHS, Bethesda, MD; §Science Applications International Corporations-Frederick, Frederick, MD; ¶University Hospital of Vienna, Austria; ∥CCMD, CC, NIH, DHHS, Bethesda, MD; and #Hospital General Universitario Valle Hebron and Ciber-ehd del Intituto Carlos III, Barcelona, Spain.

"The present study demonstrates a striking relationship between baseline CD4+ T-cell counts and response to therapy with pegylated-interferon alpha-2b and ribavirin. Among HCV genotype 1 HIV coinfected patients, those with baseline CD4+ T-cell count < 450 cells/mm3 did not achieve SVR or ETR (NPV = 100%, PPV = 30%), whereas patients with baseline CD4+ T-cell counts ≥450 cells/mm3 achieved significantly higher ETR (76%) and SVR (29%) rates. Moreover, HCV genotype 1 coinfected patients with baseline CD4+ T-cell counts <450 cells/mm3 had slower first phase kinetics in response to combination therapy than did either persons with higher baseline CD4+ T-cell counts or persons with HCV monoinfection........The same trend was found among patients with HCV genotype 2, however, the sample size was too small to establish statistical significance. Note that 5 out of 6 patients with low CD4 counts were African Americans as compared with 8 of 17 patients with higher CD4 count."

"In conclusion, HIV/HCV coinfected patients with higher CD4+ T-cell counts are more likely to achieve SVR when treated with pegylated-interferon and ribavirin. However, although most of our patients were receiving highly active antiretroviral therapy and had suppressed HIV viremia and it is not clear if our results can be extrapolated to patients with high CD4+ T-cell counts before receiving HAART, it may be appropriate to treat for HCV while CD4+ T-cell counts ≥450 cells/mm3, even before starting ART for HIV. Surely, along with our previous observation that the presence of advanced liver disease in genotype 1 HCV coinfected patients is associated with lower rates of SVR15 these findings suggest that coinfected patients should be treated whether or not there is evidence of pathologic liver disease. Further validation of these concepts requires a larger, randomized study that is stratified by CD4+ T-cell counts and race."

Abstract


Background: HIV/hepatitis C virus (HCV) coinfected patients are known to have lower sustained viral response (SVR) rates than HCV monoinfected patients. However, the role of CD4+ T-cell counts on viral kinetics and outcome is not fully understood.


Methods: HCV RNA kinetics (bDNA v3, lower limit of detection [LD] = 615 IU/mL) was analyzed in 32 HIV/HCV coinfected persons treated with Pegylated-interferon-α2b (1.5 μg/kg weekly) and ribavirin (1-1.2 g daily) for 48 weeks and compared with results obtained from 12 HCV monoinfected patients treated with the same regimen.


Results: Baseline CD4+ T-cell counts ≥450 cells/mm3 were significantly (P < 0.002) associated with SVR in coinfected genotype 1 patients. First phase decline was significantly lower among patients with low as compared with high CD4 counts (P < 0.03) and among coinfected compared with monoinfected patients (P < 0.002). Second phase decline slope showed a similar trend for coinfected patients.


Conclusions: Low baseline CD4+ T-cell count is associated with slower HCV viral kinetics and worse response to treatment among HIV coinfected patients, suggesting HCV treatment response depends on immune status. HCV genotype 1 coinfected patients have slower first phase viral kinetics than HCV monoinfected patients. First phase viral decline (>1.0 log) and second phase viral decline slope (>0.3 log/wk) are excellent predictors of SVR for coinfected patients.

INTRODUCTION


Hepatitis C virus (HCV) infection is highly prevalent among HIV infected individuals, with prevalence directly related to risk categories including injection drug use and hemophilia.1 HIV/HCV coinfected patients have rapid progression of liver fibrosis when compared with individuals infected with HCV alone.2 Current standard of care (pegylated-interferon and ribavirin for 48 weeks) therapy yields only modest cure rates among HIV/HCV coinfected individuals when compared with cure rates seen with HCV monoinfected individuals.3-8 Better understanding of underlying mechanisms responsible for lower response in coinfected patients is needed. Specifically, the impact of baseline CD4+ T-cell count on viral kinetics and response to treatment has not been studied in detail.


The biphasic viral kinetics of decline in HCV monoinfected patients9 treated with combination therapy has been successfully used to predict sustained viral response (SVR).10 Two studies of 10 and 12 HIV/HCV coinfected individuals each found a slower viral decline in coinfected subjects.11,12 However, another study of 12 HIV coinfected patients did not show significant differences in key viral kinetics parameters.13 Additionally, several pharmacodynamic parameters were found to correlate with nonresponse to pegylated interferon alpha 2a (PEG-IFN) among coinfected individuals, although IFN concentrations and pharmacokinetic parameters did not differ between persons with SVR and nonresponse.14


Treatment with interferon and ribavirin carries the risk of serious adverse events such as cytopenias, depression, and irritability. Thus, the identification of markers for SVR, which equate to cure rates, would help optimize anti-HCV therapy and avoid administering toxic and costly therapy to patients with little chance of successful outcome. Presently, lack of early viral response, defined as less than or equal to a 2 log reduction in HCV viral load at week 12, has been associated with more than 95% negative predictive value for SVR and is used to advise patients to discontinue therapy.3,7,8 Development of clinically validated predictors of SVR before week 12 will help to further optimize anti-HCV therapy among HIV/HCV coinfected individuals.


Our study, involving a larger group of HIV/HCV coinfected individuals, evaluated the influence of baseline CD4+ T-cell count on HCV viral kinetics and response to pegylated-interferon and ribavirin treatment and the early predictive value of CD4+ T-cell counts and viral kinetics for SVR. In addition, these results are compared with those obtained with 12 HCV monoinfected patients treated with the same therapeutic schedule.

DISCUSSION


The present study demonstrates a striking relationship between baseline CD4+ T-cell counts and response to therapy with pegylated-interferon alpha-2b and ribavirin. Among HCV genotype 1 HIV coinfected patients, those with baseline CD4+ T-cell count < 450 cells/mm3 did not achieve SVR or ETR (NPV = 100%, PPV = 30%), whereas patients with baseline CD4+ T-cell counts ≥450 cells/mm3 achieved significantly higher ETR (76%) and SVR (29%) rates. Moreover, HCV genotype 1 coinfected patients with baseline CD4+ T-cell counts <450 cells/mm3 had slower first phase kinetics in response to combination therapy than did either persons with higher baseline CD4+ T-cell counts or persons with HCV monoinfection.


The effectiveness of interferon-alpha formulations is probably dependent on baseline immune status and hence there is a clear biological explanation for the relationship between baseline CD4 T-cell counts and first phase HCV decline. Because race is also a major factor that would influence viral kinetics, larger clinical studies will be needed to answer the independent significance of race and CD4 T-cell counts in influencing early viral kinetics and therefore SVR.


Previous studies in smaller number of patients have suggested that baseline CD4+ T-cell count is independently associated with SVR19-22; however, these studies are limited due to small sample sizes, use of nonpegylated interferon, and different dosing regimens and dosing intervals. Several larger studies have not shown low baseline CD4+ T-cell count to be a predictor of SVR probably because these studies either did not analyze baseline CD4+ T-cell count and SVR among isolated genotypes4,7,8; were confounded by a disproportionately high percentage of genotype 2 and 3 patients4,8; did not present data in which CD4+ T-cell count was used as a cut-off in analyses3,7; or used a CD4+ T-cell cut-off of 500 cells/mm3, which may have missed the effect seen at lower CD4+ T-cell levels4,8. The RIBAVIC study shows a trend among all patients in the pegylated-interferon arm toward greater SVR among patients with CD4+ T-cell counts >500 (21% vs. 33%).8 In contrast, the present study focused on intense viral kinetics of all HCV genotypes, which were analyzed by various strata of CD4+ T-cell count. Interestingly, a similar study of 21 HCV genotype 1 coinfected patients did not find similar correlation between CD4 count and SVR or rapid viral kinetics,14 possibly due to the small sample size of both studies. Both race and baseline HCV viral levels have also be shown to be independent determinants for SVR among HCV monoinfected individuals.16 In this study, the distribution of race and baseline HCV viral levels were statistically not different between those with high or low CD4 T-cell counts. However, it is plausible that these factors could play a role in influencing HCV viral load declines. In this exploratory study, we chose to use a cut-off of CD4 T-cell count of 450 cells/mm3 arbitrarily as our intention was to demonstrate the relationship between baseline characteristics and viral kinetics. Larger studies are required to determine whether race, baseline CD4 T-cell counts or HCV viral levels plays a more significant role in virologic response to anti-HCV therapy.


For HIV/HCV genotype 1 coinfected patients, a first phase viral decline of >1.0 log/wk was associated with SVR with a NPV = 100% and PPV = 83%; a second phase decline of >0.3 log/wk was associated with a NPV = 100% and PPV = 38%. Because our study is mainly intended to generate hypothesis, the NPV and PPV will need to be validated in larger treatment studies. Not surprisingly, the first and second phase viral declines were much faster in genotype 2 than in genotype 1 coinfected patients, an observation previously made among HCV monoinfected individuals receiving pegylated-interferon and ribavirin.23 A recent study examined data from 705 patients who were enrolled in the APRICOT study, came to similar conclusions that the ability to achieve SVR was significantly impaired in those who were infected with genotype 1 and had CD4+ T-cell counts less than 350 cells/mm3.24


In conclusion, HIV/HCV coinfected patients with higher CD4+ T-cell counts are more likely to achieve SVR when treated with pegylated-interferon and ribavirin. However, although most of our patients were receiving highly active antiretroviral therapy and had suppressed HIV viremia and it is not clear if our results can be extrapolated to patients with high CD4+ T-cell counts before receiving HAART, it may be appropriate to treat for HCV while CD4+ T-cell counts ≥450 cells/mm3, even before starting ART for HIV. Surely, along with our previous observation that the presence of advanced liver disease in genotype 1 HCV coinfected patients is associated with lower rates of SVR15 these findings suggest that coinfected patients should be treated whether or not there is evidence of pathologic liver disease. Further validation of these concepts requires a larger, randomized study that is stratified by CD4+ T-cell counts and race.

RESULTS

Relationship of SVR to of Baseline CD4+ T-Cell Count and HCV Genotype


The baseline demographics of these patients are shown in Table 1.15 Among the 12 HCV monoinfected patients, 10 were males; mean age 44 years (range 32-62 years). Four were infected by genotype 3, 5 by genotype 1, 2 by genotype 4, and 1 by genotype 2. Patients infected by genotype 1 and 4 were treated for 48 weeks and those infected by genotype 2 and 3 for 24 weeks. All patients completed therapy and a 24 weeks follow-up period. Of the 32 patients who participated in this study, 3 stopped treatment before week 2 due to social reasons and another discontinued due to the onset of psychosis. Of the remaining 28 patients, SVR was achieved in 5/23 (22%) of genotype 1 and 3/5 (60%) of genotype 2 patients. Among HCV genotype 1 patients with baseline CD4+ T-cell counts <450 cells/mm3, no one achieved end-of-treatment response (ETR) or SVR, whereas those with baseline CD4+ T-cell counts ≥ 450 cells/mm3 show significantly (P < 0.002) higher ETR of 76% and SVR of 29% (Fig. 1). The same trend was found among patients with HCV genotype 2, however, the sample size was too small to establish statistical significance. Note that 5 out of 6 patients with low CD4 counts were African Americans as compared with 8 of 17 patients with higher CD4 count. Although this difference was not statistically significant (P > 0.1) we cannot rule out the influence of race on outcome because African Americans have been previously shown to have slower viral declines when compared with that seen with Caucasians in HCV monoinfected patients treated with standard interferon alpha and ribavirin.16 Furthermore, the baseline HCV viral levels between HCV genotype 1-infected subjects with a CD4+ T-cell counts less than 450 cells/mm3 were not significantly different from those with a CD4+ T-cell counts more than 450 cells/mm3 (log 6,22 ± 0.21 vs. log 6.47 ± 0.08, respectively; P > 0.5). Moreover, about half of the patients with high CD4 T-cell counts are also African Americans, and nevertheless they respond better than those who have low CD4 T-cell counts. There are no statistically significant differences in nonresponders and SVR rates between Caucasians and African Americans, but there is nevertheless a significant (P = 0.002, Mann-Whitney nonparametric test) difference in nonresponse rates between patients with low and high CD4 T-cell count. Furthermore, logistic regression with race and CD4 T-cell counts shows that CD4 T-cell counts are better predictors of nonresponse, but due to the small number of patients the results are not statistical significant.


Relationship of Viral Kinetics to HCV Genotype and Baseline CD4+ T-Cell Count


HIV coinfected subjects with HCV genotype 2 had significantly faster viral decline at days 1-84 than did HCV genotype 1 coinfected subjects (P < 0.001) (Fig. 2). HIV coinfected subjects with HCV genotype 1 showed a significantly slower first phase decline (at day 3) than did subjects with HCV genotype 2 (mean 0.75 ± 0.44 log/wk vs. 2.0 ± 0.72 log/wk, P < 0.001) and a significantly slower second phase slope (0.4 ± 0.28 log/wk vs. 0.7 ± 0.18 log/wk, P = 0.03) (Fig. 3). Transient rebounds (mean 0.4 ± 0.3 log) were observed at the end of weeks 1 and 2, when viral load was measured during the week, before the next weekly PEG-IFN injection for both genotypes. Among subjects with HCV genotype 2, HCV viral load suppression <615 IU/mL was achieved in 3/5 (60%) patients by week 4. Only 2 of 23 (9%) of patients coinfected with HIV and HCV genotype 1 reached unquantifiable levels of viremia (<615 IU/mL) at week 4.


Among patients with HCV genotype 1, viral declines at both day 3 and days 28-84 were significantly faster in those with baseline CD4+ T-cell counts ≥450 cells/mm3 than with those who had CD4+ T-cell counts of <450 cells/mm3 (P < 0.03). Indeed, first phase decline was significantly faster for genotype 1 patients with high baseline CD4+ T-cell count (≥450) as compared with those with low baseline CD4 count (mean 0.87 ± 0.43 log vs. 0.42 ± 0.27, P < 0.03). When comparing the first phase decline between coinfected and monoinfected patients (treated with either 1.0 or 1.5 μg/kg weekly) with HCV genotype 1, treated with same combination therapy (Fig. 3A), coinfected patients show significantly slower declines at day 3 (mean, 0.75 ± 0.44 log vs. 1.6 ± 0.9, P < 0.002). We found that race is also a factor that affects first phase decline in HIV/HCV coinfected patients, as African Americans experienced a slower first phase decline when compared with Caucasians (P < 0.01).


Second phase decline was also faster in patients with HCV genotype 1 with high baseline CD4+ T-cell counts (≥450) compared with those with lower baseline CD4+ T-cell counts (mean, 0.44 ± 0.35 log/wk vs. 0.26 ± 0.33 log/wk), however, this parameter showed a larger variation and the difference did not reach statistical significance (Fig. 3C). In fact, there is also no significant difference in the second phase slope between coinfected patients and monoinfected patients treated with same regimen (Fig. 3C). Genotype 1 patients with low CD4+ T-cell counts continued to have a slower decline slope also during weeks 4-12 although it was not statistically significant (P = NS, Fig. 2C).


The magnitude and frequency of the transient rebound at the end of the week is similar for monoinfected patients and coinfected patients with high or low CD4+ T-cell counts (data not shown).


Also for coinfected patients with HCV genotype 2 there was a correlation between baseline CD4+ T-cell counts and first phase decline (R = 0.9, P = 0.03), however, the difference between patients with lower baseline CD4+ T-cell counts (<450 cells/mm3) vs. those with higher baseline CD4+ T-cell counts, or between coinfected patients and monoinfected patients, did not reach statistical significance, possibly due to the small number of patients (Fig. 3B). Second phase decline was similar among all patients with HCV genotype 2 (Fig. 3D).


Predictive Ability of HCV Viral Kinetics and Therapeutic Response


A baseline CD4+ T-cell count of <450 cells/mm3 shows a strong negative predictive value for SVR (no patient with SVR out of 6 patients, Negative predictive value (NPV) = 100%) among HCV genotype 1 coinfected patients, however, the positive predictive value for the CD4 > 450 cells/mm3 criteria is low (5 of 17, PPV = 29%). In contrast, a first phase HCV decline faster than 1 log IU/mL, measurable at day 3 of treatment, shows both a high negative predictive value (0 of 17, NPV = 100%) and a high positive predictive value (5 patients with SVR out of 6, PPV = 83%; Fig. 3A).


As with HCV monoinfected patients, coinfected patients with a slow second phase viral decline slope (less than 0.3 log/wk, or 0.1 ln/d) did not achieve SVR (NPV = 100%), consistent with previous studies.9,10,18 However, the positive predictive value for the second phase slope in coinfected patients was lower (PPV = 38%). The number of coinfected patients with HCV genotype 2 is too small to draw any conclusions regarding the early prediction of SVR.

METHODS

Study Design


This was a pilot, prospective, open-label trial performed at the National Institute of Allergy and Infectious Diseases, National Institutes of Health at Bethesda, MD, from 2001 to 2004. Thirty-two HIV-infected subjects were treated with peg interferon alpha-2b at 1.5 μg/kg subcutaneously every week (Peg-Intron; Schering-Plough, Kenilworth, NJ) and ribavirin daily (Rebetol, Schering-Plough, at 400 mg every Qam and 600 mg every Qpm for <75 kg, 600 mg twice per day for >75 kg) for 48 weeks and followed up for 24 weeks after the termination of treatment. All patients signed informed consent approved by the National Institute of Allergy and Infectious Diseases institutional review board before enrollment in the study. Twelve HCV monoinfected patients were treated with the same therapeutic regimen at the Hospital General Universitario Valle Hebron at Barcelona (Spain) during the same period of time. Ten HCV monoinfected patients were treated with Peg interferon alpha 2b 1 μg/kg with ribavirin at the University of Vienna were also studied.17


Patients were eligible for the study if they were >18 years of age and had CD4+ T-cell counts >100 cells/mm3, absolute neutrophil counts >1000 cells/mm3, HCV viral load >2000 copies/mL, and had histologic evidence of chronic hepatitis C. Of the HIV/HCV coinfected patients, 85% were receiving ART and 68% had HIV-RNA<50 copies/mL at the initiation of therapy for HCV.


Laboratory Studies


Liver chemistry, immunophenotyping and safety laboratory tests were performed before treatment and during each study visit. HCV RNA was performed during all study visits (days 0, 1, 3, 5, 7, 10, week 2, week 3, week 4, week 6, week 8, and then every 4 weeks until week 48. HCV RNA concentration in plasma was measured by the VERSANT HCV RNA 3.0 Assay (Bayer Diagnostics, Puteaux, France). The assay has a quantitation range of 615-7.7 million HCV RNA IU/mL. HCV RNA from HCV monoinfected individuals were measured using commercial assays that had similar quantitation range at the respective institutions.


Statistical Analyses


First phase viral decline was calculated here as the log IU/ml decline at day 3 relative to baseline of HCV RNA. It was in general equal to viral decline at day 1, but we chose to use the day 3 decline because a number of patients had a decline only after day 1. Second phase decline slope (log IU mL-1 wk-1) was calculated by log-linear regression of the viral decline at days 7, 14, 21, and 28, or until the first unquantifiable (<650 IU/mL) viral load. In 2 cases with undetectable viremia (<50 IU/mL) already at day 7 the second phase slope was assumed to be the fastest slope observed in the other patients (-0.89 log/wk).


The nonparametric Mann-Whitney U test was used to assess the significance of difference in distribution of continuous variables between groups of patients. The Fischer exact 2 × 2 test was used to compare frequencies between groups of patients. The Spearman nonparametric test was used to assess the significance of correlation between continuous variables. Significance level was assumed to be P = 0.05.