Baseline factors prognostic of sustained virological response in patients with HIV–hepatitis C virus co-infection
Dore, Gregory Ja,b; Torriani, Francesca Jc; Rodriguez-Torres, Maribeld,e; Bräu, Norberf; Sulkowski, Markg; Lamoglia, Ricard Solah; Tural, Cristinai; Clumeck, Nathanj; Nelson, Mark Rk; Mendes-Correa, Maria Cl; Godofsky, Eliot Wm; Dieterich, Douglas Tn; Yetzer, Elleno; Lissen, Eduardop; Cooper, David Aa,b
From the aNational Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney, NSW, Australia
bSt Vincent's Hospital, Darlinghurst, NSW, Australia
cUniversity of California, San Diego, California, USA
dFundacion de Investigacion de Diego, Santurce, Puerto Rico
ePonce School of Medicine, Santurce, Puerto Rico
fBronx VA Medical Center, New York, New York, USA
gJohns Hopkins University, Baltimore, Maryland, USA
hHospital del Mar, Universitat Autonoma de Barcelona, Barcelona, Spain
iHospital Universitari Germans Trias i Pujol, Barcelona, Spain
jSaint-Pierre University Hospital, Brussels, Belgium
kChelsea and Westminster Hospital, London, UK
lHospital Das Clinicas, University of Sao Paulo, Sao Paulo, Brazil
mBach and Godofsky, Bradenton, Florida, USA
nMount Sinai School of Medicine, New York, New York, USA
oRoche, Nutley, New Jersey, USA
pVirgen de Rocio University Hospital, Seville, Spain.
Received 12 January, 2007
Revised 8 March, 2007
Accepted 15 March, 2007
Correspondence to Gregory J. Dore, National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Level 2, 376 Victoria Street, Darlinghurst, NSW 2010, Australia. Tel: +61 2 9385 0900; fax: +61 2 9385 0876; e-mail: firstname.lastname@example.org
Objective: To identify baseline characteristics predictive of a sustained virological response (SVR) in patients with HIV–hepatitis C virus (HCV) co-infection treated with interferon-based therapy.
Design/methods: A stepwise multiple logistic regression analysis was used to explore the prognostic factors associated with SVR [undetectable HCV-RNA (< 50 IU/ml) at the end of untreated follow-up in week 72].
Results: In all patients (n = 853), in addition to the HCV therapy received, the factors most predictive of SVR were baseline HCV-RNA [≤ versus > 400 000 IU/ml; odds ratio (OR) 4.77; 95% confidence interval (CI) 3.15–7.22; P < 0.0001] and HCV genotype (OR 2.87; 95% CI 2.00–4.12; P < 0.0001). HIV treatment (with a protease inhibitor or non-nucleoside reverse transcriptase inhibitor; P = 0.034), race (P = 0.027), and body mass index (P = 0.039) were also weak predictors of HCV treatment response.
Conclusions: In the AIDS PEGASYS Ribavirin International Co-infection Trial (APRICOT), the predictors of SVR among HIV–HCV co-infected patients treated with peginterferon alfa-2a plus ribavirin were similar to those in patients with HCV mono-infection. The HCV genotype and pretreatment HCV-RNA level had the greatest influence on SVR.
For patients co-infected with HIV–hepatitis C virus (HCV), the current standard of care for the treatment of chronic hepatitis C is the combination of pegylated interferon plus ribavirin [1,2]. This recommendation results from studies in HIV–HCV-co-infected patients that have demonstrated superior efficacy of pegylated interferon plus ribavirin compared with conventional combination therapy [3–5]. In the largest of these trials (AIDS PEGASYS Ribavirin International Co-infection Trial; APRICOT), in patients with HIV–HCV co-infection, peginterferon alfa-2a plus ribavirin produced a significantly higher sustained virological response (SVR) rate (40%) than either peginterferon alfa-2a monotherapy (20%) or conventional interferon alfa-2a plus ribavirin (12%; P < 0.001) . These findings led to the approval of peginterferon alfa-2a plus ribavirin for the treatment of chronic hepatitis C patients with HIV–HCV co-infection in both Europe and the United States.
Although HCV mono-infection studies have shown that response rates to interferon-based therapies are influenced primarily by the HCV genotype and baseline HCV-RNA level, and to a lesser extent by factors such as age, sex, race, and the extent of hepatic fibrosis [6–9], little knowledge yet exists regarding the relative influence of baseline prognostic factors on patient outcomes in HIV–HCV-co-infected patients. This information is even more pertinent in HIV–HCV co-infection, because SVR rates are lower than in those with HCV mono-infection, and the selection of appropriate patients for anti-HCV treatment is thus especially important. Therefore, the purpose of our analysis was to identify the baseline characteristics predictive of SVR in HIV–HCV-co-infected patients enrolled in APRICOT .
Patients enrolled in APRICOT had stable HIV infection and co-existing chronic HCV infection. Patients were required to have quantifiable HCV-RNA levels, elevated alanine aminotransferase (ALT) levels and compensated liver disease. Patients were required to be receiving stable antiretroviral therapy (ART) or no ART and to have CD4 cell counts of 200 cells/μl or greater. Patients with CD4 cell counts of 100–199 cells/μl or less were also eligible but required to have baseline HIV-RNA levels less than 5000 copies/ml. Full inclusion and exclusion criteria have been published elsewhere . Patients of all genotypes were randomly assigned to 48 weeks of peginterferon alfa-2a (40 000 Mr; 180 μg once a week) plus ribavirin (800 mg/day), peginterferon alfa-2a (40 000 Mr; 180 μg once a week) alone or interferon alfa-2a (3 MIU three times a week) plus ribavirin (800 mg/day), as described in detail previously . The primary efficacy parameter in APRICOT was SVR, defined as undetectable (< 50 IU/ml) HCV-RNA at the end of follow-up or 24 weeks after the completion of treatment (week 72) using qualitative polymerase chain reaction (Cobas Amplicor HCV test v2.0; Roche Molecular Systems, Inc., Pleasanton, California, USA).
Multiple logistic regression analyses
Stepwise multiple logistic regression analysis was used to explore the prognostic factors associated with SVR in HIV–HCV-co-infected patients. Patients were only eligible for entry into the analysis if data for all the parameters of interest were available (these data were available in 853 of 860 patients treated). Variables entered into the model are those known to influence response in HCV mono-infection or those characterizing HIV disease. In the original study, the patient's race was characterized as Caucasian, black, Asian or other , and here we compared each group against all other groups. Baseline disease or demographic factors considered for entry in the model were age, sex (male versus female), race (black versus non-black; data have shown that the reduction in treatment response rates in black compared with non-black individuals is greater than in Caucasians compared with non-Caucasians ), body mass index (BMI), HCV-RNA level (≤ 400 000 versus > 400 000 IU/ml; data have shown that this is the optimal cut-off ), HCV genotype (1 versus non-1), histological activity index score, histological diagnosis (cirrhosis/bridging fibrosis versus no cirrhosis/bridging fibrosis), ALT quotient (calculated by dividing the actual ALT value measured by the maximum normal ALT value), HIV-1-RNA level (undetectable versus detectable), CD4 cell count, use of ART (yes versus no) and treatment group. For the final stepwise model building process, a variable was added to the model if the adjusted chi-square statistic was significant at the 0.25 level, and a variable was deleted from the model if the Wald chi-square statistic was not significant at the 0.20 level.
Several models were built. The initial model was developed using pooled data from all patients included in the intention-to-treat analysis of APRICOT. Subsequently, separate models were independently developed using data from all patients in each of the three treatment groups. Also, because patients with HCV genotype 1 infection have been shown in previous analyses to have different prognostic factors for SVR than patients infected with non-1 genotypes , separate multiple logistic regression models were developed for patients infected with HCV genotype 1 and non-1 genotypes within each treatment group.
In APRICOT, 860 patients were randomly assigned and received at least one dose of medication . Of these, the required data were available for 853 patients. Baseline characteristics and full efficacy data (including patient withdrawals), are presented elsewhere . Approximately 80% of patients were men, the median age was 40 years, and 79% were Caucasian and 11% black. Sixty-one per cent of patients were infected with HCV genotype 1, and 81% had a high HCV-RNA level (> 400 000 IU/ml). Most patients (84%) were receiving ART at baseline, and the majority of patients (60%) had undetectable HIV-1-RNA levels (< 50 copies/ml). The median CD4 cell count for the three treatment groups was between 476 and 512 cells/μl, with only 6% of patients having a CD4 cell count of less than 200 cells/μl.
Identification of independent factors associated with sustained virological response
When the data were analysed for all patients, a range of baseline factors significantly predicted SVR (Fig. 1a). The variables with the greatest impact on SVR were baseline HCV-RNA level [≤ versus > 400 000 IU/ml; odds ratio (OR) 4.77; 95% confidence interval (CI) 3.15–7.22; P < 0.0001], HCV genotype (non-1 versus 1; OR 2.87; 95% CI 2.00–4.12; P < 0.0001), and treatment group [peginterferon alfa-2a plus ribavirin versus interferon alfa-2a plus ribavirin; OR 6.85; 95% CI 4.25–11.06; P < 0.0001] and peginterferon alfa-2a alone versus interferon alfa-2a plus ribavirin (OR 2.18; 95% CI 1.32–3.59; P = 0.002). Baseline ART, BMI and race were also weakly associated with SVR (Fig. 1a). Baseline cirrhotic status was not a significant predictor of SVR in the analysis of all patients.
As treatment group was a strong predictor of SVR in the above analyses, separate models were constructed for each of the treatment groups. For peginterferon alfa-2a plus ribavirin, HCV genotype (P < 0.0001) and pretreatment HCV-RNA level (P < 0.0001) were the most influential factors in the final model (Fig. 1b). For patients infected with HCV genotype 1 and treated with peginterferon alfa-2a plus ribavirin, pretreatment HCV-RNA level (P < 0.0001), ALT quotient (P = 0.009) and BMI (P = 0.042) were the only predictive factors in the final model (Fig. 1c). A separate analysis of all factors by HCV genotype (1 and non-1) and by treatment group was performed. Significant (P < 0.05) predictors of response by genotype (1 and non-1) for all patients are shown in Table 1.
For genotype 1 patients treated with peginterferon alfa-2a plus ribavirin, SVR rates were 71% in those with viral loads less than 400 000 IU/ml and 20% in those with viral loads greater than 400 000 IU/ml. In genotype non-1 patients treated with peginterferon alfa-2a plus ribavirin, SVR rates were 74% in those with viral loads less than 400 000 IU/ml and 59% in those with viral loads greater than 400 000 IU/ml.
In this large-scale randomized controlled trial of HCV treatment in patients with HIV–HCV co-infection, the independent predictors of outcome were similar to those seen in HCV mono-infection [4–7]. Importantly, the HCV-RNA viral load is several-fold higher in co-infection than in HCV mono-infection , and may partly contribute to the 15–20% lower overall SVR rates in patients co-infected with HIV–HCV than in HCV mono-infection. In addition, the use of a lower dose of ribavirin for patients infected with HCV genotype 1 in APRICOT (all genotypes treated with 800 mg/day) may also be partly responsible for the lower SVR rates compared with those seen in trials with HCV mono-infected patients [6,7].
Consistent with studies in HCV mono-infection, HCV genotype and pretreatment HCV-RNA levels had the greatest impact on treatment outcome [6–8]. Results in patients with HCV genotype 1 infection and genotype non-1 infection were generally consistent with the overall analyses. BMI and the ALT quotient had a significant but minimal influence on SVR, consistent with previous data in patients with HCV mono-infection treated with interferon-based therapies [5,8,12,14,15]. No other hepatitis C-related factors had any significant effect. Similar to mono-infection , race was a weak predictor of SVR (with lower response rates in black compared with non-black individuals), although this variable did not reach significance in the analysis of patients treated with peginterferon alfa-2a plus ribavirin.
Although the use of ART (protease inhibitors or non-nucleoside reverse transcriptase inhibitors) at baseline was weakly associated with SVR in all patients, other HIV infection parameters such as the CD4 cell count and HIV-1-RNA levels were not particularly important predictors of SVR. Interestingly, in the pooled analysis of patients with genotype non-1 infection, an improved HCV treatment response was associated with no ART at baseline, and undetectable HIV-1 RNA was also significantly associated with SVR. Although these associations may appear paradoxical, a possible explanation is that in co-infected patients, either early HIV-1 disease (and no requirement for ART) or well controlled HIV-1-RNA replication may be associated with improved HCV-specific immune responses. Conversely, the CD4 cell count was not significantly associated with HCV treatment outcomes in any of the analyses.
These results have practical implications for the management of HIV–HCV co-infection. Encouragingly, the majority of patients infected with HCV genotype non-1 (independent of pretreatment HCV-RNA levels) and HCV genotype 1 with a low HCV-RNA level (≤400 000 IU/ml) treated with peginterferon alfa-2a plus ribavirin achieved SVR. Therefore, HIV–HCV-co-infected patients in these categories should be considered for HCV treatment with the currently available therapies (in line with the current European and US Consensus Guidelines) [1,2,16].
In contrast, relatively poorer HCV treatment outcomes among HIV–HCV-co-infected patients with HCV genotype 1 and high HCV-RNA levels (> 400 000 IU/ml) underscore the importance of additional factors in treatment decision-making. For example, liver biopsy staging of HCV disease is more crucial when HCV treatment is less likely to be successful. The HCV treatment decision-making process also requires careful consideration of well-documented treatment-related toxicity, in particular neuropsychological morbidity.
Some limitations of this analysis should be considered. As a result of the small number of patients (6%) with low CD4 cell counts, the findings cannot be generalized to advanced HIV disease. Other studies have demonstrated a relationship between the CD4 cell count and HCV treatment outcomes [4,17]. It is therefore possible that if more patients with lower CD4 cell counts had been included, an association between response and CD4 cell count would have been found. Second, the exclusion of patients with normal serum ALT levels further limits the generalizability of the findings to HIV–HCV-co-infected patients with elevated ALT levels only. Third, although many demographic and clinical factors were considered, additional factors such as nadir CD4 cell count, fibrosis progression rate and illicit drug use may be potential predictors of treatment response [3,4,18].
In conclusion, the most important prognostic factors associated with SVR among HIV–HCV-co-infected patients treated with peginterferon alfa-2a plus ribavirin are similar to those observed in patients with HCV mono-infection. HCV genotype and the pretreatment HCV-RNA level have the greatest influence on the likelihood of SVR.
1. Alberti A, Clumeck N, Collins S, Gerlich W, Lundgren J, Palu G, et al
. Short statement of the first European Consensus Conference on the treatment of chronic hepatitis B and C in HIV co-infected patients. J Hepatol 2005; 42:615–624.
2. Tien PC, Wright TL. Management and treatment of hepatitis C virus infection in HIV-infected adults: recommendations from the Veterans Affairs Hepatitis C Resource Center Program and National Hepatitis C Program Office. Am J Gastroenterol 2005; 100:2338–2354.
3. Carrat F, Bani-Sadr F, Pol S, Rosenthal E, Lunel-Fabiani F, Benzekri A, et al
. Pegylated interferon alfa-2b vs standard interferon alfa-2b, plus ribavirin, for chronic hepatitis C in HIV-infected patients: a randomized controlled trial. JAMA 2004; 292:2839–2848.
4. Chung RT, Andersen J, Volberding P, Robbins GK, Liu T, Sherman KE, et al
. Peginterferon Alfa-2a plus ribavirin versus interferon alfa-2a plus ribavirin for chronic hepatitis C in HIV-coinfected persons. N Engl J Med 2004; 351:451–459.
5. Torriani FJ, Rodriguez-Torres M, Rockstroh JK, Lissen E, Gonzalez-Garcia J, Lazzarin A, et al
. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV–HCV co-infection. N Engl J Med 2004; 351:438–450.
6. Fried MW, Shiffman ML, Reddy KR, Smith C, Marinos G, Goncales FL Jr, et al
. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 2002; 347:975–982.
7. Hadziyannis SJ, Sette HJ, Morgan TR, Balan V, Diago M, Marcellin P, et al
. Peginterferon alfa-2a and ribavirin combination therapy in chronic hepatitis C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004; 140:346–355.
8. Zeuzem S. Heterogeneous virologic response rates to interferon-based therapy in patients with chronic hepatitis C: who responds less well? Ann Intern Med 2004; 140:370–381.
9. Ferenci P, Fried MW, Shiffman ML, Smith CI, Marinos G, Goncales FL Jr, et al
. Predicting sustained virological responses in chronic hepatitis C patients treated with peginterferon alfa-2a (40 kD)/ribavirin. J Hepatol 2005; 43:425–433.
10. Muir AJ, Bornstein JD, Killenberg PG. Peginterferon alfa-2b and ribavirin for the treatment of chronic hepatitis C in blacks and non-Hispanic whites. N Engl J Med 2004; 350:2265–2271.
11. Rodriguez-Torres M, Torriani FJ, Lissen E, Brau N, Sulkowski M, Sola Lamoglia R, et al
. Baseline viral load as a predictor of SVR rate with peginterferon alfa-2a (40 kD) (PEGASYS®
) plus ribavirin (COPEGUS®
) in APRICOT. In: 46th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco, CA, USA. 27–30 September 2006.
12. Lee SS, Heathcote EJ, Reddy KR, Zeuzem S, Fried MW, Wright TL, et al
. Prognostic factors and early predictability of sustained viral response with peginterferon alfa-2a (40 kD). J Hepatol 2002; 37:500–506.
13. Cribier B, Rey D, Schmitt C, Lang JM, Kirn A, Stoll-Keller F. High hepatitis C viraemia and impaired antibody response in patients coinfected with HIV. AIDS 1995; 9:1131–1136.
14. Poynard T, McHutchison J, Goodman Z, Ling MH, Albrecht J. Is an “a la carte” combination interferon alfa-2b plus ribavirin regimen possible for the first line treatment in patients with chronic hepatitis C? The ALGOVIRC Project Group. Hepatology 2000; 31:211–218.
15. Zeuzem S, Feinman SV, Rasenack J, Heathcote EJ, Lai MY, Gane E, et al
. Peginterferon alfa-2a in patients with chronic hepatitis C. N Engl J Med 2000; 343:1666–1672.
16. National Institutes of Health. Consensus development conference statement: management of hepatitis C: 2002– June 10–12, 2002. Hepatology
2002; 36 (Suppl. 1)
17. Myers RP, Benhamou Y, Bochet M, Thibault V, Mehri D, Poynard T. Pegylated interferon alpha 2b and ribavirin in HIV/hepatitis C virus co-infected nonresponders and relapsers to IFN-based therapy. AIDS 2004; 18:75–79.
18. Brau N, Salvatore M, Rios-Bedoya CF, Fernandez-Carbia A, Paronetto F, Rodriguez-Orengo JF, et al
. Slower fibrosis progression in HIV/HCV-coinfected patients with successful HIV suppression using antiretroviral therapy. J Hepatol 2006; 44:47–55.
APRICOT; co-infection; hepatitis C virus; sustained virological response
© 2007 Lippincott Williams & Wilkins, Inc.
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Highlight selected keywords in the article text.
Data is temporarily unavailable. Please try again soon.