| |
Role of week 4-rapid virological response (RVR) in prediction of sustained virological response to Peg-IFN plus ribavirin in HCV / HIV co-infected individuals
|
| |
| |
Journal of Viral Hepatitis July 2008
D. O. Shea 1 , H. Tuite 1 , G. Farrell 1 , M. Codd 2 , F. Mulcahy 1 , S. Norris 3 and C. Bergin 1
1 Department of Genitourinary and Infectious Diseases, St James's Hospital, Dublin, Ireland ; 2 UCD School of Medicine and Medical Sciences, Dublin, Ireland ; and 3 Department of Hepatology, St James's Hospital, Dublin, Ireland
"We present strong evidence for the clinical application of monitoring early HCV viral kinetics in the course of hepatitis C treatment in HIV seropositive patients."
ABSTRACT
Summary. The objective of the study was to evaluate the role of rapid virological response (RVR) in predicting sustained virological response (SVR) rates to hepatitis C virus (HCV) therapy. 65 HIV / HCV co-infected patients commenced HCV treatment per protocol. HIV / HCV patients with a mean CD4 count of 502 were treated for 24-48 weeks depending on genotype. Virological response was assessed at weeks 4 (RVR), 12 [early virological response (EVR)], 24, at end of treatment (EOTR) and 24 weeks post-completion of treatment (SVR). Primary end-point was defined as undetectable HCV RNA at 24 weeks post-treatment completion. Fifty-five per cent of co-infected patients were on highly active anti-retroviral therapy. A majority of patient group were male. 60% of HIV / HCV patients achieved SVR (35% genotype 1 / 4; 77% genotype 2 / 3). 24 HIV / HCV patients achieved undetectable HCV levels compared with baseline by week 4. The positive predictive value (PPV) of RVR at week 4 for subsequent SVR in HIV-HCV co-infected patients was 100%; the negative predictive value (NPV) was 57%. Significant variables associated with SVR were: (i) lower median pre-treatment HCV viral load, (ii) genotype 2 / 3 disease and (iii) achievement of RVR. Independent variables associated with RVR were low pre-treatment HCV viral load and genotype 2 / 3 disease. Achievement of RVR, a negative HCV-PCR, at week 4 of treatment is predictive of SVR in this cohort of patients. This may be used to guide optimal treatment duration in patient groups. More significantly, the data serve to highlight the subgroup of patients who, on achieving RVR, should be actively supported to complete HCV treatment with full dose therapy, especially patients co-infected with G2 / 3 disease for whom 6 months' full dose therapy may be sufficient to obtain a SVR.
DISCUSSION
In the era of increased liver associated morbidity and mortality in HIV seropositive patients, continuing efforts are required to address barriers to successfully treating HCV infection. Effective HCV treatment is available and proven to be successful in co-infected patients, although significant barriers still exist in treating such patients [22]. Numerous variables contribute to the fact that relatively few patients successfully engage in care and complete HCV treatment. These factors include social demographics, ongoing substance use, concerns for side-effects, and co-morbid psychiatric and medical conditions. The perceived duration of therapy by patient and provider represents a significant hurdle to be overcome when counselling a patient regarding treatment commencement.
Therefore, a need exists to optimize therapy to maximize successful outcomes.
Utilizing HCV viral kinetics may play a pivotal role in successfully treating patients. In HCV mono-infected patients, the use of viral response at week 12 (EVR) has long been accepted as a juncture point at which decision to continue or discontinue treatment can be taken [14]. Numerous groups have concluded that treatment of HCV mono-infected patients can be confidently discontinued at week 12, if EVR is not reached. Such a policy has obvious benefits in terms of costs and reduction in side effects. Such an early discontinuation rule could not be automatically adopted when treating HCV / HIV co-infected patients. Moreover, the perceived slower virological response and the increased relapse rates observed in original studies of co-infected patients suggested the need for a longer duration of HCV therapy for HIV / HCV patients. In our study, failure to achieve EVR had 100% NPV for SVR. This is in keeping with similar findings by the RIBAVIC and APRICOT study groups when analysing week 12 viral response [9,10].
More recently, investigators have assessed the utility of measuring the HCV viral response at earlier time-points in the treatment of mono-infected patients. Ferenci et al. reported that SVR rates increased with the rapidity of HCV RNA suppression. Notably, the highest SVR rates were observed in patients whose HCV RNA was undetectable at week 4 (RVR) [16]. Similarly, Zeuzem et al. treated HCV genotype 1 mono-infected patients with a low pre-treatment HCV viral load with combination therapy for 24 weeks. In the subgroup that was HCV RNA negative at week 4, similar SVR rates were observed when matched with comparable historical treatment groups treated for 48 weeks [18].
A number of groups have recently investigated the possibility of shortening the duration of combination therapy in HCV mono-infected patients with genotype 2 / 3 disease based on week 4 viral response [15-21]. Mangia et al., and Dalgard et al. shortened the duration of combination therapy in patients with undetectable HCV RNA at week 4, to 12 and 14 weeks respectively. Both reported successful shortening of treatment duration without compromising SVR rates [20,21]. However, work published by the Accelerate Investigators comparing 16 weeks with standard 24 week therapy in a randomized noninferiority trial failed to support this approach further. 1469 patients were randomized. The observed SVR rate was significantly lower in the group treated for 16 weeks (62%vs 70%). The shorter duration of treatment arm was also associated with an increased relapse rate (31%vs 18%). In the sub-analysis of the group that achieved RVR, again there was significantly lower SVR rates observed in the shorter treatment arm (79%vs 85%; P = 0.02) [17].
Published data on the utility of week 4 HCV viral kinetics in HCV / HIV co-infected patients are growing. Crespo et al. recently published an analysis of early viral kinetics in a group of 35 co-infected patients with genotype 3 disease [23]. Patients received combination therapy (peg-interferon or standard interferon with ribavirin) for a duration of 24 weeks only. The authors observed a high SVR rate in those patients with a RVR (HCV RNA <600 iU / mL) independent of type of interferon used. Conversely, failure to achieve rapid viral decline in this genotype 3 co-infected patient group, was associated with significantly lower SVR rates. A study of a subgroup of patients previously enrolled in the RIBAVIC study identified similar findings regarding predictive value of week 4 HCV viral response [24]. 323 HIV / HCV co-infected patients received either peg-interferon or standard interferon with ribavirin for 48 weeks. The authors reviewed viral kinetic response at differing time-points in the treatment schedule. Undetectable HCV RNA at week 4 was associated with a 97% PPV for SVR, regardless of which type of interferon was administered.
A more recent publication by Nunez et al. evaluated factors predictive of relapse following initial response to anti-HCV treatment. They concluded that failure to achieve RVR and a pre-treatment HCV viral load of greater than 500 000 IU / mL were the best independent predictors of HCV relapse. In this study, extended duration of treatment did not significantly impact on the incidence of relapse [25].
From our data, achievement of RVR predicts SVR in co-infected patients (P < 0.0001; PPV 100%). The findings of our analysis are consistent with previous authors' work as outlined, but interestingly, we report data for all HCV genotypes, and in particular genotype 2 / 3 co-infected patients treated for 24 weeks only. Analysis of independent variables predictive of RVR identified low pre-treatment HCV viral load and genotype 2 / 3 disease to be statistically significant associations.
The patients attended a designated clinic on a regular basis, and received constant support from a multi-disciplinary team. Haematological support in the form of growth factors was also commenced, if necessary. This approach facilitated patient adherence to therapy and the maintenance of full dose therapy.
We present strong evidence for the clinical application of monitoring early HCV viral kinetics in the course of hepatitis C treatment in HIV seropositive patients. The utility lies in the ability to predict successful primary outcome with treatment and in allowing the individualization of patient treatment duration. Clearly, identification of patients with a RVR singles out those who should be closely monitored and supported throughout treatment with full dose therapies. In addition, adopting strategies of initial shorter periods of treatment 'trials' will prove appealing, with adjustment of overall duration pending viral response. Such an individualized approach may ultimately embrace greater numbers of patients who, to date, have been reluctant to engage in care.
This is a retrospective review of two prospective studies. Viral kinetic monitoring was conducted from the outset; however, the studies were not specifically designed to use the week 4 response as a guide for treatment. Other limitations include the small sample size studied and the recruitment of patients sequentially from the designated outpatient clinic, which identifies a possible inherent selection bias.
Despite these limitations, we report on the PPV of RVR for primary outcome in the treatment of HCV in co-infected individuals and propose a role for HCV viral kinetics as an independent marker for individualization of duration of HCV therapy in co-infected patients. In the application of findings such as our own, care must be taken not to discontinue therapy prematurely in individuals who may achieve SVR with longer durations of therapy, or who may gain secondary benefit from longer duration of HCV treatment.
Historically, international guidelines advised 48 weeks HCV therapy for all co-infected patients. Our findings lend further support to the recently updated consensus guidelines, which now recommend shortening duration of treatment in a sub-group of those co-infected with genotype 2 / 3 disease [26]. Clearly, use of RVR in the setting of maintenance of full dose HCV therapy will identify a cohort for whom 24 weeks' treatment will successfully achieve primary outcome. This finding further strengthens this groups' previously published recommendation on the need to individualize duration of HCV therapy for HIV / HCV co-infected patients.
RESULTS
Seventy-three co-infected patients were identified. 65 patients met study criteria and were included for analysis. Six of 65 patients discontinued treatment early because of side-effects or did not attend for follow up. These patients were included in the ITT analysis (Table 1). On Treatment analysis was conducted on 59 patients with complete follow up data available (Table 2). The baseline demographics of the patients are outlined. 55% of the co-infected patients were on HAART. A majority of the patient group were male and injecting drug users.
In an ITT analysis, 60% (39 of 65) of patients achieved SVR. The SVR rate was significantly higher in genotype 2 / 3 patients than genotype 1 / 4 patients; 77%vs 35% (Fig. 1a). Univariate analysis of this group identified three significant predictors of SVR; lower median pre-treatment HCV viral load (P < 0.001), genotype 2 / 3 disease (P < 0.001), and achievement of RVR (P < 0.0001). In multivariate analysis, achievement of RVR remained a statistically significant predictor of SVR, with lower median pre-treatment HCV viral load approaching significance (Table 1).
Using OT analysis, the overall SVR rate was 66%, with a 39% response rate in genotype 1 / 4 disease and 83% response rate in genotype 2 / 3 disease (Fig. 1b). Again in a univariate model, significant predictors of SVR were lower median pre-treatment HCV viral load (P < 0.001), genotype 2 / 3 disease (P < 0.001), and achievement of RVR (P < 0.0001). On multivariate analysis, no independent variable was found to be significantly associated with SVR; however, lower median pre-treatment HCV viral load did again approach statistical significance (P = 0.07) (Table 2).
In the OT analysis group, 24 of 59 patients were HCV RNA negative by PCR analysis at week 4 (RVR). All these 24 patients achieved SVR. The positive predictive value (PPV) and the negative predictive value (NPV) of SVR response based on RVR at week 4 were calculated. Achieving undetectable HCV RNA by week 4 had a PPV of SVR of 100%, with a NPV of only 57%. Both genotype 2 / 3 disease and lower median pre-treatment HCV viral load were strongly associated with achievement of RVR (p values of <0.001 and <0.01 respectively on univariate analysis). Genotype 2 / 3 disease maintained a significant association with RVR on multivariate analysis (P < 0.01) (Table 3).
All patients who achieved SVR in this study were HCV RNA negative by PCR analysis at week 12 (i.e. achieved EVR). The PPV and NPV for SVR for those patients who achieved an EVR were 95% and 100% respectively.
BACKGROUND
With common routes of transmission, chronic hepatitis C infection is frequently encountered in the setting of HIV infection. Overall prevalence of hepatitis C virus (HCV) and HIV co-infection rates approximate 30-40%; however, in intravenous drug users, much higher rates of co-infection prevail [1,2]. End-stage liver disease in the HCV / HIV co-infected population remains a leading cause of morbidity and mortality [3]. Since the advent of highly active anti-retroviral therapy (HAART), energies have focussed on improving treatment outcomes for chronic HCV in HIV patients.
There is extensive evidence in support of both the efficacy and safety of combination pegylated interferon and ribavirin, for both HCV mono and HIV / HCV co-infected patients [4-11]. Quoted sustained viral response (SVR) rates range from 27% to 44% in co-infected studies [9-11]. Until very recently, the standard recommended therapy consisted of 48 weeks of combination HCV therapy for all HIV patients irrespective of genotype [6-8].
There is growing support now for individualizing the treatment duration of chronic HCV in co-infected patients similar to that employed in mono-infected groups, namely a shorter duration of HCV therapy in specific subgroups such as genotype 2 disease or genotype 3 with low pre-treatment viral load. With respect to shortening duration of HCV therapy, this group previously reported successful outcomes when employing 24 weeks of HCV therapy in HIV / HCV co-infected patients with genotype 2 / 3 disease with use of weight-based ribavirin and maintenance of full doses of HCV therapies [12,13].
The role for monitoring early viral kinetic response has been reported for the management of HCV mono-infected individuals. An early stop rule exists for those failing to achieve undetectable HCV RNA levels by week 12 of treatment [14]. More recent efforts in this patient cohort have focussed on use of week 4 viral response to predict ultimate primary outcome and thus further enable individualization of treatment duration [15-21].
The application of the guidelines used for HCV mono-infected patients to co-infected patients is hampered by evidence suggesting slower HCV kinetic response in HCV / HIV patients. The Apricot study group and Carrat et al. concluded that SVR was highly unlikely in co-infected patients failing to achieve early virological response (EVR) [9,10]. There is considerable interest in validating the utility of HCV viral kinetics at earlier time-points to predict primary outcome in co-infected patients, thus identifying those for whom duration of HCV therapy may be shortened.
In this study, we evaluated the role of rapid virological response (RVR) in predicting SVR in HCV / HIV co-infected patients treated with combination pegylated interferon and ribavirin.
METHODS
We performed a retrospective review of two prospective, open-label, single centre studies in HCV / HIV co-infected patients who attended a speciality outpatient clinic in Dublin, Ireland. These studies were designed prospectively with a number of key objectives:
* 1) Patients with genotype 2 / 3 disease were to be treated for 24 weeks only
* 2) The maintenance of full dose treatment with haematological support and growth factors
* 3) Use of weight-based ribavirin
* 4) Monitoring of HCV viral kinetic response at various time-points.
Patients who commenced combination therapy from June 2001 to January 2006 were included in this analysis.
HCV / HIV co-infected patients received pegylated-interferon at dosing of 1.5 _g / kg / week (Peg-Intron; Schering-Plough, Kenilworth, NJ, USA) or 180 _g / week (Pegasus; Roche, Basel, Switzerland) subcutaneously, and weight-based ribavirin (1000-1200 mg / day). Patients with genotype 2 / 3 disease received treatment for 24 weeks only, while those with genotype 1 / 4 disease received treatment for 48 weeks, if they were HCV RNA negative at 24 weeks of therapy.
RVR is defined as undetectable HCV RNA level by PCR at week 4 of treatment. Sustained virological response is defined as undetectable HCV RNA 24 weeks post-completion of treatment.
Standard inclusion and exclusion criteria were used. Patients without a recorded quantitative HCV RNA level by PCR at weeks 0 or 4 were excluded from this analysis. For this analysis, the end-point was achievement of SVR. Virological response was assessed at weeks 4 (RVR), 12 (EVR), 24, end of treatment (EOTR) and 24 weeks post-completion of treatment (SVR).
HCV infection was confirmed using HCV antibodies (Abbott IMX V3.0; Abbott Laboratories Ltd, Maidenhead, UK) and RIBA (Chiron V3.0; Chiron Corp., Emeryville, CA, USA). The qualitative HCV RNA measurement was conducted using the Roche HCV PCR assay (Cobas), with a limit of detection of 100 copies / mL. Quantification of HCV RNA was performed using the Versant HCV RNA 3.0 assay (bDNA) from Bayer whose limit of detection is 3200 iU / mL.
Haematological support was instituted, if necessary, to maintain full dose HCV therapy. G-CSF was administered, if absolute neutrophil count fell below 0.5 cells / mL; and erythropoietin was utilized if haemoglobin level fell below 9.5 g / dL.
Analysis was performed on both an intention to treat (ITT), and an on treatment (OT) basis. ITT analysis included all patients who received one dose of HCV treatment, while OT included those patients who completed follow-up and those who discontinued treatment because of nonresponse.
The following variables were recorded on all patients:
* 1) Age: age at time of initiation of treatment
* 2) Gender
* 3) Baseline aminotransferase levels presented as median and range
* 4) Pre-treatment hepatitis C viral load presented as median and range
* 5) CD4 count defined as <250 or ≥250
* 6) Prescription of concomitant anti-retroviral therapy
* 7) Hepatitis C genotype, 1-4
* 8) Type of peg-interferon administered; i.e. Peg-IFN alpha 2a or 2b
* 9) Achievement of RVR.
Data were entered onto an Excel spreadsheet and analysed using SAS 9.1 (SAS Institute Inc., Cary, NC, USA). Statistical methods used for univariate analysis included Student t-test, Wilcoxon Mann-Whitney U-test, chi-squared test or Fisher's Exact Test as appropriate. Multivariable analysis was carried out using multiple logistic regression including in the model only those variables significant or close to significance on univariate analysis and any variable considered particularly relevant to the outcome of interest (SVR or RVR as applicable). Statistical significance was considered at P < 0.05 unless otherwise stated.
|
|
| |
| |
|
|
|
