This uncommissioned review article was subject to full peer-review.
Review article: the role of rapid virological response in determining treatment duration for chronic hepatitis C
Version of Record online: 17 MAR 2010
© 2010 Blackwell Publishing Ltd
Alimentary Pharmacology & Therapeutics
Volume 31, Issue 12, pages 1251–1267, June 2010
How to Cite
POORDAD, F. F. (2010), Review article: the role of rapid virological response in determining treatment duration for chronic hepatitis C. Alimentary Pharmacology & Therapeutics, 31: 1251–1267. doi: 10.1111/j.1365-2036.2010.04300.x
- Issue online: 20 MAY 2010
- Version of Record online: 17 MAR 2010
- Publication data Submitted 11 December 2009 First decision 24 December 2009 Resubmitted 22 February 2010 Accepted 10 March 2010 Epub Accepted Article 17 March 2010
Aliment Pharmacol Ther 31, 1251–1267
Background For patients with chronic hepatitis C, attaining rapid virological response (RVR) is highly predictive of attaining SVR.
Aim To consider the predictive value of RVR in terms of SVR and relapse.
Methods Data were collected from published clinical trials to define the predictive value of RVR for SVR and evaluate the proposed continuum linking RVR to relapse.
Results These data support a 24-week regimen among genotype (G)1 patients who attain RVR with positive predictive values (PPVs) of 77.8% and 85.7% in patients with G1 infection treated for 24 and 48 weeks. Conversely, failure to attain RVR among G1 patients should not be viewed as a criterion for extending treatment duration beyond 48 weeks: negative predictive values (NPVs) were 60.9% and 52.7% in G1 patients without RVR treated for 48 and 72 weeks. Among G2/3 patients, RVR has a high PPV; however, the NPV varied with treatment duration indicating that a 24-week treatment regimen is warranted in G2/3 patients who fail to attain RVR.
Conclusions The present analysis confirms RVR as a strong predictor of SVR that can be used to tailor treatment duration, but which also should be appreciated in the context of treatment duration and regimen.
Sustained virological response (SVR), the goal of therapy for patients with chronic hepatitis C, is attained by 54–56% of patients who receive pegylated interferon (PEG-IFN) alfa plus ribavirin.1, 2 Unfortunately, patients with chronic hepatitis C do not represent a homogeneous population with uniform responses to therapy. Therefore, developing an optimized therapeutic regimen suited to all patients has proved difficult. To optimize treatment outcomes and health care resources, research efforts have focused on an individualized approach to treatment in which therapeutic regimens are developed to suit the particular host and viral characteristics of each patient.3
The essence of individualized treatment is to apply the minimum possible therapeutic burden on each patient without compromising the likelihood of attaining SVR. Thus, the individualized approach allows shorter treatment durations and lower dose regimens in patients who respond quickly to therapy, longer treatment durations and higher dose regimens in patients who respond slowly and early discontinuation of treatment in patients who do not respond at all. In principle, this approach is straightforward. However, identifying patients who fall into each category with a high degree of accuracy is important for its success.
Host and viral factors such as hepatitis C virus (HCV) genotype, baseline viral load, fibrosis, body weight and age influence response to therapy and can help predict treatment outcomes.4 Once treatment is initiated, on-treatment markers of response can further aid in predicting treatment outcomes. These markers include rapid virological response (RVR, undetectable HCV RNA at week 4 of therapy), partial early virological response (pEVR, ≤2 log10 decline in HCV RNA from baseline at week 12 of therapy) and complete EVR (cEVR, undetectable HCV RNA at week 12 of therapy).5, 6
Among patients who attain an end-of-treatment (EOT) response, two different clinical outcomes – SVR and relapse – are possible. Factors influencing attainment of SVR or relapse are becoming better defined. For example, the on-treatment period of undetectable HCV RNA is one of the most important factors influencing the ratio of SVR to relapse.7, 8 In turn, two elements of the treatment regimen are critical: the time during treatment at which the patient first attains undetectable HCV RNA and the overall duration of treatment. For example, the standard duration of therapy for a patient infected with HCV genotype 1 (G1) is 48 weeks.9 Thus, G1 patients who attain RVR will experience 44 weeks of undetectable HCV RNA while on treatment; the ratio of SVR to relapse in these patients favours SVR.8 In contrast, G1 patients who do not attain undetectable HCV RNA until week 24 of therapy will experience only 24 weeks of undetectable HCV RNA while on treatment; the ratio of SVR to relapse in this case shifts dramatically to favour relapse.8
Rapid virological response is the earliest on-treatment predictor of treatment response investigated to date. Theoretically, a continuum should connect RVR, treatment duration and the ratio of SVR to relapse. The objective of this review was to consider the predictive value of RVR in terms of SVR and relapse.
Study selection criteria
Search strategy. Bibliographic searches were performed in MEDLINE and through conference proceedings for clinical trials of PEG-IFN alfa plus ribavirin in the treatment of patients with chronic hepatitis C. Reference lists of the clinical trials identified during electronic searching were also hand searched to identify additional relevant trials for inclusion.
Selection of studies. Studies were eligible for inclusion in this analysis if PEG-IFN alfa-2a (Pegasys; Hoffman-La Roche, Nutley, NJ, USA) or PEG-IFN alfa-2b (PegIntron; Schering-Plough Corp., Kenilworth, NJ, USA) plus ribavirin was used to treat patients with chronic hepatitis C. The presence of hepatitis C was defined by detectable HCV RNA [as indicated by polymerase chain reaction (PCR)–based assay] and most studies also required histological evidence of chronic hepatitis on liver biopsy before enrolment.
For inclusion, studies had to be conducted in treatment-naive patients with genotype 1, 2 or 3 infections, include an evaluation of HCV RNA levels at week 4 of therapy, include outcomes of treatment (EOT response, SVR, relapse rates) according to HCV RNA status at week 4 and designate SVR as the primary study outcome. Where stated, SVR was defined using a highly sensitive HCV RNA assay with a lower limit of detection (LLD) ≤50 IU/mL. For consideration in the final analysis, studies were required to provide data specific to genotype (studies providing data from populations of mixed genotypes were not considered in detail). Studies of patients treated with standard IFN alfa or co-infected with HCV and HIV and studies that were not written in English were excluded.
Assessment of study quality and data extraction. All studies were prospective clinical trials or retrospective analyses of primary clinical trial data if the study was conducted before the importance of week 4 HCV RNA levels was recognized. For each study, the following variables were extracted: treatment regimen, treatment duration, HCV genotype, number of patients treated, and HCV RNA status at week 4 of therapy. The sensitivity of assays used to define RVR ranged from 50 to 600 IU/mL. The following outcomes data were collected: rates of EOT response (undetectable HCV RNA at EOT), SVR, and relapse (undetectable HCV RNA at EOT and subsequent re-emergence of HCV RNA during follow-up). If relapse rates were not provided, but EOT response and SVR rates were available, relapse was calculated as follows:
Of note, actual relapse rates might differ because of patients lost during follow-up. The predictive value of RVR for SVR was calculated using positive and negative predictive values (PPV and NPV respectively). PPV was calculated as the percentage of patients who attained RVR and SVR. NPV was calculated as the percentage of non-RVR patients who did not attain SVR. If predictive values were not reported, but RVR and SVR rates were available, PPV and NPV were calculated as follows:
Outcome measures. The primary outcome measure in this analysis was the predictive value of RVR for SVR, which was calculated according to genotype, treatment duration and ribavirin dosing (for G2 and G3 patients). A secondary analysis aimed to establish a disease continuum between RVR (an early marker of virological response) and SVR or relapse (markers of treatment outcome).
Predictive value of RVR
Twenty-two studies were identified with PPV and NPV data; four10–13 contained information on only mixed genotype populations and were not included in this analysis. In addition, one retrospective study14 was not included because the data were derived from a patient group also used in a separate analysis (Table 1).3, 10–13, 15–30
Genotype 1 patients. Among G1 patients, the primary therapeutic value of RVR lies in defining patients who are suitable for a reduced, 24-week treatment duration rather than the recommended 48-week duration.9 Alternatively, some research has focused on the benefit of extended 72-week treatment duration among G1 patients who do not attain RVR. In all, eight studies provided data regarding the outcome of therapy among G1 patients who attained RVR.3, 15–21 Three studies provided data on the therapeutic outcomes of G1 patients who attained RVR and were treated for 24 weeks3, 19, 21 and two studies provided data regarding 72-week treatment duration for patients without RVR.17, 18
24 vs. 48 weeks of therapy among G1 patients who attained RVR. Overall, 5270 G1 patients received PEG-IFN alfa plus ribavirin, were evaluated for RVR and received 24 or 48 weeks of therapy (Figure 1). Of these, 891 (16.9%) had undetectable HCV RNA at week 4 of therapy; 288 were treated for 24 weeks, and 603 were treated for 48 weeks. Of the 288 patients treated for 24 weeks, 224 attained SVR, yielding a PPV of RVR for SVR of 77.8%. Similarly, of the 603 patients who attained RVR and were treated for 48 weeks, 517 attained SVR yielding a PPV of RVR for SVR of 85.7%. Overall, these observations tend to support the use of a 24-week regimen among G1 patients who attain RVR; however, there appears to be a marginal benefit associated with the 48-week treatment duration (Δ = 8%). Our data do not permit comment on the clinical significance of this difference; however, we can speculate that G1 patients with RVR who are adherent to their treatment and tolerate therapy may be treated for 24 weeks, whereas patients who are less adherent may be treated for 48 weeks to augment the opportunity for SVR.
Baseline viral load has been shown to influence whether a patient attains RVR; however, studies included in the present analysis did not take baseline viral load into account when stratifying patients to 24 or 48 weeks of treatment. For example, Jensen et al.16 showed that 9.2% of patients with baseline viral loads >600 000 IU/mL attain RVR; however, the effectiveness of a 24-week treatment duration in these patients has not been prospectively studied. Zeuzem et al.,31 in their investigation of a reduced treatment duration for G1 patients, found that SVR rates were comparable in G1 patients with baseline viral load ≤600 000 IU/mL who attained RVR when treated for 24 or 48 weeks (89% vs. 85%). Overall, the results of the present analysis indicate that therapeutic outcomes are broadly equivalent in G1 patients who attain RVR and are treated for 24 or 48 weeks. Whether a minority of G1 patients with high baseline viral loads who attain RVR may attain higher rates of SVR if treated for 48 vs. 24 weeks cannot be determined because data on baseline viral load are lacking in the evaluated studies.
48 vs. 72 weeks of therapy among G1 patients who do not attain RVR. Overall, 1550 G1 patients were evaluated for RVR and treated for 48 or 72 weeks (Figure 1). Of these, 1204 (77.7%) did not attain RVR, 872 were treated for 48 weeks and 332 were treated for 72 weeks. Among patients treated for 48 weeks, 531 of 872 did not attain SVR, yielding an NPV of RVR for SVR of 60.9%. However, of the 332 patients without RVR who were treated for 72 weeks, 175 did not attain SVR, yielding an NPV of RVR for SVR of 52.7%. Thus, among patients who do not attain RVR, 60.9% did not attain SVR when treated for 48 weeks and 52.7% did not attain SVR when treated for 72 weeks. These data indicate that 8.2% of G1 patients who did not attain RVR benefited from extended treatment. This minimal advantage, however, does not warrant treating all G1 patients who do not attain RVR for 72 weeks because many patients would be over-treated, unnecessarily exposed to adverse events and unnecessarily incur health care expenditures.
Genotype 2 and 3 patients. Patients infected with HCV G2 or G3 are considered the easiest to treat, and current recommendations suggest treatment with PEG-IFN alfa and fixed-dose (800 mg/day) ribavirin for 24 weeks. For these patients, RVR is an important predictor of treatment outcome. Several studies have assessed whether G2/3 patients who attain RVR can be effectively treated for 12–16 weeks rather than 24 weeks.23–26 A crucial element of these studies is the use of weight-based ribavirin, in contrast to the recommended fixed dose of 800 mg/day.9 Studies show that reducing treatment duration to 12–16 weeks among G2/3 patients receiving PEG-IFN alfa and a fixed 800-mg/day ribavirin dose resulted in a significant decline in SVR rates, even among those who attained RVR.25, 30 These studies emphasize the importance of both RVR status and ribavirin dose when considering reduced treatment duration for G2/3 patients.
In an initial analysis that did not account for varying ribavirin dosing regimens, NPVs and PPVs of RVR for SVR were defined according to treatment duration (24 weeks vs. 12–16 weeks) (Figure 2). Overall, RVR had a high PPV for SVR regardless of treatment duration. However, PPVs for a 24-week regimen were slightly higher (G2, 89.3%; G3, 86.4%) than a shortened treatment duration (G2, 83.2%; G3, 82.4%), suggesting that overall, the PPV of RVR for SVR remains fairly consistent regardless of treatment duration. In contrast, the NPV was clearly related to treatment duration. Among G2/3 patients who did not attain RVR, approximately half did not attain SVR when treated for 24 weeks (NPV was 44.3% for G2 and 58.2% for G3), and more than 70% did not attain SVR if treated for 16 weeks (NPV was 71.8% for G2 and 73.5% for G3). Therefore, in G2/3 patients with detectable HCV RNA at week 4, a minimum 24-week treatment regimen is warranted.
When treatment duration and ribavirin dosing regimen were considered as variables, the number of analysis groups increased to four, making a marked reduction in the number of patients available for analysis in each group a limitation of this approach. In particular, calculation of an NPV among G3 patients receiving weight-based ribavirin for 24 weeks was not possible. Several recurring themes within this analysis, however, deserve brief comment despite this limitation. The frequency of RVR was similar among G2/3 patients receiving weight-based ribavirin or a fixed 800-mg/day dose. Considering G2 and G3 patients collectively, our calculations suggest an estimated 68.8% of patients who received a weight-based ribavirin regimen and 65.5% of those who received a fixed 800-mg/day regimen attained RVR.
Weight-based ribavirin is important in optimizing treatment outcomes in G2 and G3 patients. The PPV of RVR for SVR is highest among patients receiving weight-based ribavirin for 24 weeks and lowest among those receiving fixed 800-mg/day ribavirin for 14–16 weeks (Figure 3). Observations regarding NPV are consistent with the relationship between treatment duration and ribavirin regimen. NPV is highest among patients treated for 16 weeks with fixed 800-mg/day ribavirin dosing; absence of RVR becomes a moderately reliable indicator of treatment failure in this cohort. NPV is lowest among patients treated with weight-based ribavirin for 24 weeks, indicating that many patients who do not attain RVR may eventually attain SVR using this treatment strategy.
Relationship between RVR and relapse
Relapse risk is closely related to the period of continuously undetectable HCV RNA while on treatment.7, 8 Therefore, patients who clear HCV early during therapy should theoretically have a lower likelihood of relapse than those who clear the virus later during treatment.7, 8 RVR is the earliest clinical milestone during therapy and represents the best time at which to determine early HCV RNA clearance. Whether patients who attain RVR are at lower risk for relapse because of their longer on-treatment period of undetectable HCV RNA was evaluated (Tables 2 and 3).2, 3, 10, 11, 15–20, 22–26, 28–32
|Reference||Treatment/CHC study population||Treatment duration, week||Patients with RVR, n/N (%)||Patients without RVR, n/N (%)|
|Davis et al.11||PEG-IFN alfa-2b (1.5) + RBV (800 mg/day)||48||150/511 (29.4)||–||133/150 (88.7)||–||361/511 (70.6)||–||140/361 (38.8)||–|
|Ferenci et al.10||PEG-IFN alfa-2a (180) + RBV (1000–1200)||48||302/453 (66.7)||–||227/302 (75.2)||–||151/453 (33.3)||–||39/151 (25.8)||–|
|Zeuzem et al.31||PEG-IFN alfa-2b (1.5) + RBV (800–1400)||G1 (LVL): 24||110/237 (46.4)||106/110 (96.3)||98/110 (89.1)||9/106 (8.5)||127/237 (53.6)||79/127 (62.2)||19/127 (15.0)||60/79 (75.9)|
|G1 (LVL): 48||–||–||11/13 (84.6)||1/12 (8.3)||–||–||–||–|
|Segadas-Soares et al.15||PEG-IFN alfa-2b (1.5) + RBV (1000–1250)||48†||35/103 (34.0)||–||25/35 (71.4)||–||68/103 (66.0)||–||20/68 (29.4)||–|
|Jensen et al.16Post hoc analysis of Hadziyannis et al.32||PEG-IFN alfa-2a (180) + RBV||RBV 800 mg/day: 24 (n = 101)||18/101 (17.8)||17/18 (94.4)||16/18 (88.9)||1/17 (5.9)||81/101 (80.1)||51/81 (63.0)||13/81 (16.0)||38/51 (74.5)|
|RBV 1000–1200 mg/day: 24 (n = 118)||33/118 (28.0)||32/33 (97.0)||29/33 (87.9)||3/32 (9.4)||84/118 (71.2)||59/84 (70.2)||19/84 (22.6)||40/59 (67.8)|
|RBV 800 mg/day: 48 (n = 250)||40/250 (16.0)||29/40 (72.5)||29/40 (72.5)||0/29 (0)||208/250 (83.2)||116/208 (55.8)||73/208 (35.1)||43/116 (37.1)|
|RBV 1000–1200 mg/day: 48 (n = 271)||55/271 (20.3)||51/55 (92.7)||50/55 (90.9)||1/51 (2.0)||210/271 (77.5)||132/210 (62.9)||92/210 (43.8)||40/132 (30.3)|
|Berg et al.17||PEG-IFN alfa-2a (180) + RBV (800) G1 patients with HCV RNA >1000 IU/mL at baseline||48||51/230 (22.2)||46/51 (90.2)||43/51 (84.3)||3/46 (6.5)||179/230 (77.8)||–||78/179 (43.6)||–|
|72||35/225 (15.6)||32/35 (91.4)||27/35 (77.1)||5/32 (15.6)||190/225 (84.4)||–||94/190 (49.5)||–|
|Sanchez-Tapias et al.18||PEG-IFN alfa-2a (180) + RBV (800)||No RVR: 48||–||–||–||–||n = 149||87/149 (58.3)||41/149 (27.5)||46/87 (52.8)*|
|No RVR: 72||–||–||–||–||n = 142||88/142 (62.0)||63/142 (44.4)||13/88 (14.8)|
|RVR: 24 (G2/3, G1 LVL)||n = 148||–||117/148 (79.1)||–||–||–||–||–|
|RVR: 48 (G1 HVL)||n = 36||–||23/36 (63.8)||–||–||–||–||–|
|Mangia et al.3||PEG-IFN alfa-2a (180) or PEG-IFN alfa-2b (1.5) + RBV (1000–1200)||RVR: 24 weeks||123/459 (26.8)||117/123 (95.1)||95/123 (77.2)||22/117 (18.8)||–||–||–||–|
|RVR: 48 weeks||62/237 (26.2)||60/62 (96.7)||54/62 (87.1)||6/60 (10)||–||–||–||–|
|No RVR: 48 weeks||–||–||–||–||n = 175||70/175 (40.0)||53/175 (30.3)||17/70 (24.3)|
|No RVR: 72 weeks||–||–||–||–||n = 208||50/208 (24.0)||37/208 (17.9)||13/50 (26.0)|
|Ferenci et al.19||PEG-IFN alfa-2a (180) + RBV (1000–1200)||RVR: 24 weeks||120/450 (82.8)||103/120 (85.8)||89/120 (74.2)||14/120 (10.7)||–||–||–||–|
|Sulkowski et al.20 IDEAL||PEG-IFN alfa-2b (1.5) + RBV (800–1400)||RVR: 48 weeks||116/1019 (11.4)||107/116 (92.2)||–||–||–||–|
|PEG-IFN alfa-2b (1.0) + RBV (800–1400)||RVR: 48 weeks||79/1016 (7.8)||–||69/79 (87.3)||–||–||–||–||–|
|PEG-IFN alfa-2a (180) + RBV (1000–1200)||RVR: 48 weeks||123/1035 (11.9)||–||98/123 (80.0)||–||–||–||–||–|
|Mangia et al.23||PEG-IFN alfa-2b (1.0) + RBV (1000–1200)||G2/3: 12/24||n = 133||126/133 (94.7)||113/133 (84.9)||13/126 (10.3)||n = 80||54/80 (67.5)||51/80 (63.8)||3/54 (5.6)|
|G2: 12/24||n = 102||98/102 (96.2)||89/102 (87.3)||9/98 (9.2)||n = 58||45/58 (77.6)||42/58 (72.4)||3/45 (6.7)|
|G3: 12/24||n = 31||28/31 (90.3)||24/31 (77.4)||4/28 (14.3)||n = 22||9/22 (40.9)||9/22 (40.9)||0/9 (0)|
|G2/3: 24||n = 45||42/45 (93.3)||41/45 (91.1)||1/42 (2.4)||n = 25||13/25 (52.0)||12/25 (48.0)||1/13 (7.7)|
|G2: 24||n = 35||32/35 (91.4)||31/35 (88.6)||1/32 (3.1)||n = 18||10/18 (55.6)||9/18 (50.0)||1/10 (10.0)|
|G3: 24||n = 10||10/10 (100)||10/10 (100)||0/10 (0)||n = 7||3/7 (42.9)||3/7 (42.9)||0/3 (0)|
|Von Wagner et al.26||PEG-IFN alfa 2a (180) + RBV (800–1200) 152 patients were randomly assigned to 16 or 24 weeks’ treatment based on RVR status||G2/3: 16||n = 71||67/71 (94.4)||58/71 (81.7)||9/67 (13.4)||–||–||–||–|
|G2: 16||n = 19||–||18/19 (94.7)||–||–||–||–||–|
|G3: 16||n = 51||–||39/51 (76.5)||–||–||–||–||–|
|G2/3: 24||n = 71||60/71 (84.5)||57/71 (80.3)||3/60 (5.0)||n = 11||8/11 (72.7)||4/11 (36.4)||4/8 (50.0)|
|G2: 24||n = 19||–||18/19 (94.7)||–||–||–||–||–|
|G3: 24||n = 52||–||39/52 (75.0)||–||–||–||–||–|
|Dalgard et al.22||PEG-IFN alfa-2b (1.5) + RBV (800–1400)||G2/3: 14||n = 95||95/95 (100)||85/95 (89.5)||10/95 (10.5)||–||–||–||–|
|G2: 14||n = 21||21/21 (100)||19/21 (90.5)||2/21 (9.5)||–||–||–||–|
|G3: 14||n = 74||74/74 (100)||66/74 (89.2)||8/74 (10.8)||–||–||–||–|
|G2/3: 24||–||–||–||–||n = 27||22/27 (81.5)||15/27 (55.6)||7/22 (31.8)|
|G2: 24||–||–||–||–||n = 2||1/2 (50.0)||1/2 (50.0)||0/1 (0)|
|G3: 24||–||–||–||–||n = 25||21/25 (84.0)||14/25 (56.0)||7/21 (33.3)|
|Dalgard et al. (North C)24||PEG-IFN alfa-2b (1.5) + RBV (800–1400) 298 patients with RVR were randomly assigned to 14 or 24 weeks’ treatment||G2/3: 14||n = 148||136/148 (91.9)||120/148 (81.1)||16/136 (10.8)||–||–||–||–|
|G2: 14||n = 29||–||27/29 (93.1)‡||2/29 (6.9)‡||–||–||–||–|
|G3: 14||n = 119||–||93/110 (84.5)‡||17/110 (15.5)‡||–||–||–||–|
|G2/3: 24||n = 150||144/150 (96.0)||136/150 (90.7)||8/144 (5.6)||–||–||–||–|
|G2: 24||n = 31||–||30/31 (96.8)‡||1/31 (3.2)‡||–||–||–||–|
|G3: 24||n = 119||–||106/115 (92.2)‡||9/115 (7.8)‡||–||–||–||–|
|Shiffman et al. (ACCELERATE)25||PEG-IFN alfa-2a (180) + RBV (800)||G2/3: 16||487/730 (67)||–||(79)||–||243/730 (33.3)||–||(26)||–|
|G2: 16||257/372 (69)||–||(78)||–||115/372 (30.9)||–||(26)||–|
|G3: 16||230/358 (64)||–||(80)||–||128/358 (35.8)||–||(26)||–|
|G2/3: 24||466/725 (64)||–||(85)||–||259/725 (35.7)||–||(45)||–|
|G2: 24||247/356 (69)||–||(85)||–||109/356 (30.6)||–||(53)||–|
|G3: 24||219/369 (59)||–||(85)||–||150/369 (40.7)||–||(39)||–|
|Willems et al.29Post hoc analysis of patients without RVR from Hadziyannis32 and Fried2 studies||PEG-IFN alfa-2a (180) + RBV (800)||2432||–||–||–||–||n = 21||–||14/21 (66.7)||5/19 (26.3)|
|4832||–||–||–||–||n = 30||–||20/30 (66.7)||3/23 (13.0)|
|PEG-IFN alfa-2a (180) + RBV (1000–1200)||2432||–||–||–||–||n = 34||–||22/34 (64.7)||7/29 (24.1)|
|4832||–||–||–||–||n = 37||–||28/37 (75.6)||1/27 (3.7)|
|Yu et al.28||PEG-IFN alfa-2a (180) + RBV (1000–1200)||G2: 16-week treatment arm||43/50 (86.0)||43/43 (100)||43/43 (100)||0/43 (0)||7/50 (14.0)||7/7 (100)||4/7 (57.1)||3/7 (42.9)|
|G2: 24-week treatment arm||87/100 (87.0)||87/87 (100)||85/87 (97.7)||2/87 (2.3)||13/100 (13.0)||11/13 (84.6)||10/13 (76.9)||1/11 (9.1)|
|Lagging et al.30 NORDynamIC||PEG-IFN alfa-2a (180) + RBV (800)||G2/3: 12 weeks||n = 120||71||26||41.0||46|
|G2/3: 24 weeks||n = 111||91||4||62||25|
|Genotype||RVR||Treatment duration||EOTR, % (n)||SVR, % (n)||Relapse, % (n)|
|1||Yes||24||86–97 (5)||74–89 (6)||6–19 (5)|
|48||73–97 (4)||73–92 (10)||0–10 (5)|
|No||48||40–63 (4)||23–44 (6)||24–53 (4)|
|72||24–62 (2)||18–49 (3)||17–26 (2)|
|2||Yes||12–16||96–100 (3)||78–100 (6)||0–9.5 (4)|
|24||91–100 (2)||85–97 (5)||2.3–3.1 (3)|
|No||12–16||100 (1)||26–57 (2)||43 (1)|
|24||50–85 (4)||50–77 (5)||0–10 (4)|
|3||Yes||12–16||90–100 (2)||77–89 (5)||11–16 (3)|
|24||100 (1)||75–100 (4)||0–8 (2)|
|24||41–84 (3)||39–56 (4)||0–33.3 (3)|
Genotype 1 patients. Within the G1 cohort, a strong relationship exists between RVR and relapse. Among G1 patients who attained RVR, relapse rates were 19% or less regardless of treatment duration, even among those who were treated for fewer than 48 weeks (Table 3). In contrast, at least 25% of patients who did not attain RVR experienced relapse. Relapse rates as high as 53% among G1 patients who did not attain RVR and were treated for the standard 48 weeks were reported in at least 1 study (note: this study used a fixed 800-mg/day ribavirin dose).18
Among G1 patients who attained RVR, little difference in treatment outcomes was apparent between 24- and 48-week treatment durations. Relapse rates were slightly higher among patients treated for 24 vs. 48 weeks (6–19% vs. 0–10%), but the differences in relapse rates fell within acceptable limits given the similarities in SVR rates (74–89% vs. 73–92%). Among G1 patients who did not attain RVR, relapse rates were also high, regardless of treatment duration. These data clearly reinforce the concept that patients who do not clear HCV RNA early are prone to relapse and that this relationship remains unaffected by extending treatment duration to 72 weeks. These data also support the current trend in clinical practice that leans towards the use of pEVR and not RVR for delineating the ‘slow-responder’ G1 population who derive most benefit from an extended, 72-week treatment duration.33
Genotype 2/3 patients. For G2 and G3 patients, use of 12–16 weeks of treatment among patients who attain RVR was associated with a slightly elevated risk for relapse. Relapse rates were 0–9.5% among G2 patients treated for 12–16 weeks and 2.3–3.1% among G2 patients treated for 24 weeks. Similarly, relapse rates were 11–16% and 0–8% respectively among G3 patients. However, given the universally high SVR rates associated with 12- to 16-week (G2: 78–100%; G3: 77–89%) and 24-week (G2: 85–97%; G3: 75–100%) treatment durations, the potential for increased relapse risk in patients treated for less time may be clinically acceptable.
Conversely, G2/3 patients who do not attain RVR probably require 24 weeks of treatment to mitigate relapse. Although data are limited, results from a single study confirmed that shortened treatment duration is associated with very high relapse rates in patients who do not attain RVR. Yu et al.28 treated 50 G2 patients with PEG-IFN alfa-2a (180 μg/week) plus ribavirin (1000–1200 mg/day) for 16 weeks. Forty-three patients attained RVR and SVR. The seven patients who did not attain RVR did attain EOT response; however, three (43%) of the seven experienced relapse off treatment.28 In contrast, a 24-week treatment duration is more suitable for G2 and G3 patients who do not attain RVR; however, even within this cohort, relapse rates of up to 32% were reported.22
Rapid virological response is an important and accurate positive predictor of SVR. For this reason, treatment duration can be shortened among patients who attain RVR without discernible declines in SVR rates. Results of this literature analysis suggest that treatment may be shortened from 48 to 24 weeks among G1 patients who attain RVR. This observation is consistent with the recommended use of PEG-IFN alfa in Europe, where there is an optional 24-week treatment duration for G1 patients with baseline viral load <600 000 IU/mL (PEG-IFN alfa-2b) or ≤800 000 IU/mL (PEG-IFN alfa-2a) who have undetectable HCV RNA at week 4 and who remain virus-free at week 24 (Figure 4).34, 35 RVR is a less reliable predictor of treatment failure. Approximately 60% of G1 patients who do not attain RVR will not attain SVR when treated for 48 weeks. Thus, lack of RVR represents a setback in treatment progress, but does not represent grounds for withdrawing patients from therapy because approximately 40% of G1 patients who do not attain RVR will go on to attain SVR. Furthermore, in the present analysis, the sensitivity of assays used to define RVR varied widely across the included studies, with LLDs ranging from 50 to 600 IU/mL. However, we strongly advise that wherever possible, an assay with a LLD ≤50 IU/mL be used to define RVR, thus avoiding the erroneous classification of very low levels of HCV RNA as undetectable.
Among G2 and G3 patients, the predictive value of RVR for SVR is dependent on treatment duration and ribavirin dosing. Treatment can be shortened from 24 weeks to 12–16 weeks in patients with RVR, but these data suggest it is preferable to use a weight-based ribavirin regimen alongside the reduced treatment duration. With a shortened treatment duration and a fixed 800-mg/day ribavirin dose, RVR is a weak predictor of SVR, but lack of RVR is a strong predictor of treatment failure. In contrast, RVR is a strong predictor of SVR, but lack of RVR is a weak predictor of treatment failure when a weight-based ribavirin dosing regimen is used for 24 weeks.
What ribavirin dosing schedule should be used in G2 and G3 patients? At the outset of therapy, physicians must consider whether to initiate a standard fixed-dose9 or a weight-based dosing regimen of ribavirin. Results of the present analysis indicate that RVR rates are comparable among G2/3 patients receiving weight-based ribavirin (68.8%) and fixed-dose ribavirin (65.5%). However, beyond week 4, our data suggest that the type of dosing chosen affects SVR rates (Figure 4).
This pooled analysis suggests that for G2/3 patients who do not attain RVR, fixed-dose ribavirin can be continued for 24 weeks. However, these observations suggest that consideration may be given to adjusting ribavirin dose in G2/3 patients who are started on a fixed 800-mg/day dose and subsequently attain RVR and are thus selected for shortened treatment duration. For these patients, the clinician may elect to make no change to the fixed 800-mg/day dose, a strategy which, our data suggest, is associated with slightly lower PPV than if using weight-based ribavirin; or change to a weight-based ribavirin dosing, a strategy which would necessitate an increase in ribavirin dosing for some patients, but which is also associated with a high likelihood of SVR. In the event of ribavirin-related toxicity arising from the increase in dose, ribavirin could be safely reduced back to 800 mg/day and treatment duration returned to the original 24 weeks. Alternatively, G2/3 patients could start with a weight-based ribavirin regimen. This regimen would then be maintained for 12–16 weeks in patients who attain RVR. In patients who experience ribavirin-related toxicity, the dose could be reduced to 800 mg/day and treatment duration extended to the standard 24-week period. These approaches require prospective evaluation before any decisions can be made regarding the more effective regimen. PEG-IFN alfa-2a and PEG-IFN alfa-2b have different approved regimens for the treatment of G2/3 patients. In the United States and Europe, PEG-IFN alfa-2b is approved for the treatment of G2/3 patients when combined with a weight-based ribavirin schedule (800–1400 mg/day) for a period of 24 weeks.34, 36 In contrast, in the United States and Europe, PEG-IFN alfa-2a is approved when combined with ribavirin at a fixed dose of 800 mg/day, also for 24 weeks.35, 37 However, European recommendations also indicate that a treatment period of 16 weeks (without any adjustment of ribavirin dose) may be considered with PEG-IFN alfa-2a in G2/3 patients with baseline viral load ≤800 000 IU/mL who have undetectable HCV RNA at week 4 that remains undetectable at week 16.35
Several other questions remain to be prospectively evaluated. For example, prospective data are required to determine whether treatment duration can be shortened to 12 weeks23 or whether more conservative 14-week22 or 16-week26 treatment durations offer improved outcomes. Furthermore, the relative merits of various weight-based ribavirin regimens also require assessment. Prospective comparison of the ‘full’ weight-based approach (800–1400 mg/day22) with less structured 800- to 1200-mg/day26 or 1000- to 1200-mg/day23 treatment is also of interest. Baseline viral load is also recognized as an important factor in determining appropriate treatment duration and studies have shown a close relationship between low baseline viral load and increasing likelihood of RVR, both in patients with G1 and with G2/3 infection.16, 34 However, a proportion of patients with high baseline viral load also attain RVR, with data indicating that 9.2% of G1 patients with baseline viral load >600 000 IU/mL will attain RVR.16 Clearly, there is a strong association between baseline viral load and RVR, and further study is required to establish the degree of accuracy with which baseline viral load can predict RVR.
The present analysis confirms the strong predictive value of RVR for SVR across the entire continuum of chronic hepatitis C disease and shows how the predictive value of RVR can fluctuate according to genotype, treatment duration and ribavirin regimen. However, SVR is not the only possible clinical outcome in patients who attain an EOT response. Relapse is also emerging as an important clinical end point. Patients who have relapses after full-course therapy experience emotional strain related to the late failure of therapy. Recent data from the Hepatitis C Antiviral Long-term Treatment Against Cirrhosis (HALT-C) study showed significant declines in measures of physical and sexual functioning and general health (as assessed using the 36-item Short-Form Health Survey) in patients who experienced relapse compared with patients who attained SVR after re-treatment.38 Furthermore, maximum health care expenditure (to deliver a full course of therapy and manage side effects for the entire treatment period) is lost when patients experience relapse.
Although not confirmed through statistical analysis, the results of the present analysis strongly indicate a close relationship between RVR and relapse. Across all genotypes, relapse rates are generally higher among patients who do not attain RVR than among those who do, an observation that is consistent with the theory of continuously undetectable HCV RNA. Further studies are urgently required to understand the full pathobiology of relapse. Clinical focus on EOT response rate as a marker of efficacy is misleading and can create false hope of SVR among patients who attain undetectable HCV RNA late in their treatment course. Of greater clinical value are early changes in HCV RNA, which can be used to tailor therapy according to each patient’s response. Measures to mitigate relapse should begin early in therapy, when proactive control of influencing factors, such as ribavirin dosing and treatment duration, is possible. The belief that relapse occurs simply as a function of end-of-treatment viraemia represents a passive attitude to managing relapse and does not adequately take into account the on-treatment factors that can influence this treatment outcome.
Finally, while this analysis has focused on data regarding the role of RVR in the treatment of hepatitis C genotypes 1, 2 and 3 infections, there is also a limited amount of information available regarding the role of RVR in treating hepatitis C genotype 4. Ferenci et al. reported that 30 of 66 patients (45.5%) receiving PEG-IFN alfa-2a plus ribavirin attained RVR with a PPV of 86.7% when treated for 24 weeks.19 Similarly, Kamal et al. found that 69 of 358 (19.3%) patients with G4 infection attained RVR when treated with PEG-IFN alfa-2b plus ribavirin and, as in the study by Ferenci et al., 86% of these patients attained SVR after a 24-week treatment duration.39 This study also included a control population in which 50 patients with G4 infection were treated for a fixed 48-week period. Among these 50 patients, eight attained RVR (16%) of whom seven attained SVR (88%) after 48 weeks of therapy, suggesting that there is no incremental benefit in extending treatment duration from 24 to 48 weeks in G4 patients with RVR.39
In conclusion, the present analysis confirms the importance of RVR in the individualized approach to the treatment of patients with chronic hepatitis C. RVR is a strong predictor of SVR and, as such, can be used to tailor treatment duration. However, the results of this analysis confirm that RVR is a dynamic variable influenced by other factors. Therefore, the predictive value of RVR has to be appreciated in the context of treatment duration and treatment regimen. These findings confirm that a strong continuum exists throughout a course of treatment for chronic hepatitis C and the influence that an early event such as RVR can have on SVR and relapse rates. Controlled prospective studies are now warranted to evaluate these observations in a clinical setting.
Declaration of personal interests: Fred Poordad has served as a speaker for Gilead, Novartis, Valeant and Schering-Plough Corp., now Merck & Co., Inc., Whitehouse Station, N.J. He has received research funding from Schering-Plough Corp., now Merck & Co., Inc., Whitehouse Station, N.J., Roche, Gilead, Bristol-Myers Squibb, Vertex, Wyeth, Valeant, Intermune, Idenix, Novartis, Human Genome Sciences, GlaxoSmithKline and Intarcia. Declaration of funding interests: Writing assistance was provided by Tim Ibbotson, PhD, and Maribeth Bogush, PhD. This assistance was funded by Schering-Plough Corp., now Merck & Co., Inc., Whitehouse Station, N.J.
- 6Differentiation of early virologic response (EVR) into RVR, complete EVR (CEVR) and partial EVR (PEVR) allows for a more precise prediction of SVR in HCV genotype 1 patients treated with peginterferon alfa-2a (40kd) (Pegasys) and ribavirin (Copegus). Hepatology 2007; 46: 818A. Abstract 1308., , , .
- 12Rapid virologic response is enhanced by higher drug exposure among patients receiving taribavirin in combination with pegylated interferon alfa-2b for the treatment of HCV infection. Hepatology 2006; 44(4 suppl 1): 616A. Abstract 1146., , , et al.
- 13the International Hepatitis Interventional Therapy Group. Clinical implications of testing viral response during ribavirin and peginterferon α-2b treatment for hepatitis C. Hepatology 2002; 36(suppl 1): 281A. Abstract 472., , , , , and
- 14Correlations between rapid virologic response, early virologic response and sustained virologic response in HCV genotype 1 patients treated with pegylated interferon alfa-2a and ribavirin. Gastroenterology 2005; 128(4 suppl 2): A-716. Abstract S1540., , , .
- 27Comparison of interferon alfa-2a and ribavirin for 12 or 24 weeks in patients with HCV genotype 2 or 3: the CLEO trial. Presented at: 58th Annual Meeting of the American Association of the Study of Liver Diseases; November 2–6, 2007; Boston, MA., , , et al.
- 34SP Europe. PegIntron [Summary of Product Characteristics]. Bruxelles, Belgium: SP Europe, 2008.
- 35Roche Registration Limited. Pegasys [Summary of Product Characteristics]. Welwyn Garden City, UK: Roche Registration Limited, 2008.
- 36Schering Corporation. PegIntron (Peginterferon Alfa-2b) Injection. Kenilworth, NJ: Schering Corporation, 2008.
- 37Hoffman-LaRoche, Inc. Pegasys (Peginterferon Alfa-2a). Nutley, NJ: Hoffmann-La Roche Inc., 2004.