| |
Undetectable hepatitis C virus RNA at week 4 as predictor of sustained virological response in HIV patients with chronic hepatitis C
|
| |
| |
AIDS:Volume 22(1)2 January 2008p 15-21
Martin-Carbonero, Luza; Nunez, Marinaa; Marino, Anab; Alcocer, Federicoc; Bonet,
Lluciad; Garcia-Samaniego, Javiera; Lopez-Serrano, Pilare; Cordero, Miguelf;
Portu, Josebag; Soriano, Vincenta
From the aHospital Carlos III, CIBEREHD, Madrid, Spain
bHospital Arquitecto Marcide, Ferrol, Spain
cHospital Clinico, Valencia, Spain
dHospital Son Dureta, Palma de Mallorca, Spain
eFundacion Hospital de Alcorcon, Madrid, Spain
fHospital Universitario, Salamanca, Spain
gHospital Txagorritxu, Vitoria, Spain.
Note from Jules: taking a cursory look through this study report I did not see that authors compared week 4 to week 12 response (early viral response) to see which performed better. But at recent AASLD there were several study reports in moninfected finding RVR, week 4 response, can be a reliable predictor. You can see these reports on the NATAP website in AASLD Conference Reports Section. Certainly taken together the data suggests that a week 4 undetectable HCV RNA in coinfected is a good indicator of achieving SVR, although I'm not sure this study establishes that.
Abstract
Background: Low baseline serum hepatitis C virus (HCV)-RNA and clearance of viraemia at week 4 with pegylated interferon (pegIFN) plus ribavirin therapy predict sustained virological response (SVR) and enable a shorter duration of therapy in patients with chronic hepatitis C. It is unclear whether this applies to HIV/HCV-co-infected patients.
Patients and methods: In the Peginterferon Ribavirin ESpana COinfection (PRESCO) trial, 389 co-infected patients received pegIFN-ƒ¿2a 180 μg/week plus ribavirin 1000-1200 mg/day. Patients with HCV-2/3 were treated for 6 or 12 months, whereas patients with HCV-1/4 were treated for 12 or 18 months. For each genotype, baseline HCV-RNA and rapid virological response (RVR), defined as under 50 IU/ml HCV-RNA at week 4, were evaluated as predictors of SVR in an 'on-treatment' analysis.
Results: Overall, SVR was achieved by 193 patients (49.6%), 68/191 (35.6%) with genotype 1, 110/152 (72.4%) with genotypes 2/3 and 15/46 (32.6%) with genotype 4. RVR was the best predictor of SVR regardless of HCV genotype. Only for HCV-1 patients, baseline HCV-RNA less than 500 000 IU/ml was also associated with SVR. In HCV-3 patients RVR had a positive predictive value (PPV) for SVR of 90%, with treatment for 24 or 48 weeks. The PPV of SVR for patients with RVR was 69% for HCV-1 and 83% for HCV-4.
Conclusion: Undetectable HCV-RNA at week 4 is the best predictor of curing chronic hepatitis C in HCV/HIV-co-infected patients. In HCV-1 patients, baseline HCV-RNA also predicts response. HIV patients with HCV-3 and RVR may permit shortening therapy duration to only 24 weeks of pegIFN plus 1000-1200 mg ribavirin.
Introduction
It is desirable to expose patients with chronic hepatitis C to the shortest possible duration of treatment to minimize the deleterious impact of adverse events and reduce costs [1]. Current guidelines in hepatitis C virus (HCV)-mono-infected patients recommend the use of pegylated interferon (pegIFN) plus ribavirin for 12 months in patients infected with HCV genotypes 1 or 4, whereas for subjects infected by HCV genotypes 2 or 3, 6 months of therapy are enough [2]. Moreover, although weight-based ribavirin dosing is recommended for genotypes 1/4 patients, a flat 800 mg/day seems to be enough for genotypes 2/3 patients [3]. During the past few years attempts to individualize the duration of therapy according to early virological response have been made, and most experts are currently confident to shorter HCV treatment duration in patients showing a rapid virological response (RVR), defined as the achievement of undetectable serum HCV-RNA at week 4 of therapy [1]. In such individuals only 24 weeks of therapy for genotypes 1/4 [4-6] or 12-16 weeks for genotypes 2/3 might be enough [7-9].
A significant proportion of patients with chronic hepatitis C in western countries are co-infected with HIV, particularly among individuals who are or have been any time in the past engaged in intravenous drug use [10]. In this population response to hepatitis C therapy is lower than in HCV-mono-infected patients regardless of HCV genotype [11-18]. Interestingly, high relapse rates upon completion of treatment have been reported in HCV genotype 3 co-infected patients treated for just 24 weeks with only 800 mg/day of ribavirin [19,20]. Given the concern for a greater risk of interactions between ribavirin and antiretroviral drugs, this low dose of ribavirin has been used in most trials conducted so far in co-infected patients regardless of HCV genotype. In an attempt to counterbalance this effect, HCV therapy for 12 months has been recommended for all genotypes in most guidelines designed for co-infected patients [21,22].
Attempts to move from simple but rigid schedules of therapy in co-infected patients have shifted the attention to early HCV kinetics. Consideration of early serum HCV-RNA decays on therapy might allow the length of treatment to be shortened in a subset of co-infected patients, as is currently being advised for HCV-mono-infected individuals [1,4-9]. This information is unknown for HCV/HIV-co-infected patients and clearly should impact treatment decisions and management in this population, in which premature treatment discontinuations as a result of side effects are often very challenging [13,23]. The aim of this study was to assess retrospectively the predictive value of baseline serum HCV-RNA and of RVR for sustained virological response (SVR) in a large trial conducted in HCV/HIV-co-infected patients.
Patients and methods
The Peginterferon Ribavirin ESpana COinfection (PRESCO) trial was a multicentre, open, prospective trial conducted at 29 clinics across Spain that evaluated the efficacy and safety of subcutaneous pegIFN-ƒ¿2a (Pegasys; Roche, Basel, Switzerland) 180 μg/week plus oral ribavirin (Copegus; Roche) at doses of 1000 mg/day (if body weight is less than 75 kg) or 1200 mg/day (if body weight is greater than 75 kg) as treatment for chronic hepatitis C in HCV/HIV-co-infected individuals. More details of the study have been reported elsewhere [24]. Briefly, adult patients with HIV infection, detectable serum HCV-RNA, elevated serum alanine aminotransferases, compensated liver disease, no use of illicit drugs for at least one year, and CD4 cell counts greater than 300 cells/μl with or without antiretroviral therapy were recruited into the trial.
As recommended in HCV-mono-infected patients [2], early virological stopping rules were applied. Patients who achieved a reduction of less than 2 log10 in serum HCV-RNA at week 12 of therapy were thus considered as virological failures and discontinued HCV therapy prematurely. Similarly, subjects with detectable serum HCV-RNA at week 24 were also considered as virological failures and stopped HCV treatment. Although originally the treatment was planned for 24 weeks in HCV genotypes 2 and 3, and for 48 weeks in HCV genotypes 1 and 4, an amendment in the protocol was made in August 2004 to extend the course of therapy to 48 weeks in all HCV-2/3 and to 72 weeks in all HCV-1/4 patients recruited since that date. All patients were followed by a 24-week observational period after completion of therapy.
Regional health authorities and institutional review boards at participating centres approved the protocol. All patients provided written informed consent. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practices guidelines. Demographic data were obtained at baseline. All laboratory tests were performed locally. The HCV genotype was determined using a hybridization assay (InnoLiPA HCV; Bayer, Barcelona, Spain). Serum HCV-RNA was measured using a quantitative polymerase chain reaction assay, which has a lower limit of detection of 50 IU/ml (Cobas Monitor HCV v2.0; Roche, Barcelona, Spain). Testing was performed at baseline and at weeks 4, 12 and every 12 weeks thereafter until 24 weeks after completion of treatment. RVR was defined as undetectable serum HCV-RNA (< 50 IU/ml) at week 4 of treatment. SVR was defined by undetectable HCV-RNA 24 weeks after completion of treatment. Safety was assessed at weeks 4, 12 and every 12 weeks throughout the study period. Standard stepwise reductions in the dose of pegIFN-ƒ¿2a to 135 or 90 μg per week, or ribavirin by 200 mg/day but to no lower than 600 mg/day were advised for managing neutropenia or thrombocytopenia, and anaemia, respectively [24].
Statistical analyses
To assess the crude effect of baseline serum HCV-RNA and RVR on SVR, all analyses in this study were performed in patients on-treatment. In other words, only patients who completed therapy or discontinued the study for virological failure were examined. All patients who discontinued therapy prematurely as a result of adverse events, voluntary withdrawal or were lost to follow-up were excluded from this analysis. Assessment of the impact of these variables (intent-to-treat analyses) has been the subject of other substudies [25].
Results are given as absolute numbers and percentages. Comparisons between groups were performed using the Pearson chi-square or Fisher's exact tests for categorical variables. Univariate and multivariate logistic regression analyses were performed to assess the impact of baseline serum HCV-RNA, RVR and length of treatment on SVR. Differences were considered to be significant when P values were below 0.05. All reported P values were two-sided. All analyses were performed using SPSS (version 11.0; SPSS Inc., Chicago, Illinois, USA).
Results
A total of 389 patients were included in the trial and received at least one dose of the study medication. The main baseline characteristics of this population have been reported elsewhere [24]. In an intent-to-treat analysis, SVR was achieved by 193 of patients (49.6%), 68/191 (35.6%) with genotype 1, three out of three (100%) with genotype 2, 107/149 (71.8%) with genotype 3 and 15/46 (32.6%) with genotype 4. A significant proportion of genotypes 1/4 patients allocated to the extended treatment arms prematurely discontinued therapy, mainly because of voluntary withdrawal. As result, only nine patients infected with genotype 1, two with genotype 2, 45 with genotype 3 and none with genotype 4 completed the planned extended course of therapy. To assess the crude impact of baseline viral load and RVR, 331 patients who fulfilled the criteria for an 'on-treatment analysis' were analysed. We excluded 58 patients who discontinued treatment because of adverse events (33), voluntary withdrawal (13) or were lost to follow-up (12). Genotype 2 patients were also excluded, because the study merged genotypes and only three HCV-2 patients were included in the PRESCO trial. Of note is the fact that all of them achieved SVR. Finally, 18 patients discontinued treatment before completing the extended planned schedule, but were included in the 'on-treatment' analysis as belonging to the short treatment arm (6 months for genotype 3 and 12 months for genotype 1). HCV-RNA at week 4 could be performed in only 210 out of 328 patients, because a lack of sufficient plasma volume precluded further testing in the rest.
Hepatitis C virus genotype 1
Overall, 117 patients (75%) had baseline serum HCV-RNA greater than 500 000 IU/ml. RVR was achieved in 20/78 (25.6%) of them. In contrast, it was reached by 12/21 patients (57.1%) with lower serum HCV-RNA values (P < 0.01).
In the on-treatment analysis, SVR was achieved by 68/156 HCV-1 patients (43.5%), being higher in those with baseline HCV-RNA less than 500 000 IU/ml compared with the rest: 27/39 (69%) versus 41/117 (35%; P < 0.001); as well as in patients who achieved RVR: 22/32 (69%) versus 20/67 (30%; P < 0.001; Fig. 1). The achievement of RVR had a positive predictive value (PPV) of SVR of 100% (12/12) in patients with low baseline serum HCV-RNA. In contrast, the PPV dropped to 50% (10/20) in patients with high serum HCV-RNA. Statistically significant differences were observed when comparing these two PPV (P = 0.004). The negative predictive value (NPV) was 70% and was not influenced significantly by baseline serum HCV-RNA values (Table 1).
Only nine HCV-1 patients completed the planned 18 months extended duration of treatment. The SVR did not differ significantly in this subset of patients compared with those who were treated for only 12 months (67 versus 42%; P = 0.1). HCV-1 patients with high baseline serum HCV-RNA or with lack of RVR tended to have higher SVR with 18 than 12 months of therapy, however, although these differences did not reach statistical significance probably because of the small size of the study populations: 57 versus 33% (P = 0.2) and 66 versus 28% (P = 0.2), respectively.
In a multivariate logistic regression analysis that included baseline serum HCV-RNA, RVR and treatment duration, RVR and low serum HCV-RNA were significantly associated with SVR. An extended duration of treatment was nearly associated with SVR, but the trend did not reach statistical significance (Table 2).
Hepatitis C virus genotype 3
Overall, 86 (63.7%) had baseline serum HCV-RNA greater than 500 000 IU/ml. RVR was achieved by 41/54 patients (75.9%) with high viraemia and in 33/34 patients (97.1%) with low viraemia (P = 0.08).
In the on-treatment analysis, SVR was achieved in 107/135 patients (79.3%). SVR was higher in patients with baseline serum HCV-RNA less than 500 000 IU/ml than in the rest: 45/49 (92%) versus 62/86 (72%; P = 0.007) and in patients with RVR than in the rest: 67/74 (90%) versus 8/14 (57%; P < 0.001; Fig. 1). The PPV of RVR for SVR was 90% and the NPV was 43%. Baseline serum HCV-RNA levels did not significantly influence the predictive value of SVR in HCV genotype 3 patients (Table 1).
Overall, 90 patients were treated for 6 months and 45 were treated for 12 months. Response rates were similar in both groups (78 versus 82%; P = 0.5) In patients who achieved RVR, SVR was 89% (41/46) and 93% (26/28) in patients treated for 6 and 12 months, respectively (P = 0.7). Among patients without RVR, SVR was achieved in 50% (four out of eight) and 67% (four out of six) of patients treated for 6 and 12 months. Again, no statistically significant differences were found (P = 0.6); however, the small sample number can explain this result.
In the multivariate logistic regression analysis that included baseline HCV-RNA, RVR and length of treatment (24 versus 48 weeks in genotype 3 patients), only RVR was associated with SVR (Table 2).
Hepatitis C virus genotype 4
Overall, 17 patients (45.9%) had baseline serum HCV-RNA greater than 500 000 IU/ml. RVR was obtained by 26.1% of all HCV-4 patients, regardless of baseline HCV load (two out of 10 versus four out of 13 if less or more than 500 000 IU/ml, respectively).
In the on-treatment analysis, SVR was achieved in 15/37 (40.5%) of HCV-4 patients. In contrast with patients infected with HCV genotypes 1 and 3, baseline serum HCV-RNA was not associated with SVR: eight out of 20 (40%) versus seven out of 17 (41%) in patients with less or more than 500 000 IU/ml HCV-RNA (P = 0.1). RVR was, however, significantly associated with SVR in HCV-4 patients: five out of six (83%) versus five out of 17 (30%; P = 0.04; Fig. 1). The PPV and NPV of RVR for SVR in HCV-4 patients was 83.3 and 70.5%, respectively (Table 1).
Discussion
Baseline serum HCV-RNA predicts SVR to pegIFN plus ribavirin in both HIV-negative and HIV-positive patients with chronic hepatitis C [26,27]. It was the best predictor of SVR in the APRICOT trial, even better than HCV genotype [12]. The overall greater level of serum HCV-RNA in co-infected versus HCV-mono-infected patients (1 log on average) has been postulated to be one of the main reasons to explain the poorer results of HCV therapy in co-infected versus mono-infected patients [28]. Early assessment of HCV kinetics has only recently been accepted as a key predictor of response: HCV-mono-infected patients achieving undetectable viraemia as early as week 4 of treatment is now recognized as one of the most important predictors of SVR regardless of HCV genotype [1]. More importantly, on the basis of RVR, the duration of treatment can be individualized, allowing a shortening of the duration of therapy in patients clearing HCV viraemia as early as at week 4 [1,29]. Our results confirm that this principle can be applied to HCV/HIV-co-infected patients. An important consideration, however, is that baseline serum HCV-RNA seems also to be an independent predictor of SVR in HCV genotype 1 patients. For the rest of the HCV genotypes, clearance of viraemia at week 4 seems to be enough to tailor the duration of therapy. Two recent reports examining only a limited number of HCV genotype 3 patients have already found similar results [30,31].
Although most studies conducted in the past considered a threshold in serum baseline HCV-RNA of 800 000 IU/ml as the best to predict SVR, a lower value of 500 000 IU/ml was the best predictor of SVR in PRESCO [24]. This is in agreement with recent reports from studies conducted in HCV-mono-infected patients treated with pegIFN plus ribavirin, which have highlighted that 400 000 IU/ml seems to be the best threshold to predict SVR, at least in the subset of patients with HCV genotype 1 [32,33].
The PPV of SVR assessing RVR was high for all patients, but for the subset of subjects infected with HCV genotype 1 with high baseline serum HCV-RNA. In this group of patients, who are unfortunately quite common in the co-infected population [34], alternative treatment strategies should be explored. In HIV-negative patients, extension of treatment to 72 weeks in HCV genotype 1 patients who do not reach undetectable viraemia at week 4 but show more than 2 log declines in HCV-RNA at week 12 was shown to increase the chances of cure [35,36]. Attempts to extend the duration of therapy in co-infected patients have, however, been hampered by high drop-out rates given the poor tolerance of the HCV medication in most patients [24,37,38]. In the analysis per protocol, we did not find a statistically significant difference in SVR rates in HCV genotype 1 patients who received 48 or 72 weeks. In the multivariate analysis a trend towards a better response was, however, observed in the subset of patients treated for a longer time. Clearly, studies specifically designed to test whether HCV-1-co-infected patients with slow virological response might benefit from an extended length of therapy should be conducted. In parallel, strategies to minimize drop-outs during extended therapy should be implemented in this population.
Although the duration of treatment in HIV-positive patients co-infected with HCV genotype 3 has been established in 12 months [21,22], recent reports have shown that treatment may be shortened to 6 months, at least for the subset of patients with RVR [30,31]. In HCV-mono-infected patients, even shorter periods of treatment (12-16 weeks) have proved to be enough [7-9]. It should be emphasized, however, that weight-based ribavirin dosing was used in all these trials. Likewise, patients with HCV genotype 3 in the PRESCO trial also received weight-based ribavirin, and therefore our observation that no further benefit extending treatment from 24 to 48 weeks in HCV-3 patients may only apply when weight-based ribavirin is used and RVR is achieved. Using flat ribavirin doses of 800 mg/day, high relapse rates have been observed in at least two previous trials [19,20]. In contrast with HCV-1 patients, no significant independent influence of baseline HCV-RNA was seen for HCV-3-co-infected patients. The PPV of SVR in patients with RVR was 90% in our study and no further benefit was recognized in extending treatment to 48 weeks. Genotype 3 conclusions can probably be extended to genotype 2-infected patients as HCV-2 carriers respond even better to current anti-HCV therapy. Unfortunately, we were not able to address this issue in PRESCO becausee only three HCV-2- infected patients participated in the trial.
The small subset of co-infected patients with HCV-3 who did not achieve RVR had a relatively low likelihood of reaching SVR (57%). Of note is the fact that this observation was independent of baseline serum HCV-RNA and duration of therapy. Therefore, new strategies should be considered for these patients. Although most of the new antiviral drugs against HCV seem to be quite specific for the more difficult-to-treat HCV genotype 1 patients, some experimental new HCV polymerase inhibitors may be active against HCV-3 and provide hopes for this subset of refractory HCV-3-co-infected patients.
Finally, the achievement of SVR in HCV-4-co-infected patients in our study was mainly driven by the achievement of RVR, with no influence of baseline serum HCV-RNA levels. These results, however, should be interpreted with caution because the number of patients infected with this genotype was relatively small. As for HCV-1, the NPV of RVR for SVR was relatively low in HCV-4, which means that approximately one third of co-infected patients with these genotypes were able to achieve SVR despite not experiencing RVR. This observation was even more pronounced in HCV-3-co-infected patients, in whom the NPV of RVR for SVR was only 43%. Cure was thus obtained in more than half of these patients despite not achieving undetectable viraemia at week 4. Altogether, our results support in co-infection what has already been demonstrated in HCV mono-infection, which in brief is that early virological stopping rules should remain based on the results obtained at week 12 and not shift to earlier time points [1,27,37].
|
|
| |
| |
|
|
|
