Practical Issues – Therapy in Patients with Cirrhosis Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
Antiviral Therapy in Patients with Hepatitis C Virus-Induced Cirrhosis François Bailly a–c Pierre Pradat a–c Victor Virlogeux a, d Fabien Zoulim a–c c
Hepatology Department, Hospices Civils de Lyon, b INSERM Unit 1052, Cancer Research Center of Lyon (CRCL), Lyon University, and d Ecole Normale Supérieure, Lyon, France
Key Words Hepatitis C virus · Chronic hepatitis C · Cirrhosis · Antivirals · Therapy
Abstract Background: Opportunities to treat infection with hepatitis C virus (HCV) are evolving rapidly. From the introduction of interferon (IFN)-α monotherapy in the early 1990s to the approval of telaprevir- and boceprevir-based triple therapies with pegylated (PEG)-IFN-α and ribavirin (RBV) in 2011, the chances of curing patients infected with HCV genotype 1 have improved dramatically to reach approximately 70%. Significant further improvements that may cure virtually all HCV patients with an all-oral, IFN-free regimen are becoming progressively available. Key Messages: Historically, a PEGIFN/RBV combination therapy of patients with liver cirrhosis was associated with lower virological rates and a worse safety profile. The advent of the first protease inhibitor-based triple therapy was long expected, but the promise fell rapidly because of the numerous side effects and the requirement for intensive clinical management in cirrhotic patients. The newer direct-acting antivirals (DAAs) target the viral polymerase with either nucleos(t)ide analogues or nonnucleosidic inhibitors, the viral protease and the viral NS5A protein. Several clinical trials have now shown that a combination of sofosbuvir (nucleosidic polymerase inhibitor) with
© 2015 S. Karger AG, Basel 0257–2753/15/0334–0613$39.50/0 E-Mail [email protected] www.karger.com/ddi
daclatasvir or ledipasvir (NS5A inhibitors), or sofosbuvir with simeprevir (protease inhibitor), or a combination of ABT-450 (protease inhibitor) with ritonavir (ABT-450/r), the nonnucleosidic polymerase inhibitor ABT-333 and the NS5A inhibitor ABT-267, can achieve a sustained virological response in up to 95% of naive patients or previously treated patients, even in those who failed prior treatment with first-generation protease inhibitors. The best treatment regimens enable the achievement of comparable results even in cirrhotics, while other regimens still require RBV or a longer treatment duration to achieve optimal results. This improved risk/benefit ratio justifies early access programs of IFN-free regimens for cirrhotic patients. The remaining difficult-to-treat patients are cirrhotics infected with HCV genotype 3 and those with decompensated cirrhosis, for whom novel DAA combinations should be evaluated in clinical trials. Conclusions: As new DAAs are becoming available in early access treatment programs, treatment strategy studies are being performed to optimize treatment regimens with respect to the choice of DAAs and treatment duration, based on viral genotypes, prior treatment response and the presence of liver cirrhosis. In the near future, this should allow: (i) a decrease in the complications of HCV-induced cirrhosis, (ii) liver transplantations to be performed in virally cured patients, and (iii) the rescue of patients in the worst clinical situation (decompensated cirrhosis and HCV recurrence on liver graft). © 2015 S. Karger AG, Basel
Fabien Zoulim Department of Hepatology, Croix-Rousse Hospital 103, Grande rue de la Croix-Rousse FR–69317 Lyon Cedex 04 (France) E-Mail fabien.zoulim @ inserm.fr
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
a
Hepatitis C virus (HCV) infection is a leading cause of liver cirrhosis and hepatocellular carcinoma (HCC) [1] and is the most common indication for liver transplantation in Western countries [2]. Disease progression is generally slow with cirrhosis developing several decades after infection. However, substantial variations exist between individuals depending on the presence/absence of cofactors such as alcohol abuse, hepatitis B virus or HIV infection, metabolic disorders and genetic polymorphisms, etc. Cirrhosis is highly heterogeneous with a continuous spectrum of conditions ranging from early (compensated) to more severe (decompensated) cirrhosis and terminal end-stage liver disease. Compensated cirrhosis is also heterogeneous with more severe forms observed in cases of portal hypertension, esophageal varices, thrombo- or leucopenia, or low albumin levels [3]. The need for therapy increases with disease severity, which in turn is associated with a lower probability of virological response. The objectives of HCV therapy in patients with cirrhosis differ according to disease severity. In patients with compensated cirrhosis, treatment may slow down disease progression and prevent decompensation [4]. In patients with cirrhosis decompensation awaiting liver transplantation, viral eradication before transplantation may reduce the risk of HCV recurrence posttransplantation. Finally, another main objective of HCV therapy in cirrhotic patients is to reduce the risk of HCC development.
Interferon-Ribavirin Combination in Patients with HCV-Induced Cirrhosis
Antiviral treatment of HCV-induced compensated cirrhosis started with standard interferon (IFN), which gave controversial results. Valla et al. [5] found no beneficial effect on the development of decompensation or HCC, whereas Poynard et al. [6] suggested that IFN therapy could reduce the incidence of and mortality from HCC. In 2000, Heathcote et al. [7] showed in patients with cirrhosis that pegylated (PEG)-IFN once weekly was more effective for achieving a sustained virological response (SVR) compared with standard IFN (43 vs. 15%). Almost a decade later, results from the HALT-C trial showed, however, that prolonged therapy with a low PEG-IFN dose in HCV patients with a nonresponse to a previous course of PEG-IFN/ribavirin (RBV) and with 614
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
advanced fibrosis, with or without cirrhosis, did not reduce the rate of disease progression [8]. From 2000 to 2011, the standard of care for HCV treatment was a combination of PEG-IFN/RBV, allowing an SVR rate of about 50% in patients with HCV genotype 1 [9, 10]. Response to therapy varied according to genotype, with genotype 1 and 4 patients achieving lower SVR rates than patients infected with genotype 2 and 3 (40–50 vs. >80%, respectively) [11]. Pivotal trials of PEG-IFN/RBV combination therapy showed an approximately 13% lower response rate in patients with cirrhosis than in patients without cirrhosis with an SVR achieved in about 44% of patients with bridging fibrosis/cirrhosis [9, 10, 12]. Although the SVR rate is known to be associated with disease severity [13, 14], the virological response was shown to be associated with a decreased probability of developing esophageal varices, HCC and liver decompensation [15–17]. Moreover, some studies showed that SVR was also associated with a regression of fibrosis/cirrhosis [18–20]. A study by Everson et al. [21] using a low accelerating dosage regimen of IFN or PEG-IFN with RBV in patients with decompensated cirrhosis showed that achieving SVR was highly predictive of maintaining SVR after transplantation. This study also emphasized that intensive treatment was possible and could be successful in this difficult-to-treat patient population. Similarly, another study reported that SVR obtained before transplantation was associated with a lower risk of HCV recurrence [22], and it was shown that the recurrence risk was correlated with the level of HCV RNA before transplantation [23]. Two meta-analyses assessed the potential beneficial effect of IFN therapy on HCC development [24, 25]. The authors observed a slight reduced risk of HCC occurrence and suggested that the risk reduction was probably higher in patients achieving SVR. However, it was shown that the risk of HCC development does not disappear in patients with SVR, strengthening the need for continuous surveillance in these patients [26]. Moreover, in patients with the most severe cirrhosis, it was shown that SVR was associated with a reduced all-cause mortality [17].
Viral Efficacy during Phase III Trials of Triple Therapy
The low SVR rate in genotype 1 and the poor response rates observed with PEG-IFN/RBV in special populations, such as black patients and cirrhotics (25–35%) [27– 30], have driven the development of novel antiviral therapies. The combination of boceprevir or telaprevir – two Bailly/Pradat/Virlogeux/Zoulim
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
Introduction
protease inhibitors (PIs) of the NS3/NS4A of HCV – with PEG-IFN/RBV became the new standard of care at the end of 2011 [31, 32]. In boceprevir-based triple therapy, patients normally receive a 4-week lead-in phase of PEG-IFN/RBV and then 44 weeks of boceprevir associated with PEG-IFN/RBV. The SPRINT-2 phase III clinical trial reported SVR24 rates for previously untreated adults with HCV genotype 1 infection [33]. This study compared 44-week boceprevir-based triple therapy versus PEG-IFN/RBV. Among non-blacks, the SVR24 rate was 68 versus 40% (p < 0.001) and among blacks the rate was 53 versus 23%, respectively. In the overall population, 52% of cirrhotic patients achieved SVR (vs. 38% for the PEG-IFN/RBV). The RESPOND-2 trial reported SVR24 rates for previously treated patients [34]. The overall SVR24 rate for a 44-week boceprevir-based triple therapy was 66 versus 21% for the PEG-IFN/RBV. For previous relapsers, 75 versus 29% achieved SVR24, whereas 52 versus 7% achieved SVR24 among patients with a prior nonresponse. In this trial, cirrhotic patients had the same SVR24 rate than patients with a low Metavir fibrosis score (68 vs. 13% for PEG-IFN/RBV). Viral breakthrough was infrequent during treatment with boceprevir, with rates of 2 and 4% in these two different studies, respectively. In telaprevir-based triple therapy, patients normally receive first a combination of telaprevir with PEG-IFN/ RBV for 12 weeks and then 36 weeks of PEG-IFN/RBV only. The ADVANCE study, a phase III clinical trial for previously untreated patients infected with HCV, reported for telaprevir-based triple therapy an overall SVR24 rate of 73 versus 44% for PEG-IFN/RBV [35]. Black patients had a lower SVR24 rate with 62 versus 25% for PEG-IFN/RBV. Among patients with cirrhosis, only 62 versus 33% for PEG-IFN/RBV achieved SVR24. In this trial, the overall relapse rate was 9 versus 28% and the viral breakthrough rate was only 8 versus 32% compared with the PEG-IFN/RBV regimen. For the retreatment of HCV, the REALIZE trial reported an overall SVR24 rate of 64% for telaprevir-based triple therapy versus 17% for double therapy [36]. Among previous relapsers, 83 versus 24% (PEG-IFN/RBV) achieved SVR24, whereas among previous partial responders only 59 versus 15% (PEG-IFN/RBV) and in previous nonresponders 29 versus 5% (PEG-IFN/RBV) achieved SVR24, respectively. Concerning cirrhotic patients, no clear-cut difference was observed concerning the SVR24 rate compared with patients with a low fibrosis score (83 vs. 24, 44 vs. 10 and 28 vs. 5% compared with the PEG-IFN/RBV regimen for previous relapsers, partial
Viral Efficacy in ‘Real-Life’ Studies The French Compassionate Use of Protease Inhibitors in Viral C Cirrhosis (CUPIC) study was the first ‘real-life’ study that reported viral efficacy and tolerance results among previous nonresponder patients infected by HCV and retreated with triple therapy [38, 39]. This cohort included 511 patients with HCV genotype 1, compensated cirrhosis and with no response to a prior course of PEGIFN/RBV. Boceprevir-based triple therapy was given to 212 patients, whereas 299 patients received telaprevirbased triple therapy. For telaprevir, SVR12 rates were lower than in phase III trials with 74% for previous relapsers/viral breakthrough patients, 40% for previous partial responders and 19% for previous nonresponders. The overall viral breakthrough rate was 16%. Similar differences with phase III trials for boceprevir-based triple therapy were observed with 54% of SVR12 for relapsers/ viral breakthrough patients, 38% for previous partial responders and 0% for previous nonresponders. The overall viral breakthrough rate was 9%. In the international telaprevir access program, which included 1,078 patients with bridging fibrosis (n = 552) or cirrhosis (n = 526) treated by telaprevir-based triple therapy, similar results were observed with 68% SVR24 for naive patients (63% F4), 72% in prior relapsers (64% F4), 55% in partial responders (53% F4) and 34% in nonresponders (29% F4) [40]. The Early Access Program in Spain, involving 170 patients with severe fibrosis, reported an overall SVR12 of 47% for boceprevir-based triple therapy, with 41% for naive patients, 47% for previous partial responders and 23% for previous nonresponders [41]. Other real-life studies are still ongoing, such as the noninterventional PAN study in Germany for boceprevir- and telaprevir-based triple therapy, the Open Label Early Access Program (EAP) for telaprevir-based triple therapy only, and, for boceprevir-based triple therapy, the Italian and Spanish Name Patient Program. Results relating to safety and early virological responses are already available, but results concerning SVR are still awaited [42, 43].
Antiviral Therapy in Patients with HCV-Induced Cirrhosis
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
615
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
responders and nonresponders, respectively). However, the overall viral breakthrough rate was 21 versus 52% (PEG-IFN/RBV). In all these phase III trials, SVR24 was defined as undetectable HCV RNA 24 weeks after the end of treatment. This definition has changed subsequently and HCV RNA undetectability at week 12 after treatment, also called SVR12, is considered as the main treatment endpoint [37].
Prognostic Factors of Virological Response The CUPIC study identified several prognostic factors associated with SVR12 for cirrhotic patients in both PI regimens [39]. Prior response to treatment was identified as a predictor of SVR12 with higher chance of SVR12 for previous relapse/viral breakthrough (OR = 2.96, p < 0.001) and lower chance of SVR12 for prior nonresponse (OR = 0.28, p = 0.009) versus prior partial response. The HCV genotype was also identified as a prognostic factor (subtype 1b vs. 1a, OR = 2.50, p < 0.001) as well as a platelet count greater than 100,000/mm3 (OR = 2.15, p = 616
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
0.002). The absence of a lead-in phase for both treatments was also associated with a greater chance of achieving SVR12 (OR = 1.78, p = 0.004). A nearly significant trend was noted with serum albumin levels of 35 g/l or greater (OR = 1.86, p = 0.061). A recent ‘real-life’ German study found similar associations for the HCV genotype, platelet count and serum albumin level, and also showed a higher risk of experiencing a treatment failure for patients with a Child-Pugh score above 5 (p < 0.05) [50]. The IL28B genotype was also identified as a predictor of SVR12, with a higher probability of response among patients with the IL28B CC genotype (OR = 3.54, p = 0.031). However, since other studies have reported a limited impact of the IL28B genotype on response rates [51], further studies are needed to confirm this association more precisely [47]. In both phase III studies for boceprevir [33, 34] in patients with advanced fibrosis, SVR24 rates in early responders were more than three times higher than in late responders and were comparable to SVR24 rates in patients without advanced liver disease. In the CUPIC study, similar results were observed for boceprevir-treated patients with a ≥1 log10 decrease of HCV RNA levels from baseline at treatment week 4 (the end of lead-in; 62.6 vs. 25.7%, p < 0.001). For telaprevir-based triple therapy patients with a rapid virological response (an undetectable HCV RNA level at treatment week 4 or at treatment week 8 in cases with a lead-in) were also more likely to achieve SVR12 (63.1 vs. 29.7%, p < 0.001) [39]. For cirrhotic patients, triple therapies based on both of the PIs have shown new potential benefits but also major treatment risks. In practice, the risk of developing severe adverse events, including severe complications or death, had to be balanced carefully against the likelihood of a virological response and subsequent improvement of survival. Baseline parameters, such as prior treatment response, serum albumin level, platelet count and HCV genotype, had to be considered to guide treatment decisions at the individual level, and monitoring of the potential side effects, such as hematological side effects, renal impairments and serious adverse events, was fundamental during these therapies.
Treatment of HCV Cirrhosis with IFN-Free Regimens
New direct-acting antivirals (DAA), which specifically target the NS5B RNA-dependent RNA polymerase with either nucleos(t)ide analogues or nonnucleosidic inhibitors, the NS3/4A viral protease and the viral NS5A protein, are now progressively becoming available in develBailly/Pradat/Virlogeux/Zoulim
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
Side Effects Compared with PEG-IFN/RBV treatment, triple therapy was associated in both phase III clinical trials and in ‘real-life’ studies with an increased frequency of side effects. The proportion of treatment discontinuations owing to adverse events ranged from 8 to 16% with boceprevir-based triple therapy, and from 10 to 15% with telaprevir-based triple therapy in the phase III trials in both treatment-naïve and experienced patients [33–36]. In the CUPIC study, the overall incidence rate of adverse events was 40%, with 45 and 33% for the telaprevir and boceprevir groups, respectively [38]. This high rate led to treatment discontinuation in 12% of patients. Severe infections (5%), hepatic decompensation (6%) and death (6 patients) also occurred during this ‘real-life’ study. Moreover, anemia was frequent in both regimens, affecting 30% with telaprevir and 28% with boceprevir, which induced a high rate of EPO intake (54 and 46%, respectively) and blood transfusion in the most severe cases (16 and 6%, respectively). Specific side effects have also been reported, with pruritus for telaprevir in a phase III trial (43%), and dysgeusia for boceprevir (35%) [44, 45]. In the EAP study, similar results were observed for telaprevirbased triple therapy, with anemia rates of 44%, and higher adverse events rates among cirrhotic patients (67 vs. 60%, p = 0.01). A severe grade of anemia was also more often observed in cirrhotic patients than in bridging fibrosis patients (47 vs. 41%, p < 0.01) [42]. Impairment of renal function was also observed in different cohort studies for both PIs. Indeed, a significant decrease in the glomerular filtration rate was observed after the introduction of the PI but a return to normal values occurred after the end of the PI intake [46, 47]. Furthermore, an increase in serum RBV concentration was observed in a group of patients treated with telaprevir [48]. Telaprevir-induced renal dysfunction leads to an early elevation of serum RBV concentration in triple therapy, which could explain the high incidence rate of anemia in these patients [49].
Polymerase (NS5B) Inhibitors This antiviral class is active on viral replication by inhibition of the NS5B RNA-dependent RNA polymerase. Two different classes of NS5B inhibitors have been developed: nucleos(t)ide inhibitors (nucleoside or nucleotide inhibitors), which block the active site of the enzyme, and the nonnucleoside inhibitors, which bind outside the active site. Nucleo(t)side analogues usually have a higher resistance barrier than nonnucleoside analogues. Sofosbuvir, a nucleotide inhibitor, was the first NS5B inhibitor to be licensed for the treatment of HCV infection, and several agents are currently in advanced stages of development. Sofosbuvir has a pan-genotypic activity, is well tolerated and induces no major drug interactions. Sofosbuvir has been tested in association either with PEG-IFN and RBV or with other DAAs. HCV RNA undetectability is obtained in all patients receiving sofosbuvir, usually before 8 weeks of treatment, and all treatment failures are a consequence of a virological relapse after treatment cessation. In vitro resistance to sofosbuvir is described and is associated with the development of an S282T mutation in the NS5B gene. This virological resistance is not identified in the majority of patients who relapse after a sofosbuvir exposure. The nonnucleoside inhibitors currently under development have a lower barrier to resistance and therefore must be administered in combination with other DAAs. NS3/4A Protease Inhibitors The NS3/4A inhibitors target the serine protease NS3/4A and were the first DAAs to be licensed with the first-generation PIs, telaprevir and boceprevir. Simeprevir, which belongs to the first generation but second wave of PIs, has recently been licensed and other compounds, such as asunaprevir, faldaprevir, vaniprevir and ritonavir which boosts ABT-450, are now in phase II or phase III studies. Second-generation PIs which have a broader genotype coverage are in an earlier phase of development. First-generation PIs were only active against HCV genotype 1. Simeprevir is active against genotype 1 and 4, Antiviral Therapy in Patients with HCV-Induced Cirrhosis
and newer agents have activity against other genotypes (2, 4, 5 and 6) but not against genotype 3. Second-generation PIs are usually easier to use with a better tolerance compared with telaprevir and boceprevir. However, drug-todrug interactions remain a problem with these secondgeneration PIs. Simeprevir is a CYP3A4 substrate and a CYP3A4 and CYP1A2 mild inhibitor, and can interact with antiretrovirals, immunosuppressants (including ciclosporin and tacrolimus) and other drugs. Simeprevir has been tested in association with PEG-IFN and RBV, and also with sofosbuvir with or without RBV in the COSMOS study [53]. The association of sofosbuvir and simeprevir with or without RBV over 12 weeks is one of the first IFN-free combinations used for genotype 1 and 4, especially in cirrhotic patients. Resistance mutations in the gene encoding the NS3 protease have been described with telaprevir and boceprevir. The same mutations are frequently associated with a cross resistance to secondgeneration compounds. For simeprevir, a preexisting polymorphism (Q80K) associated with a reduced response to simeprevir in association with PEG-IFN and RBV has been described. The impact of this Q8OK mutation is not observed when simeprevir is used in association with sofosbuvir. NS5A Inhibitors NS5A inhibitors are potent antivirals with a pan-genotypic activity, but with a variable virological response between genotypes or subtypes (1a vs. 1b). Daclatasvir has been recently licensed in Europe. It can be used in association with PEG-IFN and RBV for genotype 1 patients, but has a lower efficacy for the 1a subtype. It is thus more frequently used in association with sofosbuvir with or without RBV for 12–24 weeks. Daclatasvir is well tolerated and induces less serious drug-to-drug interactions than PIs. Dose adjustments are not needed in patients with Child-Pugh B or C disease. Resistance mutations in the NS5A protein have been described in clinical trials. Several other NS5A inhibitors (ledipasvir, ABT-267, GS5816 and MK-4782) are currently being tested in phase II–III studies. Clinical Use of DAAs for Cirrhotic Patients Many studies have tested the efficacy of sofosbuvir, simeprevir and daclatasvir with PEG-IFN and RBV. These 12- to 48-week regimens have shown efficacy for genotype 1, 2, 3 and 4 infections, with SVR rates of 70– 90% depending on the agent used and the patient population. However, the severe side effects frequently observed in cirrhotic patients treated with PEG-IFN, with or withDig Dis 2015;33:613–623 DOI: 10.1159/000375359
617
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
oped countries [52]. The NS5B inhibitor sofosbuvir, the NS3/4A inhibitor simeprevir and the NS5A inhibitor daclatasvir have recently been licensed. For cirrhotic patients, these new drugs improve response rates and allow well-tolerated new therapeutic regimens without IFN or with a shorter exposition to IFN. These regimens have also been successful in difficult-to-treat patients, including previous nonresponders to first-generation PIs and those who have undergone liver transplantation.
Table 1. Treatment of compensated HCV cirrhotic patients with IFN-free regimens Treatment combination
Study name
Genotype (population)
Treatment status
Duration, weeks
SVR12, %
Reference No.
Sofosbuvir + RBV
POSITRON/FUSION VALENCE VALENCE SPARE COSMOS
2 3 3 1 1 1 (15% cirrhotics) 1 (15% cirrhotics) 2 and 3 (20% cirrhotics)
Naive/pretreated Naive Pretreated Ineligible/intolerant IFN Naive/pretreated Naive Pretreated Naive
12 24 24 24 12/24 12 24 24
>90 92 62 38 95 98 98 G2: 92 G3: 89 94 86 82 65
[54 – 56] [56] [56] [57] [53] [59] [59] [59]
Sofosbuvir + ledipasvir Sofosbuvir + ledipasvir Sofosbuvir + ledipasvir + RBV Sofosbuvir + ledipasvir
ION-1
1
Naive
12
ION-2
1
Pretreated
12
ELECTRON 2
Naive
12
Asunaprevir + daclatasvir
HALLMARK-DUAL
1 (cirrhotics, Child-Pugh B) 1b
Asunaprevir + daclatasvir + BMS-791325 Boosted ABT-450 + ombitasvir + dasabuvir TURQUOISE-II MK-5172 + MK-8742 C-WORTHY
Naive Pretreated Ineligible/intolerant IFN 1a/1b (9% cirrhotics) Naive 1 Naive/pretreated 1 Naive/pretreated
out PIs, encourage avoiding the use of IFN for these patients. In this context, one of the most promising benefits of newer DAAs is the development of potent IFN-free regimens. Some of the major studies of DAAs without IFN have included compensated cirrhotic patients. Results are now also available for more severe patients, such as those with decompensated cirrhosis, patients awaiting liver transplantation or with HCV graft reinfection after liver transplantation. We will describe here the results of the different IFN-free regimens tested for genotype 1, 2, 3 and 4 cirrhotic patients (table 1). Sofosbuvir with RBV Sofosbuvir has recently been approved in combination with RBV for the treatment of genotype 2 infection (12 weeks) or genotype 3 infection (24 weeks). For genotype 2 patients, all studies show an SVR rate over 90% after 12 weeks of sofosbuvir/RBV combination. The presence of cirrhosis or failure to a previous treatment has not been shown to impact significantly on this response rate in these studies. This short regimen is well tolerated and is now the first therapeutic option for genotype 2 patients with or without cirrhosis. For genotype 3, this regimen for 12 weeks is associated with a lower response rate of 63%. The presence of cir618
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
24
12 12/24 12/18
91 87 79 90 92/96 90/97
[60] [61] [62] [63]
[64] [65] [66]
rhosis is associated with a very low response rate of 19– 21%. In the Valence study, genotype 3 patients were treated with a 24-week regimen. The SVR rate was 95, 92, 87 and 62% in treatment-naive patients without cirrhosis, treatment-naive patients with cirrhosis, treatment-experienced patients without cirrhosis, and treatment-experienced patients with cirrhosis, respectively [54–56]. These data suggest that the combination of sofosbuvir plus RBV, for 12 or 24 weeks, is not a good therapeutic option for genotype 3-infected patients with cirrhosis. Treatment with sofosbuvir and RBV for 24 weeks has also been proposed for genotype 1-infected patients who are intolerant or contraindicated for IFN treatment. In a phase II study of 60 treatment-naive patients with genotype 1 infection, 24 weeks of sofosbuvir and weight-based RBV led to an SVR rate of 68%. However, extensive fibrosis and cirrhosis were associated with a high frequency of relapse, and 8 patients among 13 with F3 or F4 fibrosis were relapsers [57]. Sofosbuvir with Simeprevir The COSMOS study (cohort 2) recently included 87 genotype 1 treatment-naive or previous nonresponder patients with advanced fibrosis (F3–F4). Patients were treated with simeprevir and sofosbuvir, with or without RBV, for 12 or 24 weeks. An overall SVR12 rate of 94% Bailly/Pradat/Virlogeux/Zoulim
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
Sofosbuvir + simeprevir ± RBV Sofosbuvir + daclatasvir
in 82% of patients compared to 100% in noncirrhotics after 12 weeks of sofosbuvir/ledipasvir/RBV combination [61]. Recently, a phase 2 study of sofosbuvir/ledipasvir with or without RBV for 12 weeks included 80 difficult-totreat patients: 19 genotype 1 patients with prior sofosbuvir relapse, 20 genotype 1 patients with decompensated cirrhosis and 51 treatment-naive genotype 3 patients. A 100% SVR rate was observed for sofosbuvir-exposed patients and genotype 3 patients treated with sofosbuvir/ ledipasvir /RBV. However, the response rate was lower for decompensated cirrhotic patients and for genotype 3 patients who were treated without RBV (65 and 64%, respectively) [62].
Sofosbuvir with NS5A Inhibitor Sofosbuvir in combination with daclatasvir has been tested in a study of 167 genotype 1 (126 treatment-naive patients and 41 previously treated with telaprevir or boceprevir) and 44 genotype 2- and 3- (treatment-naive) infected patients. Patients were treated with daclatasvir and sofosbuvir with or without RBV for 12 or 24 weeks (16 genotype 2 and 15 genotype 1 patients initially received sofosbuvir monotherapy for 1 week, all previously treated patients were treated for 24 weeks). Among patients with genotype 1 infection, 98% of 126 previously untreated patients and 98% of 41 patients who did not respond to HCV PIs achieved SVR12. Ninety-two percent of 26 patients with genotype 2 infection and 89% of 18 patients with genotype 3 infection had an SVR12. The response rate was similar with or without RBV and for naive patients after 12 or 24 weeks of treatment. This study only constituted 15% compensated cirrhotic patients [59]. Ledipasvir, another NS5A inhibitor, has been developed in a fixed dose combination with sofosbuvir and will be licensed imminently. The phase 3 study included cirrhotic genotype 1 and 3 patients. In the ION-1 study, 865 untreated genotype 1 patients received a sofosbuvir-ledipasvir combination with or without RBV for 12–24 weeks. Overall, SVR rates were high at about 97–99%. Adding RBV and/or extending the treatment duration from 12 to 24 weeks did not increase SVR12. Patients with compensated cirrhosis (16%) had an SVR rate of 94% after 12 weeks of treatment [60]. The ION-2 study included 440 genotype 1 previously treated patients who were also treated with sofosbuvirledipasvir, with or without RBV, for 12–24 weeks. High SVR rates (94–99%) were observed across all arms. Twenty percent of the patients had cirrhosis at baseline and among these cirrhotic patients an SVR12 was observed in 86% compared to 95% in noncirrhotic patients after 12 weeks of sofosbuvir/ledipasvir combination, and
Other IFN-Free Combinations Asunaprevir is an NS3/4A PI and several studies have tested its efficacy in association with daclatasvir for genotype 1 patients. After 24 weeks of this combination, without RBV, the virological response is high for genotype 1b (71%) but very low for genotype 1a (5%). The phase 3 HALLMARK-DUAL study included 745 genotype 1b patients treated with asunaprevir and daclatasvir for 24 weeks. Among them, 305 were treatment naive, 205 were nonresponders to PEG-IFN/RBV, and 235 were ineligible or intolerant to IFN. Forty-seven percent of the IFN-ineligible patients and 31% of the nonresponder patients were cirrhotic. SVR12 rates were 91% for treatment-naive patients, 82% for nonresponder patients and 83% for ineligible patients. The response rates were similar in noncirrhotic (85%) and cirrhotic (84%) patients [63]. This combination is now being tested as a triple therapy, with the addition of a NS5B nonnucleoside inhibitor (BMS791325) in HCV genotype 1a and 1b, with more than 90% of 166 treatment-naive patients (9% cirrhotic patients) achieving SVR after 12 weeks of triple therapy [64]. A combination of the NS3/4A PI ABT-450 boosted by a low dose of ritonavir, the NS5A inhibitor ombitasvir (ABT-267) and the nonnucleoside NS5B inhibitor dasabuvir (ABT-333) have been tested in several studies. In the TURQUOISE-II study this three-drug combination with RBV is used in patients with genotype 1 infection and cirrhosis. A total of 380 patients were included and randomized for a 12- or 24-week treatment. SVR rates were 92% in the 12-week arm compared with 96% in the 24-week group. SVR rates did not differ significantly between the two treatment durations. However, previous nonresponder genotype 1a patients seemed to benefit from a 24-week treatment duration. This new combination, which could be available during the next year, was
Antiviral Therapy in Patients with HCV-Induced Cirrhosis
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
619
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
was achieved. The response rate was not significantly different between F3 (98%) and F4 (95%) patients. The 12week regimen was as effective as the 24-week regimen, and the addition of RBV did not seem to improve response rates [53]. The combination of sofosbuvir and simeprevir was well tolerated in cirrhotic patients. Only genotype 1 patients were included in the COSMOS study, but both sofosbuvir and simeprevir have shown efficacy against genotype 4 viruses. On this basis, the combination of sofosbuvir and simeprevir could also be a therapeutic option for genotype 4 infections, as suggested by EASL recommendations [58].
Patients with Decompensated Cirrhosis or Awaiting Liver Transplantation Decompensated cirrhotic patients are usually excluded from phase 2 or 3 studies and very limited data are available regarding the tolerance and efficacy of new IFNfree regimens in such patients. The results of a randomized, safety and efficacy study of sofosbuvir plus RBV for 48 weeks compared to observation for 6 months in patients with HCV cirrhosis and portal hypertension (ChildTurcotte-Pugh scores 5–9) were recently presented during the EASL meeting. Fifty cirrhotic genotype 1, 2, 3 or 4 patients were included, among whom more than 70% had significant portal hypertension (hepatic venous pressure gradient >12 mm Hg). Sofosbuvir plus RBV for up to 24 weeks resulted in a high rate of virological suppression (93–100% at week 24) irrespective of the severity of liver disease. This rapid viral efficacy was associated with a decreased necroinflammation with ALT normalization, and also with improvements in platelet count, albumin, ascites and hepatic encephalopathy. The treatment was well tolerated with a low rate of treatment discontinuation due to adverse events. No patients with worsening or new hepatic decompensation were observed [67]. Posttreatment virological response data are now awaited. Cirrhotic patients with detectable HCV RNA at the time of liver transplantation universally experience recurrent HCV infection. Antiviral treatment before transplantation can prevent HCV recurrence, but existing IFN-based regimens are poorly tolerated and frequently contraindicated or ineffective [68]. In a phase 2 study, 61 HCV cirrhotic patients awaiting liver transplantation for HCC received up to 48 weeks of sofosbuvir and RBV before liver transplantation. Forty-six were transplanted. Among them, 43 patients had HCV RNA
Antiviral Therapy in Patients with Hepatitis C Virus-Induced Cirrhosis François Bailly a–c Pierre Pradat a–c Victor Virlogeux a, d Fabien Zoulim a–c c
Hepatology Department, Hospices Civils de Lyon, b INSERM Unit 1052, Cancer Research Center of Lyon (CRCL), Lyon University, and d Ecole Normale Supérieure, Lyon, France
Key Words Hepatitis C virus · Chronic hepatitis C · Cirrhosis · Antivirals · Therapy
Abstract Background: Opportunities to treat infection with hepatitis C virus (HCV) are evolving rapidly. From the introduction of interferon (IFN)-α monotherapy in the early 1990s to the approval of telaprevir- and boceprevir-based triple therapies with pegylated (PEG)-IFN-α and ribavirin (RBV) in 2011, the chances of curing patients infected with HCV genotype 1 have improved dramatically to reach approximately 70%. Significant further improvements that may cure virtually all HCV patients with an all-oral, IFN-free regimen are becoming progressively available. Key Messages: Historically, a PEGIFN/RBV combination therapy of patients with liver cirrhosis was associated with lower virological rates and a worse safety profile. The advent of the first protease inhibitor-based triple therapy was long expected, but the promise fell rapidly because of the numerous side effects and the requirement for intensive clinical management in cirrhotic patients. The newer direct-acting antivirals (DAAs) target the viral polymerase with either nucleos(t)ide analogues or nonnucleosidic inhibitors, the viral protease and the viral NS5A protein. Several clinical trials have now shown that a combination of sofosbuvir (nucleosidic polymerase inhibitor) with
© 2015 S. Karger AG, Basel 0257–2753/15/0334–0613$39.50/0 E-Mail [email protected] www.karger.com/ddi
daclatasvir or ledipasvir (NS5A inhibitors), or sofosbuvir with simeprevir (protease inhibitor), or a combination of ABT-450 (protease inhibitor) with ritonavir (ABT-450/r), the nonnucleosidic polymerase inhibitor ABT-333 and the NS5A inhibitor ABT-267, can achieve a sustained virological response in up to 95% of naive patients or previously treated patients, even in those who failed prior treatment with first-generation protease inhibitors. The best treatment regimens enable the achievement of comparable results even in cirrhotics, while other regimens still require RBV or a longer treatment duration to achieve optimal results. This improved risk/benefit ratio justifies early access programs of IFN-free regimens for cirrhotic patients. The remaining difficult-to-treat patients are cirrhotics infected with HCV genotype 3 and those with decompensated cirrhosis, for whom novel DAA combinations should be evaluated in clinical trials. Conclusions: As new DAAs are becoming available in early access treatment programs, treatment strategy studies are being performed to optimize treatment regimens with respect to the choice of DAAs and treatment duration, based on viral genotypes, prior treatment response and the presence of liver cirrhosis. In the near future, this should allow: (i) a decrease in the complications of HCV-induced cirrhosis, (ii) liver transplantations to be performed in virally cured patients, and (iii) the rescue of patients in the worst clinical situation (decompensated cirrhosis and HCV recurrence on liver graft). © 2015 S. Karger AG, Basel
Fabien Zoulim Department of Hepatology, Croix-Rousse Hospital 103, Grande rue de la Croix-Rousse FR–69317 Lyon Cedex 04 (France) E-Mail fabien.zoulim @ inserm.fr
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
a
Hepatitis C virus (HCV) infection is a leading cause of liver cirrhosis and hepatocellular carcinoma (HCC) [1] and is the most common indication for liver transplantation in Western countries [2]. Disease progression is generally slow with cirrhosis developing several decades after infection. However, substantial variations exist between individuals depending on the presence/absence of cofactors such as alcohol abuse, hepatitis B virus or HIV infection, metabolic disorders and genetic polymorphisms, etc. Cirrhosis is highly heterogeneous with a continuous spectrum of conditions ranging from early (compensated) to more severe (decompensated) cirrhosis and terminal end-stage liver disease. Compensated cirrhosis is also heterogeneous with more severe forms observed in cases of portal hypertension, esophageal varices, thrombo- or leucopenia, or low albumin levels [3]. The need for therapy increases with disease severity, which in turn is associated with a lower probability of virological response. The objectives of HCV therapy in patients with cirrhosis differ according to disease severity. In patients with compensated cirrhosis, treatment may slow down disease progression and prevent decompensation [4]. In patients with cirrhosis decompensation awaiting liver transplantation, viral eradication before transplantation may reduce the risk of HCV recurrence posttransplantation. Finally, another main objective of HCV therapy in cirrhotic patients is to reduce the risk of HCC development.
Interferon-Ribavirin Combination in Patients with HCV-Induced Cirrhosis
Antiviral treatment of HCV-induced compensated cirrhosis started with standard interferon (IFN), which gave controversial results. Valla et al. [5] found no beneficial effect on the development of decompensation or HCC, whereas Poynard et al. [6] suggested that IFN therapy could reduce the incidence of and mortality from HCC. In 2000, Heathcote et al. [7] showed in patients with cirrhosis that pegylated (PEG)-IFN once weekly was more effective for achieving a sustained virological response (SVR) compared with standard IFN (43 vs. 15%). Almost a decade later, results from the HALT-C trial showed, however, that prolonged therapy with a low PEG-IFN dose in HCV patients with a nonresponse to a previous course of PEG-IFN/ribavirin (RBV) and with 614
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
advanced fibrosis, with or without cirrhosis, did not reduce the rate of disease progression [8]. From 2000 to 2011, the standard of care for HCV treatment was a combination of PEG-IFN/RBV, allowing an SVR rate of about 50% in patients with HCV genotype 1 [9, 10]. Response to therapy varied according to genotype, with genotype 1 and 4 patients achieving lower SVR rates than patients infected with genotype 2 and 3 (40–50 vs. >80%, respectively) [11]. Pivotal trials of PEG-IFN/RBV combination therapy showed an approximately 13% lower response rate in patients with cirrhosis than in patients without cirrhosis with an SVR achieved in about 44% of patients with bridging fibrosis/cirrhosis [9, 10, 12]. Although the SVR rate is known to be associated with disease severity [13, 14], the virological response was shown to be associated with a decreased probability of developing esophageal varices, HCC and liver decompensation [15–17]. Moreover, some studies showed that SVR was also associated with a regression of fibrosis/cirrhosis [18–20]. A study by Everson et al. [21] using a low accelerating dosage regimen of IFN or PEG-IFN with RBV in patients with decompensated cirrhosis showed that achieving SVR was highly predictive of maintaining SVR after transplantation. This study also emphasized that intensive treatment was possible and could be successful in this difficult-to-treat patient population. Similarly, another study reported that SVR obtained before transplantation was associated with a lower risk of HCV recurrence [22], and it was shown that the recurrence risk was correlated with the level of HCV RNA before transplantation [23]. Two meta-analyses assessed the potential beneficial effect of IFN therapy on HCC development [24, 25]. The authors observed a slight reduced risk of HCC occurrence and suggested that the risk reduction was probably higher in patients achieving SVR. However, it was shown that the risk of HCC development does not disappear in patients with SVR, strengthening the need for continuous surveillance in these patients [26]. Moreover, in patients with the most severe cirrhosis, it was shown that SVR was associated with a reduced all-cause mortality [17].
Viral Efficacy during Phase III Trials of Triple Therapy
The low SVR rate in genotype 1 and the poor response rates observed with PEG-IFN/RBV in special populations, such as black patients and cirrhotics (25–35%) [27– 30], have driven the development of novel antiviral therapies. The combination of boceprevir or telaprevir – two Bailly/Pradat/Virlogeux/Zoulim
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
Introduction
protease inhibitors (PIs) of the NS3/NS4A of HCV – with PEG-IFN/RBV became the new standard of care at the end of 2011 [31, 32]. In boceprevir-based triple therapy, patients normally receive a 4-week lead-in phase of PEG-IFN/RBV and then 44 weeks of boceprevir associated with PEG-IFN/RBV. The SPRINT-2 phase III clinical trial reported SVR24 rates for previously untreated adults with HCV genotype 1 infection [33]. This study compared 44-week boceprevir-based triple therapy versus PEG-IFN/RBV. Among non-blacks, the SVR24 rate was 68 versus 40% (p < 0.001) and among blacks the rate was 53 versus 23%, respectively. In the overall population, 52% of cirrhotic patients achieved SVR (vs. 38% for the PEG-IFN/RBV). The RESPOND-2 trial reported SVR24 rates for previously treated patients [34]. The overall SVR24 rate for a 44-week boceprevir-based triple therapy was 66 versus 21% for the PEG-IFN/RBV. For previous relapsers, 75 versus 29% achieved SVR24, whereas 52 versus 7% achieved SVR24 among patients with a prior nonresponse. In this trial, cirrhotic patients had the same SVR24 rate than patients with a low Metavir fibrosis score (68 vs. 13% for PEG-IFN/RBV). Viral breakthrough was infrequent during treatment with boceprevir, with rates of 2 and 4% in these two different studies, respectively. In telaprevir-based triple therapy, patients normally receive first a combination of telaprevir with PEG-IFN/ RBV for 12 weeks and then 36 weeks of PEG-IFN/RBV only. The ADVANCE study, a phase III clinical trial for previously untreated patients infected with HCV, reported for telaprevir-based triple therapy an overall SVR24 rate of 73 versus 44% for PEG-IFN/RBV [35]. Black patients had a lower SVR24 rate with 62 versus 25% for PEG-IFN/RBV. Among patients with cirrhosis, only 62 versus 33% for PEG-IFN/RBV achieved SVR24. In this trial, the overall relapse rate was 9 versus 28% and the viral breakthrough rate was only 8 versus 32% compared with the PEG-IFN/RBV regimen. For the retreatment of HCV, the REALIZE trial reported an overall SVR24 rate of 64% for telaprevir-based triple therapy versus 17% for double therapy [36]. Among previous relapsers, 83 versus 24% (PEG-IFN/RBV) achieved SVR24, whereas among previous partial responders only 59 versus 15% (PEG-IFN/RBV) and in previous nonresponders 29 versus 5% (PEG-IFN/RBV) achieved SVR24, respectively. Concerning cirrhotic patients, no clear-cut difference was observed concerning the SVR24 rate compared with patients with a low fibrosis score (83 vs. 24, 44 vs. 10 and 28 vs. 5% compared with the PEG-IFN/RBV regimen for previous relapsers, partial
Viral Efficacy in ‘Real-Life’ Studies The French Compassionate Use of Protease Inhibitors in Viral C Cirrhosis (CUPIC) study was the first ‘real-life’ study that reported viral efficacy and tolerance results among previous nonresponder patients infected by HCV and retreated with triple therapy [38, 39]. This cohort included 511 patients with HCV genotype 1, compensated cirrhosis and with no response to a prior course of PEGIFN/RBV. Boceprevir-based triple therapy was given to 212 patients, whereas 299 patients received telaprevirbased triple therapy. For telaprevir, SVR12 rates were lower than in phase III trials with 74% for previous relapsers/viral breakthrough patients, 40% for previous partial responders and 19% for previous nonresponders. The overall viral breakthrough rate was 16%. Similar differences with phase III trials for boceprevir-based triple therapy were observed with 54% of SVR12 for relapsers/ viral breakthrough patients, 38% for previous partial responders and 0% for previous nonresponders. The overall viral breakthrough rate was 9%. In the international telaprevir access program, which included 1,078 patients with bridging fibrosis (n = 552) or cirrhosis (n = 526) treated by telaprevir-based triple therapy, similar results were observed with 68% SVR24 for naive patients (63% F4), 72% in prior relapsers (64% F4), 55% in partial responders (53% F4) and 34% in nonresponders (29% F4) [40]. The Early Access Program in Spain, involving 170 patients with severe fibrosis, reported an overall SVR12 of 47% for boceprevir-based triple therapy, with 41% for naive patients, 47% for previous partial responders and 23% for previous nonresponders [41]. Other real-life studies are still ongoing, such as the noninterventional PAN study in Germany for boceprevir- and telaprevir-based triple therapy, the Open Label Early Access Program (EAP) for telaprevir-based triple therapy only, and, for boceprevir-based triple therapy, the Italian and Spanish Name Patient Program. Results relating to safety and early virological responses are already available, but results concerning SVR are still awaited [42, 43].
Antiviral Therapy in Patients with HCV-Induced Cirrhosis
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
615
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
responders and nonresponders, respectively). However, the overall viral breakthrough rate was 21 versus 52% (PEG-IFN/RBV). In all these phase III trials, SVR24 was defined as undetectable HCV RNA 24 weeks after the end of treatment. This definition has changed subsequently and HCV RNA undetectability at week 12 after treatment, also called SVR12, is considered as the main treatment endpoint [37].
Prognostic Factors of Virological Response The CUPIC study identified several prognostic factors associated with SVR12 for cirrhotic patients in both PI regimens [39]. Prior response to treatment was identified as a predictor of SVR12 with higher chance of SVR12 for previous relapse/viral breakthrough (OR = 2.96, p < 0.001) and lower chance of SVR12 for prior nonresponse (OR = 0.28, p = 0.009) versus prior partial response. The HCV genotype was also identified as a prognostic factor (subtype 1b vs. 1a, OR = 2.50, p < 0.001) as well as a platelet count greater than 100,000/mm3 (OR = 2.15, p = 616
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
0.002). The absence of a lead-in phase for both treatments was also associated with a greater chance of achieving SVR12 (OR = 1.78, p = 0.004). A nearly significant trend was noted with serum albumin levels of 35 g/l or greater (OR = 1.86, p = 0.061). A recent ‘real-life’ German study found similar associations for the HCV genotype, platelet count and serum albumin level, and also showed a higher risk of experiencing a treatment failure for patients with a Child-Pugh score above 5 (p < 0.05) [50]. The IL28B genotype was also identified as a predictor of SVR12, with a higher probability of response among patients with the IL28B CC genotype (OR = 3.54, p = 0.031). However, since other studies have reported a limited impact of the IL28B genotype on response rates [51], further studies are needed to confirm this association more precisely [47]. In both phase III studies for boceprevir [33, 34] in patients with advanced fibrosis, SVR24 rates in early responders were more than three times higher than in late responders and were comparable to SVR24 rates in patients without advanced liver disease. In the CUPIC study, similar results were observed for boceprevir-treated patients with a ≥1 log10 decrease of HCV RNA levels from baseline at treatment week 4 (the end of lead-in; 62.6 vs. 25.7%, p < 0.001). For telaprevir-based triple therapy patients with a rapid virological response (an undetectable HCV RNA level at treatment week 4 or at treatment week 8 in cases with a lead-in) were also more likely to achieve SVR12 (63.1 vs. 29.7%, p < 0.001) [39]. For cirrhotic patients, triple therapies based on both of the PIs have shown new potential benefits but also major treatment risks. In practice, the risk of developing severe adverse events, including severe complications or death, had to be balanced carefully against the likelihood of a virological response and subsequent improvement of survival. Baseline parameters, such as prior treatment response, serum albumin level, platelet count and HCV genotype, had to be considered to guide treatment decisions at the individual level, and monitoring of the potential side effects, such as hematological side effects, renal impairments and serious adverse events, was fundamental during these therapies.
Treatment of HCV Cirrhosis with IFN-Free Regimens
New direct-acting antivirals (DAA), which specifically target the NS5B RNA-dependent RNA polymerase with either nucleos(t)ide analogues or nonnucleosidic inhibitors, the NS3/4A viral protease and the viral NS5A protein, are now progressively becoming available in develBailly/Pradat/Virlogeux/Zoulim
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
Side Effects Compared with PEG-IFN/RBV treatment, triple therapy was associated in both phase III clinical trials and in ‘real-life’ studies with an increased frequency of side effects. The proportion of treatment discontinuations owing to adverse events ranged from 8 to 16% with boceprevir-based triple therapy, and from 10 to 15% with telaprevir-based triple therapy in the phase III trials in both treatment-naïve and experienced patients [33–36]. In the CUPIC study, the overall incidence rate of adverse events was 40%, with 45 and 33% for the telaprevir and boceprevir groups, respectively [38]. This high rate led to treatment discontinuation in 12% of patients. Severe infections (5%), hepatic decompensation (6%) and death (6 patients) also occurred during this ‘real-life’ study. Moreover, anemia was frequent in both regimens, affecting 30% with telaprevir and 28% with boceprevir, which induced a high rate of EPO intake (54 and 46%, respectively) and blood transfusion in the most severe cases (16 and 6%, respectively). Specific side effects have also been reported, with pruritus for telaprevir in a phase III trial (43%), and dysgeusia for boceprevir (35%) [44, 45]. In the EAP study, similar results were observed for telaprevirbased triple therapy, with anemia rates of 44%, and higher adverse events rates among cirrhotic patients (67 vs. 60%, p = 0.01). A severe grade of anemia was also more often observed in cirrhotic patients than in bridging fibrosis patients (47 vs. 41%, p < 0.01) [42]. Impairment of renal function was also observed in different cohort studies for both PIs. Indeed, a significant decrease in the glomerular filtration rate was observed after the introduction of the PI but a return to normal values occurred after the end of the PI intake [46, 47]. Furthermore, an increase in serum RBV concentration was observed in a group of patients treated with telaprevir [48]. Telaprevir-induced renal dysfunction leads to an early elevation of serum RBV concentration in triple therapy, which could explain the high incidence rate of anemia in these patients [49].
Polymerase (NS5B) Inhibitors This antiviral class is active on viral replication by inhibition of the NS5B RNA-dependent RNA polymerase. Two different classes of NS5B inhibitors have been developed: nucleos(t)ide inhibitors (nucleoside or nucleotide inhibitors), which block the active site of the enzyme, and the nonnucleoside inhibitors, which bind outside the active site. Nucleo(t)side analogues usually have a higher resistance barrier than nonnucleoside analogues. Sofosbuvir, a nucleotide inhibitor, was the first NS5B inhibitor to be licensed for the treatment of HCV infection, and several agents are currently in advanced stages of development. Sofosbuvir has a pan-genotypic activity, is well tolerated and induces no major drug interactions. Sofosbuvir has been tested in association either with PEG-IFN and RBV or with other DAAs. HCV RNA undetectability is obtained in all patients receiving sofosbuvir, usually before 8 weeks of treatment, and all treatment failures are a consequence of a virological relapse after treatment cessation. In vitro resistance to sofosbuvir is described and is associated with the development of an S282T mutation in the NS5B gene. This virological resistance is not identified in the majority of patients who relapse after a sofosbuvir exposure. The nonnucleoside inhibitors currently under development have a lower barrier to resistance and therefore must be administered in combination with other DAAs. NS3/4A Protease Inhibitors The NS3/4A inhibitors target the serine protease NS3/4A and were the first DAAs to be licensed with the first-generation PIs, telaprevir and boceprevir. Simeprevir, which belongs to the first generation but second wave of PIs, has recently been licensed and other compounds, such as asunaprevir, faldaprevir, vaniprevir and ritonavir which boosts ABT-450, are now in phase II or phase III studies. Second-generation PIs which have a broader genotype coverage are in an earlier phase of development. First-generation PIs were only active against HCV genotype 1. Simeprevir is active against genotype 1 and 4, Antiviral Therapy in Patients with HCV-Induced Cirrhosis
and newer agents have activity against other genotypes (2, 4, 5 and 6) but not against genotype 3. Second-generation PIs are usually easier to use with a better tolerance compared with telaprevir and boceprevir. However, drug-todrug interactions remain a problem with these secondgeneration PIs. Simeprevir is a CYP3A4 substrate and a CYP3A4 and CYP1A2 mild inhibitor, and can interact with antiretrovirals, immunosuppressants (including ciclosporin and tacrolimus) and other drugs. Simeprevir has been tested in association with PEG-IFN and RBV, and also with sofosbuvir with or without RBV in the COSMOS study [53]. The association of sofosbuvir and simeprevir with or without RBV over 12 weeks is one of the first IFN-free combinations used for genotype 1 and 4, especially in cirrhotic patients. Resistance mutations in the gene encoding the NS3 protease have been described with telaprevir and boceprevir. The same mutations are frequently associated with a cross resistance to secondgeneration compounds. For simeprevir, a preexisting polymorphism (Q80K) associated with a reduced response to simeprevir in association with PEG-IFN and RBV has been described. The impact of this Q8OK mutation is not observed when simeprevir is used in association with sofosbuvir. NS5A Inhibitors NS5A inhibitors are potent antivirals with a pan-genotypic activity, but with a variable virological response between genotypes or subtypes (1a vs. 1b). Daclatasvir has been recently licensed in Europe. It can be used in association with PEG-IFN and RBV for genotype 1 patients, but has a lower efficacy for the 1a subtype. It is thus more frequently used in association with sofosbuvir with or without RBV for 12–24 weeks. Daclatasvir is well tolerated and induces less serious drug-to-drug interactions than PIs. Dose adjustments are not needed in patients with Child-Pugh B or C disease. Resistance mutations in the NS5A protein have been described in clinical trials. Several other NS5A inhibitors (ledipasvir, ABT-267, GS5816 and MK-4782) are currently being tested in phase II–III studies. Clinical Use of DAAs for Cirrhotic Patients Many studies have tested the efficacy of sofosbuvir, simeprevir and daclatasvir with PEG-IFN and RBV. These 12- to 48-week regimens have shown efficacy for genotype 1, 2, 3 and 4 infections, with SVR rates of 70– 90% depending on the agent used and the patient population. However, the severe side effects frequently observed in cirrhotic patients treated with PEG-IFN, with or withDig Dis 2015;33:613–623 DOI: 10.1159/000375359
617
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
oped countries [52]. The NS5B inhibitor sofosbuvir, the NS3/4A inhibitor simeprevir and the NS5A inhibitor daclatasvir have recently been licensed. For cirrhotic patients, these new drugs improve response rates and allow well-tolerated new therapeutic regimens without IFN or with a shorter exposition to IFN. These regimens have also been successful in difficult-to-treat patients, including previous nonresponders to first-generation PIs and those who have undergone liver transplantation.
Table 1. Treatment of compensated HCV cirrhotic patients with IFN-free regimens Treatment combination
Study name
Genotype (population)
Treatment status
Duration, weeks
SVR12, %
Reference No.
Sofosbuvir + RBV
POSITRON/FUSION VALENCE VALENCE SPARE COSMOS
2 3 3 1 1 1 (15% cirrhotics) 1 (15% cirrhotics) 2 and 3 (20% cirrhotics)
Naive/pretreated Naive Pretreated Ineligible/intolerant IFN Naive/pretreated Naive Pretreated Naive
12 24 24 24 12/24 12 24 24
>90 92 62 38 95 98 98 G2: 92 G3: 89 94 86 82 65
[54 – 56] [56] [56] [57] [53] [59] [59] [59]
Sofosbuvir + ledipasvir Sofosbuvir + ledipasvir Sofosbuvir + ledipasvir + RBV Sofosbuvir + ledipasvir
ION-1
1
Naive
12
ION-2
1
Pretreated
12
ELECTRON 2
Naive
12
Asunaprevir + daclatasvir
HALLMARK-DUAL
1 (cirrhotics, Child-Pugh B) 1b
Asunaprevir + daclatasvir + BMS-791325 Boosted ABT-450 + ombitasvir + dasabuvir TURQUOISE-II MK-5172 + MK-8742 C-WORTHY
Naive Pretreated Ineligible/intolerant IFN 1a/1b (9% cirrhotics) Naive 1 Naive/pretreated 1 Naive/pretreated
out PIs, encourage avoiding the use of IFN for these patients. In this context, one of the most promising benefits of newer DAAs is the development of potent IFN-free regimens. Some of the major studies of DAAs without IFN have included compensated cirrhotic patients. Results are now also available for more severe patients, such as those with decompensated cirrhosis, patients awaiting liver transplantation or with HCV graft reinfection after liver transplantation. We will describe here the results of the different IFN-free regimens tested for genotype 1, 2, 3 and 4 cirrhotic patients (table 1). Sofosbuvir with RBV Sofosbuvir has recently been approved in combination with RBV for the treatment of genotype 2 infection (12 weeks) or genotype 3 infection (24 weeks). For genotype 2 patients, all studies show an SVR rate over 90% after 12 weeks of sofosbuvir/RBV combination. The presence of cirrhosis or failure to a previous treatment has not been shown to impact significantly on this response rate in these studies. This short regimen is well tolerated and is now the first therapeutic option for genotype 2 patients with or without cirrhosis. For genotype 3, this regimen for 12 weeks is associated with a lower response rate of 63%. The presence of cir618
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
24
12 12/24 12/18
91 87 79 90 92/96 90/97
[60] [61] [62] [63]
[64] [65] [66]
rhosis is associated with a very low response rate of 19– 21%. In the Valence study, genotype 3 patients were treated with a 24-week regimen. The SVR rate was 95, 92, 87 and 62% in treatment-naive patients without cirrhosis, treatment-naive patients with cirrhosis, treatment-experienced patients without cirrhosis, and treatment-experienced patients with cirrhosis, respectively [54–56]. These data suggest that the combination of sofosbuvir plus RBV, for 12 or 24 weeks, is not a good therapeutic option for genotype 3-infected patients with cirrhosis. Treatment with sofosbuvir and RBV for 24 weeks has also been proposed for genotype 1-infected patients who are intolerant or contraindicated for IFN treatment. In a phase II study of 60 treatment-naive patients with genotype 1 infection, 24 weeks of sofosbuvir and weight-based RBV led to an SVR rate of 68%. However, extensive fibrosis and cirrhosis were associated with a high frequency of relapse, and 8 patients among 13 with F3 or F4 fibrosis were relapsers [57]. Sofosbuvir with Simeprevir The COSMOS study (cohort 2) recently included 87 genotype 1 treatment-naive or previous nonresponder patients with advanced fibrosis (F3–F4). Patients were treated with simeprevir and sofosbuvir, with or without RBV, for 12 or 24 weeks. An overall SVR12 rate of 94% Bailly/Pradat/Virlogeux/Zoulim
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
Sofosbuvir + simeprevir ± RBV Sofosbuvir + daclatasvir
in 82% of patients compared to 100% in noncirrhotics after 12 weeks of sofosbuvir/ledipasvir/RBV combination [61]. Recently, a phase 2 study of sofosbuvir/ledipasvir with or without RBV for 12 weeks included 80 difficult-totreat patients: 19 genotype 1 patients with prior sofosbuvir relapse, 20 genotype 1 patients with decompensated cirrhosis and 51 treatment-naive genotype 3 patients. A 100% SVR rate was observed for sofosbuvir-exposed patients and genotype 3 patients treated with sofosbuvir/ ledipasvir /RBV. However, the response rate was lower for decompensated cirrhotic patients and for genotype 3 patients who were treated without RBV (65 and 64%, respectively) [62].
Sofosbuvir with NS5A Inhibitor Sofosbuvir in combination with daclatasvir has been tested in a study of 167 genotype 1 (126 treatment-naive patients and 41 previously treated with telaprevir or boceprevir) and 44 genotype 2- and 3- (treatment-naive) infected patients. Patients were treated with daclatasvir and sofosbuvir with or without RBV for 12 or 24 weeks (16 genotype 2 and 15 genotype 1 patients initially received sofosbuvir monotherapy for 1 week, all previously treated patients were treated for 24 weeks). Among patients with genotype 1 infection, 98% of 126 previously untreated patients and 98% of 41 patients who did not respond to HCV PIs achieved SVR12. Ninety-two percent of 26 patients with genotype 2 infection and 89% of 18 patients with genotype 3 infection had an SVR12. The response rate was similar with or without RBV and for naive patients after 12 or 24 weeks of treatment. This study only constituted 15% compensated cirrhotic patients [59]. Ledipasvir, another NS5A inhibitor, has been developed in a fixed dose combination with sofosbuvir and will be licensed imminently. The phase 3 study included cirrhotic genotype 1 and 3 patients. In the ION-1 study, 865 untreated genotype 1 patients received a sofosbuvir-ledipasvir combination with or without RBV for 12–24 weeks. Overall, SVR rates were high at about 97–99%. Adding RBV and/or extending the treatment duration from 12 to 24 weeks did not increase SVR12. Patients with compensated cirrhosis (16%) had an SVR rate of 94% after 12 weeks of treatment [60]. The ION-2 study included 440 genotype 1 previously treated patients who were also treated with sofosbuvirledipasvir, with or without RBV, for 12–24 weeks. High SVR rates (94–99%) were observed across all arms. Twenty percent of the patients had cirrhosis at baseline and among these cirrhotic patients an SVR12 was observed in 86% compared to 95% in noncirrhotic patients after 12 weeks of sofosbuvir/ledipasvir combination, and
Other IFN-Free Combinations Asunaprevir is an NS3/4A PI and several studies have tested its efficacy in association with daclatasvir for genotype 1 patients. After 24 weeks of this combination, without RBV, the virological response is high for genotype 1b (71%) but very low for genotype 1a (5%). The phase 3 HALLMARK-DUAL study included 745 genotype 1b patients treated with asunaprevir and daclatasvir for 24 weeks. Among them, 305 were treatment naive, 205 were nonresponders to PEG-IFN/RBV, and 235 were ineligible or intolerant to IFN. Forty-seven percent of the IFN-ineligible patients and 31% of the nonresponder patients were cirrhotic. SVR12 rates were 91% for treatment-naive patients, 82% for nonresponder patients and 83% for ineligible patients. The response rates were similar in noncirrhotic (85%) and cirrhotic (84%) patients [63]. This combination is now being tested as a triple therapy, with the addition of a NS5B nonnucleoside inhibitor (BMS791325) in HCV genotype 1a and 1b, with more than 90% of 166 treatment-naive patients (9% cirrhotic patients) achieving SVR after 12 weeks of triple therapy [64]. A combination of the NS3/4A PI ABT-450 boosted by a low dose of ritonavir, the NS5A inhibitor ombitasvir (ABT-267) and the nonnucleoside NS5B inhibitor dasabuvir (ABT-333) have been tested in several studies. In the TURQUOISE-II study this three-drug combination with RBV is used in patients with genotype 1 infection and cirrhosis. A total of 380 patients were included and randomized for a 12- or 24-week treatment. SVR rates were 92% in the 12-week arm compared with 96% in the 24-week group. SVR rates did not differ significantly between the two treatment durations. However, previous nonresponder genotype 1a patients seemed to benefit from a 24-week treatment duration. This new combination, which could be available during the next year, was
Antiviral Therapy in Patients with HCV-Induced Cirrhosis
Dig Dis 2015;33:613–623 DOI: 10.1159/000375359
619
Downloaded by: Stockholms Universitet 198.143.54.1 - 7/31/2015 2:27:03 AM
was achieved. The response rate was not significantly different between F3 (98%) and F4 (95%) patients. The 12week regimen was as effective as the 24-week regimen, and the addition of RBV did not seem to improve response rates [53]. The combination of sofosbuvir and simeprevir was well tolerated in cirrhotic patients. Only genotype 1 patients were included in the COSMOS study, but both sofosbuvir and simeprevir have shown efficacy against genotype 4 viruses. On this basis, the combination of sofosbuvir and simeprevir could also be a therapeutic option for genotype 4 infections, as suggested by EASL recommendations [58].
Patients with Decompensated Cirrhosis or Awaiting Liver Transplantation Decompensated cirrhotic patients are usually excluded from phase 2 or 3 studies and very limited data are available regarding the tolerance and efficacy of new IFNfree regimens in such patients. The results of a randomized, safety and efficacy study of sofosbuvir plus RBV for 48 weeks compared to observation for 6 months in patients with HCV cirrhosis and portal hypertension (ChildTurcotte-Pugh scores 5–9) were recently presented during the EASL meeting. Fifty cirrhotic genotype 1, 2, 3 or 4 patients were included, among whom more than 70% had significant portal hypertension (hepatic venous pressure gradient >12 mm Hg). Sofosbuvir plus RBV for up to 24 weeks resulted in a high rate of virological suppression (93–100% at week 24) irrespective of the severity of liver disease. This rapid viral efficacy was associated with a decreased necroinflammation with ALT normalization, and also with improvements in platelet count, albumin, ascites and hepatic encephalopathy. The treatment was well tolerated with a low rate of treatment discontinuation due to adverse events. No patients with worsening or new hepatic decompensation were observed [67]. Posttreatment virological response data are now awaited. Cirrhotic patients with detectable HCV RNA at the time of liver transplantation universally experience recurrent HCV infection. Antiviral treatment before transplantation can prevent HCV recurrence, but existing IFN-based regimens are poorly tolerated and frequently contraindicated or ineffective [68]. In a phase 2 study, 61 HCV cirrhotic patients awaiting liver transplantation for HCC received up to 48 weeks of sofosbuvir and RBV before liver transplantation. Forty-six were transplanted. Among them, 43 patients had HCV RNA
