Abstract
Treatment of chronic hepatitis C (HCV) is currently undergoing a significant change. Traditional interferon-based therapy has been limited by both efficacy and tolerability, and many direct acting antiviral (DAA) drugs are emerging. Sofosbuvir (GS-7977) is a HCV NS5B nucleotide polymerase inhibitor that has now been evaluated extensively in phase II and III interferon-free clinical trials. The focus of this review is on the clinical potential of sofosbuvir in the treatment of HCV. Sofosbuvir has a pan-genotypic effect on HCV, although viral genotype-specific differences in sustained virological response (SVR) have emerged in phase III clinical trials. Sofosbuvir has been studied both as dual therapy with ribavirin and also as triple therapy with either NS5A inhibitors or a protease inhibitor. High rates of SVR have been reported with these interferon-free combinations, particularly with genotypes 1 and 2, and the safety profile has been very favourable in both cirrhotic and noncirrhotic patients, without issues of viral resistance. Interferon-free, once-daily treatment of HCV is now becoming a reality.
Introduction
Chronic hepatitis C virus (HCV) infection is a cause of chronic hepatitis, cirrhosis and hepatocellular carcinoma. For the past two decades, interferon-based therapy has been the cornerstone of HCV treatment, but success has been limited by poor tolerability and suboptimal sustained virological response (SVR) rates, even when combined with ribavirin. Boceprevir and telaprevir were the first direct-acting antiviral (DAA) drugs to be approved for the treatment of HCV in 2011, and resulted in improved SVR rates from approximately 40–44% to 68–75% in treatment-naïve patients with genotype 1 HCV [Jacobson et al. 2011; Poordad et al. 2011]. However, because of rapid emergence of viral resistance with protease inhibitor monotherapy, these agents are only effective when used as triple therapy in conjunction with peginterferon and ribavirin. Consequently the use of these NS3/4 protease inhibitors adds to the adverse event profile of peginterferon plus ribavirin, particularly in patients with cirrhosis where cytopaenias and other serious adverse events represent a significant safety concern [Fontaine et al. 2013].
There are three major classes of DAA drugs currently in phase III clinical trials: NS3/4A protease inhibitors, NS5A inhibitors and NS5B polymerase inhibitors, which can be subdivided into nucleoside inhibitors or nonnucleoside inhibitors. The NS5B polymerase is responsible for viral RNA replication, and the catalytic site of the NS5B protein is highly conserved across the different HCV genotypes, making nucleos(t)ide inhibitors that target this protein appealing as a treatment option. The nucleoside analogues interfere with the viral lifecycle by inducing a chain termination event and breaking transcription of the viral polyprotein [Sofia et al. 2010]. In general terms, they also have reasonably high potency and a high barrier to viral resistance. In contrast, the nonnucleoside inhibitors that target allosteric sites on NS5B tend to display lower potency and a low barrier to viral resistance.
Sofosbuvir, also known as GS-7977 (and previously known as PSI-7977), is a nucleotide inhibitor of NS5B and this review will consider its clinical potential as a promising drug for the treatment of HCV infection.
Pharmacology of sofosbuvir
Sofosbuvir is a prodrug of 2’-deoxy-2’-fluoro-2’-C-methyluridine monophosphate that is converted within hepatocytes to its active uridine triphosphate form, causing chain termination during replication of the viral genome [Murakami et al. 2010]. In vitro, the active triphosphate inhibits recombinant NS5B polymerases from HCV genotypes 1-4 with similar half maximum inhibitory concentration values for each genotype, indicating broad activity across HCV genotypes [Lam et al. 2012]. The chemistry of sofosbuvir has previously been reviewed [Herbst and Reddy, 2013] and will not be reviewed in detail in this paper.
Sofosbuvir is primarily eliminated from the body via the kidney as GS-331007 (formerly called PSI-6206), an inactive nucleoside metabolite. Single-dose pharmacokinetics of sofosbuvir were studied in subjects with normal renal function (estimated glomerular filtration rate [eGFR] > 80 ml/min), mild (eGFR 50–80 ml/min), moderate (eGFR 30–49 ml/min) and severe (eGFR < 30 ml/min) renal impairment. The area under the curve (AUC) of GS-331007 and, to a lesser extent, sofosbuvir increased with decreased renal status. There was a linear relationship between GS-331007 renal clearance and creatinine clearance. Subjects with mild, moderate and severe renal impairment had approximately 56%, 90% and 456% higher GS-331007 AUC, respectively, than subjects with normal renal function [Cornpropst et al. 2012]. Further studies are required to determine the safe use of sofosbuvir in patients with severe renal impairment.
In a study of hepatic impairment, HCV-infected subjects with moderate hepatic impairment were administered sofosbuvir 400 mg QD for 7 days; sofosbuvir was generally well tolerated and resulted in similar systemic exposure to GS-331007 as noncirrhotic subjects. Significant declines in HCV RNA were observed in all subjects over 7 days of dosing [Lawitz et al. 2012]. Therefore, dose modifications are not required in hepatic impairment.
There is no clinically significant interaction of sofosbuvir with food, or with coadministration of methadone, cyclosporine or tacrolimus [Denning et al. 2011; Mathias et al. 2012].
Clinical trial data
In the initial phase II studies, sofosbuvir was evaluated in combination with peginterferon and ribavirin (PEG/RBV). In a 28-day, dose-ranging trial in subjects infected with genotype 1 HCV, 64 patients were randomized to receive one of three once-daily doses of oral sofosbuvir (100, 200 or 400 mg) or placebo plus peginterferon and ribavirin for 28 days, after which all patients continued to receive peginterferon and ribavirin for a further 44 weeks [ClinicalTrials.gov identifier: NCT01054729]. Patients in the sofosbuvir/peginterferon/ribavirin groups showed mean reductions in HCV RNA >5 log10IU/ml for all doses versus 2.8 log10IU/ml for placebo/peginterferon/ribavirin after 28 days. Although response during the 28-day sofosbuvir/placebo phase of the study was nearly identical for all three sofosbuvir groups, differences emerged during the peginterferon and ribavirin phase of dosing, with SVR24 of 56% for the 100 mg group as compared with 83% and 80% for the 200 and 400 mg groups, respectively [Rodriguez-Torres et al. 2013]. The 200 and 400 mg doses were therefore selected for further evaluation in phase IIb trials.
PROTON
PROTON was a double-blind, randomized, placebo-controlled, dose-ranging phase II study that demonstrated that sofosbuvir was highly effective against genotypes 1, 2 and 3 HCV when used in combination with peginterferon and ribavirin as 12-week triple therapy, followed by additional peginterferon and ribavirin in the genotype 1 patients, with SVR12 results greater than 90% in all sofosbuvir-containing arms of the study [Lawitz et al. 2013c] (Table 1).
Phase II studies of sofosbuvir.
PEG, peginterferon; RBV, ribavirin; SOF, sofosbuvir; FDC, fixed-dose combination; EOT, end of therapy; RVR, rapid virological response (viral load undetectable by week 4 on treatment); SVR, sustained virological response; TVT, telaprevir-based triple therapy; BOC, boceprevir-based triple therapy; wk, week.Blank spaces in the table represent incomplete or unpublished data.
ATOMIC
The ATOMIC study explored shorter treatment durations of sofosbuvir-based triple therapy, and randomized 316 treatment-naïve patients with genotype 1 HCV into three treatment arms that included sofosbuvir 400 mg plus peginterferon and ribavirin therapy of 12 or 24 weeks duration, and one arm who received sofosbuvir triple therapy for 12 weeks, and then subjects were randomized to receive a further 12 weeks of sofosbuvir alone or with ribavirin [Kowdley et al. 2013]. A total of 11 patients with genotype 4 and 5 patients with genotype 6 were also included in cohort B of the study (Table 1). SVR rates remained greater than 90% in all arms of this study, with minimal differences in SVR seen in patients with factors traditionally associated with reduced response to interferon-based therapy such as high baseline viral load, patients with non-CC IL28B genotypes or bridging fibrosis on liver biopsy [Kowdley et al. 2013]. The findings also suggested that there was no additional benefit from extending sofosbuvir triple therapy beyond 12 weeks.
ELECTRON
Following on from PROTON and ATOMIC, the ELECTRON study evaluated sofosbuvir in interferon-sparing and interferon-free regimens for the treatment of HCV infection in noncirrhotic patients (Table 1). In the initial cohort, 40 treatment-naïve patients with genotype 2 or 3 HCV were randomized to four groups, all containing sofosbuvir 400 mg/day and weight-based ribavirin. Three of the groups also received peginterferon for 4, 8 or 12 weeks. All 40 patients had a SVR at week 24, irrespective of whether or not they received peginterferon [Gane et al. 2013c]. Two additional cohorts of patients with genotype 2 or 3 infection received either sofosbuvir monotherapy for 12 weeks, resulting in 60% SVR24, or sofosbuvir-based triple therapy for 8 weeks, with 100% SVR24.
Two groups of patients with HCV genotype 1 infection also received sofosbuvir and ribavirin for 12 weeks without peginterferon, including 25 treatment-naïve patients and 10 prior null responders to peginterferon. Among patients with genotype 1 infection, 84% of previously untreated patients and 10% of prior null responders achieved SVR24 in this interferon-free regimen. Again, a high proportion of patients had a non-CC IL28B genotype, and 89% were infected with genotype 1a which has been associated with lower response rates to other DAA therapy such as telaprevir- or boceprevir-based triple therapy. This study was proof of concept that SVR can be obtained with sofosbuvir-based, interferon-free regimens in genotypes 1, 2 and 3 HCV infection, and also demonstrated an ongoing role of ribavirin in preventing relapse after treatment.
Later cohorts of ELECTRON explored sofosbuvir/ribavirin in treatment-experienced patients with genotypes 2 and 3 and also cohorts with shorter durations of treatment and lower doses of ribavirin [ClinicalTrials.gov identifier: NCT01260350], although these results have only been partially published in abstract form [Gane et al. 2012], and weight-based ribavirin dosing has been used for ongoing clinical development with sofosbuvir-dual therapy. Patients with genotype 1 HCV were also treated with triple therapy combinations of sofosbuvir/ribavirin and additional DAA drugs including the NS5A inhibitor ledipasvir (LDV; GS-5885) in a once-daily fixed dose combination, or the HCV NS5B nonnucleoside inhibitor GS9669. With both of these triple therapy combinations, excellent SVR rates have been reported in both treatment-naïve patients and also null responders who are traditionally resistant to other forms of DAA triple therapy with first-generation protease inhibitors [Gane et al. 2013a; Gane et al. 2013b] (Table 1).
LONESTAR
The LONESTAR study continued to evaluate different lengths of therapy (8 or 12 weeks) with the fixed-dose combination of sofosbuvir and ledipasvir with or without ribavirin in patients with genotype 1 HCV (Table 1). Cohorts included both treatment-naïve GT1 patients, and also 40 patients who had previously failed therapy with an HCV-specific protease inhibitor-based regimen. Half of these patients had compensated cirrhosis. SVR4 rates of at least 95% in all cohorts have been reported in a recent Gilead press release [Gilead Sciences, 2013], and phase III trials are planned with the fixed-dose combination of sofosbuvir and ledipasvir.
Daclatasvir and sofosbuvir
Two clinical trials have evaluated sofosbuvir in combination with the NS5A inhibitor daclatasvir. The first trial reported was of treatment-naïve patients with HCV genotypes 1, 2 and 3 who were randomized to daclatasvir plus sofosbuvir with or without ribavirin for a total treatment duration of 24 weeks although some arms had a 1-week sofosbuvir lead-in phase before the daclatasvir was added. SVR12 was achieved in 100% of genotype 1 patients and 86–100% of patients in the genotype 2/3 cohort [Sulkowski et al. 2012].
The combination of daclatasvir and sofosbuvir has also been evaluated in 41 HCV genotype 1 patients who had previously failed protease-inhibitor-based triple therapy with either telaprevir or boceprevir and were treated with 24 weeks of sofosbuvir plus daclatasvir, with or without ribavirin [Sulkowski et al. 2013]. Impressively, SVR12 rates of 95–100% were obtained, whether or not the combination included ribavirin, demonstrating that this combination of sofosbuvir with an NS5A inhibitor is an effective therapy even in people with NS3A protease inhibitor resistance.
COSMOS
The COSMOS study is evaluating a once-daily regimen of the NS3/4A protease inhibitor simeprevir (TMC435) plus sofosbuvir with or without ribavirin for 12 or 24 weeks in HCV genotype 1 patients (Table 1). Cohort 1 consists of patients with prior null response to peginterferon with mild to moderate fibrosis, and preliminary results from the 12-week cohort have been recently presented showing SVR4 rates of 96% and 93% for patients with and without ribavirin, respectively [Lawitz et al. 2013b]. Cohort 2 includes both peginterferon null responders and treatment-naïve patients with advanced fibrosis and a recent press release reports SVR4 results of 96–100% in the 12-week group [Medivir, 2013].
Phase III trials of sofosbuvir
Four phase III trials of sofosbuvir have been published to date, all evaluating sofosbuvir 400 mg plus ribavirin (weight-based dosing) for at least 12 weeks in patients with chronic HCV (Table 2).
Phase III studies of sofosbuvir.
PEG, pegylated interferon alpha; RBV, ribavirin; SMV, simeprevir (TMC435); SOF, sofosbuvir (GS7977); DCV, daclatasvir; RGT, response guided therapy, EOT, end of therapy; SVR, sustained virological response.
In the FISSION study, 499 treatment-naïve patients with genotype 2 or 3 HCV were randomized to sofosbuvir 400 mg plus ribavirin for 12 weeks or peginterferon plus ribavirin for 24 weeks in a noninferiority trial. In this trial, despite a marked difference in the RVR rates (99% versus 67%, respectively), the SVR12 rates were identical at 67% [Lawitz et al. 2013d] (Table 2). Notably, in this study there was a marked difference in response rates between patients with genotype 2 (97% SVR 12) and genotype 3 (56% SVR12) HCV, and between cirrhotic (47% SVR12) and noncirrhotic (72% SVR12) patients.
The NEUTRINO study was a 12-week open label study of sofosbuvir plus ribavirin in treatment naive patients with HCV genotype 1, 4, 5 or 6 (of whom 98% had genotype 1 or 4). SVR12 rates of 90% were observed overall, with 81% response rates in genotype 1 with cirrhosis. These cirrhotic patients also had extremely low rates of treatment discontinuation of only 2%, suggesting that this combination is safe and extremely well tolerated even in cirrhotic patients [Lawitz et al. 2013d].
POSITRON evaluated sofosbuvir plus ribavirin compared with placebo in patients with genotype 2 and 3 HCV in whom interferon was not an option (previously intolerant of interferon, unwilling or unable to take interferon). Overall SVR12 rates in the sofosbuvir/ribavirin group were higher than in FISSION (78%), possibly because of the higher proportion of genotype 2 patients who again demonstrated significantly higher SVR12 rates (92%) than genotype 3 (68%). Degree of liver fibrosis did not appear to significantly influence outcomes in genotype 2 patients (94% SVR12 even in those with cirrhosis), but in genotype 3 patients, SVR12 rates were low at 21% in patients with cirrhosis [Jacobson et al. 2013].
FUSION
The FUSION study was a blinded, active-control study involving patients with genotype 2 or 3 HCV who had not had a response to prior treatment with an interferon-containing regimen. Patients were randomly assigned to either 12 weeks of sofosbuvir and ribavirin followed by 4 weeks of matching placebo, or 16 weeks of sofosbuvir and ribavirin. One third of participants were cirrhotic. Overall rates of SVR were significantly higher in the 16-week arm (73%) than in the 12-week arm (50%), and again the genotype 2 patients had higher SVR rates (86% and 94% for 12 and 16 weeks, respectively) than genotype 3 patients (30% and 62% respectively for 12 and 16 weeks) [Jacobson et al. 2013]. Cirrhosis was associated with a decreased rate of SVR, particularly in genotype 3 receiving 12 weeks of therapy rather than 16 weeks (SVR 19% versus 61%). In genotype 2 patients SVR rates ranged from 60% to 78% in cirrhotic patients, and up to 100% in noncirrhotic patients without cirrhosis who were treated for 16 weeks.
Potential for sofosbuvir in liver transplantation
One obvious clinical need is for data regarding safety and efficacy of sofosbuvir in patients who have decompensated chronic liver disease, are peri-transplant or post-liver-transplant. The excellent safety data to date and the lack of significant drug interactions makes sofosbuvir an appealing choice to be studied in these groups. To date there is one case report published of a patient with severe recurrent cholestatic hepatitis C, 6 months post-transplant, who was effectively rescued and achieved SVR with treatment with sofosbuvir and daclatasvir in combination [Fontana et al. 2013]. This is promising, and results of future trials of sofosbuvir in these types of patient groups are awaited with interest.
Adverse events
Treatment discontinuations because of adverse events have been uncommon in these sofosbuvir-based interferon-free treatment regimens. In phase III trials, treatment discontinuation rates of 1–2% were seen in the sofosbuvir plus ribavirin cohorts, as compared with 11% among patients receiving peginterferon–ribavirin for 24 weeks. Adverse events associated with ribavirin therapy (fatigue, insomnia and anaemia) were commonly reported, and headache was also frequently reported. Haematologic abnormalities were more common among patients who received interferon than among patients who did not in FISSION and ELECTRON [Gane et al. 2013c; Lawitz et al. 2013d]. Neutropenia and thrombocytopenia were not generally seen in groups who did not receive interferon. Depression is a common side effect of interferon therapy, and in the FISSION study, occurred in 14% of patients receiving peginterferon, as compared with 5% of patients receiving sofosbuvir plus ribavirin. In a recent analysis of the impact of HCV treatment on quality of life, in the FISSION and POSITRON trials, sofosbuvir plus ribavirin was associated with better health-related quality of life than peginterferon plus ribavirin, and was similar to patients not receiving active treatment. Achieving SVR on sofosbuvir plus ribavirin was also associated with improvement in health-related quality of life [Younossi et al. 2013].
Concordance data: SVR at weeks 4, 12 and 24
Traditionally, with interferon-based therapy for HCV, the absence of detectable virus 24 weeks after completing treatment has been used to define SVR24. However, recent analyses have suggested that SVR12 and SVR24 are concordant across treatments that include peginterferon/ribavirin and also peginterferon/ribavirin triple therapy including telaprevir and boceprevir, with positive predictive values of 98% for SVR12 and 91% for SVR4 [Chen et al. 2013]. The concordance of SVR4, SVR12 and SVR24 has been assessed in the sofosbuvir phase II program including 590 patients. High levels of concordance were observed between SVR4 and later time points and positive predictive values and sensitivity of SVR4 for SVR12 and SVR24 were greater than 98.5% [Lawitz et al. 2013a]. Increasing reliance on these earlier time points of SVR4 and SVR12 to determine outcomes is almost certainly accelerating the pace of drug development.
Viral resistance
Viral mutations in the HCV NS3 and NS5B regions have been associated with resistance to protease and nonnucleoside inhibitors, and can be present even in previously untreated patients with HCV infection. The S282T mutation has been identified as the common mutation selected in replicon studies with sofosbuvir, and this change can confer resistance to sofosbuvir [Lam et al. 2012]; however, this mutation is not generally present in untreated patients with HCV, and has very poor replicative fitness [Kuntzen et al. 2008]. In the initial large phase II and III studies involving sofosbuvir the S282T mutation was not detected on deep-sequencing assays in any patient receiving sofosbuvir and is in marked contrast with the rapid emergence of viral resistance that has been observed with other classes of DAA in patients who had breakthrough during treatment or relapse after completion of therapy [Gane et al. 2013c; Jacobson et al. 2013; Lawitz et al. 2013d]. Encouragingly, the emerging data suggest that sofosbuvir/NS5A inhibitor combination therapy is highly effective in NS3A protease-inhibitor experienced patients, even when they have protease inhibitor resistance [Sulkowski et al. 2013].
Conclusions and future directions
Although sofosbuvir is by no means one of the first DAA drugs to reach phase III clinical trials, the collective trial data for sofosbuvir do represent a significant paradigm shift in the management of HCV infection. With sofosbuvir-based regimens, successful interferon-free treatment of HCV is now achievable across multiple genotypes, but different patterns of genotypic response to treatment have emerged compared with those seen in traditional interferon-based therapy. It is now clear that in the current DAA era, genotypes 2 and 3 should be evaluated quite separately, as SVR rates in genotype 3 patients are significantly lower than those seen in genotype 2 [Jacobson et al. 2013; Lawitz et al. 2013d]. Genotype 2 is a readily treatable strain of HCV, with superb SVR rates with sofosbuvir/ribavirin dual therapy even in patients with established cirrhosis. In contrast, the presence of advanced liver fibrosis has a significant detrimental effect on SVR rates in genotype 3 patients when a 12-week regimen is used. The FUSION data strongly suggest that for this group of patients, at least 16 weeks of therapy is needed for the dual sofosbuvir/ribavirin regimen, and the results of ongoing trials evaluating a 24-week duration of treatment are awaited with interest. The other strategy that could be considered is addition of another DAA to the regimen, but until these data are available to clarify the optimal interferon-free treatment of genotype 3 HCV patients with advanced fibrosis or previous treatment failure, there may be an initial role for continuing to use peginterferon together with sofosbuvir in this group of patients.
For genotype 1 HCV, the NEUTRINO data have shown the highest yet reported SVR rates for patients with genotype 1 HCV and cirrhosis when treated with sofosbuvir/ribavirin dual therapy, and this is particularly notable for the lack of significant adverse events, in contrast with the ongoing emerging data about safety of protease inhibitor/peginterferon-based triple therapy. However this dual therapy is not sufficient for the genotype 1 null responder group. Emerging phase II data in genotype 1 patients of triple therapy regimens of sofosbuvir plus ribavirin combined with an NS5A inhibitor, either daclatasvir or ledipasvir, demonstrate SVR rates close to 100%, in both treatment-naïve and null-responder patients, with potential to further shorten the duration of therapy to 8 weeks with this triple therapy approach, utilizing the fixed-dose once daily regimen of sofosbuvir/ledipasvir. Data on the effectiveness of this combination in cirrhotic genotype 1 null responders is awaited. A new era of successful interferon-free DAA therapy for HCV is emerging, with potential to broaden treatment of HCV to include patient groups who have either avoided or not been suitable for previous interferon-based therapy, and it is likely that sofosbuvir will form the backbone of this treatment approach.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement
Disclosures: Grant/Research Gilead; Advisory Board - Janssen.