Sofosbuvir + GS-5816 for 12 weeks shows high HCV cure rates, but 8 weeks is less effective for some patients

A 12-week oral combination of sofosbuvir plus the experimental NS5A inhibitor GS-5816 demonstrated high sustained virological response rates for people with difficult-to-treat HCV genotype 3 and other genotypes, according to phase 2 study findings presented at the American Association for the Study of Liver Diseases (AASLD) Liver Meeting last week in Boston. Reducing treatment duration to 8 weeks resulted in lower efficacy for people with HCV genotypes 1 and 2, though genotype 3 patients still showed high response rates.

The advent of interferon-free therapy using direct-acting antivirals that target different steps of the hepatitis C virus (HCV) lifecycle has brought about a revolution in treatment for chronic hepatitis C.

Gilead Sciences' nucleotide HCV polymerase inhibitor sofosbuvir (Sovaldi) is among the most effective of these new drugs. A dual combination of sofosbuvir plus ribavirin works well against easier-to-treat HCV genotype 2. The combination of sofosbuvir plus Gilead's first-generation NS5A inhibitor ledipasvir (a co-formulation recently approved and marketed as Harvoni in the US) is highly effective against HCV genotype 1, but it is not indicated for other genotypes.

Unlike ledipasvir – which is only active against HCV genotype 1 – Gilead's next-generation NS5A inhibitor GS-5816 is 'pan-genotypic', or active against multiple genotypes. In addition to more effectively treating people known to have HCV genotypes other than 1 or 2, a pangenotypic regimen could help increase access by eliminating the need for genotypic testing prior to treatment.

Treatment-experienced patients: 12 weeks

Stephen Pianko from Monash Medical Centre in Australia presented results from a phase 2 trial testing a 12-week regimen of sofosbuvir plus GS-5816, with or without ribavirin, for treatment-experienced genotype 1 or 3 patients with or without liver cirrhosis (study GS-US-342-0109).

In the era of interferon-based therapy, genotype 3 was once classified with genotype 2 as 'easier-to-treat', but it is more prone to resistance and does not respond as well to some direct-acting antivirals, requiring longer treatment or additional drugs. It is now also known that among people with genotype 1, subtype 1a is harder to treat than 1b.

This study included 107 genotype 3 patients without cirrhosis, 103 genotype 3 patients with cirrhosis, and 111 genotype 1 patients (mostly with subtype 1a) either with or without cirrhosis. Overall, about 70% of participants were men, most were white, the mean age was 55 years, about 45% had cirrhosis and three-quarters had unfavourable IL28B gene variants.

Genotype 1 patients had previously been treated with pegylated interferon and ribavirin plus approved or investigational HCV protease inhibitors such as telaprevir (Incivo or Incivek), while genotype 3 patients had previously received pegylated interferon/ribavirin alone. A majority (72%) of prior treatment failures were attributable to on-treatment viral breakthrough or post-treatment relapse rather than null response.

Participants – stratified by genotype, and within genotype 3, by presence or absence of cirrhosis – were randomised to receive 400mg sofosbuvir plus GS-5816 at doses of either 25mg or 100mg, with or without ribavirin, all for 12 weeks.

Genotype 1 patients did well on all the tested regimens, with 96% to 100% achieving sustained virological response, or undetectable HCV RNA at 12 weeks after completing treatment (SVR12). Both GS-5816 doses worked equally well and addition of ribavirin did not improve response.

Among genotype 3 patients without cirrhosis, the SVR12 rate was 85% in the 25mg GS-5816 arm without ribavirin, rising to 96% with ribavirin. The 100mg dose, however, cured all participants with or without ribavirin (100% SVR12).

Turning to the genotype 3 patients with cirrhosis, SVR12 rates in the 25mg GS-5816 arms were just 58% without ribavirin and 84% with ribavirin. Cure rates rose to 88% in the 100mg arm without ribavirin, and to 96% in the 100mg arm with ribavirin.

Across all patient groups and regimen arms, all treatment failures were due to post-treatment relapse, save for one withdrawal of consent. Genetic deep sequencing showed that everyone who experienced virological failure had NS5A resistance-associated viral variants, but no one showed evidence of polymerase resistance variants.

"Sofosbuvir + GS-5816 for 12 weeks resulted in high SVR12 rates in treatment-experienced genotype 1 or genotype 3 patients including those with cirrhosis," the investigators concluded.

"We were able to achieve very high SVR rates in some of the most difficult patients," Pianko said, adding that a potent enough combination appears to overcome viral resistance.

Among genotype 3 patients, SVR12 rates were higher with the 100mg compared with the 25mg GS-5816 dose. Addition of ribavirin had a numerical advantage in the 25mg arms, Pianko added, but did not significantly increase response rates when using the higher dose. Adding ribavirin had no effect among genotype 1 patients using either GS-5816 dose.

However, a speaker from the audience suggested that the numerically lower response rate without ribavirin among people with genotype 3 who had cirrhosis appears to support data from other studies reported at the meeting, showing that ribavirin may still be beneficial for the hardest-to-treat patients.

Treatment-naive patients: 8 weeks

Two other presentations reported results from studies of sofosbuvir plus GS-5816 taken with or without ribavirin for 8 weeks. A shorter treatment duration is more tolerable and less expensive, but may increase the likelihood of relapse.

Tram Tran from Cedars-Sinai Medical Center in Los Angeles presented findings from a phase 2 trial (GS-US-342-0102) evaluating sofosbuvir plus GS-5816 for previously untreated patients with HCV genotypes 1 through 6 who did not have cirrhosis.

In Part A of the study, presented at the EASL International Liver Congress in April, 55 patients with genotype 1, 54 people with genotype 3, and 45 people with genotypes 2, 4, 5 or 6 were treated with sofosbuvir plus either 25mg or 100mg GS-5816, all without ribavirin for 12 weeks. As previously reported, SVR12 rates ranged from 91% to 100% using the 25mg GS-5816 dose and from 86% to 100% using the 100mg dose.

Based on these promising results, Part B evaluated a shorter treatment duration of 8 weeks in 120 genotype 1 patients (most with subtype 1a) and 103 genotype 2 patients. More than half were men, most were white, the mean age was 51 years and about one-third had unfavourable IL28B variants.

Again, they were randomly assigned to received sofosbuvir plus either 25mg or 100mg GS-5816, but this time they were also randomised to take this regimen with or without ribavirin, all for 8 weeks.

Among genotype 1 patients, SVR12 rates were 87% for those taking 25mg GS-5816 without ribavirin, 83% for those taking 25mg with ribavirin, 90% for those taking 100mg GS-5816 without ribavirin and 81% for those taking 100mg with ribavirin. Among the genotype 2 patients, SVR12 rates were 77% for those taking 25mg GS-5816 without ribavirin and 88% in the other three arms. In both groups, all virological failures were due to post-treatment relapse.

One-quarter of genotype 1 patients and half of genotype 2 patients had NS5A resistance-associated viral variants present prior to treatment. Genotype 1 patients had similar cure rates regardless of the presence or absence of resistance variants (86% vs 88%, respectively), but among people with genotype 2, those with baseline resistance variants had a lower response rate (81% vs 94%). One-third of patients with virological failure had resistance variants detected post-treatment.

The researchers concluded that "SVR rates were lower and relapse rates were higher in genotype 1 and genotype 2 patients treated for 8 weeks compared to those treated for 12 weeks."

Finally, Edward Gane from Auckland Clinical Studies in New Zealand presented findings from the ELECTRON-2 trial, comparing the same 8-week regimens for treatment-naive genotype 3 patients without cirrhosis.

In this study, which included 104 participants, about two-thirds were men, most were white, the mean age was approximately 48 years and 25% to 56% had favourable IL28B variants. Almost all had HCV subtype 3a, with a few having 3k and several having an unknown subtype.

SVR12 rates were 100% for those taking 25mg GS-5816 without ribavirin, 88% for those taking 25mg with ribavirin (with two relapses and one discontinuation due to an adverse event), 96% for those taking 100mg GS-5816 without ribavirin (with one withdrawal of consent) and 100% for those taking 100mg with ribavirin.

Genetic sequencing showed that 26% of participants had NS5A resistance variants at baseline. One relapser had a Y93H mutation at baseline, but the other had no baseline or treatment-emergent resistance variants. Conversely, 10 of the 11 patients with the Y93H variant at baseline nonetheless achieved SVR12.

In all three studies, sofosbuvir plus GS-5816 was generally safe and well-tolerated with few serious adverse events or early discontinuations due to adverse events. The most frequently reported side-effects were headache, fatigue, nausea, diarrhoea, insomnia and itching or rash. Anaemia was observed more often in the ribavirin-containing arms, but was rarely severe.

These findings indicate that some patients are likely to be cured with only 8 weeks of treatment, but shortening therapy should be done cautiously. Among genotype 1 and 2 patients in Tran's study, neither the higher GS-5816 dose nor addition of ribavirin raised response rates into the now-expected 90% to 100% range using this short treatment duration. But in Gane's study of genotype 3 patients, no one using the higher GS-5816 dose, with or without ribavirin, experienced virological failure.

Based on these results, the 100mg dose of GS-5816 has been selected and combined with sofosbuvir in a co-formulation that will be tested for 12 weeks without ribavirin in phase 3 trials. SVR12 rates have ranged from 86% to 100% for patients across all genotypes treated with this regimen in phase 2 studies.

ASTRAL-2 and ASTRAL-3 (now recruiting) will compare the sofosbuvir/GS-5816 coformulation without ribavirin for 12 weeks versus the currently approved regimen of sofosbuvir/ribavirin for 12 weeks for genotype 2 or 24 weeks for genotype 3.