T. Jake Liang, M.D., and Marc G. Ghany, M.D., M.H.Sc.
May 4, 2014DOI: 10.1056/NEJMe1403619
A book on hepatitis C would read like a marriage of an Orson Welles mystery and a Shakespearean play — awash in enigma, tragedy, despair, resilience, redemption, and triumph. It is only fitting that treatment of hepatitis C virus (HCV) infection stands at center stage of such a book. After the initial introduction of interferon alfa as the mainstay of therapy, the field stalled for more than 10 years. Although the introduction of ribavirin combination therapy and pegylated interferons had increased response rates, the real breakthrough came with the development of direct-acting antiviral agents (DAAs).1 The first generation of DAAs in combination with peginterferon and ribavirin showed improved response rates, but they were accompanied by worsening side effects that have precluded a great majority of patients from benefiting from therapy.
In a previous article in the Journal, we reviewed the current and future therapies for HCV infection and commented on the rapidly shifting therapeutic landscape.1 We speculated that highly effective interferon-free regimens would be available and should revolutionize the treatment of HCV infection in the near future. Now, just 1 year after that review, we would have to say that the future is here.
The results of several phase 3 studies of interferon-free combination regimens of DAAs reported in the Journal now2-4 and recently5-9 (Table 1) unequivocally show the superiority of two such regimens over the standard-of-care treatment (a combination of peginterferon, ribavirin, and a protease inhibitor) for HCV genotype 1 infection. A previous editorial in the Journal highlighted the significantly improved response rates (rates of sustained virologic response of 93% to 99%) with a coformulated regimen of sofosbuvir (a nucleotide NS5B inhibitor) and ledipasvir (an NS5A inhibitor) among patients with HCV genotype 1 infection, as compared with the rates with the previously approved interferon-based single-DAA combination therapy.13
Other studies reported in the Journal show similarly high response rates with a different combination of DAAs among patients with HCV genotype 1 infection.2,3,8,9 This regimen includes three DAAs — ABT-450 (an NS3/4A inhibitor) coadministered with ritonavir (ABT-450/r), ombitasvir (an NS5A inhibitor), and dasabuvir (a nonnucleoside NS5B inhibitor) — with or without ribavirin. These studies evaluated the efficacy and safety of this regimen in patients who either were previously untreated or were previously treated with peginterferon and ribavirin but without a sustained virologic response. In addition, safety and efficacy in patients with compensated cirrhosis were examined specifically in one study.3 Like the studies of sofosbuvir and ledipasvir, these studies could use historical controls (treatment responses in previous phase 3 studies with the regimen of peginterferon, ribavirin, and a protease inhibitor) for comparison because of the anticipated wide difference in therapeutic margins between the old and new treatments.
In both studies involving patients without cirrhosis who were previously untreated or previously treated, the sustained-virologic-response rates were all about 96%,8,9 findings that suggest that patients with a previous nonresponse to peginterferon and ribavirin are not particularly difficult to treat with this regimen. Patients with cirrhosis did not have quite as robust a response to this regimen, though the response rate was still more than 90%.3 The study involving patients with cirrhosis also evaluated a treatment duration of either 12 weeks or 24 weeks and showed a modestly higher response rate in the 24-week group overall (96%, vs. 92% in the 12-week group), and secondary subgroup analyses suggest a greater response to the 24-week regimen, as compared with the 12-week regimen in patients with genotype 1a infection who had had a prior null response to peginterferon and ribavirin (93% vs. 80%).3
Several factors, such as racial or ethnic background, IL28B genotype, and baseline HCV RNA level, have been shown to influence treatment response to interferon-based therapy.14 As in the studies of sofosbuvir and ledipasvir,5-7 these factors do not play a prominent role in determining treatment response in these newer studies.2,3 Probably because these factors are specifically linked to the actions of interferon in the treatment of HCV infection, they do not appear to affect the response rates of the more potent DAA-containing regimens. The only factor that was modestly associated with treatment response in the trials of combination regimens of DAAs was body-mass index in one study.8 HCV genotype 1a infection has been shown previously to respond less well to DAA-based therapy than genotype 1b infection,15,16 but in these recent studies, HCV subgenotype did not seem to matter, other than in patients with cirrhosis. In one study, patients with genotype 1a infection seemed to benefit from the addition of ribavirin, whereas no significant difference was observed in patients with genotype 1b infection.2
The concept of response-guided therapy was introduced previously to tailor treatment duration on the basis of virologic response during treatment.16,17 In the era of potent DAA combination therapies, the decline in serum viral levels was rapid and dramatic: by week 4 of treatment, 99% of patients had nonquantifiable HCV RNA in the blood. Therefore, response-guided therapy is no longer necessary with these interferon-free DAA-based regimens. All patients could probably be treated with a single duration of therapy, which will certainly simplify and facilitate monitoring during treatment.
An interferon-free regimen has also been developed for the treatment of HCV genotype 2 or 3 infection. In two trials reported in the Journal last year,11,12 12 weeks of treatment with sofosbuvir and ribavirin resulted in response rates of more than 90% among previously untreated patients with genotype 2 infection but only about 60% among previously untreated patients with genotype 3 infection. The authors of one of the studies also examined the response rate with the same regimen among patients who did not have a response to prior treatment with peginterferon and ribavirin and found lower sustained-virologic-response rates (86% among patients with genotype 2 infection and 30% among patients with genotype 3 infection) than observed among previously untreated patients.12 In the same study, extending the treatment duration to 16 weeks resulted in a doubling of the sustained-virologic-response rate over the 12-week regimen (from 30% to 62%) among patients with genotype 3 infection.12
As reported in the Journal, 4 a follow-up study extended the treatment duration of patients with genotype 3 infection to 24 weeks and included both previously untreated patients and previously treated patients (with a nonresponse to peginterferon and ribavirin). Among previously untreated patients with or without cirrhosis, the longer duration regimen resulted in a response rate of more than 90%. However, among previously treated patients with cirrhosis, the response rate was significantly lower (62%). A close examination of the data suggests that relapse appeared to be the major reason for the nonresponse; among previously treated patients, extending therapy reduced the relapse rate from 66% with the 12-week duration to 20% with the 24-week duration.
Drug resistance against these DAAs is common in preclinical studies and with single-drug regimens in early clinical trials. Mathematical modeling has been applied to predict how many of these drugs are needed to minimize the drug-resistance problem.18 Practically, the number of drugs needed in a treatment regimen depends on their anti-HCV potency and the genetic barrier to the development of resistant mutations. In the case of sofosbuvir and ledipasvir, a two-drug combination is sufficient; in the other regimen, a three-drug combination appears to be necessary to achieve high response rates without selecting for resistant mutants. For the small number of patients who did not have a sustained virologic response, sequence analysis of the prevailing viral strains at the time of relapse showed the presence of previously described mutations that are resistant to each of these drugs, with the exception of sofosbuvir. Sofosbuvir seems to have a high genetic barrier to resistance, which probably explains its notable efficacy in DAA combination regimens.
The side effects associated with interferon-based therapy have prevented many patients from undergoing treatment and are a major reason for treatment failure. Perhaps the more important achievement of these interferon-free regimens is the lower rate and severity of side effects associated with treatment. The duration of treatment is shorter, and although constitutional symptoms of fatigue, headache, pruritus, and nonspecific gastrointestinal symptoms are common, most patients do not rate them as severe. With ribavirin-containing regimens, anemia is a common but manageable problem. Elevated bilirubin levels are often observed and can be attributed to inhibition of the bilirubin transporter by one of the drugs in addition to ribavirin-associated hemolysis. Serious adverse events, although uncommon (affecting <5% of study participants), were reported. Some of the events could be related to the treatment regimens. One death was reported in the trial involving patients with compensated cirrhosis who received the regimen containing ABT-450/r, ombitasvir, dasabuvir, and ribavirin, although it is unclear whether this event was related to the treatment. Further monitoring will be necessary.
At this juncture, we are certainly not ready to close the book on the treatment of HCV infection. The regimens have been tested predominantly in middle-aged, white men without cirrhosis. More-difficult-to-treat patients, such as those with cirrhosis, human immunodeficiency virus and HCV coinfection, or renal failure, remain a challenge. It is also not clear whether these regimens will be effective in those infected with HCV genotypes 4, 5, and 6, which are common in many parts of the world. Finally, the cost of treatment, which was highlighted in a recent editorial in the Journal, 13 will continue to be a deterrent for population-wide applications of these highly effective regimens. This dilemma is not only a topic of ongoing debate in the more developed countries, such as the United States and western European countries, but it is also a truly global public health problem of enormous impact — the majority of people with HCV infection live in lower-income, resource-constrained regions of the world. As pointed out in our previous review article and a recent Perspective article in the Journal, 19 the challenge will indeed continue to be how we can leverage modern medical advances, such as the treatment of HCV infection, to benefit those who are most in need.
Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.
This article was published on May 4, 2014, at NEJM.org.
From the National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD.