|Home | About | Journals | Submit | Contact Us | Français|
While we continue to wait for small molecule hepatitis C virus (HCV) lifecycle inhibitors to truly increase sustained viral response (SVR) rates and enlarge the pool of potential treatment candidates, we are obliged to investigate other means to improve outcomes with the therapies currently at hand. Improved outcomes may take many forms, as follows:
Each of these goals is potentially achievable and impacts the way we approach patients with chronic hepatitis C infection. What is the evidence supporting these modes of treatment?
The key to unlocking the evidence for individualized therapy came not from identifying better pre-treatment predictors, but rather from an improved understanding of how HCV responds during the first few weeks of treatment, and then studying the downstream effects on treatment outcome. Clearly, genotype and viral load affect initial responses, but it is the early viral kinetics that ultimately drives subsequent response rates. Fried and colleagues recently confirmed this in a retrospective analysis of a large multinational HCV database [Fried et al. 2008a]. Shiffman et al.  showed that time to HCV RNA unde-tectability was more important than baseline factors in predicting SVR in genotype-1 patients. Regardless of genotype or viral load, the ability to achieve rapid virologic response (RVR, undetectable HCV RNA at week 4) was the most important predictor of subsequent SVR [Fried et al. 2008a]. Across all genotypes, RVR was associated with a high SVR rate [Fried et al. 2008a]. It has been the ability to utilize this information to develop improved strategies for HCV treatment that has led to the concept of individualized therapy, or ‘response-directed therapy’.
It would seem, on the surface, to be an obvious strategy: give more interferon at early time points to drive the virus more quickly to undetectability, thus providing more patients with RVR and improving SVR. Unfortunately, ‘induction’ dosing of interferon has proven to be an elusive therapeutic panacea. Numerous, well-controlled investigations have demonstrated that, whereas induction interferon may provide an increase in early on-treatment responses, this has not translated into improved SVR rates. Why? It is unclear, but may relate to the duration of the higher dose given. Most investigations have applied induction dosing for only four to twelve weeks, then reverted to standard dosing. This may be too short. In the REPEAT trial of genotype-1, peginterferon/ribavirin nonresponders, 360mg of peginterferon alfa-2a with 1200mg ribavirin was administered for twelve weeks in two arms of the trial and compared to standard dosing [Jensen and Marcellin, 2007]. After 12 weeks, the two induction arms had achieved a greater number of responders, but this initial benefit was lost once the dose was reduced after week 12 for the remaining 36 or 60 weeks [Jensen and Marcellin, 2007]. In this regard, Fried and colleagues recently demonstrated that higher doses of both peginterferon (270mg vs 180 mg) and ribavirin (1600mg vs 1200 mg) given for 48 weeks was superior to standard doses in particularly difficult-to-treat individuals (genotype 1, high viral load, weighing more than 85kg) [Fried et al. 2008b].
There is evidence from several investigations that higher doses of ribavirin appear to improve SVR rates by decreasing virologic relapse, albeit at the cost of increased anemia. Jacobson and colleagues demonstrated that the use of weight-based ribavirin was associated with improved SVR rates in individuals with high body weight [Jacobson et al. 2007]. In fact, the use of very high ribavirin doses (averaging 2500 mg/day), when supported by growth factors and blood transfusions, has been associated with a 90% SVR rate in a small cohort of difficult-to-treat genotype-1 patients [Lindahl et al. 2005].
Evidence for the use of higher-dose peginterferon plus ribavirin is less clear for other genotypes. Although Hadziyannis et al.  demonstrated that 24 weeks of therapy, incorporating 800 mg/ day of ribavirin, gave equivalent treatment results to 48 weeks and 1000–1200 mg/day of ribavirin in non-1 genotypes, the non-1 genotypes were not evaluated separately for differences in response. Subsequent studies have borne this out – genotype 2 does respond better than genotype 3 to the same treatment regime [von Wagner et al. 2005]. It is conceivable that specific subgroups of genotype 2 and 3 might benefit from more intensive therapy; for example, genotype 3 with high viral load. Further studies are needed.
Perhaps one of the more important observations coming out of the registration trials of the two commercial peginterferon compounds was the ability to define a point in the therapy at which continued treatment would be associated with such a low SVR rate as to make continued treatment unnecessary. Manns et al.  noted that if the HCV RNA was still detectable at week 24, continued treatment of genotype 1 was associated with a <1% SVR rate. Fried et al.  showed that if the HCV RNA did not decrease at least 2 log10 from baseline by week 12, continued treatment was associated with only a 3% SVR rate. Therefore, the concept of ‘stopping rules’ grew out of these observations: stop treatment if HCV RNA has not declined by at least 2 logs by week 12, or is still detectable by week 24. Whether earlier time points might prove as accurate in defining treatment futility remains to be tested, but it seems likely. The advantages of defining stopping rules are obvious – it limits the burden of therapy to those most likely to have a long-term response. It is important to remember, however, that these rules only apply to patents treated with approved doses of medication over approved treatment durations, and are generally most relevant to patients infected with genotype 1.
Extending treatment in patients with genotype 1 from 48 to 72 weeks is associated with improved SVR, but would mean treating all patients for this duration [Berg et al. 2006]. As more and more patients have been treated with peginterferon plus ribavirin, more robust datasets have allowed identification of additional patient subsets based upon viral kinetics. A particularly intriguing group are those in whom a slow or partial reduction in viral load during the first twelve weeks leads to an eventual end-of-treatment response but a high relapse rate. Would these individuals benefit from an extended treatment duration? Several groups of investigators have tested this hypothesis using different randomization points. Sánchez-Tapias et al.  randomized HCV patients after four weeks: those with detectable HCV RNA were randomized to receive either 48 or 72 weeks of total therapy. This study demonstrated a clear benefit to extending treatment duration in those who were still HCV RNA positive at week 4, but suffered from using a low dose of ribavirin (800mg instead of 1000– 1200 mg). Other investigators have confirmed the value of extended therapy for slow responders [Mangia et al. 2008; Pearlman et al. 2007; Ferenci et al. 2006]. This is an important observation, which appears to confirm the analysis of Drusano and Preston  that the duration of HCV RNA undetectability is an important determinant of SVR.
Perhaps one of the more exciting concepts to arise from on-treatment viral kinetics was the possibility that treatment duration could actually be shortened in some individuals who responded most dramatically to therapy. This observation, however, entailed earlier routine assessment of HCV RNA than had previously been standard. Ferenci and colleagues revealed a close relationship between time to HCV RNA undetectability and subsequent SVR rate [Ferenci et al. 2005]. Studies by Zeuzem et al. , and later by Jensen et al. , demonstrated that 24 weeks of treatment provided similar overall SVR rates when compared to 48 weeks in genotype-1 patients who had no detectable virus at week 4 (RVR). These ‘super responders’ were generally comprised of a subset of patients with a low baseline viral load. In a further assessment of the effect of RVR upon treatment duration, Yu et al.  performed a prospective, randomized trial in 200 HCV genotype-1 subjects and demonstrated that SVR rates were lower for those treated for 24 weeks, even in those achieving an RVR. This difference was primarily the result of patients with high baseline viral load, since there was no statistical difference in those with low baseline viral load.
Several studies have evaluated the possibility of shortening treatment duration in genotypes 2 and 3 [Lagging et al. 2008; von Wagner et al. 2005; Shiffman et al. 2007; Mangia et al. 2005; Zeuzem et al. 2004]. Patients infected with genotypes 2 and 3 might respond comparably to a shortened treatment duration, compared with 24 weeks, based upon initial rapid viral response [Mangia et al. 2005; von Wagner et al. 2005; Zeuzem et al. 2004]. It is interesting to speculate why some studies conclude that if an RVR is achieved, treatment of genotype 2 and 3 may be shortened without significant loss of SVR [Mangia et al. 2005; von Wagner et al. 2005; Zeuzem et al. 2004], while others fail to confirm such noninferiority [Lagging et al. 2008; Shiffman et al. 2007]. There are clearly many differences in study design, week 4 HCV RNA cut-off for definition of RVR, and dose of ribavirin. Furthermore, these studies suggest that genotype 2 responds more favorably than genotype 3, but loses this difference when treatment duration is shortened following RVR [Shiffman et al. 2007]. The interrelationship of genotype, viral load, liver histology and age, as well as dose of ribavirin, could be explanations, but further investigation is required.
Genotype-4-infected patients may also benefit from shortened treatment duration based upon the clinical trial results of Kamal et al. . Genotype-4 patients who achieved an RVR were randomized to receive either 24 or 48 weeks of treatment. Those achieving an RVR had similar SVR rates with either treatment duration.
How much of an impact will these changes in treatment make? Mangia et al.  performed a prospective, controlled trial of RVR-based treatment duration in genotype-1-infected patients. These investigators found that shortening treatment duration from 48 to 24 weeks in those achieving an RVR was associated with a quantitatively lower SVR, which did not reach clinical significance. Likewise, extending the duration of therapy from 48 to 72 weeks in those subjects still HCV RNA detectable at week 12 was associated with a small but notable gain in SVR. Therefore, it appears that altering treatment duration in response to early or rapid viral response may not necessarily be associated with significant overall improvements in SVR rate, but may certainly allow more targeted treatment durations – limiting the burden of therapy in those with an RVR, while improving SVR in those with a slow initial response at the cost of additional treatment duration.
HCV cirrhosis represents another subgroup of patients for whom treatment may require individualization [Strader et al. 2004]. Not only is there concern for increased side effects, decompensation and infection risk [Crippin et al. 2002], but patients with cirrhosis seem to respond less well to conventional doses and durations of peginterferon and ribavirin [Crippin et al. 2002; Fried et al. 2002; Manns et al. 2001]. Perhaps more than any other subgroup of patients, cirrhotic patients have the most to gain from successful viral eradication. Several studies have reported a lower incidence of hepatocellular carcinoma [Shiratori et al. 2005], and possibly a decreased requirement for liver transplantation in successfully treated HCV cirrhosis [Everson et al. 2005]. How then should we treat patients who have cirrhosis? Well-compensated cirrhotic patients with preserved hematologic parameters probably require no revision of current treatment guidelines. On the other hand, we are in need of good prospective studies of treatment approaches for patients with decompensated liver disease and those with significant cytopenias.
Obesity and hepatic steatosis have been associated with a decreased response to interferon plus ribavirin therapy [Westin et al. 2007; Charlton et al. 2006; Negro 2006; Patton et al. 2004; Bressler et al. 2003]. Several mechanisms may be responsible for this interference, including a decrease in hepatocyte interferon signaling. It remains less clear, however, whether weight loss and/or treatment of the metabolic syndrome will improve subsequent interferon plus ribavirin therapy. Although it seems intuitive that it should, few studies have examined the effect of weight loss on steatosis in HCV-infected subjects. Hickman and colleagues demonstrated that a 10% reduction in body weight was associated with significant reductions in hepatic steatosis [Hickman et al. 2004, 2002]. Should we advise patients who are overweight and have evidence of steatosis to lose weight prior to treatment? Assuming the treatment is not urgent, such a recommendation seems to make sense on a number of levels.
Although we have made significant improvements in HCV therapy, we still have many unanswered questions:
Clearly there is still much work left to do, the results of which should enhance our ability to more effectively treat our HCV patients.
Research support: Roche, Schering-Plough, Boehringer-Ingelheim, Vertex, Globeimmune Consultant: Abbott, Roche, Boehringer-Ingelheim, Vertex, Globeimmune