According to international guidelines, in principle, every patient with chronic HCV infection is a candidate for antiviral treatment . Because of the results of phase III trials, patients infected with HCV GT1 should receive a triple regimen including either BOC or TLV [9, 10]. However, not all patients with chronic HCV GT1 infection are eligible for these therapies . The risk/benefit ratio can be poor in certain patient cohorts. Therefore, at least three key factors should be considered when selecting the best patients: Therapy-associated risk, chances of achieving an SVR and treatment urgency.
Therapy-associated risk/adverse events
Current antiviral therapies are still based on interferon (IFN), which is associated with several AEs, i.e. flu-like symptoms or fatigue [5, 11] and with the addition of PIs more side effects will occur [12-15]. In certain cases, AEs can be dangerous or even life threatening, i.e. systemic infections or severe anaemia. Various possible risk factors for serious adverse events (SAEs), i.e. cardiovascular or autoimmune diseases need to be considered. A very important risk factor is liver cirrhosis, in particular in the advanced stages with signs of portal hypertension. At week 16 of triple therapy in the French CUPIC study including only non-responder patients with cirrhosis, a higher frequency of SAEs of 38–49% was found compared with phase III trials (9–14%); 1.3–2% of the patients died, emphasizing that safety must be carefully considered when evaluating the benefits and risks of treatment [12-15, 6].
Chances of a sustained virological response
Pretreatment predictors of response should significantly influence treatment decisions. Chances of SVR range from more than 80% in relapse patients without significant fibrosis to only 15% in cirrhotic null-responders. Still, accurate prediction can be difficult because there is no valid scoring system and experience with new regimens is limited in particular in difficult to treat patients. Some of the baseline markers for SVR that have been identified for PEG-IFN/RBV are still valid for the new triple therapy regimens. However, because of the higher antiviral potency, the impact of several predictive factors is less relevant for triple therapy. Useful markers include the stage of liver fibrosis, platelet count, HCV subtype and certain genetic polymorphisms within the IL28B promoter, which are also associated with spontaneous clearance of HCV infection [12, 13, 16, 17]. At baseline, the previous response to PEG-IFN/RBV has the highest predictive value, while on-treatment ‘lead-in’ response and extended rapid virological response (eRVR)1 are the most important markers of treatment response [13-16]. As there is a separate review focusing on the predictive factors of response to treatment, this will not be discussed in more detail .
The next generation of direct acting antiviral agents (DAAs) is in various stages of clinical development . Less complicated dosing regimens, a better safety profile and a lower incidence of AEs as well as an IFN-free treatment regimen can be expected in the not too distant future [19, 20]. The development of cirrhosis in chronic HCV infection is usually a long process. Patients with no or only mild liver fibrosis can wait until new drugs are approved. Still, delaying treatment until fibrosis has reached stage F3 (METAVIR) may be a less effective strategy than treating patients as soon as they have reached F2 . However, the ideal strategy for each patient may better be determined depending on the individual disease progression, which is influenced by the prevalence of risk factors such as steatosis or older age [22, 23]. Monitoring should include the regular assessment of the stage of fibrosis to determine disease progression and the best moment to initiate treatment. The introduction and increased use of non-invasive tests such as transient elastography have made this approach more feasible, although unfortunately the accuracy of these tests is limited, in particular for the intermediate stages of liver fibrosis [24, 25]. In addition, there might be non–liver-related issues motivating immediate treatment, i.e. extrahepatic manifestations or at the patient's request.
Based on the above mentioned factors, the ideal candidate for triple therapy should have the following characteristics: F2/F3 fibrosis, previous relapse and no significant co-morbidities, which means the need for treatment is high, the safety profile is reasonable and there is a significant chance of achieving an SVR. Unfortunately, the decision to treat is not necessarily that easy in most patients. Patients who are most urgently in need of treatment often present with cirrhosis and are older, which are both associated with higher risk. In addition, a history of previous treatment failure is associated with a poor chance of SVR except in cases of relapse from previous therapy. On the other hand, young, treatment-naïve patients with F0/F1-fibrosis have a moderate risk and a good chance of SVR, but can easily wait for better therapies.
Defining the ideal candidate for boceprevir
Choosing the optimal PI can also be difficult as there are no studies available directly comparing the two PIs approved for HCV. However, the phase III trials did not seem to show any significant difference in efficacy and only minor differences in the safety profile of these two drugs [12-15]. Nevertheless, because of the special characteristics of BOC and TLV one or the other of these PIs may be preferable in certain cases.
Because the ‘lead-in’ phase is part of the BOC treatment regimen, this drug offers advantages in certain ‘easy-to-treat’ patients who are treatment-naïve, without cirrhosis and with low pretreatment viral load . A certain number of these patients will become HCV RNA negative after the 4-week ‘lead-in’ and will thus achieve rapid virological response (RVR).2 About 90% of these patients achieve an SVR after 24 weeks of therapy if the treatment with PEG-IFN/RBV is continued alone. This rate does not increase by adding a PI as recently confirmed in a randomized trial [12, 14, 47]. Shortening treatment in these patients following response-guided therapy (RGT) is possible with dual therapy according to European guidelines. Thus, treatment with dual therapy should carefully be considered in these cases as it is associated with fewer AEs and is more cost-effective . However, this decision can be difficult in patients with a high baseline viral load (>800 000 IU/ml) as the recommended treatment duration without a PI would be longer (28 vs. 48 weeks respectively) . Nevertheless, it is possible to use TLV after a ‘lead-in’ phase as well, since this was tested in the REALIZE study . However, this makes it difficult to determine stopping rules and RGT criteria in those without a RVR.
During this era of limited resources, the cost of treatment is increasingly important. Costs per month are higher for TLV than for BOC (12 154€ vs. 3987€).3 On the other hand TLV is taken for a shorter time period of only 12 weeks, limiting the total cost to 36 463€3. In treatment-naïve patients without cirrhosis the cost of a full 48 weeks of treatment is similar for both PIs because of the 32 week course of BOC (31 894€ for BOC vs. 36 463€ for TLV). However, in those who meet the criteria for shortened BOC regimens, BOC is less expensive (23 920€ compared to 36 463€ for TLV)3. Boceprevir is also less expensive in case of early treatment failure because of AEs or virological non-response and the associated stopping rules. On the other hand in Europe, shorter treatment in relapse patients based on RGT has only been approved with TLV therapy, which may therefore be preferable in these cases. Treatment of non-responders or those with cirrhosis requires BOC for 44 weeks in addition to the 4 week ‘lead-in’ phase with PEG-IFN/RBV, which is again more expensive (43 855€ compared to 36 463€ for TLV)3 than 12 weeks of treatment with TLV (Table 1).
So far only limited data are available in patients with advanced liver disease. In the CUPIC study, TLV was associated with more SAEs, a higher rate of infections and a higher mortality rate than BOC. It should be noted that this was an interim-analysis after 16 weeks of treatment and unlike TLV, BOC must be continued for the entire treatment duration . Thus, no strict recommendations can be made based on safety issues. Prevalence of Anaemia seems to be the same for both first generation HCV PIs, BOC and TLV [12-15].
Although they have only been approved for HCV GT1 the antiviral potency of BOC and TLV is not strictly limited to HCV GT1. TLV has been shown to have good antiviral potency in HCV GT2, but not in HCV GT3 infection . In contrast, there are some data showing that BOC has limited, but greater efficacy than TLV in HCV GT3 [28, 29]. Thus, in patients co-infected with more than one HCV genotype, the specific HCV genotypes can support the use of one protease inhibitor of another.
As PIs are added to HCV therapy, physicians must be aware of DDIs. When TLV is chosen, this challenge is limited to a maximum of 12 weeks. In patients with certain co-morbidities, in which several DDIs are possible TLV may be the better choice because management is easier during the last 36 weeks with the PEG-IFN/RBV tail of treatment. In post-transplant patients, however, BOC has a less challenging extent of DDIs.
Normally, either PI can be used in patients eligible for triple therapy. The risks and potential benefits of both PIs seem to be balanced. Nevertheless, the patient may have a completely different point of view. Some subjects may find the strict 8-h regimen difficult and prefer a shorter period of PI treatment with TLV, which in addition seems to be as effective if taken twice daily at least in treatment-naive patients . Others may find the fatty meals required for sufficient absorption of TLV to be difficult and may therefore choose BOC. Patients may not tolerate a visible rash because of their social or work environment, while others may not tolerate dysgeusia. As treatment adherence is a key factor for successful therapy, the patient's personal preference based on the side effect profile will certainly strongly influence the choice of PI.