How to optimize current therapy of HCV genotype 1 infection with boceprevir

Authors


Abstract

Treatment with first generation protease inhibitors (PIs) is a milestone in the history of HCV therapy. Triple therapy with boceprevir (BOC) improves sustained virological response (SVR) by 30% in treatment naïve genotype 1 patients and by 50–60% in relapsers, 40–45% in partial responders and 25% in null responders compared with the Pegylated Interferon (PEG-IFN) and ribavirin regimen. To optimize BOC treatment, screening and access to treatment must be improved in genotype 1 patients. To select the ideal candidate for immediate treatment with triple therapy, an individual risk/benefit ratio must be assessed. Recent data have shown that patients with compensated cirrhosis and more advanced disease may also benefit from this regimen. Moreover, in HCV patients with extrahepatic manifestations, patients with HCV recurrence after liver transplantation and HIV-HCV co-infected patients, immediate treatment with triple therapy should be discussed. There is growing evidence that triple therapy with BOC is cost-effective in genotype 1 patients. Finally, the treatment design of BOC must be optimized in relation to baseline characteristics, so that optimal stopping rules can be followed, Drug-drug interactions (DDIs) can be prevented and AEs can be accurately prevented and managed.

Abbreviations
AEs

adverse effects

BOC

boceprevir

DAAs

direct acting antiviral agents

DDI's

drug-drug interactions

EPO

erythropoietin

HCV

hepatitis C virus

HTA

host targeting agents

PEG-IFN

Pegylated Interferon

PIs

protease inhibitors

RGT

response guided therapty

SAEs

serious adverse effects

SVR

sustained virological response

Chronic hepatitis C virus (HCV) infection is a major cause of cirrhosis and hepatocellular carcinoma and is the main cause of liver transplantation worldwide. This global health problem affects more than 160 million people in the world [1]. Liver related morbidity and mortality from all causes can be significantly reduced by the eradication of the virus [2]. Until 2011 the treatment of HCV was based on a combination of pegylated interferon alpha and ribavirin (PEG-IFN/RBV) and less than 50% of HCV genotype 1 infected patients achieved a sustained virological response (SVR) [3-5]. In the last decade, antiviral therapy reduced the cumulative incidence of cirrhosis by 7% and of liver related deaths by 3.4% overall in western European countries [6]. The launch of first generation protease inhibitors (PI) boceprevir (BOC) and telaprevir (TLV) in 2011 dramatically improved SVR rates in genotype-1 patients [7-13]. In phase III studies, SVR rates improved by nearly 30% with triple therapy with both PI compared with PEG-IFN/RBV, reaching 63 with BOC and 75% with TLV in treatment naïve genotype-1 patients. The benefit was even greater in treatment-experienced patients: the chance of a SVR increased by 50–60% in relapsers, 40–45% in partial responders and 25% in null responders compared with the PEG-IFN/RBV regimen. The global SVR rate of triple therapy is 59–66% in treatment-experienced patients. However, this fantastic progress is accompanied by additional as well as more severe adverse events (SAEs), increased costs of treatment and new drug–drug interactions.

The aim of this article is to review available data to optimize the current use of BOC in HCV Genotype-1 patients in relation to patient selection as well as treatment design. We will discuss the individual interests of the patients as well as the socioeconomic point of view.

Selection of ideal candidates for treatment according to a benefit/risk ratio

According to European Association for the Study of the Liver guidelines, almost all patients with chronic HCV are candidates, if they are willing and in the absence of contra-indications to PEG-IFN/RBV and antiviral treatment [14]. However, the dynamics and natural history of HCV infection as well as access to screening and treatment vary widely in the countries in Western Europe. Therefore, the impact of antiviral therapy, i.e. reduced incidence of cirrhosis and liver related deaths also varies substantially. The estimated HCV screening rate varies from 34% in United Kingdom to 64% in France [6]. Access to HCV treatment among screened patients varies according to the stage of fibrosis, is generally easier in those with significant fibrosis, and depends on country's guidelines and resources. According to country-specific Markov models, the use of triple therapy in genotype-1 patients reduces the cumulative incidence of cirrhosis by 17.7% and overall mortality by 9.7% if a public health policy based on population guided therapy is implemented in addition to common guidelines based on virological response-guided therapy (RGT). Thus, to optimize the current use of BOC or TLV the first goal is to improve screening and access to treatment.

However, not all genotype-1 chronic HCV patients should receive a triple regimen [15]. In a German cohort in a real world setting only 41% of HCV patients eligible for treatment initiated triple therapy mostly comorbidities, mild fibrosis or the patient's wishes. This shows that despite the improvement in SVR, more efficient triple therapy and safer treatment options are urgently needed. Several factors should be considered when selecting the best patients: therapy-associated risks, chances of achieving SVR and treatment urgency.

Therapy-associated risk/adverse events

Triple therapy with first generation PI increases the rate of adverse events (AEs) compared with PEG-IFN. The use of BOC in phase III trials is associated with an increase in the occurrence of fatigue, anaemia, nausea, diarrhoea, dysgeusia and neutropaenia. Closer monitoring, management of AEs and careful patient education is important to improve compliance and reduce premature treatment discontinuation. In addition, interactions among different drugs in the regimen may affect the efficacy and/or safety of therapy and must be managed appropriately in each patient.

Patients with cirrhosis are especially vulnerable. Patients with cirrhosis treated with BOC in phase III trials have a higher frequency of serious AEs than either patients with cirrhosis and patients with milder fibrosis treated with PEG-IFN/RBV: 18, 6 and 12% respectively [16]. In the real world setting of the French CUPIC cohort of 212 treatment-experienced patients with cirrhosis who were mostly Child Pugh A (93%), with a mean MELD score of 8 and with portal hypertension in less than 1/3 of the patients treated with BOC, 44.3% of the patients had developed SAEs at 12 weeks post treatment. Three of212 patients (1.4%) died, severe infections occurred in 8 (3.8%) and hepatic decompensation occurred in 9 (4.2%) [17]. Only two baseline factors were associated with the occurrence of death in multivariate analysis; infection and/or hepatic decompensation: platelet count <100,000/mm3 and serum albumin <35 g/L [18]. An analysis of the SVR at 12 weeks post-treatment with BOC or TLV according to baseline albumin levels and platelet count revealed a subgroup of patients (8.3% of the population) with a high risk (51.4%) of severe complications and a low rate of SVR12 (27%). These results suggest that this population should not be treated with triple therapy. In contrast, a large subgroup of patients with platelet count >100,000/mm3 and serum albumin >35 g/L (68.3% of the patients) were shown to have a low risk of severe complications (6.2%) and a high SVR 12 rate (54.9%). These patients could benefit highly from triple therapy. Patients with either platelet count <100,000/mm3 or serum albumin <35 g/L, have a moderate risk of developing severe complications (12–16%), associated with a moderate efficacy of triple therapy (29–36%). These patients should be treated with caution and carefully monitored. Prophylactic antibiotic treatment with norfloxacin should be proposed in patients with advanced cirrhosis treated by PEG-IFN [19]. In this particular population the risks and benefits should be discussed in detail with the patient, as well as the option of delaying treatment to wait for new more potent and better-tolerated DAA regimens.

Chance of sustained virological response (SVR)

Baseline predictors of a response to triple therapy with PIs are slightly different than with PEG-IFN/RBV. In treatment naïve patients, predictive factors of a SVR with BOC triple therapy are: low viral load (<400,000 IU/ml), IL-28B genotype 12979860 CC vs. non-CC, the absence of cirrhosis, HCV subtype 1b and non-black race [20]. In treatment-experienced patients the only baseline predictor of response is a previous response to PEG-IFN/RBV. Previous relapsers have a higher rate of SVR than previous non-responders [20]. However, on-treatment predictive factors of SVR are stronger than baseline parameters. A ≥ 1-log10 IU/ml decrease in HCV RNA at week 4 of treatment is the strongest predictive factor of SVR whatever the IL-28B polymorphisms. In the real world setting of the French CUPIC cohort, previous treatment response to PEG-IFN was the strongest baseline predictive factor of SVR12 but platelet count >100,000/mm3 and subtype 1b were also associated with SVR 12. Again, on-treatment virological response at treatment weeks 4 (at the end of the lead-in-phase) and 8 also influenced SVR rates. Patients with undetectable HCV RNA at treatment week 8 achieved a SVR12 more frequently than those with detectable HCV RNA, 71.6% vs. 25.2% (P < 0.0001). SVR rates were higher in patients who exhibited a ≥ 3 log10 decline in viral load from baseline or had undetectable HCV RNA at treatment week 8 (58.4%) than in those with a < 3 log10 decline in viral load from baseline 6.3% (P < 0.0001).

Treatment urgency: treat or wait? Which patients?

More than 30 direct acting antiviral agents (DAAs) or host targeting agents are currently in various stages of clinical development [21]. Simple dosing regimens, better safety profiles and a lower incidence of AEs as well as IFN-free regimens can be expected within the next 1 or 2 years. However, the availability of these drugs will differ from one country to another and the price of this regimen could be an issue in each country. Therefore, it is very difficult to determine when these regimens will be available to patients on an individual basis. Different groups of patients could benefit from immediate treatment with the most potent existing regimen.

Patients with no or mild fibrosis can wait until new drugs have been approved, although they have the most favourable response to triple therapy of all the subgroups. On the other hand it could be detrimental for patients with more advanced fibrosis to wait, as they could progress rapidly and reach more advanced stages of disease associated with the development of cirrhosis-related complications, a lower response-rate and less tolerance to triple therapy. Patients with extrahepatic manifestations of HCV could benefit from the triple regimen whatever the stage of fibrosis. In a small on-going prospective cohort study, 23 patients with HCV-mixed cryoglobulinaemia vasculitis were treated with triple therapy. Nearly 70% of these patients had a virological response at week 24. Moreover, the cryoglobulin level decreased and the C4 level increased during treatment. However grade 3 and 4, mostly haematological, adverse events were observed in 43.5% of patients and 48% of the patients developed infection, mostly urinary and pneumopathy, during treatment [22]. Therefore, although triple therapy seems to be highly effective in this population it should be administered cautiously because of the high rate of side effects.

HIV-HCV coinfected patients could also benefit from the triple regimen even though HIV antiretroviral treatment must be adapted to avoid an interaction with HCV protease inhibitors. Recent studies have shown that a SVR24 was achieved in 63% of treatment-naïve coinfected HIV-HCV patients with BOC triple therapy with an identical safety profile as in monoinfected patients [23]. Moreover, drug-drug interactions between BOC and HIV protease inhibitors did not have a clinically significant effect on HCV response or HIV control in this study. Similar benefits were observed in treatment-experienced co-infected patients treated with BOC [24]. HCV recurrence after liver transplantation is a challenge with poor results with PEG-IFN/RBV. A recent cohort study has assessed the efficacy and tolerability of triple therapy with either BOC or TLV in this setting. Among 37 liver transplants recipients, half of them treatment-naïve and half of them treatment-experienced, 18 were treated with BOC and 19 with TLV. And end of treatment virological response was obtained in 89 and 58% of the patients treated with BOC and TLV respectively (P = 0.12). Treatment was discontinued in 16 patients owing to either treatment failure (n = 11) or AEs (n = 5). Infections developed in 10 patients (27%) with three fatal outcomes. Anaemia was observed in almost all patients (92%). Thirty five percent of the patients received red blood cell transfusions despite erythropoietin injections and RBV dose reductions. Cyclosporine and tacrolimus doses were reduced by 1.8 ± 1.1- and 5.2 ± 1.5-fold, respectively, with BOC. These results show that triple therapy is highly potent in this setting but is associated with frequent side effects and significant but manageable drug-drug interactions requiring caution and close monitoring. Data on HCV treatment with triple therapy in haemodialysis patients are limited. Results suggest that a triple regimen with BOC is feasible in this population with a low dose of RBV [25]. Finally, young women planning pregnancy may wish to be treated to prevent HCV mother to child transmission.

Choice of protease inhibitors: BOC vs. TLV

There are no published studies comparing BOC and TLV in HCV genotype 1 infection. However, phase III trials of each drug did not show any significant difference in efficacy between the two drugs. Safety profiles appear to be comparable for haematological side effects. However, there are certain differences in the occurrence of more specific side effects such as dermatological side effects and anorectal disorders with TLV and dysgeusia with BOC [26]. The duration of the triple regimen is different between the two drugs: 12 weeks with TLV, 2444 weeks with BOC. Moreover, the total duration of treatment may be different depending on the label. Treatment naïve-patients with undetectable HCV RNA after 4 weeks of triple regimen receive a total of 24 weeks of treatment with TLV [or 12 weeks in case of favourable IL28-B CC genotype [27]] and 28 weeks of BOC. Previous relapse patients with undetectable HCV RNA after 4 weeks of triple regimen receive a total of 24 weeks of TLV and 36 weeks in the USA or 48 weeks in Europe of BOC. Despite the small difference in tolerability and treatment schedule and duration between the two drugs, either of these PIs can be used in HCV patients eligible for triple therapy. Even if the potential risks and benefits of both PIs seem to be similar, patients may prefer one or the other for timing, treatment pill burden, and dietary requirements. Some patients may find the strict 8 h regimen with BOC more difficult than the twice-daily TLV regimen [28]. Other patients may find the fatty meals required for sufficient absorption of TLV difficult and may therefore choose BOC. Patients may worry about rashes and the more severe dermatological side effects while others may worry about dysgeusia. Because adherence to treatment is a key factor for successful therapy the patient's personal preference in relation to side effects profile, treatment duration, practical considerations such as pill burden, dietary requirements and mode of daily administration will strongly influence the choice of PI.

Cost-effectiveness studies: the socioeconomic point of view

Cost per month is higher for TLV than for BOC, but the duration of triple therapy is shorter with TLV [29]. In treatment-naïve patients without cirrhosis the cost of a full 48 weeks of treatment are similar for both PIs. BOC is less expensive in those who meet the criteria for shorter treatment. In contrast, TLV is less expensive for the treatment of previous null responders and in those with cirrhosis [29]. Data on the cost effectiveness of triple therapy are emerging. A cost-effectiveness analysis was performed in the Spanish national healthcare system, based on demographic characteristics and treatment results of naïve genotype-1 patients in the clinical SPRINT and ADVANCE trials, and on base-case analysis using dual therapy as the comparator. This study found favourable incremental cost-effectiveness-ratios for triple therapy (either TLV or BOC) especially in ILB28-guided therapy in patients younger than 60 years [30]. A Portuguese analysis found that adding BOC to PEG-IFN was cost-effective for both previously treated and untreated patients in a model based on Portugal-specific annual direct costs of HCV health states evaluations [31]. In a recent analysis performed in the USA in treatment-experienced genotype 1 patients, BOC increased quality-adjusted life-years and reduced the lifetime incidence of liver complications [32]. In addition, the BOC-based treatment is projected to be cost effective compared with PEG-IFN/RBV alone at common willingness-to-pay thresholds. Moreover, in another study, both universal triple therapy and IL-28B-guided therapy are cost-effective when used in patients with advanced fibrosis [33]. It is still unknown if treating patients with mild fibrosis is cost-effective. Recent data suggest that lifetime costs with triple therapy at a moderate stage of fibrosis (F2) are lower than at later stages (F3, F4) suggesting that this will be more cost-effective [34].

Optimal treatment design

According to the label, all patients treated with BOC are supposed to be treated with a 4-week lead-in phase with PEG-IFN/RBV. In naïve patients without cirrhosis RGT is possible with 24 weeks of triple regimen following the lead-in phase (28 weeks total treatment) if HCV RNA becomes undetectable at week 8 and until week 24 of therapy, which is defined as an eRVR. In contrast, previous non-responders and patients with cirrhosis need to be treated for a fixed duration of 48 weeks including the lead-in phase and 44 weeks of triple therapy with BOC. There is a difference between the Food and Drug Administration (FDA) and European Medicines Association (EMA) labelling for patients with a previous partial response or relapsers. FDA recommends a RGT approach while EMA suggests a 48 week fixed duration of treatment with a lead-in followed by 32 weeks of the triple regimen and ending with 12 weeks of PEG-IFN (Fig. 1). It is difficult to judge which of the FDA or EMA labels represents the optimal strategy. Although the EMA approach appears to be reasonable to avoid relapse the decision is based on just seven patients who relapsed in the RGT arm of the RESPOND-2 study.

Figure 1.

Recommended treatment regimen with boceprevir.

‘Lead-in’ phase or no ‘lead-in’ phase?

The value of the lead-in phase with PEG-IFN is controversial because it is only used with the BOC regimen. The initial rationale for a lead-in strategy was to lower the likelihood of resistant variants by reducing viral load prior to BOC administration. In the SPRINT-1 study, SVR rate was higher, but not statistically significant in the lead-in arm and there were also fewer relapses and breakthroughs. Based on this, all the other studies with BOC included a ‘lead-in phase.’ The lead-in strategy was subsequently studied with TLV in treatment-experienced patients in the REALIZE study. The lead-in phase with PEG-IFN/RBV offers no clearly proven virological benefit, except for a shorter duration of treatment in naïve patients without cirrhosis treated with BOC or in relapse patients treated with TLV [35]. The lead-in phase differentiates between two groups of patients in the prediction of the effectiveness of triple therapy: (a) those who could benefit from 24 weeks of standard therapy with PEG-IFN/RBV because of low baseline viral load, favourable IL-28B CC genotype and mild fibrosis and (b) those who are poor IFN responders (<1 log10 IU/ml decline at week 4), and especially prior non-responders in whom initiation of triple therapy should be discussed [36]. This approach could also be used to tailor PEG-IFN/RBV dosage during the lead-in phase in patients with poor tolerance although this strategy must be studied further in patients with advanced fibrosis and cirrhosis [35]. There is no clear evidence that justifies the use of lead-in phase in any of the other categories of patients.

Non-responders

Data on BOC in non-responders to PEG-IFN are limited. The results of the PROVIDE study showed that 40% of non-responders achieve a SVR after 48 weeks of treatment including 44 weeks of triple therapy with BOC. However, in non-responders with cirrhosis the response rate appears to be much lower and alternative strategies are urgently needed [37].

Optimal stopping rules

Treatment should be stopped if HCV RNA is >100 IU/ml at week 12 or with detectable HCV RNA at week 24. Post hoc analyses of these stopping rules have clearly shown that the combination of these two stopping rules maximized early discontinuation of futile therapy and minimized premature treatment discontinuation in both treatment naïve and treatment experienced patients [38]. Pole analysis of the five phase III studies provides some support for a new stopping rule at week 8 in patients with severe fibrosis or cirrhosis. None of the 22 patients who were above the threshold of a < 3 log10 decline in viral load at week 8 achieved a SVR [16].

Optimal therapy management

Prevention of DDI's is a new challenge with DAA regimen. BOC is an inhibitor and substrate of the p-glycoprotein and cytochrome P450 3A4/5 which metabolizes more than 50% of the clinically used drugs and is often involved in DDIs [39]. Knowledge and awareness of drug interactions are therefore crucial. All physicians should be aware of potential DDIs and some help can be obtained at different websites as www.hep-druginteractions.org. Patients should also be informed that they must not self-medicate and must avoid herbal products.

Adherence to therapeutic regimens affects the efficacy of treatment. Achieving a SVR with BOC is more dependent upon adherence to the assigned duration of treatment than adherence to the three times daily dosing interval. Adherence to >60% of three times a day dosing with BOC is important in patients who failed previous therapy [40].

Management of AEs is essential to optimize safety and adherence. Anaemia is highly prevalent during triple therapy. RBV dose reduction and erythropoietin (EPO) use are the two main options to manage anaemia. In the era of PEG-IFN/RBV the IDEAL study showed that early RBV dose reduction could impair SVR and that use of EPO was associated with a higher SVR [41]. However, a recent randomized trial in patients treated with BOC, shows that RBV dose reduction and EPO use did not influence the SVR and were equally effective in managing anaemia whether HCV RNA was detectable or not [42]. Therefore, RBV dose reduction should be the first line approach in case of triple regimen-induced anaemia.

Conclusion

Triple therapies with PIs are a major advance in the history of HCV treatment. Optimal patient selection is crucial to achieve high SVR rates with a reasonable safety profile. Real-life data on patients with cirrhosis have shown the benefit of triple therapy in the most difficult-to-treat patients. They also define the limits of this treatment. Patients with cirrhosis, a platelet count <100,000/mm3 and serum albumin <35 g/L should not be treated. Recent data have shown the benefits of the triple regimen with BOC in HCV patients with extrahepatic manifestations, in post liver transplantation HCV recurrence and in HIV-HCV co-infected patients. Optimizing BOC treatment includes optimizing the treatment design according to baseline characteristics, following optimal stopping rules, preventing DDIs and preventing and managing AEs.

Disclosure

M. Bourlière has served on the Speaker bureaus of Roche, Bristol Myers Squibb, GSK, Gilead, Merck, Janssen; Consultant: Roche, Bristol Myers Squibb, GSK, Gilead, Merck, Janssen, Boehringer Ingelheim, Vertex, Novartis and Abbvie. X. Adhoute, A. Wendt, C. Ansaldi, V. Oules and P. Castellani have no disclosures.

Ancillary