Sofosbuvir / Velpatasvir / Voxilaprevir (1) – Vosevi®

• Clinical trials
  • The POLARIS-2 trial was an open-label, randomised, multicentre Phase III trial involving 943 DAA-naïve adults with chronic HCV infection of all genotypes, with or without compensated cirrhosis.
  • The POLARIS-3 trial is an open-label, randomised, multicentre Phase III trial which enrolled 220 DAA-naïve adults with chronic HCV infection of genotype 3 and compensated cirrhosis.
  • The POLARIS-4 study is an open-label, randomised, multicentre Phase III study which enrolled 333 treatment-experienced adults with chronic HCV infection of genotypes 1, 2 and 3, with or without compensated cirrhosis.
  • The POLARIS-1 study is a randomised, multicentre Phase III study which included 416 adults with chronic HCV infection who had previously received NS5A inhibitor therapy.
  • The TRILOGY-3 study is an open-label, randomised, multicentre Phase II study which enrolled 49 DAA-treated adults with chronic HCV infection of genotype 1.

a) DAA-naïve patients without cirrhosis or with compensated cirrhosis, genotype 1

• The additional benefit is not proven.
• The pharmaceutical manufacturer has not provided any comparative data for patients with genotype 1 HCV infection. The descriptive analysis provided of a study arm within the patient population of the POLARIS-2 study for patients without cirrhosis is insufficient to demonstrate additional benefit compared with the appropriate comparator therapy. No data were provided for patients with compensated cirrhosis.

b) DAA-naïve patients without cirrhosis or with compensated cirrhosis, genotype 2

• The additional benefit is not proven.
• Overall, no additional benefit of sofosbuvir/velpatasvir/voxilaprevir compared with the appropriate comparator therapy, sofosbuvir/velpatasvir, can be established for this patient group.
• The pharmaceutical manufacturer has not provided any comparative data for patients with genotype 2 HCV infection and compensated cirrhosis. Overall, therefore, no additional benefit can be inferred for DAA-naïve patients with genotype 2 infection.
• mortality
  • No deaths occurred in the patient population under consideration.
• morbidity
  • Morbidity was assessed based on sustained virological response 12/24 weeks after the end of treatment. The endpoints were achieved by 47 out of 49 patients (95.9%) in the treatment arm and 40 out of 40 patients (100%) in the comparator arm. There is no statistically significant difference.
• Health-related quality of life
  • Health-related quality of life was assessed using the SF-36 questionnaire. No statistically significant differences were observed, except for the mental health summary score. The difference in this subscore is not considered clinically relevant (standardised mean difference, Hedges’ g = –0.43 [95% confidence interval: –0.84; 0.00]).
  • The pharmaceutical manufacturer also considers the CLDQ-HCV (HCV version of the Chronic Liver Disease Questionnaire), the FACIT-F (Functional Assessment of Chronic Illness Therapy – Fatigue) and the WPAI Hepatitis C (HCV-specific version of the Work Productivity and Activity Impairment questionnaire). These instruments cannot be used for the assessment. The validity of the CLDQ-HCV is not proven. No validation of the FACIT-F for patients with CHC is available. The assessment of work absences and occupational limitations using the WPAI Hepatitis C is insufficient for evaluating health-related quality of life; furthermore, there is a lack of information on its validity.
• Side effects
  • No statistically significant differences were observed in adverse events, serious adverse events or discontinuation due to adverse events. No differences were identified in specific adverse events; therefore, there is neither an advantage nor a disadvantage for this endpoint.

c) DAA-naïve patients without cirrhosis or with compensated cirrhosis, genotype 3

• An additional benefit is not proven.
• Based on the results regarding adverse events, therefore, neither less benefit nor an additional benefit can be inferred.
• mortality
  • No deaths occurred in the patient population without cirrhosis under consideration. One death occurred in the active treatment arm among patients with cirrhosis (not statistically significant).
• morbidity
  • Morbidity was assessed based on sustained virological response 12/24 weeks after the end of treatment. The endpoints were reached by 91/92 patients (98.9%) in the active treatment arm and 86/89 patients (96.6%) in the comparator arm of the patient population without cirrhosis, and by 106/110 patients (96.4%) in the active treatment arm and 105/109 patients (96.3%) in the comparator arm of the patient population with cirrhosis. In each case, there was no statistically significant difference.
• Health-related quality of life
  • Health-related quality of life was assessed using the SF-36 questionnaire. No statistically significant differences were observed for either of the two patient populations.
  • The pharmaceutical manufacturer also considers the CLDQ-HCV (HCV version of the Chronic Liver Disease Questionnaire), FACIT-F (Functional Assessment of Chronic Illness Therapy – Fatigue) and WPAI Hepatitis C (the HCV-specific version of the Work Productivity and Activity Impairment questionnaire). These instruments cannot be used for the assessment. The validity of the CLDQ-HCV is not proven. No validation of the FACIT-F for patients with CHC is available. The assessment of work absences and occupational limitations using the WPAI Hepatitis C is insufficient for evaluating health-related quality of life; furthermore, there is a lack of information on its validity.
• Side effects
  • With regard to adverse events, on the one hand, a statistically significant difference in favour of sofosbuvir/velpatasvir/voxilaprevir was observed in the ‘psychiatric disorders’ system organ class among patients without cirrhosis (p=0.009, ARR=14.8%); on the other hand, in patients with cirrhosis, there was a statistically significant difference to the detriment of sofosbuvir/velpatasvir/voxilaprevir in the events ‘nausea’ (p=0.015, ARR=–11.7%) and ‘diarrhoea’ (p=0.008, ARR=–10.9%). No statistically significant differences were observed for serious adverse events or discontinuation due to adverse events.
  • The results regarding adverse events must be regarded as potentially highly biased, as there is a 4-week difference in the duration of treatment – and thus the observation period – between the active treatment arm and the comparator arm. Furthermore, in the absence of blinding, the subjective assessment of the endpoints nausea, diarrhoea and psychiatric disorders may also lead to bias. Owing to these uncertainties, it is not possible to establish with sufficient certainty either a hint of less benefit or a hint of increased harm.
  • Furthermore, it is evident from both the EMA’s assessment report and the written and oral submissions that the gastrointestinal side effects are predominantly mild in nature and do not have a therapy-limiting impact. The side effect of psychiatric disorders, which is largely attributable to the PT ‘insomnia’, is also to be regarded as minor given the short duration of treatment.

d) DAA-naïve patients without cirrhosis or with compensated cirrhosis, genotype 4

• The additional benefit is not proven.
• The pharmaceutical manufacturer has not provided any comparative data for patients with genotype 4 HCV infection. The descriptive analysis provided of a study arm within the patient population of the POLARIS-2 study for patients without cirrhosis is insufficient to demonstrate additional benefit compared with the appropriate comparator therapy. No data were provided for patients with compensated cirrhosis.

e) DAA-naïve patients with or without compensated cirrhosis, genotypes 5 or 6

• An additional benefit is not proven.
• The pharmaceutical manufacturer has not provided any comparative data for patients with HCV infection of genotype 5 or 6. The descriptive presentation provided of a study arm comprising patient populations from the POLARIS-2 study for patients without cirrhosis is insufficient to demonstrate additional benefit compared with the appropriate comparator therapy. No data were provided for patients with compensated cirrhosis.

f) Patients with prior DAA treatment, with or without compensated cirrhosis

• The additional benefit is not proven.
• Overall, no hint of additional benefit can be derived for patients with prior DAA experience.
• mortality
  • No statistically significant differences in mortality were observed in the patient population under consideration.
• morbidity
  • Morbidity was assessed based on sustained virological response 12/24 weeks after the end of treatment. These endpoints were achieved by 97.4% (genotype 1) and 100% (genotype 2) of patients in the active treatment arm, and by 90.9% (genotype 1) and 97% (genotype 2) of patients in the comparator arm. There was no statistically significant difference in patients with genotype 1 or 2 infection. In patients with genotype 3 infection, there was a statistically significant difference between sofosbuvir/velpatasvir/voxilaprevir and sofosbuvir/velpatasvir (96.3% vs. 84.6%, p=0.041; ARR=11.7%). Due to the aforementioned uncertainty regarding the possible omission of ribavirin in the comparator therapy, no additional benefit can be inferred from this.
• Health-related quality of life
  • Health-related quality of life was assessed using the SF-36 questionnaire. No statistically significant differences were found in any of the patient populations.
  • The pharmaceutical manufacturer also considers the CLDQ-HCV (HCV version of the Chronic Liver Disease Questionnaire), FACIT-F (Functional Assessment of Chronic Illness Therapy – Fatigue) and WPAI Hepatitis C (the HCV-specific version of the Work Productivity and Activity Impairment questionnaire). These instruments cannot be used for the assessment. The validity of the CLDQ-HCV is not proven. There is no validation of the FACIT-F for patients with CHC. The assessment of work absences and occupational limitations using the WPAI Hepatitis C is insufficient for evaluating health-related quality of life; furthermore, there is a lack of information on its validity.
• Side effects
  • With regard to adverse events, a difference to the detriment of sofosbuvir/velpatasvir/voxilaprevir was observed in patients with genotype 1 or 2 infection for the endpoint ‘diarrhoea’.
  • However, due to the uncertainty surrounding the subjective recording of events described above – which was also identified in the POLARIS-4 study – there is no hint that there is any evidence of less benefit.
• Overall assessment
  • For patients with hepatitis C infection of genotypes 1, 4, 5 and 6 (without compensated cirrhosis) who have not previously been treated with DAAs, the pharmaceutical manufacturer has provided descriptive analyses of patient populations from individual study arms without comparison to the appropriate comparator therapy. No additional benefit can be inferred from these data in any case.
  • No data are available for DAA-pretreated patients with compensated cirrhosis (genotypes 1, 2, 4, 5 and 6).
  • For patients with genotype 2 hepatitis C infection who have not previously been treated with DAAs, the pharmaceutical manufacturer has submitted data from the POLARIS-2 study. No statistically significant differences were observed between sofosbuvir/velpatasvir/voxilaprevir and the appropriate comparator therapy. The study included only patients without cirrhosis. No data were submitted for patients with compensated cirrhosis. There is no hint of additional benefit for patients with genotype 2.
  • For patients with genotype 3 hepatitis C infection who had not previously been treated with DAAs, the pharmaceutical manufacturer has submitted data from the POLARIS-2 and POLARIS-3 studies. Statistically significant differences were observed in the adverse events ‘psychiatric disorders’, ‘nausea’ and ‘diarrhoea’; however, due to uncertainties in the study design and the predominantly mild nature of these events, these cannot be regarded as sufficiently meaningful in terms of additional benefit. Overall, therefore, no additional benefit can be inferred for patients with genotype 3.
  • For patients with prior DAA treatment, the pharmaceutical manufacturer presents results from the POLARIS-4 study involving patients with genotypes 1, 2 and 3. Statistically significant differences in the morbidity endpoint category (SVR 12 in patients with genotype 3) cannot be used to infer additional benefit due to uncertainties regarding the implementation of the comparator therapy. Similarly, due to the uncertainties described, no hint of harm can be inferred for the endpoint ‘diarrhoea’. The study included only patients who had not previously been treated with an NS5A inhibitor. No comparative data were presented for patients with other genotypes or who had received prior treatment with an NS5A inhibitor. Overall, therefore, no additional benefit can be inferred for patients who have previously been treated with DAAs.

Courtesy translation only, please refer to the German original.