Regorafenib (3) – Stivarga®

• Clinical trials
  • The CORRECT trial is a randomised controlled trial designed to investigate the therapeutic effects of regorafenib (505 patients) compared with placebo (255 patients) in the treatment of patients with metastatic colorectal cancer who had previously been treated with approved standard therapies and whose disease had progressed during or within 3 months of their last standard therapy.
  • The CONCUR trial is a randomised controlled trial in which regorafenib (136 patients) was also compared with placebo (68 patients).

Patients with metastatic colorectal cancer who had previously received fluoropyrimidine-based chemotherapy, anti-VEGF-based therapy, and, where KRAS wild-type, anti-EGFR-based therapy, or who are not eligible for such therapy

• mortality
  • overall survival
  • The results of the CORRECT trial show a statistically significant prolongation of overall survival for treatment with regorafenib compared with best supportive care. As at 21 July2011, the median survival in the regorafenib group was 196 days compared with 151 days in the best supportive care group, representing a median survival benefit of 45 days (hazard ratio: 0.77 [0.64; 0.94], p-value = 0.011).
  • This result from the CORRECT trial is not, in itself, confirmed by the results for the relevant patient population of the CONCUR trial, which show no statistically significant difference in overall survival (median 183 days vs. 203 days, hazard ratio: 0.68 [0.40; 1.18], p-value = 0.186).
• morbidity
  • Progression-free survival
  • In the CORRECT trial, the endpoint ‘progression-free survival (PFS)’ for regorafenib showed a statistically significant prolongation of progression-free survival compared with best supportive care (absolute prolongation: median 7 days, hazard ratio: 0.49 [0.42; 0.58], p < 0.001).
  • For the relevant patient population of the CONCUR trial, regorafenib also demonstrated a statistically significant prolongation of progression-free survival compared with best supportive care (absolute prolongation: median 5 days, hazard ratio: 0.32 [0.18; 0.57], p < 0.001).
  • The PFS endpoint is a composite endpoint comprising endpoints from the categories of mortality and morbidity. The ‘mortality’ component of the endpoint was assessed as a standalone endpoint in the CORRECT and CONCUR studies via the ‘overall survival’ endpoint. The morbidity component was not assessed on the basis of symptoms, but exclusively by means of imaging procedures.
  • Symptoms
  • The available analyses show, based on the relevant symptom scales, a statistically significant disadvantage of regorafenib compared with best supportive care in terms of its effects on the symptoms of fatigue, pain, loss of appetite and diarrhoea.
  • With regard to assessing the relevance of the statistically significant difference, the negative effect on diarrhoea is interpreted as a relevant effect. For the scales measuring fatigue, pain and loss of appetite, however, the 95% confidence interval of the standardised mean difference does not lie entirely above the irrelevance threshold used; consequently, it cannot be concluded that the effect is relevant.
  • With regard to the effects on symptoms, an overall disadvantage of treatment with regorafenib compared with best supportive care is observed.
• quality of life
  • The available analyses show a statistically significant disadvantage of regorafenib compared with best supportive care, both in the results for overall health status and in the results for the scales physical functioning, emotional functioning, cognitive functioning, role functioning and social functioning.
  • With regard to assessing the relevance of the statistically significant difference, the negative effect on role functioning and social functioning is interpreted as a relevant effect. For the scales measuring overall health status as well as physical functioning, emotional functioning and cognitive functioning, however, the 95% confidence interval for the standardised mean difference does not lie entirely below the irrelevance threshold used; consequently, it cannot be concluded that the effect is relevant.
  • With regard to the effects on quality of life, an overall disadvantage is observed for treatment with regorafenib compared with best supportive care.
• Side effects
  • Severe adverse events of CTCAE grade ≥ 3 occurred in significantly more patients in the regorafenib group compared with the best supportive care group in the CORRECT trial (78.0% vs. 49.0%). This difference is largely attributable to individual, severe adverse events of CTCAE grade 3, which occurred significantly more frequently with regorafenib – these include hand-foot syndrome (16.6 vs. 0.4%), fatigue (15.0 vs. 8.3 per cent), diarrhoea (8.2 vs. 2.0 per cent) and rash (5.8 vs. 0.4 per cent).
  • When considering the endpoints relating to side effects as a whole, a significant disadvantage of treatment with regorafenib is evident, as a considerably higher number of patients experienced severe adverse events whilst on regorafenib compared with best supportive care.
• Overall assessment
  • Results on mortality, morbidity, quality of life and side effects are available for the benefit assessment of regorafenib following the expiry of the authorisation period.
  • The available results on mortality show a significant prolongation of survival, which is offset by a significant disadvantage in terms of side effects, alongside negative effects of regorafenib on quality of life and disease-related symptoms.
  • The positive effect of regorafenib on survival is based on the available results from the CORRECT study, in which patients treated with regorafenib achieved a survival benefit compared with best supportive care, amounting to a median of 45 days (1st data cut-off) and 42 days (2nd data cut-off), respectively. This effect is not confirmed, on its own, by the results for the relevant patient population from the CONCUR study.
  • In the overall assessment, the positive relevant effect on survival is offset by negative relevant effects on the other patient-relevant endpoints: disease-related symptoms, quality of life and side effects. The G-BA has weighed up these positive and negative effects in a balancing decision, taking into account the advanced stage of the disease.
  • Taking into account the opinions of medical experts and the stage of the disease in question, the G-BA concludes in this balanced assessment that the patient-relevant negative effects of the new INN are so severe that the advantage in terms of the minor prolongation of survival achieved does not outweigh them in the overall assessment.
  • Consequently, following the renewed benefit assessment after the deadline has expired, it is concluded that the additional benefit for regorafenib in the treatment of adult patients with metastatic colorectal cancer who have previously been treated with available therapies or who are unsuitable for such therapies is not proven.

Courtesy translation only, please refer to the German original.