Vismodegib (1) – Erivedge®

• Clinical trials
  • To demonstrate additional benefit, the ERIVANCE trial (SHH4476g) is cited in the pharmaceutical manufacturer’s dossier. This is a single-arm, open-label Phase II trial.
  • In addition, the pharmaceutical manufacturer has cited further non-comparative, open-label Phase II trials (MO25616 (STEVIE), SHH4811g (US-EAP)) and a Phase I trial (SHH3925g) to assess adverse events.

a) Patients with symptomatic metastatic basal cell carcinoma

• The studies submitted by the pharmaceutical manufacturer for the medicinal product under assessment, vismodegib, are not suitable for demonstrating additional benefit, as the studies on vismodegib do not meet the inclusion criteria for the target population.
• According to the SmPC, vismodegib has marketing authorisation for the treatment of symptomatic metastatic basal cell carcinoma. The study’s inclusion criteria only considered patients with metastatic basal cell carcinoma, without taking symptoms into account.
• The pharmaceutical manufacturer did not present a separate analysis of the target population with symptoms in the dossier; consequently, no additional benefit can be inferred for this patient population.
• Furthermore, when considering the results, additional benefit could only be inferred on the basis of the overall clinical response rate. In patients with metastatic basal cell carcinoma, this was determined solely via imaging procedures, which in itself is not sufficient for classification as a patient-relevant endpoint.
• No patient achieved a complete tumour response.

b) Patients with locally advanced basal cell carcinoma for whom neither surgery nor radiotherapy is suitable

• Mortality – overall survival
  • In the ERIVANCE study, ‘overall survival’ was assessed as a secondary endpoint. At the time of the final confirmatory analysis, nine deaths (12.7% of 71 patients) had been recorded. The median overall survival had not been reached.
  • To assess the extent of this effect, the pharmaceutical manufacturer carried out a historical control with best supportive care based on case reports. The effect of best supportive care on overall survival is defined as unknown, as the pharmaceutical manufacturer was unable to identify any observational data for quantification. On the basis of the available data, it is therefore not possible to draw any conclusions regarding the extent of the additional benefit for the endpoint of overall survival compared with the appropriate comparator therapy (best supportive care).
• Morbidity – Objective Response Rate
  • The endpoint ‘objective response rate’ (ORR) was assessed as the primary endpoint in the ERIVANCE trial and showed a response to treatment with vismodegib in 27 of 71 recruited patients (final confirmatory analysis; 38.0% (95% CI: [26.9; 50.3])). The data on patients who responded to treatment are based on the efficacy-evaluable population (n = 63), as only in these patients was basal cell carcinoma confirmed at baseline by an independent pathologist or according to archived biopsy results. Of the 27 patients, 13 achieved a complete response (CR) and 14 achieved a partial response (PR).
  • The objective response rate represents a composite endpoint comprising the following morbidity parameters: assessment of clinical response, comprising external tumour size, degree of ulceration (in patients who presented with ulceration at baseline) and the occurrence of new lesions, and was additionally assessed using imaging procedures in accordance with RECIST criteria (RECIST 1.0) for lesions that could be visualised in this way. Furthermore, histological tumour biopsies of the target lesions taken at baseline and during the study were used to assess response.
  • Objective response is defined as complete or partial response, stable disease or progressive disease, assessed at two consecutive time points at least 4 weeks apart by an independent review panel (IRF). Clinical response was assessed on the basis of the three sub-endpoints using a defined algorithm. All three dimensions were initially assessed separately. The combination of the three individual parameters yields the Overall Clinical Response. The emergence of new lesions (≥5 mm) is the component with the highest weighting.
  • Due to the operationalisation of the Overall Clinical Response, a partial response could also be attributed to a 30% reduction in a single component of the endpoint. The clinical relevance of the resulting wide range in the quality of response, as captured by the ‘partial response’ endpoint, cannot be assessed.
  • Externally visible tumours and tumour ulcerations represent a burden for the affected patient. A considerable reduction in externally visible tumours and tumour ulcerations, up to and including complete remission, is to be regarded as clinically relevant. At the start of the study, 40 patients had one lesion, twelve patients had two lesions and only eleven patients had three or more lesions.
  • However, the pharmaceutical manufacturer has not provided separate results for the individual components of the ORR endpoint, nor has it provided any information on the location, extent of the lesions or the degree of ulceration at the start of the study and their progression during the study. In accordance with the study’s inclusion criteria, patients with small and therefore potentially less burdensome lesions (≥ 10 mm) were also eligible for inclusion. Consequently, the extent and relevance of tumour remission or ulceration remission cannot be clearly deduced from the data presented in the ERIVANCE study.
  • Furthermore, the data on the duration of the objective response to vismodegib are not yet sufficient to assess the sustainability of remission following a response. According to the operationalisation of the endpoint, a response lasting four weeks was sufficient to be classified as a complete response.
  • To assess the extent of the effect of treatment with vismodegib, the pharmaceutical manufacturer carried out an historical control. The efficacy of best supportive care in achieving complete or partial response was deemed to be non-existent due to a lack of reports on spontaneous remissions, and it was concluded that there was an increase in objective response rates with vismodegib treatment compared with the appropriate comparator therapy.
  • The available literature does not provide sufficient certainty as to whether, or to what extent, spontaneous remissions of advanced basal cell carcinoma can be expected. A rate of spontaneous remission on the same order of magnitude as that observed in the ERIVANCE study for the patient groups with complete remission (18.3 per cent) is not proven in the literature.
• Health-related quality of life
  • To assess health-related quality of life, the ERIVANCE study used the generic SF-36 questionnaire, measuring the change in score compared with baseline. Analysis of the ‘Mental Component Summary’ score shows a change from the baseline value at the start of the study of 1.2 points after 12 weeks and 1.0 points after 24 weeks. For the ‘Physical Component Summary’ score, changes in the point scales were -1.9 points after 12 weeks and -2.6 points after 24 weeks. The data at the end of the study are not used to assess any effect due to minor questionnaire response rates.
  • To assess the extent of the additional benefit, the pharmaceutical manufacturer carried out an historical control with best supportive care based on case reports. The pharmaceutical manufacturer defines the effect of best supportive care on health-related quality of life as unknown, as no observational data could be identified for quantification. On the basis of the available data on quality of life, it is therefore not possible to draw any conclusions regarding the additional benefit of vismodegib compared with the appropriate comparator therapy.

Courtesy translation only, please refer to the German original.