VALIDATION OF SURROGATE ENDPOINTS IN MELANOMA THERAPIES AND USE IN HTA
Author(s)
Hopkinson D1, Chadwick C1, Bibi M1, Bastian A2
1McCann Health, Macclesfield, UK, 2Incyte Corporation, Wilmington, DE, USA
OBJECTIVES: The use of surrogate endpoints in oncology trials may allow for a smaller sample size, a shorter study duration, and a more rapid time to market, if proven to be valid. We aimed to investigate the validation of surrogate endpoints in melanoma trials, and their use in health technology assessments (HTA). METHODS: We conducted a targeted literature review to identify studies assessing the validity of surrogate endpoints in melanoma, using the key words ‘surrogate endpoint’, ‘correlation’ ‘regression’ and ‘melanoma’. Searches were conducted on MEDLINE, EMBASE and Cochrane Library (2012-2017), ISPOR, ASCO and ESMO congresses (2012-2017), and HTA websites (NICE, SMC, HAS, PBAC, IQWIG and pCODR). RESULTS: Four meta-analyses and one phase III trial were identified that assessed the following surrogate measures for overall survival (OS): progression free survival (PFS), 1- and 2-year OS, recurrence-free-survival (RFS), early-tumour-response, time-to-progression (TTP), objective-response-rate (ORR) and disease-control-rate (DCR). Methods used for statistical validation included correlations between surrogate endpoints and OS (outcome level surrogacy, R), correlations between the treatment effects on surrogate endpoints and OS (trial level surrogacy, R), and calculation of surrogate threshold effect (STE). Analysis of immunotherapies in metastatic melanoma suggests that 1-year OS may be an appropriate surrogate for OS (R = 0.75); however, no clear correlations were observed between DCR/ORR/PFS and OS (R = 0.03/0.03/0.19). TTP was more strongly correlated with OS than early-tumour-response to vemurafenib in metastatic melanoma (τ = 0.655). Correlation between PFS and OS was reported for dacarbazine in metastatic melanoma (R = 0.71); however, IQWiG deemed the methodology insufficient for surrogate validation. Analysis of adjuvant treatment in resectable melanoma suggested that RFS could be an appropriate surrogate for OS (R = 0.91) and calculated an STE of 0.77. CONCLUSIONS: Robust and appropriate methodologies are required in order to validate surrogate endpoints in melanoma for use in HTA.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PRM7
Topic
Clinical Outcomes
Topic Subcategory
Clinical Outcomes Assessment
Disease
Oncology