REVIEW OF NICE HTA SUBMISSIONS INCLUDING METHODOLOGIES ADJUSTING FOR OVERALL SURVIVAL IN THE PRESENCE OF TREATMENT SWITCHING
Author(s)
Gurskyte L1, Muresan B2, Kulakova M2, Postma M3, Ouwens MJ4, Heeg B2
1Ingress-Health, Rotterdam, ZH, The Netherlands, 2Ingress-Health, Rotterdam, The Netherlands, 3University of Groningen, Groningen, The Netherlands, 4Astrazeneca, Mölndal, Sweden
OBJECTIVES: Often, technology appraisals (TAs) submitted to the National Institute for Health and Care Excellence (NICE) included methodologies that adjust trial overall survival (OS) for treatment switching. The objective is to assess how NICE responds to these methods, their criticism and recommendations for future submissions. METHODS: All NICE TAs that dealt with any stages of cancer and had been completed as of December 2001 were reviewed. We included reports published on the NICE website by the Independent Assessment Group or Evidence Review Group on TAs including naïve and more complex methodologies that adjust for treatment switch, such as inverse probability of censoring weights (IPCW), rank-preserving structural failure time (RPSFT) models, iterative parametric estimation (IPE) and the two-stage method. The following information was extracted: methodology used, acceptance by NICE of the adjusted outcomes, and overall criticism/recommendations. RESULTS: Adjustment for treatment switching was identified in 29 submissions. In most TAs, several methods were presented. The most prevalent combination was RPSTF and IPCW (12 submissions). Overall, the most used method was RPSTF (22 submissions) followed by the IPCW (13 submissions) and the two-stage method (8 submissions). Naïve and IPE methods were presented in five submissions each. In six TAs only a RPSFT model was presented. The stated preferred methodology by NICE was RPSTFM. The two-stage method was considered a better fit in six cases, because the common treatment effect assumption was not met or tested. IPCW was deemed inappropriate in the presence of a high percentage of switching patients. Using more than just one adjustment method was recommended. Reimbursements were not correlated to adjustment method(s) used or their number. CONCLUSIONS: These methods seem to be well accepted by NICE. For future submissions it seems important that multiple adjustment methods are presented.
Conference/Value in Health Info
2018-11, ISPOR Europe 2018, Barcelona, Spain
Value in Health, Vol. 21, S3 (October 2018)
Code
PCN64
Topic
Clinical Outcomes
Topic Subcategory
Relating Intermediate to Long-term Outcomes
Disease
Oncology