Are Landmark Survival Models Accepted in National Institute for Health and Care Excellence Health Technology Evaluations?

Speaker(s)

Davies C1, Emerson A1, Porteous A2
1Costello Medical, Boston, MA, USA, 2Costello Medical, London, UK

OBJECTIVES: Novel immuno-oncology therapies have necessitated increasingly complex survival models in health technology evaluations. The National Institute for Health and Care Excellence (NICE) Technical Support Document 21 (TSD21, November 2020) discusses several flexible models, including landmark survival models (LSMs). Considering these advancements in survival modelling, we investigated the use of LSMs in NICE Technology Appraisals (TAs) in oncology.

METHODS: The NICE website was searched on 13 May 2022 for completed oncology TAs. Information regarding survival modelling approach and External Assessment Group (EAG) and Committee opinion was extracted from all TAs that contained a LSM stratifying patients by treatment response. Analyses were undertaken to examine the rationale for LSM acceptance or rejection.

RESULTS: A total of 253 TAs were reviewed. Of these, only 5 (2%) contained LSMs, the majority of which (4/5) evaluated immuno-oncology therapies. In 2/5 TAs, LSMs were used in the base case analysis within the manufacturer’s cost-effectiveness model, while the EAG suggested use of LSMs in 2/5 cases. In the remaining TA, LSMs were explorative but received no further comment. The EAG disagreed with manufacturer use of LSMs in the base case in TA530 (May 2018), citing insufficient justification for complex LSMs over conventional models. The Committee shared the EAG opinion, noting it lacked evidence that LSMs were adequate for modelling long-term outcomes. In contrast, and though the Committee noted uncertainty, manufacturer LSMs were accepted for modelling long-term survival in TA763 (December 2021) after observing that cost-effectiveness results were similar with conventional models.

CONCLUSIONS: Despite advancements in survival analysis methodology, LSMs are not widely used in NICE TAs, which could reflect a lack of guidance for determining when LSMs should be considered. However, the recent acceptance of LSM estimates for long-term survival in TA763 following TSD21 publication may indicate that NICE is more willing to consider this approach.

Code

EE87

Topic

Economic Evaluation, Health Technology Assessment, Methodological & Statistical Research

Topic Subcategory

Cost-comparison, Effectiveness, Utility, Benefit Analysis, Decision & Deliberative Processes

Disease

STA: Drugs