NICE Reconsideration for Oncology Drugs Based on Single-Arm Trials Referred to Cancer Drug Fund

OBJECTIVES: Submissions based on single-arm trials (SATs) often raise concerns from the National Institute for Health and Care Excellence (NICE), requiring additional evidence generation within the Cancer Drug Fund (CDF). NICE published the approval rate of re-appraisals of drugs referred to the CDF since 2016. The objective of this study was to review SAT-based re-appraisals referred to the CDF during the same period and compare them with the NICE publication.

METHODS: Published NICE oncology technical appraisals (TAs) based on SATs from July 2016 through August 2023 were reviewed to identify re-appraisals using evidence from the CDF. Full-text screening of the committee papers and guidance was conducted to extract information on new evidence identified within the CDF and committee commentaries.

RESULTS: Among 44 SAT-based TAs identified, 16 were referred to the CDF and five were fully re-appraised by NICE with additional clinical trials and real-world data from CDF. Four (80%) of the fully re-appraised TAs were recommended for routine use (versus the NICE published overall rate: 87%; 26/30) and one was declined. For the approved TAs, data from the CDF could demonstrate better clinical effectiveness with extended follow-up from the real world, provide evidence on the treatment’s life-extending potential and justify acceptability of cost-effectiveness estimates, and better reflect the local patient population and clinical practice. However, limitations due to a lack of comparative evidence normally remain upon exiting the CDF. For the rejected TA, new evidence was more representative of the local population but failed to resolve the uncertainty around the magnitude of the clinical benefit and justify high cost-effectiveness estimates.

CONCLUSIONS: For re-appraisals using CDF evidence, SAT-based submissions have a comparable approval rate for routine use versus overall oncology submissions. New evidence from the CDF can reduce uncertainty around data immaturity and generalizability but is insufficient to inform comparative effectiveness.