Presentation (CTI)

OBJECTIVES: Since 2016, the Cancer Drugs Fund (CDF) enables conditional access to cancer medicines in England where clinical uncertainty prevents the National Institute for Health and Care Excellence (NICE) from making a routine commissioning recommendation. Real-world data collected via the Systemic Anti-Cancer Therapy (SACT) dataset and 'Blueteq', and analysed by Public Health England (now NHS England) informs re-appraisal. Recent evidence questions the utility of real-world data in resolving uncertainties that triggered CDF entry. This project demonstrates a proof-of-concept using scalable methodology and routine data, without Blueteq, to replicate and extend real-world effectiveness analyses that might inform NICE Technology Appraisals (TA).
METHODS: Analyses conducted in NICE re-appraisals were replicated for four CDF topics (TA766, TA795, TA780, and TA870) using linked SACT, cancer registry, and admissions data (provided by NHS England under agreement DARS-NIC-701654-Q5Z9T-v0.2). Cohorts were derived using eligibility criteria in CDF documents. Kaplan-Meier curves were plotted over equivalent follow-up periods, and extended to March 2024. Concordance with re-appraisal results was assessed using patient characteristics (age, gender, and performance status) and survival outcomes.
RESULTS: Across the four topics, 21,912 patients were identified using SACT data, with final eligible cohorts of 472 to 2,230. Demographics were broadly comparable to re-appraisal analyses for TA780, TA795, and TA870 with closely aligned survival estimates and overlapping confidence intervals. Extended follow-up showed a stabilisation of survival curves over time. Conversely, TA766 diverged from re-appraisal analyses in patient characteristics and survival outcomes, reflecting challenges of applying complex eligibility criteria to routine data.
CONCLUSIONS: Real-world datasets can feasibly reproduce survival estimates and key elements of CDF re-appraisals for carefully selected topics, and usefully extend surveillance where long-term outcomes are required to fill evidence gaps. Despite some limitations, this approach demonstrates scalable and efficient evaluation of cancer therapies and outcomes monitoring over time using real-world evidence, supporting the lifecycle approach in health technology assessments.