Transporting Evidence Across Borders: A Targeted Review of the Use and Acceptability of Non-Local Data to Derive Real-World External Control Arms in Submissions to NICE

OBJECTIVES: Real-world external control arms (RW-ECAs) are an increasingly common source of comparative evidence in HTA submissions. Payers prefer local data, citing concerns over the transportability of evidence from other countries. However, limitations in the quality/quantity of local data in some countries often necessitate reliance on non-local data. This study reviews current practices in the use of non-local data to derive RW-ECAs and its acceptance in HTA.

METHODS: RW-ECAs submitted to NICE were identified through screening company submissions of all single technology appraisals with draft guidance from 2019-2024. Information on justifications for using non-local data, evidence presented to support transportability, and payers' views on the limitations of non-local data were extracted.

RESULTS: RW-ECAs were identified in 18/267 submissions, of which 16 were in oncology. RW-ECAs were derived solely from non-UK data or alongside UK data in 15/18 submissions. Justifications for not using UK data included limited sample size/follow-up and absence of data on outcomes, confounders, or trial eligibility criteria (e.g., genetic information). Of submissions using non-UK data, only 4/15 provided empirical evidence to support transportability, and presented only simple comparisons of average patient/clinical characteristics. No study applied methods to re-weight/adjust data to increase transportability. 4/15 submissions provided only verbal arguments to support transportability, supported by clinical expert opinion in 3 cases. In 5/15 submissions, the Evidence Review Group explicitly stated that data used was not reflective of UK clinical practice.

CONCLUSIONS: Use of non-local data to derive RW-ECAs is common in NICE submissions, and concerns over transportability are regularly expressed by NICE. Although the recently updated NICE RWE framework recommends formal methods to assess and correct for a lack of transportability, these methods have not previously been used in practice. Increased confidence in RW-ECA evidence could be achieved through improved quality/accessibility of UK data and use of best-practice methods where non-local data is used.