OBJECTIVES: Health technology assessment bodies prefer local data to answer questions about the use and outcomes of therapies in routine practice but it may not always be available or sufficient. Data from other countries could be used to fill evidence gaps but the transportability across countries is not well understood.

METHODS: We used the US-based, nationwide EHR-derived de-identified Flatiron Health database to emulate a recent UK study on the treatment patterns and outcomes for patients receiving first-line therapy in advanced non-small-cell lung cancer (aNSCLC) between June 2016 and March 2018. For this treatment period, we used matching-adjusted indirect comparison to standardize the US data to reflect the average characteristics of the UK cohort (for age, sex, ECOG, histology). We compared patient characteristics, treatment patterns, and overall survival (OS) estimates before and after population-adjustment between the US and UK cohorts.

RESULTS: There were 1,003 patients in the UK cohort versus 3,819 in the US receiving 1L therapy for aNSCLC. After weighting, median OS in the US and UK were similar across 1L drug class: chemotherapies (7.7 [7.1-8.3] vs. 8.1 [7.4-8.9] months), immunotherapies (13.9 [11.0-17.1] vs. 14.0 [10.7-20.6]), and targeted therapies (21.6 [18.5-23.7] vs. 20.2 [16.0-30.5]). After weighting OS curves were almost overlapping for 1L immunotherapy and targeted therapy and for around 12 months for 1L chemotherapy after which OS tended to be higher in US compared to UK patients. Of those receiving 1L chemotherapy, the proportion receiving any second-line therapy was higher for patients in the US vs. UK.

CONCLUSIONS: The results suggest that for patients with aNSCLC receiving 1L treatment, US data, with appropriate population adjustment, has potential to be used in technology evaluations to understand long-term OS where UK data is unavailable or sparse.