Real-World Study of Breast Cancer Epidemiology and Treatment Patterns: A Pilot Study to Assess EHR-Derived Data in the United Kingdom
Author(s)
Peter McMahon, MSc1, Kevin Nolan, MSc1, Natalia Sadetsky, PhD1, Matthew Hodgeson, MSc2, Blythe Adamson, MPH, PhD3, Amit Samani, MBSS, PhD4.
1Gilead Sciences, Inc, Foster City, CA, USA, 2Gilead Sciences Ltd UK & Ireland, London, United Kingdom, 3Flatiron Health, New York, NY, USA, 4Flatiron Health, London, United Kingdom.
1Gilead Sciences, Inc, Foster City, CA, USA, 2Gilead Sciences Ltd UK & Ireland, London, United Kingdom, 3Flatiron Health, New York, NY, USA, 4Flatiron Health, London, United Kingdom.
OBJECTIVES: Robust real-world evidence (RWE) is critical for advancing breast cancer care in the UK, yet most data are derived from clinical trials, limiting generalizability. Existing UK data sources, including national registries, are constrained by limited clinical detail and delayed data availability. To address these gaps, Flatiron Health has developed oncology electronic health record (EHR)-derived RWD in the UK. This study evaluates the Flatiron UK dataset from one NHS Trust with planned expansion to include additional Trusts. Objective: To evaluate the Flatiron Health Research Database UK (FHRD-UK) breast cancer dataset by characterizing patient demographics, clinical and tumor features, treatment patterns, and outcomes.
METHODS: A retrospective cohort analysis was conducted using EHR-derived data from patients diagnosed with early (eBC) or metastatic breast cancer (mBC) between 2016-2024. Demographic, clinical, and tumor characteristics were described. Real-world overall survival (rwOS) using the Kaplan-Meier method was measured from eBC/mBC diagnosis and 1L, and disease-free survival (DFS) measured from eBC diagnosis to recurrence or progression. Analyses were stratified by molecular subtype.
RESULTS: Of 1,665 patients in the analytic cohort, 1,451 met inclusion criteria (mean age 62.6 years; 99.1% female). Most patients (91.9%) had eBC. Subtype distribution was HR+/HER2− (73.7%), HER2+ (15.7%), and triple-negative breast cancer (TNBC; 10.5%), mirroring national epidemiology. Among eBC patients, 88.7% underwent surgery. 1, 3, and 5 year rwOS was 96.3%, 85.4%, and 75.5% in the early setting (eBC) and 56.8%, 28.8%, and 16.6% in the metastatic setting (mBC) respectively.
CONCLUSIONS: This represents the first published analysis evaluating outcomes in BC patients from the FHRD-UK EHR-derived data, and demonstrated clinical validity with subtype distributions and outcomes aligned with national benchmarks. OS estimates reflect temporary oversampling of deceased patients at initial data integrations. As dataset coverage expands across NHS Trusts, this EHR-derived RWD can generate timely, robust evidence to inform breast cancer care and policy in the UK.
METHODS: A retrospective cohort analysis was conducted using EHR-derived data from patients diagnosed with early (eBC) or metastatic breast cancer (mBC) between 2016-2024. Demographic, clinical, and tumor characteristics were described. Real-world overall survival (rwOS) using the Kaplan-Meier method was measured from eBC/mBC diagnosis and 1L, and disease-free survival (DFS) measured from eBC diagnosis to recurrence or progression. Analyses were stratified by molecular subtype.
RESULTS: Of 1,665 patients in the analytic cohort, 1,451 met inclusion criteria (mean age 62.6 years; 99.1% female). Most patients (91.9%) had eBC. Subtype distribution was HR+/HER2− (73.7%), HER2+ (15.7%), and triple-negative breast cancer (TNBC; 10.5%), mirroring national epidemiology. Among eBC patients, 88.7% underwent surgery. 1, 3, and 5 year rwOS was 96.3%, 85.4%, and 75.5% in the early setting (eBC) and 56.8%, 28.8%, and 16.6% in the metastatic setting (mBC) respectively.
CONCLUSIONS: This represents the first published analysis evaluating outcomes in BC patients from the FHRD-UK EHR-derived data, and demonstrated clinical validity with subtype distributions and outcomes aligned with national benchmarks. OS estimates reflect temporary oversampling of deceased patients at initial data integrations. As dataset coverage expands across NHS Trusts, this EHR-derived RWD can generate timely, robust evidence to inform breast cancer care and policy in the UK.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
RWD163
Topic
Clinical Outcomes, Epidemiology & Public Health, Real World Data & Information Systems
Topic Subcategory
Distributed Data & Research Networks
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, Oncology