Abstract 153P
Background
A previous trial-level meta-analysis validated DFS as a surrogate endpoint for OS in HER2– BC. To shed light on the patient-level associations between DFS endpoints and OS, this study evaluated the correlation between DFS and OS, invasive DFS (IDFS) and OS, and distant DFS (DDFS) and OS in elderly patients with HER2– eBC.
Methods
Patients aged ≥66 years with stages I–III HER2– BC were identified from SEER-Medicare data (2010–2019). DFS was defined as time from primary surgery to earliest of carcinoma in situ (CIS) of breast, locoregional/distant recurrence, new primary cancer, or all-cause death. IDFS was defined like DFS, excluding CIS. DDFS was defined like DFS, excluding CIS and locoregional recurrence. OS was defined from primary surgery to all-cause death. All endpoints were assessed using Kaplan-Meier analysis. Normal scores rank correlation between each DFS endpoint and OS was estimated separately in hormone receptor-positive (HR+) and triple-negative BC (TNBC) patients.
Results
Of 28,655 patients, 90.4% had HR+ BC and 9.6% had TNBC (median follow-up: 4 years). In TNBC, median DFS, IDFS, and DDFS were 3.1, 3.8, and 4.4 years, respectively. In HR+ BC, median DFS, IDFS, and DDFS were 4.5, 5.9, and 6.3 years, respectively. Median OS was not reached (5-year OS, TNBC: 67.7%; HR+ BC: 83.7%). Across cohorts, a significant positive correlation was observed between each DFS endpoint and OS (Table), with the strongest correlation observed between DDFS and OS in TNBC (correlation coefficient: 0.69; 95% confidence interval: 0.65–0.71; p<0.001) and in HR+ BC (0.60; 0.57–0.62; p<0.001). Table: 153P
Patient-level correlations between DFS endpoints and OS in patients with HER2– eBC, stratified by breast cancer subtype
| DFS endpoints | Overall HER2– eBC cohort (N =28,655) | Stratification by breast cancer subtype | ||||
| HR+/HER2– BC (N = 25,899) | TNBC (N = 2,756) | |||||
| Normal scores rank correlation [95% CI] | P-value | Normal scores rank correlation [95% CI] | P-value | Normal scores rank correlation [95% CI] | P-value | |
| DFS | 0.53 [0.50, 0.56] | <0.001 | 0.51 [0.48, 0.54] | <0.001 | 0.62 [0.58, 0.66] | <0.001 |
| IDFS | 0.59 [0.57, 0.61] | <0.001 | 0.58 [0.55, 0.60] | <0.001 | 0.66 [0.61, 0.70] | <0.001 |
| DDFS | 0.61 [0.59, 0.63] | <0.001 | 0.60 [0.57, 0.62] | <0.001 | 0.69 [0.65, 0.71] | <0.001 |
Conclusions
We observed positive, patient-level correlations between intermediate endpoints and OS in HER2– eBC, highlighting the potential utility of these endpoints as surrogates for OS in this population, given validation in trial-level meta-analyses.
Legal entity responsible for the study
Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., and AstraZeneca UK Ltd.
Funding
Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., and AstraZeneca UK Ltd.
Disclosure
J.R. Earla, K.M. Hirshfield, J.A. Mejia: Financial Interests, Personal, Full or part-time Employment: Merck & Co., Inc. K. Kponee-Shovein: Financial Interests, Institutional, Sponsor/Funding, Employee of Analysis Group, Inc., a consultancy that received funding from Merck & Co., Inc. to conduct this study: Merck & Co., Inc. Y. Song, M. Mahendran, Q. Hua, A. Hilts: Financial Interests, Institutional, Sponsor/Funding, Employee of Analysis Group, Inc., a consultancy that received funding from Merck & Co., Inc. to conduct this study: Merck & Co., Inc. All other authors have declared no conflicts of interest.