Background: Cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) combined with endocrine therapy (ET) is the most common first-line (1L) treatment for hormone receptor-positive (HR+) and human epidermal growth factor receptor 2-negative (HER2?) metastatic breast cancer (mBC). Registrational studies for palbociclib (P), ribociclib (R), and abemaciclib (A) have shown consistent PFS benefit; nevertheless, overall survival (OS) results have been inconsistent. No head-to-head studies directly comparing CDK4/6i have been conducted, and although cross-trial comparisons can provide insights, they are limited by differences in trial design and patient population. We have used real-world evidence (RWE) data to directly compare real-world OS (rwOS) with different CDK4/6i when used as 1L treatment for patients with mBC treated in the USA.
Methods: This retrospective open-cohort study used data from two US de-identified electronic health record-derived data sets: primary analyses were conducted using data from the Flatiron Health database, which were then verified using data derived from the Tempus database. Eligible patients had HR+/HER2? mBC and received 1L metastatic treatment with CDK4/6i + ET between January 1, 2016, and April 30, 2023 (Flatiron Health), or December 31, 2022 (Tempus). Patients were followed from the first date of 1L treatment until death or last known activity. Median rwOS for each CDK4/6i was estimated using the KaplanMeier methodology. Treatment effects were analyzed using adjusted Cox proportional hazards models, including covariate and propensity score adjustments. Subgroup analyses were also conducted on patients treated with CDK4/6i + aromatase inhibitors (AI) as well as in patients who had a treatment-free interval (TFI) >12 months between last adjuvant ET and metastatic diagnosis date and were then treated with CDK4/6i + AI.
Results: 4,567 patients were selected from the Flatiron Health database (3,504, 575, and 488 treated with P, A, and R, respectively) and 612 from the Tempus database (494, 65, and 53, respectively). In the Flatiron Health database, median (95% CI) rwOS was 43.8 (41.746.6) months for P, 43.2 (34.548.8) months for A and 44.2 (37.351.1) months for R. Adjusted hazard ratios: 1.09 (95% CI: 0.891.33) for P versus A and 1.07 (95% CI: 0.901.28) for P versus R. Subgroup analyses also showed a similar risk of death between CDK4/6i for patients who received CDK4/6i + AI (2,334, 341, and 355 treated with P, A, and R, respectively. Adjusted hazard ratios 0.90 [95% CI: 0.661.23] for P versus A and 1.08 [95% CI: 0.881.33] for P versus R) and those with a TFI>12 months (2,226, 349 and 232 treated with P, A and R, respectively. Adjusted hazard ratios 0.91 [95% CI: 0.661.25] for P versus A and 1.11 [95% CI: 0.901.37] for P versus R). Findings were confirmed with Tempus data; adjusted hazard ratios indicated a similar risk of death between CDK4/6i: 0.92 (95% CI: 0.511.68) for P versus A and 1.14 (95% CI: 0.64 2.05) for P versus R.
Conclusions: Our findings, using mature follow-up data, showed similar rwOS outcomes when different CDK4/6i are combined with ET as 1L treatment for patients with HR+/HER2? mBC. Results were consistent across RWE databases. As 1L treatment evolves, these findings contribute to the ongoing development of combination strategies, such as novel oral selective estrogen receptor degraders, to improve clinical outcomes in patients with HR+/HER2? mBC and guide clinical decision-making.
VIEW THE PUBLICATION