Prognostic profiling of hormone receptor–positive, HER2-negative (HR+/HER2-) metastatic breast cancer (MBC) treated with CDK4/6 inhibitors beyond progression (BP): A synthetic control arm biomarker analysis.

e13068 Background: Second-line treatment for HR+/HER2- MBC is evolving, requiring integration of new treatment strategies such as CDK4/6i BP novel targeted agents in association to endocrine therapy, and antibody-drug conjugates (ADCs). The aim of this study is to explore ctDNA biomarkers impacting second progression free survival (PFS2) in patients (pts) treated with CDK4/6i BP. Methods: This study used the nationwide (US-based) deidentified Flatiron Health-Foundation Medicine clinicogenomic database (FH-FMI CGDB), comprising data originated from ~280 US cancer clinics (~800 sites of care). In this study a synthetic control arm (SCA) was constructed from the FH-FMI CGDB (n = 145) as comparator for a prospective multi-institutional cohort of patients treated with CDK4/6i BP (n = 65). Pts in both subgroups were profiled through ctDNA using the FoundationOne Liquid NGS panel. A 1:1 Propensity Score Matching (PSM) was applied to balance covariates, including age at diagnosis, ECOG status, bone-only or lymph node metastatic involvement, and PFS1 CDK4/6i treatment duration. Pathogenic alterations with a > 10% prevalence were analyzed individually and according to oncogenic pathways (Sanchez-Vega et al., Cell , 2018). The prognostic impact of single nucleotide and copy number variations (SNVs and CNVs) was assessed in terms of PFS2, defined as the time from second-line treatment initiation to progression or death. Results: A total of 130 pts were analyzed after PSM. Liver was the most frequent metastatic site, observed in 30.8% of the fulvestrant + palbociclib BP cohort (F+P) and 36.9% of the FH-FMI CGDB cohort (fulvestrant monotherapy; F). ESR1 was the most frequently altered gene (38%), followed by BRCA2 (32%), TP53 (29%), ATM (28%), and PIK3CA (28%). PFS2 was comparable between the two cohorts, with no significant difference (Hazard Ratio (HR) comparing F+P vs F HR = 1.36, P = 0.1141). In the overall population, cell cycle CNV significantly impacted PFS2 (HR: 2.03, P = 0.005), while in the F+P cohort, CNV in the P53 pathway had a significant impact (HR: 5.6, P = 0.024) with MDM2 CNV as main contributor. Multivariable analysis confirmed that in the F cohort, SNVs in the RTK/RAS pathway and CNVs in the cell cycle pathway significantly impacted PFS2 (HR: 2.31, P = 0.014 and HR: 2.16, P = 0.016, respectively). At the single-gene level, CNVs in the chromosome 11q13.3 region (i.e. CCND1 , FGF19 , FGF3 , and FGF4 ) were associated with PFS2 in the F group but not in the F+P group. Conclusions: The SCA developed in this analysis confirmed findings from the PACE prospective trial and highlighted a differential prognostic impact of ctDNA features across treatment subgroups. These results provide valuable insights for personalized treatment algorithms and could guide therapeutic decisions and future clinical trial design.