Paired Tissue and Circulating Tumor DNA Profiling Reveals Novel ESR1 Amplifications and CDK4/6 Inhibitor Resistance Mechanisms in Hormone Receptor–Positive/HER2–Negative Metastatic Breast Cancers

PURPOSE: With the availability of new treatment options, such as selective estrogen receptor degraders, cyclin-dependent kinase 4/6 inhibitors (CDK4/6i), and PIK3CA inhibitors for hormone receptor-positive (HR+)/human epidermal growth factor receptor 2-negative (HER2-) advanced and metastatic breast cancers, the identification of ESR1, PIK3CA, and biomarkers associated with CDK4/6i resistance has become increasingly critical. Liquid biopsy has emerged as a convenient and valuable tool for molecular profiling and disease monitoring. In this study, we evaluate real-world genomic profiling in both tumor and liquid biopsies to determine how the complementary use of both specimen types optimizes the identification of resistance biomarkers. MATERIALS AND METHODS: We performed a real-world analysis using comprehensive genomic profiling (CGP) of tumor tissue and liquid biopsy (circulating tumor DNA [ctDNA]) samples from patients with HR+/HER2- advanced or metastatic BC. Forty-five patients with ctDNA results were analyzed, including 19 with matched tumor tissue CGP results. RESULTS: The most frequently altered genes were TP53, PIK3CA, and ESR1. ESR1 alterations were detected in 33% (15/45) of HR+ BC patients in ctDNA testing. In the 19 tissue CGP and ctDNA pairs, ESR1 missense mutations were found in 53% (n = 10/19) of cases, increasing to 79% (n = 15/19) when including ESR1 amplifications identified in tissue only (n = 5/19, 26%). In our study, ESR1 missense mutations and ESR1 amplifications were mutually exclusive. Copy number variations (CNVs) frequently detected included CCND1 (n = 7/19, 37%) and FGFR1 (n = 6/19, 32%) that are associated with CDK4/6i resistance and co-occurring with ESR1 amplifications. Those copy number amplifications were detected in tissue only. PIK3CA variants showed high concordance between tissue and ctDNA (91%), whereas 67% of TP53 variants were detected exclusively in ctDNA. CONCLUSIONS: ctDNA effectively detected ESR1 and PIK3CA mutations relevant to targeted therapies, whereas it failed to identify a subset of patients with ESR1 amplification, as well as other CNVs potentially mediating CDK4/6i resistance. TP53 mutations observed at high frequency in ctDNA without tissue confirmation may reflect clonal hematopoiesis, and their prognostic significance requires careful interpretation. As a liquid biopsy is valuable for disease and treatment monitoring, understanding its strengths and limitations is essential. Tissue CGP remains a complementary tool for establishing baseline tumor profiles and enhancing CNV detection in comprehensive molecular characterization.