Lorenzo Gerratana, Rossana Roncato, Mattia Sturlese, Andrew A. Davis, Marko Velimirovic, Carolina REDUZZI, Katherine K. Clifton, Whitney L. Hensing, Ami N. Shah, Charles S. Dai, Paolo D’Amico, Arielle J. Medford, Alessandra Franzoni, Linda Cucciniello, Firas Wehbe, Seth A. Wander, Barbara Belletti, William Gradishar, Amir Behdad, Giuseppe Damante, Cynthia Ma, Fabio Puglisi, Aditya Bardia, and Massimo Cristofanilli
Background: ESR1 hotspot mutations (HS) (i.e. 380, 536, 537, and 538) are important drivers of resistance to aromatase inhibitors, but the differential impact of genomic variants (HS vs non-HS) on response to endocrine therapies (ET) under clinical development, such as novel oral Selective Estrogen Receptor Degraders and Modulators (SERDs and SERMs), is not known. The aim of the study was to evaluate the impact of non-HS ESR1 mutations on the pharmacodynamics of SERDs and SERMs as an additional ET resistance mechanism. Materials and Methods: The study analyzed a multi-institutional cohort of 1008 patients with hormone receptor positive metastatic breast cancer characterized by circulating tumor DNA (ctDNA). Pathway classification was defined based on previous work (i.e. RTK, RAS, RAF, MEK, NRF2, ER, WNT, MYC, p53, Cell Cycle, Notch, PI3K). Single nucleotide variations (SNVs) were annotated through OncoKB; co-occurrence was tested by Fisher’s exact test. A structure-based computational strategy was used to create 3D-models of ESR1 mutants and predict changes in binding affinity (dAff) across approved and experimental drugs. A positive dAff reflects a lower affinity of the drug for mutant ESR1 compared with wild type and thus a potential for a reduced response. Results: Among the total 680 detected ESR1 mutations, 633 were missense, and 631 were gain-of-function. The most frequent mutations were in codon 537 (N=305), followed by 538 (N=224). No significant MAF differences were observed across ESR1 variants (P=0.0829). The L391F mutation resulted in an increased binding affinity for Lasofoxifene (LAS) (dAff -0.34), Giredestrant (GIR) (dAff -0.18), Elacestrant (ELA) (dAff -0.08) and Amcenestrant (AMC) (dAff -0.41), while a decreased binding affinity was observed for 4OH-Tamoxifen (TAM) (dAff 0.01), Imlunestrant (IML) (dAff 0.15), Fulvestrant (FUL) (dAff 0.43), and Camizestrant (CAM) (dAff 0.02). V392F decreased binding affinity for TAM (dAff 0.05), LAS (dAff 0.13), IML (dAff 0.11), GIR (dAff 0.11), FUL (dAff 0.04), CAM (dAff 0.05), AMC (dAff 0.06) but not for ELA (dAff -0.01). F404L decreased binding affinity for FUL (dAff 0.07), ELA (dAff 0.73), and CAM (dAff 0.26), while it increased binding affinity for TAM (dAff -0.27), LAS (dAff -0.02), IML (dAff -0.05), GIR (dAff -0.69), and AMC (dAff -2.01). G415E increased binding affinity for LAS, (dAff -0.15) GIR (dAff -0.02) and ELA (dAff -0.08), while it decreased binding affinity for TAM (dAff 0.11), IML (dAff 0.09), FUL (dAff 0.29), CAM (dAff 0.19) and AMC (dAff 0.10). Mutations in codon 537 did not affect dAff for TAM, GIR, and ELA; a significant decrease in binding affinity was observed for FUL and AMC, whereas it was increased for LAS. Mutational co-occurrence was tested between ESR1 mutations in FUL docking sites and oncogenic pathways. Significant associations were observed for cell cycle SNVs (P=0.047), Notch SNVs (P=0.020), and ER SNVs (P< 0.001). Within these pathways, significant single-gene associations were observed for FBXW7 SNVs (P=0.020), ESR1 SNVs (P< 0.001), and GATA3 SNVs (P= 0.016). Given the highly significant co-occurrence of non-HS with other ESR1 mutations, combined models were examined. The Y537/F404 combination resulted in decreased binding affinity for FUL and increased binding affinity for LAS, while L536/F404 decreased binding affinity for TAM and increased binding affinity for IML, ELA, and AMC. Notably, L540/F404 restored the FUL-ESR1 interaction resulting in an increased binding affinity (dAff -2.1). Conclusions: The study suggests that genomic variability in drug targets detectable through ctDNA may modulate therapeutic response. Preclinical models are under development to investigate the combined endocrine resistance mechanism suggested by the significant co-occurrence between ESR1 mutations in SERDs/SERMs docking sites and ESR1 hotspot mutations and provide valuable additional insights for drug development and future treatment algorithms. Citation Format: Lorenzo Gerratana, Rossana Roncato, Mattia Sturlese, Andrew A. Davis, Marko Velimirovic, Carolina REDUZZI, Katherine K. Clifton, Whitney L. Hensing, Ami N. Shah, Charles S. Dai, Paolo D’Amico, Arielle J. Medford, Alessandra Franzoni, Linda Cucciniello, Firas Wehbe, Seth A. Wander, Barbara Belletti, William Gradishar, Amir Behdad, Giuseppe Damante, Cynthia Ma, Fabio Puglisi, Aditya Bardia, Massimo Cristofanilli. PD10-01 Impact of ESR1 mutations on Selective Estrogen Receptor Degraders and Modulators: an integrated liquid-biopsy and pharmacodynamics approach. [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD10-01.