Background: CDK4/6 inhibitors (CDK4/6i), like palbociclib, ribociclib, and abemaciclib, along with antiestrogens, have revolutionized treatment for ER+/HER2- metastatic breast cancer (MBC). Although most patients initially respond, almost all eventually progress, and ER+ HER2- MBC remains incurable. There is an urgent need to understand the molecular processes that drive resistance in order to improve survival. The landscape of acquired somatic alterations causal to CDK4/6i resistance remains unknown. Here we report differences in mutational landscapes between ER+ HER2- MBC patients treated with and without CDK4/6i.
Methods: Deidentified data from 780 and 1073 ER+ HER2- MBC patients (solid tumor or ctDNA liquid biopsy sequencing respectively) with at least 6 months between diagnosis of Stage 4 disease and biopsy were analyzed. Patients were divided into either treated or untreated with CDK4/6i prior to biopsy. Sequencing was performed using the Tempus XT tumor assay (DNA sequencing of 595-648 genes at 500x coverage) and Tempus XF liquid biopsy (ctDNA sequencing of 105-523 genes). Gene alterations (consisting of pathogenic/likely pathogenic short variants and copy number alterations) were compared between groups by Chi-squared/Fisher’s Exact tests and p-values adjusted for false-discovery.
Results: We first analyzed sequencing data of both solid tumor and liquid ctDNA from ER+/HER2- MBC patients. ESR1 mutations were significantly more frequent in those that received CDK4/6i than those that did not (Solid tumor 33% vs 16%, p < 0.001, q = 0.001; Liquid biopsy 32% vs 16%, p < 0.001 and q <0.001). We also saw more frequent mutations/amplifications in the following genes in the CDK4/6i treated cohort vs. those that were not. These results trended towards significance in our solid tumor, but not in our liquid biopsy cohort: CCND1 (18% vs 11% p = 0.028 q =0 .3); FGF3 (17% vs 9.5% p = 0.010 q = 0.2); FGF4 (17% vs 11% p = 0.035 q = 0.3), GATA3 (17% vs 8.9% p = 0.008 q = 0.2), PTEN (12% vs 6.1% p = 0.030 q = 0.3) and FGF19 (8.2% vs 1.7% p = 0.002 q = 0.12). Interestingly, 96-98% of CCND1, FGF3, FGF4 and FGF19 alterations were copy number amplifications. Conversely, we saw a trend towards significance for more mutations in TP53 (37% vs 27% p=0.008 and q=0.2) in those that had not received a CDK4/6i than those that did.
Conclusions: Here we present the landscape of somatic alterations in ER+/HER2- MBC patients with and without prior CDK4/6i therapy from our large real world de-identified data set. Patients with prior CDK4/6i therapy harbored significantly more ESR1 somatic alterations, demonstrated in both solid tissue and liquid biopsies. In solid tissue biopsies, patients with prior CDK4/6i therapy harbored more CCND1, FGF3, FGF4, and GATA3 alterations and less TP53 alterations. These trends were not significant after adjustment for multiple testing. CCND1, FGF3, FGF4 and FGF19 alterations were copy number amplifications, which may be consistent with 11q13 amplification. Further studies will provide insights into how these trends translate towards our understanding of CDK4/6i related resistance mechanisms.
VIEW THE PUBLICATION
VIEW THE POSTER