Abstract Introduction: Immune checkpoint inhibition (ICI) in advanced malignancies offers a novel therapeutic approach, often with a more favorable toxicity profile and, in a subset of patients, the allure of a durable response. Unfortunately, the majority of patients with metastatic breast cancer (MBC) treated with a PD-(L)1 inhibitor progress rapidly. Mechanisms of resistance to and emerging genomic alterations after receiving ICI in MBC have not been well described. We aimed to describe the genomic profile of post-ICI therapy for MBC through the lens of circulating tumor DNA (ctDNA). Methods: Under an IRB-approved protocol, we retrospectively collected data from patients (pts) with MBC who had ctDNA analysis by the Guardant 360® assay (Redwood, CA) as part of their routine care between January 2015 and December 2020. We abstracted the demographic, clinical, pathologic, and therapeutic history for these patients. We identified patients who received ICI therapy and a ctDNA analysis within 6 months of completing therapy and, when possible, a paired ctDNA assay no more than 6 months prior to starting ICI. CtDNA variants were annotated by pathogenicity based on the OncoKB classification (Chakravarty et al, JCO PO 2017, date accessed 5/2021). The variant allele frequency (VAF) for a ctDNA sample was determined by the VAF for the dominant alteration in that sample. Results: Of 473 pts with MBC and ctDNA analysis, 49 received ICI therapy. Twenty-four pts had post-immune therapy ctDNA analysis, of whom 17 had paired pre- and post-immune therapy samples. For those with post-ICI ctDNA, the median age was 55 years, and breast cancer subtype was hormone receptor-positive, HER2-negative in 6 pts (25%), HER2-positive in 2 pts (8%), and triple negative in 16 pts (67%). Patients received a median of 2.5 lines of therapy for MBC prior to receiving immune therapy. Eighteen pts (75%) received pembrolizumab and 6 pts (25%) received atezolizumab as standard therapy for 4 pts, a clinical trial for 12 pts, and off label (based on molecular tumor board considerations or physician discretion) for 5 pts. Concurrent therapies were nab-paclitaxel (n=4, 17%), capecitabine (n=10, 42%), carboplatin (n=4, 17%), anti-HER2 (n=2, 8%), liposomal doxorubicin (n=1, 4%), and monotherapy (n=3, 13%). The median time from completing ICI to ctDNA collection was 16 days. In the post-ICI samples, the most common alterations were in TP53 (22%), PIK3CA (7%), KIT (4%), MYC (4%), and PTEN (4%). Across post-ICI samples, 72.46% of alterations were oncogenic or likely oncogenic (n=100). The median VAF in the post-ICI samples was 9.7% (interquartile range [IQR]: 2.2%-23%) compared to 2.8% (IQR: 0.5%-4.8%) in the pre-ICI samples. The median number of detected alterations per ctDNA analysis was 5 (IQR: 3-7) in the post-ICI samples and 2 (IQR: 2-5) in the pre-ICI samples. Among the 17 pts with paired samples, 15 pts had new alterations (44 total alterations gained), of which 68% were oncogenic or likely oncogenic. Twenty-nine alterations detected in the pre-ICI samples were no longer detected in the post-ICI samples. The new alterations were 50% SNV, 48% amplifications, and 2% deletions, and the most common new oncogenic or likely oncogenic alterations were in TP53 in 29%, PTEN in 18%, PIK3CA in 18%, KIT in 18%, and EGFR in 18%. Conclusions: Pathogenic alterations, particularly in the PTEN-PIK3CA-AKT pathway, continue to emerge after ICI treatment. Prospective, standardized evaluation of post-ICI tumor molecular profile in MBC can give insight into rational therapy sequencing approaches and/or novel combinations aimed at delaying resistance to ICI therapy. Citation Format: Anthony K Kang, Ami N Shah, Neelima Katam, Andrew A Davis, Lorenzo Gerratana, Shruti Chandra, Saya Jacob, Meilynn Shi, Jeeven Srivastava, Paolo D’Amico, Carolina Reduzzi, Michael Gurley, Firas Wehbe, Qiang Zhang, Amir Behdad, Massimo Cristofanilli. Emerging molecular variants by circulating tumor DNA after immunotherapy in metastatic breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P2-14-05.