Background: ET is the mainstay treatment of ER+ aBC. Giredestrant is a highly potent, nonsteroidal, oral selective ER antagonist and degrader (PO SERD) that achieves robust ER occupancy. acelERA BC (NCT04576455) is a Phase II, randomized, open-label study that evaluated giredestrant vs. PCET in the second- or third-line ER+, HER2– aBC setting. The study did not reach statistical significance for the primary endpoint of investigator-assessed progression-free survival, although giredestrant demonstrated a more pronounced effect in patients (pts) with ESR1-mutated (ESR1m) tumors. We present an exploratory biomarker analysis. Methods: Pts (N = 303) were post- or pre-/perimenopausal women, and men, with ER+, HER2– aBC who had received 1–2 prior lines of systemic therapy for aBC. Randomization was 1:1 to giredestrant (30 mg PO daily) or PCET (fulvestrant/aromatase inhibitor). Baseline plasma circulating tumor (ct)DNA (n = 232) was evaluated with the FoundationOne Liquid CDx (n = 229) or PredicineCARE (n = 3) assays; ESR1 and other gene mutations in ctDNA were defined as variants with known or likely impact on protein function. Gene expression and PAM50 molecular subtype were assessed by RNAseq from tumor tissue (n = 184) which, based on collection date relative to aBC, was defined as representing early BC (eBC, n = 67), first line (1L, n = 64), or post-1L (n = 46). ER pathway activity was quantified by expression of a predefined set of 38 ER-induced or -repressed genes. Results: The majority of tumors were of the Luminal A (51%) and Luminal B (43%) PAM50 subtypes. Post-1L ESR1m tumors (n = 17) were significantly enriched for Luminal B (64%) compared with eBC (39%), 1L (44%), and post-1L tumors with no ESR1m detected (ESR1nmd; n = 29; 45%) (p < 0.001). Post-1L ESR1nmd tumors, but not ESR1m tumors, had significantly lower ER pathway activity compared with eBC or 1L tumors (p = 0.0001, 0.001). Within the post-1L subgroup, those with ESR1m had a distinct ctDNA mutational profile with more GATA3 (n = 29) and fewer CHEK2 mutations (n = 0) than ESR1nmd (n = 3 and n = 11, respectively). Tumor ER pathway activity and ESR1 expression in tumor tissue, as well as pre-treatment ctDNA composite tumor fraction and ESR1 variant allele frequency, were significantly correlated with best overall response to single-agent ET (partial response vs. progressive disease; p < 0.001, p = 0.022, p = 0.001, and 0.045, respectively). Conclusions: These data show that tumor response to single-agent ET is correlated with baseline ctDNA features and with tumor ER pathway activity. Interestingly, ER pathway activity was significantly lower in late-line tumors, except in ESR1m tumors. Overall, these results support the continued investigation of improved and personalized treatment options for pts with ER+, HER2– aBC. Clinical trial information: NCT04576455.