Exploratory biomarker analysis of trastuzumab deruxtecan (T-DXd) vs physician’s choice of chemotherapy (TPC) in HER2-low/ultralow, hormone receptor–positive (HR+) metastatic breast cancer (mBC) in DESTINY-Breast06 (DB-06).

1013 Background: DB-06 (NCT04494425), a Phase 3, randomized, open-label study, demonstrated a clinically meaningful progression-free survival (PFS; 13.2 vs 8.1 months [hazard ratio: 0.64]) benefit with T-DXd vs TPC (capecitabine, nab-paclitaxel, or paclitaxel) in patients with HR+, HER2-low (immunohistochemistry [IHC] 1+ or IHC 2+ / in situ hybridization–negative) or -ultralow (IHC 0 with membrane staining) mBC after ≥1 endocrine-based therapy (primary data cutoff: March 18, 2024). Here, we report an exploratory circulating tumor DNA (ctDNA) analysis based on baseline genomic status. Methods: Baseline ctDNA profiling in blood samples was assessed via Guardant OMNI 500-gene liquid biopsy assay. In total, 625 patients had evaluable ctDNA samples and putative tumor content, and comprised the biomarker evaluable population (BEP) presented herein. Baseline characteristics and efficacy outcomes were evaluated in key genomic subgroups (PI3K pathway, ESR1 m, BRCA1/2 m), including confirmed objective response rate (cORR) and PFS, both by blinded independent central review. Results: Genomic alterations were observed in 45.0% (PI3K pathway, n=281), 51.5% ( ESR1 m, n=322), and 7.7% ( BRCA1/2 m, n=48) of patients. The median PFS (mPFS) for each mutational subgroup was 13.2 (T-DXd) and 7.1 (TPC) months (PI3K pathway), 11.3 (T-DXd) and 7.0 (TPC) months ( ESR1 m), and 21.4 (T-DXd) and 5.6 (TPC) months ( BRCA1/2 m). T-DXd improved PFS and cORR outcomes compared with TPC across all mutational subgroups reported (Table). Conclusions: In this exploratory ctDNA analysis, T-DXd demonstrated a greater clinical benefit vs TPC regardless of PI3K pathway, ESR1 , or BRCA1/2 mutation. Clinical trial information: NCT04494425 . BEP (N=625) subgroup (n=T-DXd/TPC) T-DXd cORR, % TPC cORR, % T-DXd mPFS, mo* TPC mPFS, mo* PFS hazard ratio PI3K pathway † (139/142) 57.6 [48.9, 65.9] 41.5 [33.3, 50.1] 13.2 [9.9, 15.5] 7.1 [6.0, 9.5] 0.65 [0.48, 0.87] ESR1 m (166/156) 60.2 [52.4, 67.7] 32.1 [24.8, 40.0] 11.3 [9.8, 13.5] 7.0 [5.6, 9.3] 0.64 [0.49, 0.83] BRCA1/2 m (20/28) 80.0 [56.3, 94.3] 39.3 [21.5, 59.4] 21.4 [15.2, NE] 5.6 [4.1, 6.9] 0.14 [0.05, 0.33] Square brackets = 95% CIs (based on the Clopper-Pearson [cORR] or Brookmeyer-Crowley method [PFS]). PFS hazard ratios and CIs based on Cox proportional hazards model with no stratification factors, and ties handled by Efron approach. A hazard ratio <1 favors T-DXd vs TPC. No formal testing of significance was performed; *Number of PFS events: 89 (T-DXd) and 92 (TPC) in the PI3K pathway group, 115 (T-DXd) and 107 (TPC) in the ESR1 m group, and 7 (T-DXd) and 23 (TPC) in the BRCA1/2 m group; †includes AKT m, PIK3CA m, and PTEN m; CI, confidence interval; m, mutation; mo, months; NE, non-evaluable.