Neoadjuvant trastuzumab deruxtecan alone or followed by paclitaxel, trastuzumab, and pertuzumab for high-risk HER2-positive early breast cancer (DESTINY-Breast11): a randomised, open-label, multicentre, phase III trial

BACKGROUND: Neoadjuvant standard-of-care for HER2-positive early-stage breast cancer is trastuzumab + pertuzumab with polychemotherapy; however, existing regimens have high toxicity burdens and suboptimal outcomes. DESTINY-Breast11 assessed efficacy and safety of neoadjuvant trastuzumab deruxtecan (T-DXd) alone or followed by paclitaxel + trastuzumab + pertuzumab (THP) versus dose-dense doxorubicin + cyclophosphamide (ddAC) followed by THP for high-risk (≥cT3cN0 or cT0-4cN1-3) HER2-positive disease. PATIENTS AND METHODS: This open-label, phase III trial (147 sites, 18 countries) randomised adults 1 : 1 : 1 to T-DXd (×8 cycles), T-DXd-THP (4 + 4 cycles), or ddAC-THP (4 + 4 cycles). T-DXd-alone arm enrolment closed early following the Independent Data Monitoring Committee recommendation. The primary endpoint was pathological complete response (pCR; ypT0/is ypN0; intent-to-treat population). Secondary endpoints included event-free survival (EFS; intent-to-treat population) and safety (safety analysis set). RESULTS: Between 25 October 2021 and 12 March 2025, 286 (T-DXd), 321 (T-DXd-THP), and 320 (ddAC-THP) female patients were randomised. pCR rates were 43.0% (T-DXd, n = 123), 67.3% (T-DXd-THP, n = 216), and 56.3% (ddAC-THP, n = 180). T-DXd-THP versus ddAC-THP absolute pCR rate difference was 11.2% [95% confidence interval (CI), 4.0% to 18.3%, P = 0.003], with benefit in hormone receptor (HR)-positive [61.4% (n/N = 145/236) versus 52.3% (n/N = 123/235); difference in pCR (ΔpCR) 9.1% (95% CI 0.2% to 17.9%)] and HR-negative [83.1% (n/N = 69/83) versus 67.1% (n/N = 57/85); ΔpCR 16.1% (95% CI 3.0% to 28.8%)] subgroups. Median EFS (T-DXd-THP versus ddAC-THP, maturity 4.5%) hazard ratio was 0.56 (95% CI 0.26 to 1.17). Grade ≥3 adverse events (AE; T-DXd, 22.6% (n = 64); T-DXd-THP, 37.5% (n = 120); ddAC-THP, 55.8% (n = 174)], serious AE [T-DXd, 10.2% (n = 29); T-DXd-THP, 10.6% (n = 34); ddAC-THP, 20.2% (n = 63)], and all-grade left-ventricular dysfunction [T-DXd, 0.7% (n = 2); T-DXd-THP, 1.3% (n = 4); ddAC-THP, 6.1% (n = 19)] rates were lower for T-DXd and T-DXd-THP than ddAC-THP. All-grade adjudicated drug-related interstitial lung disease/pneumonitis rates were low and similar across arms [T-DXd, 4.9% (n = 14); T-DXd-THP, 4.4% (n = 14); ddAC-THP, 5.1% (n = 16)]. Three treatment-related deaths occurred [T-DXd-THP, 0.3% (n = 1); ddAC-THP, 0.6% (n = 2)]. CONCLUSIONS: Neoadjuvant T-DXd-THP demonstrated statistically significant and clinically meaningful pCR benefit and improved safety versus ddAC-THP.