Trastuzumab deruxtecan (T-DXd) + durvalumab (D) as first-line (1L) treatment for unresectable locally advanced/metastatic hormone receptor-negative (HR-), HER2-low breast cancer: updated results from BEGONIA, a phase 1b/2 study

ABSTRACT

Background: Patients with HR- advanced/metastatic breast cancer (a/mBC) with a low level of HER2 (immunohistochemistry [IHC] score 1+ or IHC 2+ and negative in situ hybridization [ISH]) have poor prognosis. Combining 1L chemotherapy with immune checkpoint inhibitors can modestly improve outcomes vs chemotherapy alone, but treatment benefit is largely seen in patients with PD-L1+ disease. BEGONIA (NCT03742102) is an ongoing 2-part, open-label platform study, evaluating safety and efficacy of D, an anti-PD-L1 antibody, combined with other novel therapies in 1L triple-negative a/mBC, including HR-, HER2-low disease. T-DXd is a trastuzumab-topoisomerase I inhibitor antibody-drug conjugate that improves survival in patients with previously treated HR-, HER2-low mBC (NCT03734029;Modi NEJM 2022). Here, we report updated results of the T-DXd + D combination from BEGONIA. Method(s) Patients with unresectable HR-, HER2-low (per local testing, IHC 2+/ISH-, IHC 1+/ISH-, or IHC 1+/ISH untested) a/mBC were enrolled in the T-DXd + D arm. Patients eligible for 1L treatment, regardless of PD-L1 status, received intravenous T-DXd 5.4 mg/kg + D 1120 mg every 3 weeks until progression or unacceptable toxicity. PD-L1, assessed using the VENTANA PD-L1 (SP263) Assay, was defined as high if >= 5% of the tumor area was populated by PDL1-expressing tumor or immune cells. Primary endpoints were safety and tolerability. Secondary endpoints included investigator-assessed objective response rate (ORR;RECIST v1.1);progressionfree survival [PFS];and response duration. Patients included in the efficacy analysis had >= 2 ontreatment disease assessments, progressed, died, or withdrew from the study. Result(s) As of April 8, 2022, 56 patients received T-DXd + D (34 ongoing) and 46 were included in the efficacy analysis. Median (range) follow-up was 10.1 (0-22) months. Median age was 53.5 years, 71% had received prior treatment for early stage BC, and 64% had visceral metastases at baseline. Confirmed ORR was 26/46 (57% 95% CI, 41-71) and unconfirmed ORR was 33/54 (61% 95% CI, 47-74);1/46 patients (2%) had complete and 25/46 (54%) had partial responses. Confirmed response occurred irrespective of PD-L1 expression (PD-L1 high ORR, 5/7 [71%];PD-L1 low, 13/21 [62%];PD-L1 missing, 8/18 [44%]). Median duration of response was not reached;however, 64% of patients remained in response at 12 month follow-up and 73% had an ongoing response at data cutoff. Median PFS was 12.6 months (95% CI, 8-not reached). Adverse events (AEs) were consistent with the agents' known safety, with treatment-related AEs occurring in 49 patients (88%), any Grade 3/4 AEs in 18 patients (32%), and any serious AEs in 10 patients (18%). The most common all-Grade AEs were nausea (41 [73%]), fatigue (26 [46%]), and vomiting (17 [30%]). Adjudicated treatment-related interstitial lung disease/pneumonitis occurred for 5 patients (9%), which were mostly Grade 1 or 2 and 1 case of Grade 5 associated with COVID pneumonia. Seven patients (13%) and 21 patients (38%) had T-DXd dose reduction and dose delay, respectively;22 (39%) had D dose delay. Seven patients (13%) discontinued treatment due to AEs. Conclusion(s) For patients with HR-, HER2-low a/mBC, T-DXd in combination with D in the 1L setting shows manageable safety and promising efficacy including durable responses and an encouraging PFS. Although subgroups were small, responses were observed irrespective of PD-L1 expression. Analysis of additional translational data is ongoing. Funding(s) AstraZeneca/Daiichi Sankyo.