Background: Improving outcomes in patients (pts) with TNBC remains a high unmet need. Immune checkpoint inhibitors (ICIs) + chemotherapy (chemo) is approved for newly diagnosed pts with PD-L1+ tumors. Further, single-agent sacituzumab govitecan (SG), a Trop-2–directed antibody-drug conjugate, is approved for pts with mTNBC who received ≥2 prior systemic therapies (≥1 for metastatic disease). However, more options are needed for pts with PD-L1–negative mTNBC and for those with disease progression on chemo ± ICI. Magrolimab is a monoclonal antibody that blocks CD47, a “don’t eat me” signal often overexpressed on TNBC cells. Magrolimab blockade of CD47 induces macrophage-mediated phagocytosis of tumor cells and has shown preclinical activity and promising clinical efficacy in hematologic malignancies. Certain chemos, including taxanes, enhance prophagocytic signals on tumor cells, which may lead to synergistic antitumor activity with magrolimab. This study is evaluating the safety, tolerability, and efficacy of magrolimab + nab-paclitaxel/paclitaxel or + SG in mTNBC. Methods: This open-label study has 2 cohorts (C) with safety run-in and phase (ph) 2 portions. Eligible pts are ≥18 y with measurable disease per RECIST 1.1. C1 pts have PD-L1–negative untreated mTNBC. C2 pts have mTNBC and received 1 prior line of therapy in the advanced setting, a taxane in the neoadjuvant/adjuvant or metastatic setting, and, if PD-L1+, an ICI. Exclusion criteria include active central nervous system disease, RBC transfusion dependence, and prior treatment with CD47/SIRPα-targeting agents. C1 assesses nab-paclitaxel/paclitaxel + magrolimab (safety run-in) or ± magrolimab (ph 2; randomized 1:1). C2 assesses magrolimab + SG (safety run-in and ph 2). In C1 safety run-in, magrolimab is given intravenously (IV) as a 1-mg/kg priming dose on day (D) 1 of cycle 1 to mitigate on-target anemia, followed by 30 mg/kg (cycle 1: D8, 15, 22; cycle 2: D1, 8, 15, 22; cycle 3+: D1, 15) (28-d cycles). In C2 safety run-in, pts receive a 1-mg/kg priming dose on D1, followed by 30 mg/kg (cycle 1: D8, 15; cycle 2: D1, 8, 15) and 60 mg/kg (cycle 3+: D1) (21-d cycles). The recommended ph 2 dose (RP2D) is determined in the safety run-in, with de-escalation if prespecified dose-limiting toxicity (DLT) criteria are met. Once RP2D is determined, the ph 2 cohorts will follow their respective dose schedules. Nab-paclitaxel/paclitaxel and SG are given IV per standard of care. The primary endpoints are incidence of DLTs, adverse events, and abnormal lab results by CTCAE v5.0 (safety run-in); progression-free survival by RECIST 1.1 (ph 2 C1); and confirmed objective response rate by RECIST 1.1 (C2; pooled safety run-in and ph 2). Planned enrollment is ≈144 pts. Clinical trial information: NCT04958785.