At the 2025 ASCO Annual Meeting, new data from an ongoing phase I study shed light on a targeted immunotherapy approach for patients with extrapulmonary neuroendocrine carcinomas (epNEC), a group of aggressive tumors often marked by poor prognosis and limited treatment options. The investigational therapy, obrixtamig (BI 764532), is a novel DLL3/CD3 T-cell engager designed to redirect T cells toward DLL3-expressing cancer cells, exploiting a vulnerability that is common in high-grade neuroendocrine tumors.
Delta-like ligand 3 (DLL3) is an inhibitory Notch pathway ligand that is minimally expressed in normal adult tissues but aberrantly overexpressed on the surface of many neuroendocrine tumors, including small cell lung cancer and extrapulmonary neuroendocrine carcinomas. By linking DLL3 on tumor cells to CD3 on T cells, obrixtamig facilitates direct cytotoxic engagement, potentially enabling a focused immune attack on DLL3-positive cancer cells.
This multicenter, dose-escalation trial (NCT04429087) evaluated obrixtamig across a series of intravenous regimens, ultimately focusing on two optimized protocols involving step-up dosing followed by weekly or less frequent administrations (RB2 and RB3). The current analysis examined differences in efficacy and safety outcomes based on DLL3 expression levels, stratifying patients into DLL3-high (≥50% tumor cells stained by immunohistochemistry) and DLL3-low cohorts.
Among 60 patients with heavily pretreated epNEC, roughly equally split between DLL3-high and DLL3-low groups, the tumors originated primarily from gastroenteropancreatic (45%) and genitourinary (30%) sites, with the remainder classified as other or unknown. All patients had received prior systemic therapy, and a substantial proportion—30% in the DLL3-high group and 50% in the DLL3-low group—had progressed on at least two previous lines of treatment.
Efficacy outcomes highlighted a pronounced difference linked to DLL3 expression. In patients whose tumors harbored high DLL3 levels, obrixtamig achieved an objective response rate (ORR) of 40%, with responses most commonly observed in gastroenteropancreatic and genitourinary primaries. The median duration of response was reported at 7.9 months, indicating a potentially meaningful period of disease control in this difficult-to-treat population. Disease control rates and the overall depth of responses were also higher in DLL3-high patients, and at the data cutoff in June 2024, seven patients in this group remained on therapy.
Conversely, patients with low DLL3 expression demonstrated less robust activity, underscoring DLL3 as a likely predictive biomarker for response. Importantly, the safety profile of obrixtamig was manageable and comparable between DLL3 expression groups. Most treatment-related adverse events were mild to moderate, aligning with the safety expectations for T-cell engagers and supporting continued investigation.
These findings reinforce DLL3 as a promising therapeutic target in extrapulmonary neuroendocrine malignancies and highlight the potential of obrixtamig to deliver clinically meaningful benefit in DLL3-high disease, even among patients who have exhausted standard therapies. The encouraging ORR and durability of response in this phase I setting provide a compelling rationale for advancing obrixtamig into later-phase trials, where its role in the management of high-grade DLL3-positive epNEC can be further elucidated.