4004 Background: Alveltamig (ZG006) is an innovative trispecific T cell engager (Tri-TCE) targeting two distinct DLL3 epitopes on tumor cells and CD3 on T cells (DLL3/DLL3/CD3), designed to bridge tumor cells and T cells, thereby mediating T cell specific killing of tumor cells. Here we conducted a dose optimization study in refractory NEC. Methods: This is a randomized, multicenter, open-label phase 2 study evaluating ZG006 in NEC patients (pts) who failed at least one prior line of therapy. Pts were randomized 1:1 to receive ZG006 at either 10 mg or 30 mg every two weeks, following a priming dose of 1 mg. The primary endpoint is objective response rate (ORR) assessed by IRC per RECIST1.1. Results are reported for 64 pts who received at least one dose of ZG006. Biomarker based efficacy analyses were performed based on a threshold of ≥50% of tumor cells stained at any intensity for DLL3 with an investigational antibody against DLL3 (SP347, Roche Diagnostics). Results: As of Sep. 10, 2025, a total of 64 pts (32 at each dose group) were randomized and received at least one dose of ZG006. The median age was 56 years (range: 25-72), 35 (54.7%) were male. All pts had received ≥1 prior line of systemic therapy, with 65.6% had prior anti-PD-(L)1 treatment. The confirmed ORR in 10 mg and 30 mg groups were 21.9% (7/32) and 37.5% (12/32), respectively; Disease control rate (DCR) were 40.6% (13/32) and 62.5% (20/32), respectively. Among the 19 responders, primary tumor sites included cervix (8), gallbladder (4), stomach (3), rectum (3) and colon (1). Pts with ≥50% DLL3 staining in tumor cells had greater ORR, DCR, and progression-free survival (PFS). The confirmed ORR in 10 mg and 30 mg groups were 33.3% (6/18) and 56.3% (9/16), respectively; DCR were 50.0% (9/18) and 75.0% (12/16), respectively; With a median follow up of 6.4 months, median PFS were 3.02 and 8.41 months, respectively. Median DoR was not yet mature in both groups at data cut-off. Treatment-related adverse events (TRAEs) occurred in all pts. The most frequent events were pyrexia, cytokine release syndrome (CRS) and anaemia. Most CRS were Grade 1-2, with only four grade 3 CRS, resolved with supportive care. TRAEs led to treatment interruption in 15 pts (23.4%) and permanent discontinuation in 2 pts (3.1%, one in each dose group). No grade 5 TRAE. No significant difference was observed in the safety profile between these two dose groups. Conclusions: ZG006 demonstrated a robust and durable response and a manageable safety profile in previously treated NEC pts, with superior efficacy observed at the 30 mg dose. The confirmed ORR of 56.3% and median PFS of 8.41 months in pts with ≥50% DLL3 positive tumor cells are particularly encouraging, and support further development of ZG006 for this patient population at the 30 mg dose. Clinical trial information: NCT06440057 .
Zhao et al. (Wed,) studied this question.