Henlius pushes tetra-specific T-cell engager into the clinic for SCLC, NEC

Shanghai Henlius Biotech has registered a first-in-human Phase I clinical trial evaluating pipeline candidate HLX3901 in adults with advanced small cell lung cancer (SCLC) or neuroendocrine carcinoma (NEC) who have progressed on prior platinum-based therapy. The open-label study, posted to ClinicalTrials.gov under identifier NCT07416695, plans to enroll up to 138 patients at Shanghai Chest Hospital. HLX3901 is a proprietary DLL3 × DLL3 × CD3 × CD28 tetra-specific antibody — designed to deliver two mechanistically distinct T-cell activation signals while anchoring to two separate epitopes on the same tumor-associated antigen. No other molecule with this four-specificity architecture has entered human testing for SCLC or NEC.

HLX3901’s tetraspecific design rationale

HLX3901 is an intravenously administered multispecific antibody engineered by Shanghai Henlius to bind four targets simultaneously: two distinct epitopes on DLL3 (delta-like ligand 3), plus CD3 on the T-cell receptor complex and the co-stimulatory receptor CD28. The biparatopic DLL3 engagement is intended to increase binding avidity to tumor cells, while the simultaneous ligation of CD3 and CD28 provides both the primary activation signal and the co-stimulatory signal needed for full T-cell effector function — a pairing often described in immunology as “signal 1 plus signal 2.”

DLL3 is a Notch pathway ligand expressed on the surface of the majority of SCLC tumors and a substantial fraction of extrapulmonary neuroendocrine carcinomas, with minimal expression on normal adult tissues. That expression pattern has made it a focus of drug development across multiple modalities, including bispecific antibodies, antibody-drug conjugates, and CAR-T cells. Preclinical data from independent groups have shown that DLL3-targeted bispecific antibodies can redirect T cells to kill SCLC cells, and that combining DLL3 bispecifics with PD-1 blockade enhances anti-tumor activity in mouse models.

The addition of CD28 co-stimulation is the design element that most clearly distinguishes HLX3901 from existing DLL3-directed T-cell engagers. First-generation CD3-only bispecifics can induce T-cell exhaustion over repeated stimulation cycles because they lack the co-stimulatory input that physiological T-cell activation requires. Whether incorporating CD28 into a multispecific format resolves this limitation in patients without introducing disproportionate cytokine release or other immune toxicities is the central question that this Phase I trial is designed to address.

Trial specifics

The study is structured in two parts. Phase Ia uses an accelerated titration design transitioning to a conventional 3+3 dose-escalation scheme across seven prespecified dose levels, with a 28-day dose-limiting toxicity (DLT) observation window after first administration. A backfill cohort of two to three dose groups will collect additional safety and pharmacokinetic data. Phase Ib is a dose-expansion study at the recommended Phase II dose (RP2D) identified during escalation, intended to generate preliminary efficacy and confirmatory safety data in both SCLC and NEC.

The AllSci BriefSystematic R&D and deal news. Daily.

Eligible patients are adults aged 18 to 75 with histologically or cytologically confirmed disease and life expectancy exceeding three months. Patients with SCLC must have progressed on or been intolerant to platinum-based chemotherapy combined with an immune checkpoint inhibitor. Patients with NEC must have failed platinum-based therapy. Combined SCLC and mixed neuroendocrine–non-neuroendocrine neoplasms are allowed. Patients must provide tumor tissue for DLL3 expression testing, though the protocol as registered does not specify a minimum DLL3 expression threshold for enrollment.

The DLL3 competitive landscape

HLX3901 enters a DLL3-targeting field that has matured rapidly. Amgen’s tarlatamab (IMDELLTRA), a DLL3 × CD3 bispecific T-cell engager built on the BiTE platform, received FDA accelerated approval in May 2024 for previously treated extensive-stage SCLC based on the DeLLphi-301 Phase II trial, which reported an ORR of approximately 40% and a median duration of response of roughly 9.7 months. Tarlatamab is now the benchmark against which all subsequent DLL3-directed agents will be measured, and Amgen is running Phase III trials evaluating it in first-line and later-line settings.

Beyond tarlatamab, the DLL3 pipeline includes Boehringer Ingelheim’s BI 764532, a DLL3 × CD3 bispecific in Phase I; HPN328, a DLL3-targeting tri-specific T-cell activating construct acquired by Merck through its purchase of Harpoon Therapeutics; and several DLL3-directed ADCs, including ifinatamab deruxtecan (I-DXd, DS-7300) from Daiichi Sankyo (Phase II/III), ZL-1310 from Zai Lab (Phase I/II), and QLS31901 from Qilu Pharmaceutical (Phase I/II). AbbVie’s earlier DLL3 ADC, rovalpituzumab tesirine, was discontinued after Phase III failures.

Chinese companies are particularly active in this space. In addition to Henlius, Qilu Pharmaceutical, Chia Tai Tianqing (TQB2930, a DLL3 ADC), LaNova Medicines (LB2102), and Phanes Therapeutics (PT217) all have DLL3-targeted molecules in early clinical development.