Preclinical Characterization of XB002, an Anti-Tissue Factor Antibody-Drug Conjugate for the Treatment of Solid Tumors

Abstract

Tissue factor (TF) is overexpressed in various cancers and is typically associated with poor clinical outcomes. XB002 is an anti-TF antibody-drug conjugate designed to selectively deliver a cytotoxic payload to TF-expressing tumors while minimizing TF-related adverse events, particularly bleeding. The conjugate consists of a zovodotin linker-payload attached to a mAb (clone 25A3) that binds TF with high affinity (KD = 0.86 nmol/L). In vitro coagulation assays confirmed that 25A3 does not interfere with the clotting cascade; even at a concentration of 100 nmol/L, it did not affect the activation of coagulation factor X or thrombin generation. XB002 demonstrated efficient internalization in TF-expressing cancer cell lines, exhibiting potent cytotoxicity at subnanomolar concentrations. In the HPAF-II xenograft model, a regimen of XB002 (1.5 mg/kg, i.v.) administered once weekly for two weeks achieved complete tumor regression, with no detectable tumor growth up to five weeks after the second dose. In murine patient-derived xenograft models, a single dose of XB002 (10 mg/kg, i.v.) inhibited tumor growth across multiple cancer models, including bladder, cervical, gastric, head and neck squamous cell carcinoma, and non-small cell lung cancers. Remarkably, complete tumor regression was observed in the cervical cancer and head and neck squamous cell carcinoma models within 30 days of treatment. In nonhuman primate studies, XB002 demonstrated favorable pharmacokinetics with exposure in the desired therapeutic range and no signs of bleeding or neutropenia. Collectively, these data highlight XB002's broad-spectrum antitumor activity and strongly support its further clinical development.

Figure 1.

Effect of 25A3 on the coagulation cascade. Effects on the coagulation cascade were assessed using a cell-based FXa conversion assay (A) and a thrombin generation assay (B). Thrombin generation was measured in NPP in the presence of relipidated TF and titrated antibodies. Peak IIa [highest thrombin concentration generated on the thrombin generation curve (nmol/L)] is reported relative to plasma in the absence of antibody. For both assays, a TF-specific antibody, TF-011, was used as the positive control.

Figure 2.

Structure of XB002.

Figure 3.

Cytotoxicity activity of XB002 in A431 cells. Graph shows mean and SD of quadruplicate wells. The same antibody (25A3) was either conjugated to zovodotin (to generate XB002) or to vedotin as a comparator. RLU, relative light unit.

Figure 4.

Inhibition of tumor growth in vivo with XB002 in the HPAF-II xenograft model. A, Comparison between XB002 and 25A3 conjugated to vedotin. B, Dose–response study with a single dose of XB002. C, Late-intervention model with a single dose of XB002. ^†Animals in the 5 mg/kg group received a second dose on day 18, once the average tumor size for the group reached 400 mm³. Statistical significance for the Kruskal–Wallis Dunn or Mann–Whitney U test: *, P < 0.05; **, P < 0.01, when compared with vehicle on day 15.

Figure 5.

Inhibition of tumor growth with XB002 in PDX models. Inhibition of tumor growth with XB002 in gastric (A), cervical (B), bladder (C), HNSCC (D), and NSCLC (E) PDX models. TGI, tumor growth inhibition.

Figure 6.

XB002 mechanism of action.