Characterization of osemitamab in pancreatic cancer models and patients

Background: TST001 (osemitamab) is a high affinity humanized, ADCC enhanced antibody targeting CLDN18.2. It specifically binds to the extracellular domains of CLDN18.2 and eliminates tumor cells by ADCC and CDC. Promising efficacy of TST001 monotherapy in late line G/GEJ cancer patients or plus CAPOX with or without nivolumab as first-line treatment has been observed and reported. TST003 is a novel humanized antibody targeting Gremlin-1, a member of TGF-beta superfamily. Gremlin1 promotes epithelial-mesenchymal transition (EMT) and cancer cell proliferation by binding to BMPs and blocking its biological activities. Here we report preclinical anti-tumor activities of TST001 monotherapy or combined with TST003 in pancreatic cancer models and TST001 monotherapy in pancreatic cancer patients. Methods: The CLDN18.2 expression on the pancreatic cancer cells was evaluated using IHC analysis with 14G11 antibody. The ADCC activity of TST001 on pancreatic cancer cells was assessed by ADCC reporter cell in vitro. Its in vivo anti-tumor activities were investigated in pancreatic cancer models. In a TST001 phase I clinical trial (NCT04495296), pancreatic cancer patients who failed prior available standard therapies were enrolled and received TST001 monotherapy at 10 mg/kg every 3 weeks. Results: TST001 displayed potent ADCC activities for two pancreatic cancer cell lines (BxPC-3-CLDN18.2 and MIA PaCa-2-CLDN18.2) in vitro. In KRAS wild type BxPC-3-CLDN18.2 model, the tumor growth inhibition (TGI) of TST001 was 61% at 3 mg/kg and 98% at 10 mg/kg. 7 out of 10 mice in the 10 mg/kg group had their tumors completely disappeared from Day 33. In MIA PaCa-2-CLDN18.2 with KRAS mutation, TST001 at 10 mg/kg led to TGI= 49% and combination with gemcitabine (30 mg/kg) improved the TGI to 67%. In the phase 1 trial, a pancreatic cancer patient with liver metastasis and failed prior gemcitabine plus S1 chemotherapy achieved durable clinical benefit. Despite its tumor has low CLDN18.2 expression (5% 1+, 5% 2+, 5% 3+ tested by a validated IHC assay in a central lab) and KRAS G12R mutation (by local test), the primary target lesion shrank 86% at week 6 and complete response was achieved after the patient received TST001 treatment for 270 days. As Gremlin1 is highly expressed in pancreatic cancer, we also tested the anti-tumor activity of the combination of TST001 and TST003 using the BxPC-3-CLDN18.2/Gremlin1 co-expressing tumor model with PBMC co-inoculation. 3 mg/kg of TST001 combined with 30 mg/kg TST003 exhibited significantly better TGI (60%) than monotherapy (34% for TST001 and 28% for TST003). Conclusions: TST001 displayed significant anti-tumor activity in preclinical pancreatic cancer models and higher efficacy when combined with gemcitabine or TST003. A pretreated pancreatic cancer patient achieved complete response with TST001 monotherapy. These findings support further investigation of TST001 in CLND18.2 positive pancreatic cancer patients.