Preliminary results of a phase 1 study of LB1908, an autologous Claudin 18.2–targeted chimeric antigen receptor T-cell product, in patients with advanced gastroesophageal adenocarcinoma.

4022 Background: Claudin 18.2 (CLDN18.2) is a potential therapeutic target expressed on most gastric, gastroesophageal, and esophageal adenocarcinomas (GC/GEJC/EC), with normal tissue expression limited to gastric epithelium. We present preliminary results of the dose-escalation trial of LB1908, a CLDN18.2-targeted autologous CAR-T cell product, in adults with advanced GC/GEJC/EC. Methods: This trial is an ongoing, open-label, multicenter, first-in-human phase 1 study of LB1908 in patients with advanced GC/GEJC/EC relapsed/refractory to ≥1 prior line of therapy (LOT), and with ≥1+ CLDN18.2 expression in ≥50% of tumor cells by central testing (43-14A antibody). A modified 3+3 design is used for dose escalation with planned dose levels of 0.5, 1.5, and 3×10 6 CAR+ T cells/kg. The primary objective is evaluation of safety and dose-limiting toxicities (DLTs) with secondary objectives of antitumor activity and pharmacokinetics. Results: As of the data cutoff of January 4, 2025, 6 patients were dosed with LB1908 at dose level 1 (0.5×10 6 cells/kg). Four patients had EC and 2 had GC, all with metastatic disease. Patients had a median of 2 prior LOTs; all had received fluoropyrimidine/platinum agents. Bridging therapy was administered to all patients between apheresis and LB1908 infusion, and patients received standard lymphodepletion (LD) with cyclophosphamide/fludarabine. All patients experienced treatment-emergent adverse events (TEAEs); the most common grade ≥3 TEAEs were hematologic and attributable to the LD regimen. Of grade ≥3 TEAEs related to LB1908, only gastritis/gastric mucosal injury occurred in more than 1 patient (n = 3), including 1 DLT. After implementation of toxicity management guidelines using prophylactic enteral beclomethasone and early systemic steroids, no patients experienced prolonged grade ≥3 upper GI AEs. CRS occurred in all patients, with no grade ≥3 events. No ICANS was observed. CAR-T cell expansion was seen in all patients, with a median C max of 1594 copies/µg genomic DNA (range, 264-6922) and T max of 14 days (range, 11-15). All 6 patients were evaluable for response, with 5 (83%) experiencing target lesion shrinkage (maximal 1%, 9%, 25%, 31%, and 41% reduction from baseline). Responses deepened over time in the 2 patients with multiple postinfusion scans (-25% and -41%, respectively), including 1 patient achieving a RECIST partial response 7 months after treatment. Conclusions: LB1908 demonstrates peripheral expansion and encouraging antitumor activity at the lowest dose tested, with a manageable safety profile. Implementation of a toxicity mitigation strategy ameliorated on-target gastric mucosal injury without compromising expansion kinetics or antitumor activity. Longer follow-up and data from patients treated with higher cell doses will be presented at the meeting. Clinical trial information: NCT05539430 .