S915 Endoscopic Ultrasound-Guided Injection of NBTXR3 Activated by Radiotherapy With Concurrent Chemotherapy for Esophageal Adenocarcinoma: Interim Analysis of Feasibility and Safety

Introduction: Despite definitive chemoradiation (CRT) being the standard of care for inoperable esophageal adenocarcinoma, only 25% of patients achieve a complete treatment response. This highlights the necessity to enhance cure rates for inoperable patients and to improve clinical outcomes for operable individuals. We report the interim results of a phase 1 trial of endoscopic ultrasound-guided fine needle tumor injection (EUS-FNI) of NBTXR3 (R3), a hafnium oxide nanoparticle that enhances the local radiation effect by 9-fold when activated by ionizing radiation. Methods: This is a planned interim analysis of the feasibility and safety of the first 9 patients who underwent EUS-FNI of R3 and received photon therapy. Patients with bulky mediastinal tumors or tumors above the carina are excluded. Patients received escalating doses of R3, corresponding to 15%, 22%, and 33% of the gross tumor volume (GTV). Investigational drug delivery of R3 was performed under EUS guidance; the tumor was injected at multiple sites to ensure uniform distribution within the mass. A 22-gauge needle was used for EUS-FNI, and the rate of injection was 1 cc/min. Verification computed tomography (CT) scan is taken within 3 days of the injection to visualize the distribution of the radiopaque R3 prior to starting CRT. Results: Only 1 patient each was treated at 15% and 22%, with 7 treated at 33% of the GTV, a total of 9 patients were enrolled. There were no periprocedural adverse events (AEs) during EUS-FNI. All patients completed CRT without any toxicity that could be specifically attributable to R3. Two of the 6 patients with surgery had a complete pathologic response, and 4 of 6 had major pathologic response. Four patients experienced local recurrence and/or metastasis following CRT, with or without surgery. R3 was consistently visualized on verification CT scans, with retained particles observed in all patients during follow-up. No delayed AEs were noted during follow-up. Pharmacokinetic study shows R3 leakage into the bloodstream to be under the limit of detection for all patients. Conclusion: This interim analysis showed that EUS-FNI of R3 is technically feasible, the investigational drug was retained in all patients, and no periprocedural or delayed AEs were noted. Initial clinical response and outcomes are promising. Ongoing accrual and expansion will allow a more comprehensive assessment of treatment efficacy and long-term outcomes.