Phase I dose-escalation study of talimogene laherparepvec (T-VEC) for advanced pancreatic cancer (ca).

Abstract

e14546 Background: T-VEC is an investigational, immune-enhanced oncolytic herpes simplex virus type I that selectively replicates in solid tumors. This was an open-label, dose-escalation study of T-VEC administered by endoscopic ultrasound (EUS)-guided fine needle injection (FNI) for advanced pancreatic ca. Methods: Eligibility criteria included ≥ 18 yrs old, ECOG 0-2, pathologically confirmed pancreatic ca with measurable disease (tumors ≥ 1 cm dia), and failure or unable to receive standard tx. T-VEC was administered at wks 6, 12, 18 to a single pancreatic tumor in 4 cohorts (C), 3 pts each enrolled sequentially (extended dosing allowed in pts thought to have benefit): C 1: 1 dose of 104 PFU/mL then 2 of 105 PFU/mL; C 2: 1 dose of 105 PFU/mL then 2 of 106 PFU/mL; C 3: 1 dose of 106 PFU/mL then 2 of 107 PFU/mL; C 4: 1 dose of 106 PFU/mL then 2 of 108 PFU/mL. 2 doses/pt were required for enrollment to the next C. A C could be expanded to 6 if 1 related DLT occurred. If 2 DLTs occurred, dose escalation was stopped. Endpoints were safety and activity as assessed by CT tumor dia changes from screening. Results: 17 pts were enrolled in C 1-3 given at least 1 dose of T-VEC; 65% were men; median age 54; 76% white, ECOG 0-1 82%. C 4 was not opened because of early study termination (not related to safety). 7/17 (41%) received all 3 planned doses and 8 (47%) had at least 1 post-dose CT scan. Only C 3 showed a median decrease in injected tumor dia with 2/4 pts achieving substantial tumor reductions (-36% and -33%). 3 pts (in C 1and C 3) showed decreases in the dia of ≥ 1 uninjected tumors (in liver, pancreas, kidney, and chest); 1 pt had disappearance of a nonmeasurable tumor in the liver. A dose trend was not observed for reductions of uninjected tumors. Most common AEs were ascites (47%), dehydration (41%), anemia, abdominal pain, constipation, and nausea (each 35%), and vomiting (29%). 2 pts (12%) had a grade 5 AE, both considered unrelated to T-VEC. Conclusions: EUS-guided FNI of T-VEC in advanced pancreatic ca, at initial doses of 104 to 106 PFU/mL followed by up to 107 PFU/mL, was feasible and tolerable. Evidence of biologic activity was observed. Future studies should be conducted in pts with less advanced disease, to allow sufficient time to receive multiple doses before PD.