Autologous multiantigen-targeted T cell therapy for pancreatic cancer: a phase 1/2 trial.

T cell therapy has proven challenging for pancreatic ductal adenocarcinoma (PDAC), partly due to heterogeneous expression of tumor-associated antigens (TAAs). To address tumor heterogeneity and mitigate immune evasion, an ex vivo expanded, polyclonal, T helper 1 cell-polarized T cell product targeting five TAAs-PRAME, SSX2, MAGEA4, Survivin and NY-ESO-1-was developed. These antigens were chosen based on their tumor specificity, oncogenicity, immunogenicity and level of expression. In a phase 1/2 trial, this autologous nonengineered T cell product was administered (1 × 107 cells m-2 per infusion) monthly to patients with advanced PDAC responding (arm A, n = 13) or refractory (arm B, n = 12) to first-line chemotherapy or with resectable disease (arm C, n = 12). Primary endpoints were safety and feasibility of completing six infusions, whereas exploratory efficacy endpoints included persistence and evaluating the relationship between clinical benefit and the expansion of the infused effector T cells, as well as the induction of de novo immune responses. Of 56 participants procured, 37 were infused, with only 1 treatment-related serious adverse event. Disease control rates in arms A and B were 84.6% (95% confidence interval: 54.6-98.1%) and 25% (95% confidence interval: 5.5-57.2%), respectively. In arm C, two of nine resected participants remained disease free after 66 months of follow-up. The infused cells persisted up to 12 months posttreatment and elevated levels of tumor-directed T cells were detected during dosing (P = 0.027) and follow-up in responders compared to nonresponders. Clinical outcomes correlated with peripheral expansion of functional TAA-targeted T cell clones and treatment-emergent antigen spreading. Thus, further investigation of this approach, either as a single agent or combined with other complementary modalities, is warranted (ClinicalTrials.gov identifier: NCT03192462 ).