477 BI-Specific T CELL Therapy for Pancreatic Cancer

Track: Contributed Abstracts
Saturday, February 16, 2013, 6:45 PM-7:45 PM
Hall 1 (Salt Palace Convention Center)
Usanarat Anurathapan, MD , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Robert C Chan, MD , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Hakeem F. Hindi, BA , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Roopa Mucharla, MSc , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Helen E. Heslop, MD , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Cliona M. Rooney, PhD , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Malcolm K. Brenner, MD, PhD , Center for Cell and Gene Therapy, Baylor College of Medicine, Texas Children's Hospital, Houston, TX
Ann M Leen, PhD , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Juan F. Vera, MD , Baylor College of Medicine, Texas Children's Hospital, The Methodist Hospital, Houston, TX
Adoptive transfer of T cells directed to tumor-associated antigens (TAAs) by transgenic expression of chimeric antigen receptors (CARs) can produce tumor responses, even in patients with resistant malignancies. However targeting a single TAA may lead to selection of the nontargeted TAA population, leading to tumor immune escape. To overcome this limitation, we have targeted two distinct TAAs, MUC1 and PSCA, both commonly express in pancreatic cancer. We first constructed a CAR targeting MUC1, which coexpressed a truncated form of CD19 (ΔCD19) as a selectable marker. After retroviral transduction, primary T cells stably expressed both transgenes (83±4% and 66±8%, CAR-MUC1 and ΔCD19, respectively). CAR-MUC1 T cells were able to specifically kill MUC1+ target cell lines, CAPAN1 and DU145, with no recognition of MUC1- targets, 293T (35±5%, 23±4% and 3±2% specific lysis, respectively, 10:1 E:T). To further evaluate the longer term killing effects, we cultured these MUC1+ targets with CAR-MUC1 T cells, in the presence of IL2 for 72h, and found that >60% of CAPAN1 and >40% of DU145 were killed after an initial treatment but were resistant to subsequent treatment. Importantly IHC analysis of these resistant tumor cells demonstrated lack of expression of the targeted antigen, MUC1. This observation was confirmed in an artificial model using 293T cells engineered to express either MUC1 (mOrange+) or PSCA (GFP+), which illustrated an exquisite specificity of T cells but at the same time a weakness when targeting a heterogeneous tumor cell population. Therefore we generated a second CAR targeting the TAA PSCA, and demonstrated stable expression on primary T cells (89±2%), which were able to kill the PSCA+ cell lines, CAPAN1 and DU145, with no effect on control PSCA- 293T,  (48±6%, 41±46% and 4±2% specific lysis, respectively, 10:1 E:T). Finally, we assessed the anti-tumor effects when both CAR-modified products were combined. When tumor cells expressing both TAAs were treated with CAR-MUC1 + CAR-PSCA T cells we saw additive anti-tumor effects with 76±10% killing of CAPAN1 compared with only 35±6% and 48±6%, respectively, using CAR-MUC1 and CAR-PSCA T cells individually. We saw similar results using our engineered 293T tumor model. Hence the combination of CARs that target two distinct TAA expressed on cancer cells (PSCA and MUC1) may prevent tumor immune escape and enhance the potency of adoptive T cell transfer.
Previous Abstract | Next Abstract >>