165 Capitalizing On Epitope Escape to Avoid T Cell Fratricide without Compromising Targeting of CD56+ Tumors

Track: Contributed Abstracts
Wednesday, February 13, 2013, 6:45 PM-7:45 PM
Hall 1 (Salt Palace Convention Center)
Denise Crossland, B.S. Biology, B.S. Biochem , Experimental Pediatrics, MD Anderson Cancer Center UTHSCH, Houston, TX
Sonny Ang, PhD , Stem Cell Transplantation and Cellular Therapies, University of Texas MD Anderson Cancer Center, Houston, TX
Simon Olivares, MS , Pediatrics, University of Texas MD Anderson Cancer Center, Houston, TX
Helen Huls, BS , Pediatrics Research, UT MDACC, Houston, Houston, TX
Laurence J.N. Cooper, MD, PhD , Pediatrics, UT MDACC, Houston, TX
The tumor-associated antigen (TAA) CD56 is expressed at high levels on multiple malignancies as well as on normal T cells. A T-cell based therapy targeting CD56 has clinical application before or after hematopoietic stem-cell transplantation to improve the graft-versus-tumor effect. Therefore, we tested the hypothesis that T cells with redirected specificity for CD56 could evade fratricide while maintaining targeting for this TAA. To this end we developed a second generation chimeric antigen receptor (CAR, designated CD56CAR) with the specificity of the CD56-specific monoclonal antibody (mAb) clone N901, already administered in clinical trials. Primary human T cells were electrotransferred with a DNA plasmid coding for this CD56CAR based on the Sleeping Beauty transposon/transposase system. CAR+ T cells were selectively propagated on γ-irradiated CD56+ artificial antigen presenting cells (aAPC, recursively added every 7 days), in the presence of exogenous recombinant human IL-2 and IL-21. Genetically modified T cells were numerically expanded to clinically significant numbers over 28 days resulting in almost 100% of central and effector memory populations of CD3+CD4+ and CD3+CD8+ T cells stably expressing the CD56CAR. In vitro studies demonstrated CD56-specific cytotoxicity and cytokine release from CD56CAR+ T cells in the presence of CD56+ tumor targets and specificity was confirmed by addition of a CD56-specific blocking antibody. A subset of CD56CAR+ T cells co-expressed CD56, yet they did not exhibit fratricide. This was demonstrated by a lack of significant lysis of autologous CD56+CD56CAR+ T cells although there was killing of (unmodified) autologous CD56+ targets (NK cells and CD56+ T cells) as well as against genetically modified CD19-specific CD56+ CAR+ T cells. The mechanism for the evasion of autolysis by CD56-specific T cells was evaluated using a panel of mAbs specific for CD56. We observed that CD56 expression on the cell surface of CD56CAR+ T cells varied with the mAb used. CD56+CD56CAR+ T cells had significantly diminished staining with the mAb clone N901 over the culture period on aAPC. The epitope escape from N901 began at day 7 and corresponded with increaseed staining for annexin V/propidium iodide inferring selective pressure on the genetically modified T cells. In vitro tumor and T-cell co-culture experiments revealed an inability for tumor cells to undergo similar CD56 epitope escape in response to challenge by CD56-specific CAR+ T cells. In vivo CD56CAR+ T cells demonstrated the ability to decrease tumor burden and increase overall survival of mice bearing systemic CD56+ tumor. In sum, CD56-specific CAR+ T cells are able to maintain CD56 antigen expression without fratricide, due to an ability to undergo epitope escape, and demonstrate clinical potential for use as an immunotherapy for treating a wide range of malignancies from neuroblastoma to NK-cell leukemia.