Abstract There is pressing need for new therapeutic modalities targeting ovarian cancer, the most lethal gynecologic malignancy. We present in this study a novel immunotherapy strategy against ovarian cancer that utilizing human cytolytic T lymphocytes (CTLs) genetically re-directed to target L1-CAM, a clinically relevant ovarian tumor associated-antigen, as it is over-expressed by patient-derived ovarian tumors and is associated with poor prognosis. L1-CAM-specific targeting is achieved by engineering CTLs genetically to express a HLA-independent chimeric antigen receptor (CAR), which enables CTL to specifically recognize L1-CAM, and initiates T cell cytolytic killing via an intracellular CD3-zeta signaling domain. We report that these genetically re-programmed CTLs specifically kill L1-CAM+ OVCAR-3 human ovarian cancer cells in vitro. Importantly, single dose administration of i.p. L1CAM-CAR+ CTL induced regression of i.p established OVCAR-3 xenograft tumor growth in mice and delayed tumor progression for 6 to 20-days compared to untreated mice. Therapeutic efficacy could be further improved by multiple CTL injections (4 weekly), which delayed tumor progression for 11- to 32-days compared to untreated mice and also conferred a significant survival advantage (median survival days: 62 versus 80.5 days). For the first time, these results demonstrate L1-CAM is amenable to targeting by CAR+ T cells and support the potential of this therapeutic approach in ovarian cancer. However, ultimately tumor recurrence was observed in mice, highlighting the need for further optimization of this therapeutic platform. To investigate involvement of ovarian cancer stem-like cells (CSCs), a population with potent tumor-initiating potential, in this relapse phenomenon, we isolated subpopulations from OVCAR-3 with CSC-like properties including the formation of spheroids in serum-free culture, self-renewal and preferential expression of stem cell markers such as Oct4, Nestin and Bmi-1. Affymetrix microarray revealed these CSCs-like spheroids have a globally distinctive genetic profile compared with serum-differentiated cells, and importantly, one of the most significantly reduced genes in spheroids is L1-CAM (23.6-fold decrease). This decrease in L1-CAM was further validated by quantitative RT-PCR, western blot, confocal microscopy and flow cytometry. Together, these results suggest that an important mechanism for immune escape from L1CAM-CAR+ CTL killing is loss of antigen expression on CSC-like population. Overall, our findings indicate that adoptive transfer of genetically engineered L1-CAM-specific T cells can be a potential therapy for ovarian cancer, and we are conducting further studies to uncover additional antigenic targets that are enriched in the ovarian CSC population. We envision this combined immune approach will have profound implications for opening a new door to treat this deadly disease. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr LB-166. doi:10.1158/1538-7445.AM2011-LB-166