Search

Adoptive transfer, persistence, and migration of autologous control gene-marked T cells.

Serum chemistry and cytokine levels in control animals.

10 million (A) CD8+ TCM-derived CD19-CAR-T-cells, (B) CD19-CAR-T-cells derived from an unselected T-cell population, or (C) CD8+ TCM-derived ROR1-CAR-T-cells were injected into tumor-free NSG mice. After 24 hours, bone marrow, spleens, and peripheral blood were isolated and analyzed for the presence of T-cells (CD45+/CD4+ or CD45+/CD8+) and the expression of activation markers CD25 and CD69 by flow cytometry.

(A) Cytolytic activity of CD8+ TCM-derived T-cells expressing long/CD28, long/4-1BB, and long/CD28_4-1BB CD19-CARs against K562/CD19, K562, and Raji cells. (B) Cytokine production by CAR-T-cells after co-culturing with Raji cells. Supernatants were obtained after 24 hours for analysis. (C) Proliferation of CD19-CAR-T-cells 72 hours after stimulation with Raji cells by CFSE dye dilution. Numbers above each histogram indicate the number of cell divisions the proliferating subset underwent, and the fraction of T-cells in each gate that underwent {greater than or equal to}4/3/2/1 cell divisions is provided in the upper left of each plot.

Genetic engineering of T cells for adoptive transfer by introducing a tumor-targeting chimeric antigen receptor (CAR) is a new approach to cancer immunotherapy. A challenge for the field is to define cell surface molecules that are both preferentially expressed on tumor cells and can be safely targeted with T cells. The orphan tyrosine kinase receptor ROR1 is a candidate target for T-cell therapy with CAR-modified T cells (CAR-T cells) because it is expressed on the surface of many lymphatic and epithelial malignancies and has a putative role in tumor cell survival. The cell surface isoform of ROR1 is expressed in embryogenesis but absent in adult tissues except for B-cell precursors and low levels of transcripts in adipocytes, pancreas, and lung. ROR1 is highly conserved between humans and macaques and has a similar pattern of tissue expression. To determine if low-level ROR1 expression on normal cells would result in toxicity or adversely affect CAR-T cell survival and/or function, we adoptively transferred autologous ROR1 CAR-T cells into nonhuman primates. ROR1 CAR-T cells did not cause overt toxicity to normal organs and accumulated in bone marrow and lymph node sites, where ROR1-positive B cells were present. The findings support the clinical evaluation of ROR1 CAR-T cells for ROR1+ malignancies and demonstrate the utility of nonhuman primates for evaluating the safety of immunotherapy with engineered T cells specific for tumor-associated molecules that are homologous between humans and nonhuman primates. Cancer Immunol Res; 3(2); 206–16. ©2014 AACR.

Gating strategy for ROR1-expressing B cells in the lymph node.

The use of synthetic chimeric antigen receptors (CAR) to redirect T cells to recognize tumor provides a powerful new approach to cancer immunotherapy; however, the attributes of CARs that ensure optimal in vivo tumor recognition remain to be defined. Here, we analyze the influence of length and composition of IgG-derived extracellular spacer domains on the function of CARs. Our studies demonstrate that CD19-CARs with a long spacer from IgG4 hinge-CH2-CH3 are functional in vitro but lack antitumor activity in vivo due to interaction between the Fc domain within the spacer and the Fc receptor–bearing myeloid cells, leading to activation-induced T-cell death. We demonstrate that in vivo persistence and antitumor effects of CAR-T cells with a long spacer can be restored by modifying distinct regions in the CH2 domain that are essential for Fc receptor binding. Our studies demonstrate that modifications that abrogate binding to Fc receptors are crucial for CARs in which a long spacer is obligatory for tumor recognition as shown here for a ROR1-specific CAR. These results demonstrate that the length and composition of the extracellular spacer domain that lacks intrinsic signaling function can be decisive in the design of CARs for optimal in vivo activity. Cancer Immunol Res; 3(2); 125–35. ©2014 AACR.

Supplemental Materials and Methods

(A) Mice were inoculated with 0.5 million Raji-ffluc cells via tail vein injection and tumor engraftment was confirmed by bioluminescence imaging on day six. On day seven, mice received a single i.v. injection of various doses of CD8+ TCM-derived T-cells transduced with the short spacer CD19/4-1BB/CD3ζ-CAR or of control T-cells (EGFRt). (B) Tumor growth was measured by bioluminescence imaging and is plotted for individual mice (4 mice per group). Arrows mark the day of T-cell transfer. Bioluminescence images from day six (one day before T-cell transfer) and day 19 (12 days after transfer) are shown.

FIGURE S4: Analysis of tonic signaling and function of trispecific and monspecific DARPin CARs

Purpose:The outgrowth of antigen-negative variants is a significant challenge for adoptive therapy with T cells that target a single specificity. Chimeric antigen receptors (CAR) are typically designed with one or two scFvs that impart antigen specificity fused to activation and costimulation domains of T-cell signaling molecules. We designed and evaluated the function of CARs with up to three specificities for overcoming tumor escape using Designed Ankyrin Repeat Proteins (DARPins) rather than scFvs for tumor recognition.Experimental Design:A monospecific CAR was designed with a DARPin binder (E01) specific for EGFR and compared with a CAR designed using an anti-EGFR scFv. CAR constructs in which DARPins specific for EGFR, EpCAM, and HER2 were linked together in a single CAR were then designed and optimized to achieve multispecific tumor recognition. The efficacy of CAR-T cells bearing a multispecific DARPin CAR for treating tumors with heterogeneous antigen expression was evaluated in vivo.Results:The monospecific anti-EGFR E01 DARPin conferred potent tumor regression against EGFR+ targets that was comparable with an anti-EGFR scFv CAR. Linking three separate DARPins in tandem was feasible and in an optimized format generated a single tumor recognition domain that targeted a mixture of heterogeneous tumor cells, each expressing a single antigen, and displayed synergistic activity when tumor cells expressed more than one target antigen.Conclusions:DARPins can serve as high-affinity recognition motifs for CAR design, and their robust architecture enables linking of multiple binders against different antigens to achieve functional synergy and reduce antigen escape.

FIGURE S3: Transduced targets for testing of multispecific CARs.

Supplementary Figure Legends

Purpose: This study examines cell surface ROR1 expression in human tumors and normal tissues. ROR1 is considered a promising target for cancer therapy due to putative tumor-specific expression, and multiple groups are developing antibodies and/or chimeric antigen receptor–modified T cells to target ROR1. On-target, off-tumor toxicity is a challenge for most nonmutated tumor antigens; however, prior studies suggest that ROR1 is absent on most normal tissues.Experimental Design: Our studies show that published antibodies lack sensitivity to detect endogenous levels of cell surface ROR1 by immunohistochemistry (IHC) in formalin-fixed, paraffin-embedded tissues. We developed a ROR1-specific monoclonal antibody (mAb) targeting the carboxy-terminus of ROR1 and evaluated its specificity and sensitivity in IHC.Results: The 6D4 mAb is a sensitive and specific reagent to detect cell surface ROR1 by IHC. The data show that ROR1 is homogenously expressed on a subset of ovarian cancer, triple-negative breast cancer, and lung adenocarcinomas. Contrary to previous findings, we found ROR1 is expressed on several normal tissues, including parathyroid; pancreatic islets; and regions of the esophagus, stomach, and duodenum. The 6D4 mAb recognizes rhesus ROR1, and ROR1 expression was similar in human and macaque tissues, suggesting that the macaque is a suitable model to evaluate safety of ROR1-targeted therapies.Conclusions: ROR1 is a promising immunotherapeutic target in many epithelial tumors; however, high cell surface ROR1 expression in multiple normal tissues raises concerns for on-target off-tumor toxicities. Clinical translation of ROR1-targeted therapies warrants careful monitoring of toxicities to normal organs and may require strategies to ensure patient safety. Clin Cancer Res; 23(12); 3061–71. ©2016 AACR.

PDF File - 122K, Supplementary Figure 5: Expression of ROR1 and NKG2D ligands on epithelial cancer cell lines.

PDF File - 169K, Supplementary Figure 4: The function of ROR1-CAR and CD19-CAR modified CD8+ T-cells against primary CLL is augmented by CAR-modified CD4+ helper T-cells.

FIGURE S5: Functional assessment of trispecific DARPin CAR T cells against multiple antigens

PDF File - 125K, Supplementary Figure 3: In vitro function of T-cells expressing CD19-CARs with IgG4-Fc Hinge vs. CD8a Hinge spacer and a signaling module with 4-1BB costimulatory domain.

FIGURE S1: Design and testing of E01 DARPin CAR

PDF File - 129K, Supplementary Figure 1: Design of ROR1-CARs with modified spacer length and derived from the 2A2 and R12 scFV with different affinity.

Purpose: The adoptive transfer of T cells modified to express a chimeric antigen receptor (CAR) comprised of an extracellular single-chain antibody (scFV) fragment specific for a tumor cell surface molecule, and linked to an intracellular signaling module, has activity in advanced malignancies. The receptor tyrosine kinase–like orphan receptor 1 (ROR1) is a tumor-associated molecule expressed in prevalent B-lymphoid and epithelial cancers and is absent on normal mature B cells and vital tissues, making it a candidate for CAR T-cell therapy.Experimental Design: We constructed ROR1-CARs from scFVs with different affinities and containing extracellular IgG4-Fc spacer domains of different lengths, and evaluated the ability of T cells expressing each CAR to recognize ROR1+ hematopoietic and epithelial tumors in vitro, and to eliminate human mantle cell lymphoma (MCL) engrafted into immunodeficient mice.Results: ROR1-CARs containing a short “Hinge-only” extracellular spacer conferred superior lysis of ROR1+ tumor cells and induction of T-cell effector functions compared with CARs with long “Hinge-CH2-CH3” spacers. CARs derived from a higher affinity scFV conferred maximum T-cell effector function against primary CLL and ROR1+ epithelial cancer lines in vitro without inducing activation-induced T-cell death. T cells modified with an optimal ROR1-CAR were equivalently effective as CD19-CAR–modified T cells in mediating regression of JeKo-1 MCL in immunodeficient mice.Conclusions: Our results show that customizing spacer design and increasing affinity of ROR1-CARs enhances T-cell effector function and recognition of ROR1+ tumors. T cells modified with an optimized ROR1-CAR have significant antitumor efficacy in a preclinical model in vivo, suggesting they may be useful to treat ROR1+ tumors in clinical applications. Clin Cancer Res; 19(12); 3153–64. ©2013 AACR.

FIGURE S2: In vitro function and in vivo persistence of E01 DARPin CAR-T cells

PDF File - 83K, Supplementary Figure 2: Analysis of cytokine production and proliferation of CD4+ T-cells lines modified with a ROR1-CAR derived from mAb R12 with higher affinity than 2A2.

TABLE S1: 9-mer peptides in the scFv and DARPin sequences with predicted binding to the MHC class I supertype with a predicted binding affinity of less than 100 nM

Supplementary Figure 1: IHC staining using previously published ROR1 antibodies on control and ROR1 transfected K562 cells, CLL and MCL lymph nodes and tonsil tissue. Supplementary Figure 2: Development of mAb 6D4 Supplementary Figure 3: Representative images of IHC scoring of ROR1+ tumors. Supplementary Figure 4: ROR1 expression in pancreatic adenocarcinomas using mAb 6D4. Supplementary Figure 5: ROR1 staining in absent in normal human cerebrum, cerebellum, heart, lung, spleen and liver. Scale bar represents 100μm. Supplementary Figure 6: Immunoblot validation of ROR1 expression in normal tissues. Supplementary Figure 7: Transcript expression of Ror1 variant 1 in CLL PBMC, differentiated adipocytes and normal human gastrointestinal tissues.

PDF File - 102K, Legend for Supplementary Figures 1-5

Purpose: The outgrowth of antigen-negative variants is a significant challenge for adoptive therapy with T cells that target a single specificity. Chimeric antigen receptors (CAR) are typically designed with one or two scFvs that impart antigen specificity fused to activation and costimulation domains of T-cell signaling molecules. We designed and evaluated the function of CARs with up to three specificities for overcoming tumor escape using Designed Ankyrin Repeat Proteins (DARPins) rather than scFvs for tumor recognition. Experimental Design: A monospecific CAR was designed with a DARPin binder (E01) specific for EGFR and compared with a CAR designed using an anti-EGFR scFv. CAR constructs in which DARPins specific for EGFR, EpCAM, and HER2 were linked together in a single CAR were then designed and optimized to achieve multispecific tumor recognition. The efficacy of CAR-T cells bearing a multispecific DARPin CAR for treating tumors with heterogeneous antigen expression was evaluated in vivo. Results: The monospecific anti-EGFR E01 DARPin conferred potent tumor regression against EGFR+ targets that was comparable with an anti-EGFR scFv CAR. Linking three separate DARPins in tandem was feasible and in an optimized format generated a single tumor recognition domain that targeted a mixture of heterogeneous tumor cells, each expressing a single antigen, and displayed synergistic activity when tumor cells expressed more than one target antigen. Conclusions: DARPins can serve as high-affinity recognition motifs for CAR design, and their robust architecture enables linking of multiple binders against different antigens to achieve functional synergy and reduce antigen escape.

Purpose: This study examines cell surface ROR1 expression in human tumors and normal tissues. ROR1 is considered a promising target for cancer therapy due to putative tumor-specific expression, and multiple groups are developing antibodies and/or chimeric antigen receptor–modified T cells to target ROR1. On-target, off-tumor toxicity is a challenge for most nonmutated tumor antigens; however, prior studies suggest that ROR1 is absent on most normal tissues. Experimental Design: Our studies show that published antibodies lack sensitivity to detect endogenous levels of cell surface ROR1 by immunohistochemistry (IHC) in formalin-fixed, paraffin-embedded tissues. We developed a ROR1-specific monoclonal antibody (mAb) targeting the carboxy-terminus of ROR1 and evaluated its specificity and sensitivity in IHC. Results: The 6D4 mAb is a sensitive and specific reagent to detect cell surface ROR1 by IHC. The data show that ROR1 is homogenously expressed on a subset of ovarian cancer, triple-negative breast cancer, and lung adenocarcinomas. Contrary to previous findings, we found ROR1 is expressed on several normal tissues, including parathyroid; pancreatic islets; and regions of the esophagus, stomach, and duodenum. The 6D4 mAb recognizes rhesus ROR1, and ROR1 expression was similar in human and macaque tissues, suggesting that the macaque is a suitable model to evaluate safety of ROR1-targeted therapies. Conclusions: ROR1 is a promising immunotherapeutic target in many epithelial tumors; however, high cell surface ROR1 expression in multiple normal tissues raises concerns for on-target off-tumor toxicities. Clinical translation of ROR1-targeted therapies warrants careful monitoring of toxicities to normal organs and may require strategies to ensure patient safety. Clin Cancer Res; 23(12); 3061–71. ©2016 AACR.

Adoptive immunotherapy with genetically engineered T cells has the potential to treat cancer and other diseases. The introduction of Strep-tag II sequences into specific sites in synthetic chimeric antigen receptors or natural T-cell receptors of diverse specificities provides engineered T cells with a marker for identification and rapid purification, a method for tailoring spacer length of chimeric receptors for optimal function, and a functional element for selective antibody-coated, microbead-driven, large-scale expansion. These receptor designs facilitate cGMP manufacturing of pure populations of engineered T cells for adoptive T-cell therapies and enable in vivo tracking and retrieval of transferred cells for downstream research applications.

Adoptive T-cell therapy with gene-modified T cells expressing a tumor-reactive T-cell receptor or chimeric antigen receptor (CAR) is a rapidly growing field of translational medicine and has shown success in the treatment of B-cell malignancies and solid tumors. In all reported trials, patients have received T-cell products comprising random compositions of CD4(+) and CD8(+) naive and memory T cells, meaning that each patient received a different therapeutic agent. This variation may have influenced the efficacy of T-cell therapy, and complicates comparison of outcomes between different patients and across trials. We analyzed CD19 CAR-expressing effector T cells derived from different subsets (CD4(+)/CD8(+) naive, central memory, effector memory). T cells derived from each of the subsets were efficiently transduced and expanded, but showed clear differences in effector function and proliferation in vitro and in vivo. Combining the most potent CD4(+) and CD8(+) CAR-expressing subsets, resulted in synergistic antitumor effects in vivo. We show that CAR-T-cell products generated from defined T-cell subsets can provide uniform potency compared with products derived from unselected T cells that vary in phenotypic composition. These findings have important implications for the formulation of T-cell products for adoptive therapies.

Genetic engineering of T cells for adoptive transfer by introducing a tumor-targeting chimeric antigen receptor (CAR) is a new approach to cancer immunotherapy. A challenge for the field is to define cell surface molecules that are both preferentially expressed on tumor cells and can be safely targeted with T cells. The orphan tyrosine kinase receptor ROR1 is a candidate target for T-cell therapy with CAR-modified T cells (CAR-T cells) because it is expressed on the surface of many lymphatic and epithelial malignancies and has a putative role in tumor cell survival. The cell surface isoform of ROR1 is expressed in embryogenesis but absent in adult tissues except for B-cell precursors and low levels of transcripts in adipocytes, pancreas, and lung. ROR1 is highly conserved between humans and macaques and has a similar pattern of tissue expression. To determine if low-level ROR1 expression on normal cells would result in toxicity or adversely affect CAR-T cell survival and/or function, we adoptively transferred autologous ROR1 CAR-T cells into nonhuman primates. ROR1 CAR-T cells did not cause overt toxicity to normal organs and accumulated in bone marrow and lymph node sites, where ROR1-positive B cells were present. The findings support the clinical evaluation of ROR1 CAR-T cells for ROR1+ malignancies and demonstrate the utility of nonhuman primates for evaluating the safety of immunotherapy with engineered T cells specific for tumor-associated molecules that are homologous between humans and nonhuman primates. Cancer Immunol Res; 3(2); 206–16. ©2014 AACR.

The use of synthetic chimeric antigen receptors (CAR) to redirect T cells to recognize tumor provides a powerful new approach to cancer immunotherapy; however, the attributes of CARs that ensure optimal in vivo tumor recognition remain to be defined. Here, we analyze the influence of length and composition of IgG-derived extracellular spacer domains on the function of CARs. Our studies demonstrate that CD19-CARs with a long spacer from IgG4 hinge-CH2-CH3 are functional in vitro but lack antitumor activity in vivo due to interaction between the Fc domain within the spacer and the Fc receptor–bearing myeloid cells, leading to activation-induced T-cell death. We demonstrate that in vivo persistence and antitumor effects of CAR-T cells with a long spacer can be restored by modifying distinct regions in the CH2 domain that are essential for Fc receptor binding. Our studies demonstrate that modifications that abrogate binding to Fc receptors are crucial for CARs in which a long spacer is obligatory for tumor recognition as shown here for a ROR1-specific CAR. These results demonstrate that the length and composition of the extracellular spacer domain that lacks intrinsic signaling function can be decisive in the design of CARs for optimal in vivo activity. Cancer Immunol Res; 3(2); 125–35. ©2014 AACR.

Purpose: The adoptive transfer of T cells modified to express a chimeric antigen receptor (CAR) comprised of an extracellular single-chain antibody (scFV) fragment specific for a tumor cell surface molecule, and linked to an intracellular signaling module, has activity in advanced malignancies. The receptor tyrosine kinase–like orphan receptor 1 (ROR1) is a tumor-associated molecule expressed in prevalent B-lymphoid and epithelial cancers and is absent on normal mature B cells and vital tissues, making it a candidate for CAR T-cell therapy. Experimental Design: We constructed ROR1-CARs from scFVs with different affinities and containing extracellular IgG4-Fc spacer domains of different lengths, and evaluated the ability of T cells expressing each CAR to recognize ROR1+ hematopoietic and epithelial tumors in vitro, and to eliminate human mantle cell lymphoma (MCL) engrafted into immunodeficient mice. Results: ROR1-CARs containing a short “Hinge-only” extracellular spacer conferred superior lysis of ROR1+ tumor cells and induction of T-cell effector functions compared with CARs with long “Hinge-CH2-CH3” spacers. CARs derived from a higher affinity scFV conferred maximum T-cell effector function against primary CLL and ROR1+ epithelial cancer lines in vitro without inducing activation-induced T-cell death. T cells modified with an optimal ROR1-CAR were equivalently effective as CD19-CAR–modified T cells in mediating regression of JeKo-1 MCL in immunodeficient mice. Conclusions: Our results show that customizing spacer design and increasing affinity of ROR1-CARs enhances T-cell effector function and recognition of ROR1+ tumors. T cells modified with an optimized ROR1-CAR have significant antitumor efficacy in a preclinical model in vivo, suggesting they may be useful to treat ROR1+ tumors in clinical applications. Clin Cancer Res; 19(12); 3153–64. ©2013 AACR.

Adoptive therapy with T cells modified with synthetic chimeric antigen receptors (CAR-T cells) is emerging as an important and promising modality for hematologic malignancies, and has the potential for broad applications in cancer therapy. We have worked on novel CAR designs by incorporating strep-tag II in various locations in the extracellular domain of the CAR. We have shown that strep-tag II does not interfere with CAR function, provides an intrinsic marker for identification of CAR expression at the cell surface, and introduces flexibility in tailoring spacer length to optimize CAR function. Moreover, this CAR design can facilitate the application of time-saving and cost-efficient technologies for manufacturing of high-purity CAR T cell products either by using anti-strep-tag II plus anti-CD28 mAb coated microbeads to selectively expand CAR-T cells, or by rapidly enriching CAR-T cells using established Strep-Tactin purification systems. Using NSG mouse/human tumor xenograft models, we show that both strategies readily generate potent CAR T cell products for in vivo therapy. This innovation in CAR design can be universally applied for CARs with different scFv specificity (CD19, CD20, ROR1) and costimulatory domains (CD28, 4-1BB, CD28/4-1BB), and endows CAR-T cells with valuable properties for cGMP manufacturing of therapeutic CAR-T cell product with high efficiency.

Abstract 951 Adoptive immunotherapy with T cells engineered by gene transfer to express CD19-specific chimeric antigen receptors (CARs) has the potential to induce remissions in patients with advanced B cell malignancies. CARs are synthetic receptors with an extracellular antigen-binding domain (scFv), a spacer domain that provides separation of the scFv from the cell membrane and an intracellular signaling module, most commonly the CD3ζ chain and one or more costimulatory domains such as CD28 or 4-1BB. Several clinical trials with CD19-CAR T cells in small cohorts of patients with B cell tumors have been reported with variable results. Although most studies have used the CD19-specific FMC63 scFv as the tumor-targeting moiety, the extracellular, transmembrane and intracellular CAR domains used in each trial have been distinct, and an emerging paradigm is that including costimulation in the design of the CAR is key to achieving anti-tumor activity in vivo. In this study, we analyzed the influence of extracellular spacer domain length on the in vitro and in vivo function of CD19-CARs. We constructed a panel of four CD19-CARs comprised of the FMC63 scFv and either a long spacer derived from the IgG4-Fc Hinge-CH2-CH3 domain (229 AA) or a short Hinge domain only spacer (12 AA). Each CAR contained a signaling module of CD3ζ with CD28 (short/CD28; long/CD28) or 4-1BB (short/4-1BB; long/4-1BB). We transduced CD8+ CD45RO+ CD62L+central memory T cells of normal donors with each of the CARs, enriched transduced T cells to >90% purity by immunomagnetic selection using a tEGFR marker encoded in the CAR vector, and expanded CAR transduced T cells using a uniform culture protocol. We compared the in vitro function of T cell lines expressing each of the CD19-CARs and confirmed specific cytolytic activity against CD19+ target cells including K562/CD19, and Raji and JeKo-1 lymphoma cells. Quantitative cytokine analyses showed higher levels of IFN-γ, TNF-α, IL-2 production in T cells expressing CD19-CARs with CD28 costimulatory domain compared to the corresponding constructs with 4-1BB, consistent with prior work. T cells expressing each of the CD19-CARs proliferated in vitro after stimulation with K562/CD19 and Raji tumor cells by CFSE dye dilution, with the strongest proliferation observed in T cells expressing the CD19-CAR ‘long/CD28’, consistent with the highest levels of IL-2 production by T cells expressing this construct. We then analyzed the in vivo anti-tumor efficacy of each CD19-CAR in immunodeficient NOD/SCID/g−/− (NSG) mice engrafted with firefly luciferase transduced Raji cells. Tumor was inoculated on day 0, and once tumor was established (day 7), the mice received a single dose of 2.5×106̂ T cells expressing each CD19-CAR, a tEGFR control vector, or were left untreated. Surprisingly, only T cells expressing CD19-CARs with a short spacer domain (short/CD28 and short/4-1BB) eradicated the Raji tumors and led to long-term tumor-free survival of all mice. T cells expressing CD19-CARs with a long spacer domain (long/CD28 and long/4-1BB) did not confer a significant anti-tumor effect and all mice expired from systemic lymphoma at a similar time as control and untreated mice. The anti-tumor efficacy in vivo of T cells modified with long spacer CD19-CARs could not be improved by increasing CAR T cell dose 4 fold, or by including additional costimulatory domains into the CD19-CAR (long/CD28:4-1BB). Serial analyses in peripheral blood, bone marrow and spleen showed dramatically lower numbers of transferred T cells in mice treated with long spacer CD19-CARs compared to mice treated with short spacer CD19-CARs or control T cells. Further analysis revealed that despite strong activation in vivo as assessed by upregulation of CD69 and CD25, CD19-CARs with long extracellular spacer domain induced a high rate of activation induced T cell death in vivo. Collectively, these results demonstrate that the extracellular spacer domain that lacks intrinsic signaling function is critical in the design of effective CD19-CARs, and illustrates that tailoring spacer length is likely to be essential for designing effective CARs specific for other tumor antigens. Disclosures: No relevant conflicts of interest to declare.

Abstract 643 The engineering of T cells modified to express single-chain antibody-derived chimeric antigen receptors (CARs) that recognize surface molecules expressed on malignant B cells independent from HLA is an area of intense research. The successful use of CAR-modified T cells to treat B-cell malignancies requires the definition of T cells or T-cell subsets that possess an intrinsic ability to confer potent and durable anti-tumor responses, and would ideally be directed against an antigen with selective expression on malignant, but not on normal B cells. Current efforts have focused on the use of bulk CD8+ cytotoxic T cells (CTL) for CAR-modification because the role of CTL in mediating tumor rejection is well established, or unselected PBMC that contain both CD8+ and CD4+ T cells. However, without selection, the progeny and composition of the final cell product is poorly controlled, and the rational use of CD4+ CAR T cells that play a central role in sustaining CD8+ T cell responses to pathogens remains unexplored. We have shown that ROR1, an oncofetal tyrosine kinase receptor, is uniformly expressed on chronic lymphocytic leukemia (CLL) and mantle cell lymphoma (MCL), and developed a ROR1-CAR from an anti-ROR1 mAb that confers specific recognition of malignant, but not mature normal B cells when expressed in CD8+ T cells [Hudecek et al. 2010]. Here, we derived CD4+ ROR1-CAR T cells from unselected, and sort purified naïve (N), central (CM) and effector memory (EM) CD4+ T cells and characterized direct tumor recognition and the ability to augment CD8+ ROR1-CAR CTL. We purified CD4+ and CD8+ T cells from the peripheral blood of healthy donors (n=3) and CLL patients (n=3) and transduced them with ROR1-CAR encoding lentiviral vector. In all cases, we achieved higher transduction efficiencies of CD4+ compared to CD8+ T cells. CAR-modified T cells were enriched from the cultures using a truncated EGFR encoded within the CAR vector downstream of a T2A sequence as a selection marker. ROR1-CAR expression on the surface of T cells was confirmed by flow cytometry using recombinant ROR1-Fc fusion protein. In functional assays, CD4+ CAR T cells conferred specific but weak cytolytic activity compared with CD8+ CAR T cells against ROR1+ tumors including primary CLL, JeKo-1 MCL, and K562 cells transfected with ROR1. Multiplex cytokine analysis detected high-level production of Th1 cytokines including IFN-g, IL-2 and TNF-α, and CFSE staining showed superior proliferation after stimulation with tumor compared to CD8+ CAR CTL, suggesting CD4+ CAR T cells exert typical helper effector functions after stimulation through the CAR. Polyclonal CD4+ T cells contain various frequencies of N, CM, and EM subsets that have distinct functional properties. Thus, we compared the cytolytic activity, cytokine profile and proliferative capacity of CAR T cells derived from flow-sort purified CD4+ N, CM and EM subsets. CD4+ CAR T cells derived from the N CD45RA+CD45RO-CD62L+ subset exhibited the strongest cytotoxicity, produced the highest levels of Th1 cytokines, especially IL-2, and proliferated most vigorously in response to ROR1+ tumor cells, suggesting they may be able to confer the strongest direct anti-tumor effect, and be best suited to provide help to CD8+ CAR CTL. Indeed, in co-culture experiments, the addition of CAR-transduced, but not untransduced CD4+ T cells derived from N rather than memory subsets also resulted in maximum proliferation of CD8+ CAR CTL. Consistent with our analysis of cytokine production, blocking with anti-IL-2 and anti-IL-2Ra mAbs was able to impair but did not completely abrogate this helper effect, demonstrating that help provided by CD4+ CAR T cells was at least in part mediated by IL-2. In vivo experiments demonstrated profound anti-tumor efficacy of CD4+ ROR1-CAR modified T cells in NOD/SCIDgc−/− (NSG) mice that had been engrafted with firefly luciferase transduced JeKo-1 MCL tumor cells. Experiments to compare the anti-tumor efficacy of CAR-modified CD4+ T cells derived from individual N, CM and EM subsets, alone or in combination with CD8+ CAR CTLs are in progress. Collectively, our data illustrate how cell-intrinsic qualities can inform the rational design of cell products containing both tumor-specific CD8+ and CD4+ T cells to improve outcomes of cancer immunotherapy, and suggest that deriving engineered CD4+ T cells from the naïve subset may be most advantageous for tumor therapy. Disclosures: Maloney: GSK: Consultancy, Honoraria.

Meeting abstracts Immunotherapy with T cells expressing chimeric antigen receptors (CARs) specific for a tumor cell-surface molecule is effective for CD19+ B cell malignancies. There is interest in extending CAR-T cell therapy to epithelial tumors, which requires identifying molecules that can be