BACKGROUND

[001] Adoptive T cell therapy (ACT) utilizing ex vivo expanded autologous and allogeneic T cells is an attractive therapeutic approach for the treatment of viral infection, cancer and autoimmune disease. Methods that enable the rapid generation of large numbers of therapeutic T cells are critical to the potency and safety of ACT. Various T cell enrichment methods, including selection of defined T cell subsets, as well as expansion methods have been used for ACT. It is desirable to employ a T cell population that permits relatively high activity in vivo and relatively high proliferation potential.

SUMMARY

[002] Described herein is a method for preparing T cell populations useful for expressing chimeric antigen receptors ("CARs") for ACT. The T cell populations are also useful for a variety of other purposes requiring a highly active, long-lived T cell population. The T cell populations prepared by the methods described herein are enriched for: naive T cells (TN), memory stem cells (TSCM) and central memory T cells (TCM). Thus, such cell populations can be described as TCM/SCM/N cells or TcM/scM/N cell populations. These cell populations can be derived from peripheral blood mononuclear cells (PBMC) by both: 1) depleting unwanted cell populations such as CD 14 expressing myeloid cells and CD25 expressing cells; and 2) enriching for CD62L expressing memory and naive T cells. Thus, the resulting population of cells includes T naive (TN) and stem memory cells (TSCM) expressing CD45RA and CD62L. It also includes the population of central memory T cells (TCM) that express CD45RO and CD62L. TCM/SCM/N cell populations differ from previously described TCM cell populations in that their preparation does not entail depletion of CD45RA+ T cells. TCM/SCM/N cell populations, upon preparation, are relatively free of effector memory cells (TEM) and effector cells (TE). Of course, as a TCM/SCM/N cell population undergoes expansion, differentiation will occur, giving rise to, for example, TE cells. [003] Patient-specific, autologous and allogeneic TCM/SCM/N cells (or allogenic

TcM/scM/N cells) can be engineered to express a chimeric antigen receptor (CAR) or T cell receptor (TCR) and the engineered cells can be expanded and used in ACT.

[004] Described herein is an isolated population of human cells comprising T cells (i.e., cells that express CD3 or CD3+ cells), wherein the T cells comprise central memory T cells; memory stem T cells, and naive T cells, wherein greater than 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the T cells are CD45RA+ and greater than 70% (greater than 75%, 80%, 85% or 90%) of the T cells are CD62L+. In various embodiments: less than 15% (less than 12%, 10%, 8%, or 6%) of the T cells are CD 14+ and less than 5% (less than 4%, 3% or 2%) of the T cells are CD25+; greater than 10% (greater than 15%, 20%, 25%, 30%, 35%, or 40%) of the T cells harbor a recombinant nucleic acid molecule (e.g., a recombinant nucleic acid molecule such encoding a chimeric antigen receptor or T cell receptor; e.g., a viral vector such as a T cell vector); at least 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the T cells are CD4+ and CD62L+ or CD8+ and CD62L+; at least 10% (greater than 15%, 20%, 25%, 30%, 35%, or 40%) of the T cells are CD8+ and CD62L+; less than 60% (less than 55%, 50%, 45%, 40%, 35%, 30%, 24%, 20% or 15%) of the T cells are CD45RO+; and at least 80% (greater than 85%, 90%, 95%, or 98%) of the cells in the isolated population of cells comprising T cells are T cells.

[005] Also described herein is isolated population of cells comprising human T cells (i.e., cells that express CD3 or CD3+ cells), wherein the T cells comprise central memory T cells and memory stem T cells, wherein greater than 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the T cells are CD45RA+, greater than 40% (greater than 45%, 50%, 55%, 60%, 65%, 70%75%, 80%, 85% or 90%) of the T cells are CD62L+, greater than 70% (greater than 75%, 80%, 85% or 90%) of the T cells are CD95+, and greater than 10% (greater than 15%, 20%, 25%, 30%, 35%, or 40%) of the T cells harbor a recombinant nucleic acid molecule (e.g., a recombinant nucleic acid molecule (e.g., a viral vector) encoding, e.g., a chimeric antigen receptor or T cell receptor). In various embodiments: less than 15% (less than 12%, 10%, 8%, 6%) of the T cells are CD14+ and less than 5% (less than 4%, 3% or 2%) of the T cells are CD25+; at least 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the T cells are CD4+ and CD62L+ or CD8+ and CD62L+; at least 10% (greater than 15%, 20%, 25%, 30%, 35%, or 40%) of the T cells are CD8+ and CD62L+; less than 60% (less than 55%, 50%, 45%, 40%, 35%, 30%, 24%, 20% or 15%) of the T cells are

CD45RO+; and at least 80% (greater than 85%, 90%, 95%, or 98%) of the cells in the isolated population of cells comprising T cells are T cells.

[006] Described herein is a method for preparing a population of human cells comprising T cells (i.e., cells that express CD3 or CD3+ ceUs)harboring a recombinant nucleic acid molecule, comprising: (a) providing a sample of human cells comprising T cells, wherein the T cells comprise: central memory T cells;

memory stem T cells, and naive T cells, wherein greater than 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the T cells are CD45RA+ and greater than 70% (greater than 75%, 80%, 85% or 90%) of the T cells are CD62L+; (b) activating the population of human cells comprising T cells; and (c) transducing or transfecting cells in the population of human cells comprising T cells with a recombinant nucleic acid molecule to provide a population of human cells comprising T cells harboring a recombinant nucleic acid molecule, wherein the method does not comprise a step of depleting cells expressing CD45RA. In various embodiments: the recombinant nucleic acid molecule is a viral vector (e.g., a lentiviral vector, a viral vector encoding a CAR, a viral vector encoding a T cell receptor); the method further comprises culturing the population of human cells comprising T cells harboring a recombinant nucleic acid molecule; the culturing step comprises the addition of exogenous IL-2 and exogenous IL-15; and the activating step comprises exposing the cells to an anti- CD3 antibody and an anti-CD28 antibody; and at least 80% (greater than 85%, 90%, 95%, or 98%) of the cells in the isolated population of cells comprising T cells are T cells.

[007] Described herein is method for preparing a population of human cells comprising T cells (i.e., cells that express CD3 or CD3+ cells), wherein the T cells comprise central memory T cells; memory stem T cells, and naive T cells, wherein greater man 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the cells are CD45RA+ and greater than 70% (greater than 75%, 80%, 85% or 90%) are CD62L+, comprising: (a) providing an isolated population of human cells comprising T cells; (b) treating the isolated population of human cells comprising T cells to deplete cells expressing CD25 and cells expressing CD 14 to prepare a depleted cell population; and (c) treating the depleted cell population to enrich for cells expressing CD62L, thereby preparing a population of human cells comprising T cells, wherein the T cells comprise central memory T cells; memory stem T cells, and naive T cells, wherein greater than 40% of the cells are CD45RA+ (greater than 45%, 50%, 55%, 60%, 65% or 70%) and greater than 70% are CD62L+, wherein the method does not comprise a step of depleting cells expressing CD45RA. In various embodiments: the method further comprises activating the population of human cells comprising T cells and transducing or transfecting the activated cells with a recombinant nucleic acid molecule to provide a population of T cells comprising T cells harboring a recombinant nucleic acid molecule; isolated population of human cells comprising T cells comprises PBMC; and at least 80% (greater than 85%, 90%, 95%, or 98%) of the cells in the isolated population of cells comprising T cells are T cells.

[008] Also described herein is a population of human cells comprising T cells (i.e., cells that express CD3 or CD3+ cells), wherein the T cells comprise central memory T cells; memory stem T cells, and naive T cells, wherein greater than 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the cells are CD45RA+ and greater than 70% (greater than 75%, 80%, 85% or 90%) are CD62L+, wherein the population is prepared by a method comprising: providing an isolated population of human cells comprising T cells (e.g. PBMC from a donor); treating the isolated population of human cells comprising T cells to deplete cells expressing CD25 and deplete cells expressing CD 14 to prepare a depleted cell population; and treating the depleted cell population to enrich for cells expressing CD62L, thereby preparing a population of human cells comprising T cells, wherein the T cells comprise central memory T cells; memory stem T cells, and naive T cells, wherein greater than 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the cells are CD45RA+ and greater than 70% (greater than 75%, 80%, 85% or 90%) are CD62L+, wherein the method does not comprise a step of depleting cells expressing CD45RA. In various embodiments: less than 15% (less than 12%, 10%, 8%, 6%) of the T cells in the population of human cells are CD14+ and less than 5% (less than 4%, 3% or 2%) of the T cells are CD25+; at least 40% (greater than 45%, 50%, 55%, 60%, 65% or 70%) of the T cells are CD4+ and CD62L+ or CD8+ and CD62L+; at least 10% (greater than 15%, 20%, 25%, 30%, 35%, or 40%) of the T cells are CD8+ and CD62L+; less than 60% (less than 55%, 50%, 45%, 40%, 35%, 30%, 24%, 20% or 15%) of the T cells are

CD45RO+

[009] Also described herein is a method of treating cancer, autoimmunity or infection comprising administering to a patient in need thereof a pharmaceutical composition comprising a human cell population described herein. In some cases the cells are autologous to the patient being treated and in some cases they are allergenic to the patient being treated.

[0010] A cell population can be depleted for cells expressing a particular marker (e.g., receptor) by removing (using for example a selective antibody and cells sorting methods) some, most or nearly all of the cells expressing the marker from the population of cells so that the relative proportion of the cells in the population expressing the marker is reduced. A cell population can be enriched for cells expressing a particular marker (e.g., receptor) by collecting (using for example a selective antibody and cells sorting methods) some, most or nearly all of the cells expressing the marker from the population of cells and discarding cells not expressing the marker so that the relative proportion of the cells in the population expressing the marker is reduced.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] FIGURE 1 depicts certain marker expression patterns for various T cell subsets, particularly expression patterns that distinguish TSCM form TN.

[0012] FIGURE 2 depicts certain marker expression data for various T cell subsets.

[0013] FIGURE 3 depicts the relationships between certain T cell subsets based on expression of CD62L and CD45RO.

[0014] FIGURE 4 depicts the results of surface phenotype analysis of cells after

CliniMACS™ Enrichment for TCM or TCM/SCM/N cells. Cells from the indicated products were stained with fiuorochrome-conjugated antibodies specific for the T cell markers CD3 and CD8, the central memory T cell marker CD62L, and the naive T cell marker CD45RA ll th t k CD 14 d l tory T cell marker CD25. Percentages of immunoreactive cells (shaded histograms) above isotype control staining (black line) are indicated in each histogram.

[0015] FIGURE 5 depicts the results of a study of expansion of TCM and TCM/SCM/N cell populations. Viable cell numbers starting at the day of lenti viral transduction are depicted for cell products manufactured from 4 of the 5 donors, with TcM-derived products represented by black circles, and TcM/scM/N-derived products represented by white squares. Log 2 scales were used for the Y-axes to better depict the doubling time of cells in culture.

[0016] FIGURE 6 depicts the results of an analysis of surface phenotype of TCM and TCM/SCM/N cell populations. TCM- or TcM/scM/N-enriched cells were bead stimulated, transduced with CD19R(EQ)28Z-T2A-EGFRt_epHIV7 or IL13(EQ)BBZ-T2A- CD19t_epHIV7, and expanded ex vivo. Cells were then analyzed by flow cytometry after being stained with fluorochrome-conjugatcd reagents specific for the EGFRt or CD19t transduction markers, or for the T cell markers CD3, CD4 and CD8, the central memory T cell markers CD27, CD28 and CD62L, and the naive T cell marker CD45RA (grey histograms). Percentages of immunoreactive cells above control staining (open histograms) are indicated in each histogram of a representative TCM/ TCM/SCM/N pair (A) or are indicated as data points for each cell line in (B) with open symbols for the TcM-derived products and grey symbols for the TcM/soi/N-derived products. Red bars depict the mean for each group, and p-values for paired Student's t-test comparisons of TCM- vs. TcM/scM/N-derived cell products are indicated for each surface marker. *, P-value was not calculable for CD27 because too-few valued. Note that only shading, but not cell-line specific symbols, are depicted for CD3 staining in (B) due to overlap.

[0017] FIGURES 7 A-C depict the results of a study assessing the therapeutic efficacy of CD 19R TCM or CD 19R TCM/SCM/N pre-treatment when challenged with Sup-B15 tumor cells. One day after irradiating the NSG mice, they were divided into groups (n=4) and left untreated, or treated i.v. with either 2.5x10 ⁶ mock-transduced PBMC (no CAR), CD19R TCM or CD19R TCM/SCM/N (approximately 0.5x10 ⁶ CAR+ cells). On day 2, mice were challenged i.v. with 0.7-l.OxlO ⁶ with ffLuc+ Sup-B15 tumor cells. M, million (FIGURE 7A). Representative mice from each group showing relative tumor burden using Xenogen Living Image (FIGURE 7B). Average Sup-B 15 tumor growth in each group of mice over time as monitored by Xenogen imaging and quantification of ffLuc flux (photons/sec) (FIGURE 7C).

[0018] FIGURES 8 A-B depict the results of study examining the presence of human T cells in the blood 41 days after i.v. pre-treatment with CD19R TcMor CD19R

TcM/scM/N. One day after irradiating the NSG mice, they were divided into groups (n=4) and left untreated, or treated i.v. with either 2.5x10 ⁶ mock-transduced PBMC (no CAR), CD19R TCM or CD19R TCM/SCM/N (approximately 0.5x10 ⁶ CAR+ cells). On day 2, mice were challenged i.v. with 0.7-l.OxlO ⁶ with tumor GFP+ Sup-B15 tumor cells. At day 41, the blood of each mouse was evaluated for the presence of human CD45+CD3+ cells (i.e., human T cells) by flow cytometry. Representative flow cytometric analyses of each mouse are depicted (Mouse 1 and Mouse 2 are depicted in FIGURE 8A; Mouse 3 and Mouse 4 are depicted in FIGURE 8B), where huCD45-gated cells were further analyzed for expression of the CD3 T cell marker and the EGFRt transduction marker (% CAR+ based on upper right quadrant). Gold boxes are used to highlight the GFP+ SupB15 population.

[0019] FIGURE 9 depicts the results of long term killing assays performed with TCM- cells or TCM/SCM/N cells that had been mock-transduced or transduced with

CD19R(EQ)28ZEGFRt_epHlV7 as effectors and either CD19-negative K562 cells (white bars) or CD 19+ SupB15 cells (black bars) as targets. Percentages of viable target cells (CD45-negative, FSC high) remaining after 72 hours compared to mock effector co-cultures that were normalized to 100% are depicted.

[0020] FIGURE 10 depicts the results of five hour degranulation assays performed with TcM-cells or TCM/SCM/N cells that had been mock-transduced or transduced with CD19R(EQ)28ZEGFRt_epHlV7 as effectors and either CD19-negative K562 cells (white bars) or CD19+ SupB15 cells (black bars) as stimulators. Percentages of CD45/CD8/EGFRt-gated cells that were immunoreactve for CD 107a are indicated.

[0021] FIGURE 11 depicts the results analysis of IFNy production by TCM cells or TCM/SCM/N cells that had been mock-transduced or transduced with

CD19R(EQ)28ZEGFRt_epH_V7 Five hour co-cultures were performed with TCM- derived orTcM/scM/N-dcrived cell products that had been mock-transduced or transduced with CD19R(EQ)28ZEGFRt_epHIV7 as effectors and either CD19- negative K562 cells (white bars) or CD19+ SupB15 cells (black bars) as stimulators. Percentages of CD45/CD8/EGFRt-gated cells that were immunoreactive for IFN-7 are indicated.

[0022] FIGURE 12 depicts the results of a study of efficacy of IL13 CAR expressed by various T cell populations.

DETAILED DESCRIPTION

[0023] The T cell compartment includes T cell subsets that are at different stages of differentiation. These subsets arise from differentiation of Naive T cells (TN), which are CD45RA+, CD62L+, CD28+, and CD95-. Among the stem cell-like subsets are Memory Stem Cells (TSCM), which are CD45RA+, CD62L+, CD28+, and CD95+. These cells differentiate into Central Memory Cells (TCM), which are CD45RO+, CD62L+, CD28+, and CD95+. TCM differentiate in Effector Memory Cells (TEM), which are CD45RO+, CD62L-, CD28+/-, and CD95+. The TEM differentiate to Effector T cells (TE) which are CD45RO+, CD62L+, CD28+, and CD95+.

[0024] Memory Stem T Cells (TSCM) are present at a low level in the T cell compartment, but appear to have significant self-renewal and proliferative potential. While they resemble naive T cells (TN) in that they express CD45RA+ and CD62L+, they can be distinguished from TN by their expression of CD95 (FIGURE 1). TSCM can be generated from TN by stimulation with CD3/CD28 beads in the presence of IL- 7 and IL-15. They also can be expanded in the presence of Wnt/p-catenin pathway activation.

[0025] Central Memory T Cells (TCM), which are more abundant in PBMC man are

TSCM, are a well-defined memory T cell subset with high self-renewal and proliferative potential. There is evidence that TCM persist following adoptive transfer better than Effector T cells (TE) (Berger et al., 2008 Journal of Cellular Immunology

118:4817; Wang et al, 2011 Blood 117:1888). TCM can be enriched from PBMC for

T cell therapy manufacturing based on their CD45RA- CD45RO+ CD62L+ phenotype (FIGURE 2) Wang X et al. J Immunotherapy 2012;35:689. There is some evidence that TCM behave as adult stem cells. Studies in mice demonstrated that: single cell t f f T th ti d t t d th t T can provide full immune reconstitution; that TCM expand to produce more TCM; and that TCM differentiate to TEM/TE (Graef et al. 2014 Immunity 41: 116; Gattioni et al. 2014 Immunity 41:7).

[0026] The TCM/SCM/N cell population described below incudes TCM, which have proven efficacy in CAR-T therapy, and TN as well as TSCM, both of which are more stem cell-like than TCM. FIGURE 3 presents expression data depicting a relationship between the various cell populations and PBMC.

[0027] The cell populations described herein can be genetically engineered to express, for example, a CAR or a T cell receptor. A CAR is a recombinant biomolecule that contains an extracellular recognition domain, a transmembrane region, and one or more intracellular signaling domain. The term "antigen," therefore, is not limited to molecules that bind antibodies, but to any molecule that can bind specifically to any receptor. "Antigen" thus refers to the recognition domain of the CAR The extracellular recognition domain (also referred to as the extracellular domain or simply by the recognition element which it contains) comprises a recognition element that specifically binds to a molecule present on the cell surface of a target cell. The transmembrane region anchors the CAR in the membrane. The intracellular signaling domain comprises the signaling domain from the zeta chain of the human CD3 complex and optionally comprises one or more co-stimulatory signaling domains. CARs can both to bind antigen and transduce T cell activation, independent of MHC restriction. Thus, CARs are "universal" immunoreceptors which can treat a population of patients with antigen-positive tumors irrespective of their HLA genotype. Adoptive immunotherapy using T lymphocytes that express a tumor-specific CAR can be a powerful therapeutic strategy for the treatment of cancer.

[0028] The CAR can be produced by any means known in the art, though preferably it is produced using recombinant DNA techniques. Nucleic acids encoding the several regions of the chimeric receptor can be prepared and assembled into a complete coding sequence by standard techniques of molecular cloning known in the art (genomic library screening, overlapping PCR, primer-assisted ligation, site-directed mutagenesis, etc.) as is convenient. The resulting coding region is preferably inserted into an expression vector and used to transform a suitable expression host cell line, preferably a T lymphocyte cell line, and most preferably an autologous T lymphocyte cell line.

Various CAR suitable for expression by TCM/SCM/N cells include, for example, those described in: WO 2016/044811; WO 2104/144622; WO 2002/077029; and

WO/US2014/0288961.

Example 1: Preparation of TCM/SCM/N cells

[0029] A variety of methods can be used to produce a population of human TCM/SCM/N cells. For example, a population of TCM/SCM/N cells can be prepared from a mixed population T lymphocytes. The population of T lymphocytes can be allogenic to or autologous to the subject ultimately treated using the cells and can be obtained from a subject by leukopheresis or blood draw.

[0030] The following method is an example of one that can be used to obtain a population of TCM/SCM/N cells from T lymphocytes obtained by leukapheresis or other means. Peripheral blood is collected by leukapheresis or peripheral blood draw. Day 1 of a typical manufacturing cycle is the day the ficoll procedure takes place. The subject's leukapheresis product is diluted with EDT A/PBS and the product is centrifuged at 1200 RPM for 10 minutes at room temperature with maximum brake. After centrifugation, the platelet-rich supernatant is removed and the cell pellet is gently vortexed. EDTA/PBS is used to re-suspend the vortexed cell pellets in each conical tube. Each tube is then underlayed with ficoll and centrifuged at 2000 RPM for 20 minutes with no brake at room temperature. Following centrifugation, the PBMC layer from each tube is transferred into another conical tube. The cells are centrifuged at 1800 RPM for 15 minutes with maximum brake at 4°C.

[0031] After centrifugation, the cell-free supernatant is discarded and the cell pellet is gently vortexed. The cells are washed twice using EDTA/PBS each time, and a third time using PBS. Cells are centrifuged each time at 1200 RPM for 10 minutes with maximum brake at 4°C. After the final PBS wash, the vortexed cell pellet is resuspended in complete X-VIVO 15 media (X-VTVO™ media with 10% FBS) and transferred to a transfer bag. The bag with washed PBMC is kept overnight on a rotator at room temperature on the bench top for immunomagnetic selection the next day.

[0032] Next, selection procedures are used to both to deplete the cell population of cells expressing certain markers and to enrich the cell population for cells expressing certain other markers. These selection steps preferably occur on day two of the manufacturing cycle. The cell population is substantially depleted for cells expressing CD25 and CD 14. Importantly, the cell population is not substantially depleted for cells expressing CD45RA. Briefly, cells resuspended in labeling buffer (LB;

EDTA/PBS with 0.5%HSA), and incubated with anti-CD14 and anti-CD25 Miltenyi antibodies for CliniMACS® depletion, and the composition is gently mixed and then incubated for 30 minutes on a rotator at room temperature on the bench top.

[0033] The depletion step is performed on a CliniMACS® device using a depletion tubing set. The recovered cells following the depletion step are transferred into tubes and centrifuged at 1400 RPM for 15 minutes with maximum brake at 4°C.

[0034] The cell-free supernatant is removed and the cell pellet is gently vortexed and resuspended. To enrich for cells expressing CD62L, the cell suspension is treated with anti-CD62L-biotin (made at the City of Hope Center for Biomedicine and Genetics), gently mixed and incubated for 30 minutes on a rotator at room temperature on the bench top.

[0035] Following the incubation period, LB is added to the tube and cells are centrifuged at 1400 RPM for 15 minutes at maximum brake at 4°C. The cell-free supernatant is removed and the cell pellet is gently vortexed. LB is added to re suspend the cell pellet in the tube and the resuspended cells are transferred to a new transfer bag. Anti-biotin (Miltenyi Biotec) reagent is added and the mixture is gently mixed and incubated for 30 minutes on a rotator at room temperature on the bench top.

[0036] The CD62L enrichment step is performed on a CliniMACS® device using a tubing set. The product of this enrichment can be frozen for storage and later thawed and activated [0037] To provide an intermediate holding step in the manufacturing, the option exists to freeze cells following the selection process. The cells are pelleted by centrifugation at 1400 RPM for 15 minutes with max break at 4°C. The cells are re suspended in Cryostor® and aliquoted into cry o vials. The vials will transferred to a controlled cooling device that can cool at about 1°C /minute (e.g., a Nalgene® Mr. Frosty; Sigma-Aldrich) the cooling device is immediately transferred to a -80°C freezer. After three days in the -80°C freezer, the cells are transferred into a GMP LN2 freezer for storage.

[0038] We have found that cryopreserved cells exhibit good recovery and viability, maintain the appropriate cell surface phenotype when thawed up to 8.5 months after cryopreservation, and can be successfully transduced and expanded in vitro upon thawing.

[0039] Alternatively, freshly enriched Tcm/scm/n cells can be activated, transduced or expanded as described in Example 3, below.

[0040] Example 2: Comparison of TCM and TCM/SCM/N Yield and Recovery

[0041] A study was conducted to evaluate the yield of human peripheral blood mononuclear cells (PBMC) that had undergone CliniMACS/AutoMACS™ enrichment of the following T cell populations: 1) CD62L+ central memory (TCM) cells, and 2) CD62L+ TCM/SCM/N cells, which includes both the central memory (TCM) and stem cell memory (TSCM) populations along with naive T cells (TN). Evaluations were performed using Guava analysis of viable cell numbers and flow cytometric analysis of phenotype.

[0042] Experimental Design: TCM and TCM/SCM/N cell lines were enriched from PBMC collected from healthy human donors. Briefly, TCM/SCM/N cells were produced by CliniMACS/AutoMACS™ depletion of the CD14+ monocytes and the CD25+ regulatory T cells, followed by AutoMACS™ selection of the CD62L+ memory population. In contrast, TCM cells were produced by CliniMACS/AutoMACS™ depletion of the CD 14+ monocytes, the CD25+ regulatory T cells, and the CD45RA+ naive and stem cell memory T cells, followed by AutoMACS™ selection of the CD62L+ population. Evaluation of the selection process included viable cell number enumeration, as well as flow cytometric analysis, described in greater detail below. The cells were then cryopreserved for future studies.

[0043] Overview of Enrichment Process: The blood products were ficolled, and the resulting PBMC underwent a series of washes in PBS/EDTA. PBMC were then resuspended in complete X-Vivo 15 media and, in some cases were transferred to a 300cc transfer bag that was then placed on a 3-D rotator overnight. The PBMC then underwent sequential rounds of CliniMACS/AutoMACS™ depletion and selection to enrich for either the CD147CD257CD45RA7CD62L ⁺ TCM or CD147CD257CD62L ⁺ TCM/SCM/N populations. The first step involved magnetic depletion (via CliniMACS™ or AutoMACS™) to remove the CD 14+ monocytes, the CD25+ regulatory T cells, and, for the TCM, the CD45RA+ naive T cells. The remaining cells then underwent a positive selection (via AutoMACS™) for the CD62L+ population using the Anti- CD62L-Biotin (Dreg 56) reagent. Final cell counts after both rounds of enrichment were recorded. Samples of the final selected cell population were then analyzed by flow cytometry. [0044] Overview of Flow Cytometric Assay and Analysis: Flow cytometric analysis of cell populations was conducted as follows. Samples were washed in FACS Stain Solution (FSS) using atabletop centrifuge, resuspended in FSS and 100 uL per sample was aliquoted into pre-labeled 12x75 mm FACS tubes (1 tube per condition). The required volume of antibody was added to their respective FACS tubes and tubes were then incubated for 30 minutes in the dark at 4°C. At the end of incubation each tube was washed twice in FSS and resuspended in either 250 μΐ FSS, or FSS containing DAPI as a viability dye. Samples were then tun and analyzed on a FACS Calibur (Becton Dickenson) or MACSQuant (Miltenyi) instrument. Percentages of viable immunoreactive cells were calculated via FCS Express software (De Novo Software, Los Angeles, CA).

[0045] Post-enrichment Yield: Recovery of CD3+ T cells from each of the cell products are depicted in Table 1, below. Overall, enrichment of the CD147CD25 ^" /CD45RA7CD62L ^* CD3 ⁺ TCM resulted in a 2-6% recovery, while enrichment of the CD 147CD257CD62L ⁺ CD3 ⁺ TCM/SCM/N resulted in a 4-30% recovery. Indeed, when comparing each matched pair, there was a 1.6- to 15-fold higher recover}' of TCM/SCM/N cells vs. TCM cells.

[0046] Post-Enrichment Flow Cytometric Analysis: After both rounds of CliniMACS™ enrichment, the phenotype of the enriched cells was determined by FACS analysis. Results of cell products derived from three representative donors are shown in FIGURE 4. The TCM and TCM/SCM/N populations from each donor exhibited enriched levels of CD3+ and CD62L+ cells, with the TCM being depleted of CD45RA while the TCM/SCM/N expressed CD45RA as expected. Depletion of the CD 14+ and CD25+ populations was also observed as expected with both selection strategies. Interestingly, percentages of CD8+ cells were often higher in the TCM/SCM/N population when compared to its donor-matched TCM population (FIGURE 4 and Table 1).

[0047] Results: The results of this study demonstrate that clinically translatable GMP manufacturing methods can be used to efficiently isolate both CD62L+ TCM as well as TCM/SCM/N cell population. Starting from PBMC 2-6% of the starting population is recovered as CD3+ TCM. Eliminating the CD45RA-depletion step allows recovery of 4-30% of the starting population as CD3+ TCM/SCM/N cells. The ability to recover 1.6- to 15-fold more TCM/SCM/N cells as compared to TCM cells would increase the number of T cells that can be used as a starting population for genetic engineering. Further, we observed that elimination of the CD45RA depletion step also appeared to increase the percentage of CD8 T cells in the enriched product.

Example 3: Lentiviral Transduction of TCM/SCM/N cells

[0048] A TCM/SCM/N cell population can be genetically engineered to express protein or proteins of interest. For example, the cells can be transduced with a lentivirus capable of expressing a protein or proteins of interests. The following is an example of one method that can be used to transduce the T cells with a lentivirus vector with protamine sulfate/cytokine solution.

[0049] If the cells are frozen subsequent to the selection process, the cells will need to be thawed upon continuation of the manufacturing process. The cryopreserved vials are thawed in a 37°C water bath until mostly liquid remains. The contents of the vials are placed in a conical tube containing cold X-Vivo 15 media. The cells are centrifuged at 1200 RPM for 10 minutes with max brake at 4°C. After centrifugation, the cell-free supernatant is decanted, the pellets ate vortexed, and all the pellets combined into one tube in X-Vivo 15 media to be transduced with lentivirus. The cell numbers are determined and the manufacturing process continues with Dynal bead activation.

[0050] The cells are centrifuged at 1400 RPM for 10 minutes at 25°C. After centrifugation, the cell-free supernatant is removed and the cell pellet is gently vortexed and resuspended in complete X Vivo 15 media to be transduced with lentivirus. If thawed cells are being used, the centrifugation step is not necessary.

[0051] Tcm/scm/n cells are stimulated with Dynabeads® Human T expander CD3/CD28 (Invitrogen) at a 1 :3 ratio (T cell:bead), then cultured in the presence of 50 U/mL rhIL -2 (Chiron) and 0.5 ng/ml rhIL-15 (CellGenix).

[0052] After a one to three day incubation, T cells are transduced with the lenti viral vector at a MOI typically ranging from 0.1 to 1.5 in X Vivol5 containing 10% FCS with 5μg/mL protamine sulfate (APP Pharmaceutical), 50 U/mL rhIL-2 and 0.5 ng/mL rhIL-15. The day following the lentivirus transduction, one volume of media is added of freshly prepared media/cytokine mastermix containing complete X-VIVO 15 media and cytokines (10 uL rhuIL-2 and 0.5 uL rhuIL-15).

[0053] Cultures are then maintained at 37°C, 5% CO ₂ in X-Vivo15 10% FCS as required to keep cell density between 3 * 10 ⁵ and 2x10 ⁶ viable cells/mL, with cytokine supplementation (final concentration of 50 U/mL rhlL-2 and 0.5 ng/mL rhIL-15) every Monday, Wednesday and Friday of culture.

[0054] CD3/CD28 Dynabead bead removal takes between days 7 to 9 after bead stimulation. The Dynal beads are removed using the DynaMag-50 or MPC magnet. After bead removal, a sample of the cell suspension is taken for counts. Based on the counts, the concentration is adjusted to approximately O.SxlOfymL.

[0055] Cultures are propagated until sufficient numbers of cells are achieved, which is typically between 8 and 18 days. During this time, a full replacement of cytokines are added to the T cell cultures. Starting on day 12 and every Monday, Wednesday and Friday thereafter, counts are taken to maintain a concentration of approximately 0.5x10 ⁶ cells/mL [0056] Final harvest and cryopreservation proceeds as follows. The cell product is cryopreserved once the expanded cells reach the necessary number of viable cells required for clinical administration or research purposes. The cells are counted and in- culture samples are harvested for mycoplasma testing. The remaining cells are centrifuged at 1800 RPM for 20 minutes, maximum brake at 4°C. The cell-free supernatant is removed and all cell pellets are gently vortexed, resuspended and combined into Isolyte buffer (Braun). The tube is centrifuged at 1800 RPM for 20 minutes, max brake at 4°C. The cells are resuspended in Isolyte for a second wash.

[0057] After the second Isolyte wash, the cell-free supernatant is removed and the cell pellet is gently vortexed and resuspended in an appropriate volume of the

cryopreservation media, CryoStor® 5 (BioLife Solutions), in order to have cells at an appropriate concentration. Cells are cyropresserved by placing in a controlled cooling device that can cool at about 1°C /minute (e.g., a Nalgene® Mr. Frosty; Sigma- Aldrich) or a control rate freezer system (Custom Biogenics). At completion of the freeze procedure, the cassette(s)/bag(s) and cryovials are immediately transferred into a LN2 freezer for storage.

Example 4: Comparison of TCM and TCM/SCM/N Cells That Have Been Bead- Stimulated, Lentivirally Transduced and Expanded

[0058] A study was conducted to evaluate the ex vivo expansion and cell surface phenotype of human peripheral blood mononuclear cells (PBMC) that had undergone CliniMACS/AutoMACS™ enrichment for either the CD62L+ TCM population, or the CD62L+ TCM/SCM/N population (which includes naive, central memory, and stem cell memory T cells), followed by bead stimulation and lentiviral transduction with either the CD19R(EQ)28Z-T2A-EGFRt_epHIV7, a lentivirus expressing a CAR targeted to CD19R or IL13(EQ)BBZ-T2A-CD19t_epHlV7, lentiviral vector targeted to IL- 13Ralpla (see WO 2016/044811 and WO/2002/077029 for details of the vectors and expressed CARs), using methods proposed for clinical use. Expansion was evaluated by monitoring of viable cell numbers and cell surface phenotype was evaluated by flow cytometric analysis.

[0059] Experimental Design: Chimeric antigen receptor (CAR) T cell products originating from the PBMC of three separate human donors were generated using methodologies expected to be suitable for clinical use. Briefly, TCM and TCM/SCM/N cells that had been enriched from PBMC (essentially as described above in Example 2) were activated by CD3/CD28-bead stimulation and transduced with either CD19R(EQ)28Z-EGFRt_epHIV7 or IL13(EQ)BBZ-T2A-CD19t_epHIV7 lentivirus. The CD3/CD28 beads were removed after 7-9 days, and T cell cultures were then expanded in accordance and maintained for up to 29 days with addition of complete X-VIVO 15 media as required by cell expansion (keeping cell density between OJxlO ⁶ and 2x10* viable cells/mL). Cytokines were supplemented every Monday, Wednesday and Friday of culture. Viable cell numbers were monitored during this expansion process. Final cell products were then evaluated for the surface markers traditionally used for cell product release, e.g., CD3 and either truncated EGFR (EGFRt) or truncated CD 19 (CD19t) as a transduction marker, as well as the T cell markers CD4, CDS, CD27, CD28, CD45RA and CD62L.

[0060] Overview of Stimulation. Transduction and Expansion: Enriched TCM or TCM/SCM/N cells were CD3/CD28-bead activated and lentivirally transduced with GMP-grade CD19R(EQ)28Z-T2A-EGFRl_epHIV7 or either GMP-grade or research- grade IL13(EQ)BBZ-T2A-CD19t_epHIV7 at a multiplicity of infection (MOI) between 0.3 and 3.0. The cells were put into culture in complete X-Vivo 15 media with rhIL-2 at 50 U/mL and rhIL-15 at 0.5 ng/mL. Cultures were then maintained with addition of complete X-Vivol5 media as required to keep the cell density between 0.3x10 ⁶ and 2x10 ⁶ viable cells/mL, and with cytokine supplementation (rhlL- 2 and rhIL-15) every Monday, Wednesday and Friday of culture. After seven to nine days of bead stimulation, the CD3/CD28 Dynabeads were removed from cultures using the DynaMag-50 magnet. Viable cell numbers were monitored during this expansion process, and viable cell counts during culture and on the final day of cryopreservation were recorded. The various products are listed in Table 2.

[0061] Flow Cytometry: Samples were washed in FACS Stain Solution (FSS) using a tabletop centrifuge, resuspended in FSS and 100 uL per sample was aliquoted into pre-labeled 12x75 mm FACS tubes (1 tube per condition). The required volume of antibody was added to their respective FACS tubes and tubes were then incubated for 30 minutes in the dark at 4°C. In some cases, samples were washed and then stained with fluorochrome-conjugated straptavidin as a secondary reagent. At the end of incubation each tube was washed twice in FSS and resuspended in either 250 μΐ FSS, or in 140 uL FSS and 70 uL working dilution of DAPI (made by mixing 6.9 μί, of the stock with 25 mL of FSS). Samples were then run and analyzed on a FACS Calibur (Becton Dickenson) or MACSQuant (Miltenyi) instrument. Percentages of immunoreactive cells were calculated via FCS Express software (De Novo Software, Los Angeles, CA).

[0062] Results of Viable Cell Number Analysis: After bead stimulation and lentiviral transduction with CD19R(EQ)28Z-EGFRt_epHIV7, the cell products were expanded under conditions that support T cell growth. Growth curves are depicted in FIGURE 5, and final cell numbers and fold expansion based on the total viable cells of each of the products are depicted in TABLE 3. Overall, fold expansion of TcM/scM/N-derived cell products was similar to that of their TcM-derived counterparts.

[0063] Final Product Flow Cytometric Analysis: Flow cytometric analyses of each of the cell products following ex vivo expansion revealed that all of the final cyropreserved cell products would have passed our traditional product specifications of > 80% CD3 for identity, and > 10% EGFRt or CD19t for potency (FIGURE 6A). Indeed, the TCM/SCM/N -derived products from each donor were very similar to their corresponding TcM-derived products in immunoreactivity for both CD3 and

EGFRt/CD19t.

[0064] Flow cytometric analysis of other T cell markers (FIGURE 6A) revealed that while all cell products have CD4 and CD8 T cell subsets, the TcM/scM/N-derived CAR T cell lines had a relatively higher proportion of CD8 cells compared to their TCM- derived counterparts. Both TcM-derived cell products and TcM/scM/N-derived cell products also expressed memory-associated markers CD27, CD28 and CD62L.

However, CD27 expression was generally higher on the TcM/scM/N-derived products. Furthermore, CD45RA expression levels were also significantly higher on the TCM/SCM/N -derived products.

[0065] Conclusion: Together, these data obtained indicate that both TCM- and

TcM/scM/N-enriched populations can be CD3/CD28 bead stimulated, lentivirally transduced, and expanded in vitro. Since fold expansion was similar between the TCM- and TcM/scwN-derived counterparts, it is important to note that the ability to start with greater cell numbers of TCM/SCM/N may allow for a shorter expansion time to reach sufficient cell numbers for clinical use.

[0066] Flow cytometric analysis revealed that both TCM and TCM/SCM/N cell populations were transduced with similar efficiency to express the EGFRt or CD19t transgenes. Flow cytometric analysis also confirmed that both TCM and TCM/SCM/N cell populations express the T cell markers CD3, CD4, CD8, as well as T cell memory markers CD27, CD28 and CD62L. The higher percentages of CD45RA+ cells observed in TCM/SCM/N populations were as expected, since TCM cells were depleted of CD45RA in the selection process.

Example 5: In Vtvo Efficacy of Intravenously Delivered TCM and TCM/SCM/N Cell Population Transduced with the CD19R(EQ)28Z-T2A-EGFRt_epHIV7

[0067] A study was conducted to evaluate the in vtvo efficacy of unselected PBMC, CD62L+ TCM cells and CD62L+ TCM/SCM/N cells transduced with lentiviral vector expressing a CAR targeted to CD19 (CD19R(EQ)28Z-T2A-EGFRt_epHIV7 lentiviral vector) and expanded using methods suitable for clinical use. In vivo anti-tumor efficacy of these cells administered intravenously (i.v.) was examined in

immunodeficient NSG mice using the CD 19-expressing SUP-B15 human acute lymphocytic leukemia cell line, which has been engineered to express the green fluorescence protein (GFP) and firefly luciferase (ffLuc) reporter genes (PMID: 22407828).

[0068] Experimental Design: TcM-derived or TCM/SCM/N -derived T cell lines that had been lenti-transduced with CD19R(EQ)28Z-T2A-EGFRt_epHIV7 and expanded for 21 days were used in this study (see above, Examples 2 and 4). Freshly thawed CAR T cells administered i.v. were then evaluated for their ability to control the in vivo growth of i.v. engrafted Sup-B15 cells. Both tumor burden, as measured by Xenogen imaging, and T cell persistence, as measured by flow cytometric analysis of peripheral blood, were examined.

[0069] Overview of In Vivo Xenograft Studies: Female NSG mice (10-12 weeks old) were irradiated with 300 rads at day 0. On day 1 mice were divided into groups (n=4) and left untreated, or treated i.v. (i.e., via the tail vein) with either 2.5x10 ⁶ HD270- derived mock-transduced PBMC (no CAR), HD270 CD19R TPBMC, HD270 CD19R TCM, or HD270 CD19R TCM/SCM/N. This translated to an average of 0.5x10 ⁶ CAR+ ceUs offCD 19R/EGFRt+ TPBMC, CD19R/EGFRt+ TCM or 0.625x10 ⁶ CD19R/EGFRt+ TCM/SCM/N being administered to each mouse as defined by the phenotypic analysis of the HD270 CD19R TCM- and TCM/SCM/N derived lines. On day 2 mice were challenged i.v. with 2 tumor cells per 1 CAR+ T cell GFP:ffLuc+ Sup-B15 tumor cells. Growth of this human acute lymphobastic leukemia cell line was then monitored overtime by Xenogen imaging and quantification of fire-fly luciferase (ffLuc) flux (photons/sec). At day 41, retro-orbital bleeding was performed, and the blood was evaluated for the presence of human CD3-exprcssing T cells by flow cytometric analysis, using a MACSQuant instrument and FCS Express software (De Novo Software, Los Angeles, CA).

[0070] Evaluation of In Vivo Anti-Tumor Efficacy: In FIGURE 7A and FIGURE 7B, which depict the results of this study, it can be seen that, i.v. administration of CD19R TCM cells showed only transient therapeutic benefit, while CD19R TCM/SCM/N cells exhibited more robust anti-tumor activity against the ffLuc+ Sup-Bl 5 tumors.

[0071] Analysis of T Cells in Peripheral Blood: To evaluate the persistence of T cells in the peripheral blood in these mice, flow cytometric analysis was performed on retro-orbital bleeds collected 40 days after i.v. T cell administration. As depicted in FIGURE 8, CAR+ T cells were only detectable at this timepoint in the mice that had received TCM/SCM/N cells. This presence of CAR+ human T cells correlated with the absence of GFP-expressing SupB15 tumor cells.

[0072] These studies demonstrate that in vivo anti-tumor efficacy can be observed with both CD19R TCM- and CD19R TcM/scM/N-derived cell lines, with the anti-tumor responses of the CD19R TCM/SCM/N cell-treated mice being greater than that of the CD19R TcM-treated mice. This efficacy correlated in part with the ability to detect CAR+ T cells in the blood of CD19R TcM/scM/N-treated mice.

Example 6: Effector Activity of TCM and TCM/SCM/N Cells Transduced with CD19R(EQ)28Z-T2A-EGFRt_epfflV7

[0073] A study was conducted to evaluate CD19-specific cytolytic activity, degranulation, and cytokine production of either CD62L+ central memory (TCM) cells, or CD62L+ TCM/SCM/N cells that were transduced with the CD 19R(EQ)28Z-T2A- EGFRt_epHTV7 lentiviral vector and expanded using methods expected to be suitable for clinical use. Cytolytic activity was evaluated using a flow cytometry-based long term killing assay and a 5-hour degranulation assay. Cytokine production was evaluated by intracellular staining of the T cells after 5-hour co-culture with stimulator cells. [0074] Experimental Design: TCM cells and TCM/SCM/N cells were lenti-transduced with CD19R(EQ)28Z-T2A-EGFRt_epHrV7 and expanded for up to 21 days (See

Examples 2 and 4). CD19-specific effector function of these cell products was evaluated by a long term killing assay which evaluates loss of target cells after 72 hours of co-incubation with effector T cells, and a 5-hour degranulation assay which evaluates cell surface CD 107a mobilization as a marker of T cell cytotoxic activity. Evaluation of cytokine production by freshly thawed final cell products after 5 hours of in vitro stimulation was then conducted by flow cytometric analysis of intracellular staining for IFN^y.

[0075] Brief Overview of Long Term Killing Assay: Samples of each T cell product were thawed the day prior to the assay, rested overnight in the presence of 50 U/mL rhIL-2 and 0.5 ng/mL rhIL-15, and a 72-hour killing assay was then performed.

Briefly, each T cell line and target cell line was counted and seeded at a 1 : 1 CAR+ T cell to target cell ratio into a 96-well tissue-culture treated, round-bottom plate - specifically, 25,000:25,000 in 200uL X-VIVO 15 media with 10% FBS per well. After 72 hours of culture, cells were collected with trypsinization, washed in cold FACS buffer, and then stained with anti-human CD45 to detect the T cells, with DAPI as a viability dye. Samples were run and analyzed on a MACSQuant (Miltenyi) instrument, and numbers of viable immunoreactive cells were calculated via Flow Jo software (Flow Jo, LLC, Ashland, OR). Final results were then graphed as percentages of CD45-negative, forward scatter high tumor cells remaining in the cultures when compared to that of mock-transduced effector cell co-cultures which had been normalized to 100%.Brief Overview of Degranulation Assay and Intracellular Cytokine Analysis: Samples of each T cell product were freshly thawed, and a 5-hour degranulation assay was then performed. T cells were counted and resuspended in media containing Golgi Stop, a protein transport inhibitor that blocks the reabsorption of CD 107a from the cell surface. The T cells were then seeded at a 1 : 1 ratio with tumor cells (with Golgi Stop) in a 96-well plate. To each well, an antibody against CD 107a was added, and then the plate was incubated for 5 hours at 37°C. Additional antibodies then were added to each condition to stain for CD45, CD4, CD8, and the EGFRl transgene. Cells were then fixed and permeabilized for staining with antibody specific for lFN-γ or isotype control antibody. Samples were run and analyzed on a MACSQuant (Miltenyi) instrument, and percentages of immunoreactive cells were calculated via FlowJo software (FlowJo, LLC, Ashland, OR).

[0076] Determination of Cytolytic Activity by Long Term Killing Assay: TCM cells and TcM/scM/N cells that had either been mok-transduced or transduced with

CD19R(EQ)28Z-T2A-EGFRt_epHIV7 were each plated with either CD19-negative K562 cells or CD 19+ SupB15 cells for 72 hours. Results are presented in FIGURE 9 and show similar loss of detectable SupBIS targets in the cultures with both CD19R TCM and CD19R TCM/SCM/N cells when compared to their respective mock-transduced controls. Indeed, this loss of SupBIS targets was significantly greater than that observed with K562 targets, further supporting the CD19-specificity of this cytolytic activity.

[0077] Determination of Effector Activity by Degranulation Assay: TCM cells and TCM/SCM/N cells that had either been mock-transduced or transduced with

CD19R(EQ)28Z-T2A-EGFRt_epHIV7 were each plated with either CD19-negative K562 cells or CD19+ SupB15 cells. Results are presented below (FIGURE 10) and demonstrate that the CD8+ CD19R TCM and CD8+ CD19R TCM/SCM/N cells exhibit similar CD19-specific degranulation (i.e., CD 107a expression) upon stimulation with SupB15 cells. The antigen-specificity of this activity is further evidenced by examining the mock-transduced cells (FIGURE 10) which are not stimulated to degranulate by the CD19-expressing SupB15 cells.

[0078] Determination of Effector Activity bv Intracellular Cytokine Assav: TCM cells and TCM/SCM/N cells that had either been mock-transduced or transduced with

CD19R(EQ)28Z-T2A-EGFRt_epHIV7 were each plated with either CD19-negative K562 cells or CD19+ SupB15 cells. Results are presented in FIGURE 11 and demonstrate that the CD8+ CD19R TCM and CD8+ CD19R TCM/SCM/N exhibit similar CD19-specific, intracellular IFN-7 profiles. The antigen-specificity of this activity is further evidenced by examining the mock-transduced cells 0 which are not stimulated to express IFN-y by the CD19-expressing SupB15 cells.

[0079] These data confirmed that both TCM cells and TCM/SCM/N cells that had been transduced with CD19R(EQ)C28Z-T2A-EGFRt_epfflV7 exhibit cytolytic activity, degranulation and cytokine (i.e., IFN-γ) production upon co-culture with CD19- expressing target cells.

Example 7: Effector Activity of TCM and TCM/SCM/N Cells Transduced with lL13(EQ)BBZ-T2A-CD19t_epfflV7

[0080] A study was conducted to evaluate the in vivo efficacy of CD62L+ TCM cells and CD62L+ TCM/SCM/N cells transduced with lentiviral vector expressing a CAR targeted to IL13Ra2. The CAR includes a human IL-13 variant and is expressed by a lentiviral vector (IL13(EQ)BBZ-T2A-CD19t_epHIV7). The cell populations expressing this vector were prepared as described above in Example 4 and assessed in a murine model of glioblastoma in immunodeficicnt NSG mice using the IL13Ra2+ primary low-passage GBM tumor sphere line PBT030-2, which has been engineered to express the firefly luciferase (ffLuc) reporter gene.

[0081] Briefly, male NSG mice (10-12 weeks old) were stereotactically injected with 1 x 10 ³ ffLuc+ PBT030-2 cells in both the right and left contralateral hemispheres on day 0 and allowed to engraft for 6 days. Groups of mice were then left untreated or treated with 1 x 10 ⁶ CAR+ IL13(EQ)BB7CD19t+ TCM/SCMTN, CAR+

+ TCM, mock transduced TCM/SCMTN or mock transduced TCM. PBT030-2 tumor growth was monitored over time by Xenogen imaging and quantification of ffLuc flux (photons/sec).

[0082] As shown in FIGURE 12, TCM/SCM/N transfccted with a CAR directed against IL- IL13Ra2 were superior to similarly transfected TCM with respect to suppression of tumor growth and with respect to survival in a murine intratumoral glioblastoma model.