Methods and materials for expanding antigen-specific T cells in culture

Abstract

This document provides methods and materials for expanding antigen-specific T cells (e.g., antigen-specific CD4.sup.+ T cells and/or antigen-specific CD8.sup.+ T cells) in culture. For example, methods and materials for performing a polyclonal stimulation step for a particular duration (e.g., from about 1 hour to about 48 hours) to increase the expansion of T cells having a desired antigen specificity are provided.

Claims

1. A method for producing a cell population comprising T cells having specificity for an antigen of interest, wherein said method comprises: (a) culturing a first cell population comprising PBMCs in the presence of GM-CSF for a first period of time to form a GM-CSF treated cell population, (b) culturing said GM-CSF treated cell population in the presence of resiquimod, E. coli lipopolysaccharide, and said antigen of interest for a second period of time to form an antigen treated cell population, and (c) culturing said antigen treated cell population in the presence of IL-7 for a third period of time to form said cell population comprising said T cells.

2. The method of claim 1, wherein the cells of said first cell population are human cells.

3. The method of claim 1, wherein said antigen of interest is a cancer-associated antigen.

4. The method of claim 3, wherein said cancer-associated antigen is MUC1, HER2/neu, mesothelin, WT1, NYEso-1, MART1, gp100, or TRP, or tumor cells processed by freeze/thawing, irradiation, homogenization, or heat killing.

5. The method of claim 1, wherein said antigen of interest is a pathogen-associated antigen.

6. The method of claim 5, wherein said pathogen-associated antigen is a cytomegalovirus, Epstein-Barr virus, human papilloma virus, Mycobacterium tuberculosis, Candida albicans, aspergillis, Mycobacterium avian intracellularis, Ebola virus, or HIV antigen.

7. The method of claim 1, wherein said first period of time is 18 to 24 hours.

8. The method of claim 1, wherein said second period of time is 16 to 30 hours.

9. The method of claim 1, wherein said third period of time is 10 to 22 days.

10. A method for providing a mammal with T cells having specificity for an antigen of interest, wherein said method comprises administering, to said mammal, a cell population produced by a method comprising: (a) culturing a first cell population comprising PBMCs in the presence of GM-CSF for a first period of time to form a GM-CSF treated cell population, (b) culturing said GM-CSF treated cell population in the presence of resiquimod, E. coli lipopolysaccharide, and said antigen of interest for a second period of time to form an antigen treated cell population, and (c) culturing said antigen treated cell population in the presence of IL-7 for a third period of time to form said cell population.

11. The method of claim 10, wherein said mammal is a human.

12. The method of claim 10, wherein the cells of said first cell population are human cells.

13. The method of claim 10, wherein said antigen of interest is a cancer-associated antigen.

14. The method of claim 13, wherein said cancer-associated antigen is MUC1, HER2/neu, mesothelin, WT1, NYEso-1, MART1, gp100, or TRP, or tumor cells processed by freeze/thawing, irradiation, homogenization, or heat killing.

15. The method of claim 10, wherein said antigen of interest is a pathogen-associated antigen.

16. The method of claim 15, wherein said pathogen-associated antigen is a cytomegalovirus, Epstein-Barr virus, human papilloma virus, Mycobacterium tuberculosis, Candida albicans, aspergillin, Mycobacterium avian intracellularis, Ebola virus, or HIV antigen.

17. The method of claim 10, wherein said first period of time is 18 to 24 hours.

18. The method of claim 10, wherein said second period of time is 16 to 30 hours.

19. The method of claim 10, wherein said third period of time is 10 to 22 days.

Description

DESCRIPTION OF THE DRAWINGS

(1) FIG. 1. IL-12 is needed to expand Her2 specific Th1 ex vivo. PBMCs from a representative patient recently vaccinated with p776-790 15-mer, but not p42-56, were cultured with p776-790+rIL-2, with or without rIL-12, for 12 days, after which T-cells were re-exposed to the relevant or irrelevant peptide in an IFN-g ELISPOT assay. The frequency of p776-790 specific T-cells increased only from 0.0001% to 0.001%. This FIG. 1 is as described elsewhere (Knutson and Disis, Clin. Exp. Immunol., 135:322-329 (2004)).

(2) FIG. 2. DC1 polarization in cryopreserved PBMC. Thawed PBMCs were cultured overnight in GM-CSF, then treated with various TLR agonists or IFN-γ, followed by ELISA. Representative of 14 different healthy and cancer patients.

(3) FIG. 3 contains a table of culture comparison schema.

(4) FIG. 4. PBMC pulsed with GM-CSF then resiquimod+LPS (GM+TLR) display superior DC activation. Representative of 3 donors and 3 different Ag pulses per donor (Candida, HER2 intracellular domain (ICD), or ICD-derived 15mer (latter not shown)). At 48 hours, all CD33.sup.+ cells also expressed the DC marker CD11c (not shown). Every GM+TLR treated (“C”) group displayed upregulated HLA-DR and B7.1. In contrast, DCs in GM (“B”) or IL12 (“A”) groups were heterogeneously activated and in some cases minimally activated for B7.1 or HLA-DR. Numbers within dot plots show mean Y-axis fluorescent index of the CD33.sup.+ subpopulation. Each shown IL12p70 or IL23 determination is the average for all donors and Ags for A, B, or C culture conditions (n=9 for each condition). BD=below detection (<75 pg/mL).

(5) FIG. 5. Representative dot plots for a healthy donor's PBMC CAN- and ICD-driven groups. IL12 vs GM vs GM+TLR culture conditions were compared by an intracellular cytokine assay (ICC) at the end of Round One, and also to a Round Two continuation of the “GM+TLR” groups (resiquimod+LPS omitted after round 1). Only the CD4 window is shown. % shown within each dotplot is the % of all CD4.sup.+ T cells producing IFN-γ following exposure to freshly thawed PBMC either unpulsed (UP) or CAN- or ICD-pulsed. Parallel trends were observed regarding CD8.sup.+ T cell sensitization (not shown).

(6) FIG. 6. Impact of GM-CSF and IL-7 on GM+TLR culture method's enrichment of Ag-specific frequency. A. Individual wells of PBMCs received or did not receive rGM-CSF day 0 (“GM”). All groups received Candida (CAN) day 1, and variably received LPS and/or R848. All groups expanded in IL7/IL2 (“2_7”). Re-stimulation occurred 2 weeks later. Fold proliferation is shown. Graph shows calculated frequency of CAN-specific IFN-γ producing CD4.sup.+ (left) and CD8.sup.+ (right) T-cells. B. All PBM5C cultures received GM day 0, CAN, R848, and LPS day 1, and variably IL2, IL-7, IL-15, and/or IL-21 (2, 7, 15, 21) day 2 through 15. Fold proliferation is shown.

(7) FIG. 7. Impact of rIL-7 vs rIL-2 on proliferation of Ag-specific T-cells in culture. CAN was the driving Ag; all groups received GM-CSF/resiquimod/LPS, then either rIL7 or IL2. Cultures were restimulated at end of 2nd week with freshly thawed autologous PBMC pulsed with CAN. Only CAN-specific T-cells making IFN-γ are shown. Cells were previously loaded with Cell Trace violet to gauge proliferation (X-axis shows violet intensity. A shift from right to left indicates greater proliferation as the violet dye dilutes in half every time a labelled T-cell divides. Cells also were exposed to a live-dead stain. Y-axis shows live-dead stain intensity (up is dead, down is live). CAN-specific T-cells in IL-7 proliferated more with many less dead cells and higher yields (e.g., “IL-7 group” 10.8 fold gross expansion, frequency of viable CAN-specific T-cells 81% at end of culture versus “IL-2 group” 1.7 fold gross expansion, frequency of viable CAN-specific T-cells 40% at end of culture, (10.8×0.81)/(1.7×0.40) equals 12.9 fold higher number of CAN-specific T-cells present in “IL-7 group”).

(8) FIG. 8. Phenotype of 2 week cultured PBMCs employing “GM+TLR” culture method. Cells were pulsed day 1 with CAN. Upper dot plot shows ICC IFN-γ response 2 weeks later to fresh CAN-pulsed PBMC. Blue spike includes 100% of CAN-specific IFN-γ production by primed/expanded CD4.sup.+ cells (39.5% of total CD4.sup.+ T-cells showed CAN-specificity). Lower graphs show the expression of the blue spike CD4.sup.+ cells of CD28, CD56, and CCR7. % in RUQ and RLQ show percentage of CD4/INF-γ.sup.+ cells.

(9) FIG. 9. A range of cytokine production can be evoked from the culture system. The pie graph shows that approximately 80% of CD4.sup.+ T-cells secreted IFN-γ, IL-2, and/or IL-17 upon exposure to relevant Ag. Slices show % of Ag-specific T-cells that made either IFN-γ, IL-2, or IL-17; or combinations of two; or all three. These results demonstrate a “default” pattern of cytokine production straddling T1-*T17 functional phenotype. This distribution can indicate a highly effective population for adoptive immunotherapy.

(10) FIG. 10-12 are acquisition dot plots that are representative of the ability to selectively expand antigen-specific T-cells (both CD4.sup.+ and CD8.sup.+ T-cells) when cultures are driven by a CMV (cytomegalovirus) peptide (FIG. 10), a MUC1-derived peptide (FIG. 11), or a HER2/neu derived peptide (FIG. 12). In all cases, PBMCs were exposed to GM-CSF on Day 1 of culture, to the peptide antigen, then resiquimod, then LPS on Day 2 of culture, then to IL-7 beginning on Day 3. Polyclonal anti-CD3 stimulation was not performed. End of culture ranged from Day 12 to Day 21. Each figure shows the frequency of CD4.sup.+ and CD8.sup.+ T-cells making interferon-gamma upon re-exposure to the driving antigen or to control antigens at the end of culture (boxes within each dot plot show % of T-cells making interferon).

(11) FIG. 13 is a graph showing that unremoved (a) anti-CD3 antibody/anti-CD28 antibody beads or (b) immobilized anti-CD3 antibody and soluble anti-CD28 antibody treatment failed to expand antigen-specific T cells. All groups stimulated with CMV peptide on Day 2 (Step 1). “Unstimulated” group received standard treatment only (GM-CSF day 1, R848+LPS+CMV peptide day 2, then expanded in IL7 to day 21). Other groups also received either Activator anti CD3/28 Dynabeads (“Beads”) at 1:1 ratio or immobilized anti-CD3/soluble anti-CD28 (“Imm CD3/sol CD28”) each at 1 μg/mL. Number of Ag-specific CD4 and CD8 T cells at Day 19 harvest calculated per million starting PBMC. Calculation=Fold Expansion×% CD4 or % CD8×producing % IFN-γ upon specific re-exposure to CMV peptide.

(12) FIG. 14 is a graph showing that removing anti-CD3 antibody/anti-CD28 antibody beads or immobilized anti-CD3 antibody and soluble anti-CD28 antibody treatment allowed expansion of antigen-specific T cells. All groups stimulated with CMV peptide on Day 2 (Step 1). “Unstimulated” group as in FIG. 13. Other groups also received either Activator anti CD3/28 Dynabeads at 1:1 ratio or imm-CD3/sol-28 each at 1 μg/mL. Removal of beads and imm-CD3/sol-28 occurred 48 hours after addition. Number of Ag-specific CD4 and CD8 T cells at Day 21 harvest calculated as in FIG. 13.

(13) FIG. 15 is a graph showing results from restimulation of CMV-specific T cells with immobilized anti-CD3 antibody and soluble anti-CD28 antibody treatment on the indicated days. All groups stimulated with CMV peptide on Day 2 (Step 1). “Unstimulated” group as in FIG. 13. Other groups also received imm-CD3/sol-28 each at 1 μg/mL. Removal of imm-CD3/sol-28 occurred 48 hours after addition. Number of Ag-specific CD4 and CD8 T cells at Day 21 harvest calculated as in FIG. 13.

(14) FIG. 16 is a graph showing results from restimulation of MUC1-specific T cells with immobilized anti-CD3 antibody and soluble anti-CD28 antibody treatment on the indicated days. All groups stimulated with MUC1 peptide on Day 2 (Step 1). “Unstimulated” group received standard treatment only (GM-CSF day 1, R848+LPS+MUC1 peptide day 2, then expanded in IL7 to day 21). Other groups also received imm-CD3/sol-28 each at 1 μg/mL. Removal of imm-CD3/sol-28 occurred 48 hours after addition. Number of Ag-specific CD4 and CD8 T cells at Day 21 harvest calculated as in FIG. 13.

(15) FIG. 17 is a graph showing cell expansion results from restimulation of MUC1-specific T cells with an immobilized anti-CD3 antibody treatment, a soluble anti-CD28 antibody treatment, an immobilized anti-CD3 antibody and soluble anti-CD28 antibody treatment at the indicated doses for the indicated times. The methods were the same as for FIG. 16.

(16) FIG. 18 is a graph showing cell expansion results from restimulation of MUC1-specific T cells with anti-CD3 antibody/anti-CD28 antibody beads or an immobilized anti-CD3 antibody and soluble anti-CD28 antibody treatment at the indicated ratios for the indicated times. All groups received MUC1 peptide Day 2. “Unstimulated” group as in FIG. 13. Imm-CD3 used at concentration of 0.3 μg/mL. Soluble CD28 used at concentration of 1.0 μg/mL. Activator Dynabeads are used at bead to T-cell ratio and for bead exposure duration shown. Number of Ag-specific CD4 and CD8 T cells at Day 21 harvest calculated as in FIG. 13.

(17) FIG. 19. PBMCs from patients with advanced breast cancer can be expanded as described herein to generate both CD4.sup.+ and CD8.sup.+ natural T-cells that recognize cancer-associated antigens such as MUC1 and/or HER2. Culture outcome for PBMC assayed on Day 19, from patient with advanced breast cancer exposed on culture day 2 to a cocktail of SEA1, SEA2, SEA3, and CMV polypeptides. Plots show % of CD4.sup.+ or CD8.sup.+ T-cells producing IFNγ at culture end when restimulated with freshly thawed autologous PBMC either unpulsed (UP) or pulsed with individual peptides.

(18) FIG. 20. Culture outcome for PBMC assayed on Day 19, comparing patient with advanced breast cancer with three healthy donors, each exposed on culture day 2 to a cocktail of MUC1-derived peptides (SEA1, SEA2, and SEA3).

(19) FIG. 21. CD3.sup.+ T cells, CD19.sup.+ B-cells, CD33.sup.+ myeloid cells, and CD56.sup.+ natural killer cells prior to and after culture.

(20) FIG. 22. CD3.sup.+ (CD4.sup.+, CD8.sup.+, and CD56.sup.+) T cells before and after culture.

(21) FIG. 23. Costimulatory receptor and Foxp3 expression on T-cells at end of culture.

(22) FIG. 24. Coexpression of LAP, GARP, and Helios on Foxp3.sup.+ T-cells at end of culture.

(23) FIG. 25. Ag-specific cytokine production at end of culture.

(24) FIG. 26. T-cell culture expansion and Ag-specific frequency in 24 well plates vs Wilson-Wolf vessels.

DETAILED DESCRIPTION

(25) This document provides methods and materials for producing antigen-specific T cells (e.g., antigen-specific CD4.sup.+ T cells and/or antigen-specific CD8.sup.+ T cells). For example, this document provides methods and materials for performing a polyclonal stimulation step of defined duration (e.g., typically ranging from one hour to about 24 hours depending on the stimulus' intensity) to further increase the expansion of T cells having a desired antigen specificity.

(26) As described herein, peripheral blood cells (e.g., unfractionated peripheral blood mononuclear cells (PBMCs)) can be obtained from a mammal (e.g., a healthy human or a human cancer patient such as a human with advanced cancer). Examples of other cell populations that can be obtained and used as described herein to make populations of antigen-specific T cells (e.g., antigen-specific CD4.sup.+ T cells and/or antigen-specific CD8.sup.+ T cells) include, without limitation, tumor, lymph node, spleen, bone marrow, cerebrospinal fluid, pleural fluid, peritoneal fluid, and joint fluid samples that contain cells. Once obtained, the cells can initially be cultured in a manner that promotes antigen presentation. For example, unfractionated PBMC can be exposed to GM-CSF, the Toll-like receptor 8 agonist resiquimod, the Toll-like receptor 4 agonist E. coli lipopolysaccharide (LPS), or combinations thereof, with or without interpolated exposure to the antigen(s) of interest. In some cases, the cells can be exposed to GM-CSF, resiquimod, and E. coli lipopolysaccharide. The amount of GM-CSF can be from about 10 ng/mL to about 100 ng/mL (e.g., about 40 ng/mL). The amount of resiquimod can be from about 1 μg/mL to about 9 μg/mL (e.g., about 3 μg/mL). The amount of E. coli lipopolysaccharide can be from about 1 ng/mL to about 50 ng/mL (e.g., about 5 ng/mL). Other TLR8 agonists (e.g., motolimod) can be used in place of or in addition to resiquimod. Other TLR4 agonists (e.g., salmonella lipopolysaccharide) can be used in place of or in addition to E. coli LPS. In some cases, interferon-gamma and/or other TLR agonists such as poly I:C (TLR3 agonist) can be substituted for the Toll-like Receptor 4 or 8 agonist. The cells can be exposed to one or more agents designed to promote antigen presentation (e.g., GM-CSF, resiquimod, and E. coli lipopolysaccharide) for any appropriate length of time. For example, effective presentation of antigen(s) added to culture can be achieved by first exposing unfractionated PBMC to GM-CSF alone for 18-24 hours, then the next day adding the antigen(s) of interest, then four hours later adding resiquimod and 30 minutes later adding lipopolysaccharide for an additional 16-30 hours before moving into step 2 of culture.

(27) While treating the unfractionated PBMCs in a manner that promotes antigen presentation, the cells can be exposed to one or more desired antigens under conditions that activate T cells reactive against that desired antigen(s). The first step of culture can be pulsed with any appropriate antigen of interest. For example, cancer-associated antigens such as peptide sequences derived from MUC1, HER2/neu, or mesothelin, or processed tumor cells themselves (e.g., freeze-thawed lysates) or combinations thereof can be placed in contact with the cells. The amount of peptide antigen can be from about 5 μg/mL to about 100 g/mL (typically about 10-50 μg/mL). The cells can be exposed to one or more desired antigens for any appropriate length of time. For example, the cells can be exposed to one or more desired antigens for about 1 hour to about 24 hours. In some cases, the cells can be treated in a manner that promotes antigen presentation and exposure to one or more desired antigens concurrently. For example, the cells can be exposed to GM-CSF, an antigen(s) of interest, resiquimod, and E. coli lipopolysaccharide in that sequence within the first 36-60 hours of PBMC culture.

(28) After treating the unfractionated PBMCs in a manner that promotes antigen presentation and exposing the cells to at least one antigen of interest, the cells can be exposed to IL-7 as the next step of culture to preferentially expand the T-cell subpopulation within unfractionated PBMCs which recognizes the pulsed antigen(s) of interest. The amount of IL-7 can be from about 5 ng/mL to about 100 ng/mL (e.g., about 10 ng/mL every three days or about 50 ng/mL in added fresh medium). The cells can be exposed to IL-7 for any appropriate length of time. For example, the cells can be exposed to IL-7 for about 10-22 days (e.g., about 17-19 days).

(29) After treating the cells in a manner that promotes antigen presentation, exposing the cells to at least one antigen of interest, and exposing the cells to IL-7, the cells can be exposed to (i) immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and/or (ii) beads or matrix containing anti-CD3 antibodies and anti-CD28 antibodies. Examples of beads or matrix containing anti-CD3 antibodies and anti-CD28 antibodies include, without limitation, Dynabeads (obtained from Life Technologies, Inc.) and Miltenyi MACS GMP TransAct CD3/CD28 matrix). Dynabeads containing anti-CD3 and anti-CD28 antibodies and having diameters of about 4.5 μm can be used as described herein. The ratio of cells (cultured PBMCs) to beads can be from about 1:1 to about 6:1. When using immobilized anti-CD3 antibodies, the amount of immobilized anti-CD3 antibodies can be from about 0.1 μg/mL to about 10 μg/mL (e.g., about 0.3 μg/mL). When using soluble anti-CD28 antibodies, the amount of soluble anti-CD28 antibodies can be from about 0.1 μg/mL to about μg/mL (e.g., about 1 μg/mL).

(30) The cells can be exposed to (i) immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and/or (ii) beads or matrix containing anti-CD3 antibodies and anti-CD28 antibodies beginning on any appropriate day of culture. For example, the cells can be exposed to immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies on day 12 of culture. The cells can be exposed to (i) immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and/or (ii) beads or matrix containing anti-CD3 antibodies and anti-CD28 antibodies for any appropriate length of time. For example, the cells can be exposed to immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies for from about 1 hours to about 48 hours. In some cases, the cells are not exposed to immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies for more than 4 hours. In some cases, the cells can be exposed to beads containing anti-CD3 antibodies and anti-CD28 antibodies for from about 1 hour to about 48 hours. In some cases, the cells are not exposed to beads containing anti-CD3 antibodies and anti-CD28 antibodies for more than 4 hours. In some cases, the more beads added, the shorter the exposure can be to avoid lethally overstimulating the desired antigen-specific T-cell subpopulation.

(31) After exposing the cells to (i) immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and/or (ii) beads containing anti-CD3 antibodies and anti-CD28 antibodies, the cells are washed or treated in a manner that removes the immobilized anti-CD3 antibodies, soluble anti-CD28 antibodies, and beads containing anti-CD3 antibodies and anti-CD28 antibodies. For example, beads that are magnetic can be removed from the cells using a magnet.

(32) After removing the cells from exposure to immobilized anti-CD3 antibodies, soluble anti-CD28 antibodies, and/or beads containing anti-CD3 antibodies and anti-CD28 antibodies, the cells can continue to be cultured (i) in the absence of immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and (ii) in the absence of beads containing anti-CD3 antibodies and anti-CD28 antibodies. The cells can continue to be cultured in the absence of (i) immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and (ii) beads containing anti-CD3 antibodies and anti-CD28 antibodies for any appropriate length of time. For example, the cells can be cultured in the absence of (i) immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and (ii) beads containing anti-CD3 antibodies and anti-CD28 antibodies for from about 7 days to about 11 days. In some cases, the cells can be cultured in the absence of (i) immobilized anti-CD3 antibodies and soluble anti-CD28 antibodies and (ii) beads containing anti-CD3 antibodies and anti-CD28 antibodies for longer than 11 days.

(33) Once a population of antigen-specific T cells (e.g., antigen-specific CD4.sup.+ T cells and/or antigen-specific CD8.sup.+ T cells) is obtained as described herein, the cells can be administered to a mammal for use in, for example, adoptive cellular therapies to treat infections and/or cancer. Any appropriate mammal can be treated with the antigen-specific T cells (e.g., antigen-specific CD4.sup.+ T cells and/or antigen-specific CD8.sup.+ T cells) provided herein. For example, humans, horses, cattle, pigs, dogs, cats, mice, and rats can be treated with a population of antigen-specific T cells (e.g., antigen-specific CD4.sup.+ T cells and/or antigen-specific CD8.sup.+ T cells). In some cases, any appropriate number of antigen-specific T cells (e.g., antigen-specific CD4.sup.+ T cells and/or antigen-specific CD8.sup.+ T cells) provided herein can be administered to a mammal. For example, between about 1×10.sup.3 cells and about 1-2×10.sup.11 T cells including antigen-specific T-cells can be administered to a mammal. Any appropriate route of administration can be used to administer the antigen-specific T cells provided herein to a mammal. For example, antigen-specific T cells can be administered intravenously, intraperitoneally, subcutaneously, intramuscularly, intrahepatically, or intranodally.

(34) The invention will be further described in the following examples, which do not limit the scope of the invention described in the claims.

EXAMPLES

Example 1—Generating Populations of Antigen-Specific T Cells from PBMCs

(35) IL-12 was shown previously to be needed to expand Her2-specific Th1 ex vivo (FIG. 1). Hypothesizing that exogenous rIL-12 might not provide the full benefit of DC1 polarization to T cell cultures, the following was performed to achieve natural DC1 polarization, with IL12p70 production, within bulk PBMC cultures. Even though a variety of paired danger signals can effectively DC1-polarize purified fresh human DCs (e.g., LPS+IFN-γ), most such signals proved surprisingly ineffective for activating robust IL-12 production in unfractionated PBMCs. Only overnight exposure to GM-CSF, followed by combined exposure to TLR4+TLR8 agonists (LPS+resiquimod (R848)) was consistently effective to DC1-polarize unfractionated human PBMC, whether fresh or cryopreserved (FIG. 2). This “GM+TLR” culturing method was compared to culturing methods omitting resiquimod+LPS (“GM”) and compared to culturing methods that use exogenous IL-12, but without GM-CSF or TLR agonists (“IL-12” as in FIG. 1) (FIG. 3). Cryopreserved leukapheresed PBMCs from multiple healthy donors were repetitively tested for their ability to give rise to T1-type CD4.sup.+ and CD8.sup.+ T-cells specifically reactive to clinical-grade Candida extract (CAN) or HER2 intracellular domain protein (ICD) after 1 or 2 rounds of Ag-driven culture. The following was observed.

(36) Only GM+TLR cultures displayed consistent DC upregulation of MHC and costimulatory molecules (FIG. 4). Only GM+TLR cultures produced IL12p70 and IL-23 (FIG. 4). Ag-pulsed GM+TLR cultures significantly outperformed Ag-pulsed GM and IL-12 cultures in: fold-expansion, Ag-specific frequency, and total yield of Ag-specific T1 (IFN-γ.sup.+) cells, both CD4.sup.+ and CD8.sup.+, by the end of round 1 culture (FIG. 5). The frequency of HER2-specific, IFN-γ producing T cells at the end of Round 1 culture was 9.0%±2.1 of CD4.sup.+ and 4.1±1.5% of CD8.sup.+ cells, with a further surge in specificity typically seen following a second round of Ag-pulsed DC-driven culture (FIG. 5). This robust Ag-specific enrichment of both CD4.sup.+ and CD8.sup.+ T cells was often achievable within GM+TLR cultures even without prior vaccination to CAN or Her2Neu.

(37) Besides enriching for HER2-specificity, two rounds of optimized DC-driven culture resulted in 100-400-fold T cell numeric expansion, allowing 10.sup.11 or more T cells enriched for HER2-specificity to be prepared from merely 10.sup.9 PBMCs.

(38) Maximally enriched Ag-specificity in PBMC T cell cultures at this stage appeared to require GM-CSF and IL-7 and TLR agonist exposure (FIG. 6). Omitting dual TLR agonists effectively halved the proportion of Ag-specific CD8.sup.+ T-cells, and omitting GM-CSF effectively cut the proportion of Ag-specific CD4.sup.+ T-cells by a third. Omitting IL-7 effectively drove specificity down by a log. Furthermore, factors such as IL-21, IL-15, and IL-4 relatively impaired the development of enriched T cell specificity compared to IL-7 alone.

(39) T-cells grown in IL-2 rather than IL-7 mostly died. Surviving Ag-specific T-cells released IFN-γ, but displayed minimum proliferation. In contrast, T-cells grown in IL-7 mostly survived and proliferated if they were Ag-specific, explaining the logarithmic greater yield and higher frequency of Ag-specific T-cells, both CD4.sup.+ and CD8.sup.+, in IL-7 based cultures (FIG. 7).

(40) The phenotype of the propagated Ag-specific T-cells in GM+TLR cultures was CD28.sup.+ and CD56.sup.−, a likely ideal state of “young” differentiation for adoptive T cell therapy (FIG. 8). Both CCR7.sup.low and CCR7.sup.high subpopulations were present, consistent with a mixed presence of memory effector and central memory subsets (FIG. 8).

(41) FIG. 10-12 are acquisition dot plots that demonstrate the ability to selectively expand antigen-specific T-cells (both CD4.sup.+ and CD8+ T-cells) when cultures are driven by a CMV (cytomegalovirus) peptide (FIG. 10), a MUC1-derived peptide (FIG. 11), or a HER2/neu derived peptide (FIG. 12). In all cases, PBMCs were exposed to GM-CSF on Day 1 of culture, to the peptide antigen, then resiquimod, then LPS on Day 2 of culture, then to IL-7 beginning on Day 3. The end of culture ranged from Day 12 to Day 21. No polyclonal stimulation step was performed. Each figure shows the frequency of CD4.sup.+ and CD8.sup.+ T-cells making interferon-gamma upon re-exposure to the driving antigen or to control antigens at the end of culture (boxes within each dot plot show % of T-cells making interferon-gamma). These figures demonstrate the ability of GM-CSF, resiquimod, LPS, and interleukin-7 to generate rapid enrichment of Ag-specific CD4.sup.+ and CD8.sup.+ T cells even without polyclonal anti-CD3 stimulation. FIGS. 13-18 demonstrate how anti-CD3 polyclonal stimulation of defined optimal duration, followed by further culture in the absence of polyclonal stimulation, can substantially increase both the absolute yields and often the frequency of antigen-specific T-cells.

(42) Taken together, these results demonstrate that large numbers of antigen-specific T cells can be achieved from PBMCs within 21 days or less.

Example 2—Generating Populations of Antigen-Specific T Cells from PBMCs

(43) The following was performed to confirm that PBMCs from patients with advanced breast cancer can be expanded as described herein (with the exception of the anti-CD3 bead stimulation at Day 12 of culture) to generate both CD4.sup.+ and CD8.sup.+ natural T-cells that recognize cancer-associated antigens such as MUC1 and/or HER2. The following also was performed to confirm that T cells can be successfully sensitized even when the antigens are provided as cocktails of polypeptides, rather than as individual polypeptides.

(44) Briefly, PBMCs were obtained from a 42 year old female patient diagnosed with right-sided ER−/PR−/HER2+ breast cancer, treated initially with neoadjuvant chemotherapy including Herceptin (trastuzumab) and surgical resection, and placed on maintenance Herceptin. Liver metastasis was identified, and she started receiving ado-trastuzumab emtansine (Kadcyla). After four cycles, brain metastasis became apparent for which she received stereotactic radiation therapy. PBMCs were collected and cryopreserved prior to further dosing with Kadcyla. Thawed PBMCs were cultured as described herein, exposing the PBMCs on Day 2 to a cocktail of MUC1-derived peptides (SEA1, SEA2, and SEA3), with or without a control CMV-derived peptide also added to the cocktail (each peptide at 10 g/mL). After culture expansion in interleukin-7 to day 19 of culture, T-cells were re-exposed to freshly thawed autologous PBMCs pulsed with individual relevant and control peptides.

(45) Results were obtained on Day 19 of culture when the Day 2 cocktail contained both SEA (MUC1-derived) peptides and cytomegalovirus (CMV) peptide, demonstrating that both CD4.sup.+ and CD8.sup.+ T-cells derived by the culture method were highly enriched for recognition of at least a portion of the driving antigens, in this case both CMV and SEA1 specificities (FIG. 19). In contrast, T-cells sensitized to a cocktail of MUC1-derived peptides not including CMV recognized MUC1 peptides at culture's end, but not CMV (not shown). Generated MUC1-specific T-cells can be therapeutically active if reinjected as autologous adoptive immunotherapy into patients with MUC1-expressing malignancies, which include the great majority of patients with breast, pancreatic, colorectal, and many other cancer types. Because the patient PBMCs in this example were limited in number (obtained by peripheral phlebotomy rather than leukapheresis), the experiment was run as a small scale screening culture in 24 well plates. The anti-CD3/CD28 culture stimulation step was omitted but can be included. PBMCs from patients with additional types of advanced cancers can be used. Successful activation of MUC1-reactive and/or HER2neu-reactive T-cells using the culture system described herein was representative for three patients with advanced breast cancer and one patient with non-recurrent breast cancer.

(46) As evidenced in FIGS. 19 and 20, the culturing did not require that polypeptides be added individually to culture. Thawed PBMCs from leukapheresed healthy donors or cancer patients were cultured as described herein, exposing the PBMCs on Day 2 to polypeptide cocktails. Data show that the T-cells derived from the culture method were highly enriched by Day 19 for recognition of at least a subset of the driving antigens for both CD4.sup.+ and CD8.sup.+ cells. Interestingly, while individually pulsed polypeptides generated a maximal enrichment if pulsed at 50 μg/mL, cocktail pulsing worked better at 10 μg/mL for each polypeptide, meaning that cocktails allowed conservation of not only polypeptide, but also of PBMCs and every cell other culture reagent, since far less PBMCs were required to generate the same absolute number of Ag-reactive T-cells when cocktails were employed. Furthermore, the cocktail approach can allow the cultures to go forward for therapeutic administration even if different donors exhibit disparate T-cell sensitizations (for example, to SEA2 and/or SEA3 rather than (or in addition to) SEA1).

(47) FIGS. 21 and 22 show representative compositions of different cell lineages before and after culture from a representative healthy donor and a cancer patient. Before culture, T-cells (CD3.sup.+) were around 50% of the total PBMC, with myeloid monocytes (CD33+), B-cells (CD19+), and NK cells (CD56.sup.+) constituting the other lineages present (FIG. 21). At culture's end (Day 19), CD33.sup.+ and CD19.sup.+ cells were scant in number (FIG. 21), whereas CD3.sup.+ T-cells represented 80-90% of total cells (FIG. 21). As shown in FIG. 22, CD4.sup.+ T-cells continued to represent the majority of CD3.sup.+ T-cells between Day 0 and 19, but CD8.sup.+ T-cells also expanded (FIG. 22) and also were enriched for Ag-specific reactivity (FIGS. 19 and 20).

(48) Whether from a healthy donor or a cancer patient, T-cells generated as described herein variably expressed CD28, PD1, CTLA4, and Foxp3 at the end of culture (Day 19). T-cells were analyzed for coexpression of various receptors for costimulatory ligands (CD28, the activating receptor for B7.1; CTLA4, the inhibitory receptor for B7.1; and PD1, the inhibitory receptor for PD-L1 (aka B7H1)).

(49) The percent of total CD4.sup.+ or CD8.sup.+ T-cells expressing these receptors and/or Foxp3 at Day 19 of culture was determined (FIG. 23). Representative isotype mAb staining (negative control) was shown in FIG. 23 as “Iso”. Further, subanalysis in FIG. 24 of the Foxp3.sup.+ subpopulation revealed that trivial numbers of the Foxp3.sup.+ T-cells coexpress LAP or GARP, meaning that the great majority of Day 19 Foxp3.sup.+ T-cells were effector rather than regulatory T-cells (FIG. 23). Similarly, the Helios.sup.+ subpopulation expressed negligible levels of LAP or GARP (FIG. 23), meaning that the Helios.sup.+ T-cell subset also constituted effector rather than regulatory T-cells. Furthermore, expression of PD1 and/or CTLA4 in FIG. 23 means that adoptive reinfusion of these T-cells may likely be therapeutically potentiated by checkpoint inhibitors that block PD1 (nivolumab and pembrolizumab) and/or CTLA4 (ipilimumab) (Topalian et al., Nat. Rev. Cancer., 16(5):275-87 (2016)).

(50) Luminex assays at culture's end demonstrated that many cytokines were secreted by cultured T-cells upon re-exposure to the culture-driving Ag, dominated by, but not limited to, IFN-γ secretion (FIG. 25). CMV-driven cultures (first two bars of each graph of FIG. 25) were re-stimulated with freshly thawed autologous PBMC, either unpulsed (first bar of each graph of FIG. 25) or CMV-pulsed (second bar of each graph of FIG. 25). Similarly, SEA2-driven cultures (last two bars of each graph of FIG. 25) were re-stimulated with freshly thawed autologous PBMC, either unpulsed (third bar of each graph of FIG. 25) or SEA2-pulsed (fourth bar of each graph of FIG. 25). Such IFN-γ dominance may indicate that the therapeutic potency of T-cells expanded by these methods is high (Schwartzentruber et al., J. Immunol., 146(10):3674-81 (1991)).

(51) In addition, bulk expansion of Ag-driven PBMCs in Wilson-Wolf culture vessels worked better than small scaled cultures in 24 well plates (FIG. 26). Wilson-Wolf G-Rex vessels with gas-permeable “base” membranes were developed by Wilson-Wolf to prevent untoward acidity from building up in bulk T-cell cultures (Jin et al., J. Immunother., 35(3):283-92 (2012)). This allowed a variety of large scale cultures to be sustained improved total T-cell yields and Ag-reactive frequency at 19 days of culture, even without incorporating an anti-CD3/CD28 polyclonal re-stimulation step at Day 12.

(52) These results demonstrate that the culturing techniques provided herein perform well with cells from advanced cancer patients as well as healthy donors and that cultures can be pulsed with a cocktail of polypeptides, rather than a single polypeptide, to develop subsets of T-cells recognizing each of the polypeptides in the cocktail. These results also demonstrate that scaling up with Wilson-Wolf culture vessels is effective for expanding antigen-specific T-cells.

Other Embodiments

(53) It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.