Derivation of human T lymphocytes from cord blood and peripheral blood with antiviral and antileukemic specificity from a single culture as protection against infection and relapse after stem cell transplantation

Abstract

Viral infections and leukemic relapse account for the majority of treatment failures in patients with B-cell acute lymphoblastic leukemia (B-ALL) receiving allogeneic hematopoietic stem cell (HSC) or cord blood (CB) transplants. Adoptive transfer of virus-specific cytotoxic T lymphocytes (CTLs) provides protection against common viruses causing serious infections after HSC transplantation without concomitant graft-versus-host disease. We have now generated CTL lines from peripheral blood (PB) or CB units that recognize multiple common viruses and provide antileukemic activity by transgenic expression of a chimeric antigen receptor (CAR) targeting CD19 expressed on B-ALL. PB-derived CAR(+) CTLs produced interferon-gamma (IFNgamma) in response to cytomegalovirus-pp65, adenovirus-hexon, and Epstein-Barr virus pepmixes (from 205 +/- 104 to 1034 +/- 304 spot-forming cells [SFCs]/10(5) T cells) and lysed primary B-ALL blasts in (51)Cr-release assays (mean, 66% +/- 5% specific lysis; effector-target [E/T] ratio, 40:1) and the CD19(+) Raji cell line (mean, 78% +/- 17%) in contrast to nontransduced controls (8% +/- 8% and 3% +/- 2%). CB-derived CAR(+) CTLs showed similar antiviral and antitumor function and both PB and CB CAR(+) CTLs completely eliminated B-ALL blasts over 5 days of coculture. This approach may prove beneficial for patients with high-risk B-ALL who have recently received an HSC or CB transplant and are at risk of infection and relapse.

Figure 1

Expansion and phenotype of PB-derived MV-CTLs genetically modified to express CAR.CD19. (A) Fold expansion of PB-derived CAR.CD19⁺ MV-CTLs compared with nontransduced control MV-CTLs. Fold expansion = total cell count at time point indicated divided by cell count at transduction. Means and SDs from 3 experiments are shown. (B) Phenotype of 3 CAR.CD19⁺ MV-CTLs at time of harvest. Percentage of total cells positive for each marker is shown. (C) CAR expression at completion of culture period in transduced MV-CTLs (top histogram) compared with nontransduced controls (bottom histogram). Representative plots from 1 of 3 experiments shown. (D) Coexpression of CAR.CD19 by CMV-pp65–, AdV-Hexon–, and EBV-BZLF1–specific CD8⁺ CTLs as detected by pentamer staining. Top 3 plots show CAR.CD19⁺ CTLs; bottom 3 plots show nontransduced CTLs for comparison. Sample plots from 1 of 3 experiments shown. TD CTLs indicates transduced cytotoxic T lymphocytes; and NT CTLs, nontransduced cytotoxic T lymphocytes.

Figure 2

Retention of native antiviral activity of CAR.CD19-transduced PB-derived MV-CTLs. (A) IFNγ production by transduced and nontransduced MV-CTLs in response to CMV-pp65, AdV-hexon, and EBV antigens and irrelevant pepmix as measured by ELISPOT assay. Means and SDs from 3 CTL lines are shown. (B) Specific lysis of CMV-pp65, AdV-hexon, and irrelevant pepmix-pulsed autologous PHA blasts, and autologous EBV⁺ CD19⁺ LCLs by transduced and nontransduced MV-CTLs at E/T ratio of 40:1 measured by ⁵¹Cr-release assay. Means and SDs from 3 experiments are shown. (C) Surface expression of CAR.CD19 on both CD8⁺ (top plots) and CD4⁺ (bottom plots) MV-CTLs producing IFNγ in response to CMV-pp65, AdV-hexon, AdV-penton, and EBV-BZLF-1 pepmixes visualized by intracellular flow cytometry. Representative plots from 1 of 3 experiments are shown. TD CTLs indicates transduced cytotoxic T lymphocytes; NT CTLs, nontransduced cytotoxic T lymphocytes.

Figure 3

Antileukemic activity of CAR.CD19-transduced PB-derived CTLs. (A) Specific lysis of allogeneic CD19⁺ (Raji cell line and primary B-ALL cells) and the CD19⁻ HDLM-2 cell line by PB-derived CAR.CD19⁺ MV-CTLs compared with nontransduced controls at E/T ratio of 40:1. Relative contribution of NK-cell activity was measured using K562 targets. Means and SDs from 3 experiments are shown (*P < .05; **P < .01). (B) IFNγ production by CAR.CD19⁺CD3⁺ MV-CTLs (left plot), upon coculture with CD19⁺ targets compared with unstimulated CAR.CD19⁺ CTLs (top right plot) and nontransduced CTLs (bottom right plot) stimulated with CD19⁺ targets as measured by intracellular flow cytometry. Sample plots from donor 3 shown. (C) Relative expansion of CD19⁺ primary B-ALL cells from an HLA-matched bone marrow transplant recipient cocultured with nontransduced MV-CTL (left plot) or CAR.CD19⁺ CTLs (right plot). (D) Pattern of cytokine release from coculture experiment after 24-hour incubation as measured by CBA in 2 donors. Transduced CTLs plus B-ALL; nontransduced CTLs plus B-ALL (***P = .04). No significant release of cytokines was observed by nonstimulated nontransduced and transduced CTLs (not shown). TD CTLs indicates transduced cytotoxic T lymphocytes; NT CTLs, nontransduced cytotoxic T lymphocytes.

Figure 4

Serial stimulation of individual CTLs through native αβTCR and CAR.CD19. (A) Secretion of IFNγ by CAR.CD19⁺ MV-CTLs in response to stimulation with YSE peptide (y-axis) and CD19⁺ K562 cells (x-axis). (i) CTLs stimulated with both YSE and CD19. (ii) CTLs stimulated with YSE alone. (iii) CTLs stimulated with CD19 alone. (iv) CTLs stimulated with neither YSE nor CD19. (B) Secretion of IFNγ by nontransduced MV-CTLs in response to stimulation with YSE peptide (y-axis) and CD19⁺ K562 cells (x-axis). (i) CTLs stimulated with both YSE and CD19. (ii) CTLs stimulated with YSE alone. (iii) CTLs stimulated with CD19 alone. (iv) CTLs stimulated with neither YSE nor CD19. TD CTLs indicates transduced cytotoxic T lymphocytes; and NT CTLs, nontransduced cytotoxic T lymphocytes.

Figure 5

Antiviral function of CB-derived MV-CTLs. (A) Expression of CAR.CD19 in transduced (top histogram) compared with nontransduced (bottom histogram) CB-derived MV-CTLs 1 week after transduction. (B-C) IFNγ production by CAR.CD19⁺ CB-derived MV-CTLs compared with nontransduced CTLs in response to CMV-pp65 and AdV-hexon pepmixes as measured by ELISPOT assay for 2 of the CB donors. (D-E) Specific lysis of CMV-pp65, AdV-hexon, and irrelevant pepmix-pulsed autologous PHA blasts, and autologous EBV⁺ CD19⁺ LCLs by CAR-transduced (left graphs for each panel) and NT (right graphs for each panel) CB-derived MV-CTLs at the indicated E/T ratios measured by ⁵¹Cr-release assay for 2 representative donors. TD CTLs indicates transduced cytotoxic T lymphocytes; and NT CTLs, nontransduced cytotoxic T lymphocytes.

Figure 6

Antileukemic function of CB-derived MV-CTLs. (A) Specific lysis of allogeneic CD19⁺ (Raji cell line and primary B-ALL cells) and the CD19⁻ HDLM-2 cell line by CB-derived CAR.CD19⁺ CTLs compared with nontransduced controls at E/T ratio of 40:1. Relative contribution of NK-cell activity was measured using K562 targets. Data represent means and SDs from 4 CTL lines given for Raji and HDLM-2, and 2 CTL lines given for B-ALL samples (*P < .05; **P < .01). (B) Relative survival of allogeneic CD19⁺ primary B-ALL cells either alone (left plot) or cocultured with nontransduced (middle plot) or CAR.CD19⁺ CB-derived MV-CTLs (right plot) in one representative CB donor. (C) Pattern of cytokine release from coculture experiment with primary B-ALL blasts after 24-hour incubation as measured by CBA in 2 donors (***P = .01). No significant release of cytokines was observed by nonstimulated nontransduced and transduced CTLs (not shown). TD CTLs indicates transduced cytotoxic T lymphocytes; and NT CTLs, nontransduced cytotoxic T lymphocytes.