Safety and feasibility of virus-specific T cells derived from umbilical cord blood in cord blood transplant recipients

Abstract

Adoptive transfer of virus-specific T cells (VSTs) has been shown to be safe and effective in stem cell transplant recipients. However, the lack of virus-experienced T cells in donor cord blood (CB) has prevented the development of ex vivo expanded donor-derived VSTs for recipients of this stem cell source. Here we evaluated the feasibility and safety of ex vivo expansion of CB T cells from the 20% fraction of the CB unit in pediatric patients receiving a single CB transplant (CBT). In 2 clinical trials conducted at 2 separate sites, we manufactured CB-derived multivirus-specific T cells (CB-VSTs) targeting Epstein-Barr virus (EBV), adenovirus, and cytomegalovirus (CMV) for 18 (86%) of 21 patients demonstrating feasibility. Manufacturing for 2 CB-VSTs failed to meet lot release because of insufficient cell recovery, and there was 1 sterility breach during separation of the frozen 20% fraction. Delayed engraftment was not observed in patients who received the remaining 80% fraction for the primary CBT. There was no grade 3 to 4 acute graft-versus-host disease (GVHD) associated with the infusion of CB-VSTs. None of the 7 patients who received CB-VSTs as prophylaxis developed end-organ disease from CMV, EBV, or adenovirus. In 7 patients receiving CB-VSTs for viral reactivation or infection, only 1 patient developed end-organ viral disease, which was in an immune privileged site (CMV retinitis) and occurred after steroid therapy for GVHD. Finally, we demonstrated the long-term persistence of adoptively transferred CB-VSTs using T-cell receptor-Vβ clonotype tracking, suggesting that CB-VSTs are a feasible addition to antiviral pharmacotherapy.

Graphical abstract

Figure 1.

Schematic of CB-VST manufacture in ACT-CAT and ACTCAT2 clinical trials. Eighty percent of the CB unit was used as the CB transplant graft, and 20% was used to expand the VST. From the 20% fraction, 5 to 10 million peripheral blood mononuclear cells (PBMCs) were used for EBV lymphoblastoid cell line (LCL) generation; the remaining cells were used to generate dendritic cells (DCs) or used as the source of T cells. DCs were matured and transduced with the Ad5f35-pp65 vector (ACT-CAT) on day 5 and/or cocultured with overlapping viral peptides (PepMix) (ACT-CAT2), then cocultured with the nonadherent fraction (containing T cells) on day 7 in the presence of IL-7, IL-12, and IL-15. At the second and third stimulation, EBV-transformed B cells were transduced with the same Ad5f35-pp65 vector 2 days before the T-cell stimulation or pulsed with PepMix the day of stimulation, which was performed in the presence of IL-15 (second stimulation) and IL-2 (third stimulation).

Figure 2.

VST product characteristics. (A) Cell yields after manufacture using APCs either transduced with the Ad5f35-pp65 vector (ACT-CAT) or cocultured with overlapping viral peptides (ACT-CAT2). (B) VST products demonstrated multivirus specificity as measured by IFN-γ ELISPOT assay in response to CMV, EBV, and adenovirus antigen. Data for 19 VST products is represented as the number of SFCs per 100 000 after subtracting the values of the actin negative control. (C) A CD4⁺ T-cell predominant phenotype was observed with similar effector memory (CD45RA^–/CD62L^–) and central memory (CD45RA^–/CD62L⁺) percentages in the VST products. Memory phenotype data were not available for products P0020, P0035, P0039, P005, and P0067 because of limited product availability for analysis after infusion.

Figure 3.

T-cell persistence and viral load in patients treated with CB-VSTs. (A) Detection of CMV-specific T cells in the peripheral blood of patient P0039 (red bars) and the CMV viral load (green line). Detection of CMV-, EBV-, and adenovirus-specific T cells at baseline compared with 3 months after VST infusion in the viral treatment group (B) and the prophylaxis group (C).

Figure 4.

Detection and persistence of unique TCRs over time using TCR sequencing. (A-C) Shown are unique clonotypes that were present in the VST cell product but not present before VST cell infusion. (D) TCR-Vβ diversity over time in patient P2891 in CD45RA⁺ (naïve) and CD45RA^– (memory) T-cell populations. (E) Persistence of unique T-cell clones over time in all patients with available data. (F) Unique T-cell clone contribution at 1 month after VST infusion separated by whether patients had detectable virus within the first month of infusion (reactivation) or not (no reactivation).