Off-the-Shelf Virus-Specific T Cells to Treat BK Virus, Human Herpesvirus 6, Cytomegalovirus, Epstein-Barr Virus, and Adenovirus Infections After Allogeneic Hematopoietic Stem-Cell Transplantation

Abstract

Purpose Improvement of cure rates for patients treated with allogeneic hematopoietic stem-cell transplantation (HSCT) will require efforts to decrease treatment-related mortality from severe viral infections. Adoptively transferred virus-specific T cells (VSTs) generated from eligible, third-party donors could provide broad antiviral protection to recipients of HSCT as an immediately available off-the-shelf product. Patient and Methods We generated a bank of VSTs that recognized five common viral pathogens: Epstein-Barr virus (EBV), adenovirus (AdV), cytomegalovirus (CMV), BK virus (BKV), and human herpesvirus 6 (HHV-6). The VSTs were administered to 38 patients with 45 infections in a phase II clinical trial. Results A single infusion produced a cumulative complete or partial response rate of 92% (95% CI, 78.1% to 98.3%) overall and the following rates by virus: 100% for BKV (n = 16), 94% for CMV (n = 17), 71% for AdV (n = 7), 100% for EBV (n = 2), and 67% for HHV-6 (n = 3). Clinical benefit was achieved in 31 patients treated for one infection and in seven patients treated for multiple coincident infections. Thirteen of 14 patients treated for BKV-associated hemorrhagic cystitis experienced complete resolution of gross hematuria by week 6. Infusions were safe, and only two occurrences of de novo graft-versus host disease (grade 1) were observed. VST tracking by epitope profiling revealed persistence of functional VSTs of third-party origin for up to 12 weeks. Conclusion The use of banked VSTs is a feasible, safe, and effective approach to treat severe and drug-refractory infections after HSCT, including infections from two viruses (BKV and HHV-6) that had never been targeted previously with an off-the-shelf product. Furthermore, the multispecificity of the VSTs ensures extensive antiviral coverage, which facilitates the treatment of patients with multiple infections.

Fig 1.

Cumulative incidence response rate that is based on time to first complete response and/or partial response in (A) overall patients (n = 37); one patient with nonevaluable data was excluded; (B) patients with cytomegalovirus infections (n = 17); (C) patients with adenovirus infections (n = 7), and (D) patients with BK virus infections (n = 16).

Fig 2.

Treatment outcomes in patients infected with cytomegalovirus (CMV), Epstein-Barr virus (EBV), adenovirus, and human herpesvirus 6 (HHV-6). Depiction of plasma viral load before (pre) and after (post) infusion of virus-specific T cells in all patients treated for (A) CMV infection, (B) EBV infection, (C) adenovirus infection, and (D) HHV-6 infection. (*) Nonresponders.

Fig 3.

Frequency of viral-specific T cells (VSTs) in vivo in the peripheral blood before (pre) and after (post) infusion, as measured by interferon gamma enzyme-linked immunospot (ELIspot) assay after stimulation with viral pepmixes. Results are expressed as spot-forming cells (SFCs) per 5 × 10⁵ input cells with specificity for patients with (A) cytomegalovirus, (B) Epstein-Barr virus, (C) adenovirus, (D) human herpesvirus 6, and (E) BK virus. (F-K) Frequency of T cells in peripheral blood as measured by interferon gamma ELIspot assay after stimulation with epitope-specific peptides with restriction to HLA antigens exclusive to the VST line or the recipient, or shared between the two.

Fig 4.

Treatment outcomes in BK virus–infected patients (n = 14). BK virus–associated hemorrhagic cystitis symptom score before infusion and at 2, 4, and 6 weeks after infusion of virus-specific T cells (VSTs); the grade is based on the National Cancer Institute hemorrhagic cystitis grading scale. Results are presented as (A) mean (± SEM) grade of all patients and (B) individually for each patient. (C) Example of a patient with BK virus–associated hemorrhagic cystitis. Ultrasound imaging depicts the blood clot in the bladder, shown (left) preinfusion and (right) 3 weeks after infusion.

Fig A1.

Decision algorithm for virus-specific T-cell (VST) selection.

Fig A2.

Examples of patients infected with cytomegalovirus (CMV): (A) Patient 2936 with CMV infection. Viral load (left y-axis) and frequency of CMV-directed T cells in peripheral blood (right y-axis) before infusion and at 2, 4, and 6 weeks after infusion as measured by interferon gamma enzyme-linked immunospot assay. (B) Endoscopic picture of patient 3840 with CMV colitis that depicts ulcers (in circled areas) in the ileum and colon before infusion. SFC, spot-forming cell.

Fig A3.

Example of a patient infected with adenovirus; patient 4002 with adenovirus respiratory tract infection; plasma viral load (left y-axis) and frequency of virus-specific T cells (VSTs; right y-axis) in peripheral blood over time. SFC, spot-forming cell.