Rapid Generation of Multivirus-Specific T Lymphocytes for the Prevention and Treatment of Respiratory Viral Infections

Abstract

Respiratory viral infections due to community-acquired respiratory viruses (CARVs) including respiratory syncytial virus (RSV), influenza, parainfluenza virus (PIV) and human metapneumovirus (hMPV) are detected in up to 40% of allogeneic hematopoietic stem cell transplant (allo-HSCT) recipients, in whom they may cause severe disease such as bronchiolitis and pneumonia that can be fatal. Given the lack of approved antiviral agents for these CARVs and our data demonstrating that adoptively transferred ex vivo-expanded virus-specific T cells (VSTs) can be clinically beneficial for the treatment of both latent [Epstein-Barr virus (EBV), cytomegalovirus (CMV), BK virus (BKV), human herpesvirus 6 (HHV6)] and lytic [adenovirus (AdV)] viruses in recipients of allo-HSCT, we wanted to explore the potential for extending this approach to Influenza, RSV, hMPV and PIV3. Thus, we exposed PBMCs from healthy donors to a cocktail of pepmixes (overlapping peptide libraries) spanning immunogenic antigens from our target viruses [Influenza – NP1 and MP1; RSV – N and F; hMPV – F, N, M2-1 and M; PIV3 – M, HN, N and F] followed by expansion in the presence of activating cytokines in a G-Rex. Over 10-13 days we achieved an average 8.5 fold expansion (increase from 0.25 × 107 PBMCs/cm2 to mean 1.9±0.2 × 107 cells/cm2; n=12). Cultures comprised almost exclusively CD3+ T cells (96.2±0.6%; mean±SEM), a mixture of cytotoxic (CD8+) and helper (CD4+) T cells, with a phenotype consistent with immediate effector function and long term persistence, as evidenced by upregulation of the activation markers CD25, CD69, and CD28 and expression of central (CD45RO+/CD62L+) and effector memory markers (CD45RO+/CD62L−), with minimal PD1 or Tim3. Anti-viral specificity of multi-respiratory-VSTs was tested in an IFNγ ELISpot assay using each of the individual stimulating antigens as an immunogen. All 12 lines screened were reactive against each of the target viruses [Influenza: mean 735±75.6 SFC/2 × 105, RSV: 758±69.8, hMPV: 526±100.8, PIV3: 391±93.7]. The expanded VSTs were Th1-polarized effector cells, as evidenced by production of TNFα, GM-CSF and Granzyme B, with only baseline levels of Th2/suppressive cytokines. Finally, the cells were tested in a standard Cr51 release assay and were able to lyse viral pepmix-loaded autologous PHA blasts (40:1 E:T - Influenza: 13±5%, RSV: 36±8%, hMPV: 26±7%, PIV-3: 22±5%, n=8) with no evidence of auto- or alloreactivity, attesting to both their selectivity and their safety for clinical use in HSCT recipients. In conclusion, we have shown that it is feasible to rapidly generate a single preparation of polyclonal multi-respiratory (multi-R)-VSTs targeting Influenza, RSV, hMPV and PIV3 by a total of 12 encoded antigens using GMP-compliant manufacturing methodologies. We expect such multi-R-VSTs would provide clinical benefit in preventing or treating CARV infections in the immunocompromised.