Safety and Preliminary Efficacy of "Ready to Administer" Cytomegalovirus (CMV)-Specific T Cells for the Treatment of Patients with Refractory CMV Infection

Despite advances in antiviral drugs, Cytomegalovirus (CMV) infections remain a significant cause of morbidity and mortality in immunocompromised individuals. We have recently demonstrated in hematopoietic stem cell transplant (HSCT) recipients that adoptively-transferred virus-specific T cells, generated from healthy 3^rd party donors and administered as an "ready to administer" product, can be curative, even in patients with drug-refractory CMV infections. However, broader implementation has been hindered by the postulated need for extensive panels of T cell lines representing a diverse HLA profile, as well as the complexities of large scale manufacturing for widespread clinical application. To address these potential issues, we have developed a decision tool that identified a short list of donors who provide HLA coverage for >90% of the stem cell transplant population. Furthermore, to generate banks of CMV-specific T cells from these donors, we have created a simple, robust, and linearly scalable manufacturing process. To determine whether these advances would enable the widespread application of "ready to administer" T cells, we generated CMV cell banks (Viralym-C™) from 9 healthy donors selected by our decision tool, and initiated a fixed-dose (2x10⁷ cells/m²) Phase I clinical trial for the treatment of drug-refractory CMV infections in pediatric and adult HSCT recipients.

To generate the Viralym-C™ banks, we stimulated donor peripheral blood mononuclear cells (PBMCs) with overlapping peptide libraries spanning the immunodominant CMV antigens pp65 and IE1. Cells were subsequently expanded in a G-Rex device, resulting in a mean fold expansion of 103±12. The lines were polyclonal, comprising both CD4+ (21.3±6.7%) and CD8+ (74.8±6.9%) T cells, and expressed central CD45RO+/CD62L+ (58.5±4.2%) and effector memory markers CD45RO+/CD62L- (35.3±12.2%). Furthermore, the lines generated were specific for the target antigens (IE1: 419±100; pp65 1070±31 SFC/2x10⁵, n=9).

To date, we have screened 12 patients for study participation, and from our bank of just 9 lines we have successfully identified a suitable line for all patients within 24 hours. Of these, 6 patients have been infused; 5 received a single infusion and 1 patient required 2 infusions for sustained benefit. There were no immediate infusion-related toxicities; and despite the HLA disparity between the Viralym-C lines and the patients infused, there were no cases of de novo or recurrent graft versus host disease (GvHD). One patient developed a transient fever a few hours post-infusion, which spontaneously resolved. Based on viral load, measured by quantitative PCR, or symptom resolution (in patients with disease), Viralym-C™ cells controlled active infections in all 5 evaluable patients; 4 patients had complete responses, and 1 patient had a partial response within 4 weeks of cell infusion. One patient with CMV retinitis had complete resolution of symptoms following Viralym-C™ infusion.

In conclusion, our results demonstrate the feasibility, preliminary safety and efficacy of "ready to administer" Viralym-C™ cells that have been generated from a small panel of healthy, eligible CMV seropositive donors identified by our decision support tool. These data suggest that cost-effective, broadly applicable T cell anti-viral therapy may be feasible for patients following HSCT and potentially other conditions.