Cliona M. Rooney

Adoptively transferred virus-specific T cells (VSTs) have shown remarkable safety and efficacy for the treatment of virus-associated diseases and malignancies in hematopoietic stem cell transplant (HSCT) recipients, for whom VSTs are derived from the HSCT donor. Autologous VSTs have also shown promise for the treatment of virus-driven malignancies outside the HSCT setting. In both cases, VSTs are manufactured as patient-specific products, and the time required for procurement, manufacture, and release testing precludes their use in acutely ill patients. Further, Good Manufacturing Practices–compliant products are expensive, and failures are common in virus-naive HSCT donors and patient-derived VSTs that are rendered anergic by immunosuppressive tumors. Hence, highly characterized, banked VSTs (B-VSTs) that can be used for multiple unrelated recipients are highly desirable. The major challenges facing B-VSTs result from the inevitable mismatches in the highly polymorphic and immunogenic human leukocyte antigens (HLA) that present internally processed antigens to the T-cell receptor, leading to the requirement for partial HLA matching between the B-VST and recipient. HLA mismatches lead to rapid rejection of allogeneic T-cell products and graft-versus-host disease induced by alloreactive T cells in the infusion product. Here, we summarize the clinical outcomes to date of trials of B-VSTs used for the treatment of viral infections and malignancies and their potential as a platform for chimeric antigen receptors targeting nonviral tumors. We will highlight the properties of VSTs that make them attractive off-the-shelf cell therapies, as well as the challenges that must be overcome before they can become mainstream.

Hematopoietic stem cell transplant (HSCT) is a curative option for patients with high-risk acute lymphoblastic leukemia (ALL), but relapse remains a major cause of treatment failure. To prevent disease relapse, we prepared and infused donor-derived multiple leukemia antigen–specific T cells (mLSTs) targeting PRAME, WT1, and survivin, which are leukemia-associated antigens frequently expressed in B- and T-ALL. Our goal was to maximize the graft-versus-leukemia effect while minimizing the risk of graft-versus-host disease (GVHD). We administered mLSTs (dose range, 0.5 × 107 to 2 × 107 cells per square meter) to 11 patients with ALL (8 pediatric, 3 adult), and observed no dose-limiting toxicity, acute GVHD or cytokine release syndrome. Six of 8 evaluable patients remained in long-term complete remission (median: 46.5 months; range, 9-51). In these individuals we detected an increased frequency of tumor-reactive T cells shortly after infusion, with activity against both targeted and nontargeted, known tumor-associated antigens, indicative of in vivo antigen spreading. By contrast, this in vivo amplification was absent in the 2 patients who experienced relapse. In summary, infusion of donor-derived mLSTs after allogeneic HSCT is feasible and safe and may contribute to disease control, as evidenced by in vivo tumor-directed T-cell expansion. Thus, this approach represents a promising strategy for preventing relapse in patients with ALL.