Oncolytic vaccinia virus (VV) therapy has shown promise in preclinical models and in clinical studies. However, complete responses have rarely been observed. This lack of efficacy is most likely due to suboptimal virus spread through the tumor resulting in limited tumor cell destruction. We reasoned that redirecting T cells to the tumor has the potential to improve the antitumor activity of oncolytic VVs. We, therefore, constructed a VV encoding a secretory bispecific T-cell engager consisting of two single- chain variable fragments specific for CD3 and the tumor cell surface antigen EphA2 (EphA2-T-cell engager-armed VV (EphA2-TEA-VV)). In vitro, EphA2-TEA-VV's ability to replicate and induce oncolysis was similar to that of unmodified virus. However, only tumor cells infected with EphA2-TEA-VV induced T-cell activation as judged by the secretion of interferon-γ and interleukin-2. In coculture assays, EphA2-TEA-VV not only killed infected tumor cells, but in the presence of T cells, it also induced bystander killing of noninfected tumor cells. In vivo, EphA2-TEA-VV plus T cells had potent antitumor activity in comparison with control VV plus T cells in a lung cancer xenograft model. Thus, arming oncolytic VVs with T-cell engagers may represent a promising approach to improve oncolytic virus therapy.