Whereas patients with multiple myeloma continue to relapse after autologous transplantation and are unlikely to be cured, the probability of progression is less after allogeneic transplantation and a proportion of patients may be cured. This is attributable to an immunologically mediated graft-versus-myeloma (GVM) effect which is akin to the well-known graft-versus-leukemia effect. The available clinical and experimental evidence strongly support the existence of GVM, but it is not known whether GVM is separable from graft-versus-host disease (GVHD) in practice. The best way to exploit GVM reactions is unclear, and the morbidity and mortality associated with GVHD undermine long-term survival. There is usually a time lag of a few weeks between immune intervention and disease response. There is a propensity for extramedullary disease recurrence in patients whose marrow disease is controlled with immunologic manipulation. Exploration of GVM outside conventional allogeneic transplantation or after autologous transplantation is necessary to increase the number of patients likely to benefit from this phenomenon and to make it safer. This article reviews the currently available literature on the subject.