Various clinical observations have implicated T cells in the control of chronic myeloid leukemia (CML). These observations have in recent years been supported by laboratory results indicating the presence of CML-specific T cells in the lymphocyte repertoire of both normal healthy individuals and disease-bearing patients. Both MHC-unrestricted and MHC-restricted immune effector mechanisms are involved. Donor lymphocyte infusion has produced encouraging GvL effects. However, future adoptive immunotherapy may depend on the isolation and generation of leukemia-specific T cells. Although many proteins may potentially act as leukemia antigens in CML for MHC-restricted cytotoxicity, the bcr-abl fusion protein has been most extensively investigated. There is now much evidence to suggest that the bcr-abl junctional peptides are capable of eliciting both CD4 and CD8 responses in normal healthy donors and CML patients. Furthermore, the T-cell lines generated react with autologous or HLA-matched fresh CML cells, suggesting that the bcr-abl fusion protein can be processed in vivo so that the joining segment is bound to HLA molecules in a configuration and concentration similar to those of the immunizing peptide for antigen recognition by the antigen-specific T-cell receptor. These results also indicate that the bcr-abl junctional peptides may be used for immunotherapy of CML. Other strategies available for immunotherapy of CML include immunologically or genetically manipulated donor T-cell infusion, the use of cytokines, adoptive immunotherapy with leukemia-reactive T-cells expanded ex vivo, and immune gene therapy. Novel and rational immunotherapy may therefore play an important adjuvant role in future in the management of patients with CML. Am.