Acute lymphoblastic leukaemia (ALL) responds well to chemotherapy and the majority of children and a significant proportion of adults are cured of their disease after primary therapy. However, a number of patients relapse and allogeneic transplantation following conditioning with chemotherapy and radiotherapy offers the possibility of long-term survival in a proportion of these patients. A significant number of patients with ALL develop disease that is refractory to further therapy. The infusion of unmodified donor lymphocytes (DLI) following relapse after allogeneic transplantation has been shown to be curative in patients with chronic myeloid leukaemia (CML). However, in ALL the success rate is much lower. The results of in vitro and limited in vivo studies suggest that it may be possible to manipulate lymphocytes from the transplant donor to produce cytotoxic T-lymphocytes (CTL) with increased effectiveness in killing patients' ALL cells. This may be done in a number of ways. For example, some strategies utilise the patients dendritic cells (DC) to present tumour antigens to donor lymphocytes and convert them into CTL either by pulsing DC taken in remission with ALL cells or lysate, fusing such 'normal' DC with ALL cells or using DC cultured from the patient's ALL cells. Other approaches include exploiting the expression of leukaemia-specific antigens such as the proteinase PR-3 or the zinc finger transcription factor Wilms tumour-1 protein (WT-1) to stimulate CTL responses. Alternatively, immunotherapeutic strategies might exploit differences in minor histocompatibility antigens such as HA-1 and HA-2 between donor and recipient. These are expressed solely on haemopoietic cells making them suitable targets for donor derived anti-leukaemic cells. In vivo studies to date suggest that educated T-cells may have a role to play in the treatment of relapsed and refractory ALL in the future.