Haematopoietic stem cell-supported myeloablative therapy appears to be a promising treatment modality for MM. It yields increased overall response and CR rates when compared with conventional chemotherapy, and seems to prolong the duration of survival. These conclusions, while encouraging, have mostly been drawn from uncontrolled studies carried out in select groups of patients and, obviously, need to be confirmed in controlled clinical trials. While the results of these studies are still awaited, the wider application of BMT should probably be encouraged. As in other malignancies, the best candidates appear to be those less heavily pre-treated, with chemosensitive disease, low tumour cell mass and other favourable prognostic features, for example low beta 2-microglobulin levels. Under these conditions, the chance of entering CR following BMT, be it autologous or allogeneic, is currently estimated to be of at least 50%, and the long-term probability of survival averages approximately 30%. However, while it appears that a plateau for progression-free survival cannot be ascertained following a single ABMT, reported observations of patients with continued disease-free survivals up to and beyond 10 years after allografting suggest that this latter procedure may be curative. It seems likely therefore that the higher mortality related to allogeneic BMT (which in recent years decreased from 50% to approximately 30% as the results of a better selection of patients and increasing experience in their management) may be offset by more durable control of the disease and cure in a certain proportion of patients. Recurrence of underlying malignant disease remains, however, a major problem and is the most common cause of treatment failure following BMT. For this reason, attempts to improve the impact of transplantation are warranted. Options currently under investigation include the development of more effective conditioning regimens, as applied in double auto-transplant or with targeted therapy using antibody-radionuclide conjugates, as well as post-transplant immunomodulation with either IFN-alpha, interleukin-2 or idiotype vaccines. In addition, many other problems regarding BMT for MM are still unresolved and could be properly addressed only in future clinical trials. The most important of these issues include the choice of the best conditioning regimen and the optimal source of haematopoietic stem cells, the nature of relapse after autografting, the need to purge or not to purge the autologous marrow, the existence of a 'graft-versus-myeloma' effect and its role in prolonging the duration of disease control and, finally, the likelihood of cure, especially for patients with molecularly-defined CR.