An evidence-based analysis of the effect of busulfan, hydroxyurea, interferon, and allogeneic bone marrow transplantation in treating the chronic phase of chronic myeloid leukemia: developed for the American Society of Hematology.

Because there are differing opinions regarding treatment of patients in the chronic phase of chronic myeloid leukemia (CML), the American Society of Hematology convened an expert panel to review and document evidence-based benefits and harms of treatment of CML with busulfan (BUS), hydroxyurea (HU), recombinant interferon-alpha (rIFN-alpha), and bone marrow transplantation (BMT). The primary measure for defining efficacy was survival. Analysis indicated a survival advantage for HU over BUS. Observational studies of rIFN-alpha suffer from numerous biases including sample size, variations in study populations, definitions of hematologic and cytogenetic remissions, and dose. That rIFN-alpha is more efficacious than chemotherapy is demonstrated by 6 prospective randomized trials. For patients with favorable clinical features in chronic phase, compared to HU and BUS, rIFN-alpha improves survival by a median of about 20 months. Most evidence suggests that rIFN-alpha is most effective when combined with other drugs and when given during the earliest stage of the chronic phase. Adding cytarabine to rIFN-alpha adds further survival benefit but increases toxicity. Limitations for evaluating the long-term benefits of allogeneic BMT include the retrospective nature of most studies, incomplete documentation of the clinical characteristics of the patients, paucity of the details on patient selection, lack of control groups, and limitations of survival calculations. Survival curves for BMT show that at least half of the patients transplanted remain alive 5 to 10 years after treatment, whereas similar curves for rIFN-alpha show a continuous relapse rate over time with the curves crossing at about 7 to 8 years. Estimates of long-term survival may be confounded by the selection biases mentioned and the analytic methods used. The magnitude of the incremental increase in benefit with BMT must be weighed against the potential serious harm and death that may accompany the procedure in the short term. The best results with BMT have been obtained when it is performed within 1 to 2 years from diagnosis. Since each treatment option involves tradeoffs between benefit and harm, patient choice must be based on the examination of facts presented in an unbiased fashion. Newly diagnosed younger patients and older patients who are candidates for BMT should also be offered information about IFN-based regimens, the tradeoffs involved, and, if possible, share in the treatment decision. Hopefully this analysis will provide the stimulus for evaluation of other important aspects of CML.