Tallimustine, an effective antileukemic agent in a severe combined immunodeficient mouse model of adult myelogenous leukemia, induces remissions in a phase I study.

Despite progress in leukemia therapy, only 20-30% of patients with acute myelogenous leukemia (AML) are cured. 1-beta-D-arabinofuranosylcytosine- and topoisomerase II-reactive drugs are the primary therapeutic agents used. The aim of this study was to evaluate the potential activity of tallimustine in leukemia. In this study, we first investigated the efficacy and toxic effects of tallimustine, a distamycin-A derivative, in a human leukemia model in severe combined immunodeficient (SCID) mice. On the basis of its dramatic activity in this preclinical study, a Phase I study of tallimustine at a starting dose of 300 microgram/m2/day for 3 days every 3-4 weeks was conducted in patients with refractory or relapsed leukemia. In SCID mice grafted with a human myelomonocytic leukemia cell line, tallimustine resulted in complete remission of disease in most mice at tolerable dosages ranging from 0.86 to 3.0 mg/kg/day for 3 days and was combined effectively and safely with a 2-day schedule of high-dose ara-C. In the Phase I study, 26 patients with refractory or relapsed leukemia were treated. The maximum tolerated dose was 900 microgram/m2/day for 3 days every 3-4 weeks. This dose was 3 times higher than the maximum tolerated dose in solid tumors and was limited by severe mucositis. Magnesium and potassium wasting were also observed, but other side effects (fatigue and gastrointestinal) were minor. Two (8%) patients with AML achieved complete remission and two achieved hematological improvement with persistent thrombocytopenia. The results of this study indicate that tallimustine has promising activity in AML. Future studies may combine tallimustine with other agents known to be active against AML, and investigate its activity in other hematological malignancies. The recommended Phase II single-agent dose of tallimustine is 750-900 microgram/m2/day for 3 days, and combination studies may start at 50-66% of this dose schedule. The SCID mouse model of human leukemia may be promising in the preclinical evaluation and selection of potential antileukemic agents.