The principal objectives of this study were to determine the feasibility of escalating doses of the hydrophilic topoisomerase I (topo I) inhibitor topotecan (TPT) as a 30-min infusion daily for 5 days in adults with refractory or relapsed acute leukemia and to study the pharmacokinetic behavior of high doses of TPT and pharmacodynamic determinants of TPT activity. Fourteen patients received 27 courses of TPT at doses ranging from 3.5 to 5.75 mg/m2/day every 3 weeks. A constellation of unusual adverse effects, consisting of high fever, rigors, precipitous anemia, and hyperbilirubinemia, was the principal dose-limiting toxicity of high doses of TPT on this schedule. These toxicities were consistently intolerable at the 5.75 mg/m2/day dose level; however, they were neither severe nor common at lower doses. Although the precise etiology of these effects is not known, high doses of TPT may induce acute hemolytic reactions in this patient population. Severe, albeit transient, mucositis was experienced by two of eight patients in 2 of 17 courses at the next lower dose level, 4.5 mg/m2/day, which was determined to be the maximum tolerated dose and the dose recommended for further trials. The pharmacokinetic behavior of TPT at high doses was not dose dependent and resembled that at lower doses. In view of preclinical data suggesting that TPT sensitivity might correlate with topo I levels, topo I content in leukemia blasts was assessed by Western blotting. Variations in topo I content were observed. Moreover, strong correlations were evident between topo I content and two markers of proliferation, proliferating cell nuclear antigen and nuclear protein B23, raising the possibility that differences in topo I content observed among various leukemia specimens might reflect differences in the proliferating fractions of cells in various leukemia samples. Although complete clearance of circulating leukemia blasts occurred in most courses, neither sustained responses nor hematopoietic recovery were observed in the heavily pretreated, poor-risk patients enrolled in this study, and it was not possible to correlate these differences in topo I content with clinical response. These results indicate that substantial dose escalation of TPT as a 30-minute infusion for a 5-day schedule above myelosuppressive doses is feasible in adults with refractory or relapsed leukemias; however, further development of alternate high-dose schedules in leukemia may be warranted in view of the nature of the dose-limiting toxicity and the lack of sustained clinical responses in this preliminary investigation.