OBJECTIVE Preclinical and clinical data suggest that topotecan may be more effective, and perhaps less toxic, when administered as a continuous intravenous infusion (CIVI). A previous Cancer and Leukemia Group B (CALGB) trial of topotecan, given on a daily bolus schedule in combination with cisplatin, produced more hematologic toxicity than expected with either drug alone. Therefore, we designed this phase I trial to define the dose-limiting toxicities (DLTs) and the recommended phase II doses of cisplatin in combination with topotecan administered as a CIVI. Population pharmacodynamic models for the combination also were investigated. METHODS Patients with advanced solid tumors and a maximum of one prior chemotherapy regimen for metastatic disease were eligible if they had a performance status of 0 to 1 and adequate renal, hepatic, and bone marrow function. Prior treatment with camptothecins or platinum compounds and prior pelvic irradiation were not allowed. The initial schedule consisted of a fixed dose of topotecan 0.4 mg/m2/d administered as a CIVI for 21 days and escalating doses of cisplatin administered on days 1, 8, and 15 of a 28-day schedule, until the maximum tolerated dose (MTD) was achieved. After severe hematologic toxicity was observed in the first two patients, the topotecan infusion was shortened to 14 days, and the total dose of cisplatin was administered on day 1 in all subsequent patients. After the MTD was defined, that cohort was expanded to include a total of 12 assessable patients. Hematopoietic growth factors were not allowed. For the pharmacologic studies, total topotecan plasma concentrations were measured by high-pressure liquid chromatography (HPLC) during infusion on days 3, 8, and 11 on the first cycle, and the median steady-state concentration (Tss) was determined. Platinum plasma concentrations on day 3 were measured by atomic absorption spectrometry. RESULTS Of the 32 patients enrolled, 28 were assessable for toxicity and 24 for response. The primary toxicity was hematologic, with both neutropenia and thrombocytopenia being dose-limiting. The MTD of cisplatin was 75 mg/m2 on day 1 in combination with topotecan 0.4 mg/m2/d for 14 days. At this dose level, three of a total of 12 assessable patients had DLT. The pharmacodynamic relationship between Tss and the absolute neutrophil count at the nadir (ANCn) was described by the following equation: log10 (ANCn)=4.23 - 0.47 x Tss - 0.01 x cisplatin dose (P < .0001; R2=0.64). The substitution of platinum concentration for cisplatin dose in this model did not result in a significant improvement. Three patients had a partial response: one with duodenal carcinoma; a second with small-cell lung cancer; and a third with melanoma. CONCLUSIONS Cisplatin can be given safely in combination with CIVI topotecan. However, toxicity was still substantial. Based on the current results and our previous trial of this combination, we conclude that, when combined with cisplatin, CIVI topotecan does not seem to be advantageous compared with the more traditional daily bolus schedule.