ABSTRACT
PURPOSE: We performed a Phase I and pharmacokinetic study to determine the maximum tolerated dose of irofulven (6-hydroxymethylacylfulvene; MGI-114, MGI PHARMA, Inc.), administered in intermittent weekly schedules in patients with advanced solid tumors. EXPERIMENTAL DESIGN: Three schedules were tested: A, days 1, 8, and 15 every 4 weeks; B, days 1 and 8 every 3 weeks; C, days 1 and 15 every 4 weeks. Drugs were administered as 5- and 30-min (schedules B and C) infusions. Dose levels of 10, 12, and 14 mg/m(2)/week were explored. RESULTS: Ninety-nine patients received 256 cycles. Fifteen of 74 patients evaluable for maximum tolerated dose experienced 16 dose-limiting toxicities (5 of 17 patients on schedule A, 2 of 25 on schedule B, and 8 of 32 on schedule C), principally treatment delay for thrombocytopenia. Schedule A was considered unsuitable because of frequent thrombocytopenia and treatment discontinuations. Twenty-three percent of the overall population (22 patients with grade 1-2, and 1 patient with grade 3), including 37% of patients on dose level 3, experienced unexpected dose-limiting visual toxicity, which included color perception and visual field alterations linked to retinal cone cell toxicity; the visual toxicity had an early onset, was mostly reversible, and was related to higher dose per infusion. Safety profiles were similar for 5- and 30-min infusions. The relationships between dose and area under the plasma concentration-time curve and maximum plasma concentration were linear for both 5- and 30-min infusions in the 78 patients evaluated for pharmacokinetics. The area under the plasma concentration-time curve and clearance were comparable between infusion durations. Responses included one complete (ovarian), one partial (renal), and seven disease stabilizations lasting >4 months. CONCLUSIONS: We recommend doses of 18 mg/m(2)/infusion for schedule B and 24 mg/m(2)/infusion for schedule C, limited to 0.55 mg/kg and a total dose of 50 mg/infusion, administered over 30-min.