Phase I clinical and pharmacokinetic study of mitoxantrone given to patients by intraperitoneal administration.

On the basis of its high degree of cytotoxicity against fresh human ovarian cancers and its relative lack of vesicant activity, mitoxantrone administered by the i.p. route was studied in a Phase I and pharmacokinetic trial. Thirty-three patients with good performance status and diagnoses of metastatic or recurrent ovarian (31 patients) and colon (two patients) cancers were treated with 12- to 38-mg/m2 doses, administered by the i.p. route every 4 wk for up to ten treatment courses. Mitoxantrone doses were escalated at 2- to 3-mg/m2 increments in groups of three to 11 patients. Thirty-eight mg/m2 (by i.p. dwell without removal) were considered the maximally tolerated dose in that, of eight treated patients, four experienced severe leukopenia and six experienced severe abdominal pain. Response to i.p. mitoxantrone was evaluable in 17 patients. None of seven patients with clinically measurable intraabdominal or pelvic tumor masses responded; however, in three (50%) of six patients with nonmeasurable disease, there was normalization of previously elevated serum CA-125 concentrations for 3, 17, and 24 mo. Additionally, two (50%) of four patients who underwent third-look laparotomies were found to have greater than 75% reductions in i.p. tumor masses with response lasting 24 and 25 mo. At 38 mg/m2, mitoxantrone was associated with a mean concentration.time product of 100 micrograms.h/ml in the i.p. space and of 0.071 micrograms.h/ml in plasma, yielding an i.p./plasma area under the curve ratio of 1408. We conclude that chemical peritonitis is the dose-limiting toxicity of i.p. administered mitoxantrone and that a dose of 23 mg/m2 every 3 to 4 wk should be used in future Phase II trials in ovarian cancer patients with minimal residual intraabdominal and pelvic disease following second-look laparotomy.