Abstract

Untreated and minimally pretreated solid tumor patients received alternating sequences of taxol and cisplatin. Sequential dose escalation of each agent using taxol doses of 110 or 135 mg/m2 and cisplatin doses of 50 or 75 mg/m2 resulted in four dosage permutations that induced grades 3 and 4 neutropenia in 72% to 84% and 50% to 53% of courses, respectively. Neutropenia was brief, and hospitalization for neutropenia and fever was required in 13% to 24% of courses. However, further escalation of taxol to 170 or 200 mg/m2 induced grade 4 neutropenia in 79% to 82% of courses. At the highest taxol-cisplatin dose level (200 mg/m2 to 75 mg/m2), the mean neutrophil count nadir was 98/microL, and hospitalization for neutropenia and fever was required in 64% of courses. The sequence of cisplatin before taxol, which has less antitumor activity in vitro, induced more profound neutropenia than the alternate sequence. Pharmacologic studies indicated that this difference was probably due to 25% lower taxol clearance rates when cisplatin preceded taxol. Although neurotoxicity was initially thought to be a potentially serious effect of the combination, mild to modest neurotoxicity occurred in only 27% of patients. Adverse effects also included myalgias, alopecia, vomiting, diarrhea, bradycardia, and asymptomatic ventricular tachycardia. Objective responses were noted in melanoma, as well as non-small-cell lung, ovarian, breast, head and neck, colon, and pancreatic carcinomas. Based on these results, the sequence of taxol before cisplatin at doses of 135 and 75 mg/m2, respectively, is recommended for phase II/III trials, with escalation of taxol to 170 mg/m2 if treatment is well tolerated.