Phase I trial of intravenous thymidine and carboplatin in patients with advanced cancer

Purpose: To evaluate the biologic interactions and toxicities of carboplati n combined with a 24-hour infusion of thymidine 75 mg/m(2) in a phase I tri al. Patients and Methods: Thirty-two patients with cancer refractory to convent ional therapy were treated. The first set of patients (n = 7) received thym idine alone 4 weeks before subsequent planned courses of thymidine combined with carboplatin followed (4 weeks) by carboplatin alone. Carboplatin was administered over 20 minutes at hour 20 of the 24-hour thymidine infusion. The carboplatin dose wets escalated in patient groups: 200 mg/m(2) (n = 3); 300 mg/m(2) (n = 7); 350 mg/m(2) (n = 4); 400 mg/m(2) (n = 3): 480 mg/m(2) (n = 10); and 576 mg/m(2) (n = 5). At the maximum-tolerated dose (480 mg/m (2)), five patients received combined therapy first and carboplatin alone s econd, and five patients received carboplatin first and combined therapy se cond. Maintenance therapy for stable or responding patients was combined th erapy. Results: Evaluation demonstrated a trend toward thymidine protection of car boplatin-induced treatment-limiting thrombocytopenia. Neutropenia with carb oplatin alone or in combination was negligible. Thymidine alone had no myel osuppressive effects and produced reversible grade 1 or 2 nausea and vomiti ng (57%), headache (25%), and grade 1 neurotoxicity (22%). Thymidine did no t enhance expected carboplatin toxicities. There was no therapy-related inf ection or bleeding. Analysis of platinum in plasma ultrafiltrate and urine showed no effect by thymidine. Similarly, thymidine pharmacokinetics wets n ot affected by carboplatin. As predicted, nicotinamide adenine dinucleotide levels in peripheral lymphocytes were increased during exposure ta carbopl atin and/or thymidine but were decreased by carboplatin alone. In three pat ients with high-grade glioma, responses included one complete remission (21 months) and one partial remission (14 months) at the 480-mg/m(2)-dose leve l, and disease stabilization (7 months) at the 400-mg/m(2-dose) level. A mi nor response was observed in a patient with metastatic colon cancer (5 mont hs) at the 480-mg/m(2)-dose level. Conclusion: The combination of carboplatin and thymidine as described is we ll tolerated. The data presented have resulted in a phase II study by the N orth American Brain Tumor Consortium. (C) 1999 by American Society of Clini cal Oncology.