Multicentre phase II study and pharmacokinetic analysis of irinotecan in chemotherapy-na?ve patients with glioblastoma.

BACKGROUND: To assess the antitumour activity and safety profile of irinotecan and its pharmacokinetic interactions with anticonvulsants in patients with glioblastoma multiforme. PATIENTS AND METHODS: This multicentre phase II and pharmacokinetic study investigated the effects of irinotecan 350 mg/m(2) given as a 90-min infusion every 3 weeks either prior to (group A) or after relapse following radiotherapy (group B) in chemotherapy-na?ve patients with glioblastoma. Preferred concomitant medication for seizure prevention was valproic acid. Pharmacokinetic analysis of irinotecan and its main metabolites (SN-38, SN-38-G, APC and NPC) was performed during cycle 1. An independent panel of experts reviewed the activity data. RESULTS: Fifty-two patients (25 patients in group A and 27 patients in group B) received a total of 191 cycles of irinotecan. Forty-six patients (22 patients in group A and 24 patients in group B) were evaluable and externally reviewed for activity. According to external review, one partial response (group B), seven minor responses (three in group A and four in group B), 12 disease stabilisations (seven in group A and five in group B) were observed. This resulted in an overall response rate of only 2.2% (95% confidence interval 0.2% to 6.5%). The median time to tumour progression was 9 weeks in group A and 14.4 weeks in group B. Six-month progression-free survival rates were 26% in group A and 43% in group B. Grade 3-4 toxicities (percentage of patients in groups A and B) consisted of neutropenia (12.5% and 25.9%), diarrhoea (8.3% and 7.4%), asthenia (12.5% and 7.4%) and vomiting (0% and 7.4%). The clearance of irinotecan was 12.4 and 14.4 l/h/m(2) in two patients who received no anticonvulsant. In patients receiving valproic acid, the clearance of irinotecan was 17.2 +/- 4.4 l/h/m(2). CONCLUSIONS: Irinotecan given at the dose of 350 mg/m(2) every 3 weeks has limited clinical activity as a single agent in patients with newly diagnosed and recurrent glioblastoma after radiotherapy. The toxicity profile and plasma disposition of irinotecan and SN-38 were not strongly influenced by anticonvulsant valproic acid therapy. Although the response rate of irinotecan as a single agent was limited, it remains an attractive drug for combination studies in patients with glioblastoma.