Pharmacogenomic and pharmacokinetic assessment of liposome encapsulated SN-38 (LE-SN38) in advanced cancer patients

2501 Background: The prodrug irinotecan is converted into its active moiety, SN-38, with wide interpatient variability, due in part to differential expression of cellular carboxylesterases responsible for prodrug hydrolysis. The enzyme UDP-glucuronosyltransferase 1A1 (UGT1A1) detoxifies SN-38 by metabolizing it to SN-38 glucuronide. Polymorphisms in the UGT1A1 promoter lead to decreased expression of the enzyme and are associated with increased risk of irinotecan toxicity. This Phase I study in patients with advanced cancer is designed to assess the pharmacogenomics, pharmacokinetics, and safety of liposome encapsulated SN-38 (LE-SN38). Methods: Patients are stratified prospectively according to their UGT1A1 genotype, as defined by the number of TA repeats in the A(TA)nTAA promoter sequence. Strata consist of homozygous wild-type, homozygous variant, and heterozygous patients, who are expected to have normal, low, and intermediate levels of glucuronidation activity, respectively. LE-SN38 is infused intravenously over 90 minutes every 21 days until disease progression or unacceptable toxicity occurs. Dose escalation is planned with separate patient cohorts receiving 2.5 to 90 mg/m2 of LE-SN38. Tumor progression is monitored radiographically after every 2 cycles. Results: As of 16 December 2003, genotype frequencies of 90 screened patients were 43% homozygous wild-type, 43% heterozygous, and 13% homozygous variant. Dose escalation for the 3 strata has reached 40 mg/m2, 30 mg/m2, and 20 mg/m2, respectively. Diarrhea and neutropenia have not been dose limiting. Drug pharmacokinetic parameters appeared to be linear over the doses assessed. At the 10 mg/m2 LE-SN38 dose level, greatest SN-38 systemic exposure occurred in homozygous variant patients: AUCs were 2- to 3-fold higher than for heterozygous or homozygous wild-type patients, respectively. Conclusions: Dose escalation and patient accrual in all strata are continuing. These data show that drug exposure varies across genotypes. Thus, pharmacogenomic assessments should be used prospectively as a component of individualized patient dosing decisions. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration NeoPharm, Inc.