Cisar, Laura A.

3520 Background: This report examines UGT1A1*28, toxicity, and outcome in trial NCCTG N9741 patients (pts) receiving combination chemotherapy for advanced colorectal cancer. Methods: Blood samples were collected from 520 pts randomized between 2/18/00 and 7/18/02. Pts were treated with irinotecan/5-fluorouracil (5FU)/leucovorin (LV) (IFL, n=114), oxaliplatin/5FU/LV (FOLFOX, n=299), irinotecan/oxaliplatin (IROX, n=107). 19% of IFL pts received full-dose (125 mg/m2) irinotecan; the rest received a reduced dose (100 mg/m2). Toxicity was evaluated every cycle. Correlation of dichotomous outcomes and genotype status used Fisher’s exact tests. Univariate logistic regression was used to test the association between continuous variables and binary outcomes, including a test of trend, and multivariate logistic regression to adjust for known risk factors: performance status, age, gender. Time to progression (TTP) and overall survival (OS) were tested for association with UGT1A1 using the log-rank test. Statistical significance was prospectively set at .01 to control for multiple comparisons. Results: Frequency of UGT1A1 genotype was 44% 6/6, 42% 6/7, and 9% 7/7. In all arms, the rate of grade (G) 4 neutropenia (neut) ( Table ) and G3+ febrile neut (FN) was higher in pts with the 7/7 than in pts with 6/7 and 6/6. No association was seen between UGT1A1 and diarrhea, tumor response, TTP, or OS overall or for any treatment arm. The odds ratios for pts with 7/7 developing G4 neut and G3+ FN compared with pts with 6/6 are shown ( Table ). Statistically significant associations between the 7/7 polymorphism and G4 neut were seen in pts treated with IROX (P=.004) and in all pts (P=.007). The association between 7/7 and G3+ FN was also significant for IROX (P=.006). Conclusions: There is a higher risk of neut in 7/7 pts. Due to low frequency of the 7/7, exclusion of all pts with 7/7 reduces the overall G4 neut rate from 18% to 17%. UGT1A1 was not a predictor of response, TTP, or OS for any arm. [Table: see text] [Table: see text]

AbstractPurpose: To determine the recommended starting doses and pharmacokinetics of irinotecan in cancer patients with impaired liver function treated on a weekly schedule.Experimental Design: Patients with solid tumors who had impaired liver function were enrolled into four groups based on baseline serum total bilirubin and aspartate aminotransferase (AST)/alanine aminotransferase (ALT): Group 1 (n = 19): total bilirubin 1.5 to 3.0 × institutional upper limit of normal (IULN) and ALT/AST ≤5.0 × IULN; Group 2 (n = 7): total bilirubin 3.1 to 5.0 × IULN and ALT/AST ≤5.0 × IULN; Group 3 (n = 6): total bilirubin ≤1.5 × IULN and ALT/AST 5.1 to 20.0 × IULN; Group 4 (n = 10): total bilirubin 1.5 to 3.0 × IULN and ALT/AST 5.1 to 20.0 × IULN. Irinotecan was given as a 90-minute i.v. infusion weekly for the first 4 weeks in each 6-week cycle at starting doses which escalated from 40 to as much as 75 mg/m2. After the first treatment, doses were adjusted based on individual patient toxicities. Starting doses for patients with hepatic dysfunction were derived from the maximum tolerated doses noted in the four hepatic dysfunction groups.Results: Forty-two patients were treated. Among the most frequent adverse events were neutropenia (41%, grades 3/4), diarrhea (15%, grades 3/4), nausea (10%, grade 3), and vomiting (5%, grades 3/4). Two patients died from drug-induced neutropenic sepsis. Two patients had objective tumor responses (complete response, liver metastases from unknown primary; partial response, colon cancer). Hepatic dysfunction reduced irinotecan clearance while increasing relative exposure to the active metabolite, 7-ethyl-10-hydroxycamptothecin (SN-38). SN-38 exposures in patients receiving doses of 40 to 75 mg/m2 were comparable to exposures in patients with normal liver function treated with a starting dose of 125 mg/m2.Conclusions: Irinotecan starting doses that seem to be safe for hepatically impaired patients treated with the weekly schedule are 60, 50, 60, and 40 mg/m2 for groups 1 to 4, respectively. At these starting doses, exposure to SN-38 and the adverse event profile are similar to that observed in patients with normal liver function and antitumor activity can be observed.