Prospective analysis of UGT1A1 genotyping for predicting toxicities in advanced colorectal cancer (aCRC) treated with irinotecan (IRI)-based regimens: Interim safety analysis of a Japanese observational study.

Abstract

3535 Background: UGT1A1*6 and UGT1A1*28 are risk factors for severe IRI-related toxicities in Asians, but recommended IRI doses based on UGT1A1 genotypes and other risk factors are unclear. We conducted a prospective analysis to examine the correlation between UGT1A1 genotypes and the efficacy and safety of IRI-based regimens in Japanese aCRC patients (pts), (NCT 01039506). Methods: Pts who had histologically confirmed aCRC, PS of 0–2, received IRI-based regimens (FOLFIRI, IRI+S-1, IRI monotherapy), were UGT1A1 genotyped, and provided written informed consent were included. UGT1A1 polymorphisms were analyzed and categorized into 3 groups: wild (*1/*1), hetero (*1/*6, *1/*28), and homo (*6/*6, *6/*28, *28/*28). Detailed toxicities in the first 3 months of treatment were prospectively recorded. For interim safety analysis, incidences of grade 3–4 (severe) toxicities were compared among UGT1A1 genotypes and a logistic regression model was used to predict the risk of severe toxicities. Severe toxicities and associated risk factors were predicted using a nomogram and bootstrap validation was performed. Results: We enrolled 1376 pts between October 2009 and March 2012. At the time of abstract submission, toxicity data of 504 pts were available; 46% pts had wild, 44% hetero, and 11% homo polymorphisms. FOLFIRI was administered to 63% pts. Severe neutropenia developed during the first 3 months of treatment in 33% pts: 36% in hetero [OR, 1.5; 95% CI, 1.0–2.3], 47% in homo (OR, 2.3; 95% CI, 1.2–4.4), and 28% in wild. Severe diarrhea incidence was 5%, which did not correlate with UGT1A1 genotypes. Multiple logistic regression model included regimen, initial IRI dose, gender, age, UGT1A1 genotype, and PS as predictors of severe neutropenia in the first treatment cycle. The resulting nomogram demonstrated good accuracy in predicting severe neutropenia, with a bootstrap-corrected concordance index of 0.74. Conclusions: Considering UGT1A1 genotype along with other clinical factors is important for managing pts undergoing IRI-based regimens. Our presentation will provide analysis of data from more than 1000 pts. Clinical trial information: NCT01039506.