Background: Second-generation BCR-ABL tyrosine kinase inhibitors (2nd generation TKIs) including nilotinib (NIL) and radotinib (RAD) are approved for the treatment of newly diagnosed and other TKI failed chronic myeloid leukemia (CML). Uridine diphosphate glucuronosyltransferase 1A1 (UGT1A1) gene promoter polymorphism, distinct number of TA repeats in promoter, has been reported for an association with hyperbilirubinemia on nilotinib therapy (Singer et al. Leukemia (2007) 21, 2311–2315). However, the distribution of UGT1A1 (TA)7 repeat polymorphism differs greatly between Caucasians and Japanese, namely, the frequency of UGT1A1(TA)7 repeat polymorphism is high in Caucasians, whereas it is low in Asians (Beutler et al. Proc. Natl. Acad. Sci. USA 95 (1998) 95, 8170–8174 ). Aims: The aim of this study was to investigate the association between UGT1A1 gene promoter polymorphism and hyperbilirubinemia in Korean CML patients treated with NIL and RAD as a frontline therapy. Methods: A total of 88 newly diagnosed chronic phase CML patients who treated with frontline NIL (n=39) and RAD (n=49) was screened for UGT1A1 promoter polymorphism genotype analysis. We used the High Pure PCR Template Preparation Kit (Roche, Germany) to prepare genomic DNA from whole blood and genotyped by direct sequencing of the 253- to 255-bp fragments produced by PCR. Results: The (TA)6/(TA)6 homozygote and (TA)6/(TA)7 heterozygote were seen in all genotyped population and frequency of (TA)6/(TA)6 homozygote was 79.5% (70/88) in our patients. (TA)6/(TA)6 homozygote predominated with 81.6% of the alleles in the RAD group and 76.9% in the NIL group. Relative riskfor each genotype presented in Table 1, with hyperbilirubinemia defined as CTCAE grade 2 or greater observed at any post-baseline point. The relative risks for 6/6 genotype versus 6/7 genotype was 3.5 with 95% CIs of (0.6, 18.9) in the RAD group compared with 8.1 (1.5, 43.8) in the NIL group and NIL group showed significantly high association with UGT1A1 gene promoter polymorphism (P Conclusions: This finding suggests that UGT1A1 gene promoter polymorphism may be an important determinant of hyperbilirubinemia in CML patients with NIL therapy, but not in RAD. However other mechanisms should be explored in patients who have significant hyperbilirubinemia with RAD therapy. Updated data with longer follow-up duration will be presented in the meeting. Disclosures No relevant conflicts of interest to declare.