NUDT15 codon 139 is the best pharmacogenetic marker for predicting thiopurine-induced severe adverse events in Japanese patients with inflammatory bowel disease: a multicenter study

• DOI: 10.1007/s00535-018-1486-7
• Publication Date: 2018-06-19T11:18:56Z
• Authors: Yoichi Kakuta; Yosuke Kawai; Daisuke Okamoto; Tetsuya Takagawa; Kentaro Ikeya; Hirotake Sakuraba; Atsushi Nishida; Shoko Nakagawa; Miki Miura; Takahiko Toyonaga; Kei Onodera; Masaru Shinozaki; Yoh Ishiguro; Shinta Mizuno; Masahiro Takahara; Shunichi Yanai; Ryota Hokari; Tomoo Nakagawa; Hiroshi Araki; Satoshi Motoya; Takeo Naito; Rintaro Moroi; Hisashi Shiga; Katsuya Endo; Taku Kobayashi; Makoto Naganuma; Sakiko Hiraoka; Takayuki Matsumoto; Shiro Nakamura; Hiroshi Nakase; Tadakazu Hisamatsu; Makoto Sasaki; Hiroyuki Hanai; Akira Andoh; Masao Nagasaki; Yoshitaka Kinouchi; Tooru Shimosegawa; Atsushi Masamune; Yasuo Suzuki
• Keywords: Thiopurine methyltransferase; Leukopenia; Genome-wide Association Study

Despite NUDT15 variants showing significant association with thiopurine-induced adverse events (AEs) in Asians, it remains unclear which of or whether additional genetic should be tested to predict AEs. To clarify the best pharmacogenetic test used clinically, we performed studies and haplotypes AEs, genome-wide study (GWAS) discover variants, ROC analysis select model severe Overall, 2630 patients inflammatory bowel disease (IBD) were enrolled genotyped for codon 139; 1291 treated thiopurines. diplotypes analyzed 970 patients, GWASs AEs 1221 using population-optimized genotyping array imputation. We confirmed p.Arg139Cys leukopenia alopecia (p = 2.20E−63, 1.32E−69, OR 6.59, 12.1, respectively), found a novel digestive symptoms 6.39E−04, 1.89). Time was significantly shorter, when diagnosed, thiopurine doses lower Arg/Cys Cys/Cys than Arg/Arg. In GWASs, no associated strong correlation frequency estimated enzyme activities based on (r2 0.926, p 0.0087), there differences AUCs from those 139 (AUC 0.916, 0.921, acute leukopenia, AUC 0.990, 0.991, alopecia, respectively). Genotyping sufficient Japanese IBD.