Genome‐wide genotype‐based risk model for survival in acute myeloid leukaemia patients with normal karyotype. (30th July 2013)

• Record Type: Journal Article
• Authors: Choi, Hangseok; Jung, Chulwon; Sohn, Sang Kyun; Kim, Seonwoo; Kim, Hyeoung‐Joon; Kim, Yeo‐Kyeoung; Kim, TaeHyung; Zhang, Zhaolei; Shin, Eun‐Soon; Lee, Jong‐Eun; Moon, Joon Ho; Kim, Sung Hyun; Kim, Kyoung Ha; Mun, Yeung‐Chul; Kim, Hawk; Park, Jinny; Kim, Jhingook; Kim, Dennis (D. H.)
• Is Part Of: British journal of haematology. Volume 163:Number 1(2013:Oct.)
• Page Start: 62
• Page End: 71
• Publication Date: 2013-07-30
• Subjects: acute myeloid leukaemia -- normal karyotype -- genome wide single nucleotide polymorphism array; Hematology -- Periodicals; Blood -- Diseases -- Periodicals; 616.15
• DOI: 10.1111/bjh.12492
• Languages: English
• ISSNs: 0007-1048

Abstract

Summary: Single nucleotide polymorphisms (SNP) are inter‐individual genetic variations that could explain inter‐individual differences of response/survival to chemotherapy. The present study was performed to build up a risk model for survival in 247 patients with acute myeloid leukaemia (AML) with normal karyotype (AML‐NK). Genome‐wide Affymetrix SNP array 6.0 was used for genotyping in discovery set ( n = 118). After identifying significant SNPs for overall survival (OS) in single SNP analysis, a risk model was constructed. Out of 632 957 autosomal SNPs analysed, finally four SNPs (rs2826063, rs12791420, rs11623492 and rs2575369) were introduced into the risk model. The model could stratify the patients according to their OS in discovery set ( P = 1·053656 × 10 −4 ). Replication was performed using Sequenom platform for genotyping in the validation cohort ( n = 129). The model incorporated with clinical and four SNP risk score was successfully replicated in a validation set ( P = 5·38206 × 10 −3 ). The integration of four SNPs and clinical factors into the risk model showed higher area under the curve (AUC) reults than in the model incorporating only clinical or only four SNPs, suggesting improved prognostic stratification power by combination of four SNPs and clinical factors. In conclusion, a genome‐wide SNP‐based risk model in 247 patients with AML‐NK can identify a group of high risk patients with poor survival.