Genome-wide genotype-based risk model for survival in core binding factor acute myeloid leukemia patients

Abstract

The present study attempted to build a single nucleotide polymorphism (SNP)-based risk model for predicting overall survival (OS) and event-free survival (EFS) in patients with core binding factor acute myeloid leukemia (CBF-AML). Adopting genome-wide SNP array using Affymetrix SNP array 6.0, we analyzed 868,157 SNPs with respect to OS and EFS in 104 patients with CBF-AML. Significant SNPs were identified from single SNP analysis. The risk model was constructed with incorporation of six SNPs and three clinical factors (age, c-kit exon 17 mutation, and LDH) for OS and six SNPs and three clinical factors (age, WBC, and LDH) for EFS. The model was further defined into low- and high-risk groups based on risk scores. The median age was 39 years, and the subgroup of t(8;21) and inv(16) or t(16;16) was assessed in 68 (65.4%) and 36 patients (34.6%). Finally, six SNPs per each OS (rs4353685, rs4908185, rs7709207, rs12034, rs1554844, and rs17241868) and EFS (rs13385610, rs11210617, rs11169282, rs7709207, rs4438401, and rs16894846) were incorporated into the risk model. OS was significantly different in favor of the low risk group (80.4 +/- 8.4%) compared to the high-risk group (22.0 +/- 7.3% at 3 years; p = 8.75 x 10(- 13); HR 8.67). For EFS, there was also a significant difference between the low- (75.0 +/- 5.8%) versus high-risk group (17.1 +/- 6.3% at 3 years; p = 5.95 x 10(- 13); HR 7.67). A genome-wide SNP-based risk model can stratify CBF-AML patients according to their OS and EFS in 104 patients.