Genetic variation in coding region of Aurora kinase A gene leads to cancer susceptibility

Abstract

Aurora kinase A (AURKA) gene, located on the frequently amplified chromosome 20q13 locus in human cancers, belongs to an evolutionarily conserved gene family and encodes a critical mitosis regulatory oncoprotein that is over expressed in a wide variety of human tumors [1]. In this issue of Cancer Epidemiology, Xu et al. performed a systematic and comprehensive metaanalysis to probe and unravel the true effect of a putatively functional AURKA rs2273535 SNP in cancer etiology using data from fifty-two published case–control studies with a total number of 74,896 subjects in the last decade [2]. The functional single nucleotide polymorphism (SNP), rs2273535 (T91A) leads to substitution of the amino acid phenylalanine with isoleucine at codon 31; and this allele ‘‘A’’ variant of AURKA was earlier identified as a low penetrance cancer susceptibility allele for skin tumor in mice and for multiple cancer types in humans associated with increased aneuploidy in colon tumors and cell transformation in vitro [3]. At the cellular level, AURKA amplification leads to deregulation of mitosis including defects in centrosome maturation and spindle assembly and correlates with activation of oncogenic as well as inactivation of multiple tumor suppressor pathways involving c-Myc, TP53, TP73 and BRCA1 [4]. Although several association analyses have reported correlation of allele ‘‘A’’ with different types of malignancies; a number of studies also failed to detect association of rs2273535 with cancers [2]. Epidemiological studies addressing the predisposing candidate genetic variations for complex phenotypes such as cancers may have limitations of power reducing the likelihood of detecting a significant association under certain parameters, and explanatory capability due to the variance of phenotype [5]. Approaches such as meta-analyses are useful for addressing these issues as these analyses increase the power by pooling the data from small scale studies and help identify the source(s) of heterogeneity as well as potential biases. Despite the conflicting results of individual association studies, this meta-analysis by Xu and colleagues [2]