Inherited coding variants at the CDKN2A locus influence susceptibility to acute lymphoblastic leukaemia in children.

Huan Xu ,Richard Williams,Kjeld Schmiegelow,Yu Liu,Mignon L Loh,Charles G Mullighan,Jinghui Zhang,Eric Larsen,Rachita Yadav,Meenakshi Devidas,William L Carroll,Jun J Yang,Stephen P Hunger,Wenjian Yang,W Paul Bowman,Maoxiang Qian,Philip J Lupo,Alanna C Morrison,Ramneek Gupta,Xujie Zhao,William E Evans,Hui Zhang,Torben Hansen,Mary V Relling,Jens Christian Holm ,Elaine R Mardis ,Ching Hon Pui ,Virginia Pérez-Andreu ,Geoffrey Neale ,Julie M Gastier‐Foster ,Robert S Fulton ,Naomi J Winick ,Elizabeth A Raetz

doi:10.1038/ncomms8553

Abstract

There is increasing evidence from genome-wide association studies for a strong inherited genetic basis of susceptibility to acute lymphoblastic leukaemia (ALL) in children, yet the effects of protein-coding variants on ALL risk have not been systematically evaluated. Here we show a missense variant in CDKN2A associated with the development of ALL at genome-wide significance (rs3731249, P=9.4 × 10−23, odds ratio=2.23). Functional studies indicate that this hypomorphic variant results in reduced tumour suppressor function of p16INK4A, increases the susceptibility to leukaemic transformation of haematopoietic progenitor cells, and is preferentially retained in ALL tumour cells. Resequencing the CDKN2A–CDKN2B locus in 2,407 childhood ALL cases reveals 19 additional putative functional germline variants. These results provide direct functional evidence for the influence of inherited genetic variation on ALL risk, highlighting the important and complex roles of CDKN2A–CDKN2B tumour suppressors in leukaemogenesis.

Highlights

There is increasing evidence from genome-wide association studies for a strong inherited genetic basis of susceptibility to acute lymphoblastic leukaemia (ALL) in children, yet the effects of protein-coding variants on ALL risk have not been systematically evaluated
No coding variants in ARID5B and IKZF1 were significantly associated with ALL susceptibility
Previous genome-wide association studies (GWAS) had identified an intronic variant in CDKN2A to be strongly associated with susceptibility to ALL in populations of European descent[9]

Summary

Introduction

There is increasing evidence from genome-wide association studies for a strong inherited genetic basis of susceptibility to acute lymphoblastic leukaemia (ALL) in children, yet the effects of protein-coding variants on ALL risk have not been systematically evaluated. The risk of developing acute lymphoblastic leukaemia (ALL) is highest between 2 and 5 years after birth[1,2], with initiating sentinel somatic genomic lesions (for example, chromosomal translocations) detectable at the time of birth in many cases[3,4] This early disease onset suggests a strong inherited genetic basis for ALL susceptibility, and recent genome-wide association studies (GWAS) have discovered at least six risk loci: ARID5B, IKZF1, CEBPE, PIP4K2A-BMI1, GATA3 and CDKN2A–CDKN2B5–10. These ALL risk genes are directly involved in haematopoietic stem cell function, lymphocyte differentiation and development, and cell cycle regulation[11,12,13,14,15], several of which are commonly targeted by somatic genomic lesions. We perform an exome-focused GWAS to systematically examine the impact of germline-coding variants on the development of ALL in children of European descent, experimentally explore the functional consequences of the genome-wide significant variant in the CDKN2A gene, and comprehensively characterize coding variation at this locus by targeted resequencing

Methods

Results

Conclusion