Abaji Rachid, Gagné Vincent, Xu Chang Jiang, Spinella Jean-François, Ceppi Francesco, Laverdière Caroline, Leclerc Jean-Marie, Sallan Stephen E, Neuberg Donna, Kutok Jeffery L, Silverman Lewis B, Sinnett Daniel, Krajinovic Maja
Research Center, Centre Hospitalier Universitaire Sainte-Justine, Montreal, QC, Canada.
Department of Pharmacology, University of Montreal, Montreal, QC, Canada.
Oncotarget. 2017 Jul 4;8(27):43752-43767. doi: 10.18632/oncotarget.17959.
Allergy, pancreatitis and thrombosis are common side-effects of childhood acute lymphoblastic leukemia (ALL) treatment that are associated with the use of asparaginase (ASNase), a key component in most ALL treatment protocols. Starting with predicted functional germline variants obtained through whole-exome sequencing (WES) data of the Quebec childhood ALL cohort we performed exome-wide association studies with ASNase-related toxicities. A subset of top-ranking variants was further confirmed by genotyping (N=302) followed by validation in an independent replication group (N=282); except for thrombosis which was not available for that dataset. SNPs in 12 genes were associated with ASNase complications in discovery cohort including 3 that were associated with allergy, 3 with pancreatitis and 6 with thrombosis. The risk was further increased through combined SNPs effect (p≤0.002), suggesting synergistic interactions between the SNPs identified in each of the studied toxicities. Interestingly, rs3809849 in the MYBBP1A gene was associated with allergy (p= 0.0006), pancreatitis (p=0.002), thrombosis (p=0.02), event-free survival (p=0.02) and overall survival (p=0.003). Furthermore, rs11556218 in IL16 and rs34708521 in SPEF2 were both associated with thrombosis (p=0.01 and p=0.03, respectively) and pancreatitis (p=0.02). The association of SNPs in MYBBP1A, SPEF2 and IL16 geneswith pancreatitis was replicated in the validation cohort (p ≤0.05) as well as in combined cohort (p=0.0003, p=0.008 and p=0.02, respectively). The synergistic effect of combining risk loci had the highest power to predict the development of pancreatitis in both cohorts and was further potentiated in the combined cohort (p=1x10-8).The present work demonstrates that using WES data is a successful "hypothesis-free" strategy for identifying significant genetic markers modulating the effect of the treatment in childhood ALL.
过敏、胰腺炎和血栓形成是儿童急性淋巴细胞白血病(ALL)治疗的常见副作用,与天冬酰胺酶(ASNase)的使用有关,ASNase是大多数ALL治疗方案的关键组成部分。我们从通过魁北克儿童ALL队列的全外显子组测序(WES)数据获得的预测功能性种系变体开始,对与ASNase相关毒性进行全外显子组关联研究。通过基因分型(N = 302)进一步确认了一组排名靠前的变体,随后在一个独立的复制组(N = 282)中进行验证;血栓形成情况因该数据集无法获取。发现队列中有12个基因的单核苷酸多态性(SNP)与ASNase并发症相关,其中3个与过敏相关,3个与胰腺炎相关,6个与血栓形成相关。通过组合SNP效应,风险进一步增加(p≤0.002),表明在每种研究的毒性中鉴定出的SNP之间存在协同相互作用。有趣的是,MYBBP1A基因中的rs3809849与过敏(p = 0.0006)、胰腺炎(p = 0.002)、血栓形成(p = 0.02)、无事件生存期(p = 0.02)和总生存期(p = 0.003)相关。此外,IL16中的rs11556218和SPEF2中的rs34708521均与血栓形成(分别为p = 0.01和p = 0.03)和胰腺炎(p = 0.02)相关。MYBBP1A、SPEF2和IL16基因中的SNP与胰腺炎的关联在验证队列(p≤0.05)以及合并队列(分别为p = 0.0003、p = 0.008和p = 0.02)中得到复制。组合风险位点的协同效应在两个队列中预测胰腺炎发生的能力最强,在合并队列中进一步增强(p = 1×10 - 8)。本研究表明,使用WES数据是一种成功的“无假设”策略,可用于识别调节儿童ALL治疗效果的重要遗传标记。