Abstract
Introduction
In recent years, whole-exome sequencing (WES) and whole-genome sequencing (WGS) have identified over 60 recurrently mutated genes in both coding and non-coding regions in chronic lymphocytic leukemia (CLL) patients. However, these studies were restricted to Europeans, and the genomic landscape of CLL in Asians remains less-well defined. To better understand the genetic basis of CLL in Asians, we used target gene sequencing to define the mutational status of 63 recurrently mutated genes in CLL and other small B-cell lymphomas in a cohort of 93 Chinese patients with CLL.
Methods
A total of 93 CLL patients were retrospectively enrolled. Genomic DNA was isolated from cryopreserved mononuclear cells or bone marrow aspirate specimens. Targeted gene sequencing was performed on HiSeq Sequencer (Illumina, Inc, San Diego, CA) following the manufacturer's standard sequencing protocols. The sequencing reads were aligned to hg19 using the Burrows-Wheeler Aligner (BWA). SNV calling was performed using both GATK and Varscan programs, and the called SNV data were then combined. The variants at a frequency of 0.1% or more in the 1000 genomes database or 1000g (CHB and CHS) (http://www.1000genomes.org), esp6500 (http://evs.gs.washington.edu/EVS) or ExAC03 database (http://exac.broadinstitute.org) were removed from subsequent analysis.
Results
A total of 195 driver mutations across 50 genes were identified in 80 patients (86.0%). 57 cases (61.3%) harbored mutations involving at least two genes. The most frequently mutated genes included IGLL5 (21/93, 22.6%), MYD88 (15/93, 16.1%), LRP1B (11/93, 11.8%), TP53 (9/93, 9.6%), NOTCH1 (8/93, 8.6%), ATM (7/93, 7.5%), CHD2 (7/93, 7.5%) and KMT2D (7/93, 7.5%). Only 4 cases with SF3B1 mutations (4/93, 4.3%) were identified in our cohort (Figure 1). By comparing the results from the study by Landau et al, our results showed that CLL in Chinese displayed significantly different mutational profiles from those in Caucasians (Table 1).
Cases with complex karyotype (CK) had higher frequencies of TP53 (7/16 [44%] vs 2/65 [3%], p <0.0001) and EGR2 mutations (3/16 [19%] vs 0/65 [0%], p=0.0066), as compared to cases without CK . By Kaplan-Meier analysis, we found that shorter time-to-first-treatment (TTT) was significantly associated with TP53 (p <0.0001), EGR2 (p=0.0019), KMT2D (p=0.047) and DDX3X mutations (p=0.0232). Multivariate analysis of TTT revealed that only TP53 mutation (hazards ratio[HR]: 2.945, 95%CI: 1.065-8.147; p=0.037) and advanced Binet stage (HR:4.464, 95%CI:2.037-9.784; p=0.0002) independently predicted inferior TTT (Table 2). TP53 (p <0.0001), SF3B1 (p=0.017), NOTCH1 (p=0.012), EGR2 (p <0.0001) and DDX3X mutations (p=0.007) predicted reduced overall survival. Multivariate Cox analysis revealed that TP53 mutation (HR: 20.361, 95%CI: 3.220-128.753, p=0.001), EGR2 mutation (HR: 15.497, 95%CI: 1.639-146.514; p=0.017) and SF3B1 mutation (HR: 20.230, 95%CI: 2.337-175.085, p=0.006) independently predicted unfavorable overall survival.
Conclusions
We found that Chinese CLL displayed a distinct mutational landscape from that of Caucasians. Our findings suggest different mechanisms involved in the pathogenesis of Chinese CLL. Besides, our results support that targeted gene sequencing can provide important values in the prognostification of CLL. Targeted gene sequencing of a large cohort and whole exome/genome sequencing of CLL cases are needed to better define the genomic landscape of CLL in Chinese.
No relevant conflicts of interest to declare.
