Introduction

Chronic lymphocytic leukemia with elevated genomic complexity (CLL-HGC) is clinically aggressive and is characterized by shortened survival. While it is known that CLL-HGC is enriched for specific aCNAs and that a subset of CLL-HGC carries TP53 mutations, it is currently unclear what other molecular aberrations cause or contribute to genomic instability in CLL with wild type TP53.

Methods

Within a cohort of 255 CLL cases previously analyzed on SNP 6.0 arrays, we identified 50 CLL-HGC cases with ≥ 3 acquired genomic copy number aberrations (aCNA) and/or acquired uniparental disomy (aUPD). Of these 50 CLL-HGC cases, 26 cases were wild type for TP53, and of these, 23 were subjected to whole exome sequence analysis (WES). Exome capture and WES was performed on DNA isolated from FACS-purified CD19+ and CD3+ cells, and massively parallel sequencing was performed using 96 bp paired-end sequencing on a HiSeq2000 sequencer. Bioinformatics analysis followed validated in-house pipelines. The following genes and exons were re-sequenced in all samples using Sanger sequencing: TP53 (exons 2-10), SF3B1 (exons 13-17), NOTCH1 (exon 34), and POT1 (all coding exons). Genetic data were complemented with assays for p53 protein expression and inducibility (following Nutlin 3 treatment of purified CLL cells), radiation-induced CLL cell apoptosis and ATM Ser-1981 auto-phosphorylation following CLL cell irradiation.

Results

Sanger sequence validation of all nominated candidate gene mutations in paired samples (T+N) confirmed 192 mutated genes in 23 CLL cases, a mutational load per case that is comparable to published unselected CLL cohorts. A gene mutated at high frequency, akin to TP53, was not identified. Recurrently mutated genes (N=2 or 3 out of 23 cases) in CLL-HGC included MYD88, NCKAP5, EGR2 and NXF1. Mutations in other genes previously suggested to contribute to genomic instability or CLL clinical aggressiveness were largely absent (ATM: N=1; NOTCH1: N=0; SF3B1: N=1 and POT1: N=0). A detailed gene-by-gene review of the mutated genes revealed that some of the genes can be grouped into functional classes that may have relevance to the observed genomic phenotype: nuclear export (XPO, NXF1), apoptosis regulation (BAX, KHDC1, PACS2, FBXW7), RB-E2F-p53 axis (DYRK1A, TRIM16, RB1CC1, E2F7), signaling (MYD88, EGR2), chromosome segregation (BSDC1, DDX46, PDIA4, BOD1L, ZW10) and p53 network (IRF2, SERTAD4, SYVN1, BAX). Notably absent were mutations in other DNA-ds-break response and repair or DNA maintenance pathway genes. Functional data uncovered 3 CLL-HGC cases with impaired p53 protein induction following chemical p53 activation and two cases with impaired ATM-Ser-1981 auto-phosphorylation.

Conclusion

We describe the results of WES in 23 CLL samples with high genomic complexity and wild type TP53. Overall, the following conclusions can be supported from this work: i) CLL-HGC with wild type TP53 is not associated with a high-frequency mutated gene; ii) the gene mutation load in CLL-HGC with wild type TP53 is similar to unselected CLL cases; iii) most CLL-HGC with wild type TP53 carry mutations in genes that fall into functional classes that may have a role in CLL genome destabilization or a permissive role in the survival of CLL cells with spontaneously occurring DNA lesions. Additional functional studies are in progress testing specific CLL-HGC-associated mutated alleles for effects on chromosomal stability and the p53 network. Overall, the results point to a multi-factorial origin of genomic instability in CLL.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution