The largest genomic studies on chronic lymphocytic leukemia to date confirm the genetic heterogeneity of the disease, reveal dozens of putative new driver mutations, and trace the evolution of tumors during treatment and relapse.
Two studies published in Nature provide the most extensive genomic characterization of chronic lymphocytic leukemia (CLL) to date. Results of whole-genome and whole-exome sequencing on hundreds of patients' tumors confirm the extreme genetic heterogeneity of CLL, reveal dozens of new driver mutations, and trace the evolution of tumors during treatment and relapse.
In one study, researchers conducted whole-exome sequencing on 278 CLL and matched germline DNA samples from patients in a phase III clinical trial. To increase power for gene discovery, they combined that data with two previously sequenced cohorts for a total of 538 tumor samples. The analysis revealed 44 putative cancer driver genes, including 26 newly recognized genes. In addition, researchers constructed a network model that describes the acquisition of mutations as the disease progresses. Their results point to pathways affected in CLL that could be targeted, including MAP kinase signaling and MYC activity.
“The further we go along with genome-level sequencing of CLLs, we begin to appreciate that for such a heterogeneous disease, larger cohorts matter,” says senior author Catherine J. Wu, MD, of Dana-Farber Cancer Institute in Boston, MA.
The clinical trial participants had a median follow-up time of 6 years, which allowed Wu's group to confirm previously identified genetic contributors to outcomes and uncover one new gene, RPS15, that affects progression-free survival. To look for markers of treatment response, Wu compared tumor samples taken before and after treatment from a subset of 59 subjects and found that relapse was associated with a diverse spectrum of genetic events.
The research is notable for providing genomic analysis on a large number of samples from a stringently controlled phase III trial rich with patient data, says Anna Schuh, MD, PhD, head of the Molecular Diagnostics Center at the University of Oxford, UK. “It's so important to begin to mine these very well-curated clinical cohorts and to begin to put the genomic profiling together with clinical outcome data,” says Schuh, who was not involved in the study.
In the second study, International Cancer Genome Consortium researchers analyzed matched tumor and normal samples from 506 patients with CLL or monoclonal B-cell lymphocytosis, a CLL precursor. In all, they performed whole-genome sequencing on 150 samples and whole-exome sequencing on 440. Their analysis identified 36 genes that were recurrently mutated in CLL, including 12 not previously linked to the disease. There was overlap in the gene sets discovered in the two studies, but also differences.
Importantly, the work uncovered novel mutations outside of the coding regions of genes. The most frequent—present in 13 of 506 cases—was a mutation in the 3′ promoter region of NOTCH1, which caused mRNA splicing errors and was associated with more aggressive disease. NOTCH1 is the most commonly mutated gene in CLL, so the results suggest that previous exome-sequencing studies may underestimate its contribution given that 20% of tumors containing NOTCH1 mutations contain mutations in the 3′ noncoding region. The researchers were able to identify these mutations through whole-genome sequencing analysis.
Mutations were also found in an enhancer region for the B cell–specific transcription factor PAX5, which has not previously been implicated in CLL.
The discovery of these noncoding mutations “demonstrates the need to explore the so-called dark side of the genome to detect new alterations implicated in cancer development,” says senior author Carlos López-Otín, PhD, of the Universidad de Oviedo in Spain.
- ©2015 American Association for Cancer Research.