NEW YORK (GenomeWeb News) – In a paper scheduled to appear online this week in Proceedings of the National Academy of Sciences, researchers from Washington University and elsewhere identified hundreds of copy number alterations in several dozen acute myeloid leukemia tumors.
The team used SNP arrays and array comparative genomic hybridization to detect and verify copy number changes in normal and tumor samples from 86 adults with AML. They found that 40 percent of the tumors contained copy number alterations, or CNAs, missed by cytogenetic approaches alone. On average, though, AML genomes contained fewer copy number changes than expected — just two or so per genome.
Clear copy number patterns for AML remain somewhat illusive, lead author Matthew Walter, an oncology researcher at Washington University, told GenomeWeb Daily News, given the heterogeneity of the disease.
Nevertheless, he and his co-authors say the approach is helping to pinpoint genomic changes important for AML pathogenesis. For example, more than three quarters of the CNAs detected in the new study affected at least one known cancer-related gene. As such, they explained, copy number studies will likely continue to complement ongoing AML genome sequencing studies.
"[T]he use of an unbiased high-resolution genomic screen identified many genes not previously implicated in AML that may be relevant for pathogenesis, along with many known oncogenes and tumor suppressor genes," Walter and his co-authors wrote.
The researchers used SNP arrays to find copy number alterations, followed by array CGH to validate these findings. First, they isolated DNA from tumor and normal skin samples from 86 individuals with AML who were at least 18 years old. They then evaluated DNA from these paired, matched samples at 1.85 million loci using the Affymetrix Genome-Wide Human SNP Array 6.0. Roche NimbleGen custom CGH 12 x 135K arrays were used to verify CNAs.
Initially, the team expected to find many recurrent changes in AML genomes, Walter said. So far, though, that hasn't been the case. Although they detected 201 acquired CNAs — ranging in size from about 35,000 bases to 250 million bases — each genome contained just 2.3 CNAs, on average.
"Because there are only a small number of CNAs in most AML genomes, we hypothesize that the genes in these regions may be important for AML pathogenesis (i.e., they are drivers rather than random passenger mutations)," the authors wrote.
Indeed, almost all of the copy number changes detected affected at least one protein-coding gene. And more than three-quarters contained a gene implicated in cancer or AML-related syndromes in the past.
Overall, the team found 51 recurring CNAs smaller than five million bases and three smaller than a few hundred thousand bases. Again, many of these copy number changes involved known cancer or AML-related genes.
Many recurrent changes were found in just a few individuals. But some seem to be more common. For instance, five or more patients had CNAs affecting the neurofibromin 1 gene NF1 on chromosome 17 or the ETS transcription factor gene ETS2 on chromosome 21 — both of which have been implicated in myeloid disease in the past. Individuals carrying these chromosome 17 or 21 changes also tended to have worse outcomes than those who didn't, the authors noted.
In contrast, the team found that the overall number of copy number changes in each genome did not provide new information for predicting patient outcomes over and above what can be gleaned from cytogenetic evaluation of the samples. In general, those with more complex cytogenetic profiles tend to have worse outcomes, Walter explained.
The work also uncovered rearrangements in AML genomes, including a gene fusion between the nucleoporin gene NUP98 on chromosome 11 and the histone methyltransferase gene NSD1 on chromosome 5 that has been identified in some samples in the past. After sifting through another 179 AML samples, the researchers found two more samples containing either a NUP98-NSD1 fusion or an NSD1-NUP98 fusion.
In the future, the team plans to continue doing copy number analyses. But since there seem to be relatively few copy number changes per genome, Walter explained, these studies will likely be done in conjunction with next-generation sequencing studies that identify nucleotide changes.
In addition, Walter said, examining more — and more cytogenetically complex — AML tumors could reveal additional copy number change patterns, particularly since the current study focused on genomes with two or less cytogenetic changes.
"As more AML genomes have their definitive acquired (somatic) CNAs catalogued, it is likely that specific CNAs will contribute to the growing list of genetic alterations that impact treatment choices and outcomes of patients with AML" the authors concluded.