NEW YORK (GenomeWeb News) – New research is providing an integrated look at the copy number and sequence changes in breast and colorectal cancers, suggesting the cancers are more complex than once imagined.
A team of researchers from Johns Hopkins University and elsewhere used SNP analysis and digital karyotyping to characterize copy number variations and sequence alterations in breast and colorectal cancer cell lines pinpointing, on average, more than a dozen alterations per tumor. The work, which appears online this week in the Proceedings of the National Academy of Sciences, also provides clues about the genes and pathways affected by these changes.
“These analyses provide an integrated view of copy number and sequencing alterations on a genome-wide scale and identify genes and pathways that could prove useful for cancer diagnosis and therapy,” senior authors Kenneth Kinzler, Bert Vogelstein, and Victor Velculescu, researchers at Johns Hopkins University, and their colleagues wrote.
The team previously sequenced 11 breast and 11 colorectal cancers. And in a set of papers published earlier this year, Kinzler, Vogelstein, Velculescu and others assessed the sequence alterations, copy number variations, and gene expression changes associated with pancreatic cancer and glioblastoma multiforme.
For the latest study, the team integrated SNP and digital karyotyping data on copy number alterations with sequence data from 18,191 genes from the RefSeq database to begin rounding out their view of these cancers.
“[T]he alterations detected by sequencing represent only one category of genetic change that occurs in human cancer,” the authors wrote. “A comprehensive picture of genetic alterations in human cancer should therefore include the integration of sequence-based alterations together with copy number gains and losses.”
In order to optimize their copy number assessment protocol, the researchers first used digital karyotyping on 18 colorectal tumor samples to find amplifications and homozygous deletions. By comparing these results with those obtained using high-density Illumina microarrays, the team designed a method for specifically detecting amplifications that led to more than a dozen copies of a gene and homozygous deletions, in which both copies of a gene are deleted.
Next, the researchers assessed 45 breast and 36 colorectal tumor samples at either 317,000 or 550,000 SNPs using Illumina high density SNP arrays. After weeding out alterations that were present in 23 matched healthy control samples or in multiple samples, they were left with 614 amplifications and 463 homozygous deletions believed to be cancer-specific somatic changes.
The majority of these alterations — 68 percent of the homozygous deletions and 81 percent of the amplifications — occurred in breast cancer samples. On average, breast tumors contained 18 copy number alterations (seven homozygous deletions and 11 amplifications) affecting 24 protein-coding genes, while the colorectal tumors tested had seven (four deletions and three amplifications) affecting nine protein-coding genes.
Using statistical approaches, the researchers integrated the new copy number information with sequence data. They also used Illumina arrays to look at copy number changes in the 11 breast and 11 colorectal tumors samples described in previous sequence studies.
From there, the team focused in on the genes and pathways that were most often affected in these cancers. That led them to some genes that were already known to have a role in cancer — including MYC and EGFR — as well as new candidates.
In particular, genes involved in intracellular signaling, cell cycle transitions, cell-cell interactions, and DNA topological control pathways tended to be affected by copy number and/or sequence changes. Interestingly though, the researchers detected both amplifications and deletions in the same pathways in some cases, suggesting that there can be consequences to tinkering with signal regulation, regardless of the direction.
Combining copy number and sequence data also holds promise for determining whether particular point mutations have a functional effect, the researchers noted. For example, if a gene turns up with a deletion in one sample and a point mutation in another, it could indicate that that point mutation is inactivating.
In the future, the researchers noted, incorporating information on other genome-wide changes such as translocations and epigenetic changes could provide even greater insight into cancer, as will trying to determine the timing with which genetic alterations occur in cells.
But there are already clues that could be useful for treating the cancers, the researchers pointed out. For instance, they found that two-thirds of the breast and colorectal samples tested in the study contain alterations to four key signaling pathways, suggesting that drugs targeting these pathways could prove useful for treating both breast and colorectal cancers.
And since several breast cancer samples tested contained changes to DNA topological pathways, they added, some of these tumors may be candidates for topoisomerase-targeted therapies.