COLD SPRING HARBOR, NEW YORK — At last count, the international research community had sequenced close to 4,000 individual cancer samples, with plans in the works to study at least 21,000 more tumor genomes by the end of 2014. These genomic data sets, which span some 50 different cancer types covering nearly every organ in the body, have opened the door to a number of new targeted treatment options by identifying major genetic changes that distinguish cancer cells from normal cells. To date, however, almost all cancer genomes have come from bulk tumor samples containing many different cells, and this cellular lumping can mask some of the fine-grained complexity of the disease that underpins much of drug resistance and susceptibility.

“The best way to begin studying populations of cancer cells is to look at what happens in single cells,” says Timour Baslan, a geneticist at the Cold Spring Harbor Laboratory (CSHL) in Long Island, New York, who published a protocol last month for how to analyze copy number variation of single cancer cells (Nat. Protoc. 7, 1024–1041, 2012).

To enable cancer genomics at the individual tumor cell level, in December 2010 Life Technologies, based in Carlsbad, California, announced a $1 million prize to the first team that could sequence the entire genome and RNA content of a single cancer cell. And now, many researchers are getting close to that goal.

At the CSHL Meeting on The Biology of Genomes here last month, Nicholas Navin, a geneticist at the MD Anderson Cancer Center in Houston, presented new data on the genetics of breast cancer at single-cell resolution. Previously, Navin had looked at copy number variants in 100 individual cells in each of two human breast cancer cases and showed that one such tumor contained three distinct subpopulations of cells, whereas the other cancer was made up of a genetically identical pool of cells (Nature 472, 90–94, 2011). Now, Navin has honed his technology to make it sensitive enough to detect single nucleotide differences, and he has found in a set of four cells from a single invasive breast cancer tumor that the individual cells often contain dozens of rare mutations that typically go undetected by bulk tumor-sequencing methods.

“It's my hope that these new tools can be used to improve diagnosis and treatment,” Navin says.

“They could become very powerful at detecting remaining cancer cells that can linger in the body for months or years,” as well as guide the appropriate course of adjuvant or chemotherapy, he notes.

Of a single mind

Meanwhile, Baslan and others at CSHL scanned for copy number variation in100 cells each from 12 breast cancer tumors both before and after chemotherapy. Reporting at the meeting on the first three such tumors, the researchers found that some people's cancers had gained genetic material after the treatment, whereas other tumors showed signs of chromosomal rearrangements. Although such analyses are possible with bulk sequencing, “this type of evaluation has been traditionally overlooked in cancer studies,” says Baslan.

Baslan and his colleagues started out by studying copy number variation because this method requires less starting material than is needed for full sequencing—and single cells don't have much DNA to begin with. In March, however, scientists from the BGI genomics center in Shenzhen, China reported a way to amplify the DNA three-million-fold, creating enough starting material to decode the entire protein-coding portion of the genome down to the single nucleotide level for individual cancer cells (Cell 148, 873–885, 2012).

Future single-cell sequencing projects should also become easier thanks to the creation last month of a new single-cell genomics center at the Broad Institute in Cambridge, Massachusetts, with equipment and staff support from Fluidigm Corporation, a microfluidics technology company based in South San Francisco. According to Fluidigm's president and chief executive Gajus Worthington, the new center will begin by focusing on cancer stem cells and circulating tumor cells. “We hope that studying these cells will lead to a new type of biopsy that identifies from the start whether a tumor will spread or is drug resistant,” he says.