To its proponents, the Human Genome Project (HGP) was the manifesto for a new era of biological research. The decade-long effort to produce the first comprehensive map of human genes would, they said, provide a reference manual – a "Book of Life" – for the workings of human cells. To skeptics, the project was more a piece of technological showmanship, a gaudy exercise in data-collection. The result would be not a useful tool, they claimed, but a costly frill, the scientific equivalent of a coffee-table book.
Today, a little more than 10 years after the completion of the first rough draft of the human genome, the impact of the project is something neither its advocates nor its detractors fully predicted. While many expected the HGP and its offshoots to spur rapid advances against inherited diseases, the true revolution has come in the understanding and treatment of genetic diseases that aren't, for the most part, inherited. The preeminent example of such diseases is cancer.
A person's genome is the full set of genetic material in his or her cells – a collection of DNA with six billion chemical units. While science has traditionally focused on genes one by one, genomics takes an ensemble approach to DNA, studying it in its entirety.
"When the Human Genome Project began in 1990, part of the motivation was to map the genome simply because it's there," says Dana-Farber's Matthew Meyerson, MD, PhD. "The project was billed as a way to increase our understanding of the structure of the genome – how genes are organized and controlled. Much of the emphasis, though, was actually on linking patterns of gene abnormalities to specific inherited diseases."
Research published just months after the unveiling of the genome map in June 2000 showed the potential of treating cancer by going after the genes responsible for the disease. In a study that gained national attention, scientists reported that a drug, now known as Gleevec, could send chronic myelogenous leukemia into remission by blocking a single malfunctioning gene. Deciphering the human genome would make it possible to start thinking about decoding the human cancer genome, a catalogue of all the gene abnormalities associated with cancer, and a guide to the most promising genetic targets for new therapies.
Still, it was far from obvious, a decade ago, that a genomic approach would be of much benefit in the fight against cancer. It was well known that damaged and deranged genes can cause cells to become cancerous. But did it make sense to use genomic techniques to scan them by the tens of thousands – or continue the tried-and-true method of focusing deeply on specific "suspect" genes to determine their connection to cancer?
"There was a great deal of skepticism about how helpful genomics would be," says Dana-Farber's Levi Garraway, MD, PhD. "There were very few drugs, at the time, that targeted particular abnormal genes. And the therapies that had been shown to be effective, like Gleevec, tended to work against ‘simple' cancers that arose from just one or relatively few mutated genes. There were many who doubted genomics would help against the majority of cancers, which are driven by much more complex repertories of gene mutations."
At Dana-Farber, a group of leaders and young scientists saw genomics as an opportunity for a major leap in the understanding of cancer and in its treatment. As early as 1999, the Institute was investing in "microarray" technology that could measure the activity of thousands of genes at a time. A few years later, Dana- Farber would partner with the Broad Institute of MIT and Harvard, which houses some of the most advanced machinery in the world for probing the human genome.
"The decision to support genomic research here – to acquire gene-scanning equipment and provide funding and lab space to scientists interested in the field – was a risk," says William Hahn, MD, PhD, Dana-Farber's deputy chief scientific officer and director of its Center for Cancer Genome Discovery. "It involved people heading in a new direction, with no assurance of success."
Today, ask nearly any cancer scientist about the impact of genomics and he or she may struggle to come up with a sufficient description.
"The whole vision of cancer therapy at Dana-Farber has been transformed," says Meyerson. Hahn believes it has "fundamentally changed how every cancer laboratory at the Institute works."
The map of the human genome has become the template for research efforts around the world. Prominent among them is the Cancer Genome Atlas project of the National Institutes of Health (in which several Dana-Farber scientists have leadership roles), which seeks to pinpoint cancer-related genes in several types of tumors by comparing DNA in normal and malignant tissues.
Genomics is changing the way many cancers are classified. Instead of identifying tumors by the organ of origin, researchers are starting to categorize them by the pattern of genetic abnormalities. Such data can help scientists gear treatments to each cancer's genetic signature.
"Genomics has also taught us that rare mutations – those which crop up in a small number of patients – can be powerful contributors to cancer," says Hahn. "And we've learned that cancers have many different malfunctioning genes, which explains why a treatment that works in one patient may not work in another. It's clearer than ever that combinations of therapies will be needed."
On the other hand, says Meyerson, "One of the surprises to come out of genomic research is that many different types of tumors have a large number of genetic alterations in common. A drug targeted at one alteration may actually be effective in a wide array of cancers." (A prime example is Gleevec, which is effective against a rare malignancy called gastrointestinal stromal tumor, as well as melanoma skin cancers that carry the same genetic alteration.)
Entirely new aspects of biology have been revealed by genomics. A prime example involves RNA, once thought of as simply a dutiful deputy of DNA, but which is in fact involved in everything from turning genes on and off to shutting down entire chromosomes. "No one knew microRNAs [switches for gene activity] existed or were important in gene regulation," Hahn remarks. "All the textbooks were wrong, because they were incomplete."
Genomics has had wide-ranging effects on cancer science, enabling researchers to more easily study human cancer tissue directly, rather than "working backward" from animal tissue. It has created explosive growth in computational biology, which helps analyze the oceans of data generated by genomic experiments. And, by its very nature, genomics has fostered collaboration among researchers.
Dana-Farber's Lynda Chin, MD, explains. "[Genomics] has encouraged team science on two levels. First, because it's a new field with novel techniques, you need input from a variety of people. Challenging one another, defending and explaining your ideas, has become part of the research process. At the same time, the sheer volume of data associated with genomics, and the nuances of its interpretation, requires input from people with different perspectives and skill sets."
If a single event heralded the potential of genomic approaches to cancer, it came in 2002, when British scientists using gene-sequencing techniques discovered that about 70 percent of malignant melanomas contain a mutation in the gene BRAF. The discovery held the tantalizing possibility that drugs targeting BRAF could be effective. "For scientists in my generation, that study came as an inspiration," Chin notes. "For people who had been considering sequencing, that study was the signal to really pursue it."
Perhaps the most fundamental breakthrough associated with genomics is a conceptual one, a shift in the way scientists understand and think about cancer. Being able to take a wide-angle view of the genes within tumors – to consider them as pieces in a mosaic – inevitably alters the questions researchers ask and the answers they reach. This comprehensive view brings scientists closer to what cancer actually is.
"If your tools only allow you to focus on individual genes, you're inevitably going to take a reductionistic view of the disease," Hahn says. "What you discover may be true, but it doesn't reflect the totality of what happens in real life."
"[The Lurie Center] has fundamentally changed how every cancer laboratory at the Institute works."– William Hahn, MD, PhD
Genomic research is also spurring dramatic advances in gene-sequencing technology, which reads the individual letters of cells' genetic code, making it possible to quickly and inexpensively identify gene abnormalities in tumor tissue. As recently as 2008, sequencing the genes in a single tumor sample took several months and equipment costing well over $1 million. Today, it can be done in two weeks at a cost of under $10,000 per sample. Further improvements promise to make even those statistics quickly obsolete.
"Advanced sequencing machinery is a classic example of disruptive technology, because it has completely transformed the way we conduct certain research," Garraway observes. "New techniques enable us to rapidly read and compare the entire genome of normal and cancerous tissues by identifying genetic differences including point mutations [misspellings of the genetic code], copy number alterations [in which there are too few or too many versions of individual genes], and rearrangements [in which genes swap places with each other].
"When the cost of sequencing fell sharply and we realized that it could be a practical research tool, it was almost as though the discovery clock had been reset," he continues. "Suddenly, you're covering frontiers no one has ever encountered. Today, it's a given that sequencing represents the way forward in cancer."
Even at this early stage, the itinerary is clear: In the Cancer Genome Atlas and other projects, scientists will continue tracking down the full set of genetic glitches associated with cancer – a project that, in Meyerson's words, "is not nearly done. There probably are many types of genes we don't know about yet."
In contrast to "pure science" techniques, genomics has an inherently practical streak: The very process of identifying genes linked to cancer produces targets for future therapies. It is in this area that some of the greatest challenges remain, says Chin. If a genomic survey of prostate cancer samples turns up, say, 500 abnormal genes, how can scientists determine which are causative – triggering and promoting tumor growth – and which are mere decoys? The answer is to determine the precise mechanism by which the genes contribute to cancer. Unfortunately, explains Chin, there is "no one way to rapidly ascertain the functions of large numbers of genes.
"Genes need to be understood in the context of their function within specific tissues and the body as a whole," she explains. "A gene involved in breast cancer may have no role in colon cancer. We need to consider not only how the gene functions within a cell, but how the cell interacts with its microenvironment [the sea of surrounding cells], and with each person's genetic makeup and bodily function."
Chin and her colleagues are working on a system to rank genes by the likelihood of their involvement in cancer. Such prioritizing would enable scientists to concentrate on the most promising opportunities. Dana- Farber has a clear advantage in this effort thanks, in part, to the Belfer Institute for Applied Cancer Science, which specializes in finding genetic alterations in cancer and identifying those which can best be targeted by new therapies. Chin serves as its scientific director. Also critical are Dana-Farber's Center for Cancer Genome Discovery, Center for Cancer Systems Biology, and Center for Molecular Oncologic Pathology.
"For all this progress, much remains to be done," Hahn says. "The lives of cancer patients depend on our work. The sense of urgency is always with us."
Paths of Progress Spring/Summer 2011 Table of Contents
Dana-Farber Cancer Institute, 450 Brookline Avenue, Boston, MA 02215 | Call us toll-free: