Physicians and scientists look for ways to use genetics to improve cancer care
Few people would describe cancer as an orphan disease—a rare illness that’s ignored by drug companies because producing and marketing medications to treat it likely won’t turn a profit. But University of Minnesota cancer geneticist David Largaespada, Ph.D., argues otherwise.
“In truth, all cancer is an orphan disease because even a common cancer like breast cancer is composed of many different subtypes,” he says.
It’s becoming increasingly apparent that cancer has many subtypes and that each person’s body responds differently to a given treatment. Just as no two people are alike, it’s possible that no two tumors are exactly alike, either.
“One person’s pancreatic tumor may be very different than another person’s pancreatic tumor,” says the Masonic Cancer Center’s Brian Van Ness, Ph.D. “Some people’s tumors are very aggressive, and some people’s tumors are not. Some people’s tumors respond to therapy, and some people’s tumors do not.”
But why? Likely, Van Ness says, there are genetic variations in the tumor that will affect the way it progresses—or doesn’t. Genetic variations also influence how a drug is distributed throughout the body and how it’s metabolized, meaning that patients who are given the same amount of a medication may process the drug differently; as a result, different doses would be effective in different people.
“The ultimate goal,” he says, “is to best define the right drug for the right person at the right dose.”
That mantra also describes a recent and widespread research focus on “personalizing” medicine. When the Human Genome Project was completed in 2003 after more than a decade of analysis and $1.5 billion, many scientists had hoped for a wealth of new knowledge detailing which genes were responsible for which diseases.
What they got was a wealth of complex scientific data that needed to be decoded. After all, the human genome consists of about 3 billion chemical base pairs that make up our DNA and contains more than 20,000 genes.
And today many scientists believe that disease may not be the result of a single genetic flaw but instead is likely the product of multiple genes interacting—along with influences from a person’s environment.
Yes, it’s complicated, agrees Largaespada, the Masonic Cancer Center’s associate director for basic science. But it’s a researcher’s reality.
“We used to think about the magic bullet that would cure cancer, and then everyone realized that, well, each type of cancer is different; the biology is different,” he says. “That complexity doesn’t mean we should give up—it just [explains] why it’s taken a long time and why it’s hard.”
Researchers are seeing signs of progress as new therapies show dramatic successes in certain patient subgroups, Largaespada adds, affirming that personalizing cancer care is indeed worth the effort. Now scientists just need to find out why that’s happening so they can help target those treatments to the patients who will benefit most from them.
Matching cancer genes to therapies
Largaespada’s lab, for instance, is using a tool called “Sleeping Beauty”—essentially a piece of DNA that researchers can make “jump” from one section in the chromosomes of cells to another—to identify genes linked to cancer.
Working with genetically modified mice that model human cancers, Largaespada and his team have identified dozens of new cancer genes with the aid of Sleeping Beauty. So far his lab has reported on finding 77 genes tied to colorectal cancer—only seven of which were previously known—and 17 genes tied to liver cancer.
“Now we have a lot of unpublished data on new cancer genes for the [gastrointestinal] tract and liver [and] also for brain tumors, sarcomas, and several other types of cancer,” Largaespada says. “It has really exploded in the lab. We used to be a lab that worked just on leukemia, and because we’ve developed this technology, it’s allowed us to study all of these different types of cancer.”
Identifying these cancer-causing genes and pathways is a big step, and these discoveries signal great progress. But now Largaespada is focused on what those alterations mean. Will the genetic profile of a given tumor determine how it responds to different therapies? Or does it affect a cancer’s likelihood of spreading or recurring?
Ultimately, Largaespada hopes that understanding these alterations might help physicians match a certain drug or combination of drugs to a patient whose tumor has a specific genetic makeup. His lab team is now testing whether Sleeping Beauty-induced genetically defined tumors in tissue cultures respond to a host of existing drugs and medications in clinical trials as well as a few preclinical compounds. The most promising drug combinations will be tested in mouse models.
“The interesting thing, I think, is that the cure for some cancers is probably already out there,” he says. “We just have to figure out what tumor and what patient to give it to and in what combinations. The chemists have done a fantastic job [of] coming up with more and more chemicals, and we need to apply these in a logical way.”
A different theme
A popular approach to drug therapy, especially chemotherapy for cancer, says Van Ness, head of the Institute of Human Genetics’ Division of Medical Genomics, has been to give patients the maximum tolerated dose of a medication to treat their disease. But wouldn’t it be better if the standard approach instead was to give patients the smallest possible effective dose so as to produce minimal side effects?
The concept gets “a lot of head nodding,” Van Ness says. “It makes sense.”
And Van Ness does see a lot of head nodding when he speaks to patient support groups about his research, which is focused on how genetics influences responses to drugs for multiple myeloma, and what aspects of drug development and testing are most important to patients.
Personalized medicine may seem like a futuristic concept, but on some fronts, Van Ness says, the research is already there. The genes labeled BRCA1 and BRCA2 have been clearly linked to breast and ovarian cancer, for example, and researchers have discovered tumor-specific factors that cause some tumors to grow faster than others.
Optimizing transplant success
Other cancer breakthroughs may be just around the corner.
With colleagues Jeffrey Miller, M.D., and Sarah Cooley, M.D., University oncologist Daniel Weisdorf, M.D., is leading a clinical trial that’s showing how a relatively simple genetic screening process can lead to lower relapse rates after blood and marrow transplants in people who have acute myeloid leukemia (AML). The trial is being conducted as part of a $13.3 million grant Miller received recently from the National Cancer Institute.
For nearly the past two decades, a person needing a blood and marrow transplant has been paired with the donor who had the best tissue match based on the human leukocyte antigen (HLA) and a few other criteria. But since HLA typing was recognized as being important to transplant success in the 1990s, Weisdorf says, there haven’t really been further advances in donor selection.
Transplanted blood and marrow are critical to carrying out an immune attack on any cancer cells remaining after chemotherapy and radiation. As Weisdorf and his group looked into the factors that were important in that attack, they found that AML patients whose blood and marrow donors had certain patterns of genes that regulate natural killer cells—specialized blood cells that target and destroy cancerous or virus-infected cells—had the best posttransplant outcomes.
“Using donors who had these specific sets of genes, which control some of the function of natural killer cells, could lead to about 10 to 15 percent less risk of relapse of the leukemia after the transplant,” Weisdorf says. “For the standard-risk population, that cut the relapse rate in half. And even for those who had more advanced disease, it still reduced their risk of relapse.”
These results were specific to AML patients, Weisdorf adds. There was not even a hint of the same improvement for people who had acute lymphoblastic leukemia.
This fall the University is launching another phase of this clinical trial, serving as the lead center in the national study.
Weisdorf believes that, with a relatively simple and quick analysis of blood samples from a few potential HLA-matched donors, University researchers could help clinicians choose the best blood and marrow donors for their patients, thereby decreasing the risk of relapse.
It’s simply being smarter about donor selection, he says. And it’s a clinically relevant way to bring genetics to patients’ bedsides.
“We’re not changing the way the patients are treated,” he says. “They’re going to have their transplant like they always did, but maybe with a better donor.”
By Nicole Endres