Collaborators at Duke and the University of North Carolina at Chapel Hill are leading the charge in brain cancer research by focusing on personalized medicine.
The research is being conducted by Donald Lo—director of the Center for Drug Discovery and associate professor in neurobiology at the University—and Albert Baldwin, associate director of basic research at the UNC Lineberger Comprehensive Cancer Center. With support from the Clinical and Translational Science Awards, the duo has worked on combatting glioblastoma multiforme, an aggressive form of brain cancer.
In the past few years, Lo said oncologists have found that treating patients with only one drug is ineffective and often leads to relapse. One of the reasons might be that very few tumors are made up of only one kind of cell, which makes decisions about therapy extremely complex.
“The [term] 'multiforme' in 'glioblastoma multiforme' goes back to histologists in the 19th century, who saw that cells look different in different parts of the tumor," Lo said. “If they look different, they probably act different, and they respond to drugs differently. This is why 80 percent of [glioblastoma multiforme] victims pass away in about a year.”
Lo and Baldwin said that traditional brain cancer research implants parts of tumors from human patients into mice that have been genetically modified to have no immune system. But that approach has its shortcomings, Lo explained.
“Although you can model the entire organism, you can’t really see what’s going on," Lo said. "You can’t really see how the cells in question are reacting to therapy or to each other."
The two developed a new technique—they implanted the human tumors into real brain slices grown in the lab. The tumors can then be studied, but in an environment that lets the researchers see how the tumor changes and becomes more complex over time.
One of their most pressing questions was how a single tumor can have different cells, referred to as "tumor heterogeneity."
“Tumors are notorious for not being able to repair DNA very well," Baldwin said. "So you get a new mutation, which might lead to a new branch of the tumor—and you can get many of these. It becomes a problem therapeutically if you’re trying to treat a tumor."
To track such DNA mutations, the team also uses genome sequencing—a technique that has gained traction as a way to identify different tumor regions. Lo said they hoped to find "smoking guns," or genetic trends that could be linked to drug resistance.
Lo said that initial research on tumor heterogeneity suggests that combining drugs will actually prove to be most beneficial for patients. Looking to the future, Lo and Baldwin noted that their approach could eventually be translated to help design treatment plans for individual patients.
“Interestingly, a combination therapy might not mean taking two different drugs at once," Lo said. "What we might find is that taking 'drug A' actually unmasks heterogeneity in a way that you don’t want to treat with 'drug B' at the same time. [Instead], you might want to wait to give drug B in sequence."
However, Lo added that cocktails of cancer drugs can become toxic very quickly, which means that oncologists' options are much narrower than they would otherwise be.
Ideally, Baldwin explained, brain cancer patients would first undergo surgery to remove their tumor, and then that tumor would be grown and observed again using the new technique. Drug combinations would then be tested in the petri dish, which could help oncologists identify the best therapies for a particular patient.
“Personalized and precise methods” are the ultimate goal, Baldwin said.
Get The Chronicle straight to your inbox
Signup for our weekly newsletter. Cancel at any time.