When the DNA "instruction manual" of life becomes damaged – as it does in human cells tens of thousands of times a day – bad things can happen. The cell may die, cause an inherited disease, or grow out of control to form a tumor. To prevent genetic catastrophe, cells have evolved a molecular toolkit for repairing their own DNA to restore normal behavior, duplication, and controlled growth.
A cell sensing genetic injury often grinds to a halt. It puts growth and division on hold and, like a ship in dry dock, undergoes repairs before setting sail again. The cell's repair kit – specialized enzymes and other proteins – can snip out and replace broken stretches of DNA, correct spelling errors in the genetic code, or simply bypass the damaged areas. Self-repair is vital to life and health, because DNA is constantly under attack from internal and external forces, including tobacco smoke, sunlight, environmental toxins, and dietary byproducts.
But when a patient has cancer, the repair team becomes the enemy, since most treatments attack tumors by smashing their cells' DNA. In response, tumor cells activate repair forces. Physicians, in turn, resort to giving higher doses of therapy to overcome the resistance, but this raises the risk of toxic side effects.
Clearly, new strategies are called for. Dana-Farber/Brigham and Women's Cancer Center's Alan D'Andrea, MD, and colleagues are developing ways to throw a wrench into cancer's DNA repair systems, making tumors more vulnerable to therapy. They're measuring "profiles" of the activity of different repair systems in a particular tumor to guide drug choices.
"This is an exciting part of the trend toward 'personalized medicine,'" says D'Andrea. "We could genetically profile tumor tissue from a biopsy and say, 'Here's the status of DNA repair in this tumor at this moment in time – this pathway is low, therefore a certain drug should work well, or this pathway is high, we'd better stop giving this drug.'"
Scientists have identified at least six major types of DNA repair mechanisms in normal cells. Each excels at fixing a different form of breakdown in the DNA molecule.
"By understanding these damage repair pathways, we can tweak them to enhance radiation therapy by protecting normal tissues or increasing the kill of tumor cells," says Jay Harris, MD, chairman for Radiation Oncology at Dana-Farber/Brigham and Women's Cancer Center (DF/BWCC).
D'Andrea, a pediatric oncologist by training, followed leads from his studies of a rare genetic blood disease, Fanconi anemia, to the important discovery of DNA repair abnormalities in a variety of cancers – and ways to use them for treatment.
Patients with Fanconi anemia have a defect in one of the components of a common DNA repair pathway. D'Andrea found that this pathway was part of a repair circuit that includes the BRCA1 and BRCA2 genes which, when mutated, raise the risk of breast and ovarian cancer. Although Fanconi anemia patients have an elevated cancer risk, their tumors are vulnerable to drug regimens containing cisplatin. It occurred to D'Andrea that certain cancer patients who lack this repair pathway would also respond well to cisplatin. His hunch was correct. Breast and ovarian cancers that had an inactive Fanconi repair pathway were highly sensitive to cisplatin.
How a gene mutation can both raise the risk of cancer and make tumors vulnerable to certain drugs is an interesting tale, explains D'Andrea.
"Tumors sometimes inactivate or 'give up' one of their DNA repair mechanisms," he says. "That raises their odds of mutations that allow them to evolve more quickly and become more aggressive." For the cancer, it is a Faustian bargain. Already lacking one of the six DNA repair kits, the tumor often becomes dangerously dependent on, or even "addicted" to, a single repair method. "The tumor says, 'This pathway is working well for me,' and gets committed to it," explains D'Andrea.
This is an Achilles' heel, a chink in the tumor's defenses, which scientists and cancer doctors are beginning to exploit. They identify the dominant repair pathway, then launch a pre-emptive strike with a drug designed to block it – lowering the tumor's resistance – followed by radiation or chemotherapy. Having now lost two critical repair mechanisms, the cancer cell may die. Normal cells, by contrast, have more functioning pathways and can tolerate the loss of just one.
Clinical trials of DNA repair inhibitors are already under way at DF/BWCC, with some early reports of favorable responses in women with advanced breast or ovarian cancer.
One promising approach involves a new class of drugs called PARP inhibitors. PARP, which stands for poly (ADP-ribose) polymerase, is a key protein involved in repairing broken single strands of DNA molecules.
Conducting trials of PARP inhibitors are Judy Garber, MD, and Ursula Matulonis, MD, of DF/BWCC's Women's Cancers Program, along with Geoffrey Shapiro, MD, PhD, director of the Early Drug Development Center at Dana-Farber. One group of patients believed likely to benefit includes women with "triple-negative" breast cancer, an aggressive form that occurs in many patients with mutated BRCA repair genes. These tumors tend to rely on PARP to repair themselves. So there is reason to believe that blocking PARP to overload cancer cells with genetic damage, then giving chemotherapy, may be a successful strategy for these tumors.
Another form of DNA repair inhibition is aimed at forcing damaged cancer cells to continue dividing, rather than resting and repairing their DNA.
If a cancer cell's checkpoint kinase can be blocked, the cell can't repair the damage caused by cancer drugs. Often, the cell's only recourse is to self-destruct – resulting in a victory for treatment.
"Checkpoint kinases are critical components of cascades that ultimately allow the cell to rest and recover from chemotherapy," Shapiro explains. "So if you block the checkpoint kinase, the cell has to continue through the cycle and it can't repair the damage." Often, the cell's only recourse is to self-destruct – a victory for treatment.
In a recent trial, some DF/BWCC patients who had advanced ovarian or endometrial cancer received a checkpoint-blocking drug along with standard cisplatin chemotherapy, says Shapiro. "We found that many patients who were previously considered to have tumors resistant to cisplatin were gaining some benefit, with minor tumor shrinkages and reductions in CA-125 [a cancer biomarker]," he says. Trials with two other checkpoint-targeting agents are under way in patients with ovarian, colon, pancreatic, or lung cancer.
D'Andrea and his colleagues are now seeking "biomarkers," or chemical fingerprints, that reveal which pathways are turned on or off in a tumor. (See related story Looking and Listening.) "Until three or four years ago, the only way to assess these was to grind up a sample of the tumor and add a piece of damaged DNA to see if the tumor can repair it," says D'Andrea, "[We're] beginning to capture snapshots showing the status of DNA repair in the tumor at a certain moment in time."
These methods could help streamline the time-consuming process of drug development, D'Andrea points out. "We think profiling could help 'stack the deck' so that when you begin a clinical trial of a compound, you would be more likely to enroll patients whose damage repair profiles make them more likely to respond to it.
"I think we're going to see the day when the diagnostic test is packaged as a companion kit with the drug from the very beginning."
With prospects like these, exploiting the expanding knowledge of DNA repair pathways could be an early and important milestone as "one-size-fits-all" therapy gives way to the vision of personalized medicine.
Spring/Summer 2009 Table of Contents
Dana-Farber Cancer Institute, 450 Brookline Avenue, Boston, MA 02215 | Call us toll-free: