By Carolyn Y. Johnson - The Boston Globe
Cancer patients and doctors often don’t know much about what they are up against: a tumor that will spread through the body, or one that probably will remain dormant. Now, Boston scientists have identified four genetic markers that predict whether prostate cancer is the aggressive type.
The work, led by researchers at Dana-Farber Cancer Institute, is part of a much larger effort to address a major problem in cancer treatment. Although screening tests can identify tumors early, doctors have limited insight into whether they need to hit the disease right away with aggressive treatment or should watch and wait.
That means some patients, especially with prostate cancer, are overtreated at great cost and often with severe side effects, while others might benefit from more drastic interventions given earlier.
“We have a big problem in not knowing how to optimally treat patients with these early-stage cancers,’’ said Dr. Ronald DePinho, director of the Belfer Institute for Applied Cancer Science at Dana-Farber, who led the research and is a cofounder of Metamark Genetics, a Cambridge company that is now working to translate the basic research into a test.
In prostate cancer, “it’s quite clear that the current method — which is almost a coin toss — is certainly not adequate.’’
In the study, published online by the journal Nature, researchers were searching for a biological marker — a signature of genes and proteins — that could help them predict what course a cancer would take.
They started by genetically engineering mice that developed metastatic prostate cancer. By studying the mice, they were able to identify four genes that play an important role in causing cancer cells to spread. Then, using the biological markers they discovered in the mice, they analyzed samples taken from 400 men who had suffered from prostate cancer.
Traditional methods for determining whether prostate cancer is aggressive are accurate 60 to 75 percent of the time, depending on the skill of the pathologist. But DePinho found that combining the traditional method with the new method in the sample he tested, the predictions were correct 90 percent of the time.
The finding needs more testing on patients, but outside researchers said it was a promising step forward.
“I think it’s a fantastic, compelling example of how useful it is to work with mice and mouse models of cancer to inform the way by which we prognosticate and treat cancer,’’ said Dr. Pier Paolo Pandolfi, director of the cancer genetics program at Beth Israel Deaconess Medical Center, who was not involved in the study.
He added that the finding will need to be tested in other groups of human patients, to ensure it truly is a predictor of how prostate cancer will progress. But if the initial result is borne out, he said, “this is exactly what we need in prostate cancer and in all cancers — to really know if the tumor presentation is going to be an aggressive one, a metastatic one.’’
Not knowing which cancers can be expected to be metastatic is an especially difficult problem in prostate cancer. DePinho noted that regular screening identifies about 220,000 cases of prostate cancer in the United States annually, but only 27,000 men die of the disease each year.
According to a recent European study reported in the New England Journal of Medicine, 48 men are treated to prevent a single death from the disease, subjecting many men to drastic side effects, such as impotence or incontinence.
If doctors could make more accurate predictions about the progression of prostate cancer, they could spare many men needless treatment and avoid accumulating health care costs.
“What we’re trying to do as our mission is move further upstream from the point of diagnosis, to help physicians and patients decide whether the type of cancer they are going to have is aggressive,’’ said Mark Straley, chief executive of Metamark Genetics.
Metamark is also working on devising tests for several other kinds of cancer.
The hope is that by separating patients with cancers that are hardwired to be aggressive from patients whose cancers are less dangerous, doctors and patients could make more informed choices about how intensively to treat a patient.
Ultimately, scientists are seeking biological markers that could help them tailor treatments to the patients by understanding who would probably respond to a particular drug or therapy.
Carolyn Y. Johnson can be reached at firstname.lastname@example.org.