Personalized Cancer Test Pinpoints Best Drug For Patients

Cancer, whether in the pancreas, the ovaries or the liver, can take on different characteristics and spread in different ways. That’s why, unfortunately, there’s no one-size-fits-all drug to help patients fight back.

But a new, quick test can personalize treatment and help oncologists choose which chemotherapy route to take.

The test, called Dynamic BH3 Profiling, quickly predicts whether or not a drug will work for a patient by first trying that drug on a tumor sample in the lab. A paper describing the method, which researchers say could become more widespread within a couple of years, was published in the journal Cell this week.

The idea echoes how we choose the most effective antibiotics, says study author Dr. Anthony Letai, a cancer researcher with the Dana-Farber Cancer Institute.

A new tool for predicting relapse in acute myelogenous leukemia (AML) was developed by Dr. Anthony Letai (Courtesy of Dana-Farber Cancer Institute)
A new tool for predicting relapse in acute myelogenous leukemia (AML) was developed by Dr. Anthony Letai (Courtesy of Dana-Farber Cancer Institute)

“When we’re trying to choose antibiotics for people … we simply isolate the bacteria that’s causing the problem and expose it to all the drugs that are available,” he says. Then researchers choose the drugs that best put a lid on the multiplying bacteria.

“That has operated for many, many decades,” Letai says, “so we thought, why not do that for cancer cells?”

Letai’s team isn’t the first to think of this strategy. “People have tried to do this kind of thing in years past but there have been a variety of advances in technology … that make it more feasible this time around,” says Levi Garraway, a cancer researcher at Dana-Farber who was not involved with the study.

What’s different about Letai’s work is its speed: It can quickly determine whether a drug, or combination of drugs, is working. The test looks not at when the tumor cells are dead, but rather when they’re beginning to die.

The 'Death Switch'

The researchers found that there is a point of no return, a threshold of doom, when cells begin to die that is indicative of their actual death. The team looked at varying types of cancer cells (breast, lung, melanoma) and saw that there was essentially a death switch that when flipped on, ensured the cell’s destruction.

Examining if a cancer drug flipped this switch, instead of waiting to see if the cells would eventually die, allowed the researchers to know, in about 16 to 24 hours, which drugs were working.

Without knowing this switch, it can take days of growing cells in the lab (not an easy task).

From there, it was just a matter of testing drugs. “What we really wanted to be able to figure out is what are the cancer drugs that make cancer cells move toward this threshold of death,” says Letai.

Responsive Tumors

In a trial with 16 women with ovarian cancer, the test found a fifth of the tumor samples were responsive to the drug carboplatin. In the clinic, they found that the same women whose tumors responded to the drug saw a slower progression of their cancer than those whose tumors did not.

In the future, the researchers say the method could be used to determine if new drugs in development work on real tumor samples.

“What is exciting about [the test] is that it’s something that we may very well see in the clinic soon,” says Michael Hemann, a professor of biology at the Koch Institute for Integrative Cancer Research and unaffiliated with the study. However, he cautions that the tricky part will be ensuring that any sample taken from a tumor is a good representative of the tumor as a whole.

It's Practical

Hemann also says one of the test’s strengths is that it’s a practical solution. “Most of us, in thinking about advancing tumor therapies, look at what are the characteristics of a tumor that are going to determine if a therapy succeeds or fails. This approach says essentially, that doesn’t matter,” he says.

Another growing field of tests looks at the genetics of a tumor, analyzing the DNA of cancer cells. This involves sequencing the genome, looking at possible mutations, and then understanding how those mutations affect how cells grow.

But Letai says unlike genetics, his test can provide immediate results for patients – either a drug works or it doesn’t. The test is already being used in clinical trials. “What we want to do with this,” says Letai, “is have [patients] waste less of their time on therapies that don’t help.”


More from WBUR

Listen Live