An artistic representation of tumor-circulating DNA in the blood
For years, researchers have been trying to figure out better ways to find out if a person has cancer and to find it early when it’s most curable. They want a test that’s easier on patients and faster for doctors. One idea that’s gaining ground is called a liquid biopsy. A biopsy is a sample of tissue or cells taken from almost any part of the body and sent to a lab to check for cancer. The liquid in this case is your blood. The concept is that your doctor may one day be able to use a small sample of your blood to screen for cancer, before you have any symptoms.
Recently, a study led by researchers at Johns Hopkins published results about how their blood test, called CancerSEEK, supports the concept for a single blood test to screen for several types of cancer. Their test studied 2 types of markers in the blood associated with cancer: certain proteins and pieces of DNA, called circulating tumor DNA. The test is far from ready for wide use though. They tested CancerSEEK on people who had already been diagnosed with cancer to see if the blood they collected would allow them to prove the presence of cancer and where it was. The test was stronger on finding some types of cancer than others and it had some false positives, meaning it showed there was cancer when there really wasn’t. Still, the study caused a lot of excitement. So, we got a perspective from Len Lichtenfeld, MD, deputy chief medical officer at the American Cancer Society. Here’s what he said about the research that’s been done to get us to this point and where research needs to go to bring us closer to the reality of such a comprehensive, simple screening test.
“Cancer’s early detection by means other than X-rays, colonoscopy, or Pap smears has been a subject of research for at least 2 decades, if not longer,” Lichtenfeld says. During that time, he says, we’ve been learning and discovering more about using circulating tumor DNA (ctDNA) to evaluate cancer and about the pros and cons of trying to find cancer early.
First, researchers discovered how to find and study ctDNA. “Before using tumor DNA to find and evaluate earlier stages of cancer, this technology was used to monitor patients after they received treatment for an advanced cancer,” Lichtenfeld explains.
Circulating tumor DNAs are tiny fragments of DNA in the blood that break away from tumors. After treatment, the ctDNA levels decrease because the tumor is either smaller or has been removed. “Researchers monitored ctDNA levels in blood to watch for increases over time. They realized they could see an increase in the ctDNA several months before they could visibly see a recurrence of cancer,” he says. And, of course, finding a recurrence sooner allows for a quicker treatment response.
“Another way researchers have used ctDNA is to monitor an advanced cancer for genetic changes that might influence treatment,” Lichtenfeld says. What they learn may help inform how a patient will respond to a certain treatment or show whether a patient is developing resistance to a treatment that has been working.
Researchers can take a biopsy of a tumor and also test the blood for ctDNA to get a more complete picture of a cancer’s genetics.
“A sample from sticking a needle into a tumor doesn’t reflect all the genetic changes in that cancer or all the places that cancer may be,” Lichtenfeld says. “But the ctDNA sort of acts like a vacuum cleaner, bringing together all the DNA fragments and finding other mutations that may be elsewhere. So, the 2 tests—a biopsy of the solid tumor and a liquid biopsy—are definitely complementary.”
We’ve learned that some cancers don’t really need to be found. A blood test to screen for and evaluate cancer has been around for many years—the prostate specific antigen (PSA) test. “Long before we had fancy diagnostics like CT scans,” Lichtenfeld says, “medical school professors taught future doctors that autopsies on men show many inconsequential prostate cancers, ones that never appeared during life.” Those men died from some other cause, he explains, so finding that cancer while they lived wouldn’t have made a difference.
This may be the case with other types of cancer, too. “Not every cancer is aggressive or lethal,” Lichtenfeld says. Some cancers are so slow growing that they would never need treatment. We need better methods to differentiate which ones can be left alone and which ones need to be treated.
We’ve learned that finding cancer early may not be enough. “I came from an era where we blindly accepted that every cancer we found early was a life saved. We thought we could prevent spread and death, but that wasn’t the case,” Lichtenfeld says.
The available technology for both screening and treatment then was different from what doctors use now. But this is still true: “Not every cancer we find can be effectively treated,” Lichtenfeld says. The underlying biology of the cancer can “sometimes dictate the outcome, despite our best efforts.”
“People tend to forget that just because you find a cancer when it’s small doesn’t mean that the patient won’t have problems in 10 or 15 years, or earlier for that matter,” he says. We’re doing better than we did, he says, but not as well as we need to be.
“It’s a lesson we need to heed: Small is not always better—it’s frequently better,” he says, “but not always better.” We don’t know yet whether screenings with a liquid biopsy could help doctors learn how aggressive a cancer is.
In the last several years, both academic and commercial researchers published reports about finding circulating DNA in earlier stage cancers. Many of the studies are trying to solve the same problems, with researchers using different technologies and approaches.
Many researchers are trying to solve the problem of tiny amounts of ctDNA. There’s only a very small amount of circulating DNA even with a medium-sized tumor, Lichtenfeld explains. Plus, ctDNA is only one type of DNA circulating in the blood. Pregnant women have DNA from their baby’s placenta in their blood. People who’ve had a heart attack or stroke may also have DNA fragments in their blood. Collectively, all those DNA fragments are called cell-free DNA. So, researchers need to be able to accurately identify ctDNA among all the other types of DNA in blood to avoid false alarms, telling someone they have cancer when they don’t.
But the amount of tumor DNA is “infinitesimal,” Lichtenfeld says. “With cancer, you’re looking for a very small quantity of ctDNA in a sea of unrelated proteins.”
Researchers are also exploring ways to identify where the cancer may be. The goal is to use the cell-free DNA in the blood to learn that a person may have cancer somewhere in the body before it’s easily visible by current techniques, such as colonoscopy, mammography, x-rays, or CT-scans. “The very fact that the cancer’s not yet visible means doctors need a screening tool that gives them some guidance about where to look. Otherwise, doctors could be going all over the body in a fruitless search,” Lichtenfeld explains.
In this recent study, researchers combined technologies. They looked at both cell-free DNA and known protein markers for several common types of cancer. “With that combination, the cell-free DNA allowed researchers to sort for circulating tumor DNA and know, ‘yes, we have a mutation that suggests cancer.’ Linking ctDNA information with a protein marker offered them some sense of where the cancer may be found,” Lichtenfeld says. For example, they look for raised levels of cancer antigen 125 (CA-125), known to be associated with ovarian cancer.
Their ability to pin down the cancer’s location varied based on where it started in the body. Their best results were with colorectal and ovarian cancers. Their least accurate results were with liver and lung cancers.
“Looking at ctDNA going forward, I have no doubt we’re going to get to a point someday – next year or whether 10 years from now, I don’t know – but we’re going to get to the point where we will be able to screen using a blood test,” Lichtenfeld says. To get there, a lot of complex issues need to be solved.
Our analysis has to catch up with the technology and data. “One of the incredible places we find ourselves,” Lichtenfeld says, “is that technology advances quickly, but our ability to analyze the impact of that technology and determine how best to use it, is much further behind. The translational science, understanding how to use the data we collect, is even more difficult and takes more time and patience.”
Some researchers, like the team at Johns Hopkins, are using artificial intelligence, what they call “supervised machine learning,” to help them. “It’s a matter of combining a lot of data to recognize or answer whether a mutation truly has a role in a particular cancer,” Lichtenfeld explains. “The machine continues to learn. The next time it makes a better choice, and it continues to make better and better choices.”
We need to translate what we know to people who haven’t been diagnosed with cancer. “I don’t want to minimize the Johns Hopkins research because it is elegant science,” Lichtenfeld says. “It takes a lot of work to get to where they are.”
But it’s one thing, he says, to use a liquid biopsy to test a patient who you know already has cancer, as they did for their study. “It’s completely different to translate that into a situation when you’re screening millions of people who have no evidence of cancer,” he says.
Researchers need to develop a meaningful test that does no harm. Lichtenfeld believes that “getting the technology piece down is going to be a much shorter period of time than determining the true value of the test.”
“We run the risk of having an available test that people use, without it being meaningful. Does it tell us anything? Does it make a difference?” Lichtenfeld says.
A meaningful liquid biopsy of the future, he says, will need to find cancer in someone with no signs of it and be able to determine where that cancer may be in the body. It also needs to tell us whether that cancer requires treatment. And if so, it needs to make use of what we’ve learned in the past to guide the best way to treat that cancer and favorably impact the patient’s future.
“One of the keys in this discovery process is not only to find the cancer cell or find the mutated DNA fragment circulating in the blood, but also to find out whether that particular person’s cancer is going to be a problem. To me, that’s going to be an even bigger challenge,” Lichtenfeld says.
We need to keep funding the research. “We need to use our excitement about the work on liquid biopsies to make certain we can continue to advance this science. We’re not at the end of the road. We definitely have come a long way, but we still have a long way to go.”