The next big breakthrough in cancer treatment is a diagnostic test.
It may not sound exciting at first, but consider that most of the progress we’ve made against cancer since the disease became a major focus of U.S. medical research in the 1970s has been as a result of catching it earlier.
“In the ‘70s, you didn’t know you had cancer until the doctor felt it,” said Dr. Leonard Lichtenfeld, the deputy chief medical officer of the American Cancer Society.
Now, we are inching closer to being able to find cancer before it’s visible to the human eye, using a simple blood test like the one that tests for high cholesterol. That has many cancer researchers pretty excited.
Dubbed a “liquid biopsy,” a cancer blood test is not just a futuristic pipe dream. Last week, Pathway Genomics, a San Diego-based diagnostic company, began marketing just such a test.
“Historically, we’ve called cancer a tissue diagnosis — you had to biopsy the tumor to diagnose it — but now we know tumors are secreting a bunch of things into the bloodstream,” explained Dr. Mark Roschewski, a staff clinician at the National Cancer Institute. “That includes DNA, RNA, and a whole bunch of other things as well.”
Johns Hopkins University’s Dr. Bert Vogelstein, one of the most prominent researchers looking into liquid biopsies, points to one of his studies in which a blood test found half of all colorectal cancers in stage 1 — early enough for doctors to cure them.
“If there were a drug that cured half of cancer you’d have a ticker-tape parade in New York City,” he told the MIT Technology Review last year. (Vogelstein did not respond to Healthline’s request for an interview.)
But a send-off party for cancer may be premature. Pathway can’t point to a single clinical trial that proves the accuracy of its tests.
Chief medical officer Dr. Glenn Braunstein told Healthline that the company has several underway. But he acknowledges that the test “is an adjunct; it’s not a diagnostic.”
The Food and Drug Administration (FDA) permits the test only if it’s done by Pathway’s own lab. It can’t be offered as part of standard diagnostic care. It’s the same gray area that allowed 23andMe to market its genetic test for a few years before the FDA pulled it for making unsubstantiated health claims.
Liquid biopsies are also poised between promise and reality. Eventually, the tests will reliably find cancers early, before they would otherwise be diagnosed, experts told Healthline. But that day is still a way off.
Dr. Cy Stein, Ph.D., an oncologist who specializes in prostate cancer at City of Hope cancer center in Southern California, knows finding a good diagnostic test for cancer can be harder than it seems.
“It’s been more than 30 years since PSA came out, and we’re still not exactly sure how it can be used,” Stein said.
The prostate-specific antigen blood test Stein mentioned was the subject of controversial screening guidelines. The test reliably detects prostate cancer. But it also sometimes picks up non-cancerous conditions and slow-moving cancers.
Testing generally means that a positive result should funnel a patient into treatment. But the PSA fails to differentiate between aggressive cancers and those that may never become symptomatic or dangerous.
In 2008, a test that inspects the blood for circulating tumor cells (CTCs), was approved to detect advanced prostate cancer. But, according to Stein, that test is no more effective than the PSA even though it’s much closer to being a liquid biopsy.
“The CTCs do matter in prostate cancer. They matter in the sense that if your CTC number goes down, you’re doing well. If it goes up, you’re not doing well. Does that differ from what a PSA tells you? Substantially, no, and PSA costs a heck of a lot less,” Stein said.
The CTC test remains in what Stein describes as “that research/clinical utility gap.” It’s hypothetically more informative than a PSA test because the cells reflect the cancer’s genetic mutations, but the information it provides doesn’t mean anything to patients because it doesn’t point them to different or better treatment.
A similar purgatory seems to lie ahead for the liquid biopsy.
Here’s what Roschewski said when Healthline asked him whether patients should be excited about liquid biopsies: “To me it’s nice to have this hope for the future but also to have this cautionary tale that even if you knew all of your information, there’d be almost nothing we could do about it.”
Liquid biopsies offer medical detectives the chance to solve a crime based on fingerprints left at the scene. But they have to do so with only a handful of reference prints on file: Researchers really don’t know that much about the human genome yet, much less the cancer genome.
“What you need to find is a driver mutation, a mutation that produces a protein that actually drives the phenotype of the cancer cell, and then there has to be an approved drug out there that’s going to treat it,” Stein said.
“Any type of screening test has to be associated with some type of intervention,” he said.
Cancer cells have anywhere between 100 and 500 mutations each, and healthy people develop some non-cancerous genetic mutations as they age. In order to work, liquid biopsies will have to identify only those mutations in a patient’s blood that reliably signal cancer.
“Just finding mutations in the blood sometimes found in cancer doesn’t mean somebody’s got cancer,” Roschewski said. “For example, B-cell lymphoma has a particular transmutation. It’s always found in lymphoma, but you can find it in half the people over 50 who don’t have lymphoma.”
Researchers have identified a few markers that do flag cancer, but their work amounts to “proof of concept,” according to Lichtenfeld. They don’t identify cancer often enough, with few enough false positives, to be ready for prime time. Vogelstein’s influential test only caught half of the early-stage cancers.
Even if Pathway or the next company to come along overcomes all of these obstacles, liquid biopsies will still face the problems current screening methods have famously foundered on: Will they find cancers that need treating, or will they funnel patients into unneeded treatments for cancers that would simply have languished?
“We’ve gotten to the point where our imaging technology can find cancers earlier than we ever would have, as we find these even smaller cancers,” said Lichtenfeld, who has been with the American Cancer Society for a decade and a half.
With mammograms, “We’re probably picking up cancers that never would have made a difference in a woman’s life. We’re going to move from the mammogram era, but as we continue on this march downstream, it’s likely that cell-free DNA or other markers will be part of that process.”
Pathway is also aware of the risks of over-treatment, and, as yet, its tests have no answers.
“One of the Holy Grails in cancer is to distinguish between cancers that are indolent — in other words cancers people will die with but not of — and those that are aggressive,” Braunstein said.
Even so, the liquid biopsy is exciting enough that Lichtenfeld confessed that he sometimes has to pinch himself to “not to seem too overly enthusiastic.”
If it’s hard to understand why, consider that Vogelstein and Dr. Murray Korc have used cancer blood tests to try to catch pancreatic cancer. Korc directs the Pancreatic Cancer Signature Center at Indiana University Department of Medicine.
Pancreatic cancer is the fourth leading cause of cancer death, even though it’s not that common a diagnosis. Seven in 10 people die within a year of diagnosis, and just 7 percent are still alive after five years.
“It’s a very nasty cancer,” said Korc. “By the time it presents symptoms, by the time it’s diagnosed, most of the time the horse is out of the barn. It has metastasized and metastatic cancer is very difficult to treat. If you could catch it earlier, it could help these people.”
But here is some middle ground for techno-optimists and those like Stein who have become skeptical of the kind of hype that can surround potential breakthroughs as they weave their way from the Eureka study to the first, 10th, and 100th clinical uses. A liquid biopsy may not be as sexy as a single blood test with a clear-cut result.
Instead, we will likely see a range of tests. All will have to look only for markers of cancers that need to be treated. The findings will probably come in terms of a risk panel based on a full genetic sequencing — something that continues to demand a lot of computer processing power.
But there are already a handful of blood tests being used in cancer centers to monitor patients who’ve had cancer to see if doctors can catch a recurrence sooner.
Sometimes, the future sneaks in through the back door.