College of American Pathologists
Printable Version

  Sizing up a role for circulating tumor cells


CAP Today




March 2011
Feature Story

Karen Titus

In the world of medical breakthroughs, there’s hype, and then there’s hope. With circulating tumor cells, momentum appears to be swinging toward the latter, with some reasonable stops for “hmmmmm” along the way.

Much of the hope has been fed by smallish studies, and researchers are still figuring out the best ways to count CTCs, characterize them, and then plug their findings into clinical practice. What they’re mostly finding, as adventurers so often do, is that the best horizon seems always to be the next one. Or, as the Spanish conquistador Coronado noted in his journal (on his way to not finding the Seven Cities of Gold): “The route is rough and long.”

At least CTC researchers are far better informed than, say, Columbus, who remained convinced his four Caribbean cruises had taken place in Asia.

Circulating tumor cells have been part of the cancer dialogue since the mid-2000s, when the FDA approved the Veridex CellSearch system for enumerating epithelial CTCs in whole blood. Subsequent approvals have allowed the test to be used in monitoring metastatic breast cancer, metastatic colon cancer, and metastatic prostate cancer.

Early on, looking at CTCs was a numbers game. A 2004 New England Journal of Medicine article (Cristofanilli M, et al. 351:781–791) showed that patients with metastatic breast cancer who had five or more CTCs in a 7.5-mL whole blood draw had worse prognosis than those who had fewer than five CTCs.

Stopping there, such information sparks curiosity, but on a practical level it’s little more than another way to give bad news. As ongoing research suggests, however, circulating tumor cells seem likely to yield more useful information. “When you start to say you can find them, characterize them, and then potentially use that characterization to change therapy, that starts getting exciting,” says Kenneth J. Bloom, MD, chief medical officer of Clarient (now a GE Healthcare company).

“Now, of course, it’s all in the future,” he continues. “No one’s doing it right this second. But the lightbulb is starting to go on in everybody’s head.”

Dr. Bloom is one of the more sanguine observers of the CTC field, alert to their rich potential but not prone to hype. In the past five years, he’s seen two areas of progress:

  • a better grasp of the different methodologies for capturing circulating tumor cells. “I think we’ve learned that it’s a little more difficult than we first thought,” says Dr. Bloom, who terms current technologies as “still fairly early on” in their development.
  • new technologies to analyze extremely small populations of circulating tumor cells. It’s quite likely pathologists will have the ability to do full genome sequencing or to look at hundreds of proteins or mutations on even a single cell, Dr. Bloom says.

Should these technologies live up to their promise, circulating tumor cells could offer insight into the poorly understood link between primary tumors and metastatic disease. Dr. Bloom makes the obvious, but crucial, point: that people don’t die of their primary tumors; they die of metastatic disease.

For pathologists, this raises profound and perhaps slightly unnerving questions. Are the CTCs that develop into metastatic lesions the same as the cells they analyze in primary tumors? Does it make sense to perform biomarker testing on CTCs to help determine treatment?

Preliminary data—from what Dr. Bloom terms limited studies—are intriguing, including those from several groups that demonstrated that circulating breast cancer cells can show amplification of the HER2/neu gene even if the primary tumor was classified as HER2/neu-negative. Given the known heterogeneity of some breast cancers for HER2 expression, the discrepancy should not come as a complete surprise. At the very least, though, CTCs could offer a more attractive diagnostic option. “But these are very few studies with very, very small numbers of patients,” Dr. Bloom cautions.

Massimo Cristofanilli, MD (the lead author on the aforementioned New England Journal article), says some researchers have noted discrepancies in ER status between primary breast tumors and circulating tumor cells. More than 60 percent of breast cancers are ER positive at diagnosis, he says; if there’s conclusive evidence that CTCs are ER negative, “clearly we’re using an ineffective therapy to eliminate those cells,” says Dr. Cristofanilli, chair, Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia.

CTCs might also be helpful in identifying recurrences sooner. By the time classic radiology monitoring techniques find a recurrence, Dr. Bloom says, it’s too late. “They already have a significant mass.”

When he thinks about how circulating tumor cells might be useful clinically, Dr. Bloom draws a parallel to monitoring and treating HIV. The goal would be to lower the burden of tumor, similar to lowering the burden of virus. By monitoring for early recurrence, then molecularly profiling the recurrence, physicians should, in theory, be able to adjust therapy based on molecular alterations. That’s been a successful model for viral infections, he says. “And there’s every reason to think that model would be equally effective with cancer therapies.”

Cancer shares with HIV the highly disagreeable traits of robustness and evolving in response to therapy. For those trying to understand disease progression, knowledge gleaned at the beginning may have little to do with what’s useful at the end. “You have these leaps of faith,” Dr. Bloom says.

John W. Turner, MD, chair of the CAP’s Technology Assessment Committee and staff pathologist, CJW Medical Center, Richmond, Va., invokes the same language to describe reliance on primary tumor analysis. “We’re making a leap of faith that excised tissue is the same as the cells we’re really worried about—the ones that could go to a lymph node or another organ. Whereas if you can analyze those cells that are systemic, you know what you’re getting. You know that you’re targeting those cells that are the most dangerous.”

Leaps of faith don’t have to lead to disaster (or, in Coronado’s case, Kansas).

If circulating tumor cells transform views of cancer, pathologists could have a few new prospects to pursue, some of which are discussed in a College POET (shorthand for the unpoetic-sounding Perspectives on Emerging Technology report, developed by the Technology Assessment Committee). Dr. Bloom, who authored the report, says the committee was intrigued not only by new technologies and applications of older methods for analyzing circulating tumor cells, but by the way CTCs could help pathologists lay claim to new territory. The two broader ends of the cancer spectrum, as Dr. Bloom notes, are early detection (currently the purview of radiology techniques) and recurrence monitoring (ditto). “This is a space pathologists need to own,” he says. “This is an unbelievable opportunity for pathologists.”


CTCs might give pathologists a chance to repair what has been an unnatural rift, biologically speaking, between anatomic pathology and clinical pathology. Billing challenges aside, says Dr. Turner, “Pathologists who are used to looking at cytology will have to spend time in the clin laboratory, merging their cytology knowledge with knowledge of immunofluorescence studies and analyzers and so on.”

He’s not suggesting a turf war, at least not among pathologists. The real tussle might be with radiologists, or even clinicians. As these tests become easier, “more Apple-like, theoretically you could put a box anywhere,” says Dr. Turner. “You want to make sure it ends up in the lab, not somewhere else.”

Dr. Bloom, who despite his cautions could pass for pathology’s Daniel (“Make no little plans”) Burnham, is already peering ahead to see what might lie beyond circulating tumor cells. “Clearly we need better ways to find circulating tumor cells throughout the whole of the circulation,” Dr. Bloom says. “And then if we can figure out how to concentrate those into the draw, then we would have a more robust representation of what’s actually going on. And then you would apply whatever the molecular tools of the day are to address that question.” At Clarient, he says, researchers are looking at ways to map intracellular pathways and examine tens to hundreds of proteins on a single cell.

Dr. Turner puts a new spin on personalized medicine: personalized testing. “As we continue down this road, we should be able to customize the technology, in some ways, to what we need to know.” If a patient with a malignant melanoma, for example, becomes eligible for a certain chemotherapy based on having CTCs with a B-RAF mutation, “then we can, theoretically, customize the test to match them. I find this incredibly fascinating,” he says.

Among translational researchers, the game’s most definitely afoot.

Martin Fleisher, PhD, chair, Department of Clinical Laboratories, and chief, Clinical Chemistry Service, Memorial Sloan-Kettering Cancer Center, has been working with CTCs for a little over four years, as part of MSKCC’s biomarker discovery program.

Dr. Fleisher and his colleagues began looking at CTCs in metastatic prostate cancer to see if they could help determine therapeutic efficacy, and their validation work with the CellSearch system was subsequently used, he says, in the FDA’s approval of CellSearch in that setting.

In four short years, CTC studies at Memorial Sloan-Kettering have mushroomed tremendously, Dr. Fleisher says. Hundreds of patients are enrolled in new drug discovery trials, which, under FDA and NIH “context of use” requirements, use CTCs to monitor endpoints in the therapeutic responses to targeted therapies.

Circulating tumor cells fascinate Dr. Fleisher. If CTCs show a genomic pattern different from that of the primary tumor, “There’s a message there,” he says. Why do 70 to 75 percent of prostate metastases occur in bone, for example? What is it about the bone matrix that permits those cancer cells to survive and grow?

Nor do researchers know if all CTCs are created equal. Do they all have metastatic potential? No one knows. Why, and when, do they emerge? No one’s sure.

“There’s something unique about a cancer cell that sloughs off and survives,” says Dr. Fleisher. “We’re just scratching the surface. There’s so much going on here.”

It’s manifest, Dr. Cristofanilli agrees, that cancer cells alone aren’t sufficient to cause the development of metastases. CTCs may be the trail to follow, however, to find out what’s happening on a deeper level. “Specifically, we need a population of cancer cells that we define as cancer stem cells. Microenvironment is also important, and we may have to incorporate information about other cells in the microenvironment.” He suspects mesenchymal stem cells of playing a role in preparing a niche for metastases to develop. “This is a complex process.”

What in the microenvironment of bone matrix, for example, protects CTCs from destruction? Presumably it responds to a signal from the CTC. “If we knew what that signal was, we could turn it off, and those cells would be destroyed,” Dr. Fleisher says.

That underscores the point that it’s no longer sufficient to enumerate CTCs. Researchers are shifting from accounting to content development—a dirty phrase in journalism, perhaps, but not in diagnostic science.

Their work has been made easier by the explosive growth in technology. If laboratories had a choice of two or three research tools a few years ago, they can now choose from close to 20, Dr. Fleisher says. (Using another informal yardstick to measure growing interest in the field, Dr. Fleisher says the number of articles with “circulating tumor cells” in their titles has gone from a half dozen a month to hundreds.)

Of course, you can’t characterize cells without tallying them first. Dr. Fleisher outlines the four current techniques for measuring circulating tumor cells:

  • cell enrichment (the method used in CellSearch), using immunomagnetic capture techniques that identify an EpCAM antigen located on the surface of the CTC; the cell is then removed via a biomagnetic process. The epithelial cells are distinguished from leukocytes through use of antibodies specific for CD45. The next step is to tag the cytokeratin protein found in the epithelial cells using a fluorescent antibody, which allows researchers to count the cells.

    The method has some drawbacks, which users readily acknowledge. “We understand what the rules of this game are,” Dr. Fleisher says. If the cells don’t make sufficient EpCAM, they won’t be captured, and thus not counted. If EpCAM is sufficient but cytokeratin level isn’t, the cells will be captured but remain unseen. “And therefore you won’t count them,” says Dr. Fleisher.

  • technologies that separate CTCs from other blood cells based on physical size. (CTCs are roughly twice the size of a typical lymphocyte, Dr. Fleisher says.)
  • negative selection process, which lyses the blood cells to eliminate anything that’s not a CTC.
  • chip technologies that can be used for EpCAM positive separation of cells.

Like the CellSearch process, the newer techniques have their limitations and drawbacks. And, unlike CellSearch, none of the other methods are FDA approved yet. They’re specialized, time-consuming, expensive. For now, CTC analyses by these techniques are not reimbursable. But Dr. Fleisher urges pathologists to follow them closely, and, if circumstances permit, become involved in CTC work. “This is translational research that needs to be done in a clinical setting,” he says.

Dr. Fleisher offers one other strong bit of advice, in triplicate, for anyone who wants to look at CTCs. “Validate. Validate. Validate.” This may not have the ring of veni, vidi, vici, but ignoring it renders data useless. He says he’s tired of seeing published data based on methods that haven’t been validated. “You must understand what you are measuring. We’re laboratorians. This is what we’re supposed to do.”

Dan Boffa, MD, is hoping circulating tumor cells “will expose the Achilles’ heel of metastatic progression.”

Dr. Boffa and his colleagues at Yale University School of Medicine are using circulating tumor cells as a marker for outcome after administering curative surgery to patients whose disease has spread to limited sites. CTCs are measured before and after surgery. Dr. Boffa, assistant professor, Section of Thoracic Surgery, is hoping the information will enable physicians to distinguish patients who have oligometastatic disease from those whose cancer is more widely disseminated.

“At its simplest, we just compare the counts in those two groups of patients,” Dr. Boffa says. “But we follow the counts through steps of treatment in both groups to see if the difference between limited spread and wide spread is simply a numbers game, or if there’s more to it than that—if there are aspects of the circulating tumor cells that would reflect their virulence or their potential to establish new sites of metastatic progression.”

“Surgeons hate dissemination,” he continues. “We have a great solution for cancer where it starts, and we have a great solution for isolated deposits of cancer spread. But we’re really powerless when cancer is spreading via the bloodstream. So I thought that the circulating tumor cell was the perfect way to expose a point of vulnerability in the metastatic progression pathway.”

In his four years of research, he’s watched the progression from counting to characterization. The next phase, he says, which is just starting, will be figuring out the traits that distinguish dangerous CTCs from those that may be less likely to lead to metastatic progression. “We have a lot of tools to profile the circulating tumor cells. The problem is, you don’t know their behavior,” Dr. Boffa says.

One model would be to separate CTCs, reintroduce them into an animal model, then try to correlate behavior with molecular aspects. Both the positive and negative selection methods of current CTC separation technologies have the potential to change the circulating tumor cell, however. The next step in the experimental model thus may not be truly reflective of what that cell is like in the circulation, Dr. Boffa says.

But, like the others, he remains in the ranks of the cautiously hopeful. “The technology is evolving, and this will be a real possibility in the future. This is what excites me—that we’ll be able to expose the vulnerability in the process.”

He too thinks circulating tumor cells may also help explain the role of dormancy in cancer progression—that is, disease progression beyond what was expected, based on imaging studies, at the time of initial treatment. He describes, as an example, a patient with a small tumor that is completely removed with a border of normal tissue, who then develops widely metastatic disease several years later, long after the “good until” date for circulating cells should have passed. “Where does that come from?” Dr. Boffa asks. “And what was it doing during that time period?”

“I think circulating tumor cells may actually unlock some of the secrets of dormancy, which is also sort of a black box of cancer care,” he says.

Even if they don’t, and even if half the hopes pinned to circulating tumor cells fall flat, it’s unlikely CTCs will slip away from the diagnostic scene completely. Dr. Bloom sees them as a wakeup call for pathologists.

“Just doing single testing on an excised specimen is not the be-all, end-all.” It may be what pathologists do today, he says—but it likely should not be what they do tomorrow.

Karen Titus is CAP TODAY contributing editor and co-managing editor.