Tumors have found a bloody new way to grow and spread

An artist’s concept of a tumor resorting to vasculogenic mimicry.

V. Altounian/Science

Tumors have found a bloody new way to grow and spread

On a cool day in March 2000, several hundred researchers jammed into a hotel auditorium in Salt Lake City, eager to see a showdown over what had become one of the most controversial ideas in cancer research. On one side stood cancer biologist Mary Hendrix of the University of Iowa Cancer Center in Iowa City, whose team the year before had reported an unusual, seemingly new way through which tumor cells can tap into the blood supply and obtain nutrients. Facing off against her was tumor vascular biologist Donald McDonald of the University of California, San Francisco, who was certain that she and her colleagues had misinterpreted their data. “This debate had the feeling of a boxing match—with the championship belt hanging in the balance,” Hendrix recalls.

Researchers knew at the time that tumors can induce the endothelial cells of normal blood vessels to form new supply lines into a tumor, a process called angiogenesis. But Hendrix and her colleagues contended that tumor cells themselves sometimes create their own blood-delivering tubes, a mechanism they dubbed vasculogenic mimicry (also known as vascular mimicry). Their 1999 paper “started lots of upheaval,” says histopathologist Francesco Pezzella of the University of Oxford in the United Kingdom. The Utah debate, held at a Keystone meeting, was the first public discussion of the concept.

In the end, neither side scored a knockout. Hendrix asserted that the loops and networks her team had observed represented a mini–circulatory system produced by the tumors themselves. McDonald countered that the patterns were folds of connective tissue, not tubes that carried blood. In the years since, the controversy has waned, and Hendrix and other researchers have pieced together a picture of how tumors build their own blood vessels and how they can affect prognosis and treatment. But some scientists continue to find the idea deeply unsettling.

Now, vasculogenic mimicry faces another big test. The first clinical trial of a drug to block the process—and thus potentially limit tumor growth—has begun in the United States and Taiwan. If the drug succeeds, it would bolster what Hendrix and other researchers have been saying about these do-it-yourself blood vessels for nearly 17 years. And it might also explain why some of the most hyped drugs in cancer therapy—angiogenesis inhibitors—have underperformed.

Vasculogenic mimicry roiled the cancer field because it undermined the leading idea for how tumors obtain their blood supply. In the early 1970s, Judah Folkman of Harvard Medical School in Boston proposed that tumors can grow large because they trigger angiogenesis, inducing new blood vessels that speed nutrients and oxygen to fast-dividing cancer cells. Halting the growth of these vessels into tumors, he suggested, would starve the masses. Folkman himself famously struggled to convince the many angiogenesis skeptics among cancer researchers. But by the late 1990s, drug companies were bustling to develop compounds that curtail angiogenesis, and DNA pioneer James Watson announced that Folkman’s approach would “cure cancer in 2 years.”

Yet if tumors have an alternative way to secure the blood they require, Hendrix and colleagues argued, antiangiogenic compounds might fail. “That put a big target on us,” she says. Later clinical trials of angiogenesis inhibitors have confirmed their doubts, however. Although several of the drugs have received U.S. approval for use in cancer patients, including Avastin and Nexavar, they only temporarily slow tumor growth; the tumors often become resistant. Pezzella says the evidence now suggests that Hendrix was right. “Vasculogenic mimicry is one of the ways in which tumors develop a blood supply independently from classical angiogenesis.”

V. Altounian/Science

Hendrix and her colleagues, including University of Iowa pathologist Robert Folberg, didn’t set out to deflate the excitement over angiogenesis. They were studying why some melanomas are virulent, spreading rapidly and often killing patients, whereas others are less dangerous. In one revealing experiment, they planted human melanoma cells on a gel that mimics the extracellular matrix, the fibrous material that surrounds cells. Aggressive tumor cells “migrate through the matrix and scrunch it up,” Hendrix says, resulting in networks of channels. “At low [microscope] power, they look like chicken-wire meshwork,” she says.

Folberg had been seeing similar patterns in melanomas that grew in patients’ eyes. “We naïvely assumed they were blood vessels,” says Folberg, dean of the Oakland University William Beaumont School of Medicine in Rochester, Michigan.

Taking a closer look at eye melanomas, the researchers noticed another similarity to blood vessels: Some of the channels contained red blood cells. But to the scientists’ surprise, the networks lacked the endothelial cells that line normal blood vessels, suggesting that the cancers themselves formed the channels. In a commentary on their paper, which appeared in 1999 in The American Journal of Pathology, noted cell biologist Mina Bissell of the Lawrence Berkeley National Laboratory in Berkeley, California, called the clinical implications of the findings “far-reaching.”

Yet in a rebuttal published a few months later in the same journal, McDonald and two other researchers knocked the paper on several grounds, such as failing to demonstrate that blood flowed through the networks. The blood cells inside the channels could have leaked from conventional tumor vessels, McDonald and his co-authors proposed. They deemed the evidence for vasculogenic mimicry “neither persuasive nor novel.”

Hendrix and colleagues acknowledge that other researchers had previously suggested the phenomenon. She also admits that during those early years, “we did not have all the answers.” But she says that the case for vasculogenic mimicry has become much stronger. For one thing, researchers have identified the tubes in many more tumor varieties, including breast, prostate, kidney, lung, and bone cancers. Moreover, growing circumstantial evidence suggests that when they appear, a cancer patient’s odds of survival plummet. A 2016 metaanalysis by researchers in China, which combined results from 36 studies on more than a dozen cancer types, estimated that patients’ chances of dying were roughly doubled if their tumors showed evidence of vasculogenic mimicry.

Researchers have also furnished evidence that the channels do transport blood. In a 2008 study, for instance, Folberg and colleagues injected a fluorescent dye into the arms of patients who had melanomas in their eyes. Within 30 seconds, the dye had traveled through the patients’ circulatory systems to their eyes and had appeared in the channels in their tumors.

McDonald, however, remains skeptical. Other researchers, including vascular biologist Drew Dudley of the University of North Carolina, Chapel Hill, accept that vasculogenic mimicry can occur but want further evidence of its relevance. “It isn’t clear whether this occurs substantially in human patients with cancer and whether this has anything to do with antiangiogenic therapies not working very well,” Dudley says.

Hendrix and other scientists are pushing ahead to learn how vasculogenic mimicry might work. They’re still not sure how the tumor-spawned networks hook up to the normal circulation, but they have determined that as tumors build blood vessels from their cells, the cells switch on many of the same genes that normally define endothelial cells. For example, in normal blood vessels, clots could form and cause blockages if endothelial cells didn’t release anticoagulant compounds. “That’s a real problem that cancer cells [forming their own blood vessels] would have to solve,” Dudley says. They apparently have. Hendrix and colleagues discovered that cancer cells involved in vasculogenic mimicry release some of the same anticlotting molecules endothelial cells do.

Only certain cells in tumors seem to have the ability to produce the blood-transporting channels, and they may overlap with so-called cancer stem cells, rare cells in tumors thought by many researchers to fuel the overall growth of the cancerous masses. In melanomas, for instance, skin pathologist George Murphy of Brigham and Women’s Hospital in Boston and colleagues reported that the cells that are capable of vasculogenic mimicry show cancer stem cell characteristics such as chemotherapy drug resistance and the ability to specialize into different cell types. “There appears to be a subpopulation of cells within a cancer that are very smart,” Murphy says.

The tubes that these smart cells build could be dangerous not just because they allow tumors to receive needed blood. Vasculogenic mimicry may also promote metastasis, the migration of tumor cells to new parts of the body, which is responsible for most cancer deaths. In a study reported last year in Nature, molecular biologist Greg Hannon of the University of Cambridge in the United Kingdom and colleagues tagged individual breast cancer cells with a specific nucleotide sequence, a DNA barcode, and injected them into mice. Some of the cells gave rise to tumors, and some of these tumors spawned metastases. The barcodes allowed the researchers to track metastases to their ancestral cells.

Hannon and colleagues found that the cells most likely to grow into metastases had cranked up two genes that inhibit blood clotting. To the team’s surprise, tumors derived from these wayward cells also showed vasculogenic mimicry. When the researchers inhibited the two genes, the number of vasculogenic mimicry channels in the tumors declined, suggesting that the genes induce formation of these blood-carrying networks. 

Hannon says that the genes may benefit a tumor in two ways—by spurring it to grow the ersatz blood vessels and by preventing clots from blocking them, thus ensuring a steady supply of oxygen and nutrients. But the vasculogenic mimicry channels would also make it easier for tumor cells to make a getaway into the circulation. “To my knowledge, this is the first time we’ve discovered that a process that promotes metastasis provides a selective benefit to the primary tumor,” Hannon says.

If vasculogenic mimicry does pave the way for metastasis, blocking it could save lives. Researchers have tested whether several standard angiogenesis inhibitors curb vasculogenic mimicry, but they seem to do the opposite. By stalling the formation of normal blood vessels and starving tumors of oxygen, the drugs appear to trigger cancer cells to build their own blood highways.

TaiRx, a biotech firm in Taipei, is exploring an alternative approach. The company originally developed the drug CVM-1118, a derivative of a plant compound, to block cancer cell growth. Yi-Wen Chu, TaiRx’s senior vice president, sent the drug to Hendrix, her former Ph.D. superviser, to determine whether it would halt vasculogenic mimicry. It did, curbing the activity of Nodal, a gene that drives vasculogenic mimicry by making cancer cells more like stem cells. This year the company launched a phase I trial to evaluate the safety of CVM-1118 in people with a variety of untreatable cancers and to assess its effectiveness.

Although CVM-1118 is the first drug that targets vasculogenic mimicry to reach clinical trials, pharmaceutical companies are trying to develop others, Hendrix says; several have sent her candidates to test, although she can’t disclose the firms’ names.

Hendrix recently became president of Shepherd University in Shepherdstown, West Virginia, a liberal arts institution that is her alma mater, and has arranged to move her lab to West Virginia University, Morgantown, where she plans to continue searching for agents that block tumors’ do-it-yourself vessels. Looking back on the controversy she and her colleagues stirred, she says it was traumatic. But “if we shook up the field enough to get people to think about a new approach, I’ll be happy."