|
||||||||
|
ONCOLOGY Mosaic Blood Vessels Could Provide Portal for MetastasisFindings Suggest New Methods for Halting Cancer Spread One of the great revelations of late 20th century cancer biology was the discovery that for a tumor to take root, it must put out a burst of nourishing new capillaries. This insight has engendered not only a new strategy for thwarting cancer—namely, cutting off the blood supply—but also a strikingly different picture of the goings on inside a tumor. Rather than a monolithic mass consisting of one cancer cell after another, a tumor appears more like a coral—dense material riddled with hundreds, if not thousands, of tunnels, each one representing a blood vessel.
A motley mass. Tumors are perforated by hundreds of blood vessels, each surrounded by a cuff of tumor cells (white dots in left diagram). Tumor cells take up temporary residence in the walls of 15 percent of tumor blood vessels (right). Munn and his colleagues estimate that a million tumor cells per gram of tumor could pass through these mosaic blood vessels daily. Adapted from originals by Judah Folkman and Donald McDonald A team of Massachusetts General Hospital research-ers has been shining a high-tech flashlight inside these blood vessel tunnels, and it appears that they too offer up a surprisingly varied topography. Lance Munn, Rakesh Jain, and their colleagues have found that tumor cells infiltrate the walls of the new vessels, creating a mosaic of cancer and endothelial cells. And these "mosaic blood vessels" appear with surprising frequency. Fifteen percent of capillaries in mouse and human cancers were found to exhibit a mosaic of endothelial and tumor cells. On average, tumor cells account for approximately 4 percent of the surface area of such mosaic vessels. Their findings appear in the Dec. 19 Proceedings of the National Academy of Sciences. How the interloping tumor cells make their way from the tumor into the blood vessel wall is not clear—do they squeeze their way in or do the endothelial cells make way for the intruders? Munn, HMS assistant professor of radiation oncology, and his colleagues are currently investigating these questions (see figure). But the discovery that the tumor-studded vessels occur with surprising frequency is providing clues to a pair of mysteries—one old and one new.
"There are a lot of labs looking at tumor cells and a lot of labs looking at blood vessel cells. In this lab we look at both cells, and that's what lets us find interesting things," says Rakesh Jain (second from left), shown with (from left) Rosemary Jones, Lance Munn, and Emmanuelle diTomaso. Photo by Graham Ramsay For years, scientists have known that tumors shed cells, each of which can give rise to a metastatic colony in a far-flung corner of the body. And they do so at an extraordinary rate. A single gram of tumor pours forth approximately a million cells each day. What they have long wondered is how this deadly avalanche makes its way through the body. Munn and Jain, the A. Werk Cook professor of radiation oncology, believe that these masses of sloughed off cells could be exiting through the mosaic vessels. They performed calculations showing that though relatively small, the 4 percent surface area occupied by tumor cells is sufficiently large to allow the transit of approximately one million cells per gram of tumor each day. "We can't say for sure that all the regions of a mosaic blood vessel are metastatic," Munn said. "It's possible that some of these regions might be rivers through which cells are just pouring into the vessel while others are static." Detecting Tiny TumorsSince shedding and metastasis happen in tumors less than a gram in weight, it might be possible to detect microscopic tumors, said Judah Folkman, the Julia Dyckman Andrus professor of pediatric surgery at Children's Hospital. For example, if one could capture and genetically analyze these sloughed off tumor cells, it might be possible to detect cancer even before tumors are spotted, said Folkman, who wrote a commentary on Munn's article in the Jan. 16 PNAS. "So this paper helps move us toward the point where we may diagnose cancer before we see it," he said.Folkman finds the discovery exciting for another reason. It helps explain a mystery raised in his own lab about a year ago by Tim Browder and others. Normally, cancer drugs are administered according to a particular schedule—very high doses every three weeks. Browder, HMS instructor in pediatrics at Children's, and his colleagues found that by using the same drug but changing the scheduling—low doses every six days—they could eradicate tumors that normally resist the drug. So successful was the approach in mice, Browder and his colleagues are currently trying it in humans (see sidebar). But the findings have also presented a puzzle: why are the normally drug-resistant tumor cells succumbing to the drug? One possibility is that the resistant cells are being killed indirectly—through the destruction of the endothelial cells that make up the tumor's blood supply. According to Folkman, Munn's findings suggest that the drug could also be killing tumors directly: by targeting the individual tumor cells exposed in the mosaic blood vessel wall. Researchers have found that drug-resistant tumor cells grown in thin layers are more sensitive to drug than cells grown in bundles. By singling out individual tumor cells and exposing them to drug, the researchers could kill resistant cells immediately. "So it's possible that cells squeezing into vessels may be very vulnerable to chemotherapy," Folkman said. "This is like separating cells from the pack and making them more drug-sensitive." Scientists first observed tumor cells residing in blood vessel walls in 1948, but only as splotches on an electron micrograph. For decades, little attention was paid to these pictures. In 1991, Jain and his colleagues saw white blood cells sticking to the walls of tumor capillaries and thought they might be adhering to exposed tumor cells. But at the time they had no way of spotting tumor cells in the vessel walls. Meanwhile, Jain and his colleagues began to notice another curious thing about the blood vessels. When they delivered stained liposomes and other large particles into the capillaries surrounding a tumor, the particles would leave the blood vessels and enter the tumor in an apparently haphazard way—only in certain blood vessels at certain sites. "When you've seen this heterogeneous leakiness—with these liposomes or gene vectors or other therapeutics coming out only from some regions and not others—it's something that you can't stop thinking about," said Jain. Captivated by the findings, Munn set out to discover what lay behind the apparently haphazard pattern of leakiness. His hunch was that the mosaic blood vessels—with their unusual patches of tumor cells—were playing a role. But it was not clear if the tumor cell patches occurred often enough to account for the observed leakiness. Lighting Cancer's PathWorking with Rosemary Jones, HMS associate professor of pathology in the Department of Anesthesia at MGH, and colleagues, Munn's first step was to find a way to identify tumor cells inside blood vessels. Using a genetic construct devised by colleague Brian Seed, HMS professor of genetics, the Munn team introduced the gene for green fluorescent protein into a line of human cancer cells. When triggered, the gene causes the tumor cells to light up.The team implanted the fluorescing tumor cell lines into mice in two different areas—the ovarian peduncle and the cecum—and let them grow a new blood supply. They then sampled a thousand of the new blood vessels to see what percentage were of the mosaic variety, and of these, how much of the vessel was infiltrated by tumor cells. The 15 percent and 4 percent statistics were observed in both locations. In fact, the proportions appear to hold up in human tumors. Munn and his colleagues analyzed human cancers from the MGH Pathology Department. Though their methods were less sophisticated—"We couldn't have green tumor cells, and the samples were not perfused with fixative"—they saw approximately the same proportion of mosaic vessels in the human samples. In Munn's hands, the numbers tell a tantalizing story. He calculates the number of mosaic vessels is sufficient to allow the exodus of a million tumor cells per gram of tumor every day. Whether they actually make that trip is another question—one he hopes to solve. Using a method invented in Jain's lab which allows a person to view tumors through a window placed into an animal's skin (see Focus, Oct. 2, 1998), he is looking at mosaic vessels as they form and function in living mice. His hope is to catch a tumor cell in the act of making its exodus. "Can we actually see cells coming into the blood vessel? That would be extremely exciting!" he said. Antiangiogenesis Shows Surprises as Approach Against CancerThe angiogenesis revolution is giving an unexpected boost to the workhorses of cancer treatment. When paired with angiogenesis inhibitors, radiation and conventional chemotherapies have been shown to exert a cancer-fighting power beyond what they had before. In the latest twist, tumor toxins such as cyclophosphamide and vinblastine are themselves being shown to exhibit hidden antiangiogenic properties. It makes sense, said Tim Browder, HMS instructor in pediatrics at Children's Hospital. "Think about it. Chemotherapeutic agents kill growing cells. If you say tumor growth is dependent on growing blood vessel cells, then the growing blood vessel cells should be as sensitive to the chemotherapy as the tumor cells," he said. "Also, if you give chemotherapy systemically, that is, through the bloodstream, the first place it's going to see is the endothelial cell before it even gets to the tumor cell." In fact, when Browder administered low doses of cyclophosphamide every six days to mice with drug-resistant tumors, endothelial cells began dying off 12 hours later and continued dying for two days. Tumor cell death did not peak until four days after treatment. "Our interpretation is that the killing of the endothelial cells led to the killing of drug-resistant tumor cells," he said. In addition, the low-dose regimen avoids two of the pitfalls of standard high-dose-every-three-week regimens—toxicity and tumor regrowth during drug holidays. "By lowering the dose and never stopping the drug, you limit toxicity and you gain a more sustained suppression of cancer cell growth." He and others are currently testing the low-dose antiangiogenic approach—using conventional chemotherapies along with new-generation antiangiogenic drugs—in children. "These are patients who have failed everything else," he said. Study Links Hemangioma Enzyme to Mental Retardation, Opens Path to PreventionAsk Judah Folkman what were some of the scientific highlights of the past year and he might point to a study published over a half year ago in The New England Journal of Medicine by a team of HMS investigators led by Stephen Huang. During a recent interview, he immediately produced a copy of the July 20 report. It reads like a scientific detective story—one that Folkman, the Julia Dyckman Andrus professor of pediatric surgery at Children's Hospital, believes could help hundreds of children at risk for developing mental retardation. The story begins in the early 1990s when Folkman and his colleagues noticed that occasionally children with very large hemangiomas—masses of tangled blood vessels and vascular tissue—exhibited very low levels of thyroid hormone before any treatment. Knowing that low thyroid function can cause mental retardation, Huang, a clinical fellow in pediatrics at Children's, administered thyroid hormone to the young patients but was surprised to find that even with high doses, their hormone levels barely budged. At the suggestion of HMS professor of medicine Reed Larsen, Huang and his colleagues began looking for an enzyme that might be destroying the hormone. They found that the hemangiomas were producing just such an enzyme, 3-iodothyronine deiodinase. "So the point is you can give high doses of thyroid hormone because most of it is degraded," said Folkman. "Then when you treat the tumor with drugs like interferon and the tumor starts to regress and the abnormal enzyme decreases, you lower the dose of thyroid hormone." "This is why science is so exciting," he said. "Who could have predicted that taking care of hemangioma patients would lead to a new way to prevent mental retardation?" —Misia Landau |
