On a cold, gray Saturday morning at Yale University in February 1993, instead of just reading his laboratory’s article in a cancer journal and scanning past the rest—cancer is a profoundly wide field, and there is much to read—cancer biologist John Pawelek made time to finish the entire issue. That simple decision changed the course of his research, toward a controversial explanation for the deadliest aspect of the disease—namely, why it spreads.

The issue contained a letter from three Czech doctors asking whether the fusion of tumor cells and white blood cells could cause cancers to spread, or metastasize. At the time, Pawelek was also reading a book by evolutionary biologist Lynn Margulis, who pioneered the idea that life on earth was revolutionized by ancient cells engulfing one another and fusing together, forming hybrids that had better chances at survival. “I was really excited by the connection,” he recalls. “Since there was a precedent for hybridization in evolution, why not in cancer?”

In the past 15 years Pawelek and his colleagues have shown that cancer cells can fuse with white blood cells and become highly metastatic in lab animals. Now they are searching for the same activity in humans. The leading cause of death in cancer is metastasis—tumors are generally treatable as long as they have not moved to vital organs. So if the research reveals that such hybrids help cancer spread, it could open up new avenues to fighting cancer. As Pawelek puts it: “You have to know how metastasis starts to properly fight it.”

But maddeningly little is known about what makes cancer spread. Explaining the capabilities that metastatic cells possess—to break away from their original tumors, migrate past other cells, travel around the body via lymph or blood vessels, invade tissues, and grow—would involve understanding how cells interact with one another. “And we don’t yet have good biological tools to investigate the interactions between different cell types and organs,” explains cancer geneticist Bert Vogelstein of Johns Hopkins University.

One theory behind the origin of metastasis is that mutations in one or a few genes cause tumor cells to gain the ability to migrate. Another idea suggests that no specific mutations are needed—rather cancer cells eventually accumulate abnormal numbers of chromosomes that break down the constraints that keep any normal cell from metastasizing. In Pawelek’s view, these theories do not explain how cancer cells would acquire the right genetic changes in the right order needed to spread successfully.

Instead the 66-year-old Paw­elek suggests that cancer cells spread after fusing with white blood cells known as macrophages. Like metastatic cells, macrophages can roam around and infiltrate most parts of the body and are naturally resistant to toxic drugs. “Metastasis is a very different phase from ordinary cancer and to me is almost like a new disease superimposed on a preexisting cancer cell—maybe cancer cells inherit all these traits at once by hybridizing with white blood cells,” Pawelek speculates. Moreover, macrophages regularly engulf germs and unhealthy cells—they might occasionally fuse with tumor cells instead of destroying them, much as ancient cells once joined together into symbiotic relationships a billion years ago, he reasons.

In their first experiments, Pawelek and his colleagues took a strain of mouse melanoma cells known to be only weakly metastatic and fused them with mouse macrophages by exposing them to polyethylene glycol, which can dissolve cell membranes. They implanted these hybrids in roughly 5,000 mice. “These were massive experiments that took four years to accomplish, and we were just going on faith,” Pawelek recounts. The results were striking—roughly 55 percent of the hybrid cells proved “really, really deadly, very metastatic,” he declares, in contrast to melanoma cells fused with one another—none of them became metastatic. “I was convinced we were on to something.”

Pawelek and his colleagues have also found molecular similarities between metastatic cells and macrophages, such as activation of genes linked with movement. In addition, they discovered that these cancer cells apparently produce organelles known as autophagosomes, with which the cells can digest chunks of themselves. Macrophages often produce autophagosomes as a means to snack on themselves while traveling, and cancer cells may do the same.

Pawelek is not the first scientist drawn to the fusion theory. Its earliest proponent was German pathologist Otto Aichel, who proposed it in 1911. It revived again some 50 years later, thanks to experiments showing that implanted tumor cells could spontaneously fuse with cells in lab animals and spread. About the mid-1980s, however, interest faded. David Goldenberg, who conducted some of those studies and is now president of the Garden State Cancer Center in Belleville, N.J., suggests that attention waned because although scientists could see hybrids of foreign and host cells, the tools at the time could not show that such fusions were actually taking place in natural settings. After all, if fusions were happening in cancer patients, both the tumor cells and macrophages would be virtually genetically identical, making it hard to prove that metastatic cells were hybrids.

But now Pawelek believes that technology has caught up and that the evidence lies with cancer patients who received bone marrow transplants. Radiation therapy and chemotherapy kill off bone marrow, which supplies white blood cells to the body. Donated bone marrow would naturally be genetically distinct from the patient, making it possible to see if tumor cells of the host fused with macrophages from the donor.

So far Pawelek and his colleagues have found two possible examples. In one, a boy with type O blood received a bone marrow transplant from his type A brother, and when the bone marrow recipient later developed kidney cancer, the scientists found tumor cells that possessed blood type A. In the other example, a woman who received a bone marrow transplant from her son later developed kidney cancer, and the new tumor contained cells with the male Y chromosome. In both the human examples, however, the investigators could not confirm that the cells contained the host’s genome. It therefore remains possible that these cells were not hybrids but simply came from the donor. On future samples, Pawelek hopes to use forensic DNA analysis techniques that can detect genes from both host and donor in the same cells.

Any searches for hybrids are highly vulnerable to error, cautions stem cell biologist Irving Weissman of Stanford University. “I’ve seen this kind of thing over and over again—when you think you’ve seen a hybrid, it turns out there’s almost always a cell with another cell adhering to it or very close to it.” (Pawelek insists the researchers took care to ensure such mistakes were not made.)

Weissman also cites other studies that show hybrids are actually less cancerous, not more, when tumor cells are fused with normal cells, apparently because the infusion of healthy DNA helps to suppress malignant activity. Pawelek suggests that the cell type used may partially explain this discrepancy—fusions with white blood cells show enhanced malignancy, whereas cell types such as epithelial cells can suppress tumors.

But even if Pawelek shows that the fused cells become metastatic, Vogelstein says, researchers still have to see whether these hybrids account for a significant fraction of metastases or whether other mechanisms trigger most spreading cancers.

If Pawelek is right, investigators will have new ways to combat cancer. For instance, they could aim to develop drugs based on antibodies that attack fused cells that may have unique chemical signatures or to devise therapies that block molecules linked with cell fusion. “Even if he’s wrong,” Vogelstein remarks, “the pursuit of unconventional ideas often leads to novel insights.”

In the meantime, Pawelek’s work has inspired other labs—75 scientists attended the first meeting on cell fusion and cancer in Sweden in October 2007, where Pawelek was a featured speaker. The hybrid theory, it seems, is spreading once more.

Note: This article was originally printed with the title, "A Theory of a Deadly Fusion".