Why brain tumors are hard to treat

New findings may help explain why glioblastomas, one of the most aggressive forms of brain cancer, are often so difficult to treat -- they can feed themselves, by differentiating into the intricate network of blood vessels that nourish the tumor, according to two studies published online today (21 November) in Nature.

Macroscopic pathology of Glioblastoma multiforme
Image: Wikimedia commons,
Sbrandner

These results may help explain the failure of some anti-angiogenesis therapies used to fight glioblastomas, as well as point to new possible cancer treatments that target tumor-derived angiogenesis. "The general idea is that the vasculature is created by the normal tissue," said cell biologist Angelo Vescovi of the linkurl:Mendel Institute for Genetics in Rome,;http://www.css-mendel.it/home.asp who was not involved in the research. But these papers show that "the tumor itself is actually making its own blood vessels. It's a very different way to look at them."Glioblastomas are noted for their unusual vasculature, characterized by large, highly proliferating cells and abnormal structures. Until now, however, these vessels were thought to be derived from pre-existing brain capillaries, but no one had ever gone in a taken a close look, said neurosurgeon and neuroscientist linkurl:Viviane Tabar;http://www.mskcc.org/prg/prg/bios/641.cfm of the Memorial Sloan Kettering Cancer Center in New York. So she and her colleagues did just that. They looked for mutations known to exist in the tumor cells, including large amplifications and chromosome abnormalities, in the cells of the vasculature to see if they contained a similar genetic signature. Sure enough, a significant proportion of the endothelial cells carried the tumor's mutations, suggesting they originated from the tumor cells.Meanwhile, on the other side of the Atlantic Ocean, physician and cell biologist linkurl:Ruggero De Maria;http://www.iss.it/site/attivita/issweb_istituto/ricercapersonale/dettaglio.asp?idana=1736〈=1 of the Istituto Superiore di Sanita in Italy was working with his colleagues to show the exact same thing. Scouring the genomes of the tumor vasculature for aberrations found in the tumor cells themselves, the team found that anywhere from 20 to 90 percent of the blood vessels were indeed tumor-derived. The group was "very surprised," De Maria said. "I did not expect to find it so obvious in a way," Tabar agreed. "It was nice to see another group independently reached the same conclusion."To confirm that the tumor cells were indeed capable to differentiating into the mature endothelium of the vasculature, the researchers isolated tumor cells containing markers suggestive of a stem-cell-like state and confirmed their ability to differentiate into endothelial cells. The teams then injected the cells into immunosuppressed mice and analyzed the resulting tumors. Once again, the results were the same: A large percentage of the blood vessels that developed in the tumors were of human, not mouse, origin, indicating that they were derived directly from the tumors themselves."The bottom line is we were able to identify blood vessels of human origins in the brains of mice carrying the tumors," Tabar said. "These vessels came from the glioblastoma cells."The findings may explain why a common drug used to inhibit angiogenesis in cancer patients may not be very effective, she added. After further experiments, "we realized that this drug is not able to inhibit or stop the glioblastoma cells from making blood vessels at all." The drug appears to stop the endothelial cells from maturing, but not the formation of the blood vessels themselves, she explained. And when De Maria and his colleagues killed the endothelial cells produced by the cancer stem cells, they "observed a stronger regression of the tumor," suggesting tumors need these tumor-derived blood vessels, he said -- and pointing to new potential targets for future drugs that block this process. The applications may be more widespread than just glioblastomas, Tabar added. "All tumors have blood vessels," she said. "I'm hoping that this will invite researchers to look in other cancers to see if other cancers can use this pathway to derive their own blood vessels."Researchers also have to learn more about the process itself. Tabar's group showed, for example, that Notch signaling appears to be important for the differentiation into endothelial progenitors, while the VEGF signaling pathway -- targeted by current anti-angiogenesis drugs -- is key for the maturation into adult endothelium. But "this is just the starting point," De Maria said. "We need to determine the pathways and the requirements that push [these cells] towards one pathway -- the production of cancer cells -- or the other pathway -- the production of endothelial cells.""These kinds of functional studies will reveal that you have very different subtypes within the glioblastoma," Vescovi agreed. "That will tell you something about how the [tumor developed] and how to treat it."R. Wang, et al., "Glioblastoma stem-like cells give rise to tumour endothelium," Nature, doi:10.1038/nature09624, 2010.L. Ricci-Vitiani, et al., " Tumour vascularization via endothelial differentiation of glioblastoma stem-like cells," Nature, doi:10.1038/nature09557, 2010.
**__Related stories:__***linkurl:Yeast: angiogenesis model? Yup;http://www.the-scientist.com/blog/display/57252/
[23rd March 2010]*linkurl:Cancer research, stimulated;http://www.the-scientist.com/blog/display/55649/
[21st April 2009]