New compound unusually potent at blocking brain cancer growth

By determining how a class of compounds blocks signaling in cells, UCSF scientists have identified what is perhaps the most potent drug candidate yet against a highly lethal kind of brain tumor. The compound, known as PI-103, shows unique potency against cancer cell proliferation in studies of mice with grafts of human glioma cells. Gliomas are the most common form of brain cancer, and have proven very difficult to treat. The unique effectiveness of PI-103 stems from its ability to attack two separate steps in the series of signals that trigger the spread of cancer. The dual blockade proved to be a safe and effective inhibitor of cancer cell proliferation in mice with the human tumors, the scientists found. The glioma research is being published online May 15 by the journal Cancer Cell. A description of the strategy used to identify the molecular level action of the inhibitors was published online by the journal Cell on April 27. Food and Drug Administration approval five years ago of the cancer drug Gleevec marked a promising new strategy against cancer. Gleevec was the first drug on the market designed to block ubiquitous signaling molecules called protein kinases – enzymes known to trigger normal cell proliferation, and in the case of cancer, the growth of tumors. Another group of kinases, called lipid kinases are now emerging as important new targets, especially PI3 alpha kinase, an enzyme often found to be overactive in brain, breast, colon and stomach cancers. But the sheer number of related kinases – 15 in the PI3 kinase family alone – and uncertainty about how each acts in the body – has stalled progress. Broad spectrum drugs that inhibit many related kinases inevitably cause toxicity and are poor drug candidates. To overcome this hurdle, Kevan Shokat, PhD, a Howard Hughes Medical Institute investigator at UCSF, and Zachary Knight, a postdoctoral fellow in his lab, developed a strategy to systematically inhibit many different but related kinases to identify which ones might be prime targets to treat brain tumors. In the Cell paper they described their success synthesizing a panel of different PI3 kinase inhibitors, showing for the first time the structural basis of the inhibitors’ abilities to block different PI3 kinases. They used the new compounds to dissect the role of PI3 kinases in insulin signaling and in cancer. Drawing on this new tool, William Weiss, MD, associate professor of neurology at UCSF and an investigator in UCSF’s Comprehensive Cancer Center, developed the strategy to treat gliomas. These cancers are the most common solid tumor of childhood, and about half of the people diagnosed with gliomas die within a year of diagnosis. Weiss and his colleagues report in the Cancer Cell paper that one PI3 kinase inhibitor in particular – PI-103 -- is unusually effective against gliomas in mice. They believe the inhibitor is a promising drug candidate, and a UCSF neuro-oncologist is developing plans to launch a clinical trial within a year, Weiss says. The Weiss team discovered that the inhibitor's effectiveness lies in its dual impact. It inhibits both PI3 kinase and a protein kinase known as mTOR which acts "downstream" of PI3 kinase and is part of the cell’s nutrient-sensing system. Clinical trials using inhibitors of mTOR alone have had disappointing results, Weiss says. One reason appears to be that the two kinases are part of a feedback loop. His group showed that mTOR inhibitors in clinical trials actually activate PI3-kinase while they inhibit mTOR. In effect, the drugs are blocking and encouraging cancer growth at the same time. The new inhibitor offers a mechanism through which to block both the PI3 and the mTOR kinase pathways, a strategy that appears to be particularly effective at slowing growth of gliomas. Lead author on the Cancer Cell paper is Qi-Wen Fan, MD, PhD, assistant adjunct professor of neurology, in the Weiss lab. Co-authors along with Weiss, Shokat and Knight, all at UCSF, are David Goldenberg, staff research associate in neurology; Wei Yu, PhD, assistant research anatomist; and David Stokoe, PhD, assistant professor in the Cancer Research Institute. Shokat, UCSF professor of cellular and molecular pharmacology, is also a faculty affiliate in QB3, the Institute for Quantitative Biomedical Research. The research was supported by the Howard Hughes Medical Institute; the Brain Tumor Society, the Goldhirsh Foundation, the Waxman Foundation, the Sandler Family and the Brain Tumor SPORE Program at UCSF.