As clinical researchers tout triumphs and divulge failures at the 2007 American Society of Clinical Oncology meeting this week, basic researchers toil to discover the targets and mechanisms that may show up in the headlines five years hence.

From bevacizumab (Avastin, Genentech Inc.) on down, VEGF targeting was the subject of a large number of presentations at ASCO. Two recent papers expand the knowledge about how VEGF signals, and how blocking the VEGF receptor might be useful in earlier stages than the metastatic cancers for which anti-VEGF therapy has shown a benefit.

In a paper now available online in the Proceedings of the National Academy of Sciences, researchers from South San Francisco-based Genentech, report on the impact of anti-VEGF antibodies on benign or premalignant tumors. The researchers used anti-VEGF monoclonal antibodies to treat mice with a mutated APC gene - the gene that is mutated in a majority of non-inherited human colorectal tumors, as well as in people with familial adenomatous polyposis. The mice develop numerous precancerous polyps, some of which eventually turn into tumors.

Short-term treatment of mice for three to six weeks reduced their tumor burden in the intestines by roughly 75 percent. The antibody reduced the size, though not the number, of tumors. Longer-term treatment, for up to a year, extended the median survival from 24 to almost 34 weeks. Mice treated long-term did show toxic effects of the antibody in their kidneys, but the authors noted that the mice also showed reduced signs of anemia and intestinal bleeding, and "the adverse effects were outweighed by the overall improvement of health reflected by the increased median survival."

The authors concluded that "our data suggested the possibility of a nonsurgical treatment for benign tumors, even without the need of chemotherapeutic agents," though they also noted that given the side effects they observed in their long-term treatment group, "selection of an anti-VEGF agent with the appropriate balance of efficacy/toxicity may be critical for such long-term treatments."

Meanwhile, a paper in the May 7, 2007 issue of the Journal of Cell Biology reported that VEGF is not confined to its own receptor, but also can signal through platelet-derived growth factor receptors. The authors, at the University of Manchester in the UK, used what a reviewer termed "an impressive array of mutually confirmatory experiments" to demonstrate that VEGF-A can signal through two types of platelet-derived growth factor receptors as well as its own VEGF receptors, leading to migration of adult mesenchymal stem cells and fibroblasts. Because the former are actively recruited during tumor vascularization, the authors concluded that VEGF signaling through platelet-derived growth factor receptors is likely "important determinant" of such recruitment.

The Ras protein is similarly familiar to clinical researchers. It is a GTP-binding protein that interacts with several different enzymes; pathways activated by those enzymes, in turn, are hyperactive in roughly 30 percent of tumors.

Companies including Genentech and Exelixis, which penned a deal for XL-518 in January, and Ecopia, whose Eco-4601 is in Phase I/II clinical trials, are trying to inhibit Ras-activated pathways to benefit patients (see BioWorld Today, Jan. 4, 2007, and Feb. 21, 2007.)

In the June 1, 2007 issue of Cell, though, researchers reported data suggesting that the interaction of Ras with the PI3kinase pathway could be a promising clinical avenue.

Ras already is well-known to control PI3 kinase, but the details of the interaction have been fuzzy. The authors of the Cell paper, who hail from the Cancer Research UK London Research Institute and the University of Helsinki in Finland, created a mouse with a mutation of p110alpha, the catalytic subunit of PI3 kinase, which prevented the interaction of Ras with PI3 kinase. While mice lacking such interaction throughout development showed defects in lymph system development, an inducible knockout, when they were bred with mice that have a defective Ras protein and that normally develop lung tumors at an early age, two wrongs made a right. The resulting offspring were highly resistant to Ras-induced lung tumors.

The authors concluded that "The strong reduction in Ras-induced lung tumor formation when [Ras-PI3 kinase] binding is prevented suggested that specific targeting of the interaction of Ras with PI 3-kinase may have therapeutic value in the treatment of tumors, such as lung, colon and pancreatic carcinoma, with high incidences of Ras mutation." They also noted that given the apparent good health of adult mice lacking Ras-PI3k binding, "drugs targeting this link might be expected to be well tolerated."