I once heard Judah Folkman talk about the discovery of Avastin and how, in some early trials, it wasn't showing any effects. It turned out that by the time they tested the drug in patients, there were four or five pathways working at the same time to grow blood vessels for the tumor—so if you looked at tumor growth, the drug failed; if you looked at the growth of the blood vessels, it failed; but it was, in fact, working on the pathway it was supposed to be working on.

Avastin is an interesting example, because it was indeed unlikely to have significant single-agent activity in the first set of patients treated. That would have been fantastic, but not expected. Most everyone assumed that the antiangiogenic therapies would be more useful in a combination setting. And, happily, that turned out to be true.

As we make progress in our understanding of angiogenesis, it is clear that the VEGF axis is going to continue to be really important. But there are other pathways that have been defined, so we'll see combinations of antiangiogenic compounds being used to treat patients. And if they're safe enough and effective, we'll see those marching up to approval of front line over a more accelerated time than more potentially toxic compounds.

This is an area of research that changes rapidly. For example, of the emerging themes now in oncology, number one on the list is cancer stem-cell population as a driver of long-term disease. The idea is that by treating the bulk tumor, you're not getting at the heart of the disease—the stem cells or progenitor cells or initiating cells driving the disease. And we're just starting to learn more about their biology. Another emerging theme is the idea of the normal tissue surrounding cancer cells—the so-called cancer stroma—playing important roles in the biology of different cancers. In so many ways, this is a fascinating field and progress truly is being made every day.