American Association for Cancer Research

Anti-Angiogenesis Drugs


Normalization of Blood Vessels


Let me give you the rationale of why I was thinking that anti-angiogenic therapies would normalize tumor vessels. In our body, normal tissues contain certain proteins called pro- and anti-angiogenic molecules. The pro-angiogenic molecules help blood vessel formation and anti-angiogenic molecules do the opposite. In normal tissues the effects of pro-angiogenic molecules are exquisitely counterbalanced by those of anti-angiogenic molecules. This perfect balance maintains the blood vessels in a “normal” state.

But a tumor is bathed in pro-angiogenic molecules, such as, vascular endothelial growth factor (VEGF). VEGF is one of the 20 or 30 pro-angiogenic molecules. As a result, the balance is tipped in favor of pro-angiogenic molecules, and the blood vessels that form are not normal. As I mentioned earlier, they look abnormal. So my thinking was to use drugs that can bring the system back to balance and try to recover the normal features of blood vessels.

In 1995-96, my colleagues and I gave Avastin to mice, and the tumors began to shrink. In mice that weren’t treated, the tumors continued to grow. The problem was, we found that if we kept giving Avastin, eventually the tumors began to grow again, because there were cells left that kept making angiogenic molecules.

In 1998, we were studying tumors in mice that were hormone-dependent, like breast cancer that depends on estrogen or prostate cancer that depends on testosterone. For prostate cancer, we looked at the effects of surgical hormone-withdrawal, which is like giving an anti-hormonal therapy. This lowers the levels of VEGF in tumors and causes the tumor’s blood vessels to regress. In 1998, we noticed that before the engorged, abnormal-looking vessels in the tumor died, they began to look more like normal vessels. So hormone therapy was showing effects on tumor vessels similar to those induced by Avastin. When I wrote the hypothesis about normalizing blood vessels in 2001, this 1998 data was in the back of my mind.

Again, in 2001-02, we were examining the effects of the drug Herceptin (trastuzumab) on breast tumors in mice. This drug also caused the blood vessels to become normalized—though not by lowering VEGF, but through another mechanism. Therefore, we hypothesized that when Herceptin is given with chemotherapy, the results should be better than when the treatments are not given together. Again, these are all animal data.