WEDNESDAY, FEBRUARY 11, 1998
"What's the Latest in Anti-Angiogenesis Therapy for Breast Cancer?"

History and Physiology:

About 20 years ago a strong relationship was found between cancer and the growth of blood vessels. Harvard surgeon Judah Folkman put forth this idea in the mid '70's and was roundly criticized by the scientific community. This was described as an "epiphenomenon", something like the alternative therapies of today, an association which defies explanation by scientific means and therefore should not be pursued in a serious way. Nevertheless the microscopic evidence of this blood vessel/tumor relationship was very compelling, and about 10 years later scientists had begun to identify growth factors which let blood vessels grow in tumor cells. We now have identified a variety of these factors such as basic fibroblastic and vascular endothelial growth factor, as well as the natural inhibitors of this process. Thus a new model has emerged that under certain physiologic conditions, like fetal development, the body could be induced to take a new layer of cells and turn them into blood vessels (vasculogenesis), or just have blood vessels mature and branch out (angiogenesis). These conditions could develop under normal condtions, or pathologic ones in which the tumor cells could induce the host to send blood vessels into the tumor, to feed it. So for a given tumor cell there is a balance pro and against angiogenesis, which are regulated positively and negatively by different sets of proteins. There is a constant balance which can be broken under certain situations. Natural inhibitory substances which counteract angiogenesis have been isolated as well. This creates both a lot of questions and a lot of opportunities for therapy.

Questions:

• How does a cancer cell induce such a reaction?
• What stimulates a cancer cell to become highly vascularized?
• What is the relationship to metastasis?
• How can interfering with this system be used therapeutically?

Right now there is an extensive cataloguing of the players in the angiogenic pathway. We have identified growth factor (made by tumor) and growth factor receptors which exist on blood vessels cells. When these substances bond, blood vessels are formed. The process also requires enzymes that breakdown basement membranes so that blood vessels can penetrate to the epithelial cells from which most cancers arise. These growth factors can be induced by the tumor. We know that if you make a tumor hypoxic (deprive it of oxygen), one of the reponses is to create vasculoendothelial growth factor. If you make antibodies to this factor, you can arrest tumor growth in an animal model. This strategy is being explored by a number of research groups, but so far there is no clear clinical benefit.

• The first experiment occurred in the laboratory. Mould was growing over a chicken egg in a petrie dish, technician noticed that the blood vessel density was much less around the ring of fungus. The active compound, TNP 470 was isolated. This was the first anti-angiogenesis agent to be isolated. Results from clinical trials have not been dramatic. Researchers are now looking at it in combination with chemotherapy.
• Two very promising naturally occurring anti-angiogenetic compounds are angiostatin and endostatin. Both of these proteins seem to be quite potent inhibitors of angiogenesis, but are not yet in clinical trials.
• Integrin, a matrix protein, is in trials and may be promising, but is still in the dose finding stage.
• Tamoxifen and interferon have been found to be mild inhibitors of angiogenesis.

Side Effects:

Angiogenesis is quiet in the adult body, but active in he developing embryo or in a healing cut or fractured bone. Vascular endothelial growth factor is expressed at high levels as blood vessels are being developed. No clear cut side effects have emerged, but as our bodies are constantly remodelling, it is possible that upsetting the balance in bone growth could cause problems. Too little could cause osteoporosis, too much could cause Paget's disease. Another possible side effect could be impairment in wound healing. There is some evidence of vision toxicity with TNP 470, this is a real concern, because in vision some new blood vessel growth is necessary to maintain the eye and its function.

Clinical Trials:

Other molecules being looked at in clinical trials in Europe include Thalidomide which is anti-angiogenic in high concentrations. At NCI they have had some success with brain tumors, which are highly vascular. In breast cancer, patients could not tolerate the high doses, they developed neuropathies in their hands and feet. The question may be how much effect must these drugs show in order to be approved. As there currently is no really good treatment for breast cancer, the answer may be that long, randomized trials are not needed. HER2/neu, for example may be approved just as an antibody alone even though the percentage of patients with positive response to it is about 12% The same thing may be true for angiogenetic agents. Very few are past Phase I trials, although VEGF is into Phase II with prostate cancer. Phase I trials take between 6-12 months to complete, Phase II trials a little longer.

Discussion:

With so much media hype about anti-angiogenesis, it is very frustrating for patients with metastatic disease to learn that the reality may be the therapy is being oversold. Patients would like the research to move faster and they want to be given facts--not hype. But they are not fragile, it is exciting to know that there is a next step, an alternative to chemotherapy.

Patients would like to have a centralized location where this type of information would be stored. It would save them a lot of time, and disappointment when many new hopes don't turn out to be real. Dr. Tripathy feels that while hype is irresponsible, scientists want to attract attention to their programs, drug companies want to please their investors. It does make it hard to prioritize the search for real information. You not only need a central repository for information, but also a guideline as to how to process what is real. Our newsletter will be an attempt to remedy part of this problem. We will try to cover individual topics, like angiogenesis or tissue research, giving a reasonable approximation of what's really happening. This is difficult because most of the information is not published until the trial is over. It is a guessing game until published results are available, but at least we can make an attempt to dispel fantasy and misleading information.

The FDA Reform Bill will have a section on a unified database for NCI and private pharmaceutical trials. Patient advocates should call and demand a seat at the table for defining the database. This type of activism certainly worked for AIDS and should for breast cancer. Although this database will have information on trials, it won't have information or analysis of new technologies. Universities or non-profits will have to step in and provide a cogent way to sort through this complex information. Another possibility is that anyone who wants to submit to the database would be required to submit a lay abstract and periodical update on results of clinical trials. Usually the drug companies have a clinical trial manager whose job it is to give the status and updates on their trials.

Why are there no clinical trials between diagnosis and recurrence? Why not treat women with another dose of chemotherapy when the remaining cancer cells are few rather than wait until they have a recurrence? Biologically, this makes sense, the problem is it is very expensive to run large trials on new drugs, and you need a large number of patients to notice an incremental change. No drug company will fund this type of research with no preliminary evidence that it will work. Vaccines may be good candidates for this type of therapy if and when they are shown to be effective. Also we might look at micrometastases to monitor how people are doing on a particular drug, develop a reliable surrogate marker, and see if administering the drug correlates with successful treatment.

Dr. Bernard Fisher almost ruined his career when he proposed randomized clinical trials to compare modified radical mastectomy with the standard of care radical mastectomy, but he completely changed the way surgery was done on breast cancer patients. Dr. Tripathy thinks we will have the same impact when we finally shift away from hormone and chemotherapy.

Clinical trials can sometimes be misleading. Trial looked at drug (Interluken) which would cause platelet counts to recover. They designed a high dose chemotherapy regimen which would get the platelet counts to come down, and then tested the drug to see if it could support the platelet count. The drug company misled patients to think it would help their cancer, when in reality, there was no direct benefit to the patient. To thwart this type of behavior there should be full disclosure of everything found in a trial-good and bad-once the data is found to be reliable. Once a trial is finished, the results should be published regardless of whether the outcome is positive or negative. This is not now a requirement.

It is very hard for physicians to ferret out information on new drugs today. Drug companies are more likely to provide this information to their investors than to patients. What will make physicians use any new technology (decision support tools) that helps this process? Our research indicates that physician barriers to recommending clinical trials for their patients are: perceived loss of patients to large research centers, changes in referral patterns, too much time expended by the staff. (From patients point of view there are also many concerns: randomization, toxicity, cost, extra visits.) Trials put on by pharmaceutical companies cover the cost of data manager support, NCI and other government trials which traditionally have many patients do not have information support. There are industry standards for clinical trials. FDA does require certain guidelines which are standardized internationally as to safety, informed consent, grading toxicity, etc. Additionally, unexpected hospitalizations and death must be reported to the IRB within 24 hours and this information is then passed on to every investigator.