Frank McCormick's Interview Transcript of 5-19-98
Intro:
Cancer is one area of medicine where treatments are needed desperately. As we saw in the media and public response to the angiogenis story. Gene therapy is being pursued very actively as a treatment of cancer. We have here one of the world's leading researchers to talk with us. Dr. Frank McCormick has had an interesting and distinguished career path. After post-doctoral research, he moved immediately to biotechnology, first at Cetus and then as a co-founder of Onyx Corporation. He is currently Professor of Microbiology and Immunology at the University of California at San Francisco and Director of the Cancer Center there, where he continues his research in gene therapy of cancer.
1. Dr. McCormick, we've been hearing about using gene therapy for treating inherited genetic disorders. What makes cancers attractive targets for gene therapy.
Frank McCormick: While we have a great deal of knowledge relating to the molecular mechanisms that cause cancer, understanding these mechanisms really offers many opportunities for intervention using gene therapy-related approaches. One set of approaches involves replacing tumor suppressor genes lost during the evolution of the tumor with their wild-type normal counterparts, resulting in death of the tumor cell, or perhaps _______ of the rest of the tumor cell. This is one approach that is being used by several different groups at the moment in various clinical trials. Another approach is to deliver enzymes which convert prodrugs into toxic drugs locally in the tumor and thus allow local tumor destruction. A third approach would be try to increase the immune response of the host to the tumor cell by introducing cytokines or co-stimulatory molecules that could effectively increase the ability of the host to recognize antigens on the tumor cell or increase the ability of the immune system to kill the tumor cell. These are rather different kinds of approaches, but they are all based on some level of understanding of what actually causes cancer and how cancer cells are different from their normal counterparts.
One fundamental difference between gene therapy for cancer and gene therapy for diseases such as Cystic Fibrosis is that the goal for the cancer treatment is to kill the cancer cell, whereas obviously for a disease such as Cystic Fibrosis the goal is to replace a defect in an effected tissue and have the cell that's being treated survive long enough to provide a benefit to the patient. This means that for survival, the treated cell has to evade any kind of immune recognition by the host. This means that the vector that's being used to deliver the gene not be recognized by the host as producing foreign material, otherwise the cell will be killed. Keeping the cell alive to produce a transduced protein is a very important issue for most types of gene therapy. For cancer, the goal is to kill the cancer cell. If the immune system recognizes the infected cell as presenting new antigens from the viral vector, then the tumor cell will be killed. Obviously this is potentially a good thing for tumor treatment. The end points and the goals are very different when treating cancer compared with other more perhaps conventional uses of gene therapy.
2. What are the genetic targets for cancer cells?
Frank McCormick: The tumor suppressor-genes offer very attractive targets. During the evolution of a tumor, the tumor has to eliminate certain functional genes in order to survive and develop as a tumor.
It's very clear from experiments in the laboratory that if one replaces these genes that are being selectively lost in tumor cells with their normal wild type counterparts, the tumor cell loses malignant or actually, in some cases, die. Perhaps P-53 is the best example of this. Most tumor cells, as they evolve, lose functional -53, generally as a result of mutations in the P-53 gene itself, but by other mechanisms. It seems that the major selection for P-53 is to prevent programmed cell death or apotosis which is driven by P-53 itself. The tumor cell escapes from cell death by and eliminating P-53. One introduces back into these cells type P-53 genes; then the tumor cell will die. Another beauty of this kind of approach is that if the viral vector delivering wild type P-53 normal tissue, as it undoubtedly would, then we can expect that this will have very few important consequences to the normal tissue because these cells in the normal tissue already express wild type normal P-53. Delivering more of the same is not expected to have much of a side effect or any kind of toxic effect on normal tissue. There is a tremendous therapeutic potential, delivering wild type tumor suppressing genes into tumor cells, resulting in their destruction.
3. Your approach is rather different from what we think of as gene therapy in that you are using viral genes rather than human genes. How does it work?
Frank McCormick: Our approach has been to engineer viruses so that they actually replicate in tumor cells selectively. When these viruses replicate in cells, they actually kill the host cell as a result of their normal replication cycle. By engineering the virus so that it only replicates in cancer cells, obviously it only kills cancer cells as a result of its natural replication process. The approach that we've developed most thoroughly so far depends on loss of functional P-53. We have a virus referred to as Onyx-015 which can only replicate efficiently in cells which lack P-53. This typically would be a tumor cell, obviously. We've achieved this level of selectively by actually engineering the virus, or taking a virus in which a particular gene is missing. This is a gene referred to as E-1B55K. This gene is essential in normal virus infection of normal cells because when viruses infect normal cells P53, which is referred to as the guardian of the genome, is alerted, activated during infection and would actually try to prevent virus replication, prevent the virus taking over the cell. The virus has evolved a strategy to eliminate P-53. It does this by making the E-1B55K protein. The virus enters the normal cell, P-53 is alerted and then the E-1B55K protein binds to P-53 and suppresses its activity, and thus allows the virus to go ahead and take over the cell and replicate. What we've done is taken a virus that did not make the 55K protein. As a result this virus cannot grow in normal cells because P-53 stops it. In a tumor cell, however, where the P-53 gene is inactivated, is missing, there is no barrier to replication of this virus. The defective virus, the E-1B mutant virus, can replicate quite efficiently in a tumor cell. As a result it kills the tumor cell through replication, spreads through the tumor, infecting one tumor cell to the next until eventually it reaches normal tissue. Then its replication stops because P-53 itself prevents further replication.
4. This is a very unique strategy, a very unique trick. Can you tell us how it works in practice and what is the current status of the clinical trials?
Frank McCormick: We showed in vitro or in cell culture that the specificity of the virus is indeed restricted to the cells in which P-53 is defective. We then showed in animals that the virus can actually kill human tumor cells grown in mice, and then proceeded into Phase I clinical testing, the point of which is to establish a safe dose of the virus, a maximally tolerated dose. We completed Phase I testing of the virus several months ago in head and neck cancer. We showed that even very high doses of the virus - 10 to the power of 11 infectious viral particles per tumor - gave us no dose-limiting toxicity and no significant side effects. Having proven that the virus is safe, we then moved along into Phase II clinical trials looking for a sign of efficacy and clear patient benefit. We have now completed Phase II trials in head and neck cancer with the virus alone or in combination with the standard approved chemotherapy for head and neck cancer. The results of these Phase II trials are due to be announced within the next month or so. That's for head and neck cancer. We've also started Phase I trials for the virus in pancreatic cancer and for cancer of the ovary. In the case of pancreatic cancer, the virus is directly injected into the tumor as it is for head and neck cancer. In the ovarian cancer trials the virus is administered into the peritoneal cavity and is allowed to infect to infect tumor cells within the peritoneal cavity.
5. This requires you administering the virus directly to the tumor. Can it be adapted to deal with metastases?
Frank McCormick: We know from our work in animal models that, if one injects virus into the bloodstream of a mouse, the virus will actually leak out of the vasculature into the bed of a tumor grown at a distant site and will actually infect the tumor cells and cause some destruction of the tumor tissue. Unfortunately, most of the virus gets cleared we think through the liver in this kind of administration. Currently the delivery of virus to a distant tumor or metastasis is not very efficient by direct injection of the virus into the bloodstream. However, it does seem that once it gets there it does actually replicate and spread quite well even though very small amounts of the virus actually find these distant tumors. We are currently trying to develop strategies for improving the efficiency of delivery from the intravenous route into tumors at distant sites. If we are successful in this, obviously then we would open up the potential applications from just local therapy through regional therapy into systemic therapy. That's obviously for the future. Currently we are focusing on local treatment. The kinds of diseases from which we can get benefit by delivering the virus locally include ovarian cancer, cervical cancer, possibly brain cancer, head and neck cancer for sure, and maybe other kinds of tumors where the effects, or some of the symptoms of the disease are caused by local obstructions and local tumor masses. This actually quite a substantial number of potential patients who could be treated with benefit if it works locally. Obviously, attacking metastases is a long-term goal for this kind of approach.
6. Finally, you've told us about one gene therapy strategy in fighting cancer. What other lines of research have been developed?
Frank McCormick: I mentioned that there are attempts underway to wild type P-53 to tumor cells using adenovirus or other kinds of to simply deliver the wild type gene to the tumor. The limitation to this approach is that one needs to infect every cell in the tumor when we expect to get significant benefit. Infecting a small fraction of the cells in the tumor you would not think would have tremendous benefit. On the other hand, increasing the doses of such an agent to saturate all the cells in the tumor, one would also expect to infect normal tissue surrounding the tumor. As I said before, we don't expect that this would have really serious side effects. There may be technical problems in infecting all the cells in the tumor, but the agent should be safe. It should be possible to go in with very high doses of virus to get the kind of efficacy that one would need to destroy a solid tumor mass. There are Phase I, possibly Phase II trials already underway at various sites to deliver wild type P-53 to tumors in which P-53 is missing. Similar approaches are being used for the RB retinoblastoma tumor suppressor, although this is not being pursued by as many investigators as P-53. There are a number of trials underway or being planned to deliver prodrug activating enzymes such as cytocendianmase into tumors, the idea being that this enzyme will then convert an inert compound into a toxic compound and in that case 5-FU. This will be converted locally and thus produce local benefit by producing the toxic agent just in the site of the tumor. Several studies are underway in this direction. A number of studies are underway to try and make tumor cells more immunogenic by introducing cytokines which will increase the local inflammatory or immune response to potential antigens or co-stimulatory molecules such as B-7 type molecules which will make the tumor cell more recognizable to the immune system and help the immune system to actually kill the tumor cell, if indeed it does recognize it. These are all studies which are underway. I should say that if one looks at all the clinical studies which have been initiated in cancer over the last couple of years, I think actually the majority involve some kind of gene therapy protocol as opposed to the more conventional spore molecule drugs or other biologicals. There's a whole range of different approaches being used using different agents, different virus vectors and different kinds of indications. We can really say that gene therapy for cancer is certainly a reality. Over the next few years we'll see exactly how safe and efficacious these kinds of treatments are.
Thank you, Dr. McCormick, for such a concise and revealing view of gene therapy. I'm sure we all hope that these treatments will come to fruition.
|
