UCSF's Frank McCormick: A Multidiscipinary Focus on Cancer

UCSF's Frank McCormick: A Multidiscipinary Focus on Cancer

What was your research strategy going in?

McCormick: At the time we knew several different pathways were mutated in cancer cells and played a causal role—ras and p53 were the major ones, then retinoblastoma, which is the rb pathway and which was thought to affect 20% to 30% of all cancers, and also the APC gene, which was known to be mutated in colon cancer. Our idea was to develop research programs in each of those areas and identify drug targets, screen them ourselves initially or, as we became more savvy, set up research collaborations with pharmaceutical companies where we would provide the protein targets and they would do the high-throughput screening.

Frank McCormick's Highest-Impact Papers
Published Since 1990
(Ranked byaverage citations per year)

Rank	Paper	Total Citations	Avg. cites per year
1	H.R. Bourne, D.A. Sanders, F. McCormick, "The GTPase superfamily: Conserved structure and molecular mechanism," Nature, 349(6305):117-27, 1991.	1,381	184
2	M.S. Boguski, F. McCormick, "Proteins regulating ras and its relatives," Nature, 734(6456):643-54, 1993.	734	163
3	H.R. Bourne, D.A. Sanders, F. McCormick, "The GTPase superfamily: a conserved switch for diverse cell functions," Nature, 348(6297):125-32, 1990.	1,123	132
4	S.J. Cook, F. McCormick, "Inhibition by cAMP of ras-dependent activation of raf," Science, 262(5136):1069-72, 1993.	463	103

SOURCE: ISI's Personal Citation Report, 1981 - June 1998

What kind of success have you had?

McCormick: We got a very good start with a five-year partnership with Bayer on the ras work. A number of compounds are being evaluated in animal models based on what came out of that program. And we have a similar kind of program, slightly smaller, evolving out of the rb pathway with Parke-Davis. We've also started another project on the BRCA gene with breast cancer with Eli Lilly.

How did this lead to the idea of using a cold virus to attack tumor cells? Tell us about that work.

McCormick: We had been trying to come up with drug-discovery approaches to p53. In late 1992, I went off to a conference in Woods Hole, Massachusetts to discuss the role of rb and different viral proteins and transformation. On the plane back, I had the idea that we might be able to make a mutant virus that would only grow in cells that lacked p53, which would be cancer cells. A different version could be made that only grows in cells that lacked rb.

The fundamental observation was that one of the viral proteins of adenoviruses, E1a, which is the first protein that the virus makes when it infects cells, binds to the rb protein. This led to the idea that these viral proteins target cellular proteins. The question being discussed at the Woods Hole conference was, how do you know the main function of E1a is to bind rb? It definitely binds it, but how do you show that it's really important? The traditional approach had been to make mutants in E1a that no longer bind rb. The prediction was that, if binding was the major function, such a mutant would be biologically inactive. Such mutants were made and everything was consistent. But it didn't prove that it was the major function. So I thought a better experiment would be to say that if the major function of E1a is to neutralize rb, then a cell that lacks rb shouldn't need E1a anymore. If the cell mutant doesn't have rb, than E1a should be dispensable. That led to the idea that in an rb-negative cell, an E1a-negative virus should grow quite happily. Of course an rb-negative cell would be a tumor cell, therefore an E1a-negative virus should grow quite happily in a tumor cell but not in a normal cell because rb would stop it. Then there's a second viral protein, called E1b, whose function is to bind p53. By the same logic, an E1b-defective virus should grow in a cell that lacks p53 and not in a normal cell.

What happened when you got back to Onyx?

McCormick: First I bounced it off some of our advisors and colleagues, and then we hired a consultant named Lori Rafield to round up as many mutant viruses from our colleagues in the field as we could. Some of these mutants of E1a and E1b had already been made for basic science purposes. Lori worked with many people in the field, particularly Arnie Berk at UCLA, to collect mutant viruses and cell lines and test whether the mutant viruses would grow in mutant cell lines. Based on those early results it looked promising. We then hired a very smart scientist named Ali Fattaey, who was finishing up his postdoctoral work in Ed Harlow's lab at Harvard. He did a more formal analysis of this idea, and spent a year or so really testing the hypothesis rigorously.

Where does it stand now?

McCormick: It's in phase II clinical trials for head and neck cancer and phase I clinical trials for pancreatic cancer, ovarian cancer, and metastasis to the liver. We established a while ago that even at very high doses it's completely safe. What we found is that the virus alone injected into these larger end-stage head and neck cancers does have biological activity, and we're getting response rates in the 30% range for patients who respond significantly. We had two complete responses in the first phase II trial, which means basically all the tumor mass injected with virus was eliminated. We also tested this agent in combination with existing chemotherapy drugs. It either sensitizes cells to the chemotherapy or the chemotherapy sensitizes cells to the virus. Now we’re seeing really amazing results with this. In the first ten patients treated with existing protocol for head and neck cancer plus injection of virus, nine out of ten showed a major response; two complete, and seven more showing greater than 50% of the tumor mass, and the remaining one had 40% reduction in the tumor mass. It's much better than the current therapy alone and doesn't have any added side effects. We’re pretty excited. Actually, we're really excited.

So why leave Onyx at this point?

McCormick: UCSF made me a dream offer: a whole new cancer research institute at UCSF, a new research building, four floors of lab space to be filled, lots of funds to recruit people—starting with 15 new faculty—a big commitment from UCSF to get into cancer in a big way. I get an endowed chair myself, a very nice professorship for life, and resources to build my own lab and also to recruit the hottest people in cancer research in a whole new institute at—in my opinion—the best university in the world. And it's close enough to Onyx that I can stay in touch and visit once a week.

What is your plan for the institute?

McCormick: The reality of UCSF is that there is already tremendously strong basic sciences doing all kinds of fantastic stuff related to cancer, and equally strong clinical activity in the hospitals, and very good cancer care, but those areas have diverged over many years, as they have in most hospital medical schools. Clinical people are so busy they don't have time to get into research, and vice-versa. We're trying to build a number of very strong programs in specific cancer areas, and to have a multidisciplinary approach to each disease, including lab sciences, clinical sciences, epidemiology, patient outreach and genetics. My vision is that in five years we'll have perhaps ten of these programs covering each of the major types of cancer, and very strong technology underpinning each of those programs in terms of the latest genetic analysis of tumors, the latest animal models, and so on. Each program will have its own leadership and its own resources to build the program as the members themselves see fit. The main goal is to make resources available to bring the basic scientists into the clinical area, and also free up the clinicians, to get more involved in the lab sciences. They can meet in the middle. That's the most efficient way to get scientific discoveries translated into the clinic and also to make sure that the clinical work that goes on has a very strong foundation in lab science.