Era of Hope 2011 Plenary Sessions

Session Title: Controversy Session – Setting the Bar

Title of Presentation: Controversy Session – Setting the Bar

Sharon Begley, Scientific Journalist, Newsweek: Our last speaker this afternoon is Dr. Glenn Begley, no relation I’m sorry to say. Dr. Begley is Vice President and Global Head of Hematology and Oncology Research at AMGEN where he’s responsible for the ongoing research effort around molecules with established clinical utility in addition to earlier stage clinical programs; Dr. Begley.

C. Glenn Begley, MD BS, PhD, AMGEN© Oncology: Thank you; thank you very much for the opportunity of participating today. It’s a pleasure and a privilege for me to be here. What I wanted to do in this presentation is address two issues. We’ve heard throughout the week about the concerns about the—the relatively small advances that have been made in terms of therapies that I would like to make it clear that that’s not by choice or by design. It’s a consequence of the fundamental biology that we’re trying to interdict. And the second thing that I’d like to do is talk about areas where I think there are opportunities for improving outcomes for patients and improving research outcomes and I’ll give you an example of something that I think that we should be able to address together.

A couple of weeks ago I had the opportunity of presenting at Yale and followed two cardiologists, and I enjoyed showing this slide and reminding them that cardiology is pretty simple. It’s essentially plumbing that cancer is evolution. And the other advantage that cardiologists have aside from dealing with something that’s pretty simple is that most people believe in plumbing.

So I know you’re aware of this but as we think about why it is that we’ve made less advances, why therapies are more incremental than any of us want, it’s worthwhile just thinking about what’s the fundamental biology that we’re trying to interdict here. So on this slide here, on the horizontal axis, you can see the number of cells. Cancer begins in a single cell and beneath that you can see the mass of the tumor. And at the top of the slide, you can see the number of doublings that’s associated with this tumor growth. And on the vertical axis here you can see the growth rate.

Initially cancer cell is growth is exponential which means that there’s a short doubling time, a high-growth fraction, and little cell death. At about the size of a million cells which is really just bigger than a head of a pin, it’s a very tiny tumor mass—that’s the—the stage at which a tumor is required to generate its own vasculature because diffusion of nutrients is no longer sufficient. It’s also the time at which cells acquire their earliest metastatic potential. Then as few as 10 cell doublings later is the time when we can detect the tumor for the first time. That’s about 10⁸ or 10⁹ cells. At this stage the tumor growth is slowed and there is more cell death. And it’s around this time, 10⁸, 10⁹ cells that we can first detect a malignancy.

Clearly untreated then again a matter of a few cell doublings later death will occur with a tumor mass of about a kilo. So as we think about this, it’s this margin, this window, this area between 10⁸ and 10⁹ that’s all important. And in terms of tumor mass on the left-hand side of this line, we regard the tumor as being in remission and on the right-hand side of this, we regard the tumor as being relapsed. But again this boundary line is about a billion cells. And frankly we’re simply unable to detect a tumor mass much less than about a billion cells.

From the physician and the patient perspective, the history of cancer begins as this threshold is crossed, about 10⁸ and 10⁹ cells. And that’s when the history of cancer begins, but in fact there’s been this pre-history that’s gone on for potentially two or three decades during which time neo-vascularization has occurred, tumor metastasis has occurred, and we can only get a sense of this in retrospect and at the moment we have very little opportunity to intervene at this point of tumor development.

So the point that I would like to make really is that the incremental developments we see a consequence of the fundamental biology but what I’ve outlined on this slide is also an oversimplification. Let me try and expand on that idea. So if we’ve got a virgin white cell on the left, and it gives rise to its progenitor cells, one of these can acquire a genetic event that predisposes to cancer. This yellow cell here will be the one that is responsible for the cancer cell clone arising.

Then additional genetic changes occur, and ultimately a full-blown malignancy is seen. But again this is an oversimplification because the yellow cell can result in—can spawn multiple different sub-clones and the primary tumor that we evaluate ultimately is much more heterogeneous. And in addition, this sort of heterogeneity is also occurring at the metastatic sites, so you can have heterogeneity both within the tumor at multiple sites of metastatic deposits and within the primary. And it doesn’t even cease there because tumor evolution continues once we institute therapy.

So again the task that we’re trying to address here is enormous, and I think the advances that have been made over the last 20 years in terms of cancer therapy are something to be celebrated. Frankly in the—over 30 years that I’ve been involved in this area I know you’re staggered to think that someone that looks as young as I am could have been in this area for 30 years; in the 30 years that I’ve been in this area I’ve been frankly humbled by a profound level of ignorance about cancer cell biology and the therapies that we’ve thought would be dramatic for patients and that have failed time and again.

Believe me; we would do better if we could but the essential nature of research is it’s research. If we knew the answer before we started then it wouldn’t be research.

I’ve already told you that I’ve been in this area for more years than I’d like to remember. I spent the first 25 years working in academia as a physician looking after cancer patients. My particular focus was leukemia and breast cancer. I also ran basic research lab for 25 years. For the last decade, I’ve been working for the company AMGEN and frankly the single biggest shock I’ve had since joining AMGEN is the publication bias that I’d like to share with you now and I think provides an opportunity for us to reset the bar.

In essence, we get what we incentivize and here is an opportunity to perhaps institute some change.

So over the last decade at AMGEN, there have been over 50 occasions when we have deliberately sought to try and validate a finding in the literature. And you can see here that the journals that are represented as some of the top-tier journals, Cancer Cells, Cell, Nature, and so on. And what we’ve sought to do here is try and validate these findings because for example they claim that there’s a new oncogene in breast cancer or a new way of approaching a particular pathway in breast cancer. So from our point of view it’s really important to know if that’s true before we lay a $100,000,000 or $200,000,000 bet on that area of research.

I want to say again that for 25 years I contributed to this so I’m not trying to make the case. I don’t want anyone to think that I’m a disenchanted industry fellow that’s trying to point the finger at someone else. This was my life for 25 years. But the vast majority we’ve been unable to reproduce.

On almost every occasion we’ve contacted the investigators and tried to understand the results that they’ve generated that we were unable to confirm. We’ve exchanged reagents and on occasions we’ve actually sent out scientists to the host laboratory and observed that in the host laboratory they’re unable to reproduce their own data with their own reagents on their own bench when we’re standing beside them.

So this is really troubling. It also is important to recognize that some of the slides—the—the papers on the previous slide have spawned entire fields, so some of them—there are actually 400 or 500 papers based on this work. There are also actually clinical studies that have been triggered based on this work.

So I think this is an opportunity for us to set the bar, and I think that we should provide incentives whereby we reward reproducible not flashy racy results. And there are a number of things that as I’ve gone through these publications that I think should be routine. I think that these experimental data should be performed blinded as would be done for most reputable clinical studies. We should only use validated reagents, so a single siRNA, the targeted gene is not sufficient. Polyclonal anti-peptide antibodies should not be used for immunohistochemistry and so on. Investigators should include and show positive and negative controls. The data should be replicated and complete data sets shown including the negative experiments.

Earlier this year I had the opportunity to have breakfast with one of the investigators whose data I showed on the previous slide, and he indicated to me that he had done the experiment 10 times but only one of them had worked, so that was what they showed in their cancer cell paper. We should have accurate and not overreaching titles and abstracts and obviously accurately referenced citations.

I would posit that the publication bias is likely to get worse as research dollars become more scant. Again, I don’t want to suggest that this is fraud; it’s not. It’s people that are trying to do the best they can but they’re cutting corners with the pressure that they feel from granting agencies and so on to try and generate positive results. Although this is something of a downer, I want to remind you or make it very clear that I’m an optimist. I actually believe that we’ve made enormous headway over the last 20 years. I mention this as an opportunity for improvement going forward—not to try and condemn or criticize anyone. I’m—I’ve been part of it for the last 30 years.

But I would also like to echo Denny’s thoughts and congratulate and thank the advocates, the NBCC, the Department of Defense for the efforts that you have made over the last many years that has brought attention to an area that was underserved before you began what you’ve accomplished and not only have you accomplished that you’ve served as a role model to your sister organizations around the world who have also drawn enormous attention to this devastating disease. Thank you very much for your attention.