Chapter 12: The Kleberg Center for Molecular Markers
Dr. Mills begins this chapter on the creation of the Kleberg Institute for Molecular Markers by commenting that, as an administrator, he has been repeatedly asked to develop an area and then step away. His ability in this area gave MD Anderson leadership confidence in him and he was asked to develop a proposal for the Kleberg Foundation for a Kleberg Center for Molecular Markers. He talks about the sources of funding at MD Anderson, noting that the Kleberg Foundation's philanthropy funded the Kleberg Center (which was the basis for the Institute for Personalized Cancer Therapy). He describes a major Center initiative of characterizing ten thousand tumors (information that fed the Cancer Genome Atlas), now expanded to include 20 thousand patients. He describes the technology used for this project, developed from an idea he encountered in a lecture. He notes, We have incredible power to leverage what we are doing. Next, Dr. Mills talks about how the focus of the Kleberg Center has shifted slightly after the founding of the Institute for Personalized Cancer Therapy, concentrating on discovery and on rare cancers, such as mall cell ovarian cancer. Next, Dr. Mills talks about the intellectual context for this new focus and describes the knowledge that can be generated from the study of rare cancers, giving examples of studies that have led to clinical trials of new drugs. He talks about using models to rationally select drug combinations.
The Historical Resources Center, The Research Medical Library, The University of Texas MD Anderson Cancer Center
The University of Texas MD Anderson Cancer Center - Building the Institution; MD Anderson Culture; Leadership; MD Anderson History; Building the Institution; Overview; Ethics; Definitions, Explanations, Translations; Discovery and Success; Donations, Gifts, Contributions; Discovery, Creativity and Innovation; The History of Health Care, Patient Care; Politics and Cancer/Science/Care; Technology and R&D
Tacey Ann Rosolowski, PhD:Okay. I wanted to maybe go back, because when we were talking, kind of doing the identifier for the last session we had together, you know, you made sure to mention that you were head of the Kleberg Center for Molecular Markers in 2004, and you're still serving a role over there. Do you want to talk about that and kind of show how it maybe was working in parallel with some of these developments? Oh, dear.
Gordon B. Mills, MD, PhD :No, no, no. I want to say this in a way that doesn't come across wrong.
Tacey Ann Rosolowski, PhD:Okay, sure, yeah.
Gordon B. Mills, MD, PhD :One of the things that I've been asked to do many times at the MD Anderson Cancer Center is to develop an area, to support an area, and then get out of the way and let somebody else move in and take it over. That's not an easy thing to do because most of the time, when you start doing something to that depth, you want to own it. So, one of the things that perhaps is one of my greater strengths or weaknesses, is a variant of adult attention deficit syndrome. I get bored real easy and want to move on to the next challenge. And so, shortly after I arrived here, I was asked to develop, as I mentioned, the ovarian cancer program, breast cancer program --we haven't talked about the clinical cancer genetics program-- all of which we built. And then my goal was then to get out of the way and let others take it over. I think that that process and showing that they were successful, resulted in the institution having confidence and perhaps trust, in what I did. So nothing predicts future success any better than current success. My boss, who loves -- my postdoc boss, who loves aphorisms, says, "A plow horse is never going to win the Kentucky Derby." You know, what you're going to be or how someone is going to behave by what they did in the last five years, predicting the next five. And so with that, I was asked to put in a proposal to the Kleberg Foundation, to build a molecular markers program and we did that. It was not required by the Kleberg Foundation, that we call it a center, or use their name in the center, but given the significant, or at least what I thought was very significant, investment, that was part of what we did, and they've supported us quite extensively over time.
Tacey Ann Rosolowski, PhD:So this was philanthropic support.
Gordon B. Mills, MD, PhD :Oh, absolutely.
Tacey Ann Rosolowski, PhD:I was going to actually ask you, I'm not meaning to shift gears here, but I was going to ask you, when we had the NIH discussion, who was funding more innovative research, and is this an instance of a philanthropical source taking that kind of risky step.
Gordon B. Mills, MD, PhD :One of the ways to describe this is we have three types of funding that are available right now; that from the major funding agencies, which in general are extremely conservative and in general very, I guess sort of cap out. Once you've got a set amount of money, it's very hard to get any more, and to try and use those funds for truly risky projects is hard. Every penny that comes in from philanthropy, into our group, we estimate that we leverage sevenfold, give or take, with NIH grants, other foundations, industry, and use, to a very major degree, the philanthropic funds, to do high risk, high yield, exploratory projects that are going to fail most of the time or much of the time, and use that to build the information that you can then get funding from the more conservative agencies. Now, the Kleberg Foundation support --
Tacey Ann Rosolowski, PhD:I'm sorry, can I just ask you one question?
Gordon B. Mills, MD, PhD :Sure.
Tacey Ann Rosolowski, PhD:Because you said there were three funding sources, and I think I missed --
Gordon B. Mills, MD, PhD :Industry.
Tacey Ann Rosolowski, PhD:Oh, industry, okay.
Gordon B. Mills, MD, PhD :I did mention industry.
Tacey Ann Rosolowski, PhD:Okay, yeah, you did.
Gordon B. Mills, MD, PhD :So what you have is a very conservative, very important, critically important funding source from the major agencies. And note, when funding is tight, they become more conservative. When funding is a more reasonable level, they can be just as innovative, just as supportive of new ideas and concepts, and indeed, the NIH pushes those but not through its standard approach. It's through U-grants and other more innovative programs, so they do try very hard, but the reviewers become quite conservative because money is so tight and we're so nervous. It's us, we're the problem, we're the reviewers. Industry provides funds, but most of those are for fairly targeted areas, to look at how you would develop this particular drug or this particular target. Those can be very high risk, high yield, but they're targeted and they're not going to say, Well just go ahead and follow your nose. Philanthropy, again sort of fits with both of those, to allow in many cases, truly high risk, high yield projects. And indeed, what we built, through the funds from the Kleberg Foundation, very much became the basis for the Institute for Personalized Cancer Therapy, that I now co-head with John Mendelsohn, with the idea that many of the things that we put in place, we expanded upon and really leveraged what we had, to go way further than what we were doing.
Tacey Ann Rosolowski, PhD:What were some of the first projects you undertook through the Kleberg money?
Gordon B. Mills, MD, PhD :Our big project in the Kleberg, that they funded, is something that we call T-9, which was to characterize ten thousand tumors, across ten thousand patients, to really, nine times by the time you go through all of this, to really begin to understand what was going on in cancer, something we didn't know at that time. That was leveraged, or not leveraged. That supported many of the efforts we put into the Cancer Genome Atlas. Then on a more patient oriented basis, what we did in the Institute for Personalized Cancer Therapy. We've now gone beyond the ten thousand we promised and we've done almost twenty thousand. The idea here is really, to get an idea not just of the baseline events that are happening in tumors, that was coming out of the Cancer Genome Atlas, albeit at that time that wasn't even started, to ask what's happening in our cancer patients, who are usually heavily pretreated, very difficult diseases to manage, a very different spectrum of patients from what was being characterized in the Cancer Genome Atlas. And so that really was what the process was and those two led together, to this program.
Tacey Ann Rosolowski, PhD:What were the kinds of things you were looking at, just so I have kind of a more concrete -- I mean these were genetic profiles of cancer tumors?
Gordon B. Mills, MD, PhD :Our initial emphasis was on DNA, and the reason it was on DNA goes back to what I talked about, the technology to do that had become sufficiently mature, that it was something that we could develop and implement. Indeed, the first technology we used to do this hadn't even been dreamed to be used for this reason. It was called a Sequenom platform, where we were able to measure mutations on a massive scale. The platform had been developed by a company, to look at what was happening in the fetus, by looking at what was going on in maternal blood, and hearing a seminar from them, I said well, if you can do this in maternal blood, just think of what I could do with a patient’s tumor. We actually built this as a platform to characterize what was happening and we built it sufficiently well and robustly, that it was adopted by our CLIA laboratory, as their standard platform for measuring what was happening clinically in patients' tumors. Since then, they've moved away from this platform, to genome ones.
Tacey Ann Rosolowski, PhD:So this was your brainchild?
Gordon B. Mills, MD, PhD :Yeah
Tacey Ann Rosolowski, PhD:Okay, well, I didn't know.
Gordon B. Mills, MD, PhD :You don't have to know.
Tacey Ann Rosolowski, PhD:I'm glad you told me.
Gordon B. Mills, MD, PhD :There's no reason to know.
Tacey Ann Rosolowski, PhD:No, but that's key, and I think also, really indicative of the sort of open territory that this is. You know idea, oh, that could be useful, adopt, put it in a new context, let's find a way to implement. What was the company that built?
Gordon B. Mills, MD, PhD :Sequenom is the name of the company.
Tacey Ann Rosolowski, PhD:Is the company, okay.
Gordon B. Mills, MD, PhD :Actually, it was one of those wonderful coincidences, in that I got a call from the development office, about ten minutes after I walked out of that seminar, saying someone wants to donate money to the institution, this is the amount, they want to buy a piece of equipment, their name would be on it, and I said well, I have got an idea for you. So, it was a very nice piece coming together. Now, we negotiated sufficiently, this company, the donor said well, you know, we're giving you the money, why haven't you spent it? I said look, be patient, give me another month or two, I'm going to get them to give me this machine. We'll use your funds then, to buy materials and make it run, but by having that money in my hand and saying I'm serious, to the company, I was able to more than double what we got back from the company, actually with another company and John Mendelsohn and a very similar type of sequencing process. I went to him and said look, I need you to back up a philanthropic check of X number of dollars. I will virtually guarantee that I am going to return it to you and get the equipment for free, but if I went to the company saying just give it to me, rather than saying I believe in this enough that I can pay for it but it's going to be to your advantage to give it to me, because we're going to do this, this, and this. I handed the full check back and John Mendelsohn said, "This is the first time anybody has ever done this, given me money back." I said look I didn't ask you for the money to spend. I asked you to let us leverage this in a way that happens. And I think we perhaps don't do that as much as we should. We have incredible power here, to leverage what we're doing, and it's not used, I think as well as we could. It takes time. You have to be patient, you have to do a lot of things around it, but it does really come back and accomplish much more.
Tacey Ann Rosolowski, PhD:As they say, money makes money.
Gordon B. Mills, MD, PhD :That's right.
Tacey Ann Rosolowski, PhD:Yeah. Well, I derailed you a little bit with the question about funding. Are we good?
Gordon B. Mills, MD, PhD :Mm-hmm.
Tacey Ann Rosolowski, PhD:I'm just reviewing where we were, because I had asked you for an example of a study that was done, that fed into the Institute for Personalized Cancer Therapy. Have we finished the story about the Kleberg Institute [Center], or do we need to talk about it before it segues?
Gordon B. Mills, MD, PhD :No, the Kleberg Institute continues, and so one of the things that has come out of this -- or the Kleberg Center, sorry, it's not an institute. That's what you wrote down before. Once the Institute for Personalized Cancer Therapy was developed, we went back to the Kleberg Institute and said okay, what's the difference? Or Kleberg Center. And the first piece was, is that the Institute for Personalized Cancer Therapy is very much an implementation arm, to take technology and approaches that have been developed in other areas and implement them at a scale that you couldn't do otherwise, and to facilitate, working with pathology and lab medicine, implementation into patient care. And so the Kleberg Center focused, for quite a while, on the discovery pieces that led into and supported the implementation, and the two worked together beautifully. We, about two years ago now, took a look at where we were in the Kleberg Center, and with funds that we had left, and said maybe it's time to look at something a little bit different, still focusing on identification of markets, but in a little different approach than we had been doing. One of the things that the Cancer Genome Atlas taught us is that most of the cancers that they looked at, big diseases, are driven by fairly diverse events across those diseases. There will be ten, twenty different types of mutations that will happen broadly across the tumors, but what was coming out in a few other studies is when you looked at what were rare cancers, things that were truly different, they fell into two classes. One of those, where you look just like the common tumors; the breast cancers, the lung, the cancers of aging, where you had this diverse pattern of mutations, and you could learn from the Cancer Genome Atlas, what was going on in those. In contrast, some of these rare tumors had a single gene mutated in almost all, and to a degree, we believe for some of them, it is all. In the cases where we can't find that is a misdiagnosis, a separate disease. An example is a very rare tumor called small cell cancer of the ovary, maybe a hundred cases in the U.S., that is caused by a point mutation in a single gene, and in almost all cases, exactly the same nucleate change, one single event. What that does is provides an absolutely smoking gun, if not more than a smoking gun, that this is a critically important gene to understand. That information can drive a lot of studying. So we have taken the Kleberg Center and now use it to focus on rare cancers, and rare here in our definition is any cancer that hasn't been studied in depth by anybody else, and using the strength of our size and number of patients at MD Anderson, to say that things that are so rare, you can't do elsewhere, we can do here. And so we have processes on really, truly unusual and rare cancers, that just couldn't be studied without a major consortia elsewhere. And by the way, we're completely willing to bring in samples from elsewhere. This is not a, we just work on our own processes, but we've used this to build a rare cancer program and find these new types of events.
Tacey Ann Rosolowski, PhD:When was that decision made? You said a couple of years ago?
Gordon B. Mills, MD, PhD :I think formalized, I would guess three years ago, two years ago, in that area, and we went back to the Kleberg Foundation and told them we were doing that and that we were reorganizing the funds that we had left. We weren't asking for new funds, simply for permission to change the direction in which we were going, and given that the Kleberg family had brought their money together based on cattle breeding and cattle genetics, this was something that they were very pleased to hear, to say okay, we get it, understanding the genetics here is important, you told us you were going to do this T-9 project, you've already told us you've done more than you had proposed and you have money left over, of course we would like to see you do more. And so that's really where that has evolved.
Tacey Ann Rosolowski, PhD:Now, what kind of -- I mean, I'm sure it's hard to say, but obviously this is not only interested in addressing these rare cancers, but this is an engine for producing new knowledge. What kind of possibilities does that new knowledge create?
Gordon B. Mills, MD, PhD :I'll give you an example again, around this small cell cancer of the ovary, this extremely rare disease. Had a particular mutation in a single gene and it is mutated across cancers, and so now instead of trying to figure out whether this was a driver or a passenger, important, not important, it's absolutely clear that this is a very important gene to study and understand not just what it does normally, but why is it abnormal in cancer. In terms of the other piece is we had been studying its closest relative in cancer, so one of these is called SMRK-4, the one we had studied is ARID-1A, and it doesn't really matter what their specific names are, but just really close family members. And because of the information coming out of the small cell ovarian cancer, we went back and redoubled our efforts to try and understand what both of these are doing, and again, in collaborative studies, some directly from my lab, some with a group in Holland, some with another group here at the MD Anderson Cancer Center, some with a group in British Columbia. We've gone in and it looks like the function of these genes in cancer may be multifactorial, again to this systems biology concept, but part of it completely unrelated to what we thought these genes did previously. They were thought to play a role and a particular functional complex and a particular function in DNA replication, and it looks like they do that, but in addition, do something completely different, that may be why they're involved in cancer, and we're using that to develop new therapy approaches and indeed, one of them has gone from discovery through to the clinic.
Tacey Ann Rosolowski, PhD:What is that approach?
Gordon B. Mills, MD, PhD :It turns out that we think that these particular abnormalities will sensitize tumors to a new drug that has come out and is looking very impressive, which are called PARP inhibitors.
Tacey Ann Rosolowski, PhD:I'm sorry, PARP?
Gordon B. Mills, MD, PhD :PARP, P-A-R-P, inhibitors, which were thought to be important as providing synthetic lethality, that is killing cells only if they had abnormalities in BRCA-1 and BRCA-2, the two breast cancer genes, and ovarian cancer genes, and it turns out that there may be many other things that these drugs work with. These two genes I was talking about look like their other ones in that family, and so there are clinical trials where if you have abnormalities in these genes, you can receive PARP inhibitors and we'll see if it works.
Tacey Ann Rosolowski, PhD:When did those trials start?
Gordon B. Mills, MD, PhD :Oh, about a year ago.
Tacey Ann Rosolowski, PhD:About a year ago, yeah.
Gordon B. Mills, MD, PhD :Yeah, we think that that's not going to be enough. Single agents are not going to cure cancer. Cancer, epithelia cancers, they're far too complicated for one agent to work. We need combination therapy, that's very much what Frei [Dr. Emil Frei] and Freireich [J Freireich; oral history interview] taught us, that's what chemotherapy has taught us. Single agents aren't enough, you need combinations and you need combinations that work, and so we have a major program around how to find those, how to develop those, and we have a number of processes where we use those two abnormalities I told you about, as a base to ask what combination therapies would I use for Mrs. Green, who has an abnormality in one of those two genes.
Tacey Ann Rosolowski, PhD:Well, I'm smiling because there was an interview I did with someone who had been hired many, many, many years ago, and was working with combinations, and I said you know, how did you put these combinations together, and he said well, we would sit down and I'd say, "What have you got?" And they would write them down on a napkin. It's a different world now.
Gordon B. Mills, MD, PhD :To a degree, we only had a napkin. No. There are now a thousand drugs, give or take, that are in or about to be in clinical trials, and if you were just to think of all of the two-by-two combinations, and then layer that on top of the DNA, RNA, and protein abnormalities, to say which patients you would like to treat, you come up with a completely impossible matrix to do. We have tried a lot of hypothesis driven research, and that's been very good, it's gotten us to where we are in many cases. What we do, as our major driver of this, is to look at one type of resistance to therapy called adaptive resistance, and try and build a platform that allows us to see that adaptive resistance efficiently. That platform is our protein array platform that I mentioned earlier, and what we do simply is say treat a cell with a drug. It can be any drug, it can be any cell, most cases it's many drugs and many cells, to get an idea of what's happening broadly, and say how does the cell adapt. What changes does it make to stay alive in the presence of that stressful drug? If I now target those changes, do I see those two drugs working together in a logical manner? And with some exceptions, that has been extremely informative and has led to multiple combinations that make sense, and to others that we would never have come across without doing this type of an analysis. Several of them are entering or in some cases supported by this result, not necessarily driven by this project, in trial. Some of the ones that are entering trials would never have come up without this study.
Tacey Ann Rosolowski, PhD:Wow, interesting. Do you call that evidence based research?
Gordon B. Mills, MD, PhD :I call that rational combination therapy. The rational is probably what you're talking about in terms of evidence.
Tacey Ann Rosolowski, PhD:I was just asking, because you know, that is obviously a term that's used a lot, but was just curious what your term would be.
Gordon B. Mills, MD, PhD :Well, I think evidence based care is the term, patient care is where it's usually used. We try and think of hypothesis driven research, rational combinations, as compared to say, some companies that have gone out and taken all thousand drugs and done every possible two-by-two combinations on cell lines and said okay, do we find anything here that we can move forward. So far, that brute force approach has proven to be not very efficient or effective.
Tacey Ann Rosolowski, PhD:I was wondering if it's, I was going to say ethical. Is it more ethical to use the approach that you're using?
Gordon B. Mills, MD, PhD :I don't think it's ethical in either way. I think in anything that leads you to something that can benefit patients is ethical. If you're saying is it a good use of resources, the two-by-two combinations were done primarily by companies, the random ones, and if they had the money to do that, that's their prerogative and actually goal. We didn't know if this would work. It was a test of the hypothesis that just doing enough of those would give you useful information and it's not proven to be as predictive as we would have liked.
Tacey Ann Rosolowski, PhD:Are we good with Kleberg?
Gordon B. Mills, MD, PhD :Yeah.
Mills, Gordon B. M.D., Ph.D. and Rosolowski, Tacey A. Ph.D., "Chapter 12: The Kleberg Center for Molecular Markers" (2016). Interview Chapters. 71.