Mitch Dowsett, PhD, BSc and
Ian E. Smith, MD, FRCP, FRCPE
The Royal Marsden Hospital
Institute of Cancer Research
London, United Kingdom
Dr. Dowsett, Head of the Academic Department of Biochemistry and Professor of Biochemical Endocrinology and Translational Research, joins Dr. Smith, Head of the Breast Unit and Professor of Cancer Medicine, to discuss their BCRF-funded collaboration.
Q: Tell us about yourself as a scientist and how you became interested in breast cancer research. Did you ever seriously consider another kind of career than that of the sciences?
Mitch Dowsett (MD):
I never seriously considered another career other than one involved in science. In fact, I can recall even before the age of 10 saying that I wanted to be a scientist. The thing that really turned me onto biological sciences and still astonishes me is the genetic code, which was unraveled not that many years before I was in secondary school. I'm fascinated and find remarkable that the makeup of not just humans but any biological living being was actually coded for in some chemical form. That really sparked my imagination so I took a degree in Zoology, which was actually a mixture of different biological sciences with a twist towards biochemistry and genetics.
The thing that got me involved with cancer and breast cancer was a course run at Imperial College, where I was studying, by the nearby Institute of Cancer Research. I was particularly attracted by the idea that one could begin to use some of this genetic information in relation to clinical treatment. As a result of that course, I applied to the PhD program at the Institute of Cancer Research, with which I am still associated. My dissertation was on trying to understand the mechanisms of why breast cancer interacts with bone and causes it to break down when the cancer metastasizes to that region. I got involved in what we call "translational research" very early where we would take small breast tumor biopsies from the next-door Royal Marsden Hospital and put them into culture with small pieces of bone and try to understand how they interacted. So, I've had a pretty longitudinal career where I started working on breast cancer very early all the way through to the present.
Ian Smith (IS): I also decided at a fairly early age that I wanted to go into medicine. I trained at the University of Edinburgh, where I sometimes had to look after patients with leukemias or lymphomas. Cancer became an absolutely fascinating challenge to me. From that experience, I decided to go to the Royal Marsden in London, which is the main and one of the very few stand-alone cancer hospitals in the UK. For me, as a young doctor, the Royal Marsden was an extraordinary stimulus, because there I found clinicians not only treating cancer but also working alongside scientists in laboratories on the same campus. I remember we used to sit together in the evenings and talk about the challenges of cancer treatment and the underlying biology. That was when I decided that I wanted to be a cancer specialist and to take a scientific approach working with scientists to tackle what were very challenging problems.
Now, did I consider any another profession? The answer is "Not really." I had considered studying English literature, which I enjoy very much, but I did not have a very good answer for what I would want to do after graduation. And that was why I went into medicine, because then my path was much clearer.
Q: Briefly describe your BCRF-funded research project. What are some laboratory and/or clinical experiences that inspired your work? What are your primary goals for this research?
IS: My interest in our current BCRF work really began more than 20 years ago, long before the Foundation was formed, when we started doing what is called "neoadjuvant" chemotherapy, or chemotherapy before surgery, for large breast cancers. We use that approach to try to shrink tumors so that mastectomy can be avoided. We did pioneering work on neoadjuvant chemotherapy at the Royal Marsden and then began applying the same approach with endocrine therapy, using tamoxifen and subsequently aromatase inhibitors including in particular letrozole. And we saw some excellent clinical results with good tumor regressions.
But what really began to interest me was how we could predict in advance which patients were going to benefit from treatment in the long term. Clinical measurement, we discovered, was a rather crude, inaccurate tool. So, working with my laboratory colleagues, we began to look at molecular markers, also called biomarkers, and particularly Ki67, which measures proliferation. A molecular marker is a biochemical test on a tumor that helps to characterize some aspect of its behavior. There are prognostic and predictive molecular markers. A prognostic marker helps to define a cancer's characteristics and its outlook, whether it is likely to behave in an aggressive or an indolent fashion. But what I think is more interesting and more important clinically are predictive markers, which tell you whether a cancer is going to respond to a treatment or not. The best example of a predictive marker is the estrogen receptor. It has been known for many years that if a breast cancer is estrogen receptor-positive, as most of them are, then there is a good chance that they will respond to endocrine therapy but if they are estrogen receptor-negative, then they will not.
In our work, we discovered that when you treat breast cancers with drugs, and in particular with endocrine therapy, the cells in many of these cancers stop proliferating within two weeks of starting treatment and "switch off." Therefore, we thought that measuring the change in the proliferation marker Ki67 might then give us a more accurate way of predicting which cancers were going to respond well to treatment. This has been one of the central tenets of our research and an area of research that BCRF has supported very strongly over the years.
MD: Our BCRF-funded project of the moment builds on some observations we made a few years ago that if we were to expose hormone receptor-positive breast tumors to an aromatase inhibitor, or other endocrine agents, we saw a profound fall in the proliferation of cancer cells, as measured by Ki67, the function of which Professor Ian Smith described above. We rationalized that the degree of fall might be indicative of the benefit that the patient will get from that specific treatment. So, Ian and I set up a clinical trial called IMPACT, which showed firstly that those changes in Ki67 did indeed predict the outcome of patients in the ATAC trial, a much, much larger study comparing the effectiveness of the endocrine therapies, aromatase inhibitors and tamoxifen, after surgery. Based on the IMPACT results, we believe that generally one can use Ki67 as an intermediate measurement of treatment effectiveness for an individual patient.
I was inspired to work with Ian on the IMPACT study, because of my involvement with the development of that ATAC trial. ATAC was the first trial conducted and reported on the comparison of aromatase inhibitor versus tamoxifen, versus the combination of these two drugs. It was the first big clinical trial that I was involved with, and I was struck by just how enormous the trial had to be. ATAC eventually required over 9,000 patients to be randomized and to be treated for at least two and a half years before we had any indication of whether one treatment was better than the other. And the measure of that success was based on approximately 900 patients having to relapse with their breast cancer before we knew whether one or the other was better to restrict that relapse. These difficulties seemed to be something that we should be able to improve upon by designing clinical trials that use markers, such as Ki67, to predict patient benefit. And that is where the IMPACT trial came in. IMPACT involved just 330 patients and three months treatment duration. In fact, because we took biopsies after two weeks, we could show changes in Ki67 at the very earliest stage, and were able to show that the combination of aromatase inhibitor and tamoxifen was actually poorer than just the aromatase inhibitor very quickly. So the ATAC experience was something that really did inspire my work on IMPACT. Currently, Professor Smith and I are again co-Principal Investigators on a new trial called POETIC, which builds on IMPACT to find whether giving patients an aromatase inhibitor prior to surgery and then measuring their Ki67 level may allow us to better predict during routine care the outcome of those patients in the long term.
With BCRF funding, specifically we are doing a study in premenopausal women, on whom you cannot use aromatase inhibitors because they are ineffective and cause significant adverse effects. But we thought the way around the problem was that we might be able to use the changes in hormonal stimulation of the breast tumor, occurring as a result of the menstrual cycle in these premenopausal women. We thought it might be possible to allow the menstrual cycle, instead of a drug, to stimulate certain genes in these women's breast tumors and indicate their responsiveness to treatment. We have some background data on our study, and a paper has just been published in the journal Breast Cancer Research and Treatment. These data were also presented at the 2012 San Antonio Breast Cancer Symposium. We have also launched a prospective trial to test this with colleagues in the UK and Vietnam.
IS: Now, another component of our research that excites me is changing the traditional way of using these molecular markers. Standard practice dictates that you look at the molecular markers before you treat a cancer and use them to predict effectiveness of therapy, and then the surgeon removes the tumor prior to starting adjuvant medical therapy. But what we have been doing, based on our original experience with neoadjuvant treatments, is that instead of immediately removing the cancer, we start with a short course of preoperative medical endocrine treatment with tamoxifen or an aromatase inhibitor, and then remove the cancer two weeks later and look at how these molecular markers are behaving. It is our hypothesis that measuring changes in markers after treatment will predict the likelihood of the therapy working more accurately than if this test was done before beginning treatment. In other words, applying the marker test after treatment has started and after surgery gives us a measure of not just the biology of the cancer but of the sensitivity to the treatment we are using. Now if we can do that, it means that for each individual patient, you can tell very early on whether it is worth giving someone five years of an aromatase inhibitor, five years of tamoxifen, and such. Hopefully, although we are not quite so far ahead in this, we can also tell after one course of chemotherapy whether six or eight courses are going to beneficial or not, because if it is not going to work, the sooner you stop that treatment, the better. Conversely if the chemotherapy is going to be beneficial, it is so much better to be able to say to the patient, "Look, this treatment is unpleasant but we know from the molecular markers after your first course that it's going to work and we know it's got a high chance of cure." At the present, despite all the developments in cancer research, a lot of the medical treatment of early breast cancer is still guesswork. Our research is to overcome that guesswork.
Q: Are there specific scientific developments and/or technologies that have made your work possible? What additional advancements can help to enhance your progress?
MD: The studies I mentioned are based on fairly standard technologies. The POETIC study, however, requires us to implement a collection of tissues essentially as soon as a patient is diagnosed and then again at the time of surgery, which in the UK has required a certain amount of change in the infrastructure. The POETIC trial involves 120 centers, and we just reached our targeted recruitment of 4,000 patients.
In terms of the future, including in our work using the menstrual cycle to stimulate the effectiveness of breast cancer drugs, we are using high throughput multiplex molecular technologies, to search a larger number of genes to see whether we can get better signals of response to therapy than what we have gotten so far. Also, we just started doing next generation sequencing on some of the samples from the POETIC trial to identify any mutations existing in those tumors related to outcome. So, we are embracing these new molecular technologies and are beginning to apply them in our studies.
IS: There certainly have been developments in laboratory technology, and my colleagues who work in the labs are the ones to elaborate on these. But for me, what is just as important is changing the mindset on clinical strategy. We spent many years thinking that everyone with breast cancer should have a mastectomy, but it turns out that was completely wrong, and we have changed that mindset. Now we need to change the mindset again, and to have surgeons, and indeed all the team looking after patients with breast cancer, thinking of using medical treatments first rather than rushing to surgery so that we can do these predictive molecular tests. To test our hypothesis, we set up a very large trial, in fact involving the whole of the UK, called POETIC. In POETIC, we aimed to recruit 4,000 patients who would be randomized to two weeks of preoperative endocrine therapy versus going straight to surgery.
Many of our colleagues initially had said that we would not be able to do the POETIC study on a large scale. They said that you could do this at a specialist cancer center but to make this trial work on a larger scale, you have to change the mindset of nearly all the surgeons in the land. And the really exciting thing is that trial has worked extremely well. In the UK, we have caused a change in the whole mindset, a change in the attitude to approaching breast cancer, so that surgeons and other professionals agree to this concept of using medical treatments first as a means to determining for the individual patient in from molecular tests which is going to be the best treatment. And although that does not sound as clever as developing some new technology, nevertheless, I think it's really very fundamental to further progress in clinical breast cancer research.
Q: What direction(s)/trends do you see emerging in breast cancer research in the next 10 years?
MD: The big thing that has been happening over the last few years is the application of next generation sequencing and massive parallel sequencing to identify genetic mutations and then to profile mutations in tumors. This work has revealed some surprises, such as that there are numerous mutations but with relatively low prevalence in any particular population of breast cancers. This "diversity" creates a real challenge as to how we can bring that information to clinical benefit. But trials are now being designed to see how we link those mutations to the applications of some targeted therapies. There is a real challenge, recognized by many people, in the heterogeneity of those mutations within any single subtype of breast cancer. This may mean that there are "escape routes" or populations of tumor cells that either do not contain those mutations or have additional mutations that make treatments ineffective. We must address those issues as we move forward.
The thing I would really like to see come forward in a big way is the measurement of tumor DNA in the blood, because I think that could be a game-changer in terms of the way we manage breast cancer. This essentially would be a very sensitive, highly specific tumor marker which would reflect whether or not after surgery significant amounts of tumor cells continue to reside in the patient and therefore need additional medical treatment. At the present time, we are treating patients after surgery essentially blind of whether they have residual tumor, and blind to response of that tumor. If we had had a marker of the sort, it would radically change the way we direct our therapies. I think the prospects for that occurring are good but probably a few years away.
IS: In my opinion, the most dramatic change in breast cancer over the last 10 or 15 years has undoubtedly been the appreciation that breast cancer is not just one disease. We know from gene expression studies and from many different areas of biological research that breast cancer is a series of different disease subtypes, each of which require very different treatment strategies. For example the most dramatic development in recent years has been the discovery of the HER2 biomarker on the cell surface of some breast cancers. Around 20% of breast cancers are of the HER2 subtype, and these respond very well to an antibody against HER2, called trastuzumab (brand name, Herceptin®), which has revolutionized the management of that particular subtype. HER2 has also made us aware of another subtype, called triple negative breast cancer, which is negative for three receptors: HER2, ER, and progesterone (PgR). And we know that for triple negative cancer we need to develop other treatment strategies. There are other subtypes as well and the trend of defining breast cancer on the basis of smaller and smaller subtypes, which all need more and more specific treatments, is going to continue. The fancy word for this is "personalized medicine," and we are well on our way towards that goal.
But there's another really important thing as well--the molecular markers that differentiate different types of breast cancer may also be targets, targets which differentiate normal from cancer cells and which may therefore be targets for new drugs. These might kill off the breast cancer but not damage surrounding normal cells, and trastuzumab is an excellent example. Chemotherapy also has made a major impact in breast cancer but it is a toxic, unpleasant treatment. As time goes on, we will gradually develop more and more of these targeted therapies for different breast cancer subtypes, and it is going to be so much better than the kind of blockbuster "one-size-fits-all" approach that we have had in the past.
Q: What other projects are you currently working on?
MD: The POETIC trial is one of our major efforts. We will have in the laboratory all the tissue samples from the 4,000 participants, which is an enormous resource for trying to work out the mechanisms of resistance to endocrine therapy. This is a unique set of materials to help us to truly understand tumor heterogeneity and the different mechanisms of resistance to endocrine therapy and the degrees to which these are important in any patient.
Another project that we are working on is predictors of late recurrence in estrogen receptor-positive breast cancer (late in this case being defined as after five years of the patient being free of breast cancer). We have just submitted our first manuscript on this and presented data for the first time late last year at the European Society of Medical Oncology (ESMO) meeting. The observation was made a few years ago that the risk of breast cancer continues for many years in the estrogen receptor-positive patients. The convention up until recently has been to treat these patients for five years. Work has now shown that continued recurrence after five years can be markedly reduced by the introduction of further endocrine treatment at the end of five years. But the challenge is to identify which patients really need that continued treatment. Five years is already a long time to be taking treatment and recommending it to be taken further is not to be made without serious consideration. And there are some markers that are coming through that we should be able to bring to bear. But specifically trying to identify molecular markers that could be useful in that scenario is a recent research development. The markers that we have been testing out so far and have valuable prognostic or predictive characteristics have been found to have some prognostic value for late recurrence but were "discovered" in a way by chance, rather than by design. We expect to be able to improve on these by dedicated research in this area.
IS: I have always been in interested in new drug development, which will continue to be a very important part of breast cancer. These new drugs are much more interesting than in the past, as drug development in the past largely was almost serendipitous stumbling on things that work. Now we are basing them on the molecular targets as I have described above.
I also have great interest in breast cancer in young women. For young women, breast cancer can be a particularly devastating disease. We have very good treatments, and patients can often be cured, but the cure comes with a price. Some of our treatments cause long-term infertility, which is a terrible thing for a young woman. Many of our treatments also cause suppression of ovarian function, so young women have very unpleasant side effects due to this highly unnatural state of premature menopause. This is an area that we have not really worked out properly, not so much in terms of research and treatment, but in terms of support, in terms of appreciating the particular problems, the psychological as well as the physical, problems. And it is an area that we have a lot of interest in at the Royal Marsden because young women tend to gravitate here since we are a specialized cancer center.
Q: How close are we to preventing and curing all forms of breast cancer?
MD: That is a tough question. If we think about "cure" at least in terms of preventing death from breast cancer, show we are making substantial progress. Some twenty years ago, 16,000 women were dying of breast cancer in the UK each year, and the most recent data from two years ago were that 12,000 women died of breast cancer each year. These survivors would have been diagnosed several years prior, so going forward patients diagnosed nowadays have much better prospects, including better treatments, which would lead to a much reduced risk of breast cancer mortality. So in terms of reducing mortality, I think the chances are that few patients will be dying from breast cancer twenty years from now. I believe breast cancer will become a disease that is much less feared than it has been as we move through the next two decades.
I do not think that the approaches we are presently taking are likely to lead to major increases in prevention over the next few years. First, detection is not the same as prevention. Being able to detect breast cancer earlier does allow doctors to treat patients earlier, giving the patients a better chance of surviving the disease. But early detection is de facto not geared towards prevention. My optimism is more directed to thinking that the disease will be managed much, much better, such that the fear will be taken out of the diagnosis of breast cancer. Prevention is likely to have some importance by being targeted at those at highest risk.
IS: I think in terms of preventing breast cancer, we may well be able to prevent some cases through drugs like tamoxifen, and possibly through lifestyle changes. But, I certainly do not think that we are going to be able to prevent all breast cancer in the foreseeable future and it will continue to be a major and very important disease in our society.
But in terms of increasing cure rates, I am very optimistic. I think the last 15 years have been very exciting. Even though breast cancer incidence is rising, deaths from breast cancer are falling at a steady rate. A major reason for this has been the use of so-called adjuvant medical therapies given after surgery. There are new therapies and new concepts (neoadjuvant therapy being one of the most important) emerging, which are being applied across the board. As new drugs continue to emerge, there is every reason to believe that deaths from breast cancer will continue to fall. We are already in an era where only a minority of patients die of this disease and I predict that in the next ten years deaths from breast cancer will become very uncommon.
Q: In your opinion, how has BCRF impacted breast cancer research?
IS: BCRF has had a dramatic and stimulating impact on breast cancer research. The funding itself is not just crucial but quickly accessible. This is important--it is very difficult nowadays to get large amounts of research money quickly through the more traditional sources. That is not to say that governments are unsupportive of research, quite the opposite, but they have limited budgets, and it is increasingly difficult to be able to obtain all the funds you need and quickly.
Another marvelous thing about BCRF is the community spirit that it engenders. BCRF brings all the top people involved in breast cancer research together, physically once a year and in spirit over the rest of the year. That leads to better exchange of ideas, and that is the way progress is made--not by people sitting and working in isolation, but by bringing large groups of people together. This is what makes BCRF so important and so valuable.
MD: BCRF is also not bureaucratic, which means much more of the money and time can be spent on the research itself rather than on reports. I think the trust that BCRF has in its grantees is well-placed and that trust imposes a degree of responsibility on the investigators.
BCRF funding has also produced some indirect benefits. For example, the types of projects that BCRF enables one to do are those sometimes considered high risk but high pay-off, and I think the study that we are doing trying to establish whether or not we can use the menstrual cycle as a probe for the hormone responsiveness of the breast tumor is very much along those "high risk, high reward" lines. If it works out, it will be a major benefit and a terrific advance, but it is high risk. The data we have derived so far supports us moving forward, and it is decreasing the risk as we move forward. I think it would have been difficult to secure funding from another agency for this project.
Another thing I might say about BCRF is the genuine feeling of family it creates amongst the investigators. At the meeting in October, the interactions are very warm, and our meeting at that particular time is welcomed by all of us. We all travel to New York and I think we both enjoy and get reward from interacting with the advocates and the fundraisers. This meeting really does stimulate people and creates a patient-orientated focus in the participants' research as well.
Read more about Drs. Dowsett and Smith's current research project funded by BCRF.
Read Dr. Dowsett's profile in its entirety
Read Dr. Smith's profile in its entirety