Richard Hill

By Arjun Sahgal and Robert Bristow

The following interview of Richard Hill, PhD, was conducted on January 14, 2016, by Arjun Sahgal, MD and Robert Bristow, MD, PhD. 

Arjun Sahgal: So, we're going to start off with just some introductions around the room first. My name is Arjun Sahgal, as the history committee member, and to my right is Professor Hill, and we have Rob Bristow who's also a radiation oncologist here and the two of us will be interviewing Professor Hill for this ASTRO History Committee documentary.

So, Richard, maybe you want to give us a little bit of background in terms of where you grew up and studied?

Richard Hill: So, yes. Actually I grew up primarily in the UK, just outside of a town called Reading which is about halfway between London and Bath. I went to school at Reading School which was a grammar school, and there I basically specialized in science. I went on to do an undergraduate degree at Oxford University. This was in St. John's College and I did a degree in experimental physics.

Out of that I really developed an interest in radiation so I moved on from there with a plan that was to study radiation physics and move into the nuclear power industry. But where I was studying, which was at St. Bartholomew's Hospital Medical College in London, they had a major focus on cancer treatment, so I got really sucked into the radiation biology side of things. I finished up doing my Ph.D. in radiation biology and my supervisor at that time was Dr. Patricia Lindop and the other adviser I had was Professor Joseph Rotblat.

Prof. Rotblat was a Manhattan Project person and had subsequently developed a strong interest in really trying to promote the peaceful use of nuclear energy. In fact he was awarded the Nobel Peace Prize in 1995 for the development of the work that he did in that area. And what is interesting from the Canadian point of view is that this was all designed around a society which was called the Pugwash Society that had its first meeting in Pugwash which is in Nova Scotia in Canada. They had annual meetings about how to really promote the peaceful uses of nuclear energy.

After I completed my PhD, I moved from the UK to the Princess Margaret Hospital or the Ontario Cancer Institute as it was called in those days, to work with Dr. Ray Bush, who, as you probably remember, was a former director of the Institute. And I worked with him for four years, when I was a post-doc, on various aspects of animal model work. This followed up my PhD. work looking at tumor response to radiation, because there was relatively little known in those days about the factors that were involved in that, and we can talk a little bit more about that in a minute. Then, after four years I went back to work in the UK for about two years. This was at the Institute of Cancer Research in Sutton, which is in Surrey just south of London. As it turned out Dr. Harold Johns, who was the head of physics at PMH had come on sabbatical there while I was there and he convinced me to come back to Toronto. So, I came back to Toronto in 1973 and I've been on the staff here since then. So, that's my history.

Arjun Sahgal: And so, when you came on staff here, were you directed towards a particular area of research or was there a particular question that you were really geared to ask?

Richard Hill: Well, as I said earlier, I had done my PhD. in animal studies of the radiation responsive tumors because there was really very little known about how tumors responded to fractionated radiation. One of the things that we showed in a study I was working with another student was that tumors were actually more sensitive when you gave them the same dose of radiation in two fractions rather than a single dose. We concluded that this was due to what's now called re-oxygenation, and this was one of the first demonstrations of re-oxygenation occurring during fractionated treatment. The initial primary work in this area was done by Robert Kallman at Stanford and he published a paper just about a year before our paper came out and that certainly helped us to understand what was going on.

On coming to Toronto, I picked up on the hypoxic area and, still working on animal models, we looked initially at various factors that might impact on the oxygen status of tumors. We looked first at anemia and then we looked at various techniques for modifying oxygen content, including hyperbaric oxygen breathing. We also looked at a technique where we would actually expose the animals to low oxygen environments for a period of time and then put them back into normal oxygen to see whether we could improve oxygenation in that way, and that certainly worked. It is really based on the fact that you get an initial change in the downloading of oxygen if you're at low oxygen levels, and then ultimately you get an increase in hemoglobin levels. So, both of those factors actually improved oxygenation in the tumours when the animals were moved back to air breathing. Ray Bush was interested in these areas as well, so that's where we initially started.

Arjun Sahgal: So, what was the team like, I guess? Because, you know, the Ontario Cancer Institute and you were here, there were several others that were working with you, what was the kind of the team environment focused on cancer research?

Richard Hill: In those days I guess there were about 15 senior research staff at the Institute, and they were nominally split between what were called the physics and biology divisions, but in practice most of the physicists were actually involved in biology of one kind or another. I was in the physics division but working at that time fairly closely not only with Dr. Bush but also with Dr. Gordon Whitmore, and then ultimately Dr. Mike Rauth as well. We were all really involved in trying to understand how cells responded to radiation and how the hypoxic environment of tumors was really impacting on that response and looking at ways in which we could potentially modify that for improved results.

And one of the values of the environment was that we had the graduate student program which operated through the department of medical biophysics (University of Toronto), and was actually based in the Institute, so most of the department students were in the Institute. This provide excellent access to graduate students, who were interested in that kind of work as well. and certainly one of the early students I had was Dr. Dietmar Siemann who is now head of a research program somewhere down in Florida.

Robert Bristow: Gainesville.

Richard Hill: Gainesville. He was actually my first graduate student. And later on, of course, Rob also became one of my graduate students.

So, that was the early environment. Over the course of time, the number of staff increased relatively slowly and there was an expansion of the old hospital. But it was recognized that the hospital was not large enough to accommodate the expected increase in cancer patients who were going to need to be treated and so there was a drive from about 1985 to actually build a new Princess Margaret Hospital. Actually Ray Bush was one of the people who was really pushing this and that culminated in the current building that we sit in today, which we moved into in 1995, as you probably know.

Arjun Sahgal: You know, from our point of view, to kind of learn from what was happening at that time, maybe you could summarize, and maybe Rob you could jump in here on the science part of things, of what were the controversies or the divides at that time. I mean, you were talking about re-oxygenation, fraction size, tumor response. Was there any splits in terms of the community or dissenting opinions of your work?

Richard Hill: Well, I think that it is certainly true that there was dissenting opinion in the clinic as well about the size of fractions that should be used. There was a divide, if you like, between traditions of English radiation oncology and North American radiation oncology The Canadian group was more influenced by the English side whereas the American group was more influenced by the French group. The French group had really pushed the concept of relatively small fraction sizes whereas the group in England, particularly Patterson at Manchester had really pushed the concept that you could use slightly larger fraction sizes. It all came together over time and fraction sizes tended to be reduced because it became clear that there were other factors that were impacting on (mostly normal tissue) response.

At the time when I started in radiobiology there was very little known about normal tissue response. Over the course of time as we started to understand normal tissue response much better, the classic alpha-beta concept was developed at the Gray Laboratory by Jack Fowler and his colleagues. This built on work that had been previously done by Frank Ellis who developed the so-called NSD concept to predict equivalent fractionation schedules for the clinic.

This is interesting because the development of the NSD model actually resulted in a lot of radiation biologists thinking that there were biology aspects of it that didn't make sense. And so, it resulted in a huge number of experimental studies of radiation fractionation in animals done to look at the response of normal tissues. And a primary focus was to really prove that the NSD concept was incorrect. due largely to the time component that really wasn't built in correctly. All that data resulted in Fowler and his colleagues really being able to develop and test the alpha-beta model which has turned out to be a much more relevant model for standard fractionation. 

Arjun Sahgal: And would you say now in retrospect, I mean thinking of the linear quadratic equation when it came out at that in incorporation of alpha-beta, do you think it’s still relevant or do you think that we finally have to change our thinking away from the linear quadratic, maybe incorporate molecular genetics into this aspect of how we determine ideal fractionation because there is so much controversy right now.

Richard Hill: I think you're certainly right that we now need to think in broader terms. One thing that has happened of course, which has made a big difference, is that the technology of delivering radiation has become so much more proficient and this greater precision means the original concepts of normal tissue toxicity have really changed. The concept that you were going to irradiate a large volume of normal tissue when you were treating a cancer has really changed. The volume of normal tissues that we're irradiating to high doses are much smaller, and what that has allowed to happen is that one can change the schedule because of the fact that for many normal tissues there is a volume-dependent component. Thus you can get away with giving much bigger doses of radiation to smaller volumes of some normal tissue, so on the whole, I guess what we'd call the SBRT movement is really revolving around the fact that we're irradiating much smaller volumes of normal tissue and we can actually give bigger doses. The whole early work on alpha-beta really revolved around the fact that, if you were going to irradiate a large volume of normal tissue, you really had to allow it to repair effectively and the repair and repopulation aspects of that became very important.

With SBRT, that is a much less an important issue for certain tissues. Lung is the perfect example. You give a large dose of radiation to lung, you basically knock out the function of the region of lung that you irradiate but, because the lung has a lot of excess capacity it doesn't impact on the overall patient function. And that's true for a number of different organs. There are certainly organs that it doesn't apply to, and you're familiar with that I'm sure, with the spinal cord, there's no way that one could give a huge dose to a complete section of the spinal cord in the same way as you could to a small region of lung. There is a concept, which was introduced a number of years ago, arguing that tissues can be largely divided into those with parallel versus series function in relation to the way they respond to radiation. Some tissues like the spinal cord have got to be functional over the whole length of the cord (i.e are series organs) whereas in a tissue like lung, if small region of the lung isn't functional, it doesn't really impact the other areas I.e it is a parallel organ). This concept is useful in the way we think about fractionation today, from my perspective anyway.

Arjun Sahgal: And Rob, maybe you can tell us what it’s like working with Dick as a student and how he influenced your career and the idea of radiobiology and how it was influenced.

Robert Bristow: Sure. So, you know, I came to the Princess Margaret Hospital to work with Dick Hill because I was interested at that time in how basically cells divide and the cellular proliferation aspects relative to radiation, because I had done a fourth year undergrad project in zoology on radiation effects in normal tissues, and I think that what was exciting at the time was that there was this increasing knowledge about the molecular biology of cancer and this was the kind of the heyday of oncogenes and tumor suppressor genes, and that was overlaid then on the tumor biology that Dick was doing. And so, I was actually quite interested in those concepts.

And where our projects kind of led was, one is looking to see to what the intrinsic radiosensitivity of cells and potentially backed up by the inherent genetics of the cells an important factor in radio response or in fact was it all the micro environment. And that was one of the controversies at that time, as to whether or not hypoxia was the dominant biology, and therefore everything should be placed into hypoxia, or there was also an inherent genetic radioresistance of the cell due to the inherent oncogene or tumor suppressor gene or other intracellular singling abnormalities of cancer cells.

And so, the first project that I did with Dick was to actually look at the in vitro radiosensitivity of a number of different model mirroring tumor cell lines and then compare that to the in vivo response and actually try to prove that there was a correlation between the surviving fraction of two Gray and then if you gave multiple fractions of two Gray, the actual final response. And we did that and we did that over probably seven different models to show that there was a very strong relationship when we used ex vivo clonogenic assays and also when we used growth delay.

I think the huge learning curve for me was quantitative biology, and this is one of Dick's strengths but also was the fear of all graduate students, that they would have to really learn and understand a little bit of calculus and understand the statistics around true results and important results and looking at data and understanding about outliers as potential manifestation of perhaps experimental error but also outliers as a manifestation of really interesting science and really interesting results.

And I've always been very grateful to Dick to really be the person who taught people to count because a lot of the biology was coming out on the molecular biology side was quite phenomenologic as well as highly descriptive biology and qualitative biology, there was a band or there wasn't a band. But, you know, Dick would always ask, "But how strong is the band? How often does the band occur? Is the band dose responsive?" And I think these are very important issues. And as we get into the current overlay of now sequencing results and in-depth genomics which I'm involved with, one can't forget the tumor biology. And I think it's this interplay that was very important when we recently published the overlay of hypoxia on top of genomics as a potential stratifier for patients undergoing prostate radiotherapy, and you need both parties.

So, I think that I wouldn't have been thinking like that had I not kind of grown up in the Hill era of quantitative biology where we looked at this. And then later on when I went back and worked with Dick during my residency for Ph.D. we looked at p53, and so we actually looked at specific tumor suppressor gene and its involvement in radioresistance at the time. So, one of the controversies was the microenvironment versus inherent genetic resistance, and I think that that paper that was published was actually one of the important publications in the field to say that, yes, intrinsic radiosensitivity was quite powerful.

I guess I want to flip it back to Dick because as a clinician scientist I mean I wouldn't be here without someone like Dick who was a basic and translational scientist driving clinicians to actually be very aggressive and trying to answer questions in the clinic. And so, I would say that the clinician scientists who grew out of the program at the Princess Margaret Hospital were really on the radiation side because of Dick's, I think, vigor and enthusiasm about really answering a lot of these questions from a clinical realm. And I'm very curious about how that developed, Dick, in the sense of clearly someone like Ray Bush was someone who straddled both empires, so to speak, but also it can be frustrating because the clinicians at that time there really wasn't a model to try and do research at the same time as doing clinic, and yet you persevered over many years even up until now to continue to drive forward that ethos. So, what was it like trying to get clinicians involved in asking questions relating to the biology of the disease and then the ability actually to carry it out?

Richard Hill: So, you're certainly right. Ray Bush was certainly a major influence on the way that I thought. He also was trained initially in physics so I think what you were saying about the quantitation of biology is correct. A lot of that actually did arise out of physicists who moved into the biology area and radiation biology was one of the first areas that really tended to focus on quantitation. And, fortunately I think, that has now overlapped into most of biology and certainly in the areas that relate to cancer , so we don't see the kind of phenomenology which we used to see in the literature. I think that's great.

Certainly when I first came to Princess Margaret, the focus was on treating patients well and the biological research aspects of it were very much secondary. But Ray Bush was certainly a person who was really pushing that forward. And people like Gordon Whitmore and Mike Rauth, were really involved in trying to push various biological ideas forward and talking about them to the clinicians. Many of our clinicians got very much involved and a number of them have passed through the laboratory over a period of time. The current head of radiation oncology, Fei-Fei Liu, as you know passed through our laboratory. She runs her own laboratory now but for a while she worked in the kind of combined group that I had with Ian Tannock. So, that was the way that process worked to develop more interest and really try to apply research concepts to the actual treatment of patients. But I think it's still true today, and appropriate, that a major focus of radiation oncologists remains to make sure that they treat patients as effectively as possible with radiation and to do it well. Research to improve this is actually what we've been hearing about this morning, really how do we do that better. That's where Fei-Fei Liu is going, if you like.

Robert Bristow: Maybe I can follow up a little bit, and in two areas where I think there, I won't say has been controversy but an ability to enact the biology into the clinic. And so, one is hypoxia. And when you started off with Mike Rauth and Gordon Whitmore and kind of that trifecta of looking at radiosensitizers, hypoxic cell toxins, and animal models, there was a great deal of work and this led to of course chemical smodifier conference with Ged Adams in this area, and yet today we still are not driving forward radiosensitizers and hypoxic cell toxins as a matter of course in the daily regimen of radiotherapy. So, do you have some editorial comments on -- because you've been in the game a long time and you've watched compounds tirapazamine and in most recent the evofosfamide, really hit kind of the zenith of it in terms of both pre-clinical and then clinical excitement based on early trials and yet we still have not enacted it. And so, maybe from your perspectives what are some of the barriers, reasons that that has happened, and if you think of that so or are clinicians just lazy? Tongue in cheek. Tongue in cheek.

Richard Hill: I might just make the editorial comment that we now recognize that the microenvironment generally is a much more important factor in tumor response than we used to and that largely came out of the early studies that were done on hypoxia. I personally think that the problem that we have with hypoxia-targeting agents is that the clinicians have failed to convince the companies that are producing these compounds that hypoxia is not a universal standard phenomenon across all cancers and that they must measure levels of hypoxia in the tumors when they do these kinds of trials. What we unfortunately have seen is that companies have not been willing to actually make appropriate measurements to allow identification and selection of the patients who could potentially benefit. And it's certainly turning out to be more complex, than just the amount of hypoxia in the tumor. However, in the sub-studies where measurements of hypoxia were made, we have seen significant benefit for patients who have high levels of hypoxia in their tumors. So what one would hope is that we can convince the companies that they're wasting their time to just treat everybody with these drugs. They've got to actually focus down on the patients who can potentially benefit. That's really where the clinical groups who set up these trials have failed and we've seen it again, exactly, with this most recent study with evofosfamide. They didn't look at the patient population in terms of levels of hypoxia in their tumours, so that they couldn’t determine which ones would benefit. But, to be fair, one could say that that's an after-the-fact rationalization.

Robert Bristow: Yeah, exactly. And then the other area I think that again you were certainly highly vocal in challenging some of the concepts and then was actually a master of debate in this arena was the concept of stem cells and tumor stem cells and what was meant by that term, and again, the quantitative aspects relative to the qualitative aspects. And maybe you want to talk a little bit about that as well because certainly when the qualitative data started to come out to assign markers to such cells after they're having quite a bit of tumor biology to assign a quantitative nature to the cells in terms of the clonogens.

The term stem cell became quite obscure between these two areas of research, what was meant as a clonogen versus what was basically a biomarker assessed a cell that was a small segment of the entire of population of cells. And do you think that those two areas now have come together in terms of people thinking about stem cells and clonogens and re-growing of the tumor in addition to the biologic markers or do you still think that there's a way to go there? Because you were involved really in trying to stress the quantitative issues relating to clonogens and stem cells and try to get people to understand that in addition to the markers.

Richard Hill: Certainly stem cells is currently a very interesting area and, as you say, relatively controversial. Way back when I was working with Ray Bush, we were thinking about stem cells in tumors because of the work that had been done by Till & McCulloch on the bone marrow stem cells. The question that we were thinking about was what does this mean for cancer. Some of the early work we did had suggested that not every cell in a tumor was really capable of re-growing that tumor and this is essentially the stem cell concept. When we did the initial studies, it's about 25 years ago now, we chose to study spontaneous mammary tumors in mice and we determined the big single dose of radiation which was required to cure the tumors, and compared this to the number of cells which were required to actually transplant that tumor if you injected the cells into another animal. In a sense we were using an early isograft model similar to the current work with early passage xenografts from patient’s tumors. We were able to show that there was a correlation between those two numbers and that this was consistent with what we had expected in terms of differences in the radiation response of the different tumors. Furthermore that tied up very nicely with work that had been done by Luka Milas at the MD Anderson Hospital as well.

So, our initial work in that area was really consistent with the concept that different tumors might contain different fractions of stem cells. The current interest in stem cells has really, as you said, arisen out of the fact that you can now identify surface markers which allow enrichment of the cell population for stem cells or if you like to call them tumor-initiating cells because ultimately what we're doing is we're injecting numbers of cells back into the animals and asking how many cells do we need to initiate a tumor.

What is evolved out of these studies is that it's a lot more complex than we thought. It's not just an issue of stem cells and non-stem cells because it's now recognized that the stem cell phenotype is actually quite plastic. It's probably plastic in normal tissues as well as tumors, but it's definitely plastic in tumors. And so, cells which are often called progenitor or transit-amplifying cells apparently have the ability to re-express the stem cell phenotype if they're in the correct environment. And certainly one of those environments appears to be hypoxia, which is very interesting. But it's also been shown that at least for some types of tumors treatment itself can affect the phenotype of these cell populations or essentially push them back up the differentiation pathway,if you want to think about it that way.

So, it's becoming a much more complex issue. And what we still don't really understand is what the process involves and what kind of timeline is involved. When you give a big single dose of radiation capable of curing the tumor, as we did early on, there may be no issue associated with cells being pushed back up the pathway but if you're giving fractionated treatment, which is, of course the current therapy strategy, then maybe this is a significant factor. Certainly there's data in the literature which indicates that, over a course of days, cells can re-express the stem cell phenotype. But how that impacts on tumor response we don't understand yet. We know that it can happen but we don't know to what extent it happens in different tumors. Until we have some idea about how to quantitate what's going on in individual tumors, it will be very difficult to deal with that problem in terms of therapy application.

The other side of the coin of course is that maybe we can develop strategies which directly target the phenotype of stem cells and it may be that the combination of these with radiation will help to alleviate that problem, but there's no data that I am aware of, that allows us to really know how to do that. We also have to be aware that many of the markers of putative stem cells that were identified as appropriate for tumors are also the markers that are expressed on normal tissue stem cells. How we deal with understanding the possible effects of drugs targeting the stem cell phenotype on normal tissue response to radiation is also completely up in the air.

There's really no information in that area at all but one could certainly imagine that it will be much more complex than the current way that new drugs are introduced into the clinic. Currently we look short term to see whether there is sign of toxicity and then we move forward if we don't see it. But in the case of slowly proliferating tissues, the ones that usually show the long-term late effects in radiation, that strategy is clearly not appropriate for understanding what would happen if we eliminated all the stem cells in that tissue. We're going to have to develop a strategy which allows us to get at that issue if we're really going to be able to introduce drugs which are targeting stem cells, particularly if they have any possibility for targeting the normal stem cell populations. Having said that, I think it's very likely that over the next 20 years we actually may be on the other side of the coin with normal tissue. The development of so-called induced pluripotent stem cells and the research going on in that area will probably allow us to look at ways that we might actually alleviate severe radiation effects by transplanting stem cells or induced stem cells from that patient back into organs of interest in order to try and regain their function. We're probably 10 years away from doing that but I think it will be something that's going to happen.

Arjun Sahgal: I just wanted to kind of get your opinion also. Because recently there’s been a lot of debate between the Stanford group and Sloan Kettering about this kind of query new biology associated with stereotactic radiation. We’ve read the Red Journal back and forth between the two groups. And you were there and you saw that kind of evolution particularly from Sloan Kettering as they came out with single malonate pathway. Do you really think that there is a fundamental different biology associated with these high dose per fractions or is it dose?

Richard Hill: I think it's very much an issue of dose. I don't see a major conflict actually in understanding what's going on, in the sense that we now recognize I think, from the work that was done at Sloan Kettering, that the vasculature is radiation responsive but you've got to give a relatively large dose to see a big early effect on the endothelial cell population. So, the introduction of big, single dose fractions certainly changes the way the overall tissue will respond to radiation. And it comes back again to what we were talking about, it is the microenvironment.

We now understand that the critical aspects of the microenvironment in tumours is not just hypoxia, it's a whole range of other factors which impact on the environment that the tumor cells are growing in, and certainly when one uses large radiation doses, the issue of how the vasculature responds to those kinds of treatments is going to be important. In practice, I think that’s what the Stanford group have really been emphasizing as I understand it. One of the things you have to think more about when you give big, single doses is hypoxia, because you don't get re-oxygenation to the same extent as if you give conventional fractionation. This might be thought of as a negative but, on the other hand, there's a potential gain in the sense that you're damaging the vasculature to a larger extent and maybe, if you can prevent it from repairing you will get improved response. That's one of the things that Martin Brown has been publishing on recently; there may be ways to actually prevent the vasculature from repairing the damage that's caused by the radiation. You may then get further cell killing as a result of the fact that there's now difficulty in providing the nutrients necessary for the tumors to keep growing. This may build on the top of tumor cells continuing to die because they are in regions of severe hypoxia. How is that going to play out? I don't think we know the answer to that yet.

The other factor which I think is also potentially important is that we know that if you irradiate a tumor, actually not only radiation but radiation is a good example, you actually modify the influx of bone marrow-derived cell populations which actually get into the tumor, and how that is going to play out in terms of tumor response and the potential for inducing immunotherapy-type effects is still unclear. There isn't any doubt that that's another area of research which is going to develop over the next few years to help understand how the influx of these f bone marrow cell populations is impacting on the tumor response to treatment and potentially how one might manipulate that to really develop immunotherapy-type applications. Also, as you know, there's a lot of interest in the possibility that radiation might be modifying tumor cell expression of cell surface markers which could be recognized by the immune systems. So, we're certainly going to see a big increase in work trying to understand how to bring immunotherapy approaches to bear on combinations with radiation.

Arjun Sahgal: Do you think that based on kind of all the animal work and all the clinical data, there is an ideal fractionation or do we really have to use, say, what Rob does in personalizing the molecular aspects of tumors to then determine what is the ideal fractionation. Because you did some very interesting work with alpha-beta for inpatient trying to figure out whether or not you can tailor the fractionation size accordingly.

Richard Hill: I think the answer to your question is yes and no. We need to understand better how normal tissues and how tumors are responding to SBRT type treatments. For normal tissues I think what is going to play a major role is volume, as I was saying earlier,. The truth is that researchers haven't really put a huge amount of effort into studying volume except in organs like the lung and skin. Maybe that's true clinically as well since we don't really understand fully how that plays out. But what we are, I think, starting to understand is that normal tissue response to radiation is much more complex than we thought it was. It's not just a matter of how many cells you've killed in the normal tissue. We've already talked about the issue associated with vascular damage but there are a whole range of other factors including inflammation and the influx of bone-marrow-derived cell populations associated with normal tissue response to irradiation.

As we understand better how these various things come together, which I think still needs a fair amount of effort, we'll have a better idea of how to manipulate fractionation schedules so that we're minimizing normal tissue effects without necessarily compromising response of the tumor. We certainly know, as I said earlier, that radiation can affect the stemness of tumor cell populations and we also know that proliferation is a major factor. You can't extend radiation treatments over too long a period otherwise you start to lose the benefit of them. We are moving in the other direction at the moment towards fewer larger fractions and I think that we're going to see that applied to many more tumor sites. What is still unclear I think is how well we really understand normal tissue responses in those situations.

Arjun Sahgal: Do you think there’ll be like in 10 years any more of 1.8 to two Gray, like do you think that fraction will be gone and no more six weeks of radiation, seven weeks, putting people though just protracted schemes?

Richard Hill: I think the answer to your question is no. At least part of the reason for that is that radiation oncologists are cautious people, and rightly so, because what they can do is cause serious damage. So I think it's going to take a lot more than 10 years before we actually understand the whole situation well enough to really move people away from an effective schedule that they feel comfortable with and they know that the patient can tolerate.

Arjun Sahgal: Have you been surprised at the relatively low toxicity rates that are occurring, say, with stereotactic radiation? Have we always been too conservative in your opinion or still at the cusp?

Richard Hill: No. Lung is the site where this is really taking off and that's I think is due to the fact that lung has a lot of repair capacity and is a parallel organ. The radiation oncologists have recognized that for small volume radiation, it doesn't matter if you wipe out the function of the tissue in that volume. In essence they're no longer bothered about normal tissue response. And that's fine for small volumes or for regions which don't have large vessels where damage might cause a major problem. So, we still haven't really moved -- as far as I can see anyway. Maybe I'm not up to date with the list here but I think it is still lung where this is really being done in a very aggressive way.

Arjun Sahgal: It’s coming slowly.

Richard Hill: It's coming slowly but I think what we're going to see is that for some tissues it's not going to be as easy to just disregard normal tissue responses as it is for lung and that relates to the volume issue.

Arjun Sahgal: I guess I would say in your career, and Rob, maybe you can jump in also in terms of your perspective, but what do you think have been some of the more startling achievements that you would remember that occurred say since your Ph.D. that have influenced you and your research and thinking?

Richard Hill: There are a number of those. Particularly recently there's no doubt that the development of genetic analysis is really changing the whole way biology is being studied, so there's no question that that's a major issue. It's not particular for radiation, of course.

I think the other thing that's happened which has been very important for really studying tumor response to radiation was the development of immune-deprived animals and the ability to actually study human tumors growing in the animal model. As Rob pointed out earlier, when we were doing the initial work that he did, we were basically using cell line models and they were all animal models, because it was the only thing we could do to study tumor response in vivo. What's clearly come out over the last, I guess, 10 years of studying human tumours growing in immune deprived mice is that long-term cell lines are quite limited, and this is not only true for animal, mostly rodent, tumor cell lines but also for human tumor cell lines. They really do change a lot while they've been growing in culture and this means that we really have got to recognize that using early xenograft models are more likely to be reflective of human tumor response.

But the other thing which has also become very clear and that has come out of the genetic analysis work is the concept that one breast cancer is different from another breast cancer. And that's true for all cancers. So we now recognize that every cancer is slightly different from every other cancer and that has driven a new view of how we are perceiving cancer. And so, the early concept that the pathologist looked at the tissue biopsy and told you it was a breast cancer and that this implied a certain treatment approach is less and less correct.

I think it's somewhat disappointing that cancer diagnoses haven't changed faster in this area, but this is happening, as the pathologists move into the genetic area. In a way their movement into this area reminds me a little bit of the fact that radiation biologists were also a bit slow getting into this area. For example, there was a concept that cell killing was the critical issue and that genetics wasn't that important because certainly in terms of radiation most genetic changes don't make huge differences to the actual radiosensitivity of the cells. That transition has I think now been effectively made in the radiation biology community. I think it's still in the process in the pathology community. But again, it's an issue associated with caution. You need to be very sure when you're treating patients that you know what you're doing. It's a little bit easier in the animal model system to make these kinds of changes and investigate them.

Arjun Sahgal: So you made that transition towards more molecular aspects of research in the form of more classical experiments. I mean, what was that like at that time because you know, the whole processes of PCR and DNA and RNA, all of that was transcending?

Richard Hill: One of the values of having graduate students is they push you in certain directions. and you move on the basis of their ideas as well. These are the kinds of things where there is real benefit from having trainees who get stuck into thinking about different issues. However, it's a slow process and I wouldn't say that I've made that transition as effectively as some other people. It is a process which is ongoing.

Arjun Sahgal: What would your -- and Rob, if you want to jump in, because we wanted to kind of talk about the future a little bit more, but what do you feel about the field of radiobiology? Do you think it’s dying off slowly or do you think it’s morphing? And for someone who’s interested, what advice would you give them?

Richard Hill: There are lots of interesting issues that still need addressing in radiobiology but one current problem -- I guess, it's not actually a current problem, it's been a problem for a long time - is that most of the research money is actually going into developing new drugs, rather than really trying to understand the technologies which we already have on-site if you like. So, I do think that radiobiology suffers a little bit from difficulty of actually raising funds to push some of the new ideas forward. However, I think that's changing and I hope it will continue to change. Clearly we're going to move forward with genetics to a much greater degree and the hope is that this will in fact inform us substantially about radiation response. Particularly in the context of cell-cell and cell-microenvironment interactions.

One of the issues which I think will come out over the next 10 years is much more understanding of why people have different responses to radiation. Different patients respond differently and presumably there's a genetic underlie to that, at least part of it anyway. As the studies, which are going on now looking at the genetics of patients who responded poorly or well or had severe side effects, go forward, I think we'll get a better understanding of what genetic features might be important there. How that will play out, I don't think we really know at the present time. Maybe Rob has a better idea on that.

Robert Bristow: I think you've painted a very strong picture of where we've been and where we're going. I think now in the clinic, really we have a multidisciplinary approach to cancers, and Ray Bush was one of the first to really be described as the complete oncologist because he thought not only of radiation resistance but also systemic spread and adjuvant therapies, and I think in radiation oncology and radiobiology we have to continue to think along those lines because we have always been in a curative modality. And so, what does it take to cure a patient? I mean, one is to treat the local tumor as well as possible to prevent local recurrence but also importantly to prevent a secondary wave of metastasis which otherwise would kill the patient. But clearly there are some patients who will have a occult metastasis at the time of treatment, and given that radiation oncology is one of the first local modalities that's incumbent upon us to think about what is the systemic treatment that is not only going to sensitize the tumor to radiotherapy but also to offset the potential for mortality from occult metastasis.

So I think we continue to embrace that, and then the role of radiation therapy in addition to surgery and chemotherapy is also I think again one that has almost hit almost every tumor site that in order to effectively have increased local control or increased cure with metastasis, that we have to see radiation therapy as a component of therapy and when to bring it in, and whether it'd be the major modality or secondary modality is something that we need to think about. And that also goes back to what Dick was saying with respect to potential side effects. I mean, we talked about large volumes and then we're in the era of now partial volumes, and now we're in the area of radiotherapy given in concert with other modalities. And so this concept of toxicity is actually quite broad.

I think there continues to be an interest in using self-signal inhibitors but now checkpoint inhibitors with immunotherapy and this is now kind of a heyday for that type of research given the fact that we understand that radiotherapy could augment immune responses within patients and again its controversial and it's incumbent upon radiation oncologist. I think to drive that research in additional to medical oncologist to understand the biology behind it and that again which patients will respond or not respond.

And one more concept again is what we talked about is which of the patients require radiotherapy alone and which of the patients require combined modality, and how do we choose those patients appropriately upfront. And again, I think the onus is and our specialty too, to do that, to do that well, and to give really guidance to the oncology community of how to do that. It's a sobering result now after multiple institutions including this one that have looked at the molecular genetics of tumors when they have failed secondary or tertiary chemotherapy and are being tried to be matched to a molecular targeted agent. The ability to do so is probably only about five percent of patients for practical as well as molecular biologic reasons, which says that that's late in the game to try to affect cure, or let alone try to prolong life. So, our ability to take that type of approach and bring it earlier in the disease when there are not as many perhaps parallel pathways to resistance I think is very important. And so I think the pharma companies and the industry will be paying attention to that because cure is still an important concept in oncology and I think we have to take that back and really show that we can improve cure by tackling some of the molecular genomics upfront and again trying to match those tumors to appropriate strategies when that might be the driving force to resistance. Whereas down the road in many of the ways the medical oncology driven trials are when there might be multiple pathways so attacking one pathway is just fraught with failure.

So, I think these are some of the concepts that are coming up in the future, so I only see positives for radiobiology and positives for radiobiology layered on to radiation oncology but we must think broad and we must think about how to cure patients and that includes metastasis and preventing metastasis and treating all of the metastases. And these are all concepts that are in the lexicon I think of radiation oncology now.

Richard Hill: I certainly agree with that. I think what has to be emphasized is that more and more we're combining drugs with radiation, and we've got to look at this combination really in the frontline. It is not enough to just give radiation and then wait and get drugs later. That's got to be done together. So, understanding how to do that well and how to deal with potential extra toxicities is certainly going to be one of the questions. I think we're just going to have to engage radiobiologists to really understand how to do that effectively, to build on top of known tolerated treatments, if you like, to really attack specific aspects of the cancer.

And this comes back, as Rob said, to getting in as early as possible. And so, I think the concept that we can give the standard treatment and then if there's a recurrence, we could come in with something else, has limited efficacy It's pretty rare that you can come back with a second or third strategy and actually really prevent the patient from dying of the disease. So, we've got to move that up into the frontline therapy. Whether we can get the pharmaceutical companies on board to really test this approach is another issue because it implies, of course, that you're going to use drugs only for a short term rather than for a long term.

Arjun Sahgal: I guess what we could do is just kind of maybe end off a little bit, and we’ve been asking several people also just because of the era of particle therapy and our traditional use of photon therapy, in the next 10 to 20 years I mean there’s certainly been an increased uptake in proton but what do you think about carbon? Do you really think that we are going to be really different in 15 to 20 years in terms of our use of particle therapy? And why, maybe?

Richard Hill: I think the answer to that question is no, I don't see that carbon therapy is likely to be taken up in a very large scale even though it has potential biological advantages. Protons have been fairly widely introduce in North America, but many people would argue that that has more to do with the fact that you can charge more for it than it really is beneficial. But there clearly are sites where proton therapy is beneficial and it comes back to what we're talking about earlier which is reducing the volume of normal tissue which is irradiated rather than necessarily being able to improve the response of the tumor. It's really a situation where you can potentially give a larger dose because you're irradiating a smaller volume of normal tissue..

I think the physics of protons is still not fully understood and the exact concept of what fraction of normal tissue and what doses are actually being delivered. I think that's still a bit of a controversy in the literature. As I understand the literature that there are different physicists who think that the doses are not quite so tightly defined as the advocates of proton therapy would like to have us believe. But it's clear that protons can be beneficial. We've seen that and that's going to happen. I just don't see in the present climate that the introduction of carbon ions, which is another order of magnitude in cost, is likely to happen in the near future.

Arjun Sahgal: If the RB really is different, do you think it could be worth it though to explore that technology on a greater level?

Richard Hill: Well, we would certainly then have to think about which patients would benefit from carbon ion therapy and this comes back a little bit to our discussion of hypoxia. We know that one of the potential benefits of high LET radiation is higher RBE and that there's a much smaller oxygen effect. So, if we were going to do such a study, we would want to look at the various factors that we know impact on differences in RBE and really try and play that out in terms of selecting the patients who could benefit from it. But we can't at this time easily do that except with hypoxia. Maybe there are other factors that potentially are an issue. Carbon ion therapy does exist in other countries and there's a lot of data coming out which suggests that there are positive differences in patient outcome. But I don't think there's any clear message at the present time as to which patients will respond better to such therapy. I haven't got a clear message out of that yet.

Arjun Sahgal: Okay. Wonderful. Is there anything you’d like to add just at the end in terms of editorial comments?

Richard Hill: I don’t think so, we’ve had a good discussion and I think it is an evolving area.

Arjun Sahgal: Rob?

Robert Bristow: No. I think we covered it all.

Arjun Sahgal: Okay. Wonderful. Well, thank you very much, Professor Hill.