Integrating Radiotherapy with New Therapeutic Agents for the Treatment of Locally Advanced Non-Small Cell Lung Cancer: What Do We Know?&n`sp; What Do We Need to Know? by C. Norman Coleman, MD

SLIDES & TRANSCRIPTS
Wednesday, June 14, 2000

Integrating Radiotherapy with New Therapeutic Agents for the Treatment of Locally Advanced Non-Small Cell Lung Cancer: What Do We Know? What Do We Need to Know?
C. Norman Coleman, MD

Slide 1:

DR. COLEMAN: Thank you very much for the invitation. Let me first begin by wishing Drew Turrisi good luck. If he is watching this in real time, he is probably going to inhibit his recovery. So I recommend that he not watch this meeting. But the meeting is never quite the same without Drew, who not only adds great clinical perspective, but adds a good sense of humor and balance. So we wish him very well in his recovery.

Now the good thing is that Dr. McKenna and I did not coordinate these talks, and we don't overlap too much, as you can tell here. So let me tell you a little bit about where I am from now. Some people know I have moved in the last year to a new program called the Radiation Oncology Sciences Program, which is a multifaceted program at the National Cancer Institute, put together by Ed Liu, Division of Clinical Sciences, Bob Wittes, Division of Cancer Treatment and Diagnoses, and Rick Klausner, as director of NCI.

Slide 2:

So my expertise is not necessarily in lung cancer, but I'll try to talk to you a little bit about combined modality therapy, and some of the new potential applications and targets for radiation. I plan to cover where we stand at present in combined modality therapy for non-small cell lung cancer. Dave Gandara has summarized that all very well.

For those of us who worked with Dave back in the Northern California Oncology Group days, we are really glad that we trained our medical oncologists to appreciate radiation. Dave, we knew you wanted a cure, so you're still on the right way.

How we do really know, or do we really know how, radiation works? What are the available clinical strategies for combined modality therapy? I'll talk a bit about the reductionist versus the integrated or heterotopic approach to cancer; some new paradigms for how radiation oncology works, again, some new concepts. We now have so many molecular targets -- how do we credential and understand them? And lastly, some of the roles that the radiation oncology sciences program can have in helping individual institutions and the cooperative groups potentially advance new therapies to the clinic.

Slide 3:

So this is how most medical oncologists, and as some of you know, I am one, view radiotherapy. This is a reconstruction of someone having a prostate treated, but it's a two dimensional reconstruction. Multiple beams are used, and all focusing on the target. In fact what has happened now, for those of you who don't know, intensity modulated 3-D conformal radiotherapy, as Gillies just talked about, you can shape the fields to surround the tumor. You can use multiple fields and you can have a heterogeneous intensity within each beam. So you can essentially treat almost a field within a field by giving a higher dose to centers, or areas you want to give higher doses to, then lower doses to others. It is a much more elegant way of delivering radiation, and it is really just a matter of computer applications that is going to get this further advanced in the clinic.

Slide 4:

So my only discussion of physical aspects of radiation were that slide. What's coming forward is target definition. This is both imaging of a shades of gray, but also an imaging of functional status of tissues and tumors. Hitting the target, we can now use intensity modulated radiation. We have gated treatments, and we can use electronic portal imaging devices, so you can actually see in real time what you are treating. The real issue now is you get tighter and tighter in your fields, and you're getting more precise, but less accurate.

We can use intensity modulated radiotherapy. One new project we are developing, where we use mathematical models and all kinds of guesses for what radiation dose is. And we're now developing an optical probe, a 100 micron thick optical probe that can measure radiation dose and dose rate in real time. I think we will soon have the ability to actually know what we are doing with clinical radiotherapy, and brachytherapy is obviously useful for short distances.

Slide 5:

Now it took me a while to finally understand track structure experiments. This is a picture of ionizations and electron in a radiation beam when it is giving off ionizations with the electrons as it goes through tissue. This is a high linear intensity radiation. We won't focus on this. But this is where little energy deposits are given, and this is the nanometer size of DNA. Although radiation is not limited pharmacologically, you can penetrate membrane barriers very easily, radiation is not given off homogeneously in tissue. There are packets of high densities of ionization. It is this high density that can potentially activate targets, not just in DNA, but elsewhere in the cell. So there is heterogeneous distribution of radiotherapy when you get down to the nanometer level, and that's where really all the action is.

Slide 6:

In the past we used to just talk about ionizing radiation, that you had to have electrons energetic enough to actually create ionizations and molecules and DNA. There are a number of papers in Science now showing that when you use low energy electrons that do not produce ionization, you can produce transient molecular anions in DNA. These can move from molecule to molecule to molecule, and you can potentially produce single and double strand breaks with lower energy electrons than people thought about before publication of this in March. So exactly how radiation creates damage now is being brought to a whole different hypothesis.

Slide 7:

What happens when you give very, very low doses of radiation that doesn't kill cells? And this is from Sally Admundson and Al Fornace from NCI, and this is looking at CIP-1 and WAF-1 induction and GADD-45 induction, two gene inductions at very low doses of radiation, at a fraction of the 200 centigray of clinical dose. These are doses of 2 centigray, 1% of the clinical dose. This is at 25% of the clinical dose.

You can see gene inductions at very low doses, doses that do not cause cytotoxicity, that barely cause any cell cycle delay. But you can see gene inductions that are not only up in large orders of magnitude, but also persist for many hours. What this tells us is that radiation can perturb targets, can potentially give us targets that we can then use for molecular modification.

Slide 8:

So where are we with combined modality therapy at present? Again, not being an expert, fortunately I read the right papers just in time. It appears that concomitant chemoradiotherapy is probably better than sequential, although the window of opportunity approach is obviously interesting.

The one point I really wanted to make is that we have learned in the past from induction chemotherapy in head and neck, that when you give chemotherapy and wait three weeks and give chemotherapy, although the tumor may be responding in size, it's also has increasing proliferation. You may end up having a tumor that you think responds, but it in fact will do worse. One has to be very careful about using the window of opportunity when you have potentially effective curative modality, because you may end up stimulating cell proliferation.

Local control seems to improved with sequential combined modality therapy and concomitant combined modality therapy. Improved local control can improve survival. It's very effective. You can probably reduce distant failure rate by a small percentage only for those patient who have persistent local disease. In prostate cancer and other models, it's been shown that if you let patients fail locally, you can probably have a showering of metastatic disease. So local control is also important in distant disease to some extent.

The use of new standard chemotherapeutic agents, taxanes and others that have already been mentioned, that are now coming forward. One point that I think Drew Turrisi would have emphasized is the concept of dose intensity. We are learning that from really hyperintense radiation schedules that it appears that if you get your radiation treatment in a shorter period of time, and the same with combined modality therapy, you may have improvement in results. So in the dose-time ratio, time may be as important a factor or more so than actual dose.

Slide 9:

What are the concepts moving forward for combined modality therapy? We talk about spatial cooperation, when Bob Wittes first heard that he thought it was NCI and NASA working together. This is where chemotherapy and radiotherapy really have separate killing fields or killing zones. The potential use of both modalities at the same time with toxicity independence, or you can use them separated in time, which we tend to do. The classic Steel classification for combined modality therapy is additive, one plus one equals two; toxicity enhancement, one plus one equals three, and that's some kind of molecular interaction. And for protectors, which I will talk about a little bit, one plus one equals 1.5 for normal tissue, where you are studying normal tissue effects. I think for lung cancer, when you are treating such great volumes of normal tissue, the protector aspect of this equation may be very important as we move forward.

We now know we can target the entire tumor, but we shouldn't target the entire tumor. Some earlier talks mentioned that, including abnormal physiology, abnormal tissue components. What's important, the cytostatic agents may allow the optimization of spatial cooperation. It's sort of a trick I learned in the streets of New York City. One holds him down, while the others kill him. We may need to use the cytostatic agents to hold the tumor down while chemotherapy and radiotherapy can take turns picking the wallet and other things.

Slide 10:

So again, Dave Gandara gave a really wonderful summary in the fall ASCO symposia. How do you integrate new agents? Well, for new agents you use them in low dose radiosensitizing ways, and also in high-dose cytotoxic ways. And the schedules will really vary from weekly to prolonged continuous infusion. The real MTD for radiation modification is really a chronic MTD. It's not a single dose MTD. That effects the design Phase I and Phase II studies. It really addressed questions a little differently than we do with just drugs by themselves.

Slide 11:

So what are some of the radiation modifiers that are available for clinical use? Again, you have heard of some of these -- the ones that increase DNA damage, the halopyrimidines, which are really no longer active. I won't talk about all the chemotherapeutic agents are available.

The hypoxic sensitizer field, people write this off as being dead. In a metaanalysis by Jens Overgaard for all hypoxic therapies and in a randomized trial for head and neck cancer using nimorazole, there was about a 5% or 7% difference in survival by antihypoxic approaches. These aren't great breakthroughs, but they are not worthless approaches. We have heard about hypoxic cytotoxins, tirapazamine, and Gillies showed the data from Australia, Lester Peters showing really incredibly exciting data for platinum plus tirapazamine and radiation for head and neck cancer.

Slide 12:

Now the paper from Hanahan and Weinberg has been shown in many talks throughout the land since it was published. In the past, we in the laboratory, and most people in cancer research, have used a reductionist view where one focus is on the cancer cells. But in fact the heterotypic cell biology view now illustrated here reminds us that cancer cells are not just all cancer. You have cancer cells, you have endothelial cells, you have stroma, you have inflammatory cells. And the stromal elements, as we have heard already today, could be very critical in the response of the cancer cells to therapy.

Slide 13:

Jim Mitchell and Laurie Hersher in Jim Mitchell's lab about ten years ago, looking at double staining techniques. This is the propidium iodine in lung cancer specimen, staining nuclei, and the yellow is antibody, a stain for white cells. You can see when you look at tumor, there are tumor cells, but there are lots of white cells.

Slide 14:

There is another slide with an admixture of malignant cells and normal stromal cells.

Slide 15:

They did some very careful quantification work on the percentage of cells in a cancer that were not cancer cells. What they found in squamous cell cancer of the lung and adenocarcinoma was that 50% or more of the cells were leukocytes, and this is not including the stromal cells and what have you. As you often know, when you do a biopsy, you sometimes have trouble finding the cancer cells for all the other stuff that is in there. And all this other stuff that is in there, I think make important therapeutic targets.

Slide 16:

Now Gillies showed a technique of oxygen imaging using tumor biopsies. This is something that is coming down the line from Jim Mitchell's group using a thing called over-enhanced MR, which is a cross between EPR and MR. They have a triple radical molecule that reacts with molecular oxygen, which gives a signal in real time where you can actually measure oxygen concentration in vivo without perturbing the animal.

You can see the bright signal is the signal from this molecule. Brighter means more oxygen. This is a mouse without a tumor. These are the kidneys. With 21% oxygen you see a brighter signal. When you lower the oxygen, you see a lessened signal. Bring it up, down, up down. So I think this may be an example of things to come, where one can actually measure molecular processes in vivo.

Slide 17:

Now the targets for radiation are numerous. We have heard about DNA, and that's probably the major target. One can potentially modify radiation response by molecules that are involved in DNA damage recognition and repair, such as AT, p53, BRCA, and all these other genes and gene products.

What is interesting is that most cancer tendencies involve errors in DNA repair. We may in fact have a great source for potential molecular targets as we really try to understand the role of DNA repair in these freaks of nature, these genetic lesions that give us cancer susceptibility. So learning how BRCA-1 works and their targets, MDS and others may be a great way of developing new targets. But one may also try to make a normal cell look like a premalignant cell. Some groups are developing tests -- Tej Pandita and Michael Kastin's groups are working on inhibitors of ATM, which may be potential radiation modifiers.

Slide 18:

We have heard briefly from McKenna on non-DNA targets, and there are many. It's almost every process in the cell. It could be cell membranes, apoptosis. The basic FGF story we heard this morning may involve the ceramide-created apoptosis pathway, the signal transduction intermediates, kinases, and phosphatases, DNA damage recognition, response, and repair, RNA stability, transcriptional apparatus, almost any biological process you can name, apoptosis, cell cycle, protein stability and degradation, other organelle, cell-cell communication, extracellular factors such as growth factors, inflammation, immunologic perturbation, and inducible transiently expressed genotypes.

Then there are lots of targets for radiation that are non-DNA related. Some work that we are doing, which I won't show here relates to COX inhibitors, which we are doing in prostate cancer. These appear to work in vivo and in vitro. Some may be mechanisms directly related to DNA damage and repair and response in the cell. Others may relate to the COX effects in normal tissue. But this is a new potential drug.

What we also talk about is the microenvironment phenotypes. What is important is that we talk about genomics, but what is really important is protein expression. A number of groups in the past have looked at p53 effect based on redox state and have shown that by altering the redox state of a molecule, you may take a perfectly normal molecule and make it relatively ineffective. As we learn about oxidative stress, microoxidative stress, a lot of the molecules that we think of functioning normally may in fact may not be functioning normally, and provide molecular targets.

Slide 19:

So the molecular targets approach to radiation will include two approaches, the class of DNA target, with the rationale of increasing DNA damage in cell killing, and the non-DNA targets. The rationale here is just to abolish cellular homeostasis. You may be able to give a drug at a fairly non-toxic concentration that has absolutely no effect, only when you add radiation or a chemotherapeutic agent will the cell actually die. So it's a way of using potentially toxic drugs in much lower and safer concentrations, and then focusing the cytotoxic therapy, which in this case would be radiation.

Slide 20:

There is quite an interest in understanding normal tissue injury. This is a cartoon from a recent paper in the New England Journal looking at TGF, and the TGF pathway are now being worked out. But people have been interested in late radiation response. Is late radiation injury, which is in fact the ultimate dose limiting toxicity, a permanent process, or is it some kind of chronic, sustained, inflammatory process that you can perhaps intervene many years after the event? You can give high doses, get people through the acute toxicity, and then fix up late toxicity down the line.

Slide 21:

There is work just published about a month ago by Michelle Martin in the Red Journal looking at TGF beta-1 and radiation fibrosis. This is looking at pig skin, looking months after radiation, and looking at TGF beta-1 expression in the skin late out, from as far as a year out, showing the sustained increase in TGF beta-1 concentration.

Slide 22:

Her group also took normal fibroblasts, and took fibroblasts from radiated skin and grew them out. And the fibroblasts in the radiated area were able to sustain increase production of TGF beta, suggesting that something about the radiation process may activate TGF beta. In data I won't show, they did a study published, I think, in JCO about a year ago using tocopherol and pentoxyphylline or two other agents, they were able to reverse four or five years out, severe late radiation injury, again suggesting this is a chronic, continuous process, rather than a permanent process, and may be a whole novel way of targeting high dose therapy, and then reversing the downstream late effects.

Slide 23:

There are a good number of agents now that may be involved in TGF beta modification, and this paper contains it in the Red Journal this year, in Volume 47, page 277.

Slide 24:

Another clever paper was published in Science earlier this year with the idea that you could potentially inhibit p53 to suppress the GI toxicity from chemotherapeutic agents. Presumably you can do the same with radiation. So you can take advantage of what the malignant phenotype gives us in resistance from cell killing, and apply that to normal tissues. This is just another clever approach that I wouldn't be surprised would be used down the line.

Slide 25:

So how do we help expedite translation from the laboratory to the clinic? This is some of the stuff that we hope to do at NCI, and do in conjunction with all of you folks.

Slide 26:

Intermediate endpoints are important for molecular targets, and this includes biopsies and laser capture, and some of the work presented today from UC-Davis is interesting, using DNA from serum.

One must identify molecular targets and assess the response of that target. One wants to know if you're doing what you think you're doing when you give your agent. These involve microarrays, which you are all familiar with, and molecular and functional imaging.

Slide 27:

What are some of the programs I will be involved in with the Radiation Oncology Sciences Program that we think can potentially help the field in general move forward? Workshops such as this, like the Radiation Research Program that we sponsored. We try to support collaborative approaches such as interactive program project grants among institutions, and a number of institutions are now working in that direction. We're developing something called the radiation modifier evaluation module with Ed Sausville's program DTP, a way of developing systems so when you have a drug that you want to test with radiation, and the pharmaceutical company doesn't want to invest in that, we will hopefully have a facility that can allow that to happen within the NCI, and I'll illustrate what we plan to do.

We, in the intramural program, have the advantage of not having to charge for health care, and there is the potential that we could collaborate with extramural investigators to do some of the Phase I and Phase II studies much more quickly. Then if they look promising, exploit them for the cooperative group. At the Radiation Research Branch we are doing molecular imaging, and that's going to be a major emphasis of ours. And we have an advanced technology center that we can do microarrays, and we look forward to doing that in collaboration with extramural investigators.

Slide 28:

This is the radiation modifier evaluation module. This is my first program, and the most important thing we do in the NCI is to develop an acronym. I thought this was important to get this through, so I named it RAMEM. Once you give something an acronym, it's like it's been there forever.

What we hope to do in this is some screening and some high frequency screens to look for radiation-drug interactions. But many of the new agents are going to be cytostatic agents, and they are going to need in vivo tests, and particularly normal tissue protectors are all going to need in vivo tests. So Rosemary Wong is putting this together. We plan to have cell lines in vivo and in vitro to normal tissue assays, where if we get this organized, we'll have an oversight committee that people can submit ideas for new drugs. It's almost like the pre-RAID concept. We can help you get new concepts tested pre-clinically and out to the clinic.

Slide 29:

The other role we can work on is as I mentioned before, the early Phase I/Phase II, doing this in collaboration with extramural investigators. Maybe someone will want to come and spend a year and help do a clinical trial, and have access to some of the things we have at NCI. But again, we want to help potentially facilitate novel concepts going from the laboratory through to where the cooperative groups can do the later Phase II and Phase III studies.

Slide 30:

We have a big molecular targets program that focuses on radiation oncology, including a RAMEM module, the microarrays, basic research groups of our own, and lots of collaborators from imaging to genetics, biology, and so forth. I think we are developing a very small, but hopefully mighty molecular targets program.

Slide 31:

What I want you to do is to conceptualize radiation in a different way. Most of you think of radiation in the macroenvironment or the normal environment, where we think of tumor and normal tissues, and physics is designed to help target the tumor and miss the normal tissue. Radioprotectors all work in this macroenvironment and this normal environment. These are predictive assays, or studies we do, pharmacokinetic studies. The microenvironment is cell-cell interaction. We heard how important that is. But when you get to nanomolecular, to the nano and pico environment, that's where radiation molecular perturbations are exactly the same as what you do with chemotherapeutic agents. And that's where these fields really come together. So I have a concept called nano intensity modulated radiotherapy. You want to really get the drugs and the radiation energy and deposition together in the right place at the right time. I think this is where all the action will be.

Slide 32:

This is one of my cartoons. The concept of this is don't think of radiation as three-dimensional colored isodose curves. Think of radiation as focused biology. It can bring the perturbation where you want it. It can stimulate pathways that your molecular modifiers can use. It can use drugs in lower doses, because this has the ability to kill. The targets are just illustrated in cartoon fashion, but you just name it and it's there from the cell to all its processes, to the extracellular environment and so forth. So there are lots and lots of targets, almost too many targets now, when 10 years ago we didn't have enough targets.

Slide 33:

What are the conclusions? Radiation therapy is focused biology. Technical improvements may allow increased dose intensity by decreasing time and by decreasing the volume of normal tissue. Imaging will be essential for target definition and for molecular processes. Intermediate endpoints will help identify and credential targets. Molecular therapeutics are rapidly emerging so that the pre-clinical models are increasingly important.

One has to really begin to interpret the pre-clinical models with cytostatic agents properly. You can hold things in check, but maybe not kill them. So you have to have the right model for the right drug. And again, many new agents will be cytostatic, so that radiation and other cytotoxic therapies will be essential.

That essentially ends my formal presentation. We are stimulated to have a little fun here and we always get discouraged about our lack of progress or our inability to sometimes move fields forward. We don't like our survival curve. But this is a very, very sobering slide.

For those who search the Internet and have found a website called AThe Onion@ B "World Death Rate Holding Steady at 100%." No matter how hard we work, there are some things we simply can't fix. Thank you very much.

[Applause.]

Slide 34:

DR. SAXMAN: Are there any questions?

DR. GANDARA: Norm, radiation pneumonitis is a life threatening problem in a small percentage of patients for essentially all the new chemotherapy radiation regimens. Are we at the point where we should be measuring some either TGF beta or something else, and is it practical? In other words, can we identify perhaps either prospectively or at least retrospectively, some patients who are particular risk independent of volume and other issues that might be related to their radiation therapy planning?

DR. COLEMAN: The answer is possibly. Mitch Antcher and his group at Duke have looked at pre-radiotherapy TGF, and those who have higher circulating levels seem to be at higher risk for radiation pneumonitis. I think these are the kinds of reasonable correlative studies that one can do. You can prove that hypothesis by in fact doing that, or test the hypothesis. And now that there are molecular interventions or therapeutic interventions, the number of patients you found that entered as TGF beta that predict for that, perhaps you can reduce the TGF beta by these other blocking agents, and prevent radiation pneumonitis. I think these really very good ideas. We forget to work on the other side of the equation as much as we can. If you can protect normal tissues by a factor of 50%, you could really increase your dose intensity.

DR. SAXMAN: Dr. Coleman, using your sort of model at the nanoscopic level, one of the things that we are finding in medical oncology is again the MTD of these drugs may not actually be the appropriate doses. The sort of paradigm with radiation oncology has always been give us as much as you can, short of harming the patient.

Do you think that there is going to be sort of an equivalent model with radiation in that giving what would have been considered "suboptimal" doses, but some maximum biologic effect at this nano level is somewhere down the line, or is it always going to be as much as you can give?

DR. COLEMAN: That's a really great question. There is some data now looking at a thing called low dose hypersensitivity, where if you give a cell a priming dose of radiation, and then give it a low dose, and then give it a second dose, it is relatively radioresistant. And if you keep your dose below a certain level, at the really top of the survival curve there is a hypersensitivity.

Sally Edmundson is looking at just a couple of genes or microarrays to see what is turned on. We may be able to understand what processes you don't want to turn on which will make a cell relatively radioresistant. Or you may understand if you turn that process on, you need to turn it off and you will sensitize the cell.

So I think low dose radiation works in magic ways a little bit. Why brachytherapy works, we haven't really understood. I think there is a little bit of hypersensitivity. There is probably something in the low dose part of that survival curve that you can potentially take advantage of, and that may be by using low dose refraction. Some groups are doing that in brain tumors and reporting anecdotal spectacular response rates in very, very low doses, but very frequently.

Monoclonal antibody therapy may in fact work by this low dose hypersensitivity region and brachytherapy may. So I think as we understand what radiation does, and what the consequence of that are, we might use radiation in different doses for molecular modification. For gene therapy you may want to use a very intense dose to activate genes. You may want to activate some signaling processes, which you can then use to kill the cell using lower doses of radiation. I think, like drugs, we are going to begin to use radiation as a drug, which is how a lot of us think about it.