An Interview with Dr. Kevin Camphausen

Dr. Kevin Camphausen

 (Interviewed by advocates Helen Schiff and Musa Mayer)

HS:  First, to introduce you, I want to ask you what you do and what your position is. 

Kevin Camphausen: I’m a radiation oncologist and Chief of the Radiation Oncology Branch of the National Cancer Institute, NIH in Bethesda, Maryland.

 HS:  What are your views on stereotactic radio surgery vs. whole brain radiation? 

KC: For many years we’ve treated patients with any histology who develop a brain metastasis with whole brain radiotherapy. Over the years, there has been a series of trials that have shown that whole brain radiotherapy improves survival, and that the addition of whole brain radiotherapy to a neurosurgical debridement (removal) of a metastasis improves survival.

Usually when a metastasis goes to the brain it has some associated side effects, either because of bleeding, or because the mass is impinging on some function.  The patient could have a seizure or some other manifestation of this thing growing in the brain that shouldn’t be there.  

Typically, patients are broken into two categories. The first is those patients whose primary disease--we call it disease below the clavicle (collarbone)--is currently under control, and they have a brain metastasis. The other group of patients has disease below the clavicle that is not controlled, and they have a brain metastasis.  The reason we split the patients into those two functional categories is that patients who have no active disease elsewhere, but only in the brain, don’t necessarily need a whole bunch of chemo at that time, whereas patients who have active disease below the clavicle need chemo that’s directed against their primary tumor. So, that’s a division point that we make.

For a patient who has a metastasis growing in her brain, typically she would be evaluated by a nurse or by a general practitioner. That’s typically after an MRI of the brain. The MRI of the brain can help with diagnosis of very small lesions that would be hidden if only a CAT scan of the brain is used. A neurosurgeon may evaluate the patient. Usually, if there are one, two or potentially even three metastases in a resectable location, the neurosurgeon may take the patient to have a resection, for one of three reasons. One reason is to remove the tumor, the second is to make the diagnosis, and the third is to reduce the symptoms that the patient might be having at that time, which can be relieved immediately at the time of surgery.  

There have been a series of clinical trials to find out what is the best procedure following a neurosurgical procedure. We’ve found over time that the best thing to do next is whole brain radiotherapy.  Usually there are several different fractionation schemes and the treatment is usually given over the course of one to three weeks. We target the entire brain, not just the area where we saw the lesions. That’s sort of standard of practice in the United States.

If the patient is not able to have neurosurgery, typically she would be a candidate for whole-brain radiotherapy alone, which is also the standard of care in the United States. We can also use another procedure called stereotactic radiosurgery, which can either use a gamma knife, or be LINAC-based radiosurgery, or SRS, in order to treat these metastases much like a neurosurgical procedure. We use a single dose of  radiation, a much higher dose than we use in whole-brain radiation, in order to treat these local areas as if it were neurosurgery.  

The question that arises right now, and that several groups have attempted to study, is whether a neurosurgical procedure is equivalent to a radiosurgical procedure. There have been several trials that have attempted to answer this question. However, the trials have failed to accrue enough participants and have been closed; therefore that question has never been answered.

For breast cancer, it’s a slightly different problem. For the most part, metastatic breast cancer patients who have controlled disease below the clavicle, who have developed brain metastasis can live a number of years, whereas a person who has a lung cancer or another primary cancer that develops a brain metastasis--their survival can be measured in months. So the question has come up, “what are the side-effects of giving somebody whole brain radiation?”  It is thought there would be changes in neuro-cognitive function. In breast cancer patients who have no active disease, should we offer them a stereotactic radiosurgery and not whole brain radiation? That is opted for by many people and it’s thought to be standard practice, but it’s actually not standard practice. Standard practice is still to do whole-brain radiation followed by some sort of radiosurgical boost--and that’s sort of the debate right now.  

HS: What is a radiosurgical boost?  

KC: For the patients who have one to three metastases, but are not good surgical candidates for neurosurgery, often what we will do is give them whole brain radiation followed by a radiosurgery boost. Because either it’s against something that’s not resectable or the patient is not a good candidate for surgery, we’ll substitute radiosurgery for that neuro- surgical procedure.


HS: Do you do this for something that’s left behind after the whole-brain radiation?  


KC: We do it typically for any patient with one to three metastases that are smaller than three centimeters that are in a non-vital area of the brain. So we wouldn’t do it against the ciasm or right against the pituitary stalk or something like that. We do it after the whole brain radiation and so it’s basically adding the boost that we think is comparable to the neurosurgical debridement.

HS: The purpose of the stereotactic SRS is to prevent local recurrence and the whole brain radiation is to prevent recurrence in the entire brain? 

KC: Within-organ recurrence.

MM: Why is dealing with the whole brain important?  

KC: You have to be careful that you classify patients into those who have active disease below the clavicle and those who don’t.  It’s thought for the patients who don’t have disease below the clavicle that some of them may have an isolated brain metastasis.  They may not actually be showering their brains with metastatic foci from an active primary tumor, so some people think that if we have a single focus that we can irradiate, that the rest of the brain is not harboring micro-metastatic disease.  So therefore the patient would not need whole-brain radiotherapy.


In order to see something on imaging, even MRI, it has to be 5 mm. to 1 cm. A 1 cm tumor has one times 10 to the ninth, or 1 billion cancer cells in it. One could argue that a metastasis smaller than that—that has 100 million cancer cells which can’t be imaged—is still  a problem. So we have a problem with our imaging not being sensitive to see these small lesions.  


MM: When you say no disease below the clavicle I assume that you do not mean metastatic disease that is under control in treatment? So you mean there has been no prior manifestation of metastatic disease?  


KC: Well, they can have had prior metastatic disease and it’s currently under control, but not under treatment.  So if they have active disease under the clavicle, I think that puts them in a different group. Where we know that they have either lung disease or liver disease  that are also showering metastatic foci every day, that really puts the whole brain at risk from not only the ones that we see, but from new foci that are spreading every day.


MM: How would you answer the argument those on the other side make, that if women are living longer with controlled disease with Herceptin, the late effects of whole-brain radiation in maybe a year or two years can be dementia?  Maybe if they meet the criteria, it’s better to just use the stereotactic radiosurgery and then carefully monitor the people and use whole-brain radiation if they have a recurrence. 


KC: Absolutely.  We just have to be careful about the vernacular we use and the apple and oranges we are trying to compare. So the data that we have about “dementia” in patients that got radiotherapy, many of those were done in an era where we gave much larger doses of radiation and much larger fractions of radiation. And so this “dementia” is not a well-studied thing with current modern radiotherapy techniques. We are getting ready to open up a protocol where we are going to monitor everyone who comes through with about 15 different neuro-cognitive assays  to try and determine what the best assay to use is and to see if there is a decrement, or an increase in dementia. We know that the worst thing that can happen to somebody’s neuro-cognitive function is to have a recurrence of the metastasis or a new metastasis. No other treatment has a bigger decrement in their neural functions than a recurrence.


MM: Even if it’s caught early, with 3 months scanning?   


KC: Absolutely.


MM: is there data on that? 


KC: Sure. There’s lots of data. But again it’s a little bit of apples and oranges. We can’t just scan patients whenever we want. It’s not paid for. So for brain mets follow-ups, they usually get one month, 3 months, 6 months, and then a year. And if it’s much more you have to pay for it. And there’s a big difference between catching something small and catching something big.


HS: how much does do the scans cost? 


KC: Anywhere between $1500 and $2500 apiece. And again it’s only going to image something that’s 100 million cells and up. Nine hundred million and 999 thousand cells it won’t see, which I think is plenty of cells.


HS: Have you ever seen anyone who lived a long time with whole-brain radiation and then got real dementia? 


KC: Oh yeah, I’ve seen it definitely.  But the thing we have to remember and we are specifically talking about breast now, is that there’s a whole new dementia we’ve only been talking about the last few years, and that’s Adriamycin dementia. Ladies used to get treated with very high doses of Adriamycin.  The first problem we found out was it is cardio-toxic. Then, we decreased someone’s lifetime exposure to Adriamycin to 350 or 300.  Then what we found out years later is that these women were getting what we call late Adriamycin dementia. They’ve never had whole-brain radiation, yet they developed all the symptomology of dementia. So one could argue in all these cases that have been thrown up about dementia we can say, “did they get Adriamycin?”


HS: Is that just true for Adriamycin or is it chemobrain in general? Is this something different than chemo brain? 


KC: It’s similar for chemo brain but Adriamycin is specific to the breast cancer ladies.  Mostly because that’s where it has been used and that’s where we have the longest follow up. A lot of patients we treat, say in lung cancer, don’t have long term survival so we don’t have late dementia to worry about.


HS:  What is radionecrosis (death of brain tissue) from this stereotactic radio surgery and can you get dementia from SRS? And is there a way besides a biopsy to see if it is a recurrence or radio necrosis? 


KC: Sure. So the incidence of radionecrosis with radio surgery is pretty low right now. We’ve learned pretty well about the volume and doses we can use, so the incidence of radio-necrosis is probably less than 5% using modern techniques. Radionecrosis itself doesn’t necessarily cause dementia, because it is usually a very focal area, but what it can do is lead to swelling which requires steroid use. So sometimes we do have to have a neurosurgical procedure to remove that area of necrosis so we can get the patient off the steroids.


For a patient who has a primary brain tumor called glioma, those are usually a lot bigger than a brain metastasis.  So sometimes we have the same problem: that we don’t know if it is a recurrence or a necrosis, and we use PET scanning to differentiate. The problem using with using PET scan for metastases is that they tend to be a lot smaller. And so if you get a study and it lights up, you’d say “oh, that’s a recurrence.” But if it lights up, you don’t know if it’s truly a necrosis or if it’s just too small to see. PET scans are pretty expensive right now and as far as I know they are not approved for radionecrosis vs recurrence for metastasis.  


MM:  Is the risk of radio-necrosis the reason why larger tumors aren’t treated with SRS? 


KC: Yes. So you can imagine that it is a volumetric issue. If you have a marble sized area and you need to treat the marble sized surface to a dose, and if you had an orange and needed to treat the surface of that orange to the same dose, the center is going to be very, very hot. There are basically concentric circles of dose. It’s a volumetric growth, from a bb, to a marble, to an orange. Even though the outside dose, the surface, is getting the same dose. The inside is going up by a factor of three. So volume is the biggest issue there.

HS: What’s your cut point (beyond which you won’t use stereotactic radiosurgery)? I notice in the literature some people use 3 some 4… 

KC: For the most part they use three. I think that’s the radiation oncology therapy group recommendation.

MM: And what about multiple metastases?  We’ve been talking about single or isolated mets for stereotactic radiosurgery.  

KC: We try and stay similar to the neurosurgery literature, we do one to three.  It sort of depends. If we can do them all at one sitting, we’ll do them all at one sitting. If the headframe has to be placed multiple times, we may not be as interested in that.

MM: Do you consider multiple metastases as prime facie evidence that you may have many micro-metastases that you just don’t see yet?  

KC: I do. Yes.

 HS: About screening: I’ve read that they are putting two machines together, something with spectroscopy to try and figure out if it’s radionecrosis or a metastasis.  

KC: That’s basically a software program for use with an MRI. You can use it with an MRI, or you could look at just the regular T1, T2 images, or you could do something called perfusion which looks at blood flow, or you could use something called spectroscopy.  Again the issue with all of these is that the technology has not quite caught up with the smaller size we need. 

 MM: Is there technology that will be able to detect much smaller brain mets on the horizon?  

KC: You know I think the higher Tesla magnets that we’ve got out there—the standard magnet is about 1.5 Tesla.  We’ve got a bunch of threes, people are looking at 7 even 11, people who serve on the research arm, so maybe as we go up in the magnet strength we’ll be able to find smaller and smaller things. That’s one side. The other side is that people are doing a lot of things like gene array profiles, where they say which group of patients are more likely to have brain metastasis.  And those patients are getting more in-depth screening. Maybe something like that would be more beneficial than a higher Tesla magnet.

 HS: We know that the incidence in HER2-positive metastatic patients of brain mets is higher than for other metastatic patients. What do you think about screening asymptomatic metastatic patients, and doing prophylactic radiotherapy as they do for lung cancer?  

KC: There’s a couple of small studies out there, patients who have brain metastasis, their HER2 status from the brain met to the breast primary are only 50% linked.  So their primary tumor may have been HER2, but the metastasis may no longer be HER2; which is why it doesn’t respond to Herceptin, and there’s the blood brain barrier.  So the other question you asked is whether we would consider doing prophylactic radiation in these patients.

 HS: Well, also whether you would do more scanning.  

KC: Sure. I don’t have a problem with more scanning. I always think more scanning and screening is better, as long as you are going to do something with the information. So if we scan somebody and we find very early that they have a small focus of metastatic disease, what are we going to do about it?  Are we going to offer them whole-brain radiotherapy? Are we going to do a stereotactic procedure? Is it any different than what we were going to do if we found it when they become symptomatic? Or, are we going to wait until they become symptomatic, in which case we shouldn’t be doing the screening? 


Before you set up a screening program, you have to decide what you are going to do with a positive finding. And I don’t have a problem with doing that. We just have to decide what we are going to do ahead of time. But for prophylactic radiotherapy, it is a very difficult thing in the United States because of neuro-cognitive issues. So there have been two large trials using prophylactic cranial radiation in a different disease called small cell lung cancer and it’s been shown beyond a doubt not only is there a decrease in local failure in the brain, but there is  actually an improvement in overall survival, which is rare for a radiation alone study. And still most people don’t offer their patients prophylactic cranial radiation because of the fear that they will get dementia.  


HS: But that’s in lung cancer?  


KC: Yes. That’s all the data we’ve got to go on.

MM: And there is a higher incidence of brain mets in lung cancer than in metastatic breast cancer?  

KC: Yes. Much higher. Now the standard is typically 3 if not 5 different chemotherapy agents, all of which have neuro-cognitive toxicities themselves. So you are setting up the patients to have some problems.

 HS: You say you don’t have any problem with scanning so long as you know what you are going to do, but what would you do with something very small?  

KC: The standard of care answer is easy. The right answer is very, very hard. We don’t have all the information yet. If I scan somebody and I find a very small lesion what I’d do is recommend that she go to one of the places that’s studying on protocol an SRS-alone protocol. So do a radio-surgery into this area without whole brain radiation, and then scan her very, very closely to see what the patterns of failure look like. Is treatment failing to control the first metastasis?  Is there treatment failure elsewhere? What does that pattern look like? What sort of neuro-cognitive effects are there from SRS alone?  So that would sort of get back to possible toxicity from Herceptin that we don’t know about and/or the Adriamycin she already had.

 HS: We know that in a lot of these people, 20-30 percent have asymptomatic brain mets you don’t necessarily want to treat these people if it’s toxic. 

 KC: Yes. They’ve been walking along fine.

 MM:  Will we have actual data to reassure us about long-term toxicity from whole brain radiation. Is that being carefully studied?  

KC: We’re working on that right now. Lots of other people are too.

 MM: And when might we expect some good neurological or neuropsychogical data on this?  

KC: Well, that’s always a moving target. So if we finish the trial, we are looking at studying 60 patients, probably take about a year and a half to accrue. When should we present that data? 3 months, 6 months 2 years, 10 years?

 MM: Probably all of the above. 

KC: Absolutely.  So all those things are related. From this point I’ll give you six months data in 3 years. We’ll accrue off of that. Lots of other people are studying this. Mine is not the only trial. There are other trials out there. It’s expensive. There’s neurocognitive testing and all these MRIs. And there are issues that I’m learning about that I didn’t know, how you even test people. They can actually learn the test that you are giving them, if you give it to them before the thing. You know whole-brain radiation takes only two weeks. So if you do the test before they start treatment and you want to do the test within a week after finishing the whole brain, it’s only 3 weeks between. Many of these patients are fine and they remember the test from the previous time they took it before the radiation. So they actually scored better. It doesn’t mean the whole brain made them better. They knew how to do the test. So there are tests out there that they call multiple versions and that’s even more expensive.

 HS: That’s interesting because I read about patients saying that their cognition was better after the tumor is taken out.  But it could be that they just remember the questions. 

KC: When you have a patient sitting in front of you and they have just been told, after they thought they were cancer free, that they now have a brain met and we need them to meet a neurosurgeon, a radiation oncologist, a medical oncologist, we’re going to dig in your brain, oh, and by the way can you take this test? You know, like, who am I? What in the world is going on? But when they finish their treatment, they are much calmer, they’ve done some reading about it. They are a week out when they sit down and take the test. I think some patients do better.

 MM:  So what you’re saying is that there’s more than one baseline. That it should perhaps immediately follow the treatment compared to six months a year later…  

KC: Maybe, or maybe not. What we have to do is look at trends. Somebody might need steroids before treatment because they presented with a seizure, and they may not need steroids when we’re done. Steroids affect neurocognition too.  So I think that’s a difficult thing to decide which one to use.  We have to get the data and look at it.

 HS:  I’d like to ask about two experimental treatments. Efaproxiril  and the other is moxifin?  People are looking these up. They are searching on the web and it’s already been shown that there were just subgroup findings that weren’t borne out. 

KC: As it stands right now there are no radiation sensitizers or radiation modifiers that have been proven to be of benefit in whole brain radiotherapy for metastasis.

 HS: Are there any ongoing trials now?  

KC: Lots of them.

 HS:   One last question. What do you see as the way forward for treating brain metastasis for women with breast cancer?  

KC: I think the way forward is science-based.  It’s biology-based.  So, when a patient is diagnosed with this we go back and try to get her breast primary tumor.  If they are going to do neurosurgery, we compare that to the metastasis. We need to document all of the prior Adriamycin and other treatment toxicity, so we get a handle on everything. We need to match HER2 status between primary and metastatic disease. We need to figure out what’s the best treatment for patients is that crosses the blood-brain barrier. We need to figure out what neuro-toxicity really means. As women enter menopause they also have neuro-cognitive changes; so how many of these women are either in a chemical menopause and a natural menopause, and how does that effect their status? We really need to deal with better informing our patients what the toxicity is of the treatment we offer them. And so being able to be part of a neuro-cognitive function study or an imaging study, I think that people need to get involved into clinical trials to answer this. If not for them, then for the next generation.

 MM: Anything you’d like to add on the work the Center of Excellence is doing?  

KC: We are highly intrigued with the work that the COE is doing examining the differences of tumors growing in the CNF environment vs. in the breast or other sites of metastatic disease.  Weve been a firm believer that the micro-environment alters the growth patterns of tumors and alters their phenotype and growth patterns depending on where they grow.

 HS: Some people say there is no real blood-brain barrier by the time the brain mets open up the blood-brain barrier.  

KC: There is evidence to support that in glioma. In glioma, when you find it, the blood-brain barrier is pretty much broken down, but in brain metastasis I don’t think that’s the case. When a neurosurgeon goes in and takes the metastasis out they basically have a shell around them. You can basically lock these things out, like scooping ice cream. Whereas in glioma you have no idea where the brain starts and stops and where the tumor is. So I think they are very different things. Metastases take up contrast, but when they are really small they don’t usually take up a lot of contrast. I think there’s still some blood-brain barrier there.

 MM and HS:  Thank you, Dr. Camphausen!