• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Medical Devices

  • Print
  • Share
  • E-mail

Thermal Aspects of Radio Frequency Exposure: Transcript for January 12

U.S. FOOD AND DRUG ADMINISTRATION

MOBILE MANUFACTURERS FORUM

GSM ASSOCIATION

WORKSHOP ON THERMAL ASPECTS OF RADIOFREQUENCY EXPOSURE

January 12, 2010

Gaithersburg Hilton
620 Perry Parkway
Gaithersburg, Maryland

“This transcript has not been edited or corrected, but appears as received from the commercial transcribing services. Accordingly the Food and Drug Administration makes no representation as to its accuracy”

TABLE OF CONTENTS

Session 5:Localized Heating

Arrenhius Relationship from the Molecule to the Cell to Clinic: Bill Dewey

Entire Group 5A: Effect of Heat on the Nervous System

Vascular Effects: Hari Sharma

Direct Damage, Metabolic Effects: Jack Hoopes

Lessons from Clinical Hyperthermia in Brain Tissue: Nathan McDannold

Discussion on the Brain/CNS

Entire Group 4B: Effects of Heat on the Eye:

Per Soderberg

Relationship Between SAR and Temperature Elevation: Akimasa Hirata (Question and Answer only)

Session 6: Group Discussion

P R O C E E D I N G S (8:30 a.m.)

Agenda Item: Session 5: Localized Heating

DR. MORRISSEY: I would like to introduce Dr. Bill Dewey from the University of San Francisco, emeritus. He will be giving our opening talk today, Arrhenius Relationships: From the Molecule and Cell to the Clinic.

DR. DEWEY: Thank you, Joe. Thanks for inviting me. It has been informative to hear these various things discussed, things I haven't thought about in a long time. I retired in 2004. Still doing a few things every now and then, but anyway.

This is the title of the talk today, the Arrhenius Relationship: From the Molecule and Cell to the Clinic. By the way, that screensaver up there is very illustrative of what is happening in the hyperthermia field, the vicissitudes up and down and so forth, that beautiful road there. I don't know where we are now, if we are down in the valley or up on the hill, but that is how it has been going for many, many years.

Today I am going to summarize some work that we and others have done starting in about 1971 until our recent review in 2009. This summarizes the talk I am giving today in a general sort of way.

If anyone would like copies of any of these slides, you can send me an e-mail. My e-mail is listed in the list of participants. I would be happy to send you an attachment with these slides if you want them.

This guy right here, Arthur Westra, was a postdoc in my laboratory in Colorado in 1969. He had received a fellowship to look at radiation effects on cell cycle progression. He came to my laboratory and said, I think we ought to look at hyperthermia. Hyperthermia? I said, why do you want to do that? We have enough problems with radiation. So he twisted my arm and I agreed to do it, and that is what got me started in this whole thing. These things happen in life. Take advantage of them, good or bad. He was from Holland.

Arthur Westra did these experiments, clonogenic survival as a function of minutes of heat treatment at different temperatures, 46 and a half, 43 and a half minutes of heat treatment. These are asynchronous Chinese hamster ovary cells. Nice survival curves.

What we do with these curves, we take this slope here. The slope is called the D-37. That is the amount of heat which we call the D-37 it would take to reduce the survival to 37 percent of initial value. This is the D-37 value on these different curves here.

We plot those slopes, one over that slope, as a function of absolute temperature. You know absolute temperature, zero is 273 degrees, 20 degrees, 293. So one over the absolute temperature is what we plot. Here is what it is in Celsius, up here. So this would be the inactivation rate for Chinese hamster cells. I will show you in a minute, from this slope you get the activation energy.

Now, pig kidney cells which are resistant, have exactly the same curve, except it has shifted by a factor of ten. Notice that? Ten-fold less. They are tenfold more resistant than the Chinese hamster ovary cells, but they have the same slope. That is the important thing. The same slope, the same activation energy, which I will show you in a minute how to get that. Very significant Arrhenius plots.

Here is a thermodynamic equation, one over D sub naught inactivation rate per second. This is the classic equation you see in the textbooks. This is the absolute temperature times E to the minus delta H over 2T. So this is an entropy term, this is an activation term. This A here equals all this right here. This is effectively a constant. From 43 to 46 degrees, this only varies by two percent.

This other term over here involves the activation energy, which varies by 70 percent. The reason for that is related to exponential temperature, this is linear temperature. It comes out as an equation. So this equation in terms of logs, one over the D sub naught is equal to the log A minus this. So this right here is the slope of that curve. You get the slope, this delta H over two. So that is how we get the activation energy, from those slopes.

Steve Sapareto, when he was in my lab, he followed up on these curves a little bit, different temperatures. The last curves I showed you only started at 43. Now he goes from 46 and a half to 41 and a half. Some of the data points have been left off for clarity, but now you see what happens here. It is a very important phenomenon. At these lower temperatures, 42 and a half and below, these curves start to flatten out. This is what we call chronic thermal tolerance. That was mentioned yesterday several times. So this is a significant observation in many cell lines and in vivo systems as well, the flattening of the curve, called chronic thermal tolerance.

The other thing I want to point out, the pig kidney at 46 degrees would look like this, about the same as the Chinese hamster at 43. That goes along with that Arrhenius plot I showed you where there is a factor of 10 difference on the survival in terms of Arrhenius plots.

The pig kidney cells are very resistant. We are going to talk more about that. The pig kidney cell is more sensitive than the Chinese hamster cell.

Also shown here are cells in G1. When you heat cells in G1, it is very similar to the asynchronous population, 42 degrees for G1 and 42 for the asynchronous population. As I will show you later, that is because the cells that survive are the cells in G1. The cells in S phase in mitosis are killed off; they are gone. So the ones that predominate and control the survival curve are the G1 cells. So whether you take an asynchronous or a G1 you get about the same sort of curve.

From that last set of data we can get the Arrhenius plot again. One over D sub naught is a function of one over absolute temperature or a temperature here. This is very similar to the plot I just showed you down at this point.

This is break point. This break point is seen in many cellular systems, animal systems as well. In this slope, the activation energy would be 140 kilocals, down here 365, this break. The main thing is the break.

I should mention, if you have step-down heating, step-down heating means you might heat at 45 degrees for five or ten minutes, something like that. Before you go to a temperature of lower, it takes this break away. It doesn't really change the curve, it eliminates the break. So step-down heating, you go from a higher temperature to a lower temperature, where the higher temperature is for a very short period of time not giving much killing, but it eliminates this break.

Now we talk about thermal dose. Many people have the idea, or some people have the idea that -- let's talk about the change in temperature, 37 to 42, ten minutes. We multiply these together, it would be five degrees times ten minutes, that is 50 degrees minutes. Does thermal dose make any sense?

You can also do it this way, 37 to 46, five and a half minutes, same thermal dose, but this gives lots and lots and lots and lots of killing. This doesn't give very much. So obviously this for a thermal dose doesn't make much sense, because of that exponential relationship with one over T.

We have to have some other means of thermal dose. This just doesn't make any sense.

Now we come back to using these Arrhenius relationships. We talk about two temperatures. We have temperature one, time one at temperature one, time two at temperature two. These are related by an R factor here, where R is governed by this activation energy here. We can integrate this over a period of time. We had this varying over time, and you can take an integral of this relationship right here, like 43 minus this temperature here. For each interval you can integrate that and add it up. That is what we do in thermal dose. That I will tell you more about in a minute.

Again, above 43 degrees, the activation energy is about 140. That would be an R value of .5. That .5 comes out of this. So that means that if you decrease the temperature by one degree, how much do you have to change the time. That is what this is telling you. So .5 means that if you increase the temperature by one degree, you have to decrease the time to .5. On the other hand, if you decrease the temperature by one degree, you have to increase the time to one over .5 which is two. So sometimes we talk about R of .5 and an R of two. It depends whether you are talking about going up or down in temperature.

For less than 43 degrees, you have an activation energy of about 365, the R value is .25. Then in this situation, if you decrease the temperature by one degree Celsius, you have to increase the duration of heating by a factor of four. That is the basic relationship you have here for thermal dose, doing thermal dose this way. This would be equivalent time of 43 from these different temperatures here. I will show you in a minute how we use this.

Steve Sapareto when he was in my lab said, let's just model this. So he took different profiles of heating here, temperature as a function of time. One situation is A, coming up like this, B like this, C like this. All of these when integrated for these periods of time would give the same thermal dose at equivalent time of 43, which would be like this, 43 for 120 minutes.

All of these different profiles in terms of this equation I just showed you would have 120 minutes at 43, all these different profiles.

This is used in the clinic. First of all, I will show you what happens when you take those different survival curves. I will show you those different survival curves for CHO cells going all the way from 41 and a half to 46 and a half. If you do that type of calculation, converting to equivalent minutes of 43, all of these points fall on the same line.

You start to see some deviation here. This deviation you get from this curve, the thermal dose calculation, this deviation is caused by the development of chronic thermal tolerance. You get the chronic thermal tolerance, and those curves start to level off. You will see it in here. So those calculations doesn't really apply when you get to this chronic tolerance situation, an important point, I believe.

This Arrhenius plot has been seen in many in vivo systems as well. This is from some work done in pig skin probably back in 1947, exposure for epidermal necrosis in pig skin. This is the temperature, one over T here. These points fell nicely on a straight line. The reason they start to curve here is because of the time it took to warm up to the temperature. Other than that, it was a nice straight line, all the way from 45 to about 60 degrees, something like that. So this relationship applies over a temperature range like that.

Mike Borrelli, when he was in my laboratory, wanted to look at this very, very carefully, so he had some cells growing in mylar, where he could change the temperature very, very quickly. He looked at temperatures all the way from 43 and a half to 57 degrees. He had the same straight line curve from 43 and a half to 57. So, a few seconds at 57 is the same as several minutes at 43. It fit the curve nicely. That was in Chinese hamsters.

This is now in pig skin in vivo. Activation energy, 150 kilocalories per mole, the R value would be two. Now we come to the subject of chronic thermal tolerance, I mean acute thermal tolerance, acute. Acute means that we give one heat treatment for 45 degrees in 7.5 minutes. It takes survival down to here.

If you wait a period of time before you give a second heat dose, now these curves change their shape. This is work done by Dennis Leeper after he left my laboratory, by the way. I don't know what funny thing happened here, but nevertheless these curves flatten out. This is the time between the treatments. This would be the curve 12 hours between the two heat treatment, 24 hours, 48. Seventy-two, it comes back down again.

This shows the development of acute thermal tolerance, that means acute heat dose, which is distinct from the chronic thermal tolerance. So development of thermal tolerance at 12 to 24 hours between the two treatments, and then going out to 72 hours, then it decays away, and finally it comes back as a shoulder on this curve. Development and decay of thermal tolerance. This is seen in many, many systems including in vivo systems, the kinetics like this, so this may be very important.

What does this do to Arrhenius plots? This is work done by Marian Law, an Arrhenius type lot, duration of heat treatment in minutes for 50 necrosis in mouse ear, 50 percent necrosis. Normal ears with just one heat dose, has an Arrhenius plot like that. When she waited 24 hours after initial treatment, 43 and a half at 20 minutes, it shifted up. No change in the slope, no change in the activation energy. It shifts the curve. The entropy curve is changing from the thermal tolerance. But the basic relationships, time, temperature, still hold. Again, this is seen in many systems.

So this chronic thermal tolerance is something we can live with, we just have to understand it. It is only shifting the Arrhenius curve, it is not changing the shape. It is a shift of about one degree. Thermal tolerance shifts these curves about one degree to a higher temperature, because they are more heat resistant.

Now let's go back to thermal dose again in the real world, equivalent minutes at 43, EM 43. If you do this calculation for a short interval of delta T times R, T one minus 43, and we can integrate this over a period of time from zero to some final value. That is called the -- you heard it mentioned several times yesterday -- cumulative equivalent minutes at 43 degrees, in this case for T 90, where you get 90 percent of the points in the tumor up to a certain temperature. That is 43 T 90, for 90 percent of the points in the tumor. You can also do it for 50 percent of the points in a tumor.

I have to acknowledge Paul Stauffer for coming up with this term. When I was writing this paper, I was using these long words and all that, and Paul said, let's have a nice little acronym for it. Cumulative equivalent minutes at 43. That is what an engineer can contribute to a biologist. Well, basically I was a physicist before, but here is who I am talking to. If I have to know something about physics, I'll say I am a biologist. On the other hand, if I have to know something about biology, I am a physicist. So Paul was very nice in coming up with this idea of CEM 43. That is Paul Stauffer.

Above 43, we assume that R is .5, below 43 it is .25.

Jim Oleson several years ago when he was at Duke, he is still at Duke, but he is not doing hyperthermia anymore, did this clinical study where they looked at probability of response to soft tissue sarcoma or non-adeno carcinoma to CEM 43 290. This was combined with radiation, by the way. So what is the probability of response, how it relates to the cumulative equivalent of 43, adding up all these different heat doses over there in different parts of the tumor.

He got this sort of relationship. After this study, the Duke group with Jones and Don Thall in animals and humans, did a phase III study. They get to about 20 minutes, something like that, you have a significant increase in the cure rate. I think you even had survival, didn't you? Equivalent minutes, CEM 43 about 20 minutes, a randomized phase III trial, great, human and animal both. So this extended this concept related to the CEM 43 290.

Now let's start looking at the heat sensitivity of different cell types. Some work that Pete Reforst and other people, Bill Bedford and others at Colorado State took these seven different cell lines, the Chinese hamster cells, the mouse cells, clonogenic survival as a function of heating time at 45.5.

This curve right here is the mouse cell, right here human cell. This right here is the pig kidney cell, resistant pig kidney cell. The muntjac is pretty resistant, too, so there is a big difference in these cell lines in terms of the thermal sensitivity. Clonogenic survival is a function of heating time 45 and a half, big differences. Resistant pig kidney cell, muntjac, sensitive mouse cell and other cells in here as well, but let's concentrate between the mouse cell and the pig kidney cell.

These cell lines have been grown at 37 degrees. They are cultured at 37 degrees, and then they were heated at 45 and a half at different times. You can't argue the big difference.

The next slide shows the same cell lines heated at 42 and a half, hour to heat treatment, clonogenic survival. Up here at the top we have the pig kidney cell, here is the muntjac, here is the mouse right here, Chinese hamster. So the mouse cell is quite sensitive, pig kidney cell quite resistant at 42 and a half, just like it was at 45.

What could this be related to? Pete Reforst and those collaborators looked in the literature and said, wait a minute, there seems to be some correlation with body temperature. They looked at other things, chromosome number, all different parameters they looked at. This is the only thing that had some correlation. So order of thermal resistance, pig kidney cell would be number one, mouse cell would be most sensitive.

So in general, as you look at this in relation to body temperature, body temperature at 39.4, mouse cell ear 35.7, 37. So in general there was a correlation with the core temperature of the animal from which the cells were derived, a general correlation.

At this point, I would like to say, it is a real problem in my mind at least, what is the appropriate core temperature? That was discussed some yesterday. As I understand it, if you take the mouse at ambient temperature, the core temperature is something like this. You have the mouse in an ambient temperature of 37 degrees, the core temperature is something like this. So it seems to me that the core temperature seems to depend on the ambient temperature where the animals are maintained. So we have some difference here, what is going to be an appropriate core temperature. The pig kidney cell, pig kidneys are at a higher core temperature than muntjac.

So we have this general correlation, but I want to look at this more precisely by taking those survival curves and using the equivalent minutes at 43 calculation for delta T, the increase above a certain core temperature or body temperature.

When this is done, this is from the 45.5 data I showed you. Heating time for a delta T of five minutes for the cell lines here, the muntjac and the mouse. This is the delta T corresponding to 45.5. For the mouse you take 35.7, for the core temperature the delta T is 9.8. If you assume the core temperature to be 37, this would be a delta T of 8.5. It makes a big difference. If you assume the core temperature of 36, this is what that survival curve would look like, equivalent for a heating time for a delta T of five, using this basic equation here. A T of five would be the T of 45.5 times R, where R is two, then delta T minus five.

The pig kidney was like this, the muntjac like this. The pig kidney has a delta T of 6.1, muntjac 7.0. So these curves look pretty good when you use whatever the appropriate core temperature is. If you assume the appropriate core temperature for the mouse, it is at 37, that looks pretty good in relation to the other two cell lines, the resistant cell lines, sensitive cell lines all about the same in terms of the elevation of the core temperature.

But which is the appropriate core temperature? If you took the T 35, it doesn't make much sense.

This is from the T 45.5 data I showed you. Now let's look at the same thing. This is from the 42.5 data. That is where we had that chronic thermal tolerance developing for the pig kidney and muntjac. Now if you take the mouse, the 35.7 is way up like that. Mouse at 37 has a core temperature like this. Not very good correlation, but it is certainly better for a core temperature of 37 than a core temperature of 35.7.

We assume here that this is an R value of four because it should be below the break point, supposedly, below the break point in the Arrhenius curve. That would be the assumption. We are below the break point, and if the R value of four were appropriate, this is the calculation you would have.

But let's say it should be two. Now let's calculate that for an R value using R of two, because we have T minus five. The mouse at 35.7 doesn't look very good, but now all of these curves are about the same in here, the pig kidney and the muntjac and the mouse. You can't go any further than this with the pig kidney and the muntjac because of that chronic thermal tolerance that developed. But now it looks like this would be a more appropriate dose calculation using the value of two, even though it is a delta T of -- for muntjac it would be four, pig kidney 3.1, mouse 5.5. But we wouldn't expect this to be true.

The bottom line that I want to make out of this, I think we need a lot of data in this area, where we look at different situations for core temperatures in animals, in different animal systems, different end points. Then maybe we can make some extrapolation to humans. But you can see the dependence on the core temperature that you assume is appropriate. That I think is the dilemma. We need data on that in the real world, in the animal world, various animal species, maybe different organ systems, and try to determine what is appropriate.

I don't know what is appropriate now. All I can show you is the data. We should have made this calculation years ago, but we weren't smart enough to do it then. You know how it is; you have to wait until all of a sudden you get the inspiration to do it. So this is a challenge we have to have, to know what is the appropriate core temperature.

The other challenge I think we have, these cell lines were derived from these animals, and then they were cultured a long time at 37 degrees. So there has to be some memory in there. What is that? What happened? We probably haven't changed the DNA supposedly, except maybe epigenetic changes in the DNA, that is methylation of the cytosine that can have effects. But there is some memory in there if this makes any sense at all, because these cell lines have not been maintained at these different core temperatures. They are maintained at 37. There has to be a memory of some sort.

To my way of looking at it, we need a lot of real solid research in this area to understand what is going on. It has to be something, but I certainly don't know what is going on.

This is 42 data, 42.5 at core didn't fit. This is work done by Mort Miller, who accumulated some data from the literature. This is one of the references I have. The data that he obtained on the percent embryological defects, rats, rats and mice. I don't think you have the guinea pig in there, do you, Mort?

He made these calculations for a thermal dose four degrees above the core temperature, for four degrees above the core temperature, percent embryological defects from the maximum temperatures they had, 4.5 above the core temperatures.

The series of data here from these different investigators look like this. If you want to get 20 percent embryological defects at seven or eight degrees, there seems to be a dose response here. I don't know how to explain this, maybe you do, Mort, why does this investigator have these data that look much different than this?

If you started here for a zero dose of about ten percent abnormality, here it started at zero. What is the appropriate dose response? What do you think, Mort? What is the difference? What is going on? He didn't show you this yesterday, so we are going to talk about it today.

DR. MILLER: The short curve on the left side is a very steep one. It is from the work of Carol Kimmel. That stands apart from the other data, and that was with rats.

DR. DEWEY: You don't have any idea why?

MR. MILLER: No, I don't, but I plotted it.

DR. DEWEY: Another reason for more research.

DR. MILLER: You can't just ignore that. Just to reiterate, the problem this entire literature has is that it is working with very small numbers of subjects. Kimmel's work on the left is 13 litters per regimen. You can't get much out of 13 litters unless you do very high doses. The other is of comparable sizes, but they were more comparable slopes, but there were definitely two lines there.

DR. DEWEY: My plea is, let's get more research along here, so we can understand what is appropriate. We want to extrapolate to humans. I don't know anything about it. This is not my area of expertise, but I think we need more research.

Now I get into some of the mechanisms of how cells are killed. Going back to our original experiments that Arthur Westra did, clonogenic survival as a function of where the cells were in the cycle when heated at 45.5 for 15 minutes, ten minutes, five minutes.

What you see here, in mitosis they are very sensitive, resistant in G1, sensitive in S phase, at all these different temperatures. So the bottom line is, they are sensitive in mitosis. What we found here is, these mitotic cells did not complete division. They went all the way through the cycle, and they appeared in the next cycle as tetraploid cells. Something happened to the mitotic apparatus, the mitotic cells.

We have done several experiments to show that heat affects microtubules, cause the microtubules to be dissembled. In mitosis the critical target is probably proteins associated with the microtubules.

In G1, I'll show you data in a minute that suggests this is related to damage to the centrosome. S phase, Rosemary Wong and others in my laboratory showed this is related to effects in DNA synthesis. It inhibits the replication in the core migration in S phase. The next slide will show what happened in S phase when you look at these cells. They come through mitosis. They don't die immediately. They go all the way through the next mitosis, then you see chromosome aberrations like this.

These are Chinese hamster cells. Different types of deletions, exchanges, and right here, the secondary construction of the X chromosome. When the cells are in the new part of the cell cycle, you get some deletions here. This is the nuclear organizer region, has a lot of repetitive DNA that codes for ribosomal RNA.

Quantifying these sorts of data, clonogenic survival is a function of aberrations per cell. There is a lot of data that we have for Xray. Radiation in mitosis in G1 are lying like this. When you heat cells in S phase, it is very similar. The aberrations per cell are about the same after heating in S phase and radiation in any parts of the cell cycle. So again, it looks like the S phase cells are very sensitive. They are killed because of chromosomal aberrations related to effects in DNA synthesis.

I don't have time to go into the experiments, but Rosemary is sitting back there, she can tell you the details about it.

These cells in M, mitosis G1, they don't correlate very well. They don't have aberrations. If you take the secondary constriction, exclude that data, these points move clear over. So these cells heated in mitosis in G1 obviously are not killed because of chromosomal aberrations. Cells in S phase are.

You would expect some of these cells in S phase that have these chromosomal aberrations might have some long-lasting effects, not from point mutation, but from rearrangements of the DNA, if they survive. They are very difficult experiments to do, because if you have any contaminating G1 cells, they survive. So you have to get a very highly asynchronous population at S phase. Then you may find some long-lasting effects, which would be genetic changes related to inversions or something like that, not point mutations. You would expect that.

I don't know anybody who has done the experiments. They are very difficult to do. You are going to have to have a population 99.9 percent S phase.

What happens in G1? These are experiments done by Mike Borrelli, who said, let's look at what really happened. Put cells in G1 and then look at them after ten minutes of 45.5. You see this blebbing of membranes here.

He followed these individual cells and then looked at them in G1, saw this going on, and then followed them all the way through the cycle, and found out if they form colonies or not. One cell by cell to find out what happened.

What he found was that the cells that looked like one, two or three, they would form a colony. Cells that looked like four or five, lots of blebs, they wouldn't form a colony. So very clearly on a one to one basis, he could relate the killing of cells in G1 to these blebs.

It has to be membrane damage, right? Well, unfortunately we couldn't show that, because Mike Borrelli, Charles Vidair and others looked at ion concentrations inside the cell with microelectrodes. The ion gradings were maintained just fine, sodium, potassium, even transport of calcium, sugar, glucose, all these things, no evidence of any membrane damage at all, but we had this sort of correlation.

In other studies these people did, they found cortical microfilaments next to plasma membranes were detached, so they thought this was related to detachment of microfilaments in the cell, which would let the membrane bleb out. Probably again effects on protein, cortical microfilaments, something like that.

In the next slide, they saw something else. This was done by Charles Vidair. They looked one cell at a time, following all the way through. If you have a cell that has never been heated, you have a centrosome like this on two, and your cell divides. It is the centrosome that divides.

If you heat a cell however, you have a lot of cells that show this phenomenon here, something that happened to the centrosome dividing in different parts. It goes into mitosis, you have an aberrant division, and these cells are always non-clonogenic. On a one to one basis, he was able to show that the cell killing was related to damage to the centrosome, leading to aberrant divisions that would not let the cell form a colony.

What he wanted to do was pursue this now in detail, look at the effects on the centrosomal protein, relate that to blebbing, cell to cell basis. The study section didn't see fit to fund him, so he gave up. He is now doing great work for the California EPA. It is much less stress. It is too bad he didn't continue, because he wanted to follow this all the way through the blebbing, to the -- structures, and do it on a very precise mechanism. But according to the study section, that is descriptive biology, you are supposed to be doing molecular signals. He was going to look at proteins and collaborate with some of the people at UCSF looking at centrosomal proteins.

Let's get into some of the factors that affect heat sensitivity. People have looked at specific proteins like enzymes. If you replace one amino acid by another for denaturation of a specific protein, the Arrhenius curve can be shifted by three degrees, changing just one amino acid in the protein. Three amino acids, nine degrees. So this again goes along with the idea of, our target is probably a protein, directing protein structures, and possibly in these cells that have differences, there are differences here; how do we know?

I should say, the evidence for the protein being involved. In mitosis we see effect in the spindle. In S phase or any part of the cell cycle, we find there is a lot of non-histone protein that is aggregating with DNA. Rosemary did some of those experiments. They also have thermal tolerance for that phenomenon. So proteins seem to be very important, either in S phase, G1 or wherever it is, in G1 on the centrosome.

Acute thermal tolerance, I showed you that, second heat treatment at two hours after initial heat treatment decreases sensitivity. It shifts the Arrhenius curve. It is also seen for chronic thermal tolerance that develops a few hours after low temperatures, and this decays in two or three days.

Step-down heating, a relatively low temperature immediately after higher temperatures, that synthesizes cell by eliminating the break in the Arrhenius curve. Heat sensitivity is different for different tissues in the same animal, some of the factors that affect heat sensitivity.

Going on with that, for different mammals the heat sensitivity appears to relate more closely with elevation above the baseline core temperature of mammals instead of the absolute hyperthermic temperature. Lower pH, lower intracellular pH, we showed that, increases the sensitivity. What does that do? It just shifts the Arrhenius curve with the lower temperature, one degree or so. It is a shift in the Arrhenius curve.

Heating cells in the presence of glycerol. Glycerol stabilizes proteins, decreases sensitivity, shifts the Arrhenius curve to higher temperatures, there is no change in slope. I showed you heat sensitivity varies during the cell cycle, S phase in mitosis being most sensitive and G1 most resistant.

Summary. To attain the same lethal effect, that is an isoeffect, you have to decrease the temperature about one degree and increase the duration of heating about twofold, below 43, activation energy about 140 kilocalories per mole above the break point. This relationship holds between 43 and 60 degrees or higher.

Recently people in various parts of the world have been looking at thermal ablation, things like that, very high temperatures, 60 degrees or so. That again seems to fit the Arrhenius relationship quite well.

Protein denaturation, aggregation seem to be critical, activation energies. We can look at the non-histone protein associated with DNA; you can see this effect.

For different biological systems, different cell types, absolute heat sensitivity varies depending on the entropy term. That makes the curve go up and down, but delta S is fairly constant. That is the slope of the Arrhenius curve.

People who contributed to this study, here are a few of those. Arthur Westra, Steve Sapareto, Hatsumi Nagasawa, the chromosome person, Dennis Leeper, Mike Freeman, pH, Jim Oleson, the clinical work, Dewhirst, clinical work, Diederich, Stauffer, clinical, 743, Raaphorst, Mort Miller, Mike Borrelli, Rosemary Wong, she has done a lot of this fork migration studies with Monty Kapp. I forgot to put Monty Kapp's name down there. That is the rate at which the fork migrates in cell synthesizing DNA; hyperthermia really slows that way down, the blockage of the fork. Single stranded DNA results and probably leaves the chromosome aberration. Ron Coss, microtubules, Joe Dynlacht did the calcium work, Charles Vidair. Then we have these two people here. Chelsea Landon worked with Mark Dewhirst, student. Without her help I couldn't give his lecture. She made the PowerPoint slides, most of them. Then I had to get my son involved. He is an engineer now, so he helped me with some of these PowerPoints.

I really appreciate all the help you guys gave me. As I said, if you want any of these slides, my e-mail is listed in the list of contributors. I would be happy to send you anything you want.

Thanks a lot.

DR. MORRISSEY: Thank you, Bill. That looks like a lifetime of work there. That is quite a bit of information.

We have time to take a couple of questions, if anybody wants to ask one.

DR. ZISKIN: The chronic thermal tolerance may be a very important factor for a biological species to survive heat. I was wondering, what is your explanation for what is happening there to produce it.

DR. DEWEY: Well, that is a very good question. I can tell you what we noticed - effects on non-histone proteins is tolerant for that. We get less aggregation of the protein. Almost any chromosomal aberration, there is tolerance for that.

I don't really know what the mechanism would be for the chronic thermal tolerance. Something has happened to that cell. It is a very good point. I don't know. It would be nice to have some more research. But you can't get funded, at least I couldn't anymore, for that. I had to look at gene knockouts. So I did computerized video time lapse so you could follow cells after radiation. I think we need more work on that.

DR. VEYRET: A very interesting talk. I wonder if you could tell us a bit more about the fact that the slope is more or less constant in terms of mechanisms, protein maturation and so on. Have you considered other phenomena such as membrane activation, phase transition and so on?

DR. DEWEY: I would say it all points toward effect on protein. If you look at activation energy for different proteins, it is pretty close to these values, 140 kilocals, something like that, different proteins. That is what we find in our cells; we affect the mitotic spindle. We did effects on spindle itself, microtubules, looks like it is affecting the protein.

So if you have effects on different proteins, say mitotic spindle for cells in mitosis, centrosomes, cells in G1 and non-histone proteins associated with the replication complex for S phase, it is still protein in all cases. Protein should have an activation energy of 140, 160 calories per mole.

That is what we hypothesized in 1971 in our first paper, and ever since then, all the data seems to support that. Even though we had blebbing of membranes, we couldn't see any effect from the membrane. The ion gradients were still the same, no change in ion gradients. Glucose transport was the same, amino acid transport. So it didn't look like there was actual membrane damage itself. But the filaments below it, cortical microfilaments of the protein is what caused it.

DR. WONG: To answer the first question, in terms of the thermal tolerance development, when you do the heat shock after heat treatment, you are inducing heat shock proteins. Depending on the dose, the rate of the induction of the heat shock proteins, you now induce a whole number of signaling changing signaling pathways.

Part of the thermal tolerance probably involves the heat shock protein and its ability now to stabilize proteins at a subsequent treatment. That is why you don't get the increased damage like in the step-down heating or step-up heating that Bill talked about.

So protein's denaturation is very critical in terms of the thermal damage one sees. That can be modulated by heat treatment, low doses and whatnot.

I think understanding the difference between your core temperature perhaps in the pig kidney, they already have a much more stable protein, or their heat shock protein induction is much more induced because of the higher temperature or core temperature, so that when you do heat shock protein induction is much more induced because of the higher temperature or core temperature, so that when you do heat, you are not getting the destabilization and the protein denaturation one sees at a lower temperature animal.

DR. DEWEY: Thanks a lot, Rosemary. I forgot to mention heat shock proteins. There is a lot of data on heat shock, HSP-7 in particular, associated with tolerance. Laurie Lee has done some real nice work on that. She is in our laboratory in San Francisco. There is a lot of work related to heat shock protein. Thank you, Rosemary.

DR. DEWHIRST: I was just going to mention more evidence for protein being the target. That is the work of Jim Lapaque. Jim did extensive work using differential scanning calorimetry to look at the heat of inactivation. It is very clear that protein is the target. So if you are interested, you can look up some of his papers.

DR. DEWEY: Thank you, Mark. Very good.

DR. MORRISSEY: Thank you, Bill. I had one real quick question, and then we have to get on with the schedule here. I am bringing it more in terms of the cell phones and the very low level. Your Arrhenius plots suggest that you can increase four minutes every time you drop a temperature. When does that drop off? Does that go all the way down to 37?

DR. DEWEY: I should have mentioned that as well. Mark Dewhirst and others have data that look at liposome delivery and things like that and physiological changes, like 40 to 41 degrees. That seems to fit the Arrhenius relationship too, down at 40, 41 degrees.

DR. MORRISSEY: But it is 38 degrees or so, like a degree of an increase, is there any evidence to suggest that that extrapolates right back down?

DR. REPASKY: I think there is a lot of work that needs to be done at those lower temperatures with regard to the heat target. Our work would certainly suggest a change in membrane lipid organization as well, but at lower temperatures.

DR. MORRISSEY: But not necessarily lethality.

DR. REPASKY: No.

DR. DEWEY: I expect you have got a very good point.

Agenda Item: Entire Group 5A: Effect of Heat on the Nervous System

DR. MORRISSEY: Thank you, sir, thank you, Bill. We will now go into the next session and Hari Sharma, effects on the nervous system. Our first speaker will be Hari Sharma, followed by Jack Hoopes and Nathan McDannold.

DR. SHARMA: First of all, this is a great honor to be here and invited by the organizers.

After this great talk of altered thermal sensitivity, I can only add that brain temperature or body temperature is very sensitive, and it can be altered by a wide variety of factors, both exogenous and endogenous. I will share very briefly my observation with you.

This is a work in collaboration with Wright Patterson Air Force Base, who introduced nanoparticles, and CTR. They are introducing drugs of abuse, how temperature can change, and NIDA.

We believe the very old philosophy that when we would like to tackle the problem, I think we should go to the root cause rather than treating symptoms. We believe that our central nervous system is very important, because not only neurons we have, we have also glial cells and endothelial cells. So far, people are talking about neuroprotection, but they do not put a lot of attention on endothelial cells as the catchment of the blood-brain barrier. In normal situations, the blood-brain barrier is in endothelial cells, and they are leaky in almost all disease conditions, and they could be affected by both drugs and temperature.

This shows that all elements, neurons, glial cells and also we have a blood barrier that is not in micro vessels. Here the point is that the blood-brain barrier is very tight and the other barriers between CSF and blood could be milder, but they are still controlling the micro environment of our central nervous system.

Here I would like to point out another thing, what will happen if we put localized heating to the nervous tissue. I think we are also going to alter not only blood-brain barrier, but also brain blood barrier, transport of various materials from brain to blood, and they can also have a general systemic effect. But this aspect has not been discussed before either in terms of pharmacology or in terms of brain injury.

It is very difficult as we have discussed here, to disassociate between any effects either they are producing directly, for example, temperature, or they are mixing with stress. Still we are not able to disassociate these things, so we believe that stress plays an important role.

Just to show here that our CSF barrier is also very important in regulating brain function.

Brightman and Reese at NIH showed the first time that the blood-brain barrier, to lengthen them is -- at a tight junction. Therefore, lots of studies have carried out what will happen to these tight junctions or what will happen to the membrane.

Why hyperthermia? This is very important, because everybody here -- it is great to hear our mentors talk the last two days, with Gordon, Dr. Dewhirst. What we feel that temperature is very important and increase of .5 degrees Celsius is very crucial.

At this moment, I am not going to debate that global warming as increased or we have so much cold weather here, but I believe that in summer, still temperatures are very high, and that results to more intensity and frequency of summer heat waves in various parts of the world.

We tried to put some basic information by Elsevier in 1998, where we tried to have some data on hyperthermia and brain function. Nowadays, people get startled after hyperthermia effects, when in France, more than 10,000 cases of death were reported. Previously it was believed that hyperthermia is a problem for only developing countries, but now the situation is changing, and so many deaths are reported also in the United States.

Still we do not know that weather, this hyperthermia or temperature, they affect equally to various populations in a country or in our world. Likewise, military personnel. I am not going to discuss because of time the Gulf War syndrome. Exercise and heat involvement are very sensitive to temperature changes.

We tried to study for the first time details of central nervous system changes. Now we are concentrating on the modulating factors, both exogenous and endogenous, because no population, either military or civilian -- they are having populations from diabetes, hypertension, so many diseases and many other physiological modulation by drugs.

So what is happening about that? We tried to also cover some of the areas in this volume, and there are several people here that contributed to this and we are grateful to them. Therefore, we use a model in animal laboratory situations where the rats are exposed to 38 degrees Celsius for four hours, and you can see the rise in temperature is quite linear in our hands.

Then we believe that hyperthermia is a problem of brain function, so what we measure is blood-brain barrier with the leakage and brain edema, and as you can see, there is a very tight correlation between leakage of blood-brain barrier and brain edema in relation to temperature.

The leakage of the dye as a protein complex is not superficial; it is deep. I am just showing you some examples. Most of you are familiar with the brain part, hippocampus, hypothalamus, cerebellum. They are deeply stained. It means that it is not a joke, hyperthermia when temperature reaches in conscious animals about 41.2 or 3 degrees Celsius, it opens the barrier.

The point is, what will happen? Why do we have a blood-brain barrier, and what is its function? The point is, can we say that when our barrier is broken down, we are going to have cognitive changes or some other changes? The answer is yes. Because of lack of time, I only want to show that at the time of increased permeability there are changes in behavioral function. Also heat treatment of drugs can reduce it.

The other important point is the EEG changes. We can see here that in this model after three hours, the brain waves change, and it is still not normal even two hours after discontinuation of hyperthermic insult.

We are aware that regional blood flow has a great role in hyperthermia and its modification. So we tried to analyze changes in regional blood-brain barrier with changes in regional cerebral blood flow. Although a tendency of lowering of blood flow in almost all brain regions after four hours. It is time relevant. Nothing happens one hour and two hours; four hours. But they are not correlated with increased blood-brain barrier permeability. So ischemia alone is not the reason here in this case.

Then we just tried to understand actual changes on the blood-brain barrier. What we can see here is that, as you compare the other picture, the endothelial cell membrane is infiltrated in hyperthermia, even though the tight junction. So this is a different aspect of temperature induced endothelial cell damage.

Regarding normal morphological changes, the important point I would like to note is that the most vulnerable regions in the hippocampus is CA-3 and CA-4 area, not CA-1 and CA-2 that has been discussed a lot in ischemia, where vessels are occluded. So every physiological phenomenon has different aspects.

The -- axis is also degenerated. This is just an example. Regarding the specific cell damage, I don't believe that when we have hyperthermia or temperature, all neurons are going to die in one area. This is an example. In parietal cerebral cortex one neuron has a very dark and dense appearance, and the nucleus is very eccentric. But in the same area other neurons could be normal. We still do not know, and we have to discuss, that where there are specific temperature sensitive neurons and how they behave under hyperthermic conditions. This is another example of cell damage and leakage of blood-brain barrier. Even synaptic damage looks like very similar to excitotoxic damage.

Obviously I am not going to discuss in this model, you can see heat shock protein up regulation. But now the question is, we cannot find clinical cases that could be simulated in laboratory situations. So therefore we do not have any drugs developed to treat our patients, no matter what the case is.

So therefore, what we did, we apply same magnitude of hyperthermia, 38 degrees Celsius, four hours, in animals that were made hypertensive or they were made diabetic by drugs in laboratory, and you can see here, the same amount of heat dose produces larger damage in hypertensive animals and also in diabetic rats. So it means that the brain could be more sensitive to heat if our physiological or pathological conditions are altered.

We heard discussion yesterday whether the young and old animals could be different or very similar. I am not going to any controversy, but I can only tell you that using in vivo model, the same amount of hyperthermia induced to young animals and old animals, for example, Y means young and O means old, but not geriatric. You can see the differences are very clear in neuronal cell damage, and these are showing other things. So there are certain differences. The hyperthermic effect can be modulated by disease and age.

Our problem to relate clinical cases versus laboratory investigation lies mainly to the fact that we cannot find two identical clinical cases.

We have heard here that many thermal doses do not produce any change in the brain. I fully admit that. But what we don't know that at what stage we are looking, for example, if we put animals to stress, maybe it is thermal stress, the first stage is alarm reaction, where you can see the symptoms. But if the magnitude of the stress is continued, there is a large stage where you can't see the change, but it doesn't mean that the stress effect is not there. The only thing we can do is continue to a certain stage where the absorption stage will appear, and that leads to damage.

These are the consequences. I am not going in details, but this is very important. If we don't see symptoms, it doesn't mean that it is not going to produce any symptoms.

People are not similar in living in various parts of the world. Think about our military persons are exposed to the area, for example, Middle East, where so much sands are there, and they are breathing lots of silica dust. They can reach to their central nervous system, and many of the nanoparticles from gunpowder or something like that.

So we just added this question also, because nowadays people are using nanoparticles for drug delivery, but together with USAFD, we are doing some experiments to find out a neurotoxicity approach, at the National Center of Toxicology.

What we found, obviously I must confess that we cannot reproduce human situation, but these are laboratory investigations. We have treated silver nanoparticles, 50 to 60 nanometers, small doses, for one week, and we can see that there is cell produced leakage in the brain. I can show you, this is albumin immunostaining, leakage in the blood-brain barrier, but they are different in copper and silver, although they are very similar in size. So it means that probably the inherent metal property can have a specific effect rather than the size of the nanoparticle. These are also telling you that glial cells are intact.

But this is not the point I am showing you. I am showing you that when these rats are inoculated with these nanoparticles, subjected to same degree of hyperthermia here in the lower case, you can see much more severe damage can be seen as compared to the -- group. So it means we are always modifying the brain temperature. We are always modifying the body functions if we change the situation.

I am very briefly concluding. This is an example of more leakage in blood-brain barrier by silver nanoparticle treatment. This is a complicated slide, but I can only tell you that not only that, the nanoparticles exaggerate the brain damages; even the drug effects are altered, because if the normal animals were treated with this antioxidant, they are rather neuroprotective. But when they are treated to nanoparticles in affected animals, they are not able to do that.

This is just an example here, telling GFAP and normal animals treated with normal drugs cannot do that. This is the example of nickel staining; the same thing can occur.

So what we believe is that our central nervous system is very sensitive to temperature changes. Its first reaction is to open the barrier because it is quite universal. Then the consequences will follow in various ways that leads to cell death.

I believe I will only stop here. This is also a very old saying. Of course, we are responsible for the things we have done, but we have to think that we are fully responsible if we have not done anything proper at the right time.

Thank you very much for your kind attention.

DR. MORRISSEY: Thank you, Hari. What I would like to do is go through the next couple of talks, if that is okay, and then we will have a large discussion section. So the next speaker will be Jack Hoopes from Dartmouth College.

DR. HOOPES: Thank you very much for inviting me. I have been doing hyperthermia even in my graduate work -- sorry. Thank you for inviting me to talk about this. Joe and I have had a lot of discussions about RF. I find it is pretty challenging, actually.

I am going to go through some work that we have done at Dartmouth and with Mark even before that a little bit, where we were looking at higher temperatures. In some ways it was so much easier, because we got effects, some pathology based. We could see things and quantify things and correlate them to temperatures. Even though we didn't think it was easy at the time, it seems like now this challenge of RF in these lower doses is greater. Luckily, I think we have better biological assays and ways to look at say microarrays and other fishing expeditions that I don't believe we have done. I think there is some weakness here. I am going to try from a pathology standpoint point that out a little bit, of where at least I feel changes might be made.

This has been shown a couple of times, but I just wanted to reiterate that obviously, even at low temperatures for short periods, things happen in cells. I think Hari Sharma just showed that, and others. We worry about these signal transduction pathways. We all talk about this. We have not identified those pathways, so until we do that and learn how to quantify it, we are not quite there yet.

As Mark Dewhirst said earlier and others, it does appear protein denaturation is the principal cell morbidity and mortality certainly at moderate temperatures, and maybe at these low temperatures. Once again, hard to identify morphologic change at these levels of one or two degrees above basic temperature.

Heat shock proteins, for a long time we understood them in the terms of thermal tolerance. Now it looks like they have a bigger role. So that will be interesting to see where that goes.

I do think that just physically, the evidence suggests that virtually all cellular responses probably result from some sort of translational or vibrational or rotational activity. That is going down to a pretty small level, but I do think we have to try to understand what is the pathogenesis, or is there a pathogenesis if these low doses are going to cause change.

I think that hyperthermia, like a lot of injury cascades, is complicated. I think at RF doses it will continue to be complicated, but first we have to identify what are the factors at these low doses. I think we are probably not there yet. But I think that likely the cellular targets are these molecular transitions and these lipid bilayers seem to be one of the targets. I think DNA and RNA, unlikely at these low doses, the denaturation temperatures with these molecules seem to be pretty high, although maybe there is some unwinding of DNA at these low doses. So once again we are back to the protein effect, I think.

Here is a few things that I think we have learned as a starting factor. We learned that rodent tissues are slightly more sensitive to heat than humans, we think. The extrapolation of low to high or high to low in various species can be complicated, but we were forced to do that because that is what we have to work with. I think these complications of blood flow differences, tissue geometry and all that, we have shown in this meeting, that is what is going to be important. We still don't quite get this pathophysiology as well as we need to, to understand it and to modify it or put regulations on it.

I am an animal model person. These are a few things I think might be important in developing of animal models. You first of all have to have a size and spatial appropriateness to be able to give some sort of insult and to assess the insult. I think mice, clearly the model we use the most, can be challenging in this area, but I think it is where we are.

You have to understand, obviously the dose that you are going to give and the dose rate, which we have always talked about that in radiation. It seems like maybe that is going to be a factor, although I don't think we quite understand it the way we do in radiation.

The ability to assess temperature over time as non-invasively as possible. In the old days or maybe in the current times, if you are treating cancer in animal models, you have some sort of thermometry mechanism to assess how things are going to get a thermal history. That is hard to do in mice, and maybe MRI can do that. I think we need to try to work on something like that. As Mark has shown and others, if we don't measure those temperatures and we don't know those thermal histories, you try to co-register that with the path change, and it gets a little complicated, and not very meaningful. So we need to understand that.

We need to develop these pathophysiology end points at these low exposures, especially of RF. Then what are the tissue end points, and I will talk about that in just a second.

I just wanted to talk about one model that we used at Dartmouth years ago when we were treating brain tumor patients. We were trying to understand what were the end points, what were the hyperthermia-only end points of the brain.

We had a model. This is a radiation hyperthermia model, but this was the large animal model. This was a microwave antenna at 915 megahertz, then these were Teflon catheters that used lextron probes, so we continuously monitored the heating throughout the experiment in most of these patients, which had a similar setup to this, but even more thermistors or thermocouples, continuously monitoring throughout the one-hour treatment. So I am going to try to show you how I think that was effective.

We put these catheters in, and then we did continuous pullbacks of these to measure the three-dimensional map. As you might guess, the hottest place was where the antenna was. As one pulled back, the temperature went from normal 37 degrees or 38 degrees to a high right at the antenna of maybe 60 degrees. Then our cutoff point was that we wanted to achieve at about 1.5 centimeters or 7.5 millimeters, we wanted to achieve a certain temperature, and that was 41.5 degrees. So we were able to give a dose that we understood.

This is what happened. When we did that, we created a fairly large area of necrosis. That is maybe not so important, you might expect that at a high heat dose, and we were 60 degrees in the middle. What was important is where the tissue survived. At the inner border of this contrast enhanced ring of the CT, we were able to measure that. Then we could measure where the blood-brain barrier breaks down, which is the outer layer of that contrast enhanced ring, and get an exact thermal dose at a certain time period for what those changes were. Our goal was 43 degrees for 60 minutes, 1.5 centimeters from the antenna, which was situated in the center there.

This is a little higher magnification. This is what the brain looks like on a real path slide. It looked like this. It is not exactly spherical as you might think, because gray matter as we know compared to white matter has more blood flow and more thermal washout, so it was less sensitive than the white matter to the heat.

This is what always amazed me. This is about 1.5 millimeters from here to here. This was right at the margin of the killing. This is normal brain here, which is very little effect, mostly normal effect. Here is the transition zone right here, and this is completely necrotic here. This is probably a range of less than a millimeter, this transition zone.

So obviously the tissue was very specific. If you hit a certain thermal dose you killed it, and if you hit another thermal dose that was very close, you didn't. I think this is the thing we have to understand about hyperthermia.

We made these iso dose curves, like this. Here again is the brain section, the whole brain section. We made these iso dose curves and put these numbers on these different areas of heating so that we could show in each one of these zones what happened to the individual cells.

The brain is very simple tissue overall; there are only about four cell types, four or five cell types. We saw that at the highest doses here -- of course in the necrotic area there were no cells obviously, it was just coagulated protein. So astrocytes were elevated at the highest levels and came down, where neurons and oligodendracites, those are the cells that make myelin, they went up at the lower temperature. So there is a little bit of difference. Astrocytes are very reactive.

Maybe the most important information was here. We needed it at three different times, 30, 40 and 60 minutes. We assessed by CT MR and by histology. We showed that the dose for necrosis seven days after treatment on a histology basis was about 42.2 degrees, about 45.2 at the 30 minutes, and then at 60 minutes you can see that it can down to about 43.2. So a good example of what we had always thought, that has been challenging to demonstrate, that 43 degrees for 60 minutes really does kill -- according to this study, it kills most of the cells.

Then the blood-brain barrier you can't assess with histology because that is a physiologic change. We use that contrast enhanced ring. So the temperature based on MR and CT was pretty close. At 30 minutes of heating it was about 42 and didn't really change much. So that temperature for blood-brain barrier breakdown was about 42 degrees.

As I said, when we are thinking about the brain for this RF, there are only a few cells. You have neurons, lowly or non-proliferating cell, oligodendrocytes, they make the myelin in the white matter and then astrocytes, very reactive glial cells, an inflammatory cell, endothelial cells as Dr. Sharma just showed, a very important cell, lines the blood vessels. It is what is a capillary. A capillary is only an endothelial cell, no other cell present in the capillary. Then the ependymal cell, which lines the cavities where the CSF is, the ventricles.

I used to think about this in radiation biology terms. When you damage a tissue, there is what we used to call a target cell. It is the cell with the weakest link. So for every insult, something is going to give up first. Generally every cell doesn't have the same tolerance. So I think it is worth looking at this, trying to figure out with microarray technology or other kinds of technology what is the cell that we are the most worried about.

Dr. Sharma might say it is endothelial cell; I think he did. So I think maybe that is an issue. So once we understand that, maybe we will have a better idea of, if that is the most sensitive cell, maybe that is where we should look. If those cells go, the rest of the cells are going too, because there is no blood supply.

Here are a few other things. I think what I am going to do in the rest of the few minutes here is try to look at what might be possible, how can we use animal models and come up with some parameter to assess these low heating levels that we previously thought -- I think during my career, if somebody said what is going to happen at 40 degrees or 39 degrees, I would say nothing. Well, nothing is probably wrong. Here are a few ways that we can look at that.

What I think is unlikely to happen in a pathology is irreversible changes like necrosis, gliosis, demyelination, hemorrhage, DNA strand breaks. Those seem unlikely to occur, so I'm sure we are not going to find those. What is possible is some irreversible changes like apoptosis, which is a biological cascade event that we can assess after the fact. I think that probably is an area which could be looked at more closely.

Cognitive changes as Mark has shown is obviously an area that we are starting to get into and could be extremely sensitive, because it is so dependent on neurotransmitters and other types of changes. Then maybe some sublethal DNA alteration that could be assessed by microarray technology, laser capture dissection or other methods of looking at DNA.

Then reversible changes are even more difficult. That is a really difficult thing to do, because you have to get it before it changes. It doesn't mean that even if it changes back that it hasn't had a long term effect. Things like minor edema or some kind of edema as Hari Sharma has shown, cognitive changes, blood flow changes, alterations in neurotransmitters, cytokines.

There are a few papers in the literature that are starting to look at these kinds of changes. They are just hard to assess. I look back at the easy things we did. We just heated tissue and sacrificed the animal and took the tissue and put it on the microscope. It was pretty straightforward. It doesn't seem that straightforward now.

So how do we assess these things? Imaging instruments we used before. CT, probably not sensitive enough. Something like CT probably can't really work. MR, with some of the new MR technology and the high field magnets, especially some of the blood flow techniques, that seems worth looking at. It is very good for edema, demyelination, and as Mark does in his lab, temperature assessment.

PET scanning, low resolution sort of technology, but now the ability to use radioisotopes with antibodies, looking at things like apoptosis. Maybe that technology could be good, especially with some of these small animal scanners.

Fluorescence is very interesting, fluorescence confocal microscopy, even in living animals, very shallow penetration, but the ability to look at really small spatial molecular change differences, which could have a big cascade of events. Maybe it is these small things, but we do have to do those.

Pathology, which is what I did, my bread and butter, seems like conventional history is not so good for this. The typical morphologic change is not very useful. Transmission electron microscopy, immunohistochemistry, apoptosis assays, there again, a little bit better and maybe possible that we can identify things.

I do think, as my students have said to me 500 times today, they are in the back of the room here, many of them, this idea of doing these fishing expeditions with microarrays to look at gene changes, everybody that does those -- I have only done a few -- realizes how complicated they are, because you get 2,000 genes back, and the expression level goes from fivefold to 5,000 fold, and you don't know whether fivefold matters or 5,000 fold matters.

But anyway, I think that is a very sensitive technique. Then you can look at a lot of genes at the same time. I think some type of that work can be extremely useful.

Then the behavioral change as Mark has shown, I think is a really innovative new technique that could be the most sensitive of all these, but has to be pursued, I guess.

This was one of -- Joe gave us these four points that we should look at, targeted and cost effective research to better define thresholds. I don't know any better than anyone else, but I would think that the exposure techniques of trying to do these short exposures, fairly short exposures, maybe a few hours, and then these longer exposures, we have done a little bit of this with extremely low frequency electromagnetic fields. We did things like this, which did show some changes. But here again, almost a little bit of a fishing expedition to decide what it is. With all these animal models, you tend to want to go over the top a little bit to identify a change and then scale back to something that is more realistic.

This issue about repeated exposure. Having an assay that you can use at the end is critical in measuring temperature, as Mark has said many times. This inability to measure temperature or to get something to correlate with your pathology change is really critical. I think I said that.

In summary, what it seems like is important after listening to this today is to get some appropriate animal models. Mice are the most common model and the one that we might think about using the most, but also more difficult, because you have to be able to give the energy in a consistent and confined manner. Then some sort of imaging and biology assays, whatever; these have to be worked out. Small animal models are more challenging, but that has to be done.

The ability to measure this temperature, to assess the temperature, to know what you are doing in terms of thermal dose. We have typically used invasive technology even in mouse studies I do in my lab now with nanotechnology, heating. Even in a mouse tumor we use an invasive -- although the probes are quite small now, maybe .3 millimeter, but putting that in a mouse brain is not going to work. So there has to be some other technique that we use, or we measure it in some other way and then try to co-register it.

Then this to me is the most interesting part in some ways, and the most challenging, is to figure out what are sensitive assays that are consistent, that can be used to pick up changes that we haven't previously identified, and to pick up molecular changes. I think that is where we are in these levels of two or three degrees above baseline temperature. They are more sophisticated and they have to be quantifiable. So I think those are important.

That might be it. I will say one thing. I will just plug my own work, which I think is interesting. We are now engaged in this technique of using iron oxide nanoparticles that are activated by an altering magnetic field. This is a portion of a cell, and these are clumps of particles.

We are a little bit in the same boat that the RF is in, because now these particles become intercellular very fast, and we activate them. We can kill cells with what we believe is heat, although there are doubters. But we can only measure heat. But once they get intercellular and you activate them, you can't measure heat, because we don't have anything sensitive enough to measure.

This is what one of the particles looks like. They are antibody directed. I will show you one more slide. These are phagosomes with nanoparticles in them. The larger the aggregates, the higher the SAR and the more heat you get. But even though these particles individually and in clumps get very hot, it is likely that in some situations, especially antibody directed, there is no temperature elevation outside of an individual cell. It may be completely confined in individual cells, which is attractive in some ways for metastatic tumors and others. But anyway, that is a little off the point.

Thank you very much.

DR. MORRISSEY: Thank you, Jack, that was great. Right on point as well. Some thresholds from that One of the things that work and some other work and targets and things, so that will be great for discussion after the final talk by Dr. Nathan McDannold, Lessons from Clinical Hyperthermia and Brain Tissue.

DR. MC DANNOLD: First I would like to thank you for the opportunity to talk to you today. The title of my talk is about lessons from clinical hyperthermia and brain tissue. Unfortunately we don't have a lot of clinical data at this point to discuss, so I am going to just talk about the preclinical work that we had to do before getting to the clinical patients with the technology

that we are working with.

Our group investigates the use of ultrasound for non-invasive ablation of tumors. We have a big interest in going after brain tumors.

Just to give you a quick overview of what we are doing, we have a device that we use, a focused ultrasound in the brain. It turns out to be a fairly complex device to get around the problems of the skull, which heats up with the ultrasound beam and also distorts the field. So we use this large hemisphere array and focus it into the brain. We are going after brain tumors, and other groups are going after other targets.

So far we have only done three patients with the device. These are some of the tumors, where we come in from all directions and focus to a small point. We are concerned with both heating at the focal point, but we are also very interested in protecting the brain surface. So we have to be able to show that we can insure that we are not overheating the brain surface. I think that is fairly relevant for the discussions that we are having here today.

These are some examples of the heating. We see the focal point in our three patients, and we get some artifacts. These are MR temperature mappings that we do during the procedure. The whole thing has constant thermometry.

As these patients' numbers grow, we are going to have data on the heating on the brain surface, and we will be able to establish that certain levels are safe, at least with respect to the exposure levels and durations that we are using in these treatments.

As I mentioned, we are very interested in looking at the brain surface. With these three patients I went through and we analyzed this. We can generate curves and using different metrics for the brain surface heating and the focal heating. We can do some simple extrapolation, and we can use this data to determine whether -- going forward our initial patients were very conservative, and we had a limitation on the power -- that we could safely do this using known relationships based on the hyperthermia literature, on thresholds, and the temperatures that we are measuring, and it did indicate that we could go forward.

But before we got to this point, we had to go through and validate these measurements, and the ability of MRI temperature mapping to accurately measure the temperature and predict thresholds. I am going to talk a little bit about that.

This is a study that we did about five years ago. We wanted to demonstrate that we could use MR temperature mapping to determine these thresholds, but we also wanted to compare the predictive value of input parameters, the power that we were putting into the animal in predicting the outcome compared to the measurements we did with the MRI, with the peak temperature and the thermal dose. We were also interested in evaluating the sensitivity of MR to histological changes.

We had a secondary goal. Our group is also interested in using ultrasound to temporarily modulate the blood-brain barrier to enable drug delivery. We have heard in our previous talks that heat is known to disrupt the barrier, so we wanted to see if we could achieve blood-brain barrier disruption without any irreversible tissue effects.

Then finally, we hope that we can use the advantages of the technology that we are using to focus ultrasound phased arrays that we use to generate the heat, and the MR thermometry to improve our ability to measure thresholds in the brain and other tissues.

As has been discussed a little bit, these experiments measuring thresholds for heat damage of any type can be very challenging, especially if you are using an implantable heat probe, some sort of heating probe, and temperature probe also. The temperature probe can interfere with energy deposition and vice versa, the energy deposition can interfere with the ability to measure temperature.

It becomes very difficult to accurately correlate the damage effects to the probe location. If you are looking at high temperatures, you are very interested in what is happening at the edge. At the edge of your heated zone, the thermal gradients are very sharp, and your error in your temperature measurement will be large in those locations. You may also be confounded by things like secondary damage. Tissue that is not directly heated could have a reaction that you may interpret at damage, and you may end up getting an incorrect threshold.

If you are looking at later times, tissue swelling also confounds your measurement. You think the tissue was at one place where you heated it, and later due to swelling it has moved, so you extrapolate what the temperature there was, and you could be incorrect.

We hope that with MR temperature mapping we can get around most of these problems. With ultrasound phased array we can tailor the temperature profile so that we have a fairly flat profile at the center.

This is a very simple phased array that we use that creates a ring-shaped focus, and over time the heating at the center fills in. You can use the temperature at the center where it is flat to get a very good precise temperature measurement. Since it is an external source, you can easily move this source to a different location and get larger numbers to improve your threshold detection.

This is just an example of the profile you can get with this very simple phased array. You can get better results with more complex devices that are available now. You can see at the center of the heated zone, it is very flat, so you can do averaging and get a precise temperature estimate.

Furthermore, since we can move this to different locations, we will have one measurement for each point. We are not trying to correlate the temperature distribution to the effect of damage, so we can tailor the dose at each location, and get a precise measurement, and then look for any effect there. It is fairly reasonable to assume that any effect we get is going to occur where the maximum temperature was. This we hope will improve our ability to measure thresholds for damage.

We did the study. This is in rabbit brains. We went through a craniotomy. We did four locations in each brain, and then we looked at the resulting tissue damage in MRI and histology.

Back to our original question, if we could disrupt the barrier, in some cases right after the exposures, we did see a small area where we had MR contrast enhancement, which indicated blood-brain barrier leakage.

We didn't see any effects in T2. In our previous studies, any time we saw anything in T2 we saw damage, so we were encouraged. However, when we went back -- this was two days later -- we didn't see any blood-brain barrier opening, which we might expect, but we did see T2 indication of damage, and in histology that is also what we found. So in this study with these exposure durations, we were not able to achieve blood-brain barrier opening without any damage.

This is a close up.

In other cases we saw both blood-brain barrier breakdown and immediate tissue damage in the brain. When we looked two days later, we saw enlargement of the lesion that we produced. This dark area indicates hemorrhage, so we saw different effects that occurred, delayed effects, and an increase in the size of the lesions.

Looking at this, we can then generate our threshold plots. We can plot -- for each location, each of these is one lesion in the brain, and we can correlate.

In this case, the numbers of experiments we did, we couldn't break down the different types of damage. If we had larger numbers, we could eventually do that. Then we could generate these plots of whether we had damage or no damage.

In a certain region here, this was about equivalent minutes of 43 degrees. Where there was an uncertain region, this was most likely due to noise and biological uncertainty in the experiments. Then we could fit this to a probability curve and give a threshold level.

What value to choose as your safe threshold level is open to discussion. All we can do is generate these curves. We had a 50 percent probability at about 18 minutes at 43 degrees. If you wanted to look for an absolute safety threshold, you might go back to where the probability is very small, which is a few minutes. Or maybe within the error bounds, which would be down maybe at one minute at 43 degrees.

So this uncertainty with these exposures correlated to an uncertainty in temperature of about three degrees. This is about what we would expect with the technology we were using at the time. We were at a one and a half tesla MR scanner, given the noise and the imaging, so this was the level of certainty we could get with these numbers. Presumably if we increased the number of experiments, we should be able to get a tighter threshold.

We then also compared the ability of different parameters to predict when we would get damage. The thermal dose wasn't significantly better than the peak temperature, but the exposure durations were not that different for these experiments. But we did see a significant improvement in our ability to predict the threshold based on the measurements compared to the input parameters.

So at least with ultrasound, looking at the input parameters is not a very good prediction of what the temperature is. So I think that having measurements is going to be key, or taking that into account, the uncertainty in ability to predict the temperature rise for a given energy deposition, is going to be very important.

We also looked at the increase in the diameter. I am not going to go into too much detail on this plot, but as I mentioned, we did see an increase in the diameter of the lesions. So if you are looking at a later time point in histology and trying to correlate that back to where you had different temperature iso contours, you could run into problems due to this increase in size and damage. So I think the technique that we are using, where we look at each location as one data point, instead of trying to correlate the entire lesions at different temperature values, I think is a valuable way to go.

As I mentioned, we didn't have enough data to look at different effects of the heat exposures. We did see different effects. In a lot of case we just saw basic thermal necrosis, but in other cases the lesions near the threshold were associated with hemorrhage.

This goes back to what cells are going to be most sensitive to damage. I think that this data and some other data I will show does suggest that blood vessels are very sensitive to thermal insult.

This is an earlier study we did, where we are looking at the peak temperature and correlate to that. We did find a correlation between different types of thermal damage. These are different histology slides from the three most severe type of damage. Interestingly, the earlier types of damage were generally thermal necrosis, but around 60 degrees Celsius we started to see onset of hemorrhage, but then at higher temperatures we don't see that hemorrhage. So again, at some point without cooking the tissue we can damage the blood vessels and get hemorrhage as a result.

Our conclusions from this are that we always saw damage when we had DBD disruption. So we weren't able to achieve that. The thermal dose was best for predicting the damage. This goes back to the talks earlier today. These were better than the applied ultrasound parameters. We also were able to show that we were sensitive to the threshold level tissue damage. This should hopefully allow for a longitudinal study, where we can look at using MR and evaluating the effects to tissue over time, without having to sacrifice the animal at different time points.

We did a similar study a few years earlier in muscle, where we were looking at MR changes. The values that we got for the threshold, at least the 50 percent thresholds, were pretty much identical to what we saw in the brain. We didn't look at histology in the muscle.

I think this is a good indication that MR is a useful tool for this sort of analysis. These values are also close to what other people have found using MRI, and it is consistent with the prior hyperthermia studies, which are written up in the report by Dr. Dewhirst.

Going back more to the clinical studies, this technology that was first tested in our group is now being used in Switzerland. I think that there is going to be some data coming out that will provide a lot of missing data in humans. They are using this technology not for tumor ablation, but for functional neurosurgery for chronic pain. They are targeting areas. These are healthy tissues, they aren't tumors. They are not using very high temperatures. They are going to have a lot of data on what the thermal dose was and the resulting tissue effects, at least as measured by MRI. I think this data is going to be fairly valuable.

Speaking with the investigators on this, they are also reporting that their patients do feel sensations as the ablation occurs. They are going to have the correlated temperature maps with that. So studies like this in the coming years are going to provide useful information that is currently lacking in the literature.

This is just some of the data. Using different MR techniques, this diffusion rate in imaging, we can look. It is very sensitive to small vascular changes. T2 weighted imaging is looking at -- you can see edema and you can see DBD breakdown with the T1.

To summarize. This combination of using MR temperature mapping and focused ultrasound provides a good platform for defining thermal thresholds. MRI may also be useful for looking at more subtle effects than the irreversible tissue damage that we looked into. As the previous talk mentioned, MRI is sensitive to a lot of things, including function, that we may be able to probe for thresholds.

I didn't get into the detail on this, but there is a very nice report on nerve excitation and function changes. These thresholds may be lower. The current temperature sensitivity of the MR is thought to be about half a degree or one. In principle you can lower this by imaging for longer times, but we may reach some fundamental limits on MR temperature imaging.

Finally, the ongoing clinical trials with focused ultrasound and laser, which I hadn't mentioned, is going to hopefully provide more human data on brain thresholds that are currently listed.

With that, I am going to acknowledge the supporters of this. Thank you.

DR. MORRISSEY: Thank you, Nathan. That was again a very good talk, right on target.

We are up to the break right now. We are a little behind schedule, but we will break for a few minutes. I am probably not going to allow you 15, but enough to stretch your legs and get something to drink.

We will come back here. Get your questions ready. What we want to do from now on, after the break until lunch, is discuss some of the things that the presenters have presented -- the most sensitive cell type, what are the target health effects, what are some of the thresholds that we know, what are some ways to get where we need to be.

(Brief recess.)

Agenda Item: Discussionon the Brain

DR. MORRISSEY: I am going to open the discussion, which is dependent upon participation from the group, especially the expert group. I think Mark wanted to start off with showing some slides that are relevant to what we were talking about.

DR. DEWHIRST: Jack and some of the others brought up the idea of doing genomic analyses to see if we could understand how hyperthermia might affect tissue function. It turns out we have done studies like this, so I thought I would show this, not so much for the meat of it, but to show you how this could be done.

What we are looking at here is a genomic analysis done on canine sarcomas. These are pets that have cancer. What we did was to take tissue biopsies on these tumors before treatment, and then they underwent one hyperthermia treatment, and temperatures ranged from about 40 to 45 degrees within the tumor. Then we did a second biopsy 24 hours after this first heat treatment, and did a second genomic analysis on them.

At the same time, we also did functional imaging on these animals. We measured ADC, which is the diffusion coefficient of water. That is measured with MRI, as Nathan mentioned earlier. We also measured interstitial fluid pressure in these animals.

So if we take the genomic data and do what is called an unsupervised analysis, we come up with two clusters of animals here, the red cluster and the blue one. These clusters differ in terms of their gene expression, that is highly enriched in mediators of inflammation. So one cluster has a higher grouping for inflammation versus the other one.

It turns out that that clustering of inflammation related genes links to changes in interstitial fluid pressure. So this is a change in interstitial fluid pressure for the group one versus group two. This group here has much less of a change in interstitial fluid pressure compared to that one. So that is linking the genomic expression data back to the physiology.

We also showed that the change in the diffusion coefficient of water, that is, from pretreatment and then 24 hours after heat, was linked linearly to the expression level of FLT-1 and KDR. These are the receptors for VegF. So it suggests that when you do hyperthermia treatment, tumors that have a higher expression of these receptors for VegF are going to respond more greatly in terms of their change in interstitial fluid pressure.

Anyway, I will just skip over this part because it is not so important for this discussion. But the point is that you can link these genomic data back to functional imaging. I think it is a great area for exploration at lower temperatures and normal tissue. Clearly there is opportunity to do that if we are careful, using the techniques for example that Nathan talked about and Jack, of linking back your thermal history to the actual tissue biopsy that you take.

DR. ZISKIN: I'm not familiar with the use of interstitial fluid pressure. The only thing that I can think of is edema formation. But what is it actually relating to?

DR. DEWHIRST: I'm sorry, it is related to edema. In fact, the diffusion coefficient of water and IFP are both related to edema.

So when you have an increase in ADC, that means that there is more space for the water to diffuse. It is diffusing faster, if you will. So that is definitely related to edema, and so is this.

DR. HOOPES: Mark, what made you want to do interstitial fluid pressure? What made you think to do that? DR. DEWHIRST: We thought that hyperthermia might be changing water content, but we didn't know if there would be variation between animals or if there would be any kind of link back to anti-tumor effect. So that is how we did it.

DR. HOOPES: I was curious. We found too that it seems like it is one of the biggest indicators for tumor response for nanotechnology.

DR. DEWHIRST: Oh, really?

DR. HOOPES: And it is related to VegF, and there is even potential to use anti-VegF to affect this, which affects other things. So it is odd you said it. We thought we were novel about it, but I guess you are there, too. A lot of people are.

But it does turn out to be something we have known, And really, interstitial fluid pressure is just the natural leakiness of the tumor capillaries. It is maybe a step less than a true edema situation, but it is high in most tumors. It is higher than in normal tissue.

DR. DEWHIRST: I think if you injure a normal tissue, interstitial fluid pressure might go up. It is a very sensitive method if it is done properly. You might even in a normal tissue see a change, I don't know. But certainly you could do ADC, and Nathan showed data like that. So I think ADC is another way to get it without having to poke a probe into the tissues.

DR. HOOPES: There are probes now that you can measure, very sensitive small probes in the same order of magnitude as these thermistors, .3 millimeter that measure interstitial pressure.

DR. DEWHIRST: I would like to emphasize too that there are optical techniques now that might be very useful for a variety of normal tissues. For example, we at Duke have the capability of measuring hemoglobin saturation, total hemoglobin content. We can measure back scatter which is related to cell killing. That can all be done with optical spectroscopy. You don't have to take a tissue out to get this kind of information, which means you can look at it over time without having to sacrifice an animal. You can even do this in humans.

Then with some of the new infrared technologies that are coming out, you can probe even deeper. There are a number of different molecular probes that are becoming available, apoptosis markers, markers of matrix metalloproteinase activity, things like that, that can be detected optically at reasonable depths, so that you get tissue information on animals, anyway.

So I think the optical is one way to go. You kept pushing on the idea of cost effectiveness. Optical is cheap compared to MR.

DR. MORRISSEY: Cheap is good. Interstitial pressure, does that seem like the most sensitive end point? Is that relevant?

DR. DEWHIRST: I think it is one end point that could be looked at, but I think some of this optical stuff might be worth doing as well. It is an extremely sensitive technique.

DR. HOOPES: Interstitial fluid pressure is a pathologic condition that is due to something else. It is that tumor vasculature that allows that to occur.

I agree with Mark, these techniques such as FRET, which is this fluorescence resonance technique, which looks at the spatial ability between molecules that are changed. It is very sensitive. It has low penetration power, but I think those are the kind of things that are relatively inexpensive, but highly sensitive and can see changes, sub-anatomical changes.

DR. MORRISSEY: So what exactly are you looking for with the optical spectroscopy? Are you looking at flow rates?

DR. DEWHIRST: With spectroscopy you get information about the oxygenation of the tissue.

DR. MORRISSEY: So it is like PET?

DR. DEWHIRST: Yes. But again, this back scatter effect I am talking about relates to the size of the scattering objects in the tissue. If you have a lot of apoptosis going on, you get a change in this back scatter factor which reflects that.

We have got papers published on this. It is a very, very sensitive technique, and I think it could be useful for some of this lower temperature stuff.

The activity of matrix metalloproteinase, this was something that Ralph Weisletter's group led the effort on. He has got a near infrared probe that can get down two or three centimeters. So you have got the possibility of getting information at depth in a rodent anyway with this near infrared technique that Weisletter has got. That is an example, where you have to have a probe, but you get very valuable information out of it.

There was a question in the back.

PARTICIPANT: I wanted to make a couple of comments to follow up to what you were mentioning about optical techniques. There is a variety of optical techniques that have been published to monitor thermal damage in a variety of different ways. One is polarized light microscopy and looking at changes in birefringence of tissue induced by denaturation. This can either be done histologically.

There are also techniques like optical coherence tomography, which is a high resolution imaging approach that has been used. It has about maybe a couple of millimeters penetration depth, so it is pretty shallow compared to MRI. But it can provide real time information on changes in polarization and scattering due to denaturation.

There has also been some studies published on looking at endogenous fluorescence in addition to FRET that also tracks changes in denaturation of proteins.

DR. DEWHIRST: You can actually do what is called redux ratio imaging, which is FAD over an ADPH, both endogenous fluoroflores, that tell you about mitochondrial function. So that is definitely possible with optical.

Actually, Bruce Tromberg at UC-Irvine has been doing a lot of work looking at the water signal. The biggest absorber in tissue is water. In fact, you can use that to measure tissue water content. So that is another technique. You have to go down into the infrared to get the information, but it is there for the taking, with very cheap technology.

PARTICIPANT: I have a question to the last three speakers. I am quite new to the field. Although this workshop is more review of the thermal effects of the blood-brain barrier, or brain itself, also we are looking at cell phone radiation.

Most of the work that has been presented here is increasing the temperature by different ways. So is it possible, do you think if you increase the temperature using any kind of means, what we have done with drugs of abuse, and then on top of this you have radiation effects, do you think that will compromise the blood-brain barrier permeability? Even at the smaller scale, because Dr. Dewey did a good job showing the .5, the next step you do this. But in some cases after 40 degrees, even .2 degree become important to show the changes in the brain when you are going beyond that.

So I think I just want to get some points from some of other experts in this field to see if we need to be looking into that, too.

DR. MORRISSEY: I think the question was to the speakers, and whether other drugs or other things could sensitive to blood-brain barrier leakage.

DR. HOOPES: Was that it? Was that what you were thinking? Or is there non-thermal effect? Are you talking non-thermal effect?

PARTICIPANT: We are all talking about thermal effects and using different ways to increase the temperature and looking at adverse effects with the temperature. But in the back of the mind, this whole thing about cell phone radiation can increase also the temperature. So I think we need to keep that in mind when we are looking at increasing temperature, especially I am talking about the brain. Can it become important, even increasing .2 or .5 after 40 degrees, and can it be possible to produce the adverse effects in the blood-brain barriers or some of other pathology or edema and all those things?

DR. HOOPES: Eric maybe is the best to answer this question, but that study has to be done specifically. It is going to matter, about how much field do you give and what is the strength of that field and all those variables have to be taken into account. But I think conceptually, the fact that RF can generate some heat is possible.

PARTICIPANT: That is what my point was. I think if you increase the 40 degree and you see the changes in cell signaling, morphology, pathology, everything. But then on the top of this, if radiation can increase two degree or .5 degree, is there any studies being done showing --

PARTICIPANT: (Off Mic.)

PARTICIPANT: But it also depends on the exposure. Fine, I think dose always depends on the duration.

PARTICIPANT: Not with a phone. It is going to be steady state. It will be less than .02 of a degree or something like that.

PARTICIPANT: So I guess if somebody is on the phone for 15 minutes versus 30 minutes, temperature will not increase?

PARTICIPANT: (Off Mic.)

DR. MORRISSEY: I think we are making the assumption that RF certainly can deposit heat in tissues at some level. What we are doing right now is just trying to find the thresholds of heat that are required to trigger these adverse effects, whether it comes from a cell phone or a tower or whatever source, RF or not.

DR. D'ANDREA: In thinking about translating some of this kind of work to safety standards, my mind goes right to the mechanisms of coupling energy to the body. For these safety standards, we are talking about fields that go from kilohertz all the way up to many gigahertz, millimeter waves and up to terahertz, even.

So is it going to be possible to think of a metric on which to base the standards that is going to be viable for skin, for muscle, for bone? As frequency goes up, penetration is going to be more towards the surface. So in order to support a limiting dose of RF, you have to have some metric that covers many tissues, and that seems to be kind of difficult to do.

Maybe the speakers could comment on that aspect of translating this work into safety standard metrics.

DR. HOOPES: My own thought is that it is a lot like radiation. It is figuring out what are the safe limits for radiation, what are the normal tissue -- it only took about 100 years, and we are not quite there yet.

I was just being facetious, obviously. That is the concept, and different tissues are going to respond slightly differently. I don't think one size will fit all, for sure. So those studies will have to be done in individual tissues, I think.

DR. MC DANNOLD: I don't know if the data we have now, how relevant it is. But the temperatures that you are discussing, I don't know if any data out there looking at such low temperatures. You would need such a long exposure to get the effects that we have been looking at in hyperthermia.

The question I have is, what should we be looking for in these sorts of exposure levels? We wouldn't expect, based on the studies that have been done so far that we would get irreversible tissue damage, but are we looking for temporary changes, or are we looking at potentially at-risk groups and how they would respond? Maybe the experts here would know what should we be looking for in these very low temperatures for potentially a long time, these low thermal doses. I don't know of any data out there on that.

DR. MORRISSEY: Nathan, that is a great question. That has been a problem, what is the target for these very low level effects. I think we have come to the conclusion that we need to go to the Hyperthermia Society, what effects are you seeing at modest temperatures, and just titrating down dose response, to see where do they fall off.

There really are no repeatable, reproducible targets at very low, low level RF exposure that aren't sufficient to cause heating, so we are left with what are the targets.

DR. MC DANNOLD: I think if you could assume that the Arrhenius plots work at these lower temperatures, you could exclude a lot of things based on data that is existing now. But beyond that I think it is an open question.

PARTICIPANT: Well, I would like to add a little additional question to this question from you. What I have seen up to now being brought up as being effects is that it is either done by the whole brain heating or like you are doing with very high temperatures in a small volume. How do these high temperature small volumes experiments in thresholds convert to the large volume at large temperature increases?

So how much tissue damage do you need in order to see it as it grows effect, which has not been addressed. Whereas, you look to the exposure at lower frequency, we will have whole brain exposure, and if we go to a higher frequency, where you have local deposits, it will not enter the brain. It will be the skin.

So there might be a difficulty in translating our experience with hyperthermia towards the area that we are dealing with now.

DR. MC DANNOLD: I think that is a very good point about differentiating between effects of heat on cells and then effects of heat on metabolism and the body's response to it. That could also be problematic.

PARTICIPANT: So in fact, you need a thermal threshold, but also a volume threshold.

DR. ELDER: A question for Dr. McDannold. If I wrote down correctly or remembered correctly from your summary slides, you had a very short statement that said that since the temperature sensitivity was on the order of .5 to one degree C. Would you revisit that statement and tell us where that comes from and what was the basis for that very short sentence?

DR. MC DANNOLD: That is based on what I have seen in the literature, using this temperature mapping for hyperthermia or ablation monitoring. They basically look at areas that are not heated at all and look at just the noise level in these.

These are maps of temperature change, so what is the distribution, where you know the temperature change is zero, the standard deviation is on the order of one degree.

DR. ELDER: What is the physiological end point that is sensitive here?

DR. MC DANNOLD: That was an image noise level, what we typically use now. That is acceptable for us for what we are using it for. In principle, with MR you can improve the signal to noise ratio by scanning for longer times. There are a lot of things you can do with a higher field, better coils. But what the actual limit is, if we can be sensitive in vivo to the levels that you are talking about, I think in principle we should be able to do that. I think that there are still a lot of unknowns. I don't know of anyone who has done that yet.

DR. ELDER: I would like to make another comment while I've got the microphone. Robert McIntosh made the comment about the rise in temperature in the brain tissue from using the cell phone being an order of .1 to .3 degrees C. I just wanted to remind the audience that that is all based on mathematical models and has not been verified in a live body, to my knowledge.

More than likely, that temperature range is probably maximum to what you might expect in a live body. In other words, the temperature rise in the brain from using a cell phone is probably much less than those values.

DR. BERGERON: And in addition to that, it may only be that much on a normal temperature brain, if the brain were at an elevated temperature for whatever reason, either through passive heat exposure or exercise or some bias because of some clinical condition, it may indeed be different. So we have to recognize that as well.

DR. CHOU: An additional point to Rob McIntosh, the degree temperature rise is only a very small region. We are talking whole brain, this is a very small region in the brain during the wireless communication.

DR. FOSTER: I just want to point out that calling for physiological studies to look for effects at low levels of exposure are incredibly controversial. Funding agencies should be prepared to spend a lot of resources on specific effects.

To give you an example, we had a talk earlier about the effects of heating on the blood-brain barrier. It turns out that there has been almost 40 years of controversy about whether low level microwave radiation damages the barrier.

It started off in the late 1960s when a very poor study by a Ukrainian group followed up in the early '70s by equally poor studies by an American group, which claimed that there is an effect. Then after this created an international controversy, a number of well-funded and expert groups got into this, including University of Rochester and Brooks, doing exhaustive studies that found nothing at all in exposure levels below international limits. So the subject pretty much died in the early '80s. It was resurrected in the early '90s when a group led by Salford started reporting effects of mobile phone radiation at low levels below exposure limits. This led to a series of well-funded and highly detailed studies which could not confirm these results.

So after 40 years of research, we still can't say for sure whether low level microwave radiation affects the blood-brain barrier. We do know it is a very complicated thing to measure, subject to many induced artifacts including local blood flow changes and so forth. But I think after all this work, very few health agencies would use this effect in setting exposure limits.

I'm not sure that we should be calling for more exploratory research at low exposure levels as trying to find hazard mechanisms that are clear from higher exposure levels, and trying to find out how they might apply to setting exposure limits.

DR. SHARMA: You pointed out a very good thing regarding effect of any kind of radiation or exposure on blood-brain barrier. But the point is that when we try to study blood-brain barrier, it should be coupled with the tracer that we used.

The most established research in blood-brain barrier is done by Stanley Rappaport here at NIH, what he has done. Whatever knowledge we are having so far is largely derived on his studies. Even if you use osmotic opening of the barrier, you can open for a certain period, for example, 30 minutes, to proteins, large molecules, but still it is open for four to six hours to sodium.

So those studies, either they fail in detecting blood-brain barrier leakage, it must be coupled for what they are measuring. So this is very important. As a pathologist, our concern is the leakage of protein. When protein goes to the brain, it produces vein edema.

So I think this is a theoretical question. Whenever we speak about opening of the blood-brain barrier or breakdown of the blood-brain barrier, I think a tracer study must be informed for which tracer they are not open. It could be they are not open for proteins, but they are open for small molecule tracers that we don't know.

DR. FOSTER: But the point is, we have had 40 years of controversy about this effect, and nothing has come out of it. People are doing poor studies. People are arguing about whether a small effect exists or not. This kind of controversy will keep on going forever and will use up enormous resources. It is not clear that changes in the blood-brain barrier at a tiny level are even physiologically significant.

So the question is, what end points should we be looking at which are useful in setting the exposure limits, and how do the thresholds extrapolate down at lower exposure levels.

DR. SHARMA: I was just reminding you about this story only that in the '70s people were looking for ischemic injuries, and from '70 to early '90s they don't even care to do a study of blood-brain barrier. So there were no studies. But after '90s it is quite clear now that all ischemic injuries opens the barrier.

I am interested in the question in the laboratory where they are studied. This is also very important. That is my only point.

DR. BERGERON: Correct me if I'm wrong, but are those 40 years of experiments on normal brain, and maybe now it might be more focused if we were to examine the effects on a brain that is potentially biased for some other reason or clinical reason.

DR. FOSTER: If you are looking for small effects from small exposures, it is a hopeless thing, unless the funding agency wants to massively support the research. It would not appear to me after all this controversy to be a very productive scientific end point. I would be very reluctant to call for more exploratory studies involving different end points at low exposure levels unless they are informed by some clear hypothesis that something might possibly happen.

DR. MORRISSEY: A lot of good points being made.

PARTICIPANT: Yesterday we saw data on pain levels versus necrosis in skin cells. I would like to ask the question of the experts, is there any evidence of any kind of damage that occurs from multiple occurrences of pain and not damage? Is there some underlying effect that can occur from that?

DR. DEWHIRST: That was me that made that statement. There is no evidence, because we have virtually no data on repeated exposures. So the answer is, we don't know.

DR. SHARMA: Your question is very valid, do we have any changes in the brain or central nervous system after pain or pain related mechanisms. Is this your question?

With Professor Thorsten Gould who is an expert in pain mechanism, we are using an animal model of neuropathic pain. The paper is published in the Journal for Pain in 2006. What we have seen is that neuropathic pain either produced by lesion of the sciatic nerve or a crush of the nerve from two to ten weeks there is specific breakdown of the blood-brain barrier in the spinal cord associated with neuronal changes.

So this is now established fact. Therefore people are thinking about that pain itself can produce either short term or long term changes in the central nervous system. I have only this to add. You can look for the literature and you will find this paper. Thank you.

DR. MORRISSEY: These are all great question, what is an adverse end point, does pain qualify, is it followed by lasting damage, a lot of the stuff that Ken brought up.

I think what we also do need to keep in mind here is that this is not necessarily feeding into a cell phone standard. It is feeding into a radiation, an RF standard. I don't know, but I have heard if you put these little poodles in a conventional microwave, they will die; that is an adverse effect. If you look at these blood-brain barrier studies at very low SARs, the weight of evidence starts to suggest of the good studies that at very low SARs, there is no effect.

Some are in the middle, there is an adverse health effect. That is what we need to find. We need to find what is the most appropriate and relevant parameter that we should look at. The blood-brain barrier has been mentioned. There are pros and cons. Other potential parameters would be welcome. What are other things that may be used as a parameter to look at with regards to the brain and the central nervous system? Sooner or later we need to identify the adverse effect so we can set limits.

DR. D'ANDREA: I might suggest as I have in the past that the integrated output of the central nervous system by measuring behavior might be the most sensitive measure of changes due to RF exposure. At least, that is the experience that we have had in the past. With the new cognitive measurement techniques, I think those can be even better than some of the animal work that has been done so far.

So I would suggest looking at cognitive impairment as one measure of subtle RF exposure, and heating.

DR. MORRISSEY: Comments? Tomatoes?

DR. DESTA: Quick question to that. Would that be applicable to very small partial body exposures?

DR. HOOPES: That might be difficult. I had the same comment. In a rodent brain there is not a lot of foliage. It looks like a little bowling ball up there. So there are areas of that brain that aren't being used too much. So if you think you are having a really focal area of some sort of effort, whether that is going to translate into behavioral change is a question mark. But I agree that the sensitivity of it, that is a physiologic change that may be sub-morphologic.

PARTICIPANT: Yes, but certainly if there were a hot spot formed in the brain, where it is formed is very important. For example, the implanted thermodes have shown how sensitive the brain can be to small changes in temperature.

DR. HOOPES: I hate to go back to radiation and biology roots, but that is what I know, anyway. If you look at the most sensitive thing, it is astrocyte activation. It is probably one of the most interesting things, and changes in endothelial cells.

When we look at changes in radiation, we learned that a long time ago, about late effects. You give a little radiation of the kidney, and you can see a little swelling on day three in the endothelial cells, and nothing happens until five years ago, and all of a sudden your kidney is fibrotic because you caused late effect damage in those endothelial cells, which translated into vascular change.

So there are these historical issues like that. But I think the activation of astrocytes can easily be seen with GFAP type stains and things like that.

DR. D'ANDREA: Do you think that would occur with low level RF exposure?

DR. HOOPES: I don't know. But it is certainly one of the most sensitive changes that you see in the brain overall, is astrocyte activation.

DR. MORRISSEY: One of the questions was, would it occur with significant thermal exposure, and then could you titrate down.

DR. HOOPES: We showed that. We showed that at low levels. We showed that at 40 degrees or something like that, there was a significant activation of astrocytes.

DR. MORRISSEY: But those are the questions. Will it occur at a thermal level, and how low can you go, and what is the threshold.

DR. HOOPES: I don't know anything lower than that.

DR. MORRISSEY: Questions, comments? Behavior versus pathology.

DR. ZISKIN: You mentioned 40 degrees activates the astrocytes. I always think from the point of view of clinically that at 41 degrees, that relates to a Fahrenheit temperature of 106. That is lethal. So that is a big effect.

It seems to me we need something that would be able to pick up something earlier than that, than 40. I tend to agree with John that behavioral things are probably more sensitive to physiology here on this. So I think his point is well taken.

DR. HOOPES: But even a behavioral change has to have a physiologic basis, doesn't it?

DR. ZISKIN: If you can measure it.

DR. HOOPES: That's what I mean. It has got to have a neurotransmitter change or something, because you don't get a behavioral change without physiology.

DR. ZISKIN: But it may be a lot easier to measure a behavioral change than some internal physiologic change. I think that was part of your point, wasn't it?

DR. MORRISSEY: And the baby doesn't necessarily have to be thrown out with the bathwater. You can kind of do both in parallel, but we are trying to narrow down the end points to make it practical.

DR. CHOU: Currently our IEEE and ICNIRP standards are based on behavioral change, so we are on the right track, at least from an historical point of view.

I just wanted to comment about the terminology. We have a lot of radiation biologists here who keep talking about radiation. Here we are talking about non-ionizing radiation, but meantime we are still using the term radiation. This is okay for the people in the science, and we understand what radiation means, radiation pattern and all that.

But I think for the general public or people who are non-specialists, when they hear radiation immediately they are going to think this is dangerous. So in terms of our IEEE document, and also from Bob Griffin from FCC and WHO, we try to avoid using the terminology radiation for non-ionizing. We say RF exposure, electromagnetic exposure instead of radiofrequent radiation. People cannot mix up with the other radiation.

PARTICIPANT: Just a short question. When you expose a brain, you first have to go through the skin. When we analyze this with human, it is maybe different from animal. The exposure of the brain is quite small compared to the skin itself.

How do you do to discriminate in such exposure what is entering the skin and what is entering the brain. We can analyze specially the brain exposure, but before to have any raise in brain temperature you have a lot of deposition in the skin at higher frequencies and at lower frequencies. At higher frequency it is most the case, so how to answer this question?

DR. MORRISSEY: Clearly in a practical sense -- I think Mike Bergeron brought this out as well -- lab studies don't always correlate what happens in the field or in practice.

I think right now, to try to keep things as simple as possible, we are stratifying tissue by tissue, and trying to look for the thermal thresholds. We leave it to the modelers to tell us.

PARTICIPANT: I understand. We are doing some things with animals. In this case we inject some physical adjunct like EMS, but in real case with real animal, the distribution is not exactly the same as with the average human. So what we deduce from the animal experiment, how to extrapolate to human.

It is not a question of dosimetry. We are not dealing only with cell brain, we are dealing with brain, the in vivo brain.

DR. SHARMA: This is a very important question that you raised. I can say only the following. Of course, every tissue is important in terms of radiation or heat exposure. But the brain is important not only because we are here for that. The reason is that brains normally do not have neuro regenerative capacity normally.

So therefore, by any chance if brain is damaged, even a small part or for a short time, it can produce long term consequences on it. Whereas skin and muscle tissue can very well regenerate. That is one thing.

The second thing, regarding extrapolation of animal research to human cases, of course everything is impossible. But you must consider that numerous laboratories are using cell culture as a model, and is studying on these things and trying to extrapolate. Even whole animal model versus human cases is much more natural to understand than even from cell or molecular biology studies.

Therefore, the point is that brain is really a special organ. The idea that one should try to find whether brain is damaged or not, because it has no regenerative capacity. And other tissue can regenerate. That is my point.

DR. GORDON: I have something different about the nanoparticle research that was presented briefly in Dr. Sharma's and Hoopes' talk.

The EPA, and I'm sure other federal agencies, are redirecting a lot of research into nanoparticle toxicology, which apparently very little is known about. That is why they spend a lot of money on it, because now suddenly all these nanoparticles are being dumped in the environment. So it seems to me that the possible interaction between these iron and titanium oxides that are apparently being deposited in our bodies as we speak, and the interaction with cell phone RF, is anyone looking at that?

The overall whole body toxicity of these particles, I think there is very little known. There is some interesting in vitro work I have seen, where it gets taken up in the neurons and so forth. But 20 years ago there apparently were no nanoparticles, or very few, and suddenly the industries are now just pouring them out, but the toxicology of them is just beginning.

So perhaps there is a mixtures effect, the synergism between RF and nanoparticles. So maybe you could comment, if you have anything to say. That is all I know about it.

DR. HOOPES: We use iron oxide, for the reason that it is not very toxic in itself, by itself. At least, that is the hope. But I appreciate your question, though. Eery particle that is developed, depending on multi domain or single domain particle, has an SAR value that is tied somewhat to the field strength and the frequency.

So what we find anyway, depending on the particle that we use, if we use a frequency outside of the zone, we don't get any heating. But there might be other motion artifacts and other types of things. So from a strict hyperthermia standpoint, we have a fairly narrow range for the particles we use, for heating anyway. But what about the other effects? I don't think we know that. The typical RF frequency wouldn't heat the particles we use. But whether it does other things, motion, vibrational, other types of things, there are lots of people who think the heat is a side effect, that it is not a major effect anyway. So we have arguments on that issue.

DR. DEWHIRST: We have been doing some stuff with nanoparticles, too. We have been using iron. We have also looked at some yttrium oxide nanoparticles. What we find is that these are taken up by cells clearly, but cells don't keep them, they get rid of them.

So if you expose cells to a nanoparticle, they will take them up, 24 hours later they are gone. I think this is an important thing to consider. These don't accumulate in the body permanently. We probably have some kind of steady state, if we are breathing these things in through the atmosphere or whatever.

DR. HOOPES: We are looking at tumors. We don't do a lot of normal tissue work with it yet, but our tumors, we haven't seen that. They go in phagosomes. They get on the membrane and the membrane itself is phagocytizing, as I showed. They get in those phagocytes, they can last a long time. They are not very toxic to the cells once they get inside these cytoplasmic phagosomes. So we haven't seen them go out.

DR. DEWHIRST: Our stuff is only in vitro so far. Actually, no, we have done window chamber stuff with them, too.

DR. HOOPES: These are just animal tumor studies.

DR. DEWHIRST: They just push them back out.

DR. HOOPES: We haven't seen that. I don't think they have seen it, either. Even if you have, you shouldn't say it.

PARTICIPANT: The nanoparticles and the intermediate phagocytes were known in the cancer therapy. Last BEMPS meeting we had two talks. But for that we need a few hundred kilohertz and very, very strong field, ten kilocalories per mole is needed. But this question is well known for local heating by nanoparticles, but other issue is not known.

DR. VEYRET: I have a very basic question, maybe naive, to all speakers about the doses and thresholds. It goes beyond the topic of the nervous system.

I just heard as an example that activation of astrocytes start at 40 degrees. That is a threshold in terms of temperature. But what about dose, in the sense as it was defined today several times, like in the CEM 43? Could you get activation of astrocytes at 39 degrees by just waiting longer?

That is a very crude way of asking the question, how do you relate the effect of thermal injury load in terms of threshold, that is, temperature threshold, and dose, that is, temperature and time and duration?

DR. HOOPES: Ours is fairly crude. You saw, we had those isodose curves we positioned on the heat side of the brain, and we did the exact same isodose curves. So we took the tissue in the contralateral side of the brain and just literally did a GFAP stain and just did a point count technique and quantified the amount of astrocyte volume in either side of the brain. One was heated and the other wasn't.

DR. VEYRET: I understand this. These are experiments that we do. But I am asking you and all other speakers about this definition. You gave an example of this in terms of temperature threshold and then of dose. So I would like you to comment on the difference between the two, because this is very relevant to RF research.

If I expose cells for 48 hours at 37 degrees with RF load and if my exposure system is not working, thermoregulated properly, then I might be at 38. Then should I expect heating effects from this one degree temperature elevation if I wait long enough? You see what the question is?

DR. HOOPES: Yes, it seems like that should be done, or even if you didn't get any heat but you exposed, would astrocytes increase their process and would there be activation. We didn't look at that, but I understand what you are saying.

DR. VEYRET: Yes, but I would like to have the answer from other speakers who might be concerned with this issue.

DR. DEWEY: I think your point is very well made. Coming back to my background as a radiation biologist/biophysicist, we used to talk and still do talk about what we call early effects and late effects in radiation, and what causes vascular injury, that whole argument. You know about that, Mark.

We still have those arguments, until Reuben got funded for many years to study late effects versus early effects. It seems like this question still pertains. You want to know what the late effects are, not just what you see in the first few days, but what you see a year after or something like that. I'm not close to what you are doing, but it sounds to me like a very important question.

DR. MANTIPLY: When I started reading the papers about CEM 43, I was struck by, up front it didn't put a temperature range limit on that. My understanding was, at 39, 38, at lower temperatures, you can't keep projecting this down to lower temperatures and long time periods. There is the adaptation that occurs over a long time.

So I think we need to be explicit when we introduce that, that this is valid over a specific frequency or specific temperature range.

DR. DEWHIRST: This is something that was in the original paper that we published in IGH in 2003. We had a long discussion about that. So if you go back and look at that paper, it is pretty well described.

DR. HOOPES: You're right, once you got below 41 and a half or something like that or above 48, then it broke down. It was a range when the CEM was accurate.

DR. VEYRET: I am not getting a clear answer to my question, so let me say it again in a different way.

I learned today that the main heating effects might be related to protein denaturation with a delta H of 140 kilocalories per mole. So this means that if I know what the temperature is, I can tell how long it will take to get so much above the barrier. If I work with 38 degrees, it might be years or centuries, so it is not relevant.

But my question is, from your experience, can I expect to have a one degree above nominal temperature affect this type of reaction if I wait two days? That is a very basic pragmatic question.

DR. HOOPES: So you are asking is there a threshold. I don't think we know that answer yet.

DR. VEYRET: No, I am not asking that. You said there is a threshold at 40 degrees, but --

DR. HOOPES: No, I didn't say that.

DR. VEYRET: No, that is how I interpreted it.

DR. HOOPES: I don't have that data. But you saw in the brain, we had six different levels that we measured, all the way from 60 degrees in the center of that necrotic lesion out to what looked morphologically normal. It looked morphologically normal at 40, and we stopped. There was also a 39 and a 38; we didn't do that, but was there some activation of astrocytes out there? We just didn't do it. That is the problem. But is there a threshold where at some point those astrocytes, they don't care, and they are not going to do anything. But that is still not the answer.

DR. BERGERON: Is what you are asking, does the threshold change with elapsed time?

DR. VEYRET: No, not really.

PARTICIPANT: Yesterday we had a very nice talk on the immune system. We have heard that 2.5 gigahertz makes very clear response.

DR. VEYRET: It might be caused by a phased transition in the membrane, which is a different process. I was asking about the change in the therapy for the protein denaturation, which is a very, very simple chemical change, while the first transition is another business.

DR. DONALDSON: To make a comment here. From my field of dealing with environments or heat and cold, we are commonly asked for thresholds. In the U.K., if the temperature falls below a particular temperature for too many consecutive days, old age pensioners get an extra bonus in their social security. Or if it gets too hot, advice is given out to people in the London Underground, take water to them.

When you set a threshold, you have to also give advice for people who exceed this threshold. You say we don't know the threshold and all the rest of it, if you give a threshold, how are you going to advise people not to exceed it?

DR. DEWHIRST: Can I just show you this data up here? This was from our original review. This is time to reach an end point. It doesn't matter what the end point is. The original data which was developed by this gentleman here was at 43 degrees, so we knew that in order to get this level of damage in muscle that you needed to be at 43 degrees for a certain amount of time, 30 minutes, it turns out.

So what I did was, I took the Arrhenius equation and I used the parameters for both mouse and man, and then generated false data or computer generated data for what it would look like above and below 43. So for example, if you get down to 38 degrees, then to get this damage you would have to go for tenth to the fourth minutes or ten to the sixth minutes, a million minutes.

So I think you can talk about thresholds in terms of time of exposure, it seems to me, when you get down into the range of say 40. Here you are talking about this kind of damage, something like 500 minutes at 40. So you could put in some kind of information about how long to be exposed at these elevated temperatures, given whatever the end point is. This happens to be the relevant one probably for what we are talking about. But if you are talking about brain or whatever it is, you could generate these kind of things.

Why do I have confidence that you can do this? Because of the Arrhenius plot. We know the Arrhenius formulation works very well. Bill talked about it this morning, about how these curves are parallel. We know they are parallel. So you should be able to do these kinds of things with a fair degree of certainty, put some error bars around it and say, if you are working out on a power line you shouldn't be up there more than three hours, you should come down. You could put limits on this kind of thing.

DR. DEWEY: Mark, do you have actual data there?

DR. DEWHIRST: This data point right here is real data.

DR. DEWEY: Yes, but how about 38 degrees?

DR. DEWHIRST: This is using the Arrhenius --

DR. DEWEY: (Off Mic.)

DR. DEWHIRST: I agree with that, but you can't do the experiment. How are you going to have somebody sit at 38 degrees for a million minutes?

DR. DEWEY: I don't know what your end point is. Is this human you are working with?

DR. DEWHIRST: Yes. This is human data here, and there is mouse data. It doesn't matter. It is not that much different.

DR. DEWEY: Do you have data at 38 degrees? How many hours after --

DR. REPASKY: (Off Mic.)

DR. HOOPES: That is a little awkward, because that is his normal temperature, is 38.

DR. DEWEY: That is the whole point

DR. REPASKY: (Off Mic.)

PARTICIPANT: It seems to me we have good evidence and very good reasons to say that the underlying mechanism that causes this damage has no threshold and goes all the way down. But in reality, there is other evidence that the body has protective mechanisms that cause an effective threshold, if you like.

The question that then leads to is, is there a side effect of those protective mechanisms that could be harmful in the long run. We know for example that if you have sub-traumatic damage to the body, the body will react say by producing a callus, for example, which you could consider a long term health effect. If we continually stimulate heat shock proteins, does that lead to some long term pathology? Or would we believe that some of the other protective mechanisms, if they are continually stimulated instead of just occasionally, could that cause a health effect? I would be interested in the views of the biologists and the physiologists.

DR. MORRISSEY: Great question. And Mark, building on Lindsay's point, one of the goal questions that we wanted to get out of this workshop was, what are the appropriate time conditions that we would consider acute and chronic, and whether there is a threshold or a falloff at 38 degrees, 39 degrees, or whether it is simply irrelevant, because the cumulative time would be impractical. It may not be a real big issue. If we can define what is an appropriate chronic exposure time, a lot of these end points may just fall off the chart.

DR. DEWHIRST: (Off Mic.)

DR. MORRISSEY: Right, but getting more specifically to the goal questions, for the standard, again, the standard has to be done with various parameters, because sooner or later somebody has to test something. So what is an appropriate chronic exposure time?

DR. DEWHIRST: Can't you take it around and look at it the other way and say what are the ranges of exposures that humans have, exposure times? Work backwards from that. What is practical in terms of workers or people who are around these kinds of facilities, and figure out how long they are exposed and then work backwards from that.

DR. MORRISSEY: Right now what has been done is what is practical in the FDA and in the laboratory for testing.

DR. DEWHIRST: I'm saying turn it around the other way, the clinical side. We do this all the time when I do work for the lab. We say, we are going to try a certain experiment, but the first thing we want to know in that experiment is how is it done in the clinic, and then we try and do it like that. So rather than coming up with an idea that has no clinical relevance, it is the same kind of idea. DR. DEWEY: I would like to make a comment going back in history. You guys are all too young to remember this, but we used to test atomic weapons in the atmosphere. About that time, there was an argument going on between Edward Teller and Linus Pauling. They could look at the data, very low levels of radiation exposure, strontium-90, and all the data that was coming in.

But the data was there indicating that there was a small increase, or could be a small increase in cancers or something like that. So these two gentlemen -- you know who Linus Pauling is, won the Nobel Peace Prize, got a Nobel Prize a couple of times, and Edward Teller, he wants to dig canals up in Alaska with A-bombs. So it is a different position. They could look at the same data and basically the data would say that theses low levels of radiation that people were getting around the world.

By the way, this led to the cessation of testing. Eisenhower was President then. He said we are going to stop. I think it was Eisenhower. But they had the same data. It was a very, very small percentage increase, I can't quote what the values were, but a small increase, let's say .00 percent or something like that. Teller, who wants to dig canals up in Alaska would say, don't worry about that, .00 percent of the people, that is nothing. On the other hand, Pauling would take this data and .00 percent and multiply it by the number of people in the world or something, that is a hell of a lot of people.

So it seems to me, you have got the same kind of concern, what is the probability of affecting a certain number of people, and how are you going to interpret it? How do you evaluate that? But that was the argument that was going on in this country and around the world in the late '50s, when finally Eisenhower -- didn't it happen when Eisenhower was President, they finally stopped testing, no more nuclear testing? It was based on those sorts of things. I think you have got the same problem.

DR. DESTA: Joe, I have a quick question. I think pH was mentioned as a sensitizer to hyperthermia. Is that a normal physiological change, or are there other things that would occur with the normal physiological parameters that would also affect the sensitivity to hyperthermia?

DR. DEWHIRST: Cells that are deprived of nutrients like glucose or amino acids also are greatly sensitized to heat killing, so a fairly large effect. The pH effect involves an acute intercellular drop of pH. It depends on how far you drop it down, so it is a complex relationship. But it depends on how far the pH goes down. It is an acute effect.

DR. MORRISSEY: I think Bill's story was right on. It is very difficult to assign hazard, assign risk.

One thing that we also lack is an end point. When you are dealing with radiation and cancer, cancer is a fairly reliable end point to gather. What we are looking for now is, tissue by tissue, what are the end points that we should consider.

I have heard behavior. I have heard astrocyte activation. Those are very good end points. How can we follow those up? What are cost effective ways to follow those up? Can they be done in vivo? Can they be done in vitro? Do they require animal studies?

Who else has other suggestions as to how to look at these parameters, or how they should be looked at?

DR. MILLER: I just wanted to first mention what Dr. Sharma mentioned. My perception, and I think he confirmed it or reconfirmed it, was that the brain in adults is post mitotic, so it is fully formed. So the only thing left for it to do is die. When you get to be up in age, you start thinking about that.

The Arrhenius equation is of course based on the Kelvin scale. This is somewhat personal, but I began to think, damn it, Arrhenius is working here. I had an occasion to have an MRI of the head, and the radiologist read it and he said, we don't see any problem at all. You have got the normal amount of gray matter for a man of your age. I said, what does that mean? He says, you lose as you go on.

I couldn't help but think, here we are. We have an active system. We are all losing gray matter with time. I begin to think Arrhenius is working here. Those long times that Mark put up, it is happening. So to me personally, it made some sense, what he was saying. It is personal and not scientific, but it is something we are all going through.

DR. REPASKY: I wanted to ask a quick question. What are the consequences of astrocyte activation? Does anybody know? If they become activated, what is the end point of that? What is the physiological response after astrocyte activation?

DR. HOOPES: I don't think people know entirely. A lot of the astrocyte in processes actually attach to blood vessels, so they can change, they can dilate vessels and do other types of things like that. But I think they are just a non-specific stimulated cell, that appears to be the most sensitive.

So when you look at them and they have been activated, the cell body is a little larger and the processes are more obvious. Like I said, they have these foot processes that looks like they attach to capillaries and blood vessels that may change the dynamic. But they seem sort of non-specific in a lot of ways.

DR. REPASKY: So it might not be a bad thing.

DR. HOOPES: Exactly. I don't know, if somebody said I want to activate all your astrocytes, do you want that, I would say I probably don't want that, but I don't know.

DR. REPASKY: But we are not sure.

DR. HOOPES: But we're not sure.

DR. REPASKY: If there was a way of getting more blood to my brain once in awhile, I would appreciate it.

DR. HOOPES: Yes, and certainly in traumatic injuries and things like that, in the peripheral areas where there is no hemorrhage, there is a lot of activation of astrocytes in those areas.

DR. REPASKY: But in terms of designing an end point, if we don't know what that means, I think it is important, because it is measurable.

DR. HOOPES: It is measurable, yes.

DR. REPASKY: But we don't know whether it is bad or possibly even good. I wouldn't want to be interfered with either, but still it might not be a bad thing.

DR. HOOPES: Exactly.

DR. DEWHIRST: How about doing cortical thinning?

DR. HOOPES: Do what?

DR. DEWHIRST: Just measure the thickness of the cortex. If you took a bunch of mice and put them in a hot room and left them there their whole life, compared to control mice who weren't in a hot room their whole life, and just see whether their cortex gets thinner faster. Just a thought.

DR. HOOPES: They don't have very many folia, I know that. There is not a lot of interaction to the neurons, it doesn't look like.

DR. MORRISSEY: But that is a great point, because the standards are based upon adverse health effects. Like we discussed in some of the sessions, we have seen effects that are not necessarily considered adverse health effects. So that is a great point.

I had always heard of astrocyte activation with inflammation and things. So there may be some corollaries there. But point well taken. We are basing the standard on adverse effects, so we need to try to have as much information.

DR. ZISKIN: Dr. Dewey brought up data showing chronic thermal tolerance. The survival curves then level off, they then continue on with temperature increase. Doesn't that in effect state that there is a threshold? If you have the chronic tolerance existing, doesn't that really imply that there is a threshold?

Thinking in terms of Arrhenius, where it just keeps going down and down and down and down, as Mort said, you are losing. But the body has so many ways of toleration of things like adaptation and so on, that we somehow are able to survive in our environment.

So I feel that maybe we are not giving the body enough credit for being able to adapt to these small temperature rises.

DR. DEWEY: Again, this is a cellular system I shall discuss, some work done by Bill Bedford, and we did some of that stuff in our lab as well.

Basically they looked at effects on cultures in terms of the chronic thermal tolerance. What happened, that curve goes down and flattens off, and then later it goes down again. What do you think of that? Chronic thermal tolerance is flat, and then down it goes.

You know what that was related to? The cells were held in G1 and finally they overcame that block and they went into S phase. If you have stationary culture, to keep them stationary without too much lowering of pH, it might just be flat forever almost.

But again, that comes back to the system. In that case it was going into some resistant G1 into sensitive S phase. We did some polymerase alpha and all sorts of stuff like that. It is a lot of fun.

DR. ZISKIN: But wasn't the concept, the heat shock proteins stabilizing, the proteins? If that is taking place, then there would be some justification for assuming that there would be a long term stabilization of cell survival at a slightly elevated temperature.

DR. DEWEY: Yes, that is very true. I didn't mention that in my talk. Fortunately Rosemary Wong brought it up, but many of us have done a lot of work with heat shock protein, HSP-70 in particular, and that does certainly protect the denaturation of proteins apparently.

So what happens after a long period of time with HSP-70 levels and so forth, you go on and on and on. If it is there all the time protecting or restoring damage, they think the big thing with heat shock protein is probably restoring the damage that is already there, I guess.

I think some of these questions have not really been addressed. If it lasted forever, obviously you would have some sort of a threshold, I guess.

DR. MILLER: I think Bill's point is well spoken, and Marv raised an interesting question. Is there also a threshold for the induction of heat shock proteins. If you have very low levels of stress, are there heat shock proteins coming along? I don't know.

DR. REPASKY: I think we have data on that, that you do induce heat shock proteins at 38 degrees. It takes longer, but they come up 24 hours after the heat exposures. It takes longer.

A question is whether or not the heat shock response at mild temperatures is even HSF-1 dependent. It may not be related to protein damage, in other words, at mild temperatures, but there is a heat shock response to fever, a range of temperatures, but it does take much longer to come up.

DR. SHEEHAN: I can say that heat shock protein is not only activated by heat or hyperthermia. It is activated by even injury and even cold. There are two or three papers already there. So that is only my addition to this comment.

DR. WONG: I need to clarify that when I mentioned the heat shock protein, there is a constitutive level of heat shock protein that is resident in your cells, in your body. With heat shock you are inducing additional heat shock protein, which then triggers a whole array of signaling pathways that are protective in terms of survival, for proliferation, et cetera.

So when I mentioned that there may be a difference in terms of the core temperature for the pig kidney, they may have a higher level of constitutive heat shock protein that stabilizes a variety of their proteins, so that when the additional heat shock we are challenging them in, then they are much more resistant to that.

I don't think there has ever been a measure in terms of the constitutive and inducible level for the different organ systems, animal systems, that we have used in the data that has been presented. So that has to be considered in terms of what model that you are going to choose to measure the guidelines in terms of the effects that you are going to be looking at.

DR. MORRISSEY: That is a great point. If we are going to look at threshold levels, certainly the system that we are using, the differences in the system, are going to --

DR. ELDER: I would like to ask Betsy's question that she asked about astrocytes. That is, what is the consequence of having heat shock protein induction? My impression is that it is a beneficial effect.

DR. REPASKY: I was just going to mention, when people set the legal limits for alcohol, one of the standards was dizziness and being unable to drive well, and a variety of other assays we all know about in terms of telling whether someone has had too much or not. That is because the cerebellum is unique in the brain in having very low levels of heat shock protein, and they are induced, thank goodness, because if they weren't induced, alcohol at a much lower level would kill even more brain cells.

So I think the idea of looking at the threshold level of proteins like heat shock proteins, which are able to be induced, because in that particular tissue, as Rosemary just pointed out, the cerebellum is able to protect itself very quickly from alcohol through this rapid induction of stress proteins.

But I was just thinking that there have been some standards set for some substances like alcohol based on whether or not the brain can protect itself or not. So that is in part the basis for the legal limit for how much alcohol you should be able to tolerate and still drive well, for whatever that is worth.

DR. DEWEY: Elizabeth, your comments about alcohol, I want to tell you an anecdotal story related to that.

When I was in Colorado, we had known at that time that alcohol can induce thermal tolerance. If you treat a cell with alcohol and wait a little bit, they are tolerant to heat. So alcohol induces heat shock proteins and makes them more resistant.

So I don't have to mention names, but a couple of guys were going to test this out in the hot tub. We had hot tubs in Colorado in those days, cold in winter. So we went out to the edge of town and a bunch of us got in the hot tub. One of these guys was going to see if maybe he couldn't get some exposure to heat and then somehow later do some experiment on his hand or some damn thing, looking at the effects of heat.

The problem of it is that all of a sudden the guy drinking alcohol, look around, and he had dropped down in the water. At hyperthermic temperatures, alcohol does much worse. So yes, it can cause tolerance, but it is a very interesting story. That was done experimentally by accident.

DR. REPASKY: (Off Mic.) Inappropriate expression of heat shock proteins is a very bad thing, but it is a very good thing if you are having stress that involves adult tissues, for example, such as heat or alcohol.

DR. MORRISSEY: These are all great questions. Let's bring this back around to the workshop and the goals that we set out, what are the health end points.

So far I have heard several things, the endothelial cells may be fairly sensitive, astrocyte activation may be a possible end point to look at. I heard some discussion about behavior versus pathology. I don't know that we ever really got to the bottom of that one, but it was a good discussion.

Does anybody want to summarize where we are? Does anybody think they can cap off what end points they think we should look at, just make a cut?

DR. DEWHIRST: I think this is very difficult, because it really depends on whether you talk about total body versus partial body. So if it is partial body, then you need to consider whatever organs are in the field, and maybe be specific for that.

Then you have got to consider what organs are in the field. So if you are heating the abdomen, you don't want to look at the brain, I don't think.

DR. MORRISSEY: For this hypothetical, for the brain.

DR. DEWHIRST: For the brain, okay.

DR. REPASKY: (Off Mic.)

DR. MORRISSEY: Not necessarily, but you can go with that. I know the brain is very complex.

DR. HOOPES: It just seems like from what we have heard, the two cell types that we focused on are endothelial cells and astrocytes, and they do make the most sense. They are the most responsive of all the cells in the brain. So one is the blood-brain barrier looking at things like interstitial pressure, minor edema, blood flow, and then astrocytes is just a general sensitive cell in the brain that seems like it is stimulated easily without known consequences. But it is fairly easily stimulated to have a morphologic change that can be quantified. That is important, too, to be able to quantify whatever we do.

Then the behavioral aspects, I think people are getting much more sophisticated with rodent behavior now, by doing these things and putting them in mazes and other kinds of things to quantify subtle changes in their ability to respond. So I think people are much better at that now.

DR. MC DANNOLD: From the report it looked like the thresholds for stimulation or suppressing neural function could be lower. Would that be an adverse effect that you would want to include? That could maybe happen before anything we would detect otherwise.

DR. MORRISSEY: What would your study design be?

DR. MC DANNOLD: I don't know if you would look at some of the stuff that is in the report, how they did those studies.

DR. DEWHIRST: They were looking at nerve transmission velocities and things like that. Those are quantifiable end points.

DR. SHARMA: This may be out of context, but so far we are discussing the basic effect of radiation, thermal dose or hyperthermia towards changes in the brain, and every model looks complex. So my only point is, coming from a society called neuroprotective organization, why not we should try to study in any model that we are studying some effects of good drugs that they are trying to attenuate or prevent it?

Anyway, we have so much limitation from animal model to human cases. Some good quality of drugs might be working, for example, drugs having neurotropic factors or a combination, or antioxidant. So everybody I saw, the results were presented fantastic, and they have a very nice model. So it is only my question regarding pharmacokinetics that we should also consider the influence of drugs that might be helping or alleviating some of the changes. Most of the clinical cases, people are taking sometimes some drugs, so that would advance our knowledge and may shed some new light on this.

DR. MORRISSEY: That is a great comment, Hari. Does anybody else have any comments? I think we have discussed this now for quite awhile. I think Jack summed it up fairly well for my understanding as well.

PARTICIPANT: I would just like to point out that from the point of view of regulation and protection, it is of fundamental importance whether or not there is a threshold. With ionizing radiation we have effects with a threshold and without. We have erythema and we have intelligence deficit, et cetera, from doses we can see, and then we have the assumed zero threshold for cancer induction. That governs the whole radiation protection regime.

When we go to EMR and NIR, the assumption, presumption, evidence, observation, whatever you want to call it, is the ease of threshold. Hence we use specific absorption rate, and we assume that that is what is important. In ionizing radiation we use integrated absorption. We use a different unit. In the one case we are interested in the integrated dose over life or whatever, in the other case it is just the rate, it doesn't matter. Once you get below that rate, we assume.

So it is really important. It is not just a little detail, it is fundamental to the whole protection philosophy. That is probably why -- the other reason of course is to be able to interpret experimental data, so you can look at how you are getting a thermal artifact.

But that is possibly why some of the standards people are so interested in this question. We are not really interested in killing 50 percent of brain cells as an end point. That is not going to be important for regulators. But it is really important if you extrapolate that back and there is an effect over the long term.

DR. VEYRET: That's right. It seems to be that there is a contradiction between the usual approach of ICNIRP and IEEE in saying that this is a threshold effect, that we want to avoid going over one degree C. temperature elevation. At the same time we learn here that the thermal effect that we assume to exist is based on a simple process which is protein denaturation with a known entropy change, which means that there is no threshold, in terms of dose.

At any temperature, if you wait long enough, then the effect will occur. If we are thinking about chronic exposure at temperature above nominal core temperature, then the effect will occur, in principle.

DR. MORRISSEY: These are great questions. I don't know that we are going to get to an answer today, because it does seem like we are going back and forth. We are at the point where we are going to hear from Dr. Per Soderberg.

What I would like to say about lunch, we are going to shift now. We have taken a number of notes, Joe and I, so we will try to distill that. We will probably send them back out to the experts for their comment and critique. But we are going to shift now and move towards the eye, for a 15-minute talk. Then we are going to go get lunch, but today it is going to be a true working lunch.

So what I would like you to do after the talk, we are going to break. We are going to go out and get sandwiches, come back in, and we are going to have a discussion on the eye, similar to the brain but on the eye, and we are going to try to address end points on that tissue as well.

Agenda Item: Entire Group 4B: Effects of Heat on the Eye

DR. SODERBERG: I will try to talk a little bit about effects of heat on the eye. To those of you that would like to see effects from very low temperature rise, I will probably have to disappoint you, because I will not discuss that.

I think it makes some sense. I am from the optical radiation field. The experience from that field is that very subtle temperature rises don't really matter in the long run. Anyway, I am from Uppsala University, from the Gullstrand Lab. I am also heading the optical radiation subcommittee in ICNIRP.

I wanted to start by just reminding you from a heat point of view quite complex anatomy of the eye. The eye is partly exposed directly to the air, partly covered by the eyelids, and looking at the anatomy of the eye itself it is again very complicated from a heat point of view.

The eye is essentially like a football with a transparent window at the anterior surface, without any vascular supply, but with a tear film on the anterior surface that may evaporate. Then the transparent window continues around and closes up the eye to the exterior.

On the inside of this sheet there is a vascular membrane which we usually call the uvea, which consists of the iris, the ciliary body and the choroid. On top of the choroid like a wallpaper sits on the retina. You can see that these structures have two different vascular supplies, the superficial surface of the retina supplied by its own vascular supply coming from the nerve head, while the deeper part of the retina is supplied from the uvea by a parallel blood flow.

If you look at the uveal blood flow, it is a network of vessels that become capillaries, and they are then drained out through the vortex veins. If you look into the eye, you will see a reddish appearance. What you are seeing is the blood flow in the choroid right through the retina, which is essentially transparent in the normal eye. But you also see the retinal vessels that come out of the optic nerve head and distribute blood on the surface of the retina, which is then taken back through the nerve head out to the eye.

If you make a cross section of the retina, you will find that the outer portion of the retina is a non-vascularized tissue, nourished by the capillaries of the choroid. The inner parts of the retina are vascularized tissue nourished by the retinal vessels. These vascular supplies are quite different. The retinal vessels have an autoregulated flow, and the uveal vessels have probably no autoregulation. There is some question mark about that, but probably no. But the flow of the choroidal vessels is on the order of ten times that of the retinal vessels, making it the highest blood flow in the human body probably.

So what happens if we apply heat to the eye? How does that heat get out to the eye? If heat is implanted in the eye, the heat that is loaded into the cornea to some extent disappears by evaporation. This of course conducts the heat to tissues adjacent to the eye, and very important for the posterior part of the eye, there is convection. There are two ways that the heat may be convected out of the eye, the retinal circulation, but more importantly the choroidal circulation through the vortex veins.

What kind of thermal damage do we get in the eye? These are intentional heat exposures to the cornea, created to change the shape of the surface of the cornea. You can see that the heat exposure has created cloudiness spots in the corneal surface. If you look at that in the microscope, what you see is shrinkage of the collagen, which is a major protein of the cornea.

Joe asked me to have some opinion about end point for damage and some other things that we have discussed for other tissues. I would argue quite strongly that the end point for damage in the cornea is simply clouding. You may actually be interested in looking at spatially resolved specular reflection, because that is an indirect way to look at the refractive properties of the cornea, which are of course important from an imaging point of view.

We can define acute and chronic exposure. I take the clinician's very simplistic opinion, saying that acute is for me probably below one day and night, and chronic is above. But this may be discussed.

When should we look for this threshold? I think the optical radiation experience is that it is only relevant to look immediately after the exposures. Do we need more research? I think we do need more information about how the ambient temperature interacts with the threshold.

If we now walk into the eye, this is an image of an iris, where heat was applied on the iris with a laser. In this case quite high intensity, so vaporization is formed, but you can see that there is also a whitening of the iris tissue. If you apply enough energy, you can drill a hole which is used in the clinic.

What is the proper end point for eyes damage? Again, I would say coagulation of the iris or whitening, visible whitening. The definition of chronic and acute, I have the same opinion. What is the time for observing the threshold? I would say only immediately after the exposure.

Is there any need for research? In fact, there is not so much data on this. I think the iris has not been considered a very important tissue. If you get a small scar on the iris, that is not a big deal.

If we then walk behind the iris we have the lens. The lens is a non-vascular cellular tissue with a cellular turnover, but with the particularity that the cells are stored throughout life inside the lens. So the lens is I think the only organ in the body that grows throughout life.

What is the appropriate end point? I would say clouding, which is cataract, which destroys the visual function of the eye. Same opinion about acute and chronic. What is the appropriate time point to look for the end point? For intense exposures it is of course immediate, but for the lens it has been found epidemiologically that even slight temperature rises may on a long term basis cause cataract, which is known as glassblower's cataract. Is there any need for further research? I do think we need more information about these long term effects.

Let's now walk into the retina. I showed you a picture before of the retinal surface, which is usually reddish. Here heat has been applied intentionally. What happens is that white spots are created in the retina. This is simply coagulation, of course, of the retina.

If you wait sometime, these coagulated spots will fibrotize and pigment will agglomerate in clumps around these coagulated spots. This is a response to the trauma.

What is the appropriate end point for damage? In the optical radiation theme, many things were tested, but we actually ended up using directly observable damage. The current safety standards for optical radiation are built on directly observable damage, where you look ophthalmoscopically into the eye.

Recently we have had a new tool, the optical clearance tomography. I don't think there is enough experience of that yet to say if that will add any valuable information to end points.

If we look at time after exposure, the experience from the optical field would indicate that observation less than a week is absolutely relevant. Do we need more research? There is one problem that we still have, and that is multiple high frequency exposures with the intermediate spot sizes, where we need more information.

That was what I was going to say. Thank you very much.

DR. MORRISSEY: With that, what I would like to do is break for lunch. We will all go out and get sandwiches. Then we will come back and we will have a discussion.

(The meeting recessed for lunch at 12:12 p.m., to reconvene at 12:25 p.m.)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A F T E R N O O N S E S S I O N (12:25 p.m.)

(Presentation by Dr. Hirata.)

DR. MORRISSEY: With that, let's feel free to open this also up to discussion. Dr. Hirata, with regards to the tearing and the evaporative heat loss from the eye, how effective is that? How much heat can be dissipated? Do you take that into account? The tearing in the eye.

DR. HIRATA: During the microwave exposure, in some cases the surface of rabbit eye is cooled down by eye tearing. One degree Centigrade or more are very rapidly changed in the region of the cornea, but in the lens it is not so affected by the eye tearing.

DR. CHOU: In addition to the tearing, the anesthesia, all that can change the outcome, depending on the temperature. Also, that is one of the reasons we discuss here, should we use SAR or temperature. It is also relating to the study from the University of Washington. At that time we were able to demonstrate that it caused cataract very easily in the rabbit eye with about 41 and a half degree cause cataract.

But you can do the same thing with the same exposure. You drop the body temperature of the rabbit, put the rabbit in the ice bath and lower the body temperature. With even a higher SAR you don't get any cataract because your baseline temperature drops. So whatever determines the damage is the absolute temperature to cause the damage.

That just gives you another example of SAR. Really what is happening is, the temperature is causing the damage, not the SAR.

DR. MORRISSEY: Certainly what I was getting from your talk was that temperatures of at least 40 degrees or so are required for any detectable damage. Am I correct in that? Temperatures of at least about 40 degrees or so.

DR. HIRATA: Yes.

DR. MORRISSEY: So those are fairly significant. Would you say that lens opacity is the most appropriate end point? You had looked at retinal damage, damage to the cornea, others. As far as microwave insults in the one or two gigahertz range, you are getting penetration throughout the eye, but what is the most sensitive?

DR. SODERBERG: If you look at whole exposure of the eye, I would probably agree that the lens is the best detector, because it is closed up in the system and doesn't have the cooling of the retina with the blood vessels, very little of the evaporative effect. So I think the lens is probably a good detector.

DR. MORRISSEY: I think that is what we have been basing it on, right, C.K., is lens opacity. So not a lot of surprises there. It is good to know that the ceiling is at least 40 degrees or so.

DR. CHOU: The other thing to point out, when you deliver -- to the eyeball, you also damage the surrounding eyelids instead of the eye. Similarly in the University of Washington experiment, we showed the same thing on rabbit. You can cause cataract very easily in the rabbit.

We did the same thing on our monkeys, and could not produce any cataract without burning the nose ridge and the eyelids and all that around the eye and causing thermal injury without causing any cataract.

DR. MORRISSEY: C.K., you were talking about the facial structure of the rabbit. I remember that as being a significant issue. Bernard, ICNIRP I guess has the same basis for the standard of the lens opacity?

DR. VEYRET: With the ICNIRP we have to look at the science again when we revise the guidelines. But in the Blue Book there is a short description of the effect on the eye.

DR. MORRISSEY: I think it is a fairly straightforward threshold, lens opacity and 42 degrees. Does anybody else have any questions on the eye, or comments or questions?

PARTICIPANT: I was out a few minutes, this point may have been addressed. I am just curious, what is the normal or ambient temperature of the lens as compared to the back of the eye? And is there a normal temperature gradient from the back of the eye toward the cornea?

DR. SODERBERG: Yes, and the temperature of the lens is strongly dependent on the ambient.

DR. MORRISSEY: Is it dependent on the core temperature as well, or is that just the heat sink?

DR. SODERBERG: That I cannot say.

DR. MORRISSEY: If there are no more questions, we will just continue with our lunch. We have got about 20 minutes or so. At 1:15 we are going to start with our breakout sessions. The discussion on behavior will be in the Rockville Potomac Room, Dr. Kyatkin. The discussion here will be Gary. You will do this discussion, and make sure that we can get your slides.

Agenda Item: Session 6: Group Discussion

DR. MORRISSEY: When we started out yesterday, the goal was to try to gather information to facilitate the evolution or the strengthening of the RF exposure standards. And certainly there is a lot of interest in these standards now because of mobile phones. They are used as a compliance requirements for mobile phones, so clearly there is a lot of interest there.

Also, we made the case that there are a lot of medical devices, imaging, ultrasound and MRI has a very hefty RF component in some cases. The terahertz imaging; we will all go home at the airport and we will stand in line and we will wonder, would I rather just squirt through here and not worry about someone seeing me in my skivvies, or do I really want to go through all the long lines and the checks. That may or may not occur, but terahertz technology is another form that forum that falls within here.

We have been very lucky to have a lot of expert hyperthermia scientists help engage us and provide insight. I think we went through a lot of great talks.

I am going to just summarize, if I can. I was not at every session, so I may have to enlist your help, but I want to make sure that I have captured the essential parts, or at least the gist of what went on in the different sessions, including some of the breakout sessions that some of you may not have been in.

The first session that I was in, and Dr. Christine Wang from UCSF had a shuttle to the airport, the first breakout session was Dr. Christine Wang, who reported on the testes. As we know, the testes is a fairly thermo sensitive organ. We can get decreased sperm count and sperm concentration, something that we want to protect. In fact, this was one of the discussions in that breakout session.

Since this is in many cases a transient effect, decreased sperm count is often a transient effect at 43 degrees and can recover, is this an adverse health effect? The standards are predicated on adverse health effects. We all agreed that it was. I just throw that out. Would everybody agree that a decreased sperm count would be considered an adverse health effect? So we agreed that that was an adverse health effect.

Most of the studies involved looked at about 43 degrees, which is certainly above perhaps the threshold. There were some human studies that looked at say a degree or two above ambient, either putting the testes in the inguinal canal or hot water immersion. But it took at least a degree or so for these effects to occur. So the threshold that was agreed upon was about a degree or a degree and a half. So energy that would be required to increase the testes temperature a degree to a degree and a half is about the threshold for the testes. So that was fairly straightforward.

The process of germ cell death in this case was apoptosis. Dr. Wang went through a lot of nice studies looking at the cascade, the phosphorylation, the BCL-2 and the fact that it was a cytochrome C release from the mitochondria that activated the capspace cascade into apoptosis, and not death receptors or paroophoron granzyme type granular release, but mitochondrial mediated apoptosis.

I believe that was pretty much the long and short of it. There have not been studies that looked at the minimal amount of exposure. Most studies have been done for at least 30 minutes or so, many for days or weeks when you get to a degree or a degree and a half to look at the effects.

So that was the information from that breakout session that will feed into the standard, that we are looking at about a one to a one and a half degree increase in the testes and germ cell death by apoptosis as the end point.

The next session, I was not at the cardiovascular system session, but can I ask someone to comment? Mort, how did the cardiovascular session go? Were there any health end points? I'm sorry, Gavin, for the cardiovascular session. I know I am putting you on the spot here.

DR. DONALDSON: The cardiovascular reported that the body's changes for what we would expect to see in terms of changes in blood pressure and heart rate, et cetera. I don't think there has been any recent work which has changed our opinion of that.

I think the extra stuff may be to do with heat related stuff. The threshold temperatures when people change and vary by country are depending on whether you are traumatized in a hot country or cold country. So if you are worried about additional exposure due to RF, it a little bit depends on where you are. You might be paradoxically less susceptible in a warm country than you might in a cold country.

Maybe Chris has got something to say, because you were also --

DR. GORDON: People climatized to a warm tropical climate probably might not respond to a heat load behaviorally like someone in a temperate climate, yes.

DR. DONALDSON: I think the other thing is, from our perspective of whole body stuff, the power loads that people are talking about, we don't consider significant effects on the human cardiovascular system. The amounts of energy you have to put in has to be a lot greater.

DR. ELDER: I would just like to add a short comment. Ken Foster asked the question about whether or not we could use the standards that had been developed, the recommendations developed by ACTIH, which limits worker exposure to one degree, to 38 degrees C. I believe your response was that ACTIH had done our work for us. So his recommendation was yes, we could look at ACTIH.

DR. MORRISSEY: In several of the other sessions, I think what was boiling out of that was body temperature. This is not considering hypothetical scenarios of the pregnant woman on the desert island that is really hot and humid, and she is doing jumping jacks. But for most people, temperature rises of a degree or two, up to 39, but certainly 38, did not overly stress these systems. Would that be what you would feel with the cardiovascular system as well? Good.

So we went through the cardiovascular system. The next breakout session was on children. I thought this was very fascinating. Michael Bergeron did a really nice job.

What we tried to do in this workshop as many of you know was bring in people that had not sat through our meetings year by year by year, get some different perspective, and I think it was great. Michael has done some really nice work on children, albeit children eight years and up, but looking at exercise, performance physiology.

In fact, some of the dogma that I had assumed was true I think Michael challenged quite a bit. I think we all assumed that children are a little bit more prone to hyperthermia or have a little less ability to dissipate heat. In some cases that is true. Michael, jump in if I say something wrong. Upon good hydration and factors like sleep and a reasonable fitness level, Michael made the case and provided a lot of references for the fact that children aren't as thermally regulatorily challenged as we might think. They are fairly capable of dissipating heat, and we shouldn't look at them as a population, at least from the eight and above population, as something that we need to overly protect.

Certainly we need to protect children just because they are children, but they do have a fine capacity to dissipate heat. So that was very interesting.

As far as end points to look for, cognition and neuromuscular control, Michael went through some studies on that that he had looked at. From a perspective of looking at end points that might start to falter upon excess heat, cognition and neuromuscular control were suggested as potential end points.

Again, even up to 40 degrees may be theoretically tolerable, but certainly the 38 to 39 degree range would be something that most fit hydrated well-rested children should be able to tolerate for reasonable periods of time.

DR. BERGERON: Thank you, Joe. Yes, what you said was correct, that there doesn't seem to be a maturational disadvantage from eight years and up. Certainly any healthy well-hydrated, fit, not overly burdened with clothing or protective equipment, and not working out at an unusual intensity or insufficient rest periods between bouts can tolerate quite a bit.

But again, the main point of the session was that there wasn't a maturational disadvantage. But there are a number of factors we tried to focus on (recording interruption). There are a bunch of other clinical conditions, whether it is type two diabetes or cystic fibrosis, obesity, a number of factors that again would modulate a body temperature response to exercise and such.

What we don't know, and we had a lot of discussion on this, we are looking at thermoregulatory or thermal strain response to passive and exercise heat exposure. We don't know the differences in bone metabolism, let's say, during rapid growth. There may be a particular vulnerability there, we don't know, or there may be a particular vulnerability to the skin itself, I don't know; no one has done that work, or to the brain itself.

So I think we want to be careful with how we interpret where the discussion went, that kids don't seem to have a thermoregulatory disadvantage, at least down to eight years old. But that is as far as it goes.

The other point was that there is a host of other modulating factors, if you are going to look at whole body heat exposure. You need to consider all those other things, too.

DR. ZISKIN: I did not have the benefit of participating in your session, but from a medical viewpoint, the concern that I would have is in the infant and the neonate, where they are known to have an inadequate thermoregulatory system. This would be the concern that I would have, rather than in the eight-year-old.

DR. BERGERON: Right, and we had that discussion, that we need to be careful that the data that we have reflect eight years and up. Certainly there is probably a greater risk at even four years and below, let's say.

DR. MORRISSEY: In addition to Mike, we did have Dr. Hirata, who gave a very nice talk with the data that he had on children through three years, the three-year-old. Just summarize it.

DR. HIRATA: From my computational results, for the same SA, whole body average, the power of absorption, the child can dissipate the heat, because of a large body area to mass ratio. So the temperature for RF exposure probably be smaller than the adult.

The child's thermoregulatory response is not good compared with the adult. But from our computational fitting, they are almost comparable with each other. Thank you.

DR. BERGERON: The important point with that surface area to mass advantage that was just described was in a fairly moderate environment, so that the gradient between skin temperature and the environment was fairly large.

Theoretically it hasn't been shown, but you can imagine at very high temperatures, where all of a sudden there may be an absorption of heat from the environment, that advantage may go the other way. That has not been demonstrated, but thinking about it logically, theoretically it would play out that way probably.

DR. MORRISSEY: Right, but at modest heating loads, the child, although his thermoregulatory system is not mature, he has the advantage of having a favorable surface to volume area, which helps him to dump heat by conduction and convection.

DR. BERGERON: And sweating may be more efficient, too.

DR. MORRISSEY: We talked about that as well. While the mature adult can sweat liters and liters, the fact that children, even though their sweating capacity may not be as robust, the fact that it doesn't just fall off immediately, but is dissipated or more evenly distributed, I guess, can in some cases be even more beneficial. So that was interesting.

We did work more towards thresholds, looking at this 38 to 39 degree whole body levels that weren't necessarily to cause alarm in healthy children.

We then had a talk by Betsy Repasky that I thought was very interesting. Everybody was here, but do you want to just summarize it again in a couple of sentences?

DR. REPASKY: I think there is a great unexplored area of natural responses to temperature that relate to immune regulation. From our work and from a group of individuals that is working in this area, it looks like there is thermal sensitive signaling events which potentiate the immune response in response to not only fever, but also inflammation.

Inflammation is not generally considered systemic. It may be quite local in the skin or in some region of the body. One of the cardinal signs of inflammation from the ancient times has been calor or heat. When you feel a red bump on your skin, for example, it feels warm, but that is because the rest of your skin is quite cool.

There seems to be a necessary relationship between elevation and temperature, whether you are starting from 28 degree skin or a 37 degree core temperature, that appears to be strongly correlated with survival in nature from infections. Those thermally sensitive steps in the immune response really are just now starting to be identified at the molecular level.

From my own research, we are interested in lipid and membrane protein organization. The data that I showed indicates a fairly rapid, within two hours you can see by eye effects in terms of reorganization of membrane, lipid domains and proteins.

I suspect that with more sensitive assays, these changes could occur much sooner than that. But over the course of several hours you can realize a very significant change in the organization of proteins, which are similar to the reorganization that occurs upon full activation with antigen, from a lot of in vitro studies.

DR. MORRISSEY: That's fine. What I gathered from that talk with regards to the workshop goals is that it may not be a minimal threshold, but many of the observations you had were at a degree or two, and they took at least two hours, if not the more robust effects took about six hours to occur.

So with regards to the standard, when we are looking at whole body temperature increases of a degree, and perhaps looking at aberrant stimulation of the immune system to some of these autoimmune or hypersensitivity reaction types, 38 degrees for shorter periods of time might not be as problematic.

DR. REPASKY: No, or 39.5. There is a lot of literature now using hyperthermia in cancer therapy. As far as I know, there isn't evidence for potentiation of autoimmunity, and certainly not in our own work.

DR. MORRISSEY: Thank you. Then we went to dinner, which was nice, came back and heard a great talk by Bill Dewey. Every time I hear those foundational talks, you always pick up something more. You are always learning something that you didn't learn the last time.

Effects on the nervous system. This was complicated. I don't pretend to have absorbed it all. But what I did understand is that a common thread seemed to be that with regards to pathology -- and of course there is an issue with behavior and pathological end points, which is the most sensitive, or which should the standard group be looking at, or both -- but with regards to pathology, what I understood was that endothelial cells may pose some of the more sensitive cell types. Hari, does that sound reasonable, that the endothelial cells may present some of the more sensitive cell types in the brain?

DR. SHARMA: I believe so. I believe that endothelial cells are highly vulnerable to any insult to the CNS, including temperature. After that, astrocytes are also equally sensitive. Neurons probably comes third.

DR. MORRISSEY: Right, exactly. When we were in the behavior session with Eugene Kyatkin, it was a similar type of conclusion, that the endothelial cells and the astrocytes seemed to be -- if one was looking at a cell type that was the most vulnerable to heat insult, those would be the cell types to look at.

But neurons may be more refractory to frank temperature, but they are not replaceable, so there is that issue. But if one is looking at the most sensitive end point for temperature, endothelial cells and astrocytes.

We had a lot of discussion on methods to pursue this. I don't want to go over all of them; it is going to take me awhile to digest it, anyway.

Hari and Nathan, did you want to say a concluding sentence? Remember, what we are trying to do here is, we are trying to set standards, so end points and thresholds, as close as we know. If there are none, you say there are none, but understanding that what we are trying to do here is, for the brain we are trying to identify target end points and maybe temperature thresholds, what you distilled, not only from your talks, but form the discussion. It is okay to say we don't know, but try to get as close as we can.

DR. MC DANNOLD: I think that we can go look at the hyperthermia literature and get an idea of threshold. But the absolute temperatures we are talking about are much higher. We know there are metrics we can use, the thermal dose, to try and extrapolate that, but we have to take into account things like thermal tolerance. The data as far as I know is just not there at these lower temperatures that we are talking about here.

DR. MORRISSEY: That is what I gathered. We did a lot of brainstorming and we set some foundations, we identified some target cell types. We identified perhaps some ways forward, but we didn't have a lot of concrete conclusive evidence at the lower temperatures. So that may be an area that perhaps we need to dig into the hyperthermia literature even harder, or perhaps do some legwork, do some wet bench work.

Following that, we had effects on the eye, which I thought was very good, and again thank Dr. Hirata and Per for helping us with that.

That was fairly straightforward. Per, do you want to just summarize that? You understand the standards, so target end points, thresholds.

DR. SODERBERG: My conclusion was essentially that the thermal load that RF from mobile phones will provide is not enough to cause eye damage. Apart from that, it seems reasonable to use the lens as a detector for heat damage, because that is probably the most sensitive.

Now, when it comes to calculating heat relaxation of eye tissues, it is extremely complicated. It is very variable. It depends on the curvature of the eye, the eyelids, and there are many different factors that have an implication.

DR. MORRISSEY: We also had a talk by Dr. Hirata, who did some SAR temperature mapping. Just a sentence or two, just to summarize the main conclusions.

DR. HIRATA: The current international guidelines on standards are based on guide work. Microwave induced cataract is formed with intense microwave exposure.

However, in that study a systemic anesthesia was applied. Then the thermoregulatory response was not activated. Based on a recent study, no cataract was found under the condition without anesthesia due to the increased blood perfusion rate due to the temperature elevation. For minimum wave exposure, fissure binding was observed, but no ocular injury was observed. Thank you.

DR. MORRISSEY: Thank you very much. We also had a talk by Rob McIntosh -- that was on the first day just before we went to dinner -- on SAR temperature correlations. So go ahead and give a summary of that.

DR. MC INTOSH: Thanks very much. Apart from all the technical aspects of modeling, some of the outputs from it, I guess the main one is the blood flow, how critical blood flow is. The higher the blood flow, the effects of RF are going to be very low, in particular in tissues like the brain. If you put in a field say at the limit of public exposure, you can get temperature changes in the brain of only about .02 of a degree or .03 of a degree or something of that order. So in that regard, those sorts of effects of RF are quite minimal.

The most complex organ in the body as seen by the modeling is the eye. You have got again the lack of blood flow by the elements of the eye, like the lens and the aqueous humor and vitreous humor, which lead to the higher temperature changes happening in the eye. I think Dr. Hirata calculated temperature changes of about .3 of a degree. So that would definitely be the one to consider the most in terms of what the modeling showed.

DR. MORRISSEY: Thank you, Rob. I thought that was very interesting, and emphasized how far we have to go, not only with the biology, but modeling dynamic blood flow and getting more accurate numerical models.

So we went through the nervous system and the eye. We then had a session on the skin. I was not in on that session, but Gary, do you think you can give a sentence or two, the main points of any targets and thresholds?

DR. FISHER: I think the main points are that we did not observe adverse molecular changes at elevated temperatures of 38 degrees for one to two hours. We did observe adverse molecular changes at 41 degrees.

So I think that is about all we can say about that, within the range of 60 to 90 minutes, 41 degrees elicited adverse responsible, 38 degrees did not.

DR. MORRISSEY: That is exactly the kind of information that is needed.

Again, I wasn't in the session, but is skin fairly uniform in terms of its response, the areas of the skin, the different regions?

DR. FISHER: (Off Mic.)

DR. REPASKY: I think there are some differences, but just as you have indicated, the palms, the hands and the wrist, and actually the face is very sensitive as well. There are differences in TRPU receptors throughout the body that make skin either more sensitive or less sensitive to coldness or warmth.

DR. MORRISSEY: Gary, since I wasn't here, what was the adverse effect at 41?

DR. FISHER: The end points?

DR. MORRISSEY: Yes.

DR. FISHER: The end points that we measured had to do with adverse events in collagen metabolism.

DR. MORRISSEY: Which brought us to the session on behavior, Eugene Kyatkin, which I did attend. Again, very interesting. I think a lot of the individuals within the standards group held to the traditional dogma that the brain was held at a fairly constant temperature. It receives about a third of the blood flow. One would assume that it is held at fairly constant temperature. The body certainly fights to try to thermoregulate its core and its brain. But there were exclusions that could occur with normal activity of a degree or two, not only in animals, but in humans. Eugene, you want to give just a few sentences on the discussion?

DR. KYATKIN: I tried to emphasize first that it is very important not to assume about temperature which are occurring in some organs, but to record the temperature. Especially it is important with the brain. When we start to record with high resolution in freely moving animals, we can see many unexpected things. Dogma will not stand.

The second point, I tried to demonstrate that normal types of physiological activity of animals, at least animals, and we have some proof that something similar could happen in other organisms, accompanied by pretty robust changes within three degrees Celsius of brain temperature. For example, in sexual behavior and peak of male ejaculation, temperature in rats can reach 39.5, with some animals going a little bit higher. It is accurate measurement, and it was the highest levels in physiological condition.

The third thing that may be important for our meeting is that there are specific situations and conditions which make the individual very vulnerable to even small thermal impact, high physical activity with inability to lose heat, combination with drugs, combination of drugs with activity, plus small thermal impact. This is the most important point.

DR. MORRISSEY: That was great. I think one of the most beneficial things that came out of that session was, we discussed a lot of John D'Andrea's old work. The current standard for RF exposure is largely based, at least the whole body standard, and to some extent the local limits, but we are working towards the local limits, improving them, but the whole body standard is based largely on work that came out of Brooks Air Force Base, John D'Andrea in the back there, and several of his colleagues, on work stoppage in animals including monkeys that received radiofrequency exposure resulting in a thermal load that resulted in about a one degree C. temperature increase. That is what the standard has been predicated on for quite some time.

I think after lengthy discussion, we felt that this was an appropriate end point, an appropriate adverse health end point, behavioral work stoppage. We confirmed that we were on fairly solid ground.

There were some suggestions to perhaps try to replicate these studies in humans. John, do you want to say anything about that, or conclusions of the meeting in that regard?

DR. D'ANDREA: I have written a couple of papers summarizing behavioral work stoppage type literature. Much of that work was done by my colleague, John DeLorge. Ellie Adair did some of that work, and others. It seems to be a fairly consistent outcome in three different species. Those were some of the elements that allowed the committee to use that as a basis to support a standard. The effects on behavior were robust at four watt per kilogram, which was the threshold.

The only thing that is really missing is something that Ellie Adair has requested over the years. That is to look at the human in this regard. The thermoregulatory capability of the human far surpasses the animals used in those behavioral studies, and perhaps even behavioral studies with humans exposed to RF, much in the same way that Ellie Adair conducted her physiological thermoregulatory studies, would be of probably great benefit to setting the safety standard.

DR. MORRISSEY: So that was the meetings. I think we have a lot of information now to move forward and bring to the IEEE standards group. I think the ICNIRP representatives may get value out of this information as well, to help to move forward those standards.

I am going to try, with the permission of the presenters, I would like to try to put all the presentations up on the Internet, on an Internet site. Would that be okay? I can certainly provide a password, or I can make it public. Passwords are kind of annoying, but I can certainly understand if you would like to password protected. But is that okay with all the presenters, to make these presentations available at least to the group here? Good.

So I will do that. I have all of your e-mails. I will send out a final e-mail in the coming weeks, giving you the address of the website where we can download these presentations.

I want to thank you all for coming. I think it has been a very useful workshop.

DR. ELDER: We didn't talk very much about cancer, but I believe one of the outcomes of the 2003 publication in the International Journal of Hyperthermia, which was the proceedings of the first workshop earlier, was that in reference to cancer, heat has not been shown to be a carcinogen.

So the question I would like to ask is, is there any data that has happened since that workshop to this time that would change that conclusion?

DR. DEWHIRST: The answer is, I don't know. That was not part of what I was charged to do in the review. Probably it needs to be looked at before we come to a conclusion. But at the time of the original review, it was not known to be a carcinogen, except in circumstances where it is combined with a known carcinogen; it could accelerate carcinogenesis. But by itself, there wasn't any evidence that it was. Probably I could get one of my students to do a quick Medline and see what we can find.

DR. ELDER: Just an additional comment relative to what you said about, the heat may enhance the carcinogenic process. In the radiofrequency studies with animals that I am familiar with, where the animals had been given a carcinogen, the exposures up to four watt per kilogram have not been shown to enhance the carcinogenic process.

DR. OHKUBO: I have a short comment regarding the blood-brain barrier function due to the exposure to the RF field.

Our model has been -- five or six CS to evaluate the RF exposure effects on the blood-brain barrier by using the -- microscopy which utilized the closed cranial window for more than one month. The study is still ongoing with Professor Bernard. We are thinking about the threshold of exposure levels.

For instance, even in more than 10 brain SAR exposure at the 1.5 gigahertz, which induces more than 40 Centigrade in the surface of higher micro circulation, we failed to get a disruption of the blood-brain barrier. So it is very unexpected -- has confirmed by several doses using the cranial window method. The result will come out soon. The preliminary result has already published in Journal of In Vivo in 2007.

DR. FOSTER: I just have one comment. I think you deserve a round of applause for organizing this, and just being very effective.

DR. MORRISSEY: Thank you very much, Ken. Thank you, Chiyoji. I want to thank you all for coming. It has been a very useful workshop. There has been a lot of information.

Mort, I forgot the teratogenecity. No, you can't escape.

DR. MILLER: I thought I was going to get out of this.

DR. MORRISSEY: No, no.

DR. MILLER: What struck me about the meeting was the vast difference in what we were addressing in terms of health risks and everybody starts off except the infant with obviously no apparent health risk, except that the infant that is born, there is a ten percent chance that it is going to have a defect. That is the tip of the iceberg.

The interesting part is that 30 percent of all pregnancies are aborted due to defects. The body has a system for getting rid of things, and it does, but it doesn't get them all.

Anyway, I was thinking about this, because I did have some conclusions, but I reminded myself that we had 310.15 Kelvin if you are at 37 degrees. That temperature, there is an average distribution of energy, but there is also this Maxwell-Balsman distribution and activation energies are possible at higher levels. They are certainly available at Bill Dewey's 265 kilocalories per mole.

But something is happening, because we are getting bad results right at the start when the child emerges. Ten percent of them are defective. It seemed to be to fit with the Arrhenius equation. That was the hypothesis that I was testing, that any temperature for any time has an effect, and the problem of threshold seems moot because it is occurring at background levels.

That is about the sum and substance of what I had to say. I did have a conclusion slide. I think I have said what I wanted to say. Thank you very much.

DR. MORRISSEY: Thank you all for coming. I did learn from Bill Dewey, you always have these gems, I think a lot of us assumed this dogma of a threshold at 38 or 39 degrees. I think we have now brought that into question with the Arrhenius extrapolation, and that is certainly something we need to keep in mind.

With that, I will adjourn the meeting. Thank you all for coming, and have a good trip back. Make it through the security okay. You are responsible for confirming your shuttles with the phone number that I have placed up there, because I won't be at the hotel much longer. Thank you.

(Whereupon, the meeting was adjourned at 4:05 p.m.)