Thanks, it’s a pleasure to be here.  I think one message from today’s meeting would certainly be that the management of these problems is multiple different teams and those teams are critical to have the success with patients both from the surgical option the radiosurgical options and endocrine management and these are fascinating problems and the best results are going to occur when all these options are available for our patients.

So the first part I’m going to talk about is nonfunctional pituitary tumors, hormonally inactive tumors, that’s the  more common of the types of problems that we’re asked to get involved in and we also have a team of  people with whom we’re working and several of our Fellows help to put together this data, Dr. Park, Dr. Mew who’s actually here in the room. I’d  like to thank them for their help in this as well.

So the goal of this is you want to get rid of the tumor or make it inactive, stop it from growing as we’ll see in the hormone active tumor we want to stop hormone production and try to maintain the pituitary function.  And together it’s the combination of  all of these things that is the best result.  A  lot of people get confused about what is this all about and we broadly lump surgery, medical management and radiation all into the same portfolio.  But really there are different forms of how radiation is delivered and what’s commonly thought of as radiation therapy which is done with multiple different treatments over the course of say 25 to 30 fractions is widely different, wildly different from what we’re doing with radiosurgery.

And while this represented a standard of management for patients in years past when really not outcome data existed related to options such as radiosurgery, many patients ended up with similar types of improvement perhaps but the cost was much higher in terms of delayed hormonal function loss and I’ll show  you some of the data related to that.

The concept of fractionated radiation therapy is it’s being done over the spreading of time and it’s being done in order to reduce morbidity, not being done to improve response rate of the tumors that are treated.  It’s sort of based on the concept that if we’re giving everything in the same general area the same dose, tumors that have a higher cell division rate than do late responding normal structures such as the blood vessels, brain and cranial nerves we’ll get away with it but we won’t need to worry as  much about that for radiosurgery.

So what is radiosurgery?  Radiosurgery is the single surgical procedure that uses a highly conformal delivery of multiple radiation beams that are cross-fired on a target.  It is defined by intraoperative imaging with a goal of trying to biologically inactivate the target tissue in a single  procedure so the wheels in the wheels out like we do other surgical  procedures. 

The device that we put in Pittsburgh in 1987, the first of 8 subsequent types of modifications was a device called the Gamma Knife, that was the fifth one in the world at that time, there’s very little outcome data related to it. The original device had 201 x-ray beam with cobalt sources that cross-fired these gamma rays emitted from the cobalt sources on the target.  A general concept of why this works is that in a single treatment the radiobiological effectiveness the kill rate is about 4 times what we would see in the conventional fractionated radiation therapy.  So if we can give a tumor in a single gamma knife  procedure 12 Gy at the edge of this tumor, that’s essentially equivalent to giving 48 Gy the standard fractionated dose we give with radiation therapy.

But the reality is since we’re treating a tumor on the edge giving 12 Gy, if within the tumor we’re getting four times that amount at some point, in essence a dose equivalent to almost four times what we can safely give by fractionated radiation therapy.  And that’s why it has a superior response.

So this paralleled the development of imaging tools and it was critical to have the right imaging tools and many of you aren’t old enough to remember like me the first CT scanner that arrived at UPMC in 1975. This was a very rudimentary tool compared to what you’ve seen today in terms of MRI imaging.  But the first therapeutic device which was linked with these kinds of things was the gamma knife which we put in in 1987 and of course now we see with high resolution 3 T imaging, we needed a knee device which was a parallel development in this field and that’s what the current gamma knife looks like.

So I’m going to show you – we don’t have a lot of movies to show you so I want to show you one, it shows you a little bit about what the gamma knife actually is.  Within this thing that looks like the old mercury space shuttle, there are 192 cobalt sources which generate photons beams that are cross-fired on a target point inside the skull.  And this has robotic movements which can actually allow us to change the beam diameters between 4,8 and 16 millimeter beams which combine to destroy the target. These beams pass through the head, through the scalp, through the skull, there’s no surgical incision, there’s no opening because photon beam calculations can be used to determine what is the attenuation of each beam as it passes through the tissue before it summates on the target. We can actually mix beams together using a combination of very small beams in a single isocenter or what’s called shots in our terminology to cover the 3D geometry of this type of tumor.  And this effect is designed to be able to cause the tumor to inactivate.

So over the years the devices have changed, we have two devices here the 4C device and the current Perfexion device which as you see  has a lot more room in it and allows us to treat tumors that are much more widely scattered in the head.  This particular discipline is a classic example of disruptive innovation and in our field what we started out thinking was that this would be something used for tumors that were really unsuitable for surgery and so that’s what it’s initial role was.  But then as the results began to accumulate over 25 years, it became well we can use it for residual tumors, vascular malformation and sometimes as a primary management and now its growth in – is really a primary management for many benign skull-based tumors.

In a 24 year experience we’re over 11,000 cases at this particular center. As  you can see the pituitary tumors are not the most common indication and that’s because of the success that Paul has  outlined related to surgical removal of these tumors. So in many ways we’re like the clean up hitter for patients whose surgery can’t be completely successful. Whenever you’re the cleanup hitter that means that your results are never going to be as good as the primary management option because you’re ending up treating patients who have generally failed the  initial surgical  procedure.

So in our pituitary tumors, 83 percent of these patients had prior surgery, relatively few of them had already failed radiation therapy as well. And these tumor volumes tend to be relatively small, we’re trying to give dosages that are in a single treatment at the edge covering the tumor wall. 

Originally we thought that we could only treat small tumors that are far away from critical structures such as the optic nerve and chiasm.  So a tumor like this that’s grown back after the original transmodal type of operation and we can craft using the current technologies, very sharp treatment plans which confine the dose to the 3D geometry of this tumor seen in these various planes. But at the same time restrict the dose so that it sharply falls off, that the dose within critical structures such as the optic nerve and chiasm is extremely low. 

There’s another example of this in a patient where is regrowth of the tumor, we can outline the critical structure such as the optic nerve and then we can track the dose which covers the 3D geometry including the component that goes out into the cavernous sinus and treat this particular tumor. 

Here’s a residual pituitary tumor with right cavernous sinus and Paul has outlined beautifully the new advances that the endoscopic approach allows.  But in those cases where there has been either residual tumor or regrowth of this particular location, you can now use radiosurgery to treat this tumor in a very conformal way.  So these two procedures fit together hand in hand. 

Similarly a more difficult outline tumor, we’re also keeping dose low in the pituitary stock because we want to make sure that the dose in that is restricted so we don’t  have a  high risk of developing further endocrine dysfunction.

A typical example of a nonfunctional tumor might be a 29-year  old patient with residual nonfunctioning tumor, he’s already had two prior surgical procedures extensively  lateral to the carotid here and 40 months later we see significant regression of the tumor. 

So in our series of 125 nonfunctional patients up to 2009, these patients mostly were after surgery.  Relatively few of them are having primary radiosurgery.  Most of these were residual over current tumors and it’s being used as an adjutant or salvage method in these patients. 

We look at the structure of where it is, the involvement as you see, many of these are patients who have had lateral cavernous sinus involvement or suprasellar extension and the structures which we’re keeping a close eye on is related to how close it is to the visual structures, we want to track the visual function before and after the procedure as well as cranial neuro problems and of course we want to look at endocrine function.  Only about a third of these patients have normal endocrine function prior to radiosurgery, many of these patients already have hypopituitarism of one degree or another.

We’re looking at the dose and we’re trying to follow these patients in this case over about 5 years or so in terms of treatment.  We look at long term tumor control for the nonfunctional tumors and we see that over the course of time at one year about 99%, at five years 94% but there is some late failures as we get out to ten years of patients using this ____ type of analysis for tumor control.  

And we’re looking it related to tumor types, a new tumor versus residual tumor versus a recurrent tumor.  Long term tumor control rates are in the close to 90% range.  We didn’t find that how close they were to the optic nerve was important to this, they’re still overall in the 80 to 90% range.  When they’re a little bit closer to the optic nerve we may have to restrict the dose a little bit for safety and perhaps in this group of patients because of dose restriction their tumor control rate is about 10% less.

Then we looked at the multivariate analysis in terms of what are the factors that can allow us to predict better tumor control. And clearly larger tumors are more of a problem in both multivariate univariate analysis and is significant.  And also whether this represents the first or second recurrence of their tumor was also highly significant which could be looked at from the Kap-Meier approach.  So smaller tumors 95% tumor control rate, larger tumors down to 78% tumor control rate and that by the time you get out to 10  years when there’s a larger tumor, down to about half the patients who have long term tumor control rate. 

This is similar to patients with first recurrence versus two or more recurrences after two or more failed operations. Then we can see that the results are better when the patient is treated soon after the first definition of recurrence or residual tumor. So the point of this is don’t wait.  In our view since the tumor control rates are much better  in this situation, it doesn’t make any sense to us to say do subtotal removal or see that the tumor is growing and then just watch the patient to see what’s going to happen. 

If we look at the visual function in this – no patient has had a worsening visual function who had normal vision before, four patients with visual dysfunction because the tumors are very close to the optic chiasm – four patients had worsening vision but three of these were related to continued tumor progression and only one in a patient that had a stable tumor volume.

What about hormone function which of course is the most important to the patient in our endocrine teams.  And in general for patients who had normal hormonal function before radiosurgery, 12 patients had worsening hormone function and for patients who either had partial deficits again about 25% of patients and these are the most common deficits that were noted over the course of time,  ACTH thyroid stimulating hormone and growth hormone and very low incidence of developing any evidence of DI. 

So what is the overall rate over time of developing some additional hormone loss over the course of time, it actually builds over the course of years, at 5 years it’s about 29%, by the time you get up to 48% about half the patients at 10 years have developed some degree of a new at least single axis reduction so that they may need to be replaced either with cortisol primarily or a Synthroid.

There are factors that could affect the additional risk of hormone loss.  Certainly if they had prior radiation therapy so this now represents a cumulative dose to the pituitary gland, that was a significant factor that increased the risk of subsequent hormone loss in this group.  So again, in a Kap-Meier, prior radiation therapy versus no prior radiation therapy much higher rate of developing at least one new hormone axis lost. 

If we look at his compared to fractionated radiation therapy, there are a number of papers that expand many, many years and this is only looking at really the most recent interval because this represents the interval of time when radiation therapy, fractionated radiation therapy had quantum leaps in terms of how it was being delivered, using intensity modulated radiation therapy techniques and  computer imaging. Still the rate of hormone loss with hypopituitarism over the course of patients was in 50 to 70% of patients. As opposed to other reports from other centers using gamma knife radiosurgery where the rates range from only about 6% up to about 40% so overall about 50% less risk of developing a new pituitary hormone loss compared to fractionated radiation therapy. 

So in summary for the nonfunctional tumors, if we look at this tumor control rate initially very high at 10 years goes down to about 75% of patients.  And the poor response factors are the tumors that are already at larger volume and have had two or more prior operations and have residual tumors. In terms of the risk of developing a new hormone deficit 25% is an average of about two years but significantly increase with  prior radiation therapy. And it goes up over the course of time so maybe by the time ten years are out, almost half the patients may have at least one new axis loss. 

The risk of other complications is extremely low.  So no patients, only the one with the decline in visual function related to stable tumor treatment, again a very low risk even though most of these patients are patients being treated for tumors that are invasive of the cavernous sinus of developing other cranial nerve dysfunction related to locations in the cavernous sinus. 

So for the  latest in nonfunctional tumors, we think it’s a safe and effective treatment, actually if we look at data on terms of Centers that are using this technology, over 25,000 pituitary tumor patients who have had the gamma knife worldwide.  There are now hundreds of peer review publications, a book was just released by Sheehan and Lawson sellar, suprasellar tumors which has a very nice listing of all the publications related to various treatment strategies including surgery and radiosurgery. 

The most common complication over the course of time at least by 10 years is the risk of at least one new pituitary hormone axis but the rate of pituitary hormone loss is significant lower with radiosurgery compared to radiation therapy.  And just remember that – of our tendency is to try to lump everything into nice little pots, that this not the same type of procedure as what is being done with conventional fractionated radiation therapy. 

So we think it’s indicated in residual, progressive and recurrent tumors and we’ll talk a little bit about the hormone reactive tumors next.   It can be considered in certain newly diagnosed tumors, that is  with patients who don’t have optic neuropathy related to visual compression but who may have major other medical comorbidities for open surgery.  For rejection fraction of 20%, 5% or something like that, you may not want to do even transmodal surgery.  And we now know after years of experience that even tumors that are adjacent to the optic nerve can be treated effectively with the current strategies which allow us to spare dose into critical structures even very close to the tumor itself.

So now we are going to give you some data from our endocrine active tumors, this isn’t as beautiful perhaps because of the difficulty of these tumors and why combination treatment is extremely important and I’m very grateful for Dr. Liu who helped put together this data.

So we talked a little bit about the nonfunctional tumors for trying to attain tumor growth control, preserve function, preserve existing endocrine function and eliminate some of the long term risks of radiation therapy.  But there are two other important factors related to hormone active tumors that we have to consider.  One is we need to try to get rid of this excess hormone secretion, or at the very least we’d like to reduce or eliminate the use of suppressive medications that cause, as I’m sure you’ve heard, the very long term costs of this. 

And we should remember that this is a single procedure done in a single day, it’s based on a frame application, this is a guiding device for stereotactic frame, every patient comes in the morning, we give them a little bit of a relaxation technique, we use local anesthetic and we attach this guiding device to the head and then we go over and do an MRI scan and then we bring the patient back. We do a computer plan, we do the treatment and the patient goes home in the same day.  In the older days when the planning systems were a little bit slower the patient could sit there and read the New York Times while we did the planning and then we’d do the treatment.  Now we often treat 5 patients in a morning, we can – we start early in the morning putting the frames on, we get the patient’s image with MRI scan by about 7:00 or 8:00 o’clock in the morning before the radiologists wake up, and that way we don’t have any problem to get the imaging that we need. We are there supervising the type of imaging that is being done. We don’t have to do an hour and a half MRI scan on most of these patients, we are doing a scan which allows us to localize the target within the head and almost always it’s being done with MRI.  And that’s because it’s only with the MRI scan that we really can see the tumor. 

So over the years there has been a lot of confusion about things and one of the reasons that Sue talked about is that as surgeons we always could be thought well we are doing a relatively good job and then every couple of years the endocrine teams would get together and change the rules.  So they would say you know 25 years ago you know growth hormone less than 10 was great.  So there were a lot of papers published in our literature saying well we’ve got all these patients with growth hormones less than 10. Then somebody said well, actually it needs to be less than 5.  And then somebody said no, actually it needs to be less than 2.5 but you need to have it less than 1 when you do a glucose tolerance test and besides that the IGF1 that we cooked up, also that needs to be normal for age.  And so the bar kept getting raised for this. And as you read the literature you got more and more confused because what happens, somebody was talking about one set of values and also the type of values that were being used weren’t understandable from one place to another. 

But one of the things that came out was that Alex Landolt is a great microsurgeon, done thousands of pituitary tumors and then started using gamma knife as a cleanup hitter, and started to think well, actually we found that it works much better and faster than radiosurgery but we are little worried that some of the suppression medications that are being used actually reduce the rate of response to gamma knife. So a sort of story emerged in our world which we actually think is probably not true, that patients on suppression medication don’t do as well as patients who are off suppression medication and that for patients like growth hormone secreting, most of whom are now on Cabergoline, they need to be off the suppression medication before you do gamma knife and the same for patients with Prolactin on Cabergoline or growth hormone patients on Sandostatin, so it turns out at least in our data that those, usage of those agents on or off medication does not affect the results.   So what we need to do to achieve the goal is we have to give much higher dosages because here we are not only trying to stop the tumor from growing, inactivate the tumor, but we actually want to shut down the biochemical protein synthesis of the cell. And that requires a much higher dose in order to achieve that, that effect.

So if we look at growth hormone and we are trying to achieve this particular high benchmark growth hormone values less than 1 and the glucose tolerance test with a normal IGF1 and we have to get the normal pituitary profile.  So here’s a typical patient with a growth hormone secreting residual tumor that’s out lateral on this particular patient and after a period of time a significant reduction in this particular tumor, which is not an overnight effect, it may take many months or even years before we get that type of response.  Here is another patient with this type of tumor with a smaller tumor over the course of time in about 3 years.

So 58 patients we’ve looked at now with clinically active acromegaly, we lost 11 patients to follow-up and we excluded 4 patients who had less than 2 year follow-up, more women than men in this with a average age of about 40 or so.  Most of these patients it is expected, they had prior surgical resection, a few had had surgery followed by radiation therapy, and the reasons to consider were not only residual tumor but also that these patients were resistant to suppression drug or simply couldn’t tolerate the drug, only one patient. And by the time they are treated 58% of these patients already have some form of pituitary axis loss, many of them have invasive tumors and we looked at this value of the IGF-1 index which is the value of the patient divided by the upper limit of normal for the patient, which is age and gender specific.

So again we started out with the various models of the gamma knife that we’ve used, we put on the frame, we do local anesthesia with sedation, the imaging is required and these are postoperative patients because as you’ve heard the gland enhanced more than the tumor does, we want to actually give a low dose of the contrast, about half of the dose of the usual dose of Gadolinium or other paramagnetic agent.  And then we use a varied volume scan to be able to get sort of a 3-D picture of this and then new planning systems for this.

So the endocrine follow-up in these patients is now out over 6 years or so, we are using higher doses for this but we are keeping the dose to the optic nerve using gamma knife well below 8, 9, 10, which is actually well within the tolerance of these types of tumors.  And we get these types of high resolution imaging, this is target based treatment, we do the treatment plan here as you see on the coronal and the lateral view with significant regression of this tumor over the course of time, and this page IGF-1 level is significantly better although not probably age normal for this 51 year old patient, but he is off of any suppression therapy at this time.

And if we look at the response rate for acromegaly we can see that tumor control that is prevention of tumor enlargement of further growth can be done in the vast majority of patients. And this either means further shrinkage of the tumor, measurable eye imaging or that the tumor shows no further signs of growth with only one patient showing further growth. But that’s the imaging result and from the endocrine perspective what’s more important of course is what happens to these patients in terms of their endocrine activity.

So in this group of patients with residual recurrent tumors we could obtain remission defined as a growth hormone level less than 2.5 and a normal IGF-1 for age off of medication in about 44% of patients. And control was achieved in another group of patients but these patients remain on suppression medication over the course of time.  A process that takes time to develop so that these patients can’t be told well it’s going to be done in X number of months, or we’re going to see it, we begin to see effects certainly in most patients within a couple of years but the overall effect will build up over time and so we can’t sort of say that there is a single point in which the patient may or may not have responded to the, to the procedure.

If we look at the significant factors in a multivariate type of analysis we did not find that patients on suppression medications for acromegaly had any different outcome than the patients who were off suppression medication. And yet certainly if they required medication afterwards, suppression medication, those patients were going to do better with lower values. Tumor invasiveness as we might expect was an important aspect in terms of response and the more invasive the tumor the more difficult.  Looking at the IGF-1 index before the procedure, if the index was less than 1.5 then the tumor, then the hormonal relapse rate was much, much better than if the patients had very high levels beforehand with much higher growth hormone and IGF-1 levels.

And similarly the response rate was much better for patients with less invasive adenomas compared to patients who had more invasive adenoma complications. No patient developed any optic nerve loss, our plans were always based on very specific sparing of optic nerve structures.  No patient developed a new deficit of extraocular movement, no patient developed diabetes insipidus, but if we look at patients with the eventual development of at least one additional pituitary hormone axis this was seen in about a third of patients.

One new axis in 6 patients, 2 new axis deficiencies in 9 and 1 patients with since a pan hypopituitarism with those particular findings over the course of time.  When does it occur?  It occurs anywhere from 9 to 145 months after the procedure, but on median about 3 years after the procedure. 

Did they need something else done?  Not very many. One patient had repeat surgical resection, four patients had repeat radiosurgery for evidence of tumor growth. And if we look at other reports of gamma knife for these types of tumors we see similar rates of new hypopituitarism in about 1 out of 3 patients over the course of time.

So for these patients with acromegaly we think it’s effective in many patients for these residual or recurrent tumors and think it should probably be done early in the patients who have this particular disorder because of the lifelong risks related to excessive growth hormone hypersecretion so we don’t think that these patients we should wait. 

Prolactinomas are an even greater challenge, Paul has talked about that.  And the reason they are because of the reasons we end up treating a patient is because they are very invasive, number one, and because other treatment has already failed, because clearly the initial management of these patients is medical with suppression, suppression therapy. And it’s being used in these group of patients who have generally failed that.  In our series of 22 patients with tumors that are laterally invasive in the cavernous sinus interesting male and female about the aside, all of these patients have had medical treatment or radiation therapy and 15 had had prior surgical resection.  So 13 of these patients had resistance to suppression medication or 6 who had had intolerance of it, and 8 of these patients already had some form of endocrine loss after the initial management of the patients.

The same way that we do these patients with the gamma knife treatment as we talked about, and tumor volumes, larger tumor volumes which require some lowering of the dose to preserve optic nerve function is going to be necessary in these invasive types of tumors that are larger. But a typical finding might be this 62 year old patient with a residual left cavernous sinus prolactinoma, serum prolactin level was 258, on Cabergoline and 22 months later there is significant tumor shrinkage, the prolactin level has now gone down to better, not normal.  So if we look at this we can see prevention of tumor growth in most patients, however or regression, but 3 patients had delayed tumor growth over the course of time and if we look at variables in this relatively small series, age was important in looking at the sort of response rate.  The younger does better. 

What happens to the endocrine outcomes?  Well in terms of complete normalization of this most of these patients will need to remain on drug, some will have, or most will have improvement in terms of their hormone prolactin level but persistent elevation may remain in these particular patients and again tumor volume was probably the most important but also that the initial prolactin level before the procedure was in the lower range rather than some of these patients with sky high levels. So if the prolactin level is quite high over 200 here then the tumor control rate as we see here in terms of normalization is very, very low compared to patients where the prolactin level is statistically significant if it’s less than 200.  Similarly for tumor volumes which allows us to give a higher dose, smaller tumor volumes are going to have a much more likelihood of normalization of prolactin level than the bigger tumors. 

Again no patient developed optic neuropathy, no other new cranial nerve deficits and one patient developed a new hormone axis loss, some of these patients required additional resection, one went on to radiation therapy with a large invasive tumor and one patient had repeat radiosurgery.  If we look at prolactinomas in terms of hypopituitarism risk, it’s still in the range of about 1 out of 3 patients over the course of time, but overall we think for prolactinoma that it’s an important adjuvant management for patients who have failed medical and surgical options, usually with laterally placed tumors in which we can use it.

I’ll conclude with just a few comments about this disease which in fact extremely difficult as well because we are now dealing with patients who have failed surgical management and who are either intolerant or can’t take suppression medication in which we are trying to achieve this type of, this type of goal.  51 patients with clinically active Cushing’s disease, 10 were lost to follow-up so we can’t tell you about those, so we have 41 patients left in the series.  95% prior surgical resection and 5 who had had radiation therapy, most of these patients of course are used with suppression of cortisol production with Ketoconazole and about half the patients, about half the patients already had significant pituitary hormone loss with this prior, due to prior treatment. Trying to give much higher doses, these tumors tend to be smaller, symptoms are quite dramatic for tumors that are often so small and sometimes hard to image on the postoperative scan.  So tumor control has been quite good in terms of prevention of further growth of the tumor either seeing shrinkage or no further growth with 4 patients showing progression.  However in terms of the endocrine outcome, total remission in 44% of patients with 40% showing persistent elevation of factors measurable for Cushing’s disease.

Again no complications related to this type of – however about 25% incidence of a new pituitary hormone axis in patients who have had the procedure, that’s sort of in the ballpark of what some of the other series are in terms of this. This is probably the one with the largest reported delayed hypopituitism risk which may be related to using very high dosages in a single treatment to this tumor equivalent to basically giving almost 130 grade of fractionated reduction therapy, so 4 times what we can give by usual radiation therapy techniques.

So for Cushing’s disease we can attain tumor growth control in these refractory tumors in most, endocrine remission is more difficult to achieve but when it does it may eliminate the need for lifetime suppression therapy in patients and radiosurgery certainly seems to have a reduced risk of additional hormonal axis loss in comparison to more conventional radiation therapy types of techniques, and perhaps over the course of time this may reduce lifetime management costs and of course all of us are trying to spend a lot of time trying to reduce the cost to our government so that it doesn’t go bankrupt and we are no longer at 15% medical healthcare needs. 

So that’s basically a summary of the work related to these patients that we’ve done with our technology as an adjuvant to microsurgery and to endocrine management of these patients. The procedure itself appears to be low risk of comorbidity, has the best results in patients in terms of those who do not have functional tumors and has a moderately good success in the patients who have failed surgery with, with endocrine active tumors.  Thanks very much.