I feel a little bit like I'm carrying coals to Newcastle because there are people here with really amazing expertise in the radiosurgical management of patients with pituitary adenomas, greater than mine, I dare say, but I'll do my best. So people have trouble adapting to technical changes, disruptive changes, even to some extent radiation oncologists will have a hard time accepting the concept of low volume, high dose radiation, which is how they think of radiosurgery, but it's more of an issue on the neurosurgical side because neurosurgery is often organized in a paramilitary way and we're going off to war, we beat our chests when things go well with surgery, but the fact is if there's a better way of winning the battle, we should embrace it, and radiosurgery often is that, and this is not a repeat of vestibular schwannoma talk, I just want to bring this up to show how these battles can take on a life of their own and harden along lines. This was a paper from Norway in neurosurgery a couple of years ago about vestibular schwannoma radiosurgery, it purported to show that radiosurgery was better in a prospective randomized trial than microsurgery, and unusual for neurosurgery, which usually has several comments after the papers, they had seven, and guess what, the four radiosurgeons all said, well great, this shows that radiosurgery works, and the three microsurgeons all said, well that just shows they didn't know how to operate. So that's the kind of controversy that exists in vestibular schwannoma radiosurgery, and Dr. Friedman touched on that earlier, and fortunately, for pituitary adenoma management, I don't think that's really the case, because there's a general consensus that surgery is the best treatment for patients who have symptomatic macroadenomas, and who have hypersecreting tumors that are symptomatic, not because of their size, but because of the substance that they produce. So the goals of management of these patients are to completely remove the tumors for oncologic cure, that is to prevent them from growing, that could be surgery or radiosurgery, and or to achieve endocrine remission, because hypersecreting tumors can be macroadenomas as well. There's also a consensus that for patients with prolactinomas that are either large, and or have a prolactin level greater than 500 nanograms per milliliter, that the treatment of choice is medical management with a dopamine agonist, and that basically works all the time for those patients. So this is an algorithm that originally came out of the work from the Stanford Endocrine Group, led by Larry Katznelson, who's a neurosurgeon, and you can see it happens to be about incidentally found tumors, but this is a useful way of stratifying treatment for pituitary tumor patients in general. If you have a functioning tumor, that is a hypersecreting tumor, if it's a prolactinoma, very high prolactin level, or a macroadenoma, older person, undoubtedly you treat those patients with dopamine agonists. The exception to that would be a young woman of childbearing age, in whom the prolactinoma is interfering with conception, in whom, well, there there might be some dispute. I think most neurosurgeons would definitely think that surgery is the best option. There are some endocrinologists who prefer to try to manage those patients medically, that's still probably a minority of those physicians, whereas if the patients have Cushing's disease or acromegaly, pretty much no question, everyone agrees you should aim for a surgical cure, whether it's incidentally found or not. For a non-functioning tumor, if it's a small tumor, no question, the patient should be followed. The tumors may never grow to a symptomatic size. If they are symptomatic, from visual loss would be probably the most common symptom, especially a visual field loss, classic paradigm would be a bitemporal hemianopsia from chiasmal compression, although not exclusively that. They can have extraocular movement deficiencies because of involvement of the cranial nerves of the cavernous sinus, and those patients should be managed almost definitely starting with surgery, and we'll see what role radiosurgery has to play in those patients. If the macroadenoma is large enough to be compressing the chiasm but is asymptomatic, then also some debate and difference between physicians about whether you would take those patients for surgery at that point or follow them, but that's sort of a dealer's choice that we don't have to dwell on too much here. So transphenoidal surgery is the mainstay of management of patient's pituitary adenomas. This is a classic example, somebody with a non-functioning, large macroadenoma causing visual loss, and these are preoperative images obtained in the OR with intraoperative MRI. That happens to be something I'm interested in, so I'll show a few examples of that. Here's the ENT search preparing the nose on the left, and you can see a better view of the magnet under the table there, and on the lower right, these are images taking what's called a compare function off the intraoperative MRI, preop, intraop where I thought that the tumor was basically all out, which is a little embarrassing because it looks like it was barely touched, and then I found a few septations behind which the majority of the tumor still lay, and that's without and with contrast, and you can see the tumor is out, and you get a surgical cure that way. No one would really think of using radiation therapy or radiosurgery in a patient like that. I'm only showing this particular case because the previous surgery was done microsurgically, and this was done endoscopically, and I just did it two days ago, so it's sort of hot off the press, but anyway, very nice, and it's fun surgery, especially nowadays with the endoscope, you get a nice view of the cella. If the tumor's large enough and the diaphragm cell is disrupted, you can see up to the optic chiasm before and after a section with intraoperative MRI, and there are various techniques for repairing the CSF leak using a fluorescein injection through a spinal catheter, and then confirming that there's no leak at the end, different grafting materials, fat graft, and the surgery goes. So again, there's a consensus across the board that if you can cure patients with these tumors surgically, that's what you should do. So what is the role of radiosurgery? It primarily is an adjuvant treatment for patients with pituitary tumors, especially those who have incompletely resected non-functioning adenomas, and the goal is to stop tumor growth. Doesn't matter if the tumor's still there as long as it's not growing and not compressing the optic chiasm, or if you have an incompletely resected or endocrine-persistent hypersecreting tumor where the goal is very different because the tumor may not grow, but if it's still producing your excess ACTH or growth hormone, then the patient's life is at risk from that, and there you really have to achieve an endocrine remission. So in a sense, it's analogous to AVM radiosurgery. Let's assume you have to really obliterate an AVM and not just subtotally obliterate it. You know, in a way, all or none, and that's the same thing with hypersecreting pituitary tumors, at least those causing Cushing's disease and acromegaly. It's sort of all or none. For an occasional patient, and I'll show an example of that later on, who has a tumor with a primary cavernous sinus component, then radiosurgery can be the primary treatment modality for a patient with a pituitary tumor. That would be relatively unusual. So here's a hoary chestnut from the radiosurgery literature. Almost all pituitary tumors are benign, so if you subscribe to the linear quadratic formula of radiation therapy, and I'm heading out on maybe some thin ice here with Dr. Schlesinger is in the office, in the audience. I don't want to misspeak in terms of physics, but from a primitive neurosurgeon's understanding of the formula and these curves, the more benign a tumor, or late responding, or slow growing, or put it in the formulas terms, lower alpha beta ratio, the bigger bang you get for your radiosurgical buck. So if you're going to give 15 gray to an early responding tumor, one with a higher alpha beta ratio, the equivalent dose in fractionated radiation terms is okay, but it's not as high. Much, much better if you're treating a benign tumor. So single dose radiosurgery really favors the treatment of benign tumors, which basically all pituitary tumors are, practically speaking. So what would be the criteria for treating a pituitary tumor patient with radiosurgery? Surgical treatment should be maximized. If the whole tumor's out, then that's it. You don't have to do anything else as long as they also have achieved endocrine remission. If it's not, then perhaps sometimes the patient should be followed and reoperated on when the time comes. So surgical treatment should be maximized. Medical treatment should be maximized. If they have a prolactinoma, or if they have growth hormone secreting tumor that is not ideal for surgery, then they should be treated medically. But if you're going to do radiosurgery, you should stop the medical treatment in advance of the radiosurgery, probably for about eight weeks. Because analogous to surgery, your remission rates are better if you stop your medical treatment enough in advance of the surgery to maximize the surgical results. And there's a literature that shows that that's true as well for radiosurgery, both for patients with prolactinomas and growth hormone secreting tumors. You have to keep your maximum dose to the optic chiasm less than eight gray. There is a small literature that shows you can go up to 10 or even 12 gray. But one of the goals we were charged with was to talk about indications and complications. So I would tell you if you're going to exceed that eight gray threshold for pituitary tumors for the chiasmal dose, you should do it with real caution. And it's really best if you stick to that eight gray maximum. Practically speaking, that generally means that your tumor margin should be at least three millimeters from the optic chiasm. So you could also treat patients with hyper secreting microautonomas in whom surgical risks are high if the tumor is small enough and otherwise favorable for radiosurgery, because again, they have a lethal disease, Cushing's disease, or acromegaly. You also want to make sure that after your surgical treatment is maximized, that enough time has elapsed. You would never treat these patients early on after surgery, because things change a lot if you wait a little bit after surgery. Here's an example from a paper we published in 1999. Here's what the tumor looked like originally after surgery. It almost looks like a peak and shriek operation, but believe in yourself, wait three months, and that's what it looks like. Now you have a radiosurgical target. So the decrease in volume of residual tumor and stuff after transvernal surgery is measurable, predictable, and it could be quite dramatic. So you should wait at least three months for radiosurgery after surgery. Make sure your imaging is as good as possible. Potentially try to plan off of a 3-Tesla magnet where you'll see things with somewhat more clarity than with a 1.5. Now you could do pituitary radiosurgery with any available device and basically it'll work if you do it carefully, you plan the treatment properly, and you make sure that it's delivered accurately. This was a patient treated when I was at New Jersey Medical School using a great system. This was a marriage of the X-knife, the late lamented X-knife made by the company formerly known as Radionics. What's that? Male Speaker 1 No, not lamented. Male Speaker 2 Well, but, well, hold on a second, joined to the University of Florida floor stand. That was a partnership between Radionics and UF that lasted about three weeks, I think. But it was great while it lasted. So here's a patient. Male Speaker 1 That's why they're variant partners. Male Speaker 2 What's that? I'm talking about heading out to thin ice. Here's a patient with a recurrent non-functioning adenoma. She had had prior transgenital surgery and radiation therapy. Tumor regrew by temporal hemianopsia. This is what it looked like after her surgery and tumor control five years later after radiosurgery. You could do it with a cyber knife, even though Accurate decided not to support this meeting at all. I guess no reason to change the slides. This was somebody who actually never had surgery. She was followed for a small tumor that looked like a pituitary adenoma originating in the right cavernous sinus and enlarged. She had ophthalmoplegia, especially cranial nerve six, and numbness in the right face and was treated with a cyber knife at Overlook Hospital. These are pictures at the time. The only way I could capture images was to photograph the screen, but anyway, you could see the treatment plan, and we did indeed treat her with two fractions, and I'll talk about that at the end of the talk, as you can see here, nine gray times two, with shrinkage of the tumor and symptomatic improvement half a year later. You could certainly use the cyber knife in an adjuvant setting. This is a patient with a large, non-functioning adenoma who had intraoperative MRI-guided surgery. As you can see, it looks like it's hard to tell what that residual is, but there was some residual tumor there. He received 14 gray in a single fraction and has remained controlled since then. You could treat the patients with a Novalis MMLC inverse planning type system. This patient had had two prior transvenodal operations elsewhere and radiation therapy. The Novalis system, as you either know or you'll see, is a LINAC-based system with a robotic couch that positions the patients with a combination of stereotactic coordinates, infrared optical tracking, and registration between fluoroscopy and preoperative digital imaging. That's what the treatment plan looked like. In her case, the prescription dose was 12 gray, and as I'll allude to later on, this is one of the things that you have to take into account, prior radiation. So you can't have some simple, fixed, abstract notion of what the dose is. If it has volume, location, and other potential morbidities, it's that, it's prior radiation as well. So this is how she did overall. This is what her MRI looked like when she presented with apoplexy, a not very impressive looking operation there, okay, but just wait, wait, see? Now after three months, you have your radiosurgery target, we're over four millimeters away from the optic chiasm, and she remains controlled after having failed all of her previous treatments. And you can also treat the patients with a gamma knife. This patient had a very partial previous transfenylol resection of his tumor, which continued to enlarge, and he was symptomatic. After his repeat operation, waiting a few months, that's what the residual tumor plus gland looked like. It's a picture of our new gamma knife, which we installed at the initiative of my radiation oncology colleague, John Knisely, a little different from most situations where the neurosurgeons are pushing for it, but nonetheless, we're doing great and happy to be working with this. And we were able to treat this patient with 16 gray and keep the eight gray line below the optic chiasm. So if you look at the literature on the treatment of patients with non-functioning adenomas, this is from a chapter in the textbook of stereotactic and functional neurosurgery from a few years ago. You could see that unsurprisingly, for the small volume benign tumors, that the tumor control rate is quite high, on the order of 95 percent or even more, if you sort of add up all the series. This is not true for patients with hypersecreting tumors. The rate of endocrine remission is just not that high. This is from Dr. Steiner's chapter in the same book, and there certainly are many other reports that look at this and the reviews, and the rates of remission really remain kind of the same over multiple series, even with, as you'd expect, much higher doses than you would use to prevent a non-functioning tumor from growing. And as you can see, the remission rate, especially for prolactinoma patients, is quite low. So that's why no one, I think, would really suggest that radiosurgery is the primary treatment for patients with hypersecreting tumors. It should be reserved for patients who failed other treatments. And the question is how to make this better, whether it's dose escalation or some combination of medical treatment and radiosurgery. So resonance, what does this patient have? Hirsutism, gaining weight, diabetes mellitus, abdominal striae, easy bruising, poor healing of wounds. What has she got? What's that? Cushing's disease. This lady has had thickening of her facial features, enlargement of her hands, had to take off her wedding ring. Her voice has gotten deeper. Yeah, very good, okay. What about complications of pituitary radiosurgery? Whether hypopituitarism is really something you'd call a complication, I don't know. Because if you're treating this small area in which you can almost never differentiate the tumor from the gland on imaging, to say that getting a decrease in pituitary function is a complication, it's almost part of the treatment itself. Nonetheless, it's not inevitable. It is dose dependent. So in that sense, we have an ability to lower that rate of complication. And it's as high as 72%. That was an incidence at a series at the Karolinska Institute, the home of the Gamma Knife. Certainly in general, you can assume it's going to be on the order of 30%. So about one third of the patients whom you treat with pituitary SRS will get hypopituitarism over several years. Now what about visual loss? This will be, should be, a very rare, bad, but very rare complication if you keep your maximum chiasmal dose to less than 8 gray. But keep in mind, any prior treatments that the patients have had, obviously with radiosurgery, but also certainly with radiation therapy, no matter how far back in the past. And future neurosurgeons don't think that you can simply shove this off and let the radiation oncologists and physicists worry about it. If you're doing radiosurgery, you're responsible, and you have to make sure that that treatment plan meets these criteria and that the patients are appropriately treated. And then the secondary neoplasia, it was alluded to earlier, I'll just say a brief word about it in a bit. So what about dose selection? There's not much hard data to really guide one on this. It's based on the notion that, well, benign, slow-growing tumors can be treated with a lower dose, but hypersecreting tumors, where you have to obliterate and not just stop, need a higher dose. So for non-functioning tumors, the rates that have been described range between 13 and 16 gray. The bigger the tumor, the closer you are to the chiasm, prior treatments, you choose a lower dose, and the converse presumably is true. For hypersecreting tumors, the prescription dose can be as high as 30 gray, again, keeping within those chiasmal constraints. The cavernous sinus, which primarily contains motor efferent nerves, there's an efferent component to the trigeminal nerve, but it's not a special efferent nerve, so for whatever reason. It's also a peripheral nerve, not like the optic nerves, which are a central nervous system histologically. The cavernous sinus can take a bigger hit. You can certainly give over 20 gray and even potentially as high as 40 gray into the cavernous sinus and not see a complication, although you don't want to push it too much. What might make this better in the future? This is an idea promoted by Brain Lab, the idea that intraoperatively, as you're doing surgery, this happens to be for meningioma, but it could be for anything else, including a pituitary tumor, that based on your updated navigation, you'll create a model of the tumor. You could do with intraoperative imaging as well. Do your radiosurgical treatment planning on the fly and say, okay, we still can't do a safe treatment plan, too much dose in the brain stem, or all right, down here, this is small enough. We can treat with SRS or fractionated SRS or RT, whatever the case may be. That could be one way that neurosurgeons can maximize their role or the bank for their buck in pituitary radiosurgery. What about neoplasia? Dr. Freeman quoted a paper by Patel and Chang, and they estimated 1 in 2,500 risk. I mentioned this from the chapter by Dr. Steiner, 0.7 percent, one out of 140 patients. That seems awfully high, and I don't think that jibes with the experience of anyone in this room or also in the literature, but it's worth including just as a caution to know that there is going to be some risk. Radiation therapy has a risk of secondary neoplasia. It stands to reason that radiosurgery is going to do it at a much lower rate, but still it's going to be something. It's true that the group in Sheffield, England, headed by Jeremy Rowe, did find 0 percent secondary neoplasia in 1,200 patients followed for 10 years, but a couple of years after that paper came out, they did report a case of a probable vestibular schwannoma malignant transformation. So it's rare, but not zero. A few brief words about frame versus framelessness. Are frames the gold standard? This picture by John Adler showing a bent pin in a CRW frame, Bob Masunas, I think we can agree he's late in lamenting, showed how there's potential inaccuracies in stereotactic frames. There were studies showing the accuracy of frameless systems. We compared patient perceptions of frame versus frameless for people with benign and malignant disease, and frameless mattered more to them in a perception of discomfort and pain in radiosurgery than malignant versus benign disease. So why does that matter? Because what's the role of hypofractionation for pituitary radiosurgery? This is a paper by John Adler using a cyberknife, and the histologies are mixed. Fewer than half of the patients had pituitary tumors, but nonetheless the same principles presumably apply. Patients were treated with an empirically selected mix of sessions and doses and so on, and you could see high rate of tumor control, and only one of those patients had visual loss because of the radiation as opposed to tumor growth. So it's not great, obviously, for that patient, but overall not bad statistics. These are patients all of whose tumors were within two millimeters of the optic chiasm. So is there a role for fractionated radiosurgery? Maybe you'll hear something about that from Dr. DeSalis when he speaks. So in sum, radiosurgery provides excellent control of pituitary tumor growth, although keep in mind that endocrinological remission is less likely to occur. You have to protect the optic apparatus, keep the maximum dose to less than 8 gray. The role of hypofractionation is uncertain, and that even after these many years of pituitary radiosurgery is still a subject for a future study. Thank you.

radiosurgery

pituitary adenomas

technical changes

treatment option

endocrine remission

devices