This lecture will review the management of lacrimal gland tumors, epithelial and lymphoproliferative, by looking at natural history, imaging characteristics and treatment modalities. We will discuss the improved survival and ability to preserve the eye with current treatment modalities including a combination approach of surgery and adjuvant radiation. Furthermore, we will review the evidence for ultra low dose radiation for MALT lymphoma of the lacrimal gland. Lastly, we will discuss possible future direction in using genetic mutation profile to guide management of epithelial tumor of the lacrimal gland.
Lecturer: Dr. Vivian Yin, Memorial Sloan Kettering Cancer Center, New York, USA
DR YIN: Hi, everyone! Thank you very much for joining us. We’ll be talking about the evidence-based management of lacrimal gland tumor today. I’m just gonna pull up the Q and A. So I will try my very best to try to field some questions at the same time as giving this lecture. So feel free to type in your Q and A box as we go, and feel free to stop me. A little introduction about myself. My name is Dr. Vivian Yin. I’m an oculoplastic and orbital oncologist at Memorial Sloan Kettering. None of my financial disclosures are relevant for today. So for today, we’re gonna go over a little bit about the epidemiology and the basic background on — we’re gonna focus mostly on malignant tumors of the lacrimal gland. And then we’ll go over a little bit about the treatment for lymphoma, as well as epithelial tumor, the controversy over exenteration, over globe sparing, as well as intra-arterial, and we’ll end the talk a little bit about what may be coming up on the horizon for treatment of lacrimal gland tumor. Some of these are still very early, in their investigational phases, but we’ll talk about gene mutation and how we may try to use these for targeted therapy in treatment of lacrimal gland tumors. So as a background, we know that lacrimal gland tumors are divided into epithelial and lymphoproliferative. As I mentioned before, I’m not gonna talk about the benign inflammatory categories. Within the epithelial, we tend to divide them into those that are malignant and benign, and roughly I would say half are benign, and of those the most common is pleomorphic adenoma. Of the malignant, adenocystic dominates. You’ll notice that based on the percentages, 25% is missing, because that’s filled in by everything else that’s listed here. Among the lymphoproliferative disease of the lacrimal gland, MALT lymphoma predominates, about 75% of lymphproliferative and disease, followed by follicular and diffuse large cell lymphoma. And we’ll go over that as we go through these slides. So just to get you thinking, we’re gonna start with a case. This is a case of mine, 69-year-old female, who presented with the story of two months of bulging eyes. And I’ve given you two slices of the MRI here. One is a coronal on the top, and an axial on the bottom. Now, if you were looking at these, and trying to describe this lesion, a couple of terms may come to mind. One is that you can see on the coronal that the globe is displaced inferiorly, and you can see the enlargement of the lacrimal gland, especially if you compare it to the fellow side. So whenever you’re looking at an image, I always say: If you don’t know what you’re looking at, just go side to side and see what looks abnormal. So this is the rough size of a normal lacrimal gland, whereas on the other side, you can see how it is much more prominent, and the center of the globe is here. You have the center of the globe slightly higher on the other side. Now, another term that may come to mind as you look at this on the axial is that it molds around the globe and also has a little tail. We’ll go over those later on in terms of the significance of them. Looking at those two images, what would you like to do next for this patient? Would you like an excisional biopsy of the lacrimal gland, meaning you remove the entire gland and try to get a diagnosis? Would you like to do an incisional biopsy? Do you want a PET/CT first? Do you want a bone marrow aspirate? Or none of the above? There is something else you would like to do first? So I’ll give you a couple more seconds to fill that out. And I know some of you may be thinking: One of the questions that was posted on the registration was a question about incisional versus excisional biopsy. We’ll talk about that a little bit. So you can see here that I see most of the people want an incisional biopsy. There are some that want excisional biopsy. And I’m glad to see that the differences are not huge, because that allows me to talk a little bit about the controversy over incisional biopsy. So in this case, we did an incisional biopsy, and the reason is because I was suspicious of this being a lymphoma. And lo and behold, the incisional biopsy shows these small blue cells you see on the left, and these — I’m sorry — these little glandular tissue, which are your lacrimal glands. And when you stain them for CD20, which is what the slide is showing on the right, CD20 is a stain for B-cells. So as you can see on the report underneath, it’s positive for CD20, so it’s telling us this is a type of B-cell lymphoma, and it’s negative for CD5 and CD3, which are the markers for T-cell. So this is consistent with a MALT lymphoma, which as you know is the most common lymphomatous disease of the lacrimal gland. So now that you know the diagnosis is a MALT lymphoma, what do you think is the appropriate next step? Would you like a complete excision, possibly with frozen section or margin control, or would you like to initiate induction chemotherapy, followed by maintenance? Would you like a PET/CT first? Would you like to do a CSF tap for flow cytometry? Or none of the above? Okay, so it looks like it’s split equally between complete excision and initiation of induction chemotherapy. Now, this is a little bit of a trick question. Because I didn’t tell you, besides the orbit, was there any other systemic involvement. So the correct answer is actually a PET/CT. And the reason that that has to be the first step is because we do not treat lymphomatous disease with surgical excision. Instead, the treatment is either radiation, if it’s limited to the orbit, or if it’s involving the systemic system, then you consider chemotherapy. And with MALT lymphoma, sometimes you don’t even — if there’s low burden of disease in the rest of the body, you actually can observe those instead of treating them with chemotherapy. So like I mentioned, orbital lymphoma is predominantly MALT, which constitutes about half or just over half. Now, depending on which literature you look at, there may be a split between what is the second most common. So it could be either follicular in some studies, in some regions of the world, or it could be diffuse large B-cell lymphoma as a predominant one. Now, this is an American-based journal that says that diffuse large B-cell for the North American population is a little bit more common than follicular, but I can tell you in my practice, I see the two sometimes being almost exactly the same. And then you have a small percentage of mantle cell. I didn’t put a percentage of NK-T-cells, because it’s relatively rare, and we don’t really know the exact incidence of that contributing to orbital lymphoma. But one of the things that is unique about NK-T-cell is that they are very chemo-resistant. So you treat them in a different fashion than the rest of the lymphomatous conditions. Now, these typically present as a painless mass. To have bony erosion on imaging is quite uncommon. There is a small percentage, though, that can present with bilateral. So if you see a patient with bilateral lacrimal gland enlargement, that doesn’t rule out orbital lymphoma. It just makes it a little bit less likely. And lymphomas are quite common in the orbit, and accounts for 20% of all orbital masses. This is not 20% of all malignant lacrimal gland tumor. But all orbital masses, 20%, approximately, is orbital lymphoma, which is a big chunk. And only about half of them involve the lacrimal gland. The others can have systemic involvement or present as a plasmacytoma or discrete mass in the orbit. And the treatment is actually radiation. Most recently, there’s been a push to decrease the dosage of radiation. As we know, radiation causes a lot of ocular side effects, and this is an attempt to decrease the side effect profile to the globe. Now, I notice someone had raised their hand. I’m gonna encourage you to actually type in the Q and A instead, as we’re not able to hand over the mic to the participants. Now, with ultra low dose, what we mean by that is actually total radiation. You may remember back from medical school that the amount of radiation required to induce cataracts is 10. This is 8 grey, so it’s less than the dose to induce a cataract. You’re giving it on consecutive days, 2 grey on each day. This has shown quite a good response. In one study, it showed that the overall response rate of complete regression or complete response of as high as 86%. Now, that is very impressive when we’re talking about the world of oncology. And partial response, meaning that you may have shrinkage of the lacrimal gland size, but you can still see slight enlargement or slight fuzziness around the edges, showing maybe some residual disease, about 14%. Like I mentioned, MALT lymphoma sometimes with systemic involvement — medical oncologists will even say to observe, because these tend not to be fatal, and they tend not to cause morbidity or mortality for the patient. Now, someone is asking: Do we need to take biopsy on both sides if it’s bilateral if we’re suspicious for lymphoma? And the answer is no. We don’t need to biopsy both sides. With that being said, there are situations where I may biopsy both sides, if I go to one side and find that I’m not getting a good yield, or the gland looks largely normal and I think I’m not gonna get a good result, we will sometimes go to the second side in order to get the diagnostic results that we want. So the strict answer is no, we don’t need to go to both sides. I usually go after the one that is larger, assuming that there’s more disease there, for a definitive diagnosis. So what about this patient? Who is a 53-year-old lady, presented to me with just swelling of the upper lid. And she says that she’s had headache for three days. Because of the headache, she received a CT scan in the emergency room, and you can see here the coronal and the sagittal presented on the right for you. Now, one of the things you may notice is that if you look very carefully, the bone seems to be missing here, or there’s something going on there. The mass is a lacrimal gland mass. But it seems to go beyond the lacrimal gland, and even into potentially the extradural-intracranial component. And you can see even more clearly on the sagittal, we have this ring enhancement, and it doesn’t look like the last one, cleanly, where it’s just lacrimal gland involvement. So it looks a little bit different than the last one. But in the same vicinity. Now, this was a diffuse large B-cell lymphoma. These are very common to have systemic involvement. So unlike MALT, where there’s a low risk — about 15% to 30% risk of systemic involvement — these are much higher, pushing 50% to 60%. They tend to be more aggressive and more rapidly growing, and they can be associated with visual loss, because of compressive mechanism on the optic nerve. So you saw the previous MALT lymphoma, where it was so big that it was causing the displacement of the globe. But even then, the patient has zero visual symptoms, meaning no visual loss, no color vision, no optic neuropathy. But with these ones, even if they’re not significantly large in size, they can — depending on the location — they can cause visual loss. These, however, the good news is they’re very, very chemosensitive. They are radiosensitive as well. Maybe a little less so than MALT lymphoma. So these we tend not to treat with low dose radiation, if we’re gonna irradiate the orbit. But these are exquisitely chemosensitive, and because of that, we tend to recommend chemotherapy first if there’s any systemic involvement at all. Here I list the regimen. I know people are gonna ask me: Which regimen do you use, in terms of chemotherapy, and my answer is that it really depends on the region and the degree of systemic involvement. So CHOP tends to be the first line for diffuse large B-cell lymphoma, but if there is significant systemic involvement, R-CHOP tends to be the first line for most centers. So check with your center locally in terms of what is the preferred first line therapy. Now, I mentioned that diffuse large B-cell lymphoma can cause optic neuropathy, and here’s an example now. This is not a case of lacrimal gland, but I think it’s important for you to recognize the difference between diffuse large B-cell lymphoma and MALT lymphoma and how distinctly different the two are. So this was a patient that had a mass wrapped around her optic nerve, as you can see here. The nerve is not displaced, but rather encased, and on the axial, you can see it’s definitely not the nerve itself that’s involved, but rather something around the substance of the nerve itself. And on the optic nerve imaging on the right there, you’re already seeing some subtle pallor on the left side, which is the top photo there, compared to the normal side. So this is an example of a patient who actually presented with NLP vision, solely from this mass that is involved. Now, this patient does have systemic involvement, and had received chemotherapy, R-CHOP specifically, and the orbit component did not respond, so they had to go on to second line chemotherapy in order to get orbital response. And any questions so far? Okay. If not, I’m gonna move and switch gears a little bit, to talk about epithelial tumors of the lacrimal gland. This is probably what most of the excitement around the lacrimal gland evolved, because of the new targeted therapy that is available on the market, as well as now that we’re understanding more and more about cancer genomics and its influence on malignancy as well as its influence on metastasis, by and large, in terms of epithelial tumor, we say about half of them are malignant and half of them are benign. The true number is a little bit less clean cut like this. So in some areas of the world, you’re seeing a little bit more benign, where it’s a 60-30 split. And in others, you’re seeing a little bit more malignant tumors, and malignant is 60% and benign is 30%. But I think thinking about it as roughly 50/50 is still a good rule of thumb. And of the malignant, we typically talk about adenocystic and benign pleomorphic. I’m not gonna go into too much detail about the other malignant ones as well as mucoepithelioid so we have time to talk about the exciting developments in epithelial tumors. So a little bit of background of adenocystic. We know 20% to 35% of lacrimal tumors are adenocystic. So this is definitely less common than lymphoma in the orbit. They tend to have more early PNI, meaning perineural invasion. There’s a high risk of recurrence with adenocystic and a high risk of mortality. In textbook, people tend to say 50%, but in some of the new treatment modalities or new literature coming out with survival of pushing 80%. And we’ll talk a little bit later on, when we go through the specific details of those studies why we think that may be, and whether that is generalizable to the whole world or not. In terms of subtypes of adenocystic, we know there’s cribriform and solid. And sometimes you’ll see terms like sclerosing, tubular — now, by and large I think clinically speaking, we tend to care about two patterns specifically, which is the cribriform plate, cribriform form, which is the most common, and it’s the one that is considered a little bit more good in terms of prognosis. And the solid, which is also called basaloid, is the one that we worry about a little bit more, since there’s some studies suggesting it’s the form that tends to recur or the form that tends to metastasize. Now, here’s an example of those two forms that I mentioned, so this was a 24-year-old patient of mine who presented actually with ptosis, and she said she had some intermittent sharp pain, and you can see on this photograph you may be able to — sorry, the facial photograph — you may be able to see some fullness in the sub-brow area. She’s pharmacologically dilated, so that is not a blown pupil. And on the CT scan, I give you two coronal only, because there are specific characteristics I wanted you to pick out. You can see that there may be some deficit in bone, in the upper photos here. So you can only judge bony defect on CT. You can see this mass is also pushing the globe down, but it’s not indenting the globe, and it kind of has these kind of scalloped edges, as I would like to call them, where it’s not so clean cut. You can’t really tell where the muscle ends and where the tumor begins. So this is one of the patients that had cribriform subtype of adenocystic carcinoma. Now, there was a question that is coming in, asking: Are traditional clinical and radiological features enough to differentiate pleomorphic adenoma from adenocystic? So I’m guessing that the question that Kupra is asking is: Can I tell whether it’s pleomorphic or adenocystic based on a clinical exam and based on, let’s say, a CT or MRI alone? And the answer is not always. There are some features that are useful to know. So I’m trying to point out some of them as we go along. So like I mentioned, if you see breaks in bone, like this patient, where there is very high suspicion of potentially a bony defect, these tend to be adenocystic rather than pleomorphic. The other thing is we’ll talk about some of the signs of adenocystic that you may try to pick up on imaging. This edging looks a little fluffy, so this may be part of the tumor since we can tell the muscle is here, and those tend to be things we tend to look for, to tell us that they may be adenocystic. Pleomorphic tends to be more well circumscribed. It’s a clear oval, the edges are sharp and distinct, and they tend to have molding, meaning the bone may have edges where it thins out, but there’s not a break in them. It’s smooth, not scalloped or moth eaten. But there are a portion of patients where you can’t tell. Those are patients you tend to do an incisional biopsy on. While we’re on the topic, because I know one of the questions that was sent in through the registration was: Should you do an incisional biopsy on lacrimal gland tumor? And the answer is: If you are not sure, you should. So let’s say that the imaging and the clinical exam is ambiguous, and you really are not 100% sure whether this is adenocystic or pleomorphic, and the tumor is really big, and you’re worried about potential posterior extension to the apex and potential vision loss just from your surgery alone… I would say in those cases I would state that you would need to do an incisional biopsy in order to consent the patient properly. Now, the question becomes: Well, what about recurrence rate for pleomorphic adenoma? So yes, there is a high risk for recurrence of pleomorphic adenoma, and when they recur, they tend to be multifocal. Meaning that what you end up seeing is these little foci that are left behind and regrow, which makes it harder to go back and clean out completely. And I have seen multiple recurrence of pleomorphic adenoma that get sent in to us in the end that are incredibly hard to clean out perfectly, but I think the risks of what people are worried about is truly malignant transformation. I think most of us who treat a lot of lacrimal tumors would agree that malignant transformation does exist, but they’re not as high as what we used to believe in. I think the transformation probably is somewhere around 15% to 20%. And some of the literature is starting to suggest that. The problem is we don’t have good prospective or retrospective studies to tell us the exact incidence of malignant transformation. But I think the risk is probably relatively low, and the risk of malignant transformation is not dependent on whether you did an incisional biopsy or not. I don’t have any evidence or proof of that, but this is anecdotally what we see. Whereas recurrence is much more tied to whether you have removed things completely or not. There’s one more question about the use of fine needle biopsy for pre- or intraoperative… I’m not sure why the question is intraoperative. But fine needle aspirates or biopsy for diagnosis. So there have been attempts at biopsying the lacrimal gland without an incisional biopsy. Most of us tend to move away from that for two reasons. One is that you don’t get as good of a sample, and you may end up needing to go in for incisional biopsy anyways. The other risk that is always in the back of our mind is the risk of an orbital hemorrhage. These adenocystic or pleomorphic are highly vascularized, and if you don’t have an incisional biopsy, you can’t really obtain hemostasis. So it’s a bit of a dilemma, and for me, the biggest reason I’m not much of a proponent for fine needle biopsy is because of diagnostic yield, where there is still, I would say, 30% of the time where you would need to do an incisional biopsy anyways. And these — you can do them through relatively small incisions. So I use a 5 millimeter to 7 millimeter incision, to do these biopsies, to get good specimens, in the lid crease. And those heal really well, so you really don’t have any reasons, in essence, not to just do an incisional biopsy. So here is a contrast of the previous patient. This is a 46-year-old female who presented with proptosis for about two months, and you see that clinical photo — there’s some displaced — you can sort of see some chemosis, and on the MRI to your right there, you can see a large mass that’s occupying I would say half if not more than half of the orbit. And you can see it’s very indistinct. So you kind of see this little tail. And this tail is very important, because this is probably the most telltale sign of adenocystic carcinoma. So if you see something like this, this is definitely not a pleomorphic adenoma. And you can sort of see that it’s quite large. It’s occupying the entire lateral orbit. It’s pretty rare for lacrimal gland tumor to cause displacement of the optic nerve, so these patients tend to have relatively good vision at presentation. And this is an example to you of a basaloid or solid adenocystic carcinoma. You can see that it’s very, very blue. So when I’m teaching my residents and pathology is always something that is really hard for them, I try to simplify to say: Is there a lot of blue or a lot of pink? A lot of blue means there are lots of cells there, because those are the nucleus. If there’s a lot of pink, those are usually the extracellular matrix or scar. This is definitely a highly cellular pathology, and that’s what solid is trying to describe. Or basaloid, meaning blue, meaning lots of blue on the slides. So moving on to a little bit about pleomorphic, they are about 50% of all lacrimal gland epithelial tumor, way more common than adenocystic. The old school data says there’s risk of transformation of 10% at 20 years and 30% at 30 years. I’m not sure if that’s true anymore, based on what we know today. Recurrence I do agree with. I would say recurrence is probably higher than reported in the numbers we have in the old literature, meaning literature that’s greater than 20 years. That recurrence can recur in 3% of what is presumed to be complete excision, and with incomplete excision, the recurrences are a lot higher. I would say 30% is probably correct in terms of incomplete excision. Now, the question of complete excision is, I think, what this number is telling us — is that even if you at the time of surgery think you’ve gotten complete excision, if pathologically it’s closer to the capsule, you have to assume it’s not, and then the question becomes complex in terms of… Do you go back in and clean things out more, or do you do frozen section at the time of surgery? The strict answer to that is we don’t really know, and it’s really dependent on the person who is treating this person and what they’re comfortable with. I don’t mean excisional biopsy. This is separate from biopsy. Let’s say you didn’t really know at the time of seeing this patient whether it was pleomorphic or adenocystic. What I tend to advocate for if you’re going back to do excision — you have to do complete excision, you have to excise out your previous incisional scar, for its entire depth, and remove any area that you have “touched”, quote-unquote, before, in order to make sure that you truly have gone around any potential for seeding. So a lot of the argument against incisional biopsy — and interestingly enough, I was just at a conference for orbital oncology, and even among experts who treat cancer only, this was a discussion point where people were arguing about what should be done. So I think the take-home message that I keep pointing out about this is that if you did an incisional biopsy, it’s not the end of the world, but you do have to consider anything you’ve touched from the first surgery as dirty, and if you weren’t the first one to do an incisional biopsy and then you see this patient, you have to assume that anything near your incision ends up being bigger and the reconstruction ends up being bigger. So this is an example of a patient with pleomorphic adenoma. This is a relatively young gentleman who you can see has significant displacement of the globe, and fullness in the sub-brow or superior temporal aspect of the orbit. Now, you can see that the imaging here shows you something completely different. So very well circumscribed. These tend to be very firm, even when you’re cutting it out, and it can indent the globe significantly in the imaging, which also is a bit of a telltale sign of being pleomorphic. Now, even higher up, it’s still showing a very distinct, well circumferenced, very clear cut. One question is: When you look on the coronal, Dr. Yin, you said this is a little fluffy. How do you know this is not an adenocystic? This is one of the commonest questions, when residents or fellows present cases to me, that they often will hang up on one image, and say… Well, but this proves you wrong. This has to be adenocystic, because on the coronal it looks a little fluffy. And this is when I tell them: You have to put the whole picture together. We never know for certain. Tissue is king. Until you get the tissue on the slide, you’re never 100%. There are always surprises. But you have to use your best judgment to put the story together. The clinical story of how long this has been going on, looking at patients’ old photos, and putting it all together instead of getting it hung up on one image alone. There are always artifacts. Maybe the patient moved a little bit in that one slice. So you put the whole picture together. Two out of the three of the images is telling you this is pleomorphic. This is a patient that I would remove the entire gland, assuming this is a pleomorphic adenoma. And they look a little different from the previous pathology I showed you. Now you see a lot of pink. So this is what we call bland-looking. So not a lot of cells. These little blue dots are the cells. It’s predominantly pink. You can still see preserved glandular architecture. The epithelium around the gland looks nice and thin and normal. So this is a benign tumor of the lacrimal gland. So in terms of adenocystic carcinoma, a big series out of Moorfields looked at their survival all the way back to 1972 to 2014, and showed that survival is still relatively poor with malignant epithelial gland tumors. Only about 50% at ten years and 60% at 5 years. Why is it that their latest series is showing a lower survival rate compared to some of the quote-unquote “globe sparing” surgery you might be seeing coming out of North America? We don’t really know, but my suspicion and my guess is that this is older patient data. So let’s say from ’72 up to 2000s, the patients were treated very differently, patients may have presented later, our techniques, chemotherapy and radiation, was very different. So you have to take this with a grain of salt. But in my mind the true survival or prognosis for these patients is a little bit higher than this. Now, a little bit about imaging. Besides CT and MRI, is there something else we can use to help us distinguish the different lacrimal gland tumors to help guide us in terms of what we tell patients? There is some interest in ultrasound, and I have to say that I haven’t used ultrasound as widely as some other people do, but I think it’s very interesting to note that for some of you who have difficult access to MRI or CT scan, that this may be a technique that you want to develop or get good at. One caveat is that ultrasounds are very technician-dependent, meaning that you have to train a very good technician, and be consistent with the technician, in order to get good yield of your ultrasound result. So sometimes I’m lucky to have a very good ultrasound technician who gives me beautiful photos and gives me very diagnostic pictures, but sometimes they are completely off the mark, and it’s really depending on who is doing the ultrasound that day. And there is some suggestion to think that at least for distinguishing epithelial versus non-epithelial tumors of the lacrimal gland, that the lymphoproliferative diseases tend to have a more preserved vascular architecture, and that’s what the septa and the tree-shaped vascularizations are talking about. You see this nice thing that looks like blood vessels. This shows you arterial flow as well as venous flow. With epithelial, there’s more disruption of the natural vascular architecture, where you see these blobs that don’t look like blood vessels, and those tend to be more epithelial tumors. You’re gonna ask me: You told me epithelial or non-epithelial. You didn’t tell me malignant or benign. That’s exactly the point. This ultrasound does not tell you or does not have enough evidence to show whether they can accurately distinguish between malignant versus benign. In this paper, they did try to talk about using combination of all the factors that are listed there, meaning resistance, septa entry, tree-shaped vascularization, in order to try to figure out: Can you subcategorize the epithelial into malignant or not and the non-epithelial into malignant or not? But I think the data is a little bit softer in that aspect, versus just distinguishing between lymphoproliferative versus epithelial tumor. So I hope as we get more experience with this, we can have better characterization of how to use ultrasound and potentially color Doppler with the ultrasound. Now, this is a correlation to tell you that these are the patients that they’re showing the lymphoproliferative and epithelial changes on. But I’ll say just looking at this MRI, I don’t need an ultrasound to tell me that this is not likely to be a pleomorphic adenoma. Ultrasound is great, but we don’t really know how to use it, in addition to our imaging technology that we have available to us right now. So what is available to us is CT and MRI, that we do have lots of experiences with. And as I alluded to earlier, there’s a posterior extension or this kind of tail, as I call it, in the back, and this was beautifully talked about in a paper by Jeff Rose, and they termed it a Wedge sign. And they were able to show that the Wedge sign is quite distinct and accurate in telling you that this is potentially adenocystic, rather than the pleomorphic adenoma. Now, how useful is the Wedge sign in distinguishing lymphoma versus adenocystic? Not as useful. Lots of lymphoma patients do actually have this Wedge or Tail sign, but it’s useful to know that the Wedge sign is correlated with carcinoma relatively well, so if you see that, you can go ahead and not worry about this being a pleomorphic adenoma. One of the other characteristics we talked about was bony erosion or excavation, excavation — the other word for it is remodeling. Meaning that it didn’t eat through the bone. Rather, it remodeled bone. Whereas erosion can be thought about as actually eating through bone or actually involving the bone itself. Calcification has been pointed out in several studies to be correlated with adenocystic carcinoma. Molding to the globe is a classic word that is used to describe lymphomatous lesion of the lacrimal gland, but I would have to say that adenocystic carcinoma also has a tendency to do that. Even though textbooks tell us that it tends to be indicative of lymphomatous proliferation of the lacrimal gland, that can also happen with epithelial tumors of the lacrimal gland. So with all of that information, now we’re gonna move on a little bit to treatment. So let’s ask the audience: Which of the following treatment modalities for adenocystic carcinoma — or let’s generalize it and say epithelial malignant lacrimal glands — which of the following do you think would improve survival? Is exenteration or cranio-orbital resection — meaning you not only remove orbital contents, but bone in addition to it — story of no prior biopsy, predominant basaloid histology, the use of adjuvant chemotherapy, or none of the above? So I’m glad that I’m seeing kind of… Oh, there’s more and more people thinking adjuvant chemotherapy. This is very interesting, because I was expecting you guys to tell me that exenteration has been shown to improve survival. I was not expecting people to pick adjuvant chemotherapy. But I can understand where the rationale is coming from. The reason I’m posting this question as such is because this is the biggest thing we talk about. So this was a beautiful study out of Moorfields that was recently published. Finally we have a good study to show us that exenteration, which is this line right here, actually does not have any prediction compared to globe sparing surgery and radiation. Which is this top line here. Now, are you gonna interpret this as globe sparing does better than exenteration? I would not really do that. Rather, I would just say that these two are statistically not significant. So they’re equivalent. But interestingly enough, the patient who got cranio-orbital resection, the line on the bottom, actually did worse. Is this because cranio-orbital resection is bad for the patient? And it causes the patient to have higher mortality? I wouldn’t draw that conclusion. Rather, I think the patients who ended up with cranio-orbital resection are from an older series, as this is from 1972, so what I would draw from this is that survival is equivalent whether you get an exenteration or not. What about factors that do affect overall survival? (audio dropped) one study that… Sorry about that. Sorry, guys. How do I go back? Thank you. So one study that looked at the correlation between residual tumor — they classified it as R0/1 together versus R2. R0 or 1 is the AGC way of categorizing how much residual tumor. We talk about gross residual, meaning that you know at the time of surgery that you can physically see tumor left behind, versus microscopic residual, meaning at the surgery you thought you took it all, but the pathology shows that there was actually tumor left behind or tumor extending to margin, so you’re assuming there’s tumor left behind, versus R2, where you can see that you didn’t take it all. When you have gross residual tumor, the survival was worse than those who had just microscopic residual, and this is kind of intuitive for most of us who treat cancer. And also, intracranial extension is kind of intuitive. That if there is intracranial extension, that we’re gonna assume that survival is worse, since we can’t get complete surgical excision in that sense. So in that sense, you’re having gross residual. So it’s kind of intuitive to think that these two will have worse prognoses, compared to those who don’t. But it’s nice to see a study that actually shows this to us. And then the other thing that does show worse survival that I alluded to earlier is the basaloid subtype, or the solid subtype. Whether that’s because they behave a little bit like morpheoform and it’s harder for us to get what’s left behind or not, we don’t know. Now, T staging: Most of the North American studies have shown that T2 and below versus T3 and above is kind of a cutoff point for us to prognosticate survival for these patients. So those who are T3 and less did a lot worse and were around the 50% percentile, in terms of overall survival at 5 years, whereas T2 and below were around 80%, sometimes 90% survival in North American studies. However, in the study by Moorfields, they did not see that distinction, and some of the other studies from other countries did not find the distinction between T staging being prognosticating. We’re seeing a bit of a split there and we’re not sure at this point why some studies are showing significance of T staging, whereas others are not. So what about innovative treatment modality? I think that’s what most people are most excited about. And we’re going a little bit over time, but I hope you’ll bear with me. I’ll say that before we even talk about this, you have to think about: Why are we innovating? The goal of innovation should not only be improvement in mortality, but more importantly for us, as ophthalmologists, we care about saving the eyeball and sight as much as we can. But one thing we sometimes forget to think about is the quality of life for the patient. Even if I can improve or extent the patient’s life by one year of life, you have to think about how much you’re impacting the patient’s quality of life. So if that improvement of one year of life is a trade-off of being hospitalized for six months, is that really still worth it for the patient? So I would caution people who are trying to push the boundaries of what we’re doing in terms of treatment to also think about balancing those three factors, so that you are truly designing an innovation that is good for the patient. So one of the most exciting things I will say for the past ten years is a number of studies coming out, looking at globe sparing surgeries. Now, Dr. Esmaeli is probably the first one at MD Anderson to start treating patients in this manner, and had very good results, with survival being at least equivalent to exenteration. What do we mean by globe sparing? We don’t mean you do a partial excision of the tumor. We mean that you just don’t do an exenteration, but you still do a big surgery, to make sure you do a proper lateral orbitotomy, and remove the cancer, unblock, with the purpose of reaching no residual tumor, and then followed by adjuvant radiation. Now, in her series, which — the latest publication actually combined her results with those from Korea, with Dr. Kim’s series, and this is probably the largest series that is so far today published on the outcome of these patients, and they actually showed that the patient has really good survival at around 80%. There is local recurrence, as you can see in the chart on the bottom, if there is gross residual, as we mentioned. So this kind of confirmed some of the other studies, looking at prognosticators for recurrences. That if you do have gross residual, the recurrence rate is really high. Whereas no or microscopic residual, the local recurrence drops to around 15%. Now, in terms of survival, you can see that disease-free survival, which is the one we probably care about the most, meaning that not only do patients live, but they don’t have any recurrences, is around 72%, so that’s great. But even more impressive is the overall survival. You see that we’re now pushing 90%. Now, this is a disease that we used to say 50% of the patients are gonna die at five years. Now we’re showing 90% or greater than 85% surviving past five years, so this is really, really great to know. Now, what does RFS mean? RFS means recurrence-free survival. Meaning that this is even more specific than disease-free survival, because disease-free survival encompasses those who have recurrence, as well as who we know we left disease behind. So there never was cure of disease and we never had complete response in essence. So this is more specific to: People may have had no disease after surgery and radiation, then had a new full sized disease — it’s around 44%. And in 90% of this, the patient had eye preservation. Why is that not 100%? It’s not 100%, because this is a study that looked at the final outcome. So let’s say if the patient recurred, in this category, this group, then they may have gotten an exenteration as a rescue therapy, so that’s why some of these patients did end up getting exenteration. One of the concerns when this study was getting designed was the risk of radiation to the eye. A lot of radiation oncologists, even some of those of that I worked with at first may be resistant to radiating the orbit that has a globe there with high dose. And part of the reason is that they’re worried they’re gonna cause permanent damage to the eye and you end up losing the eye as well. Now, I think this strategy of keeping the eye in and radiating is possible because our radiation technology has also improved. So I would say 50 years ago, we probably wouldn’t be able to save the eye if we had radiated with the eye in the socket. But now with all of our IMRT, IMPT, and all the stereotactic radiation we have available to us, we’re able to achieve a radiation field that minimizes damage to the eye. So if you look at some of the more recent studies in terms of side effects to the eye, we’re noticing that dry eye is by and large pretty universal. But you can see that retinopathy has dropped significantly, the need for enucleation is very rare, in terms of a globe that is painful and blind, that needs to be removed. And more than 68% of the patients have final vision that is even better than 20/40. One of the important things to remember out of this is that a lot of patients would rather have their eye in, even if it doesn’t see, versus having an exenteration. So even though only 60% or 68% or 70% of patients are getting more than 20/40 vision, there’s a greater number of patients who are happy even if they’re not seeing as good as 20/40, or not having a functioning eye, to just have a cosmetically pleasing and symmetrical eye in their socket. Now, what about specific to the cornea? Because I would say corneal side effects are what leads to enucleation or a non-functioning eye in these patients. Risk of optic neuropathy is relatively low with these patients, and because we’re radiating the lateral quadrant of the eye only, even if you have radiation retinopathy, for that to impact your central vision is low. So really the biggest side effect we’re worried about is actually cornea. So corneal perforation is only about 13% of these patients. Although acute keratoconjunctivitis, meaning dry eye, painful eye, conjunctivitis is actually quite common, and in my patients, I tend to tell them actually it’s universal. Usually when I’m counseling a patient who is about to undergo radiation, I tell them to expect it, and if they don’t have it, it’s rare. What about intra-arterial chemotherapy? The debate between globe sparing alone versus intra-arterial has been ongoing. And I would say the big group driving intra-arterial chemotherapy is a group from Bascom Palmer. Originally he did intra-arterial, followed by exenteration, followed by systemic chemotherapy with radiation, followed by systemic chemotherapy. So I think I’m belaboring the design of their study, because I think a lot of people assume that intra-arterial chemotherapy means you do intra-arterial and that’s it. You’re done. But that is actually not what his study was designed to do. His study was designed to actually improve survival and in one essence I sort of use this term: That he basically threw the garden sink at the patient. So he used everything we possibly have, to try to increase that survival from 50% that we have historically to greater. And if you look at his study, he then subdivided in his second paper on this protocol to two groups. I’m gonna focus on the overall one first, because the separation between group one and group two is slightly artificial and retrospective in nature. So the study was not designed from the get-go to say there’s a group one and there’s a group two. But rather, after the completion of the study he went back and looked at the data and tried to parse out the two groups to figure out which group did better. As an epidemiologist and as a scientist, we always put a lot of caution on people trying to reparse data, because you’re building in inherent biases. I’m not saying the study is not valid. I’m just saying take it with a grain of salt in terms of how much we believe in the study. However, you can trust the overall. So overall, he did improve survival, where 84% of the patients are now alive at 5 years, compared to 50%. So I would say intra-arterial chemotherapy has been able to achieve, in his protocol, as high of a survival as the globe sparing series out of MD Anderson. However, you have to remember these patients had exenteration and had systemic chemotherapy and had radiation. So once again, just kind of belaboring the point that this is a significant improvement compared to what’s conventional, conventional meaning exenteration. And these two lines is then parsing out whether patients actually followed strict protocol or not. As I mentioned, the protocol is quite hard. So they had a lot of patients falling off protocol because of intolerance to chemotherapy or intolerance to radiation. And the next kind of little three slides is gonna talk a little bit about personalized medicine. So for those of you who had… (audio dropped) sorry. I keep accidentally muting myself. Just to get you guys started about lacrimal gland epithelial tumor and the genetics of it, I’m gonna start off with a question. So the question is: Whether you think the genetic profile — the genetic mutational profile of lacrimal gland epithelial tumor, namely adenocystic, is the same profile as salivary or parotid gland. We’re kind of split on this. Half of you think it’s true, half think it’s false. So the old school thought is that parotid gland tumor is similar to adenocystic tumor. However, the clinicians who treat this a lot, namely the orbital surgeons, we tend to think they behave a little bit differently than parotid gland tumors. There are some similarities, but there are some differences. It was reassuring to see some studies from a pathology group looking at microRNA expression, and showing that there is actually difference between lacrimal gland versus parotid gland tumor. So you can see the clustering here with the blue on the bottom — these are the lacrimal gland tumors. The yellows are salivary gland tumors, and the greens are breast. You can get adenocystic carcinoma of the breast as well. So you can see how the breasts cluster to one end, the salivary glands cluster to another, and it’s nice to see this nice clustering of the lacrimal gland tumor. So it’s showing us that even though there is overlap that there is still something different about the lacrimal gland adenocystic carcinoma, compared to the others. And there are some miRNA differences. Namely 2 that are upregulated in these lacrimal gland tumors. The black ones are the downregulated ones. You’re gonna ask me what does this mean exactly. Tell us that there’s a targeted therapy. I’m gonna tell you that we don’t quite know what these miRNA expressions are telling us in terms of which pathways specifically that we should focus on just yet. And just to cloud the picture a little bit more, there’s been three series looking at genomics of lacrimal gland adenocystic carcinoma. One out of Mayo Clinic, one out of Bascom Palmer, and one out of Anderson. But with that being said, they’re showing different mutations. The Mayo Clinic showed that lacrimal gland adenocystic is similar to parotid adenocystic. They’re showing rearrangement in MYB. But with MD Anderson, they’re showing the RAS or KRAS being mutated, and the Bascom Palmer is showing the NOTCH pathway being different. So they’re treated completely different with targeted therapy. The MYB, upstream or downstream from that, has specific targets that may be able to be targeted. There are more KRAS therapies in clinical trial, which also has an overlap with RAF or MAP kinase. Whereas NOTCH pathway is targeted completely differently, with a different set of inhibitors. So depending on which pathway is driving adenocystic, there may be completely different drugs that may be targeting these patients. Now, on top of that, can we use genetics to try to help us distinguish potential for different tumors that are implicated? Why do we care about this? And we care about this because you may have heard a term: Liquid biopsy. Going even beyond fine needle aspirates, what if we don’t have to touch the tumor? We can avoid biopsying the tumor altogether, and then looking at people’s serum or blood or maybe even tears, as a way of distinguishing what tumor we’re dealing with. That’s the so-called liquid or non-cellular biopsy. In order to do that, you have to be able to figure out what you’re looking for. So what non-cellular targets you’re looking for, in order to be able to do or design liquid biopsy. So one study looked at trying to distinguish the different epithelial tumors, based on STAT3 or the IL6/JAK/STAT3 pathway. And it’s interesting that they showed all the way on the right there a potential target for us to design liquid biopsy, so you don’t have to touch the gland, to know whether it’s a carcinoma or pleomorphic adenoma. This is just still in benchwork. It’s not ready for a pilot clinical study, but I think in the next ten years one day we can have the dream of not having to touch the tumor and being able to figure out what the tumor actually is. What is the immediate next step for us to look at? I think showing you how there are three different pathways implicated — there are some interactions between these pathways, and the pathways are quite complex, as you start to dive more and more into it. I think it’s good to leave you with this last slide, which is a good way to think about the genomics of lacrimal gland tumor. This is actually something that I’m trying to get established at MSK, with trying to establish — hopefully getting all three sites, meaning Mayo Clinic, MD Anderson, and Bascom Palmer, to collaborate on looking not only at MYB and FNYB, but at other gene mutations we haven’t even considered. NOTCH being one of them, but there are other pathways, such as how does mTOR play into this? There are targets for mTOR inhibitors currently on the market. And looking at interaction with RNA and mRNA expression. And in cases where we just can’t figure it out, whole-exon sequencing. So when you’re looking at genetic analysis of a lacrimal gland, how do you go about thinking about the interaction between all of these tests that we hear on the market? What is FISH analysis? What are next-gen assays, versus whole-exome? Thinking of things that may not be mainstream. And you have to think about microRNA impacts the expression of all three of these levels of genomics. So with that, I’m gonna leave you with a couple of summaries, just to kind of sum up what we talked about. So remember that low grade lymphoma, as well as mantle cell, because they are very radiosensitive, can be treated with low dose radiation, as low as 4 grey, and some people are even pushing the envelope to 2 greys. Exenteration actually does not increase survival for lacrimal gland epithelial tumors. And that globe sparing plus radiation has been equivalent to intra-arterial plus exenteration plus chemotherapy plus radiation. So both of those are equivalent in terms of survival, and both are valid to offer your patients. Think about, in terms of genomics of lacrimal gland that microRNA expression is distinct for lacrimal gland and we have to think of ourselves as distinct from the head and neck surgeon when dealing with lacrimal gland tumors. Kras, Myb, and Notch have all been implicated, and we may want to look at STAT3 as a possible target when dealing with these tumors. So I’m sorry I went over a little bit, but at this point, I’m gonna ask if anybody has any last minute questions that they want to ask me. And if not, I’m gonna say thank you guys very much. You have my email on the last slide, so if you have any additional questions that you think of later on, feel free to email me. I’m always happy to answer questions via email. Thank you.
February 1, 2019