 Our next presentation is by Nina Boll. She's a student at Thomas Jefferson University. She's gonna talk to us about conjunctival tumors and benign versus malignant counterparts. All right, so like you said, my name's Nina Boll. I'm from Thomas Jefferson University in Philadelphia. Today I'm gonna be talking about some research that I did with Dr. Carol Shields and the rest of the Ocular Oncology Service at Welles Eye. The title of our research is conjunctival tumors in 5002 cases. It's a comparative analysis of the benign versus their malignant counterparts. So this research was published this year in the American Journal of Ophthalmology with Carol Shields. So a quick introduction to some conjunctival tumors. There's a wide variety of tumors that can be in the conjunctiva more commonly. You can see melanocytic tumors such as PAM or melanoma, epithelial tumors like CIN, SCC, vascular tumors, lymphoid tumors, and then less commonly, chorostomatous, fibrous, neural, lacrimal, just to name a few. Obviously there's quite a range of tumors that can be seen. The three most common conjunctival malignancies arise from or resemble some sort of benign or pre-malignant counterpart. So you have primary acquired melanosis or PAM with its melanoma counterpart, conjunctival imptraepithelial neoplasia or CIN with its SCC counterpart, squamous cell carcinoma, and then benign reactive lymphoid hyperplasia, BRLH, and lymphoma. Also the purpose of our research was to evaluate conjunctival tumors in a large retrospective series of 5002 cases. Specifically we wanted to look at the common malignancies and then the benign and pre-malignant counterparts. The purpose being so that it would be easier to distinguish between the two because oftentimes it's difficult to see oh, is this CIN, is this SCC, et cetera, and really see what the major features are both. So the methods, we reviewed all of the patients that were seen in the Ocular Oncology Department that had a conjunctival tumor. It ended up being about 5002 cases. We reviewed charts from April 1st, 1974 until August 1st, 2015. So we basically took down patient tumor features such as age, race, sex, the tumor size, like the location, if it's along the limbis, if it's farther from the limbis on the tarsus, how big it was, the color, the shape, there's feeder vessels, et cetera, and then both the clinical and the histopathologic diagnosis. All these features were analyzed and we created nine tables. So the data is actually quite extensive and I obviously don't wanna go through every chart because that wouldn't be very interesting. So just some selected results that I thought were interesting. So of the 12 most common tumors, if you look at the location of them, there really is a pattern that goes along with seeing where they are. There's an equivalent nasal versus temporal location of Nevi, CIN, and SEC. So it's not more likely to be nasally or temporally that about even for both. If you look at the percentages, PAM tends to be more temporally located. BRLH tends to be more nasal. And then along the limbis, you're more likely to see Nevi, PAM, melanoma, CIN, SEC, and dermoid tumors. And then more remote from the limbis, you'll see other tumors like papilloma, lymphoma, BRLH, dermalipoma, and pyogenic granuloma. So jumping into kind of the meat of the presentation, which is the benign and malignant counterparts. So first looking at PAM versus melanoma, there were 1,224 total cases of both of these tumors seen at the ocular oncology department from 1974 to 2015. So we did a relative risk evaluation. So it's more likely to be melanoma versus PAM if you have a tumor that's more than one millimeter in thickness, 2.4 is the relative risk for that. There's no pigmentation, which is actually kind of interesting for me as a medical student. The lack of pigmentation makes it more likely to be melanoma. If there's feeder or intrinsic vessels, if there's a presence of cysts, specifically PAM doesn't have any cysts. There were absolutely no cases of PAM that had cysts and 7% of all the melanomas that they saw did have cysts. So it's not all the melanomas, but definitely no PAM. And then if there was presence of hemorrhage, the patients were older, and then if the tumor actually was along the Tarsus 1.6. So putting all of this information together, say you have a melanocytic patient and a patient who's older than 60 years old, the tumor is more than one millimeter thick, has feeder and intrinsic vessels. It's like, it's 24 times more likely to be melanoma than it is to be PAM. So it's an easy way to differentiate based on just tumor features. So these are all melanocytic tumors. The top two are just Nevi, so you can see it's the top left image. It's pigmented along the three o'clock limbus. It's pretty easy to see. The top right is a little more faint. It's also a Nevi. The middle two images are both PAM. You can see it's more widely dispersed. It's along the limbus, definitely no cysts and it tends to be flatter, less nodular. And then the bottom two images are some angry-looking melanoma. The bottom left is nodular. You can see some feeder vessels that are engorged. And then the bottom right, this is a black patient who had, so the nodular area towards the right ended up coming back as being melanoma. And then the rest of the tumor that was along the limbus was just PAM, so he actually had both in his eye. So the next benign and malignant counterpart. So for CIN and SCC, there were 715 total cases that we looked at and the greatest relative risk for being SCC versus CIN. So if it's more diffuse, if there's any sort of brown pigmentation, it didn't have to be the whole tumor, just kind of some spots. It made it more likely to be SCC if it's more than 10 millimeters in basal diameter and then finally if it's more than one millimeter thick. A lot of these things are pretty intuitive, but I think that the combination makes it more interesting, is if you have this epithelial tumor, it infiltrates the surface of the eye. It's greater than 10 millimeters in diameter, more than one millimeter at thickness, and then there's some sort of region of pigmentation. It's 4.3 times more likely to be SCC than CIN. So a little bit less impressive of a differentiation between the epithelial tumors and the melanocytic tumors. So these are some images. The top two are CIN, the bottom two are SCC. The top left is not as clear to see, but you can see there's no pigmentation. There's maybe some vessels, but it's not very clear. The top right, it's much more, you can see that there's a tumor there. That's also CIN. The bottom left, you can see there's some area of brown pigmentation, definitely thicker feeder vessels and intrinsic vessels. The bottom right, this patient has pretty extensive SCC. There's some leukoplakia. It's really involving the bulbaric conjunctiva and this patient was a heavy smoker, so it kind of makes sense that the tumor is more extensive. And then the last differentiation of benign amyloid and counterpart. So BRLH versus lymphoma, or benign reactive lymphoid hyperplasia. So there are 465 total cases that were seen in the ocular oncology department. The greatest relative risk for being lymphoma is that if the patient's older, so that was actually quite a big relative risk 2.6. If the tumor was both nasal and temporal, it's more likely to be lymphoma, which is, this is a little bit confusing, but basically BRLH tends to be nasal. So if it's just nasal, it's less likely to be lymphoma, but if it's, then it's more likely to be a BRLH, but both it's lymphoma. If it's more diffusely involved, if it's located superiorly and inferiorly along the tarsus, the tarsal conductive, then it's more likely to be lymphoma and then the large basal diameter. And taking it together, a lymphoid tumor in patients who's older than 60 years old, if the tumor is located both nasally and temporally with a basal diameter more than 10 millimeters, there's a 6.2 times greater chance of being lymphoma over BRLH. So the top two pictures, these are in a child who has BRLH. You can see it's located nasally. Honestly, these two tumors are quite difficult to differentiate. You can see there's like the salmon patch coloration, but the location does suggest that in a child it's more likely to be BRLH. And then the bottom two pictures are lymphoma. It's really extensively involving the superior nasal conjunctiva. So there are some obvious limitations to this study, one and probably the biggest one being the fact that this is an ocular oncology service. So all of the patients that we're seeing are referred to the ocular oncology service. It really over represents the serious malignancies, such as melanoma and then they're not really looking at things like racial melanosis or maybe some of the smaller nevi that are more classically nevi and not maybe it's melanoma. We also didn't look into long-term follow-up and we didn't really look at treatment differences or prognosis. There was one table that kind of just tabulated how they treated it, but didn't really do any extensive analysis on that. So these are the conclusions and basically it's just a summary of all the differentiating features that I talked about before. I won't read them out again, but I do think it's interesting to point out that the differentiation between PAM and melanoma was much more significant with those features, 24 times greater risk if you have, if you're older than 60, more than one millimeter. And then really the presence of feeder and intrinsic vessels helps to differentiate the two.