The cloud. And then I'm going to start the webinar now. And like I said, we'll delay about 30 seconds to a minute and let people join. Good luck. Thank you. Hi. Hello, everyone, and welcome to the IGCS pathology webinar titled Intraoperative Diagnosis of Ovarian Tumors, Back to Basics. We're thrilled to have representation from over 80 countries joining us today. My name is Dr. Sakina Thiriay, and I'm a gynae pathologist and cytopathologist in Toronto, Canada, and I'll be moderating today's session. The format for the webinar will be a 45-minute presentation followed by a 15-minute Q&A segment at the end. During the presentation, if you have questions, please type them into the Q&A feature found at the bottom of your screen, and we'll do our best to address as many of these as possible during the discussion. The recording will be available in the IGCS education portal. I am pleased to introduce today's presenter, Dr. Angela Rolte, who is originally from Mizoram in northeast India, and she earned her MBBS degree from St. John's Medical College, Bangalore, and MD in pathology from the All India Institute of Medical Sciences, New Delhi, and completed her training for the FRC path in Newcastle, United Kingdom. She is currently the lead gynecological pathologist for the Northern Gynecological Oncology Center and also the lead cytopathologist for the Northeast and Yorkshire Cervical Screening Center, both based in Gateshead, UK. She has a great amount of experience in the care of women with gynecologic cancers and will share tips on providing intraoperative diagnostic service, specifically in low-resource settings. Please join me in welcoming Dr. Angela Rolte. Dr. Rolte, if you'd like to share your screen, and I hand the Zoom podium over to you. Right, hi, are you able to see my screen? Can you see my screen? Yep. I'm just gonna, I just, right, I'm gonna do this again. I just shared it, but it's not allowing me to share it now, one moment. I'm going to try again. I'm clicking the share button and I could share it when we had our practice, but... I'm going to try again. Right, can you see it now? Yes, now we can see it. I just had a bit of a moment there. Right, okay. Let me start my slideshow. Right, is that okay? Can everybody see my screen? Yes, we can see it and we can hear you fine, thank you. Lovely, thank you. Many thanks Dr Thiriyayi for your kind introduction and thank you to the IGCS online education team for this exciting opportunity to talk to you on intraoperative diagnosis of ovarian tumours back to basics. I will begin with a brief overview of ovarian tumours with a short literature review on frozen sections followed by illustrative examples of selected cases that have been a recurring problem both from literature and from our own experience. I will use these examples to provide you with tools and techniques for intraoperative diagnosis where frozen section facilities are not available. Although a significant proportion of ovarian cancers present at an advanced stage, there are also considerable challenges for the diagnosis and management of early stage disease as it is often not possible to determine if an ovarian mass is benign, borderline or malignant as they may have similar tumour marker profile and radiological appearances. The majority of frozen section requests we receive offer early stage disease in women with the risk of malignancy index of more than 200. Frozen section is used in these cases to confirm a diagnosis of malignancy. If malignancy or at least borderline is confirmed, staging procedure can be performed at the same time eliminating the need for a second procedure. Staging surgery includes dotal hysterectomy, omentectomy, pelvic and pariotic lymphadnectomy and peritoneal biopsies. Whereas staging procedure is not only unnecessary but may also contribute to the subsequent mortality and morbidity in metastatic ovarian tumours. And in the presence of disseminated malignancy, frozen section is used to confirm a diagnosis to confirm gynaecological origin before undertaking radical cytoreductive surgery. These would be for cases where preoperative diagnostic biopsies were either not possible or insufficient due to lack of accessible sites for biopsy or at laparotomy is found to have an unusual distribution or odd appearance for primary gynae origin. The accuracy of intraoperative frozen section diagnosis of ovarian tumours have been examined in various studies. The largest series to date is from our centre by my predecessor Dr Paul Cross which is currently being updated and many thanks to Manisha Varnaker for this slide. A subsequent quantitative systematic review of 38 studies by my gynae oncologist Nitya Ratnavelu demonstrated that the concordance rates between frozen section and final diagnosis were 94, 79 and 99% for both benign, borderline and malignant diagnoses respectively. The challenge to diagnosis of borderline tumours even for experienced pathologists is demonstrated by the fact that 21% of borderline tumours diagnosed on frozen section turned out to be invasive cancer at final diagnosis. The discordant results between intraoperative frozen section diagnosis and final diagnosis have been either due to sampling errors or interpretation errors. The key elements to frozen section diagnosis of ovarian tumours are clinical to include imaging tumour markers and intraoperative findings, careful and thorough gross examination with selection of the most appropriate area to sample for frozen section and microscopy to include frozen section histology and intraoperative screen cytology. Lastly and in fact perhaps the most important element especially in difficult cases is the communication between the surgeon and the pathologist. Still there are challenges to ovarian tumour frozen section due to the limited number of tumour samples and tumour heterogeneity especially in large ovarian masses. To overcome some of these challenges some pathologists advocate more tumour sampling up front. Now whilst this can be helpful the morphologic heterogeneity and the morphologic overlap between the various tumour types, poor cytomorphologic details either due to freezing artefacts, nature of the tissue or variable technical quality can still make optimal microscopic interpretation quite difficult. Instead I find that recognising the range of gross appearances of various tumour types very helpful but the problem is that there are limited gross images of ovarian tumours particularly in the fresh state available both in literature and textbooks to allow for picture matching which is what most pathologists do at the start of their career and even later and this is what I will be showing you a lot of today. The other tool that I find incredibly helpful in improving diagnostic accuracy is tumour scrape cytology as one is able to obtain material from a wider area with preserved cytomorphological details. There have been several reports demonstrating the accuracy of intraoperative cytology as a complement and in some situations as an alternative to frozen section histology in the intraoperative diagnosis of tumours from various sites including several attesting to its accuracy in ovarian tumours. In fact the most recent paper is from a Japanese group it was earlier this I think was March 2021 where they have reported on the superior diagnostic accuracy of intraoperative cytology to that of frozen section histology and I must admit I have to concur with their report as I've been using intraoperative cytology as an adjunct to frozen section histology for several years now and when COVID first happened and our frozen section service had to be suspended due to it being an aerosol generating procedure we continued to provide intraoperative diagnostic service by replacing frozen section histology with intraoperative cytology alone and during the peak in the early days of the pandemic we achieved 10 concordance between intraoperative cytology and final diagnosis in all 10 cases which means 10 patients did not have to have a second procedure. All 10 patients underwent appropriate clinical staging and most of the diagnoses were made on careful gross examination and confirmed in cytology. The texture and consistency of the scraped material also contributed to the clues for diagnosis. I know the numbers are small but it still provides proof of concept and we were unable to accurately subtype one or two cases but that's not really crucial for intraoperative management and fortunately no one caught COVID during the procedure and the turnaround time was much quicker than frozen section histology. This was our first case, what I call our first COVID pandemic case examined under a ventilated hood with full PPE. As you can see the specimen is sliced under a hood, cut surface is solid cystic with hemorrhagic contents, the solid areas have this soft fleshy and slightly spongy consistency but mixed with these sort of bulbous papillary areas which were selected for scraped cytology. After blotting away excess fluid and blood which can obscure cytological details, cytology sample was obtained by pressing the edge of the glass slide gently onto the freshly cut surface and scraping the surface with the edge of this glass slide. This was then smeared gently onto the surface of another clean slide. Smears thus prepared were immersed, were air dried and immersed in DIFQUIC fixative solution A followed by staining with DIFQUIC solution B and C for about 10 to 15 seconds, each blotting out excess fluid between each step. DIFQUIC is a proprietary brand of Romanovsky stain often used by cytotechnologists for rapid one-stop assessment of cytology samples and this is quick, taking no longer than two to three minutes and is easily carried out by the pathologist without the need for experienced personnel and during the pandemic I did most of this myself. The only additional material needed for the procedure is this stain, slides and water, so relatively cheap and easy to do anywhere and these stains I put them in a Copland jar and they have a long shelf life, the only thing is the alcohol tends to evaporate over time. This specimen grossly you can see is an obviously malignant tumour with a gross differential diagnosis between two tumours associated with endometriosis, namely clear cell carcinoma and endometrioid ovarian carcinoma. In this case I would favour a clear cell carcinoma due to this characteristic spongy appearance and the cauliflower top-like intrastick bulbous projections. This is the scrape cytology showing classical features of clear cell carcinoma where the cells are loosely cohesive with abundant fragile wispy cytoplasm that is either granular or finely vacillated with minimal nucleiform pleomorphism and deep magenta coloured stromal hyaline cores. And this was confirmed on the frozen section, sorry in the permanent section. Now on to the next case, this was a solid ovarian mass from a postmenopausal lady with a slightly raised CA125 showing this fibromatous, spongy and edematous areas. The edematous areas had a grape-like vesicular appearance almost resembling an edematous hydropic or hydropic chorionic villi. The image on your left is the fresh image with the fixed image on the right for comparison and as you can observe some of the features are more subtle and difficult to appreciate in a fresh specimen, particularly this soft velvety area. This is the frozen section histology on your left and the permanent section on the right. Now note the fibromatous stroma containing the scattered widely spaced glands and cysts that are lined by a single layer of benign looking epithelium. If you look at the epithelium you can see that nuclei are small, almost pycnotic and the single layered in the cytoplasm is quite difficult to make out and this would very well do for a benign fibroma. And the screen cytology from the fibromatous areas showed this flat honeycomb monolayered benign looking happy sheets, whereas the cytology from the spongy and vesicular areas showed this the loss of the honeycomb pattern with more variation in nucleus size and shape and hyperchromasia, a bit like a drunken honeycomb but still relatively monotonous. Again I don't know if it's projecting but there are subtle stromal highline pores in the background. I'm just going to go back to the previous and I don't know if there are cytopathologists that have dialed in but this reminds me of an acinic cell carcinoma really, I think that's the best comparison I can arrive at. So the frozen section or the frozen section diagnosis was, I said I'm not really sure what this is because there are benign areas mostly represented on the histology but the cytology shows is really atypical areas that look like a clear cell carcinoma. So essentially I said it could be benign, borderline and I can't rule out a clear cell carcinoma. But anyway the patient underwent staging procedure, I think she had a total hysterectomy, bilateral cell bingofrectomy, fomentectomy and pelvic and periortic lymphadenectomy. Now this is the final histology showing the tumoral heterogeneity. In this low power you can see the adenofibromatous area, these bland glands set in this fibrous stroma merging into borderline clear cell areas and finally into the clear cell carcinoma areas, most likely the tubular cystic subtype. Clues to the heterogeneity were not well represented in the frozen section histology but is demonstrated by both the gross appearance and intraoperative cytology due to wider sampling. Again this is the clear cell carcinoma with the tubular areas with the corresponding cytology and another area from the same. Immunohistochemistry is not really needed in this instance because the morphological features are quite typical but we normally do, usually in difficult cases, HNF1 beta, amica, napsin and WTA1 and B53 just to rule out a serious carcinoma but this was done just for demonstration purposes. What was interesting was HNF1 beta were also positive in an adenofibromatous area that looked very nondescript and didn't have the typical clear cell area. The tumour showed wild type B53, negative for WTA1, they were all positive for Pax8 including the adenofibromatous area but actually what was interesting was ER or estrogen receptor was positive in the adenofibromatous areas but negative in the carcinoma and the borderline areas. So really, this is a non-difficult area of ovarian tumour frozen section and especially tumours that develop from stepwise progression from a clear cell adenocarcinoma via probably the same pathway, ARID1A, PIK3C and mTOR mutation, similar to endometriosis but without the clues provided by associated endometriosis on gross examination. And this is the one area where we've had the most concordant results between benign and malignant diagnosis at frozen section and final diagnosis. The next case is an 18 centimetre solid cystic mass from another postmenopausal lady with a slightly raised CNR. So this is a non-difficult area with a slightly raised CA125. The specimen capsule is examined for deposits, disruptions and adhesions and as you can see this is a largely intact bocellated cystic mass with a small area of capsular disruption and adhesions. Slicing reveals two distinct components, one showing this trabeculated microcystic appearance composed of numerous small thin and slightly thick walled cysts containing nucleoid material. Note that the cystic structure and integrity is retained without flopping and collapsing over when the cyst contents are released. The other component is smooth, homogeneous, lightly tanned solid mass with a very soft consistency and if you were to do the touch test for state that you do it would be the consistency of your cheek and very soft cheeks I would add. Two sections were taken from histology, one from the cystic and one from the solid area and scrape cytology was obtained from both components. This is the frozen section of the microcystic trabeculated area showing these irregular dilated glands containing bluish tinge nucleoid material within the dilated lumina. As you can see the glands are lined by single layer of rather benign looking either flat cuboidal or columnar epithelium with dark but bland non-stratified nuclei. The solid component on the other hand shows cells that are loosely arranged in sheets, quite fragile looking cells with little supporting stroma and where the stroma is present it's delicate loose canidomatous. The cells themselves are polygonal with abundant eosinophilic finely vacillated cytoplasm with a relatively uniform mild to moderate nuclear atypia. The scrape cytology reflects the cell scene on histology, again comprising these loosely cohesive acinic carcinoma like cells with relatively uniform mild to moderate nuclear atypia. Note the eosinophilic highland core globules and the mucin in the background. Based on these features a diagnosis of an intraoperative diagnosis of clear cell carcinoma cannot exclude mucinous component was made. This is the permanent section of the microcystic component containing the cystically dilated glands containing mucin. As you can see the nuclear atypia is well much better observed on the permanent section and in fact there is an area of invasion from borderline to invasive here. And this is a mucin stain called Alcyon Blue DPAS highlighting the abundant intraluminal mucin present within the gland lumina. And this also demonstrates that mucin can be present in many non-mucinous ovarian tumors including serous clear cell and endometrial tumors. This is the low power view of the solid component. On high power you can see these cells possess abundant clear cytoplasm with distinct cytoplasmic margins and mild to moderate nuclear atypia. A few cells possess small but distinct nucleoli. Mitotic activity is scanty and nuclear atypia is moderate at most. And a final diagnosis of a clear cell carcinoma arising from borderline clear cell tumor was made. Again we don't really need immunochemistry but I did this for demonstration purposes. As you can see this tumor is positive for HNF1 beta and Napsin negative for Amica and p53 is wild type. So clear cell carcinomas are usually positive for one of these three markers and they should be negative or show a normal wild type p53 although there will be rare cases where they can be positive. Right the next case is again another post-menopausal lady with a complex pelvic mass slightly raised to 125. This is another example of a clear cell carcinoma and this time it's predominantly cystic with bulbous intra-cystic polypoidal projections and the gelatinous vesicular areas and sort of edematous looking character to the tops of some of these papillae which really reminds me of the top of a cauliflower. The frozen section histology showed ribbons and glands of papillae lined by again relatively monotonous medium-sized cells with moderate eosinophilic cytoplasm and uniformly mild to moderate nuclear atypia separated by edematous pale stroma. The nuclei are round to oval hypochromatic bit small but distinct nuclear in areas. The scraped cytology shows again classical features of clear cell carcinoma papillary clusters with this beautiful magenta colored stromal highline cores the cytoplasm in this case is a bit more dense than the previous ones, and there's a little more nuclear pleomorphism. Again, beautiful cracking stromal highline cores. The final diagnosis of a clear cell carcinoma was made based on gross frozen section histology and cytology. And I think in many of these, in all of these cases, you didn't really need histology. I think you could have made a diagnosis both on the gross and on the cytology. The permanent sections show a tubular papillary pattern of clear cell carcinoma, and you can see these highline, hyalinized papillary cores that were so beautifully seen on cytology preparation as dense magenta cores. Although if you remember what the frozen section histology looked like, they were just pale and very nondescript. They are not as clear cut, and therefore it's difficult to pick them out. Again, immunochemistry not required, but performed for demonstration purposes, showing HNF1 beta, which shows a nice, crisp nuclear positivity. And napsin is positive in this case, and it's usually very patchy, and it has this sort of granular cytoplasmic positivity. Another example of a clear cell carcinoma in this instance, it's showing predominantly solid type or solid appearance, and the consistency, again, I don't know if you can almost imagine, if you feel your cheek, that is the consistency throughout, and it has this slightly yellowish tinge, and again, with gelatinous areas. There were papillary areas, but these were, again, similar, quite sort of bulbous, polypoidal intracystic projections. And this is a permanent section, sorry, fixed image to compare with the fresh. And this was much more atypical than the previous ones we saw, and I think in this, clear cell carcinoma is definitely a differential on this particular section. As you can see, the corresponding permanent section, beautiful clear cell carcinoma, which really was not obvious or quite difficult on the frozen section, and that's why the diagnosis of clear cell carcinoma is really difficult. But there was no cytology in this case, but still the diagnosis of clear cell carcinoma was made because there were other areas that had this classical papillary or tubular papillary architecture. And a bit more here, and frozen section diagnosis of clear cell carcinoma confirmed on permanent section. In this instance, Napsin against strong granular cytoplasmic positivity along with patient F1-beta, B53 was wild type, and I think this case probably warranted WT1 and B53 because of the degree of nuclear atypia, and there were some cells, but the cells were a bit more eosinophilic than clear. So, I have thus attempted to demonstrate the range of gross appearances of clear cell carcinoma to include those arising from benign adenofibroma to borderline to clear cell carcinoma, and those appearing from adenofibroma can appear grossly benign. The polyploid or the papillary areas are quite different to the serous papillary carcinoma, and I've put here some images for comparison. Again, cauliflower-like broad bulbous, whereas this is an example of a papillary serous carcinoma, and the surface or the tops of the papillae are a bit more granular, velvety, a bit more like broccoli rather than a cauliflower. So, frozen section interpretation of clear cell carcinoma can be difficult, as I've mentioned, as many of the typical features that are easily recognized on permanent sections are artifacts of fixation and are therefore not obvious on frozen section histology, like the clear cytoplasm. Now, stromal highline cores that are so characteristic on permanent sections can appear pale and unremarkable on frozen sections. However, if you combine the clinical features of marginally raised C125 with large postulated ovarian mass, most of them associated with gross examination and were available with frozen section histology and the classical cytological appearance, it becomes a slam-dunk diagnosis. And I think that one of the reasons it causes problems is the low-grade nuclear atypia, because we always associate clear cell carcinoma with being malignant, highly malignant, chemo-resistant. And I think that might actually contribute to the chemo-resistance, the low-grade looking nuclear atypia and almost benign appearances in some cases. And I think we have sent about 80 to 90 cases of high-grade, high-stage ovarian cancers for HRT tests since March of this year, and the lowest genomic instability scores have been from clear cell carcinoma. The lowest we've had so far is about two, and that was a clear cell carcinoma. And the genomic instability score is usually around 10 to 12. And that might perhaps contribute to the chemo-resistance, perhaps, I don't know. But what I find most fascinating is the consistent cytological features, irrespective of the gross and histologic subtype. Because the other reason it's difficult is they're rare, and the range of gross appearances of the cytological features are the same, irrespective of the gross appearance or the histologic subtype. I'm shifting gears slightly onto another tumor type that causes frequent problems in intraoperative diagnosis. Perimenopausal lady with the large, intact, smooth-walled cyst, what I call a happy-looking cyst, but no adhesions or disruptions to the capsule, like something that has grown slowly over a period of time. Now, current recommendation is usually to take two sections per centimeter, but obviously, this is not practical in an intraoperative setting. Therefore, to minimize the chances of targeting the most significant area, I would normally palpate the whole ovary and feel for any firm or hard areas, and specifically target those areas to make the first slice. On opening, the cyst is multiloculated, thin-walled with this watery gelatinous or mucoid material. And note how the cyst just flops and collapses as the mucinous contents are released. I then run my hand over the cut surface. I almost close my eyes when I do this because I think your senses get sharpened when you do that. And then I feel for any firm or solid areas, which I will then select for sampling for frozen section. You can get firm areas even in benign conditions, and these could be either due to small daughter cysts containing inspected mucin, or areas where the mucinous glands have ruptured, resulting in extravasation of mucin, and therefore reaction to the mucin, released mucin. But in this, the cut surface was fairly uniform, shiny, smooth cyst lining throughout, but no solid or firm areas. And I will then select the most complex area for sampling. For frozen section, of the most complex area, shows glands and dilated cysts lined by single to minimally stratified epithelium, with a uniform basically located nuclei. The scraped cytology showed, again, quite uniform monolayered and slightly crowded sheets within some growth, but again, smooth nuclear outline, no significant nuclear atypia. And based on the growth, frozen section histology, and intraoperative cytology, an intraoperative diagnosis of mucinous neoplasm borderline at most was made. And this is the fixed specimen. This was the area of most complexity, which was the area I sampled for frozen section. And this is the permanent section, which reflected the frozen section histology. Extensive sampling did not reveal any further worse areas than that were sampled for frozen section histology. In fact, the area sampled for frozen section was the most complex area, even after extensive sampling. This shows again, slightly stratified nuclei, but low grade nuclear atypia and scattered mitosis both within the lumen and within the crypts. Sorry, a final diagnosis of mucinous borderline tumor was made. The patient underwent a mentectomy and removal of the appendix and contralateral adnexae, but no further staging surgery was performed. The next case, on the other hand, is what I would call an unhappy cyst. This capsule is largely intact, with an area of yellowish discoloration, most likely an infarct. And close to this area of infarct, there appears to be a small area of disruption through which mucoid material was oozing out. And again, this has provided me with clues to make a concerted effort to look for any firm areas on slicing. On slicing, you can see this is a multiloculated, again, mostly thin-walled cyst with thin mucoid fluid, but with a bit more variegation than the previous cyst, which could also be due to the infarction and necrosis. But as I run my hand through this, I can feel a small, firm area, which measured about 2.5 centimetres, and this area was preferentially sampled for frozen section histology and intraoperative cytology. There you go, that's a frozen section histology, and you can see that even on low par, you can appreciate the architectural complexity with marked epithelial proliferation. Again, much better seen on higher magnification, where you can also appreciate the crowded back-to-back confluent glands with strong exclusion. The epithelium lining the glands are markedly atypical with goblet cells and luminal necrosis. Scraped cytology from the thin-walled area showed classical benign monolayered honeycomb sheets, but the cytology from the solid area shows necrosis and classical nuclear features of malignancy, discohesive, markedly atypical cells. An intraoperative diagnosis with at least mucinous borderline with a small area of invasive adenocarcinoma, unable to comment if expansile or infiltrative-type invasion was made, and the patient underwent a full staging procedure. And this is the permanent section of the thin-walled area showing a classical benign-looking cyst. However, the solid areas shows, again, crowded atypical back-to-back glands indicating an adenocarcinoma. Luminal necrosis, marked nuclear atypia. Again, necrosis, marked nuclear atypia. So, this was final diagnosis was of a mucinous carcinoma of the ovary. Again, another angry-looking cyst, but this time, as I feel my hand, you can almost imagine, oh sorry, this was a hard, firm area. And when you see an angry-looking cyst like this, palpate carefully to look for any hard and firm areas. And when I opened, this is a straightforward area of adenocarcinoma, which was sampled for frozen section. This was not sent for frozen section, but I wanted this case to demonstrate what a neural nodule looks like because for the longest time before I saw this case, I thought a nodule was a well-defined nodule and that I would pick it up and it would be no problem. But actually, the ones I've come across, they are very subtle. Note this granular, tan, velvety-looking lining to the cyst wall. And a section on low part of that cyst wall shows this neural nodule, which you can, which interestingly has a papillary architecture, but these multinucleotide giant cells and other areas showing markedly atypical, pleomorphic cells, and interestingly has a papillary architecture, but these multinucleotide giant cells and other areas showing markedly atypical, pleomorphic cells, and mixed with spindle giant cells and inflammatory cells, and a diagnosis of, and there were other areas of mucinous, expansile-type mucinous carcinoma in this lady, but a diagnosis of mucinous carcinoma with a neural nodule containing sarcoma-like and anaplastic carcinoma was made. And although that component was quite small, she recurred earlier this year and presented with severe pain along her lower limb, and she was found to have a large, massive recurrence, and this was the biopsy of that recurrence, and unfortunately, she died shortly after diagnosis. This is another case of a neural nodule from a carcin, mucinous carcinoma in a 22-year-old, again, very subtle. This whole thing was necrosis and carcinoma, but the neural nodule was like there, and this is the low-power view of that. Unfortunately, she recurred, the tumour recurred one year after diagnosis, and this is the cut surface of this anterior abdominal wall recurrence, and on low part, it's a solid tumour with a lot of hemorrhage, and it comprises predominantly multinucleic, almost like osteoclast- like giant cells, or maybe epulis-type giant cells. I mean, these are meant to be, well, according to literature anyway, they're meant to be reactive, but I don't think they are. It wouldn't have recurred within one year, and the whole tumour was this, it looked like a giant cell tumour of soft tissue, and in fact, I sent it to the soft tissue pathologist just to confirm that it wasn't that. They're positive for CD68 and negative for cytokeratin, and what's interesting is B53 was strongly positive in a large majority of cells, and, you know, perhaps this is a B53-driven divergent de-differentiation, but again, if you see this in a frozen section, or if you see granular areas in an otherwise borderline tumour, I think it's worth sampling that area, and if you have the chance to take a cytology, by all means, do that, because from the two cases, okay, I don't have wide experience in this, but these are rare, but the two cases that I've had, they recurred within two years, one within two years who died within two years, and another recurred within a year. Unfortunately, this lady has lost a follow-up, so I don't know what went on with her. Right, how much time do I have? Sorry, I'm very conscious of the time, because I've got a lot more to tell you, but anyway, this is, again, on the mucinous theme, but slightly different type of tumour, 50-year-old lady, but unilateral, 30-centimetre complex mass, with a slightly raised C125 and a markedly raised alpha-fetoprotein. Now, this was not known at the time with the frozen section histology, and all I can say is I'm happy, you know, I'm happy it wasn't me on the hot seat that day. This specimen was large, predominantly cystic mass, which was disrupted on receipt, and as you can see, the cut surface is cystic, multi-cystic, but it has more variegation than the previous mucinous tumours, with areas of haemorrhage and possibly necrosis. Again, small daughter cyst containing mucin, and sections of these areas show microcystic appearance, which were lined by bland, benign-looking mucinous epithelium. Another section of the stromal area shows two areas, so this section was from the cyst wall, which was slightly thickened in edematous, and that showed two areas of hypercellularity, showing atypical stromal-type cells, and another area showing pink eosinophilic cells, pink eosinophilic cytoplasm, and having glandular lumina. And my very astute colleague who did the frozen section said, benign-looking mucinous glands in cysts with areas of stromal hypercellularity showing atypia treat as ovarian tumour of borderline malignancy, but I'm not able to subtype. I think it was a good call. And this is the corresponding permanent sections of the sections taken for frozen section. Cytology was not done. Not all of us do cytology, so some of my colleagues don't do that, but I do it on all my cases. As you can see, the cystic area is confirmed to be benign mucinous cyst, and the stromal areas are the atypical stromal cells. So I went back and further sectioned, took additional sections of the tumour. And as you can see, the thickened walls contain these translucent edematous-looking wall containing these opaque, confluent white areas. And this is the mega slide of this corresponding gross picture. And these are blue, dark in both the areas. And these nodules represent areas of poor differentiation, atypical stroma, characterised by abundant immature primitive stroma growing in a story form pattern with brisk mitoses. And in some areas, these were associated with abundant eosinophilic polygonal cells with abundant eosinophilic cytoplasm. To cut to the chase, this is a malignant sertolylatic cell tumour with areas showing well, moderate, and poor differentiation. These polygonal cells are often present in sertolylatic cell tumours that secrete alpha feta protein. And although these were interpreted as lytic cells by some authors, most now view them as foci of heterologous hepatic differentiation, secreting the AFB. Adjacent to these were areas of intermediate differentiation, characterised by the sertolia tubules and immature stroma, and areas of well differentiation characterised by well-formed sertoli tubules. Calretinin is positive in, calretinin and inhibin are positive in both components. The sertoli cells are positive for cytokeratin and show membranous staining for CD99 and WTA1 shows apatite positivity. So final diagnosis, as I mentioned, was a sertolylatic cell tumour showing all types of differentiation and two DICER1 mutations were detected in this tumour with a variable allele frequency of 40 and 36% respectively. Sertolylatic cell tumours are rare and they can be solid, solid cystic, and very rarely cystic. And this was a rare example of a predominantly cystic, poorly differentiated sertolylatic cell tumour where one had to make a concerted effort to look for the poorly differentiated areas, and certainly these were not well represented in the area sampled for frozen section histology. Perhaps if we had made more slices, but I don't know, I've never made a diagnosis on frozen sections, so I don't know whether I'll make the same, whether I'll be able to do it next time round. The most common type is a solid type as seen in this tumour. Note the haemorrhage and necrosis, and this was from a 22-year-old lady a while ago who sadly died within two years of diagnosis with multiple intraperitoneal recurrences. Now I'll start the next slides by just showing this irregular crowded complex confluent glands. What do you think it is? Could do for a low-grade endometrioid carcinoma, low-grade endometrioid-bearing carcinoma with these fused complex back-to-back glands. But what if I told you this tumor is a small pale yellow nodule within a slightly enlarged ovary, and this is the full megamount of the tumor contained within the ovary? And what if I told you this was positive for inhibin and calretinin? You're right, this is a well-differentiated cytolated cell tumor. Behavior is benign with almost 100% survival rate, and as expected, this was negative for DICER1 and FOXL2 mutation. Again, gross examination is key. You have to see the gross or else you'll come unstuck. I'm going to show you a series of cases of a newly described entity in the WHO 2020, which I'll tell you up front are mesonephric-like adenocarcinomas. We've had a few of them over the past couple of years. Probably we had more prior, but we didn't recognize them as such. This was a 55-year-old lady with the large pelvic mass invading the bladder, normal CA125, and the clinical impression was that of a sarcoma. Grossly, sorry, this was solid gray-white, slightly granular texture to the gut surface, bit of papillary area. There were areas of hemorrhage and cystic areas with plaque-like thickening to the cyst lining with whitish papillary projections into these hemorrhagic cysts. Frozen section revealed a poorly differentiated non-mucinous adenocarcinoma composed of solid glandular and papillary areas, and it was called a poorly differentiated high grade non-mucinous adenocarcinoma, most likely gynecological origin. I'm just going to run through a range of morphological appearances, and this is the tumor type that I call the great architectural mimicker. It looks like anything on the planet architecturally, but the nuclear cytomorphological details are fairly uniform and consistent. Should I just finish this, Sakina? Yeah. Yeah, that'll be great. Yeah, yeah. And so, there are areas like here, long tubular glands, almost like a tram track appearance, relatively uniform nuclear atypia, marked nuclear variation in size, or a dramatic nuclear atypia is absent even in the most aggressive tumors. The chromatin, these are solid areas. They're usually hypochromatic, uniformly distributed chromatin, but they can have a variable distribution of chromatin. The cytoplasm can be scanty, but it can also be abundant, and note this rhaptoid-looking area here. So, this is an area with abundant eosinophilic cytoplasm. In some places, it can be quite papillary, almost looking like a Schiller-Duvall bodies. Note the architectural variation for the uniform cytomorphology of the nuclei. Again, it almost looks like tubulocystic architecture of the clear cell carcinoma with hobnailing, another eosinophilic area, very hypochromatic brisk mitosis. Literature seems to indicate eosinophilic secretions, luminal secretions, as the most consistent feature, but that is not my experience in the five or six cases that I've seen. In fact, you can have any material, either granular, and in this instance, this is mucinous. Again, a papillary area with almost Schiller-Duvall looking areas, another area that looks like a low-grade serous papillary tumor, tubular architecture, back-to-back glands, and I've just called this angiomatoid. I've not seen this in literature, but you have these blood-filled spaces within which you have these tumor nodules protruding in, and these are characteristically positive for GATA3 and TTF1 and PAX8. Again, literature says calretinin CD10 positive, but that is not my experience. The origin of the tumor, I believe, is mullerian, and this is because of the association with endometriosis in the few cases that I've come across. This is a good example where you can see flat negativity for ER with background endometriosis showing ER positivity within the endometrial glands and stroma, and CD10 negative within the tumor but positive within the periglandular stroma. BFRA3 was wild-type reaction pattern. Final diagnosis was mesonephric adenopasnoma, likely primary origin, ovarian origin, and tumor has invaded the uterus, mesorectin, pelvic, and periortic nodes, and it showed a canonical KRAS mutation at codon 12, and unfortunately, she's recently found to have multiple brain metastasis. Shall I end my show? I just want to come to the end of the… I'm just coming to the… I think we need a part two for this. I've just got too much to show you. Oh, my goodness. Right. How can I come to the end of… Oh, sorry. Here. So, when I said back to basics, what I effectively meant was that due to the pressures, due to the pressures, the time pressures for pathologists, I think everywhere, grossing, cutting up, and even handling of frozen section specimens are being increasingly performed by biomedical scientists or pathology assistants or more junior colleagues, and pathologists are moving further away from the grossing room, but I hope I've managed to illustrate that for intraoperative diagnosis, especially in difficult cases, the diagnostic dilemma is often resolved by extensive communication with the surgeons and pathologists being quite awful with the range of gross appearances of the different tumor types, and perhaps, I know we're all involved with molecular and everything else, but I think we shouldn't move too far away from the grossing room, and that will help you to maintain your skills with the gross diagnosis and also enable you to select the most appropriate area for frozen section or intraoperative cytology, which is simple, accurate, fast, and inexpensive, and I think, in my opinion, and in evidence-available literature, I think it can be used instead of frozen section histology where frozen section histology is not available or is not possible like it did for us during the pandemic, and where frozen section histology is available, it can help improve diagnostic accuracy, and I've got a whole load on the advantages of intraoperative cytology, but I think I'm just going to have to skip over that. Now, I would just like to acknowledge all my gynae oncologists who send me those beautiful specimens, my pathologist colleagues, and our wonderful cellular pathology staff. Thank you for your attention. I'm happy to take any questions. Thank you, Dr. Relty, for that highly informative presentation. It's a lot to get through, but I think you managed to cover a great deal. So, we're now open to questions. As a reminder to the audience, if you have a question for Dr. Relty, you can submit it by the Q&A feature found at the bottom of the screen. So, we have a few questions related to mucinous neoplasms. The first one is regarding pseudomyxoma peritoni with an ovarian mass. Does it necessarily mean that it has to be malignant? Can a benign mucinous cystadenoma present with pseudomyxoma, and in such a case, will the mucinous material throughout the abdomen be described as mucinous tumor deposits? Excellent question. I've got a whole thing on mucinous. In fact, I've got a slide of one, but unfortunately, we've run out of time. But several things. Yes, appendical, append, in fact, there was a case, I think it was about two weeks ago, it was sent as an ovarian mass, but actually, it was an appendicular diverticulum that was stuck to the ovary. And, sorry, can I just see the rest of the, sorry, has the question disappeared? There were quite a few, so I was just trying to address that. Yeah, it's in the answered questions. So is it, can I, can I see the question again? Is it, can I, oh, is it in the answer, right? Okay. Okay. The pseudomyxoma involving the ovary or pseudomyxoma, the mucin, you don't need to see cells. Can I just show you the picture, if that's okay? I don't know whether that'll help describe what I'm trying to explain better. The, firstly, the mucin consistency is very different. It's thick, viscid mucin. When we get these ovaries, they're still warm, but you almost can slice through it. They're thick, viscid mucin, and you do not need to see any cells to call it pseudomyxoma ovarii. And benign mucinous cyst can present with pseudomyxoma, but they are incredibly rare. I've seen one, but that was arising, interestingly, from a mature cystic teratoma with mucinous, you can, you saw, you had a proper mature cystic teratoma, and you had perhaps a divertical, no, perhaps an appendiceal type tissue, somatic tissue, and that resulted in pseudomyxoma. And interestingly, she also had a lamin, so she had two pathologies, but I think either ways, she was sent to our primary peritoneal cancer unit, and they did HIPEC, and they did the whole pseudomyxoma type surgery and chemotherapy. Yes, you can get it, but they're rare. And will these be, in such case, will the mucin mature distributed described as mucinous tumor deposits? I suppose it will be, wouldn't it? It depends if there are cells in it, it depends on the origin of the mucinous tumor deposit, if it is benign, and you just, benign within the ovary, and there are no cells, I suppose you can, you can describe it as mucinous tumor deposit, but I don't know whether you can call it tumor, so I'm not entirely sure of that answer, but you don't need to see cells to call it malignant. I'm so desperate to show you this case, can I just share one case, which will explain what the lady has, can I share my screen now, or is that, can I just go ahead and answer the question? Go ahead. Yeah, so can I just show this? Okay, I'm just going to share my screen. Okay, is that my screen? Okay. So, this was from a 58-year-old lady with normal tumor markers, and she had a right, the history was of a right adnexal complex solid cystic mass, and FIGO was, clinically, was a one-year primary ovarian neoplasm, and frozen section was done to confirm and proceed with debulking surgery, and this is where I wanted to talk about the clinical, or the surgeon communication. When we got this mass, the surgeon phoned me, and she said, you know, Angela, this is really weird, I really don't know what it is, there's a small nodule in the omentum as well, and this was the specimen we received, very gelatinous, high-line, and I don't know if some of you are from India, it looks like pumpkin sweets, you know, petha, you call it petha, it looks like that, and this was during the pandemic, so we did not take frozen section histology, oh shoot, but I took frozen, I took the intraoperative cytology, and all I got was mucin, I didn't, there were no cells, I did it a few times, no cells apart from the small hyperchromatic cells. Then, long communication with the surgeon, can you have a look at the appendix, first he said, well, I'm not entirely sure, then looked again, said, yeah, actually, the appendix is quite bulbous, I said, well, that's it, this is the appendix of the primary, and that was because the mucin from pseudomyxoma is so thick that it almost solidifies sometimes, and then this was the appendix, can you see the appendix? And from, as you can see, bulbous appendix, and on cut surface, it had these multi-cystic gelatinous appearance, and this is the permanent section, so although this was a mucinous tumor, most of it was extracellular mucin with few signet ring cells suspended in it, and in the actual appendix, there were more solid signet ring goblet cell areas and more poorly differentiated areas again, and this was positive for CDX2, neuroendocrine markers, and the final diagnosis was a grade 3 goblet cell adenocarcinoma of primary appendiceal origin. So I'm going to stop sharing my screen and confirm that you don't need to see cells to call it pseudomyxoma. Did you perform in vitro chemistry? No. I think, I know literature says most of the ovarian tumors are metastasis, that's not my experience, or maybe they diagnosed them before it comes to us, so I don't see many of them, and the ones I see are mostly primary appendiceal origin. We've had one case from the pancreas, and it looked benign, and I don't perform, unless it's really difficult, we've had a few mucinous tumors that have been recently quite difficult, but routinely, if they look box standard, if they don't have the features of metastasis, I don't routinely perform immunochemistry on them. I think there's a, thanks for your responses to those questions, that's great, and then there's a related question about how do you intraoperatively determine if a mucinous adenocarcinoma involving the ovary is primary or metastatic? What features would you look for? Yeah, so the first two cases I showed you, you know, they were primary ovarian. The mucin was mucoid, there was no, was watery, was, whereas a metastatic, very thick, viscid mucin. There will be instances where it's really difficult, but you can just, actually I had a picture of a metastasis, which was initially thought to be primary, and the pattern of invasion, it just looks different. There's a lot more necrosis, there's a lot more dirty necrosis, even on gross examination, and yes, morphologically, obviously, it's a gastrointestinal type, there is a lot of overlap, but I think, and when you get the ovary, the metastatic, it almost feels like, you know, you get blobs of mucin on the top, even though it looks intact, the whole ovary looks slippery, and it's this thick, I think, I can't stress the importance of thick, viscid mucin, and that will help, and, you know, there will, there are instances, I think very rarely, where we always put a caveat, if it doesn't fit neatly into an ovarian, you know, in terms of the architecture of the glands, in terms of the pattern of glands, I mean, it's very hard to describe it without showing the images, but most of them, in my experience, are primary ovarian origin. What type of, oh, sorry, go ahead. So, another attendee has asked, what type of cancer can benefit from intraoperative cytology technique? All, except serous borderline tumors. Serous borderline and seromucinous, I find them very difficult, because it doesn't, because you need to assess for invasion. I think serous borderline tumors, I've got a whole thing on them that I didn't get a chance to show you, but serous borderline, all tumors, apart from seromucinous and serous borderline tumors, if I answered it the other way around. Is that okay? Thanks for that. So, another question relating to borderline tumors is that the sensitivity is lower for borderline tumors than diagnosis of benign or malignant, and as a surgeon, is there anything that could be done to make the sensitivity higher? Sorry, can you just repeat? I can't see that question. Can you just repeat that? So, the sensitivity for diagnosis of borderline tumor is lower than for benign and malignant, and is there anything that the surgeon could do to improve the sensitivity? I don't know. I think borderlines are hard, and let's face it, even on permanent sections, experienced pathologists don't agree whether a particularly exuberant borderline tumor is expansile type, if you can call it that, of a low-grade serous carcinoma, let alone on frozen section, and I think the bottom line in these instances is to guide the intraoperative management, and again, I had a few pictures, and I think if something is really florid, imagine if a borderline tumor looks like either a pomegranate or the cross-section of a cerebellum, I think you should go ahead and stage it, because we know that 25% of these really florid borderline tumors have lymph node involvement. Therefore, it is a difficult area, not just on frozen section, but also on permanent sections, and as I said earlier, even experienced pathologists, we don't all agree, and if you don't get the classical invasive, obvious invasive islands, and it's just this exuberant proliferation, it's hard to tell whether, you know, if you call 10 pathologists, 5 will call it expansile type, low-grade serous, and half will call it borderline. I think the main message here is it's worth doing staging procedure in those cases, because the number of times we've come across lymph node involvement, and if they're, you know, I call it involvement rather than metastasis, and you know, they will be subsequent risk factors for development of serous carcinomas later on. Is that okay? Well, thanks for that, Dr. Raut. I think that that's all that we have time for today, and I'll just mention that there's been many complimentary comments as well about your presentation and some requests for a part two, so I think that's something that we might need to look into. So, Dr. Raut kindly also offered that if you, you know, we were not able to get to your question, that you could submit it by email to her, and she will try to respond, and just to mention also that the recording will be available in the IGCS education portal for IGCS members, and also as a reminder to IGCS members, they're seeking pathology advice and recommendations. There is the Ask a Pathologist resource, and you can submit an email to pathology at IGCS.org. So, I thank you all for attending, and to Dr. Raut for this great presentation, and so I wish you all continued health and safety. Take care and have a great day. Thank you very much for having me. Thank you.

Dr. Angela Rolte

pathologist

ovarian tumors

intraoperative diagnosis

gross examination

communication

pathologists

surgeons

diagnosing tumors

frozen section histology