It is a great pleasure for me to welcome you all, and thanks a lot for joining us to this webinar organized by the Mesothelioma Subcommittee of the Rare Cancer Group from the EASLC. And I'm very pleased to share this lecture with Dr. Luca Brigg from Graz, University of Graz, with Dr. Andrew Churg from British Columbia University in Vancouver. And we give a lecture on updated pathology consideration and terminology in mesothelioma. So we have prepared an agenda, and we have decided to divide in three parts. Part one, I will do the updated pathology consideration and terminology, which is done by myself. The second part is done by Dr. Luca Brigg on immunohistochemical staining for separating mesothelioma from other tumor, and a challenging question. And the third part is performed by Dr. Andrew Churg, the separation of benign from malignant mesothelioma proliferation, which is also a very challenging task. So I will start. So you know that mesothelioma is an aggressive, rare cancer, principally in the elderly patient with poor survival, but it should be a preventable pathology because it is mostly due to asbestos exposure. The optimal therapy, we know, is under dependence of age, performance status, and comorbidity, but mostly on staging and histology. The impact of histology subtyping is very important because it drives the prognosis and allows to determine specific therapeutic strategy. So that we have done during the last WHO 2021 new classification, and we also have introduced new concept and terminology. So we divide the classification in three parts, benign and pre-invasive mesothelial proliferation, including adenomated and well-differentiated papillary mesothelial tumor, and we have decided to not use anymore the terminology of well-differentiated papillary mesothelioma to avoid confusion with diffuse mesothelioma because it has not the same prognosis. So we include the terminology of mesothelioma in situ for the first time, and after we decided to classify mesothelioma and divide in localized mesothelioma because of a better survival and diffuse mesothelioma, and the terminology has changed, and we avoid now to use the term malignant because mesothelioma is definitely malignant. We have divided, we have maintained the histological classification to three classes as follow, epithelial, biphasic, and sarcomatoid. Just to work on pathological sampling because it is very important. Most patients probably feel in that presentation, but the question is how reliable is sitology specimen? It is important, it is actually accepted to identify earlier a mesothelioma when BAP1 is lost with a control positive on the slide, or if there is a pneumozygous deletion by silica and, I say by fish, I say, which could be replaced by the cytoplasmic loss using EMTA. It could also be used as a sequential cytology sample that could facilitate pharmacokinetic studies and response to biomarker in intrapleural betrayal, but it is not for routine, and however, cytology alone remains inadequate because it could not render the disease of tissue mesothelioma, it could not strictly evaluate histological subtypic and gradig, and could not evaluate biphasic and sarcomatoid components. Tissue biopsy sample is also an important point, and it is recommended to perform at least three distance sites, if possible, multiple to recapitulate tumor heterogeneity, and to consider a rare penterase viator acouscopy imaging for strict subtyping and grad. And finally, the number of biopsy facilitated extensive immunohistochemistry and molecular analysis. So, localizing mesothelioma, I would like to say that it is a rare tumor of mesothelial origin, which is mostly a circumscribed mass without clinical or histological evidence of deep serosal trait. The incidence is very low, 0.52 and 0.56. It has a median survival. Clinically, it's more male than female. The muscular and intrathoracic mass, the subtypes are identical as the diffuse mesothelioma, epithelial, sarcomated, and biphasic, the same immunophenotype, the same molecular abnormality, except that ChEORiAC showed that the TRAF7 mutation and genomic neoploidization was observed in localized mesothelioma. So, we are discussing on morphological subtyping and heterogeneity. Why? Because in 2019, we performed, we used artificial intelligence and used deep learning to pass classification of mesothelioma, and we observed that it improved prediction of patient outcome. We selected 2,300 trading set of wood, wool slide images, data set of 681 cases, and a control test of 56 cases from, as a control from TCGA collection. And what we see, we see that the model, and we built a mesonet algorithm. We see that the model was able to stratify patients even within established histological classification, but inside the epithelial curve, as you can see, we were also, the model was also to divide in three different curves of better, intermediate, and poor prognosis. So, the high-risk patients found by the model were not only sarcomated, but also biphasic and epithelioid, and the types of interest showed that there was atypia, pleomorphism, or very atypical somal response with atypical fibroblast. And the low-risk patients found by the model were only all epithelioid. So, that's what we have worked on. So, the WHO 2021 classification, we stratified with epithelioid, biphasic, and sarcomatoid, and also we described cytological feature and pattern, and we divide them in better prognosis, unfavorable prognosis, and unfavorable prognosis. As you can see, tubulo-papillary, trabecula, or adenomatoid, are actually typical pattern of poor prognosis. The same for cytological features of lafociocytoid or low nuclear growth, and for somal feature, when there is a mixed somal component, which is more than 50%, and with a solid component, which is less than 50%, it is associated with a favorable prognosis. And we do the same for the sarcomatoid, and we divide and see that some unfavorable prognosis in epithelioid, solid, and micro-papillary pattern are associated with poor prognosis because the micro-papillary pattern give lympholytic spread, and the cytological feature associated with unfavorable prognosis are rhabdoid, pleomorphic, or high nuclear growth and necrosis. For the sarcomatoid, we will see that better prognosis is associated with a lafociocytoid component, while unfavorable prognosis is associated with transitional pleomorphic, and we will discuss that. We also, a grading system was applied to divide epithelioid into prognostic groups, and it was accepted by the AGCC, ECCR, and the College of Apartheid and Pathologies, and we have divided, we have used a two-tiered system for epithelioid based on nuclear pleomorphism, mitotic count, and presence of absence of necrosis. And finally, it was divided in low grade, including nuclear grade 1 and 2 without necrosis, and high grade, which was a grade 2 with necrosis, grade 3 with or without necrosis, and you can see on the right at the bottom that there is a different curve of survival, and it fits very well with this low and high grade. Sarcomatoid mesothelioma, rare, usually less than 10 percent. It is observed mostly in men, in 96 percent. In older patients, it is, there is a frequent association with pleural plaque in 70 to 90 percent, and it is diffused, morphologically it is a diffused pleural thickening or modularity of spindle cell proliferation, sometimes very highly cellular with atypia, may look like sarcoma, or blonde looking like organizing pleuritis, and the important criteria is to show the invasion of spindle cell between adipocytes or other tissue underneath, and the desirable criteria to make a diagnosis of sarcomatoid mesothelioma is the cytokeratin positivity, less than 10 percent of sarcomatoid mesothelioma do not show positivity of cytokeratin, but you have to use different cytokeratin, or the criteria are cytoplasmic MTAP loss, or by the de-homozygous deletion by fish of the decanthoid. BAP1 loss are less frequent in sarcomatoid mesothelioma, and there is also NF2 mutation and and mutation from NGS or RNA-seq, but we will discuss that later. The main differential diagnosis, the challenging situation that faced the pathologist when he had to give an answer to the clinician, is that to differentiate sarcomatoid mesothelioma from sarcomatoid carcinoma. Sarcomatoid mesothelioma, rare, it is a diffused pleural thickening, and marked pleural thickening, and encasement of the lung parachroma, parachroma. While for sarcomatoid carcinoma, it is rare, 0.3 to 3 percent of all non-small cell lung cancer, it is usually in a small curve, but there may be asbestos exposed, and it is usually a solitary peripheral mass with a predilection for the upper lobes that invite the pleura. For us, you know, under the microscope, even on small biopsies, it is very self-challenging because the histological, cytological features are strictly the same. And if you have not, you know, the diffused pleural thickening with a fistoned area in casigula lung, compared to the solitary mass in sarcomatoid carcinoma, you have to use, after the appropriate clinical radiographic setting, a panel of immunochemical markers showing the positivity of the cytokeratin, but also using two mesothelioma markers, but cadretinin is positive only in 30 percent of the case, the same for CtK5-6, and it is less for WT1 and D2-Ca. But we showed with Marczewski that GATA3 with a cutoff of more than 30 percent positivity may help and favor sarcomatoid mesothelioma compared to GATA3 less than 30 percent or normally negative in sarcomatoid carcinoma. You can also use the desirable criteria, which, if necessary, which is a reduced cancer. And also, by NHZ, you can look for specific mutations in favor of sarcomatoid carcinoma, which is a KRAS mutation, observed in more than 30 percent of the cases, 90 percent on exome 12, mostly PZV12C or V12V. And it is true that mesothelioma can show a KRAS mutation, but it is normally G13D. And also, the skipping met on exome 14, observing 33 percent of pleomorphic carcinoma, which has never been observed in mesothelioma. So, at the end, sometimes it's so difficult that if you have the opportunity to use ARD sequencing, you can see that we can, you know, observe final diagnosis on uncertain diagnosis of sarcomatoid mesothelioma in 63 percent of the cases, while carcinoma in 7 percent, sarcoma in 13 percent, and 17 percent remain uncertain. So, that's a very difficult task, and you need an appropriate clinical radiographic setup. Biphasic mesothelioma, the definition is at least 10 percent of each component is epithelioid and sarcomatoid is required for the definitive diagnosis in resection specimen, which means EPP or EP. Any tumor can be diagnosed as biphasic on small biopsy sample, regardless of the percentage of epithelioid or sarcomatoid components, and the percentage of sarcomatoid components is evaluated in resection specimen for prognosis. So, it is very important, the value of biopsy sampling in predictive histology and biphasic histology, and that has been proven by Chirac, who showed that a concordance of diagnosis between biopsy versus resection specimen is increased with the number of biopsy, 3 versus 2, with a p value of 0.012. The diagnosis of epithelioid on the initial biopsy changed to biphasic on the resection specimen in 20 percent of the cases. The other challenging question is BAP1, nuclear loss, and silica in two homozygous deletion can be helpful to separate two sarcomatoid components from stromal reactive response, as well as the expression of cytokeratin, as you can see here, in benign reactive stroma, while it is stronger in mesothelioma, and you can also use reticulin-10 cytokeratin could be helpful to detect the percentage of the sarcomatoid component. I want to highlight two new cytological features that were very misleading. Transitional mesothelioma are characterized by elongated plumb cells that appear intermediate between epithelioid and sarcomatoid, arranged in sheet-like patterns. The main pitfall is because it may look like epithelioid, and if you do reticulin, you can see that it highlights nicely single cells in transitional and in sarcomatoid, while in the epithelioid, it is delimiting a cluster of epithelioid cells. And why it is important to make a diagnosis of transitional mesothelioma? Because the transitional feature should be classified under sarcomatoid peptide, as we were able to show it by ARD sequencing, showing that the transitional cluster altogether just close to the sarcomatoid and a distance of the epithelioid. When associated with an epithelioid component, the transitional mesothelioma component should be classified as B-phasic mesothelioma. And it is very important, it is a very poor diagnosis, as you can see, with a median survival of 6.1. There was no specific molecular associated, except that the analysis of different expression genes of 260 genes showed that there were very highly aggressive cell cycle genes. And we can observe that there was a CDKN2 over the duration of 74% of the cases, and BAP1 loss only in 50% of the cases. Pleomorphic is the second new cytologic dreadful feature observed in mesothelioma. It is made of large cell, anaplastic cell with a wide range of morphology, with at least 10% of epithelioid component. And the pleomorphic component is also associated with a very poor prognosis, which is nearly six months after the initial diagnosis. And you can see that it is a large cell with what we call emperipolesis, and a strong inflammatory infiltrate. We performed a whole exome sequence to characterize pleomorphic mesothelioma with Sobacroi, and it was a series of 24 pairs. There was no specific molecular signature, no molecular-derivable signature. And by RNA-sequencing, we were able to observe that it clustered with epithelioid, trachomatoid, and B5Z. So, we could not associate pleomorphic with a specific subtype. And it was also, we also observed that somatic alteration of BAP1 was observed in 37.5%, and there was different alteration, falchi, slici, misos, or inframed. And what was remarkable is that there was no BAP1 alteration in the star chromatin component associated with pleomorphic features. And there was a co-occurrence with deletion on chromosome 4, 19, and 13, and focal loss of chromosome 19. So, molecular heterogeneity, just a word, because I think it will be very important for the future. Somatic mutation in mesothelioma is low, usually less than 2 synonymous per megabyte. In consequence, microsatellite instability is rarely reported, except when it is a mesothelioma associated with a Lange syndrome. And there was two out of 56 cases in the ORTC Spectra Archangels Study. Or in Taglia Manto in 2021, it was reported. The molecular landscape of pleural mesothelioma is characterized by multiple copy number alteration, loss and gain, single nucleotide, or structural alteration affecting tumor surface origin. On the right, you can see the major reference in the literature from 2018 to 2024, and I will talk of that. The remarkable paper made by Dagogo Jack et al. in 2022 on a theory of 980 pleural mesothelioma was very interesting because she was saying that molecular analysis revealed similarities across site of origin and histologic type. And she also observed frequent inactivation of the alteration involving all the molecular abnormality I told to use, Diken 2A to B, BAP1 and TAP and NF2. And what was very interesting is on the right, you can see the prevalence of molecular alteration in epithelioid versus non-epithelioid. And you can see that BAP1 is mostly observed. The prevalence is greater in epithelioid while silica N2A, silica N2A, NF2 and TAP and TP53 and ZD2 are mostly observed in non-epithelioid. And a paper made by Ang and Chiriac published in Archive Patholab Med in 2020 was also regarding what molecular marker can help in diagnosis and prognosis and they observed cell cycle, DNA repair, epithelial pathway, RNA machinery, histone regulation and they observed all those alteration. And they made a table which is very interesting showing the prevalence of genetic alteration in mesothelioma. And you can see the overall of BAP1 is 40 to 53%. Silica N25 to 49 and TAP 27 to 34 and NF2 21 to 33. And you can see that BAP1 is less, is more prevalent in epithelioid compared to sarcomatoid which is the opposite for silica N2A and TAP and NF2. And also all those markers can be done by immunochemistry and you can see that the prevalence of nuclear loss by BAP1, cytoplasmic loss of M-TAP and complete loss of merline. And you can see also that BAP1 is more prevalent in epithelioid compared to B5C and sarcomatoid for cytoplasmic loss of M-TAP and complete loss of merline. But my co-author, my co-speaker will maybe talk better than me on that. I just want to also highlight the importance of germline mutation in mesothelioma and it was performed by Pano et al. in 2028 and they said that there was 12% of BAP1 in patient of younger age, less plural than peritoneum, no asbestos exposure and patient with a concomitant second cancer saying that it was a cancer BAP1-associated syndrome. But in 2020, one view show when novel germline mutation in DNA damage repair showing that BAP1 is observed in 10% in epithelioid while MSH3 and REG2L4 are observed in 10 and 20% and more observed in the sarcomatoid subtype. There are also molecular abnormality such as alkyl arrangement detected in young by immunohistochemistry or by FISH or NGS, which is rare but can given to therapy, which is very important. It is in younger, but it is more in peritoneum that in plural, which is the same with ATF1 rearrangement and AWSR1YY fusion that is mostly observed in peritoneum compared to the plural, which is very rare. And I finished by two slides showing that we try at the time of where immunotherapy is so important and we try to redefine malignant plural mesothelioma type as a continuum that in-covers immune and vascular interaction. And we were able to separate the mesothelioma in three group. One showed which was mainly epithelioid mesothelioma expressing VGFR2 and VSTAR, which is very important because there is a marker that could be applied on biopsy tissue sample and there is a therapy at the end. And the two other, which is a hot group with numerous immune cell, immune competent cell, and which was VEGFR1 and mostly observing biophysic and sarcomatoid, while the last one, which is a cold desert of immune cell group with VEGFR3, and which is mostly observed in biophysic and sarcomatoid. And those two are associated with worse prognosis, which is the case for the cold case. And the last part, we try to disentangling heterogeneity of malignant plural mesothelioma through deep integrative omic analysis. And we were able to see that there was four sorts of molecular variation, one morphology factor, which was, and all other proposite factor associated with the morphology factor, while they find some with prognostic value that can be estimated in clinics and related to therapeutic candidate, which is pleuidic capture, aneuploidy can be estimated by fish and associated with axitinib. Adaptive response factor, which separate cold and hot tumor assessed by pathologist is estimated immune content and related to immunotherapy. And the one which is CRMP index factor, we found to the CPGI methylator phenotype, and which is assessed by CMP panel assay methylate and associated with mitomycin. So, at the end, during the last five years, consider our advances and improve our understanding in the diagnosis, prognosis, and precision medicine for plural mesothelioma. For clinical practice, it is essential to properly characterize the histological subtype of plural mesothelioma to implement precision approach for molecular subset correlated with the histological subtype. And finally, the molecular profile, if we have not found the drug about mutation, plural mesothelioma largely consists of inactivating tumor suppressor gene, and it could be run when immunohistochemistry is defective for diagnosis, when clinical morpho and phenotypic correlation are equivocal for the separation of benign versus malignant mesothelial proliferation for the differential diagnosis with other neoplasm and metastasis, or for providing predictive information on how mutation that just may affect prognosis and therapeutic decision. So immunotherapy has introduced mesothelioma to not a new area, but further work are needed to select the good immunoblocker in the near future. So thank you for your attention. And this is some references or guiding to assist the decision of practitioner. Thank you for joining us, and I am very pleased to introduce my next speaker, Dr. Le Cabric, which is going to talk on immunohistochemical staining. Thanks a lot. So thank you, Francois. It's a great pleasure to talk after you. And as you already mentioned, I'll be talking about immunohistochemical staining, or better said, I'll be touching upon immunohistochemical staining, because in limited amount of time, it's only thing what we can do. In the presentation after me, is we discussed very nicely how to distinguish between reactive and neoplastic lesions in plural. And in this next few minutes, I'll be going through some of the, for me, most important or more useful immunohistochemical markers, which should help us and help us every day in differentiation between mesothelioma and carcinoma, mesothelioma, sarcoma, and some other tumors. What is very important to remember is that we are talking about very rare disease here, and metastasis to plura are much, much frequent. So in one publication, I found this ratio between mesothelioma and plura and metastasis being one to 50, which is very, very huge difference. And that's why it's not always necessary to try to put or diagnose something as mesothelioma, because other diseases in plura or other metastasis to plura are much more frequent. As many of you are aware of, once we decided that something is malignant or neoplastic in plura, then we really need to confirm mesothelial origin. And still current guidelines recommend that we should always have two positive mesothelial markers and two negative markers for carcinoma or some other tumors which come in question. However, it's very, very important to know your laboratory, to know your immunostains, and also to see what type of samples you are dealing with. Because if you are dealing with many crushed artifacts or some different artifacts in the biopsies, you have to take this into account when evaluating immunohistochemistry, and also very special problem might sometimes be caused by entrapped normal or reactive mesothelial cells or also epithelial cells. In much greater details about immunohistochemistry and everything actually you need to know about diagnosing mesothelioma, you can find in this publication. So this is just for your reference. And I will just go through a few of these antibodies, which I find very useful. So the title of the slide, the most useful staining, maybe it's a little bit too much, but I find cytokeratins very, very useful when dealing with different tumors. First of all, we can also sometimes appreciate the architecture much easier. And of course, invasion, if we are thinking of where are the tumor cells, and if we are having issues on the HME staining to identify something as a tumor cells in fat tissue or in lung parenchyma, keratin stainings will be of great help. There are many keratins you can use, pozzolipan cytokeratin, cytokeratin 818, or cytokeratin IAE1 or 3. And most of epithelial mesotheliomas will stain nice and strong positive. And most of sarcoma mesotheliomas will stain positive. And sometimes it might be, unfortunately, the only marker positive in tumors. And if we rule out some other possibilities, it might be the only clue guiding us to diagnosis of sarcomatoid mesothelioma. What's sometimes misleading is that also reactive mesotheliostroma is also positive with cytokeratins, but usually intensity of staining is less. That being said, if we just have reactive mesotheliostroma, we cannot compare intensity of staining. That's why it might cause some diagnostic dilemmas. Also, not to forget that there are some other malignancies, sarcomas, melanoma, even angiosarcoma and lymphomas, which might show very focal, but might show positive reactions with cytokeratin. And it's always depending on the size of the biopsy or the sample, how to interpret it. And also, depending on the fixation time, size of the sample, and type of fixative, we can have different staining results. Carotidinin, especially if it's a stain as it's supposed to be, so nuclear and cytoplasmic, nice, strong staining, is usually clear marker of mesothelial origin, but also don't forget that in up to 10% of lung adenocarcinoma and 40% of lung squamous carcinoma, it might be positive, but usually it's focal and not so strong, both in cytoplasm and adenocarcinoma. Cytoplasm and nuclear. Another marker which is very nice is WT1, which should be interpreted as positive in this context only as a clear nuclear staining. So that's why it's sometimes very useful not to do the evaluation just on the lower power, but really to go on a higher powers to check if the staining which we are observing is actually nuclear staining. And don't forget also that in a small amount of squamous or carcinomas, it might be positive, but in lung adenocarcinoma, it's usually negative. Cytokatin 5-6 is also very useful, especially some colleagues find it very useful in setting of spindle cell or sarcomatoid lesions because it might stain just a few cells, for example, in sarcomatoid mesotheliomas and then in appropriate context and the reactions of other antibodies, it really can help us in coming to a diagnosis. It's usually stain, staining is membranous acetylplasmic, and it can stain according to different resources up to 20% of lung adenocarcinoma. And as you are all aware, it will stain a great majority of squamous cell carcinomas. D240, also don't forget, this sometimes might be pitfall that we will forget that squamous cell carcinomas in a majority of cases will also stain positive with D240 and a small percentage of lung adenocarcinomas. And that's why all these percentages are actually the reason why we should have at least two positive mesothelioma markers and also two negative markers to be certain that we are dealing with cells of mesothelial origins. This one is rather, we can call it a new kid on the block. H1, it's a membranous staining, and what is very good is that lung adenocarcinomas and squamous cell carcinomas are negative. Some, it has been described, can show a positive cytoplasmic staining. And important is not to forget that thyroid carcinomas will show nice positivity and also half of the cerebrospinal carcinomas. So that's why, if there is a question of metastasis, we have to be very careful with interpretation of this staining. Claudine IV is also a relatively new marker and, or at least in our house, we're using it relatively recently, just a few years for confirming if something is actually not of mesothelial origin, but of epithelial origin. If you have this nice membranous staining, very strong, then it's nice, positive reaction. And you have always to be also very careful with interpretation because this is a very common situation that we will have this in mesothelial cells. So in mesothelioma, that we can have this dot-like cytoplasmic reaction, but this dot-like, very faint reaction is actually to be interpreted as negative reaction. So Claudine IV is always negative, or at least so far has been always negative in mesothelial cells or mesotheliomas. Two other markers which are expected to be positive mostly in carcinomas or two-armed thrombocytoglifectomia origin are BREP4 and MOC31. They both will stain cytoplasmic membranous staining and don't be surprised because up to one third has been described of mesotheliomas can show but only focal positivities. Just, I have listed here a few other pitfalls. So estrogen receptors, alpha and progesterone receptors are not positive in mesothelioma. And don't forget that carotene can be positive also in breast carcinomas. But Francois has nicely shown that GATA3 can be very useful marker also for sarcomatoid mesotheliomas. If it's very strong and diffuse in more than 30% of cells with nice nuclear staining, but also epithelial mesotheliomas can show focal positivity for GATA3. And PAX8 and PAX2 are also useful negative markers because they will be negative in pleural mesotheliomas. Coming almost to a end, just two or three markers which I find very useful. If you cannot afford to have huge amount of immune stains, then STAT6 is definitely one of the best you can use for diagnosis of spindle cell lesions. In this case, solitary fibrous tumor, which is also can come potentially as differential diagnosis with sarcomatoid mesothelioma. And as you know, actually STAT6, and as you know, actually STAT6, if it's nicely stained like nuclear strong staining, it is very nice specific marker for this tumor. Another marker, which is also nice to have is beta-catenin, but also you won't have to be aware or very careful interpretation because it should really be a nuclear staining. And it's also something which we see or which is typical for desmoid fibromatosis without need to go to molecular analysis. And another very useful and very good antibody is SS18-SSX to prove or exclude if something, if we are dealing with synovial sarcomas with spindle cell lesions, because the synovial sarcoma can also show positive reactions with calretinine. And in rather old publications, almost 12 years ago, they summoned nicely where in which tumors you can, other than lesothelioma, you can find positive reaction with calretinine in comparison to keratinins and WT1. And it was very interesting that also in desmode fibromatosis, you know, up to 75% they have found positive reactions with calretinine. And also in this histiocytic, or tumors of histiocytic origins, calretinine is also very commonly positive in combination, especially with keratinins, which might then cause some misdiagnosis. Very useful paper, which I like very much and use very commonly in interpretations of spindle cell lesions, which are not easy then to classify, is also by Varshensky, published in 2017. And in this table, they really put nicely together how to interpret immunohistochemical reactions with cytokeratinins, markers of mesothelial origins, and epithelial markers when dealing with spindle cell lesions. And it is very useful, and especially like if you see it here, if it's positive in cytokeratinins and negative in mesothelial and epithelial markers, then suggests that the diagnosis is probable sarcoma or mesothelioma, and one should exclude other lesions. So in real life, this is really very useful table, because not so uncommon we will find the situations with different expressions of cytokeratin or mesothelial markers or epithelial markers. This is something I really can recommend for coming to the, or closer to the right diagnosis. And with this, I will finish my presentation, and just to repeat what is the most important. We still have to rely primarily on morphology. So we have to start from morphology, and especially if we are limited in immunohistochemistry, and actually it's not good to overdo immunohistochemistry, because that will cause more problems. It's really nice to have differential diagnosis in mind before starting using different antibodies. When interpreting different states, we should think also about pre-analytical and analytical factors. And like I said at the beginning, mesotheliomas are very rare tumors, so you don't need to try to diagnose everything in plural as mesothelioma. With this, I'm really thankful for your attention, and I'm really happy and proud that I can now introduce or invite Dr. Andrew Chirk to give his presentation on differentiation between reactive and neoplastic lesions. So I'm going to talk about separating benign from malignant mesothelial proliferations. The issue that I want to address is that most of the time, mesotheliomas are readily diagnosed by microscopy. You're sure that you're dealing with a malignancy, and the question is, is it a mesothelioma or something else? But sometimes the microscopic appearances are equivocal. You're not sure, are you dealing with a malignancy of some sort, or are you dealing with a benign reaction? And benign mesothelial reactions are common. Don't forget that in theory, if you biopsied every pleural effusion, you would find some sort of mesothelial reaction, the vast majority of which are benign. It's very useful to know what the imaging shows, because nodular pleural thickening is always strongly suggestive of malignancy, not specific to mesothelioma, but suggestive of malignancy. Pleural thickening involving the mediastinal pleura is almost always malignant with a major differential of tuberculosis. If you're a clinician listening to this, tell your pathologists about this, because pathologists don't know what these findings are as a general rule, but you do. If you have no pleural thickening, and the operator has done a thoracoscopy and says the pleura looks normal, then one has to be careful. Now, how do you separate benign from malignant mesothelial proliferations with a microscope? As Luca just said, routine morphology is still the basis of what we do. In other words, H&E stained slides. But it is possible to do immunochemistry for specific markers that help separating benign from malignant, and it's also possible to do fish. In theory, you can also sequence tumors or proliferations, but that's not generally done. This is a drawing I made, and what it shows is a representative thickened pleura, where all of this at the top is a fibrotic pleura, the circles at the bottom are fat. So, we're looking at a notional biopsy of the pleura, and the distribution of mesothelial cells is very important. If the mesothelial cells are in the middle of that thickened pleura, parallel to the pleural surface, that is usually benign. That's probably the original surface of the mesothelium. On the surface of a thickened pleura, the process can be benign or malignant. It usually is benign, but not always. But if you've got a proliferation of mesothelial cells that goes from almost the pleural surface down to the interface with the fat, or enters the fat, that's a malignant process with mesothelial cells. And the general rule is invasion of tissues, which usually means fat, by mesothelial cells is always a marker of malignancy. And here's an example. So, this is a very thick pleura. The pleural surface is at the top. The chest wall would be at the bottom. We have proliferating mesothelial cells from left to right and from top to bottom. This is simply too much for a benign reaction, and what we can see, in fact, is that there's fat at the bottom, and the fat cells are surrounded by mesothelial cells. So, there's fat invasion. That means that it is a malignant mesothelioma. But here's the sort of problem one runs into with benign versus malignant. Here's a pleural biopsy. And all the cells in this biopsy, I can tell you, are mesothelial. And right in the middle, at the red arrows, we have what looks like they might be invasive mesothelial cells. But notice, on this surface and this surface, we also have mesothelial cells, which means this is an on-fast cut. And in that circumstance, we don't know if what's in the middle is really invasive or it's just a surface cut on fast. And this is where a variety of adjunctive tests are helpful, as I'm going to come to in a minute. Now, Francoise has mentioned desmoplastic mesotheliomas, and these are very difficult to deal with because they are palsy cellular tumors that look like some sort of fibrotic reaction. I've listed some of the criteria that one can use to help decide that something is a desmoplastic mesothelioma of which stromal invasion, and that usually means fat invasion, is the most common and the most useful. The others, sometimes you see sometimes, and sometimes you don't. Here's an example. This is a very thick pleura, and it looks like it's just a fibrotic pleura, but something is going on in the fat. And what we see in the fat are spindle cells going between fat cells, and they are keratin positive. And Luca has mentioned the utility of keratin. It's very useful because it tells you where those mesothelial cells are. Mesothelial cells that are keratin positive in fat are malignant by definition. So this, in fact, is a desmoplastic mesothelioma. The differential for a desmoplastic mesothelioma is organizing pleuritis, and this can be a very difficult diagnosis. One useful rule is that the cellularity in organizing pleuritis is greatest immediately under the effusion and then gets more fibrotic and posse cellular as one gets away from the effusion. This is so-called zonation. You may also see capillaries perpendicular to the surface, and that's another helpful sign of a benign reaction. But the important problem is that the cells under the effusion, the mesothelial cells, can be extremely atypical. Keratin is, again, very useful. And here's an example. So this is a patient known to have an organizing empyema. The pleural cavity would be at the left. At the right is chest wall fat. This is very fibrotic looking, very posse cellular looking, but very thick. And notice in the box here, something is going on in the fat. So we want to look at that more closely. And what we see if we do a keratin stain is that the keratin is positive in this reaction, as keratin is always positive in reactive mesothelial cells, but there are no positive keratin cells in the fat. And that's very reassuring. The keratin stops right here. The fat is free of keratin, and that provides support for the notion that this is organizing pleuritis. So just to summarize about keratin, all active mesothelial cells, mesothelial proliferations are keratin positive. You see that in benign reactions. You see that in malignant reactions. Keratin staining per se does not make a mesothelial proliferation malignant, and that confuses a lot of pathologists. But what keratin tells you is where the mesothelial cells are located. If they're in the fat, well, then that's malignant. Now let's turn to the issue of genomic testing, because sometimes, or not all that infrequently, you're stuck with a biopsy where you're not sure if it's benign or malignant, keratin is not helping you, maybe you don't have any fat, and now you need some adjunctive test. And in this context, it's important to realize the difference between lung carcinomas, which we deal with commonly, and mesotheliomas. Lung carcinomas typically have driver mutations, such as EGFR mutations. And for many of these pathways, there are drugs that can block the particular pathway, such as EGFR inhibitors. From the point of view of morphology, in general, lung carcinoma mutations are not detectable by morphologic techniques. There are a few exceptions that I've listed, but we're talking about a small percentage of cases. Mesotheliomas, on the other hand, do not have driver mutations. Instead, they have mutations or deletions of tumor suppressor genes. So, you end up with uncontrolled cell proliferation. From the point of view of therapy, these types of changes are not, as a rule, drugable, but the interesting part is you can detect them by morphologic techniques. Francois showed you this very useful article from the Go-Go Jack, and what it does is list the types of mutations found in almost a thousand pleural mesotheliomas. And notice that the vast bulk of mutations or deletions are really only in six genes, CDKN2A, CDKN2B, BAP1, MTAP, NF2, and TP53. The rest of these, although they certainly occur in mesotheliomas, are fairly uncommon. The importance of this is we can easily test by morphologic techniques everything that is to the left of the blue line that I've put on here. So, what I want to show you is the use of testing for four genes that help separate benign from malignant. BAP1 has a whole set of complex functions related to cell proliferation, cell death, and now is believed to be a tumor suppressor gene, and we can detect that by immunochemistry. CDKN2A is a cell cycle inhibitor. It's a tumor suppressor gene. It is actually the most commonly mutated or deleted gene in human carcinomas, and we can detect that by fish. But there's a closely related gene that, as I will show you, sits very near CDKN2A at 9p21, and that's methylviodenosine phosphorylase, which is frequently co-deleted with CDKN2A, and we can detect that by immunochemistry. And NF2, which is a late change in mesotheliomas, can also be detected by immunochemistry. Here are some examples. So, here is a BAP1 stain of a mesothelioma. We're looking for nuclear positivity because that's normal, and in this tumor, there is nuclear positivity. But in this tumor, the tumor here at the left is completely negative. It's lost BAP1. The arrows point out internal positive control cells, which is very important when you're interpreting these stains. So, this is a mesothelioma, and BAP1 has been lost. I should point out that BAP1 loss is sensitive in epithelioid mesotheliomas, but not in sarcomatoid, and here's a biphasic mesothelioma. On the right, we have the epithelioid component that has lost BAP1. On the left, the sarcomatoid component that has not lost BAP1. So, you can certainly stain sarcomatoid mesotheliomas for BAP1, but the rate of loss is not very high. My estimate is it's less than 20%. This is fish for CDKN2A, and when you do CDKN2A fish, we use a dual-color technique in which green is the centromere 9, and red is CDKN2A, and what the green does is it's your internal positive control. So, on the left is a benign reaction in which we've got red and green. On the right is a mesothelioma, which in this case has lost virtually all of its red signals, so it's close to 100% loss. That's a mesothelioma. Now, the problem with fish is you need specialized equipment and people who know how to interpret what they see, and that's expensive and time-consuming, and lots of labs don't have fish available. The workaround for fish is MTAP, methylthiodenosine phosphorylase, and as you can see from this schematic, MTAP sits within 100 kilobases of CDKN2A at 9p21. In fact, the commonly used fish probe, which is the VISIS probe, actually encompasses both of of these genes. Now, the nice part about MTAP is you can detect it by immunochemistry, and what you're looking for is loss of cytoplasmic staining. So, here is a mesothelioma which has retained MTAP, which shows a brown here. Here's a mesothelioma that has lost MTAP, and the brown staining cells are inflammatory cells. That's your internal positive control that you really want to see. So, this is a good substitute for fish. It's not perfect, but the sensitivity for CDKN2A deletion is 70 to 80 percent. NF2, for which the protein is named MERN, is also a useful marker. As I mentioned, it's said to be a late loss of NF2, said to be a late effect in mesothelioma. So, the sensitivity is probably 40 to 50 percent, but here is a mesothelioma with retained NF2, MERL and staining, and that can be either membranous or cytoplasmic, and here's a mesothelioma which has lost NF2, and again, we've got nice internal positive control to tell us the stain is working properly. So, all three of these markers are very useful, or actually four markers, and you can see any combination. Here, for example, is a mesothelioma that has lost BAP1, but retained MTAP and NF2, MERL. There is another piece of information that comes out of this, and that is tumors that show this pattern of BAP1 loss, MTAP, and MERL in retention are said to be sensitive to PD-L1 inhibitors. At least, that is the theory. There are a number of articles in the literature that suggests this should be true. Lastly, P53, a gene that is mutated in lots of tumors, can be mutated or deleted in mesotheliomas, and for the pathologist, you follow the same rules as you do for a serous carcinoma. Look for either overexpressor or null patterns. The problem with P53 is the sensitivity is very low. We're talking about 15%. Here's an example. So, here's a mesothelioma. BAP1 is retained, MTAP is retained, MERL is retained, but P53 stains every cell quite intensely. So, this is the overexpressor pattern, and this is an indication that you're dealing with mutated P53. How do you put these together? You don't need to see loss or abnormal markers for all of these genes. Loss or abnormality of any one of them says the process is malignant. It's important to realize that the sensitivity varies by site and by marker. BAP1 has a sensitivity probably around 70% for epithelioid mesotheliomas in both pleural and peritoneal cavity, but as I mentioned, very low sensitivity for sarcomatoid tumors. CDKN2A or MTAP has much better sensitivity for sarcomatoid tumors, but lower sensitivity in the peritoneal cavity compared to the pleural cavity. NF2, it's a late event, but if you see loss, that's malignant. P53 is a perfectly good marker, it's just very insensitive. And the other point to remember is these are not mesothelial-specific markers. Some people try and take loss of one of these markers and say, okay, that makes a tumor that I can't otherwise diagnose a mesothelioma. That's not true. All of these markers can be seen in other types of tumors. What they're important for here is saying that something is benign versus malignant. If you're interested in this whole topic, this is a review article we did a few years ago. It's a little bit out of date, but it talks about all of the markers that have been proposed. And thank you for listening.

mesothelioma

asbestos exposure

histology

WHO 2021 classification

immunohistochemical staining

genomic testing

BAP1

CDKN2A

pathology advancements

therapeutic strategies