Dr Eve Crane - EBV-related lymphoproliferative disorders

[Music] So, it's a very exciting time. I think, well, and it has been, to be part of combating pathology, particularly now as we have an incredible amount of data coming out that's really informative on the mechanisms underlying these disorders. And that is the reason why the classification schemes have really undergone such significant revision. It's because we're really knowing more about these disorders and better ways that we may want to so classify and and target therapy for them. So today, I don't know if it's going to be everything. Uh, there's always more to know about EBV associated with disorders, but, uh, some of the emerging, uh, areas that I think there's gonna be some new data that may be of interest, as well as some of the important revisions to the diagnosis coming out within the WHO, as well as the International Consensus Classification. Um, let's see if I can get it to work. All right. So, yeah, just a brief overview. I'll start with the case presentation, discuss some of these proposed revisions of classification, particularly relevant to Burkitt and immunosuppression, immune dysfunction related lymphoproliferative disorders. And there are actually some very interesting ways we're discovering that lymphoma disease subtypes are linked to specific hosts and clinical factors. And finally, some emerging concepts, uh, which are not quite yet, uh, ready for the WHO, but things that may be particularly relevant to viral-driven lymphomas that may allow us to better target them, including viral effects on host cell metabolism, manipulating the viral latency state to make these, uh, lymphocytic processes more recognized by the host immune system, and using, um, improved methods of viral DNA detection as a better screening tool.

So, case presentation. This is one that I had seen a while back. A young woman who had basically a series of episodes of fairly significant, uh, lymphadenopathy, cervical lymphadenopathy, always had resolved spontaneously, either with antibiotics or on their own. But she had one that was just a little bit worse with fevers and things, uh, unresponsive to antibiotics than she typically had. So a biopsy was performed. She had had a prior history of mononucleosis as well, so it was not thought to be acute EBV in this setting. And what she has is extensive necrosis that we can see on either side of the biopsy, and then kind of a little bit of a mixed atypical lymphoid infiltrate in here. We can see quite a bit of its positive for CD20, including some of these areas of necrotic cells were also significantly staining for for B cells. Looking at them a little bit closer, some of these cells look quite atypical with some vesicular nuclei and irregular nuclear contours, and we did find that they had kind of an activated B cell phenotype. In addition, many of them, although kind of sub-optimal in this tissue with extensive necrosis, did have an EBV association by the non-coding RNA EBER. Because the biopsy was so scant, they went ahead and got an excisional biopsy, and we get a better look at some of these large atypical cells that were interspersed here. This kind of partial nodal involvement by the node that they went after, but we can see that there was some extensive EBV positivity there, essentially getting all of these large cells. And overall, um, in addition to that, it did have complex cytogenetics. So kind of the idea that it could be just a reactive process really went out the window with this, and it was clonal, um, by Ig rearrangement as well. So a diagnosis at that time was made of an EBV-positive diffuse B-cell lymphoma. And I will just note that the WHO Fifth Edition now just is dropping this "not otherwise specified" and just making these EBV-positive diffuse B-cell lymphoma. Unfortunately, uh, the patient was still a little bit reluctant because she had had prior bouts of this sort of similar, uh, lymphadenopathy and wanted to kind of hold off and on therapy, and she had received some steroids as well. And so obviously, you know, you can get a short-term response to the steroid administration. And so they were monitoring her lactic dehydrogenase as well as quantitative EBV levels, which initially diagnosis were quite high, and of course dropped after the steroids. Then her LDH started rising again, although still within normal levels, but the EBV did not. Um, and so they went ahead and performed a PET scan, another CT PET scan to clarify before initiating therapy. And while the neck nodes had substantially improved, she had substantial infradiaphragmatic lymphadenopathy with these very high SUVs up to 52. And so they went after one of these nodes, and indeed we see something that looks relatively similar with these large, you know, vesicular nuclei, looks like a large B-cell lymphoma, quite a bit of pleomorphism in here, and more diffuse the basement than we had seen initially. And the overall phenotype was pretty similar. They had their, obviously, she had not yet been treated, so positive for CD20 and had a similar activated B-cell phenotype. However, uh, looking at EBER and consistent with kind of the drop in her EBV titers and her DNA, this was now negative, and only isolated lymphoid cells, you can see kind of have EBV positivity here. And, and so that was quite striking, um, especially as we then got cytogenetics back, and it is definitely the same process. However, it had a new deletion in 2q and had become tetraploid, um, rather than the complex diploid, uh, karyotype, and indeed the same clone. So this appears to be, um, an example where we actually were able to catch it in the act of being an EBV-positive diffuse large B-cell lymphoma that subsequently underwent loss of its EBV dependence, which is something that has been rarely reported but very difficult, um, historically to prove. And her, she was otherwise immunocompetent patient, and so we believe that in the setting, probably the immune selection had driven the loss of Epstein-Barr virus, which is normally not incorporated into the chromosome but maintained as an episome separate, uh, from the the normal patient DNA. So it is possible that you could undergo loss of EBV, and then it may only be there initially, um, as a part of tumorigenesis, and may drive, uh, the initiation of lymphomagenesis. So it's quite an interesting, uh, case, and it kind of makes you wonder, well, you know, how many of these are there that maybe we just don't catch that you initiate, um, the tumor genes with EBV, particularly in like a transplant setting where you have EBV reactivation, but could you catch it? Um, and so as we know, EBV is present in a latent state in most of these lymphoproliferative processes, which involves methylation of the viral DNA. Only a few of the viral products are transcribed, such as such as EBER. But this methylation of the DNA of the virus also can result, and does result, in epigenetic silencing of host, uh, genome. And so potentially these changes, and which often include some tumor suppressor genes, you could see a signature, a methylation signature that would indicate that this process did come from an initial viral-driven process, or at least hypothesized that it did. There's been another study as well that there may be some residual viral products that we can detect in these instances. Um, of course, this is, uh, just a table. Don't get too overwhelmed by it, but just a table, uh, showing you kind of, and this is one that we've published if you want to take a look at it, kind of all the different products of EBV that are expressed in the different latency states. And as you note, a non-coding RNA is present in all of them, and, and it's present at high levels, so that's how we normally look at that. Um, but EBNA1 is also present. It's what actually tethers the EBV to the genome. And so this is another good thing to potentially look at, but it's expressed at very low levels. It's kind of hard to normally detect. But a paper came out relatively recently in Modern Pathology where they looked at EBNA1 using actually a very sensitive technique, RNAscope, and they found, for example, in EBV positive, uh, cases, this EBV positive Burkitt case, you can see lots of these little dots that are consistent with kind of tethering it, uh, in the nucleus there. Um, but in cases where there had been a history of EBV relation, you would actually find isolated cells showing these these little dots. So there's little traits of EBNA1 that you could actually detect and find these hit-and-run cases. So we'll have to see. You couldn't see it in cases that were EBV negative, and then also where PCR negative by quantitative PCR looking for this protein. So it's kind of interesting why they would necessarily retain just that mRNA and not not EBER, at least in a detectable fashion, but kind of intriguing that we could potentially detect these cases. And then you say, well, okay, maybe you could detect them, but it doesn't really matter. Um, most, in most instances, in the current state, we don't necessarily treat an EBV-positive disorder. For example, in Hodgkin lymphoma or Burkitt lymphoma, we don't necessarily treat these, uh, differently if they're EBV-related or not. Um, but the question is, maybe we should, and maybe we will. Um, and this is one of the changes that's arisen out of the the Fifth Edition of the WHO is that maybe we do actually in some instances want to recognize based on EBV status, particularly for one of our more most characteristic, uh, lymphoma processes, Burkitt lymphoma, which has some of the most distinct clinical, morphologic, and genetic features of any thing that we attempt to diagnose. And of course, this is a nice cytology image, I believe this is from Amy Chadburn at Weill Cornell, looking at this, the classic morphology of Burkitt lymphoma with this very deep, uh, basophilic cytoplasm and cytoplasmic vacuoles that are quite characteristic. And of course, the tissue section is also quite characteristic with your medium-sized lymphoid cells showing its starry sky appearance because of their very high mitotic rate, you get a lot of debris that these things are eating up here. They usually have a MYC translocation, typically to an Ig locus, and a very characteristic immunophenotype with a germinal center B-cell phenotype, but negative for BCL2, and basically every cell is undergoing division with Ki-67. Now, there's some variability in terms of EBV association, and that depends on, uh, subtype, with no single feature really being definitive. So historically, and in the Fourth Revised Edition of the WHO, we think of kind of three epidemiological variants of Burkitt lymphoma, which is certainly a very reasonable way to think about it. There's an endemic form in equatorial Africa, Papua New Guinea, a non-endemic kind of more sporadic form with a more worldwide distribution, and immunodeficiency-associated in equatorial Africa. That is the most common childhood malignancy, peaking around ages four to seven years with a male predominance, and a very characteristic appearance, kind of often in the jaw or facial bones or the orbit, with very unusual involvement found in bone marrow or leukemic phase. And the vast majority of these cases are EBV positive, and EBV is in almost all of the lymphoma cells, and there appears to also be an overlap or potential interaction with endemic malaria. So quite an interesting and quite, um, characteristic process. Meanwhile, sporadic Burkitt, we get kind of still get the MYC translocation, same phenotype, but a very different kind of presentation. It is very common in children as far as lymphomas go in in the United States and Western Europe as well, around 30 to 50% of the lymphomas in that population, but more often EBV negative. Can be EBV associated with, tends to be more in older individuals, intensive presenting areas such as the abdominal cavity, ileocecal area, ovaries, kidneys, breasts, etc. And around 20 to 30% have EBV. Finally, immunodeficiency-associated is also quite interesting. We know that a variety of forms of immune deficiency will increase risk of certain types of lymphoproliferative disease, but often this is not as commonly seen, for example, in transplant. It's very kind of characteristic in the setting of HIV, and unlike things like primary central nervous system lymphoma, we see this when the immune system is relatively intact, CD4 count greater than 50. And unfortunately, the risk persists even in patients on adequate antiretroviral therapy. And there's a number of hypotheses related to this, potentially even related, uh, to effects of the HIV Tat protein within the nucleus increasing the risk of a potential MYC translocation. And in contrast, uh, to what we see in the endemic form, only a subset of these are EBV positive. So it's not necessarily, you know, just extremely poor control of the virus that leads to it, but potentially other factors are at play.

So, Burkitt lymphoma is particularly interesting also because we always think about the characteristic MYC translocation, but it also, uh, that's not sufficient, and it's also of course not specific. We see MYC translocations in a variety of large B-cell lymphomas. Um, so it also shows a very distinct molecular, uh, as well as gene expression profile. And the molecular profile, and this is for sporadic Burkitt, is quite interesting and distinct from that, for example, of other germinal center large B-cell lymphomas. Some of these are shown in brown here, DLBCL, large B-cell lymphomas often have these mutations in EZH2, BCL2, whereas the most commonly mutated gene in Burkitt is MYC. So in addition to having a translocation, there's typically quite a few mutations in this gene as well. And then in sporadic Burkitt, we see mutations involving TCF3 and ID3. These two proteins normally interact to inhibit, uh, B-cell receptor signaling, and so those are seen there. Okay, so where are we going with this? Well, is there a difference between in these kind of characteristic mutations between EBV-positive and EBV-negative Burkitt lymphoma? And in fact, there is. Well, we see this like TCF3, ID3 pathway being affected in sporadic Burkitt and increasing B-cell receptor signaling, activating kind of a tonic form of antigen-independent BCR signaling. We don't really see that as much in the setting of EBV-positive Burkitt lymphoma. It has fewer driver mutations, and in certain studies, they found an almost mutual exclusivity between the presence of EBV and Burkitt and these TCF3, ID3 mutations. And instead, um, you know, chronic BCR stimulation may be from other sources, either from potentially the viral products or, you know, potentially other pathogens, because we definitely do see an interaction with other potential, uh, pathogens, certainly in, um, endemic endemic Burkitt with malaria is a possibility. The other thing, um, that's quite interesting from a treatment standpoint is that while we have decreased driver mutations, that is, you know, things that we know are oncogenic or known, um, loss of tumor suppressors, and we also have fewer mutations associated with apoptosis, if you look at all comers, they're actually quite a lot greater mutational burden in the setting of EBV-driven Burkitt lymphoma as opposed to EBV-negative. And there seems to be increased somatic hypermutation, increased activity of activation-induced cytidine deaminase, and defects in, and mismatch repair. Now, these defects in mismatch repair appear to occur in the context of retained apoptotic pathways, and so that means you may have a particular susceptibility for the EBV-associated Burkitt to DNA damaging agents. So all of these things, like looking at not just from the diagnostic standpoint, but also having that molecular picture may be helpful going forward in developing targeted therapies.

So, taking all this into account, and this is a text directly copied from, um, the Leukemia article that has recently just come out from the WHO in regards to the proposed changes to the Fifth Edition of the lymphoid monograph, the hematolymphoid monograph, that rather than now looking at the epidemiological context and geographic location of these Burkitt lymphomas, it may be more important to think about them as EBV-positive Burkitt and EBV-negative Burkitt, and they may form discrete biologic subgroups based on these molecular features, regardless of that epidemiologic context. And, you know, probably additional subgroups going forward as we look at, you know, precisely targeting what sort of mutations are there and what vulnerabilities these processes may have. So I thought that was pretty interesting, um, and certainly a kind of a revision in the way that we think about things and and really thinking more mechanistically how we can potentially come up with the best outcome for patients.

Another area that I think is particularly exciting within the revised WHO, this was not as much addressed by the International, uh, Class Consensus Classification, um, was a potential revision to post-transplant and other immune suppression-related lymphoproliferative disorders. In the Fourth Edition, the Fourth Revised Edition, we have a non-destructive category, used to be early lesions, that was kind of confusing. These tend to be, uh, hyperplasias, often associated with EBV. We've got the polymorphic category. These are destructive lesions, often associated with significant necrosis and things that would look like a lymphoma in an immunocompetent patient that we can put into a standard classification, such as a diffuse large B-cell lymphoma, and then classic Hodgkin lymphoma. With the revised classification, there's been a recognition that we can see not just things, so sorry, one thing we can see all of these processes not just in the post-transplant setting, but at, um, in a variety of settings of immune dysfunction and immunosuppression. And we have now, we have chapters on HIV, a chapter on PTLD, there's iatrogenic sort of methotrexate-related lymphoproliferative processes. But outside of the transplant setting, these were not as routinely recorded or consistently addressed in the same way, and it was very confusing. And so not necessarily like everything that has the same histologic diagnosis are we now going to treat the same way, but at least now we have a consistent, um, or they propose a consistent nomenclature to keep track of these things. And they're going to fall into hyperplasia, um, you know, such as some often EBV-related, because otherwise it's hard to, uh, really classify as that, polymorphic, which we know can happen in the setting of HIV as well as, um, in the setting of iatrogenic therapy, nicotinic cutaneous ulcer. This is a new and very important emerging category where we're seeing these not only, um, in transplant, but in the elderly, other forms of immune suppression, and, you know, things that are technically a lymphoma, such as diffuse solar species lymphoma, then we'll indicate whether or not it's associated with a virus of some kind, and what is the situation in terms of immune deficiency or dysregulation. And this is important because this list is ever expanding as we get novel forms of ways to modulate the immune system, either in a negative or positive way for autoimmune disease or chemotherapy. And again, a patient who has a transplant may be treated differently from the clinical standpoint than someone who has autoimmune disease, but at least now we have kind of an organized way to keep track of where these things are arising and how they're associated with a potential virus. And this is just an alluvial plot, um, showing kind of how the shift in the diagnostic categories looks. Again, now instead of looking at post-transplant, HIV, iatrogenic, primary immune disorders, we're now going to divide them more on a morphologic standpoint, um, rather than continuing to proliferate this list on and on and trying to describe the same things separately, um, so many different times. But there is still, um, a subset of lymphoproliferative disease that may be relatively specific to congenital immune deficiency or inborn errors of of immunity. And so this also remains a separate category.

Um, so the International Consensus Classification, which emerged out of the Society for Hematopathology and European Association for Hematopathology, as well as carrying on a tradition of a, um, of the clinical associations that had been previously, the clinical advisory committee that had been previously been used with the WHO revisions, also did promote one classification change was that they were going to recognize this EBV-positive polymorphic B-cell lymphoproliferative disorder across settings, um, not just now in PTLD, but also in HIV and other settings. And this is actually a brain section from an HIV patient with a relatively intact CD4 count that did not have primary CNS lymphoma, but instead had more of a polymorphous proliferation. These are EBV-positive cells, the mixed kind of CD20-positive cells in the background, and that patient responded very well to just rituximab. It was not an aggressive process, and, and she did not have, and she had adequate ART therapy. So, so these things kind of do need to be separately, uh, designated. And this is kind of how it would look in the two different classification schemes, which I guess we'll be using, uh, going forward, two different schemes for the time being until we reunite. But sort of a polymorphic LPD, EBV-positive, setting of HIV would be the, what I understand from the Fifth Edition, and an EBV-positive polymorphic LPD, not otherwise specified, in the International, uh, Consensus Classification. And as I mentioned, this may be particularly important because there are just so many different types of immune dysregulation that are emerging, and I think we just still need more data to understand how they're relating to lymphoma proliferative disease, and this will create a mechanism where we can keep better track of it and, you know, have this data readily searchable within our pathology archives and available for discussion going forward. And these may include things like multi-agent chemotherapy, which we know is associated with immune deficiency, but exactly how, which regimens, how long these remain areas which are a little bit unclear. Novel immune modulatory agents do result in changes in the lymphoproliferative disease spectrum, and I think including these are are going to be very interesting. And there's also other things, you know, that they mentioned, aftermath of CAR T, checkpoint, and inhibition therapies, how are these things, you know, potentially related? We don't necessarily need to know, um, from this classification scheme, but it's a way, a framework for incorporating this data along with immune senescence, and that, um, and and things along that line, which also remain fairly poorly defined. And so they have it now, immune deficiency/dysregulation to kind of incorporate this wide spectrum. And so this is just another example of where this sort of nomenclature could come in handy. We know that EBV-positive marginal zone lymphomas are now included as part of the spectrum of post-transplant lymphoproliferative disease. And this was in the Fourth Revised Edition, but it's not otherwise, um, extensively, uh, discussed within the the revised Fourth Edition. But we know that they can occur basically in any of these immunodeficiency settings with autoimmune disease, um, inborn errors of metabolism, immune senescence, plus or minus chemotherapy, and in HIV. So we now have a method to kind of recognize, um, these regardless of setting, basically for follow-up. And the importance of that is that they may respond differently to therapy. Maybe restoration of the immune system will be sufficient, or at least a component of the therapy.

So, other specific associations. One of which we discussed already was HIV in association with Burkitt lymphoma. This sort of scheme would allow you to like begin recognizing, hey, you know what, every time we see Burkitt lymphoma, it's more often in the setting of HIV than than transplant. So it's kind of, we already know that, but there may be other things arising that this will then draw attention to. And HIV, of course, was also associated with primary CNS lymphoma, that has been addressed by ART to a significant extent. But patients with untreated HIV had a very had increased risk of lymphoma around 60-fold. But the increase for this primary CNS in the absence of ART was around thousand-fold, and 30% of lymphoma cases, and otherwise rare. Similarly, in the post-transplant setting, it's very interesting. Early cases also were very likely to involve the brain, around half of the cases were in the central nervous system, often isolated to the brain. And then that kind of the whole paradigm shift. These patients were early patients were on azathioprine in the late 60s and early 70s before cyclosporine was introduced. And then with cyclosporine, we saw kind of a shift going from the brain to being more common in the graft or gastrointestinal tract or maybe as disseminated disease. However, the overall rate remained the same. And this is just a brief example that I published a little while ago looking at kind of how shifts in the medications that we use, which now, we're having a much larger explosion of the number of different medications that are used in both the transplant and autoimmune setting, but these can definitely shift the type of, um, lymphoproliferative disease that we're seeing, both the type and site. Um, so, so we saw kind of beginning to arise in the early 2000s, and again, the fraction of cases that were in the brain as opposed to systemically, and eventually being around 40% of the cases were actually primary CNS PTLD, which was sort of unheard of, kind of in the late 80s and early 90s. And trying to sort some of this stuff out ends up to be a little bit of a problem because patients are on multiple drugs at once, and these drugs may change over time to optimize other aspects of their immune suppression for a transplant. But quite strikingly, almost all of them were taking mycophenolate. But this is often in combination with cyclosporine, where we didn't see it before, or or to this, which is also a calcineurin inhibitor. This is just an example of one of these patients here who had a renal transplant and was on mycophenolate and the calcineurin inhibitor, tacrolimus, and we kind of see these perivascular, very aggressive looking B-cell proliferation in there that was EBV-positive. So in order to kind of get at this, at that time, we pulled data that came from the United Network for Organ Sharing, which is basically something maintained by the Health and Human Services division in the United States, keeps track of all of our solid organ transplant patients. So it's a very enormous database, not always entirely all the fields filled up as you'd like, but still a very large, you know, thousands of cases of PTLD along with their drug regimens and other factors. And we found using this, using a logistic regression model, I got the help from the School of Public Health to do this. We found that mycophenolate, even though a lot of patients were on it, was independent of other medications they may or may not have been taking, was significantly associated with this, uh, brain localization of the PTLD compared to other sites. While the calcineurin inhibitors, including both cyclosporine and tacrolimus, even if they were taking it together, did partially protect from getting a central nervous system, uh, PTLD, it tended to arise instead at another site. And in the opposite of what we see for most PTLDs, it was actually protective to have either a liver or thoracic transplant compared to a kidney transplant. So quite interesting, but certainly, kind of as we introduce these new drugs, you can get a shift in the type of lymphoproliferative disease. And I think, you know, we're keeping a closer track of that now. Um, and one of the interesting things about these is that they're the CNS cases of PTLD, in contrast to those that we see systemically, are always EBV-related, essentially every case, and they don't decline with time. Like even some that are 10 years out, they're still EBV-related. That's mostly 20% of systemic, uh, cases. And one hypothesis, although it's not something we can necessarily test very easily, is whether or not these calcineurin inhibitors, which are seem to be protective against EBV-related lymphoproliferative disease, is it really a risk of the mycophenolate, which I think it is, um, but it's also a decrease in the protection of the brain that we're inadvertently giving, um, because azathioprine also increased the risk of CNS lymphoma quite a bit, and, and that went away once patients were getting these calcineurin inhibitors. And, um, of course, I'm a pathologist, so I don't, I'm not on the patient's side anymore. I did, uh, brief surgical residency where took care of transplant patients and dosing cycles.

Foreign is really a pain because it's very, um, hydrophobic and it likes to all go into the brain. And so you don't get very, the systemic levels are are hard to get completely stable, but you do get high and stable levels in the brain. And interestingly, a work from Richard Ambinder's lab, he found that, whoa, um, that it actually will inhibit the lytic cycle activation of EBV. And while you don't necessarily have much lytic cycle activation, a typically latent, uh, viral cycle, that there is this, there's a certain extent of lytic cycle activation that helps continue to drive tumorigenesis in these cases. And so it's possible that this high levels and localization of calcineurin inhibitors in the brain was actually kind of protecting the brain from developing PTLD. The other thing that's somewhat intriguing and also challenging to test, but it's now, um, becoming somewhat feasible to test, is whether or not we see, because we see this very striking difference, particularly with time, an EBV association between brain cases and systemic cases not involving the CNS in terms of EBV relation, that whether or not because the CNS is a, you know, immune privileged protected site, that we don't see the immune system-driven loss of EBV in those cases. But these may be more likely to be hit-and-run cases that are systemic, but we can't have really hit-and-run in the brain because it can't really attack it there. So I think it's definitely an intriguing idea. Probably won't be all of them, but I think we may be underestimating the number of hit-and-run cases in systemic PTLD.

And of course, um, we also see this in patients with autoimmune disease, and this is something that has been harder to track because we often don't record that in our pathology report or for consultation is sent, it may or may not be mentioned. So I think that it's very nice that we're going to have the standard thing, well, is the patient on any immune suppression, any immune dysregulation present? You know, we may not know what to do with that yet, but it could be listed, and that may allow earlier recognition of kind of rare complications. For example, of mycophenolate can also result in CNS associated lymphoproliferative disease. This is a 76-year-old lady who was treated for eosinophilic granulomatosis and polyangiitis on mycophenolate and steroids for about 10 years before she developed these, um, brain lesions, one seen very well here, and developed foot drop and, and some other focal neurologic signs. What we had was kind of a Hodgkin-like proliferation with these large atypical forms and kind of mixed kind of histiocytes, small lymphocytes in the background. We got Dim Packs 5, uh, characters that could find these Reed-Sternberg-like cells, which were extensively positive for EBER, but there's also quite a bit of EBV reactivation in other cells or EBV positivity in other cells. So it's not really a Hodgkin, uh, form, but some sort of positive like lymphoproliferative disorder. But regardless of how we classify it, um, what's been found is that if you take these patients off mycophenolate, and, um, even just with rituximab monotherapy alone, they seem to do extremely well. She actually had resolution of her lesions after four months and no longer seemed to also require immunosuppressive therapy for her underlying autoimmune disease. And other small series have identified this as well, that, you know, we don't necessarily want to tell these patients that they have, you know, you have primary CNS lymphoma because these lesions seem to melt away once you get them off of mycophenolate. And in the context of rituximab, multiple cases have been reported, they're polymorphic, they're monomorphic, and the patients seem to do well. Obviously, we'll need a little bit larger experience to know that, uh, this is, you know, going to be in every case or if there are any other factors that we may want to take into account. But certainly keeping track of these sources of immune suppression may be very important for patient care. And that basically, um, says that, yeah, and then in case you're wondering, well, how often are we going to find stuff like this? This is another example that that we've known historically as well, the potential association between young men with inflammatory bowel disease, such as Crohn's disease, who are treated with thiopurines, sometimes in the context of anti-TNF alpha therapy as well, and the risk for development of hepatosplenic T-cell lymphoma. So I think there are going to be more of these, and particularly as we find more innovative ways to target specific aspects of the immune system, we're going to see more kind of unusual or characteristic settings for for these lymphocyte processes emerging. And this may have implications for therapy, but there may also be particular patients with particular susceptibilities that may inform us, you know, maybe this is not a good drug to use in this context. So I think it's going to be very exciting going forward as we learn more about the interplay between the immune system and these EBV or other lymphoproliferative processes.

So, um, just now, uh, just going over the last part of the talk, just a few emerging concepts. One area which I think is particularly exciting is the viral effect on host cell metabolism. It causes a lot of changes that are sort of a hallmark for cancer, but because it's metabolism, it's also amenable to small molecule targeting. And then a couple of other topics, including, uh, folks developing methods to manipulate the EBV latency state and measuring, um, viral DNA. So basically, we saw earlier, um, with Burkitt lymphoma that the EBV tended to lower the threshold for lymphomagenesis. You didn't need as many driver mutations for Burkitt lymphoma in the setting of EBV. That also seems to be in the case of the diffuse large B-cell lymphoma and classic Hodgkin lymphoma that that can, um, be basically one of the hits that you get. And why is that? Um, EBV itself is extremely potent. It does infect B cells. It's latent reservoirs normally in memory B cells, but you can in the laboratory infect, um, just peripheral fresh, uh, peripheral B cells that don't have a mutation, and EBV can immortalize those cells. It's quite powerful in terms of reducing proliferation. Of course, we see that with mononucleosis, and the immune system ultimately gets it under control. So EBV is doing something to alter the host cell metabolism, survival, and, and proliferation, um, which acts a little bit differently in the dish, obviously, than in the body. But many of these pathways are similar, kind of to what we think of basically effects that will be hallmarks of cancer that have recently been included in the hallmarks of cancer. Some of these metabolic changes, including, you know, increased glycolysis, increased anabolic growth, mitochondrial biogenesis, all of these are induced also by EBV. And various metabolism metabolites can alter the cell signaling or DNA methylation or even the differentiation or ability of that cell to differentiate. And this is kind of just a summary diagram that we put together from some of the major labs that are looking at this. When EBV is EBV, uh, here infecting a cell, initially does activate the NF-kappaB pathway, but it finds a way to keep that activated through LMP1 signaling, that's seen in Hodgkin lymphoma as well as latency 3, polymorphic PTLD, and other processes. And EBNA2, which is also seen in kind of latency, latency 3 settings, turn on MYC and mitochondrial biogenesis, get increased steering metabolism with purine synthesis. One thing that was quite interesting that came out recently was this HMG-CoA reductase pathway, which of course can be targeted with drugs, was that it was needed to make this particular small molecule, geranylgeranyl pyrophosphate, that was basically needed as part of a chaperone protein to get LMP1 to the membrane. And if you could actually disrupt this process, which they looked into doing, LMP1 couldn't get to the membrane, couldn't keep going with its NF-kappaB signaling, and you could actually, um, slow growth of these EBV-infected cells. So that was quite interesting. So I think this area is just really at its, um, kind of beginning. We're learning a lot about how the virus, not even just, you know, you don't have to even be making virions to disrupt the cellular metabolism. Even latent viruses are disrupting cellular metabolism, and that may create synthetic lethal opportunities to attack it that won't bother a normal cell, potentially.

Um, another area that's exciting, and this comes into play with Burkitt lymphoma in particular. So as we know, kind of depending on host immune status as well as the type of EBV-positive lymphoproliferative disorder, it's going to have a different EBV latency pattern. Sometimes people kind of blaze over, you know, once you start talking about EBV latency pattern, it's like, oh, you know, what now? You know, we just look at EBER. Um, but the thing is, for example, LMP1 is on the membrane of the cell, and it's it's immunogenic. So if you had cells that are expressing LMP1 and you can get the immune system to go after them, you know, that that may be quite powerful. And certainly with polymorphic post-transplant, they're often latency 3, and if you can kind of restore the immune system a little bit, the body may be able to help take care of it. But in Burkitt, they really don't have, um, much immunogenicity, and often, you know, you really only are going to see EBNA1, which is kind of nuclear and poorly immunogenic. But we're seeing Burkitt in HIV patients when their immune system is relatively intact. So if the Burkitt lymphoma could maybe express a little bit more EBV antigens, you know, you could get the immune system to attack it. And so we talked a little bit about methylation status, and obviously EBV is methylated enough, lighting other things, you know, maybe a hypomethylating agent could actually increase expression of some EBV-related antigens. And in fact, doing a screening method, this is at the Sussman's lab at Cornell, doing a screening method, she actually came up with the hypomethylating cytarabine, um, could induce these EBV antigens, and so you can actually put these, you know, kind of Burkitt cell line into a mouse model, give it cytarabine, and the cells could then be attacked by EBV-specific T cells. So this is a potentially very exciting way to kind of manipulate the cells and take advantage of you, and you can also administer obviously patients even these specific T cells as well, or have their own, uh, host immune system may be able to do better to attack it. Introducing EBV lytic activation has also been proposed. That's a little bit more complicated, and you may also get, um, undesired consequences with a whole bunch of virions coming out. You may get some sort of cytokine storm, or maybe possibly contribute to tumorigenesis. We don't really know. Richard Ambinder's lab has also looked at kind of another method, again, resulting from various screening assays, that an antibacterial agent plus a kinase inhibitor may allow some lytic cycle proteins to get activated, but without virions. These are potentially very exciting areas that are kind of on the horizon.

Another area which could potentially be implemented in the short, short term, I think it's very exciting. We often have, you know, patients in the post-transplant setting or with other forms of immune deficiency that are sick, fevers, lymphadenopathy. Well, is it a lymphoproliferative process? Is it an infection? You're going to often see EBV reactivation in that setting. You know, monitoring just EBV DNA may not be, um, super sensitive or super specific. But because within a lymphoproliferative disorder, the EBV DNA is methylated, you can actually look at the methylated DNA and develop sort of an, and this group, Rena Shen and others at Hopkins, also with Dr. Ambinder, we're looking at kind of a methylation index, looking at EBV and the methylated versus unmethylated fraction. And this turned out to be quite sensitive, um, in these patients who otherwise can be, you know, it kind of can be challenging, and may get biopsy, maybe non-diagnostic. So this could allow kind of early detection, hey, you know, methylated EBV DNA is rising, looks like, um, lymphoma process is brewing, you know, then maybe that can even be addressed without a biopsy. So I thought that was particularly exciting. We have not yet implemented that here, but I think it would be intriguing to do so.

So, in conclusion, um, just a very exciting time, I think, in the setting of, um, hematopathology in general, given the amount of data that we have giving us insight into genetic mechanisms into disease, but particularly looking at EBV-associated lymphomagenesis. We're kind of learning a lot more about the interaction of viral products with host cell metabolism, other genetic drivers, and the host immune system that may allow us to better and more specifically target these processes. And also these hit-and-run cases may also be interesting and may deserve some sort of special, uh, treatment or maybe, you know, particularly vulnerable to something. And, and that area remains, um, and, you know, kind of an exciting, uh, new frontier to explore, really now that we have some tools. Um, so, yeah, so I think going forward, we're going to see further revisions to our diagnostic categories that that will allow us to figure out better targeted therapy for for certain groups. We'll see with the targeted, you know, with the improved immunogenicity of these tumors or synthetic lethals with DNA damaging agents, etc.

[Music]

[Music]