Perfect, so thank you so much for joining us today. This is the ClinGen Somatic Cancer and Vick Virtual Molecular Tumor Board Case Series hosted by the ACMG Genetics Academy. Today, we're really excited to have a great team from Northwestern University in Chicago, Illinois to talk to us about therapy-related neoplasms, diagnosis, and clinical perspective. Next slide, please. So this series is intended to be an educational resource for all healthcare professionals that are at risk. Well, it's not meant to drive specific clinical decisions or anything like that. So next slide, please. This series comes as a collaboration between the Variant Interpretation for Cancer Consortium, which is a GA4GH driver project, the Global Alliance for Genomics and Health, and also the other entity involved in this collaboration is the ClinGen Somatic Cancer Clinical Domain Working Group. Both of these groups do a large amount of volunteer work helping to create standards and guidelines for cancer variant interpretation. And we couldn't do any of this without our generous hosts from the ACMG Genetics Academy. Next slide, please. So this is the planning committee that puts on this webinar series. We have Cordana Rapsack, me, I'm Beth Patel, Manuela Benari, Jason Saliba, and Alex Wagner as the participants from ClinGen and Vick. Jane Radford from ACMG is kind of the program manager from the ACMG staff. And none of this would have been possible without my previous co-lead of the Vick Virtual Molecular Tumor Board and administrator for a ClinGen, former administrator for the ClinGen Somatic Working Group, Shruti Rao, who is no longer part of the planning committee, but a very large part of why this series has been established. So next slide, please. So this session has been approved for one AMA credit. Next slide. And in order to obtain that credit, please complete the evaluation in the ACMG Genetics Academy following the presentation. Please allow three hours to view and print the certificate for one AMA credit. Next slide. A recording of the webinar will be available in two weeks online at the ACMG Genetics Academy. And there's the URL. And then to submit questions for this session, we're asking you to use the Q&A feature and not the chat. Please only use the chat for technical issues. If you're having problem with video or audio, those kinds of things. But please use the Q&A and you can use uploading if you see a question that you're interested in hearing an answer to, and we will pay attention to that at the end of the session when we move into our Q&A discussion. So now to the fun part. So thank you so much to today's speakers, Dr. Yasmin Abaza, Dr. Juhua Gao, and Dr. Medina Sukunova, all from the Feinberg School of Medicine and Northwestern University in Chicago. We're really excited to have you here and thank you for sharing your expertise with us. Next slide. Here are the financial disclosures. So today's learning objectives are, at the conclusion of this session, participants should be able to discuss the clinical pathologic features of therapy-related neoplasms of myeloid and thyroid origin, identify genetic factors predisposing to therapy-related complications, and discuss the current status of clinical management of therapy-related neoplasms. And with that, I'll hand it off to our presenters for today. Thank you so much for being with us. Good morning. My name is Yasmin Abaza and I will be starting our discussion today with my colleagues on therapy-related neoplasms. So as you all know, cancer remains the leading cause of death in the United States and there has been significant improvement in the overall survival of cancer patients over the past decade. And this has led to the rise in therapy-related myeloid and lymphoid neoplasms, which is considered a late complication of chemotherapy and radiation therapy. Historically, therapy-related neoplasms were thought to be a consequence of DNA damage from these toxic therapies. However, there's a lot of debate in the field regarding the contribution of other factors, such as inherited genetic predispositions and environmental exposure on the development of these fatal types of malignancies. So therapy-related myeloid neoplasms are the more common and the better characterized subtype of these therapy-related neoplasms compared to therapy-related ALL. Therefore, for my portion of the talk, I'll be focusing more on therapy-related myeloid neoplasms. There are two major subtypes of these neoplasms. The first is, and the most common subtype, which occurs in about 70% of patients, occurs in patients with prior exposure to alkylating agents and results in abnormalities, mostly in chromosomes five and seven, with a long latency period of about five to seven years. And they typically present with antecedent hematological neoplasm, like MDS, prior to progressing to AML. The second subtype, which occurs in about 30% of patients, occurs with prior exposure to topo-rasmase II inhibitors. And these patients typically have a shorter latency period of one to three years, and they typically present with upfront AML without antecedent MDS diagnosis. And they typically present with translocations, such as KMT2A rearrangements, core binding factor AML, and therapy-related APL. So there are a lot of pertinent questions in therapy-related myeloid neoplasms, which are still unanswered and still we need to know more about, including why do some patients develop therapy-related myeloid neoplasms while others who have solid tumors or other tumors and receive chemotherapy do not develop these malignancies? And why is the latency period very variable with some patients taking 10 years or more to develop these neoplasms? And why are these myeloid, therapy-related myeloid neoplasms so inherently chemoresistant? And my colleagues will be presenting two cases on therapy-related neoplasms. And throughout these cases, we'll be trying to address some of these questions. Good morning. My name is Zhu Ma Gao. I will start with two cases to illustrate the therapy-related hematopoietic neoplasm. So our first case is a 44-year-old female. She has a history of invasive appendicill carcinoma, received multiple resection chemotherapy. And later, she also developed breast carcinoma, also received the surgery, tamoxifen, and also had a hematopoietic myeloid neoplasm. Her hematopoietic carcinoma also received the surgery, tamoxifen, and bilateral breast radiation therapy. She has a very complicated personal medical history. Her anemia actually started in her 20s. At that time, there was some consideration about some bone marrow failure syndromes, such as Fanconi anemia, particularly when the physician noticed she was kind of short, 4'10", but she did not have other manifestation for Fanconi anemia or dyskeratosis congenita. She does not have any birth defects, no abnormal thumbs or limbs. Her head is small, but kind of proportional for her body size, no intellectual disability. She did not have any abnormal pigmentation or nails or gray hair, but she did have some infertility issue. When she was diagnosed with appendicitis carcinoma at a very young age, 39, additional study revealed mismatched repair protein expression were all normal. She was diagnosed with bilateral breast invasive ductal carcinoma, again, at a very young age, involving bilateral breast. It's very unusual. Her breast cancer was ERPR positive and HER2 positive. Because of her young age, a mirror 28 gene panel was sent, and she was negative for BRCA and other gene on the panel. Now she presented at age of 44. In addition to her long history of anemia, she also developed leukopenia and somocytopenia. A peripheral blood smear review we received in the lab showed pancytopenia with these rare circulating blasts. These are larger cells with high NC ratio, very immature chromatin. That prompt a bone marrow biopsy. You can see the bone marrow aspirate here. Shows a lot of erythroid precursors with megaloblastoid changes, some sign of dysplasia. Iron stain, we can see these ring pseudoblasts. Again, another sign of dysplasia. She also has increased blasts in her marrow. So her core biopsy is very hypercellular with erythroid precursors also increased these immature cells. We did a immunostain, a CD34 highlights these immature blasts, about 15% of her marrow. So she now has a diagnosis of myelodysplastic syndrome. Given her complicated medical history, this was classified as therapy-related MDS. All right, I can take over to this part. Hi, my name is Medina Sukunova, and I'm going to present genetic portion of this case. You can see how we can slide for that with karyotype. So of course, bone marrow was assessed for karyotypic abnormalities. And you see that altogether, this karyotype can be characterized as a highly complex. We did some additional fish work to confirm most essential abnormalities. Mecom is not involved, although one of chromosomes three is abnormal. So we can, of course, score and see that it's very, very highly complex cytogenetic result. And what is very significant is loss of material on chromosome 5, and as well as P arm of chromosome 17, a well-known location for a tumor suppressor, TP53. This material is replaced with chromosomal material of different chromosomes. And those who are familiar with fish probes, designs, can see that fish analysis is very consistent with duration of karyotype analysis with the loss of TP53 and 5q31 region on chromosome 5. We also analyzed diagnostic bone marrow sample using in-house myeloid neoplasm NGS panel, which is not super huge. It includes 40 genes, but all 40 genes have clinical implication in myeloid neoplasms. And out of all those 40 genes, so we identified the pathogenic variant in TP53 gene, notably with a very high allele frequency. We always do this exercise with allele frequencies, and Yasmin will talk much more about the application of allele frequency in treatment of myeloid neoplasms. But in this case, two possible mechanisms. Why allele frequency of some specific variant can be very high. It can be either, as you can see here, due to copy neutral loss of heterodigosity, which results from, for example, abnormal non-disjunction of chromosomes. And some daughter cell receives free chromosomes with random elimination of a chromosome. And in that case, for example, if chromosome A has a TP53 mutation, so you can see that after random elimination of the chromosome with a wild type allele, so both remaining chromosomes have now same variant. So in that case, with the copy neutral loss of heterodigosity scenario, we have a homozygous situation. But another mechanism, and that's what we see in this case, we know results of chromosome analysis in fish. What can happen is either due to loss of the second homologue with a wild type allele, or it can be a crossover, and it can be a loss of only segment or chromosomal arm. In that case, the remaining chromosome with a mutant allele in the second chromosome can be either lost, or it can be a loss of a segment. So in these two scenarios, we do see high allele frequency, and it's always worth investigation. So in this case, as I mentioned a minute before, fish analysis was positive for loss of TP53, as well as chromosome analysis. So we knew that this high allele frequency is due to loss of a second allele due to a deletion. But still, the question remains if this variant is a germline variant or also a somatic event. And we will talk about germline testing a little more. And also, as Jiwa mentioned, that the myriad 28-panel gene, which includes all breast and ovarian cancer syndrome associated genes, as well as TP53, was actually negative. So in this case, although there was a suspicion for possibly thrombinia, it was not very high. Another way how you can also rule out a possible germline origin of a detected variant is simply by following up on that variant. So in that case, you can see that the very first follow-up sample, still allele frequency was kind of high. But then a month after, even a few weeks after induction therapy, that variant was gone. So it's very suggestive that this TP53 variant is not germline, so we can rule out thrombinia. However, there is still a lot of red flags in this patient's personal history, as well as her phenotype. So I'm not going to read all of these red flags. She denied Ashkenazi Jewish ancestry. But still, multiple primary cancers in one individual is a red flag, as well as a type of colon cancer before age of 50, bilateral breast cancer, which is also worrisome, and also time of onset before age of 45. And remember, she had anemia in her 20s, and in general, her first cancer, her second cancer. At a young age, it was all still raising a possibility of a possible predisposition for cancer. So this patient was sent for cancer genetic counseling. And as you can see, this is a drafted pedigree. Although there is a history of cancer in that family, at the same time, the pedigree, again, is not consistent with lymph thrombinia, where we would probably see a lot more affected individuals. Still, this pedigree does not rule out a possible thrombinia. The fact that she had anemia and other cancers a little later in life, compared to a typical thrombinia presentation, is not very surprising. There were cases with much weaker variants and much attenuated onset. So we could not exclude that. You can also see that this individual is of Indian ancestry. So as you can see, we excluded lymph thrombinia based on follow-up studies and also myriad original genetic testing. But to really rule out funconea anemia and myriad panel did not include funconea anemia genes, the recommendation was to send a skin biopsy for expanded inherited cancerous kennel. And the reason why we cannot still exclude funconea anemia is her pedigree cannot rule it out. And although she didn't have kidney abnormalities and abnormal pigmentation, abnormalities of thrombs, still she was rather short and had a history of infertility, as well as multiple cancers at a young age. So her skin biopsy sample was sent to GeneDx for expanded cancer predisposition panel, which we didn't score how many genes. Way over 200 genes, including all funconea anemia, fat, plinth, breast, barium, all possible cancer predisposition genes. It was negative. The panel included also del dupe. So we are confident that with this negative result, the reason for possible predisposition to cancer is yet to be determined. The only variant identified on that panel was a heterozygous pathogenic variant in GNA gene, which, again, is quite an inconclusive result in the setting of this patient, because GNA gene encodes enzyme glucosamine, which is known as a rate-limiting enzyme in sialic acid biosynthetic pathway. It's one of the founder variants in Southeast Asian population. So it was not surprising, actually, to detect it in this particular patient. However, autosomal dominant associated condition with pathogenic variants in this gene is autosomal dominant C. allurea, which absolutely didn't match with this patient phenotype. And recessive condition associated with nonagamyopathy only puts her in the category of carriers with possible indication for a relative who would be interested in testing, to be tested for that. As for her cancer predisposition, as I mentioned before, unfortunately, we did not identify the cause. And it's still to be determined. So, our second case is also extremely interesting. This is a 28-year-old Hispanic female. She has a history of breast cancer involving multiple sites at a very young age, 19. So, her breast cancer is ER negative, PR negative, HER2 positive, and has metastasized in one of the lymph nodes. P53 was negative. For that, she received bilateral mastectomy and chemotherapy, reception, and radiation. And then the patient presented with pancytopenia at age 28. A peripheral smear review here can show pancytopenia with these rare circulating blasts. They have high NC ratio, open chromatin, nucleolus indicate their immature status. They also have cytoplasmic vacuum. A bone marrow aspirate showing sheets of these blasts with very minimal normal hematopoiesis, except there are a few erythroid precursors. The bone marrow co-biopsy is also extremely hypercellular for this patient and show sheets of blasts with minimal erythroid poiesis, but no other lineage. So, these blasts we did flow cytometry immunophenotyping and showed their lymphoid origin. So, this was diagnosed as B-lymphoblastic leukemia. We did the PCR and shows PCR Able 1 fusion was negative. So, this is a patient with BARL, but the question is whether this is a de novo BARL or it's a therapy related, given her history of treatment. Okay, I'm taking this page again. So, you can see results of karyotype analysis, which is very consistent with the hyperdiploidy. So, this is very typical hyperdiploid karyotype with losses of a lot of chromosomes. And in addition to this hyperdiploid clone, we also identified a clone arising as a doubling event of the original clone, which is very typical for hyperdiploid BLL. It was not surprising. We still ran micro-analysis just to confirm that this is clear-cut hyperdiploid karyotype. For this case, what's notable is also loss of one of chromosome 17, which is quite consistent with BLL with hyperdiploidy, which according to all Cog and big study publications is highly associated with loss of TP53. What's very notable is almost half of hyperdiploid BLLs with detected variant in TP53 gene have it in the germline condition associated with Li-Traumini. And if you look at this patient's personal medical history, what was also striking, a breast cancer at the age of 19, way younger than 53, with multiple sites, then hyperdiploid BLL, and I'll show results of sequencing on the next slide. And unfortunately, her history of fight with cancer did not end, so she was diagnosed soon with pancreatic adenocarcinoma, and now she's undergoing treatment for that cancer. So clearly, this personal history was significant indication to send patient for cancer genetic counseling. You can see this is her family pedigree. So most of relatives are back in Mexico, and it is unclear if there is a history of cancer or not, but you can see what is very notable is adrenocortical carcinoma. This is clearly a red flag for Li-Traumini, as well as breast cancer before age of 53, as well as hyperdiploid BLL. And her mom back in Mexico died of breast cancer at the age of 24. We were not surprised to see a Virginia TP53 variant in her bone marrow biopsy. It was altogether significant indication for skin biopsy to be sent for a targeted analysis. The level of confidence was high enough just to do a targeted analysis for TP53 variant, which was done at the University of Chicago. So we do not run normal tumor in house lab. So that's why there is, as you can see, there is a pipeline with additional germline testing performed at different vendors. And the testing confirmed that this variant is in skin, is present in heterozygous state. So clearly it's a germline, and patient has a diagnosis of Li-Traumini. For patients with Li-Traumini syndrome, due to their extremely high risk of developing all kinds of cancer, they are recommended to have strict surveillance based on NCCM guideline. So basically they should have frequent physical exam, blood test, almost annual imaging study, like MRI for brain, whole body, breast, and also neurologic and the skin exams. They also should have frequent colonoscopy. So surveillance absolutely makes a difference. As you can see, the long-time survival dramatically increased for those patients undergoing surveillance. Switch gear, I want to quickly mention the update for the upcoming WHO classification for therapy-related neoplasm. So the new, the fifth edition of WHO classification already has been published and will be widely used by pathologists next year. So the criteria for therapy-related myeloid neoplasm basically remain the same. So first of all, we need to make a diagnosis of myeloid neoplasm, including MDS, MDS-MPN, and AML. And also the patient should have a history of cytotoxic therapy, including alkylating agent, topoisomerase-2 inhibitor, and also radiation covering lung bone containing active bone marrow. But the new edition expanded this therapy to cover more medications, including like anti-metabolites and anti-tubulin agents, and also POP1 inhibitor. The same, the new international consensus classification also provide the guidance for diagnosing therapy-related hematopoietic neoplasm. Very similar to WHO, we need to fulfill the diagnostic criteria first, and then also notice the patient has history of therapy, including chemotherapy, radiation, and immune intervention. The ICC guideline actually recognize CHIP, SICUS, and also lymphoblastic leukemia and lymphoma can occur after therapy. So they mentioned these should be noted as additional comment associated with this diagnosis. So we do recognize a lot of our patients with therapy-related myeloid neoplasm may have some inherited genetic predisposition, making them more susceptible to cytotoxic treatment. How do we recognize these patients and provide appropriate clinical management? For all our patients, we do complete diagnostic workup, including the morphologic examination of these blood and bone marrow samples, immunophenotyping, molecular and cytogenetic analysis. We also pay attention to patients' personal and family medical history, history of treatment and exposure. So if there's anything suspicious about the patients, for example, like a young onset of cancer, multiple cancer in individuals, or multiple cancers in immediate families, as well as if we identify any mutation on somatic NGS panel, which could be germline, these will trigger additional genetic counseling consultation. We are lucky at Northwestern, we have a group of adult oncology genetic counselor, even with specialization in different subspecialty, we have a hemo-oncology genetic counselor. So they see patients. So these consult may be time-sensitive, some can occur as inpatient, but sometimes the patient may be too sick during initial hospitalization, or maybe there's some payment insurance issue, they will see the patient at outpatient service. The time is sensitive, particularly for those who will have planned for bone marrow transplant. Usually genetic counselors see patient and the skin biopsy were obtained. We use different vendors, including GeneDx, Invitae, and the University of Chicago Genetics Lab. When the results come back, the genetic counselor will see patient again to explain the results. Next, I will hand over to Dr. Abada to discuss the clinical implications. Thank you, Dr. Gao. So, as I alluded to in the beginning of my talk, the exact pathogenesis of the therapeutic myeloid neoplasms are largely unknown. But there are four main drivers that are thought to play a role in the development of these neoplasms. The first is inherited predisposition. With patients that have inherited mutation in DNA repair, genes have an increased susceptibility to the development of multiple independent malignancies that may not even be related to previous therapy. In addition, prior chemotherapy and radiation therapy in these patients may induce the development of mutations due to the impaired DNA pathway in these patients. The second potential pathway is the direct induction of fusion oncogenes, as I previously mentioned, especially with the toborazomase II inhibitors. The third potential pathway is the clonal selection of cells with pre-existing somatic mutations and clonal monopoiesis by the prior chemotherapy and radiation therapy. Because the thought process is that these prior therapies give these chip clones a competitive advantage over normal hematopoietic stem cells. And later on, these clones acquire additional somatic alterations and then develop into therapy-related myeloid neoplasms. In addition, chemotherapy and radiation therapy also induces damage of the bone marrow stromal niche, and that also promotes clonal selection of these leukemia-initiating cells. So, when it comes to treatment, there are numerous factors that do play a role in how and complicate the treatment for these patients, including that prior therapy does result in organ dysfunction, such as cardiomyopathy in patients with prior endocyte exposure. Prior therapies also deplete normal hematopoietic stem cells and limit the hematopoietic reserve, which therefore impairs cat recovery in these patients after treatment. These patients also have multiple comorbidities, poor performance status, and refractoriness to transfusion support, which further complicates their care and our choice of treatment. In general, the way we treat these patients is largely dependent upon their cytogenetic and molecular profile, similar to how we treat de novo AML. So, this is an overall algorithm, just to simplify how we typically look at these patients and treat them. So, in general, we always prefer clinical trials, if available. When they're not available, then the treatment option depends upon the patient's eligibility for intensive chemotherapy. And again, we treat them in a very similar manner to those with de novo AML. So, those who are not eligible for intensive chemotherapy are typically treated with hypomethylating agents, plus the BCL-2 inhibitor reninoclax, which is our current center of care for the frontline treatment of AML unfit for intensive chemotherapy. We can also use FLT3 and IDH inhibitors in those who have respective mutations. And those who are eligible for intensive chemotherapy, again, the options and the way we treat is similar to how we look at patients with de novo AML. So, those with therapy-related APL would be treated to our arsenic-based regimens. Patients with core binding factors are typically treated with intensive chemotherapy, plus the addition of gentuzumab or gozumicin. Those with NPM1, FLT3, IDD-negative patients are also treated with intensive chemotherapy with or without the addition of phenetoclax. And in those two categories of AML, the decision of transplanting and first remission is guided by the MRD status using molecular MRD assessments. For the remaining subtypes of therapy-related AML, we always aim for transplant first remission and those eligible for intensive chemotherapy. And for therapy-related MDS, we typically treat with an HMA plus or minus phenetoclax approach and proceed with transplant and those eligible for transplant. Although the therapy-related myeloid neoplasms generally do worse than their de novo counterparts, the two subtypes of AML that particularly do worse and have very poor outcomes and need and are areas of unmet clinical need are those with the TP53 mutation and KMT2A rearrangement. And I'll talk a little bit more about the progress made in those subtypes of AML. So in AML, the TP53 mutations are known to have poor response to all our current standard of care treatment options, including both conventional chemotherapy and CPX351, which is shown in the figure on the left. CPX351 is a dual-drug liposome encapsulation of donorubicin and citeribine, which is FDA-approved in the treatment of patients with therapy-related myeloid neoplasms. And in this study, the patients who had a TP53 mutation, which is represented by the solid lines, did equally bad with both forms of chemotherapy with a median survival of about five months in both arms. Treatment was also, or outcome, sorry, was not improved in those treated with hypomethylating ages plus venetoclaxis, which is our current standard of care for unfit patients, with the TP53 mutated patients having an overall survival of around six months, which is significantly worse than those who have our wild type for TP53, which is represented in the upper right figure. Overall survival did not improve with using a different formulation of HMA, a base approach like citeribine for 10 days with or without the addition of venetoclax, with the median overall survival, again, ranging around six months and not improved in that, with that treatment. And all of this underscores the poor outcomes of these patients and the need for novel agents. There's also an interest to assess the impact of the burden of mutant TP53 and how that impacts our response to frontline treatment. And our colleagues at MD Anderson did conduct a retrospective analysis of 202 patients with newly diagnosed TP53 mutant AML, and they analyzed their outcomes based on the allelic burden of TP53 and the choice of frontline therapy. And what was found in this study, which was really interesting, was that patients with very allelic frequencies of 40% or less had significant improvement in their response rates when treated with citeribine-based regimens, while those with variant allelic frequencies above 40% had higher response rates with HMA-based approaches. In subgroup analysis, there was a significant improvement in overall survival among patients with VAFs of less than or equal to 40%, which is demonstrated in figure A. And the overall survival was significantly better in those who received intensive citeribine-based regimens, which is the blue line. However, overall survival was equally poor for all patients with VAFs above 40%, which is shown in figure B, regardless of the type of treatment used. And therefore, based on these results, we only use intensive chemotherapy in patients who have lower TP53 mutational burden and overall patients who have high burden of disease do equally poor with all standard of care treatment options, and we always try to put them on clinical trials. So identifying drugs that can restore the wild type function of TP53 has been an area of ongoing interest, and APR246 is one of these drugs. It's a prodrug, which when converted to its active metabolite, which is MQ, it covalently blocks the cysteine residues in the TP53 mutant protein, resulting in conformational change of the protein to its wild type conformation, leading to, or to actually restore the protein function. And in early phase trials, APR246 plus azacitidine did show preliminary efficacy in TP53 mutant myeloid neoplasms, which are the two figures on the right. However, a randomized phase 3 trial comparing azacitidine to the combination regimen in TP53 mutant MDS patients failed to meet its primary endpoint, which was a huge disappointment for this promising agent. In an attempt to investigate the role of immunosuppressive therapy in these patients, Solomon and colleagues at Moffitt conducted in vitro studies to compare patient samples with TP53 mutant versus wild type AML and MDS, which showed that TP53 mutant patient has significantly higher PD-L1 expression in hematopoietic stem cells and a higher proportion of immunosuppressive Tregs compared to wild type patients. This was also associated with upregulation in the TP53 mutant patients, which in turn induces PD-L1 expression promoting immunovasion. And collectively, these results suggested that TP53 mutant MDS and AML have an immunosuppressive phenotype and therefore may benefit from immune-based treatments. CD47 monoclonal antibodies are among the most promising immunotherapies currently in clinical trials for myeloid neoplasms. CD47 is a macrophage immune checkpoint, highly expressed in AML and high-risk MDS. When binds to its receptor surf alpha on the surface of macrophages, it actually provides a do-not-eat-me signal inhibiting macrophage-mediated tumor phagocytosis. Migrolumab is a first-in-class CD47 antibody, which on its own had limited single-agent activity. However, in preclinical studies, azacitidin was found to upregulate both expression of CD47 and prophagocytic signals such as calverticulin in AML and MDS cell lines. And in vitro, the combination synergistically increased the macrophage-mediated phagocytosis of AML cell lines, which translated into improvement in the survival of AML xenograft mouse models. And these promising preclinical results were the basis for a Phase 1b study assessing the efficacy of migrolumab plus azacitidin as frontline treatment in patients who have high-risk MDS and newly diagnosed TP53 mutant AML unfit for intensive chemotherapy. And this is a summary slide of the outcomes of these, the most up-to-date outcomes of this phase 1b study. And among patients with high-risk MDS, which is the upper two figures, median overall survival was not reached for all of the patients and was 16.3 months among patients who had a TB53 mutation. And this actually compares favorably to historic data, where the overall survival of TB53 mutant high-risk MDS is about 6 to 12 months. Among the TB53 mutant AML patients, median overall survival was 10.8 months. Again, this compares favorably to AZFN, which, as I previously mentioned, overall survival in TB53 mutant patients is around 5 to 7 months. Patients who were bridged to transplant, which is the lower right figure, did significantly better with the median overall survival not reached compared to those who did not go to transplant, underscoring the importance of transplant in these patients. And there's currently, McRollin is currently in phase 3 randomized trials in both AML and high-risk MDS, and the results of these trials have not yet been reported. So KMT2A-rearranged acute leukemias is the other subtype of therapeutic-related neoplasms that have very adverse outcomes. These rearrangements occur in about 5 to 10 percent of AML patients and 10 to 15 percent of AML patients and are associated with very poor overall survival, hence the need for targeted therapies as well. And what has been found was that the KMT2A fusion proteins bind to menin, and that binding has been found to promote leukemogenesis through the induction of a variant expression of Hox genes, and as a result, menin inhibitors have been identified as novel targeted therapies, which are currently in clinical trials for both KMT2A-rearranged AML and ALL, and whether these agents will eventually improve the outcomes of these high-risk subtypes of acute leukemias is yet to be determined. So in conclusion, the pathogenesis of therapeutic-related myeloid neoplasms remains largely unclear. The treatment of these patients is based on their cytogenetic and molecular profile, similar to de novo AML. We always try to treat these patients on clinical trials, especially those who have high-risk disease such as the TB53 and KMT2A-rearranged diseases. There are many unanswered questions that still need to be addressed in this field, including the role of CHIP mutations in the development of these myeloid, in these therapeutic-related myeloid neoplasms, and whether we can identify patients who are at risk for the development of these therapeutic-related myeloid neoplasms prior to treatment initiation of the primary cancer, and whether or not this will allow us to predict and potentially prevent the development of these fatal neoplasms. Thank you very much for attention and time, and me and my colleagues are happy to answer questions. Yeah, that was a fantastic presentation. Thank you so much. This is a reminder for the participants who are recording the webinar. It will be available in two weeks, and please remember to fill out your evaluation. Yeah, I think that was really an awesome presentation. I definitely learned a lot, even after working a lot with different team samples. We do have several questions in the Q&A, and if you do have a question for those who are participating, feel free to enter it or look at the questions in the Q&A, and I'll vote certain ones that you are interested in. So, the first question is, if a germline variant is identified before administration of cytotoxic therapy, would it change the choice of therapy or management, and do you think this should be taken into account in different clinical management guidelines for these patients? So, if there's a germline variant, it typically doesn't impact whether or not we go with what treatment, but it does impact transplant, especially if they're from related donors. So, that's where the problem really occurs, is that if we do have a germline variant, whether it's TB53 or other variants, it may not impact how we treat the patient himself in terms of the chemotherapy, but it definitely impacts transplant, especially if we're looking for related donors for that patient, and that's where we do hasten the genetic counseling and try to get results fast, especially if we want to proceed with transplant soon. Major issues with that is that we cannot do genetic testing in the hospital. A lot of these leukemias are admitted for about four weeks, so insurance doesn't pay for it, so we have to wait until they're discharged, and it takes about, even if we get genetic counselors to see them very quickly, which they typically do, it still takes about eight weeks to get results back. So, that is the only hindrance or issue that we face when related donors are being considered for transplant. Thank you so much for that. Another uploaded question has to deal with pipelines and workflows. It says, which pipeline or workflow did you use for the detection of variants? Could you explain the steps? It sounds like most of your testing is send out, but if you have any insight on that, feel free to answer. So, as we previously mentioned, we send a lot of the sample to different vendors, like in VT and GDX, and some to University of Chicago Genetics Lab. So, these are usually based on genetic counselor's choice, because if they see a certain risk for, based on patient's history, family history, if they see their certain gene are more risky than others, they choose that particular panel. And we usually defer to them to analyze these results and come back. So, I'm not sure exactly which analysis pipeline they used in different labs. I don't know if Medina has any insight for that. So, this is typically CMG standard pipeline, but as you can see from our presentation, our molecular pathology division performs only testing of tumor specimens. We do not do germline assessment, and regretfully, we do not make an assessment and correlation of tumor normal in parallel. There is many reasons for that, including not only logistics, but also reimbursement, and it's very difficult to compete with big companies like GeneDx and VT for reimbursement. So, that's why our pipeline, our workflow typically is, as Dr. Gao mentioned, is careful assessment, personal medical history, all associated and relevant phenotypic features. In parallel, we do in-house somatic NGS testing. If some variants identified based on allele frequency and also known literature have association with possible cancer predisposition, we flag it, and on the report, we have a disclaimer that although we do not test normal tissue and cannot confirm whether this variant is germline, but it's highly suspicious genetic counseling and germline assessment. So, in case of blood cancer, clearly it has to be skin biopsy is recommended, and then genetic counselors take from there immediately. And based on the history, that's why we selected these two illustrative cases. As you could see, in first case, it was not very clear what to look for. In second case, everything was screaming, leave Romania, leave Romania, leave Romania. As you can see, in first case, it was extended, very big panel from GeneDx, and in the second case, it was targeted test for this particular variant. Great, thank you. So, there's another question about the first case that was presented. Is there any value in pursuing whole exome sequencing or whole genome sequencing, something more comprehensive after non-informative panel results for hereditary cancer genes? Yeah, I think there's always value, because the first case just screamed there is an inherited genetic component for this patient, but with our current method, we could not identify anything. So, I think there's always a value in pursuing a more extensive panel. We didn't, in this case, we didn't, in this case, I think, for various reasons, maybe the family decline or, you know, there is a, the patient was also, you know, has more important medical condition to battle. So, we didn't do in this case, but I think it is always, there's always value to pursue that. Great, thank you for that. I have another question about, that I think is really interesting, about therapy-related neoplasms. So, is there any known explanation to why therapy-related MDS and AML are really aggressive and may not respond to treatment, even those with normal TP53 results? So, we don't, it's still unclear, but one of the pathways, as I mentioned in one of the slides, was that these, there's close selection of pre-existing CHIP clones by the chemotherapy and radiation therapy given to previous, to the primary malignancy. And these clones survive these kind of chemo and radiation. So, by, they become inherently chemo-resistant. So, since they're selected and they're the ones that have an advantage, survival advantage over normal stem cells. So, eventually those are the ones that populate the bone marrow at a higher frequency and all they need is an additional mutation to develop into therapy-related myeloid neoplasms. And since they survived initial chemotherapy, they do become inherently chemo-resistant. That's one of the thought processes. And that is all supported that some, there's damage to the bone marrow stromal niche that also supports their selection of these leukemia-initiating cells. So, that's the thought process behind why they're so chemo-resistant compared to the de novo counterparts. But again, this is an area of active research. All right. Yeah. Thank you for that. I think it's really important to understand like the selection of the different clones and the tumor type and what makes certain clones, you know, survive the harsh treatments that these patients have to go through. So, another question that we have is, is there a role for lymphofractionated peripheral blood to screen for germline changes in myeloid malignancies rather than fibroblast-derived DNA? Is it faster or cheaper? Is there any advantage to that? No, we did discuss this internally. There is no advantage to that because it's getting difficult to determine whether the mutations from the leukemia cells that are circulating versus just germline. So, in cancer patients or leukemia patients, using blood is difficult. So, that's why we refer to skin fibroblasts. Another really cool question. Thank you for all the great questions in the Q&A. Do you typically send telomere lengths or chromosomal breakage analysis for Fanconi anemia diagnosis? They say it's usually hard to send it at the diagnosis of acute leukemia due to pancytopenia. So, yes, I do send all of my younger individuals, so patients who are young and present with either MDS or AML, regardless if they're therapeutic or not, I do send for telomerase lengths and to Fanconi analysis. And I do do it at diagnosis. It's a send-off. It takes a while to get back. So, I don't wait for these results typically to especially patients who need immediate therapy. And it usually doesn't impact our initial treatment. But again, it is something that we take in consideration, especially when we go with transplanting these patients and on choice of conditioning regimens and that kind of stuff. But yes, we do send it in young patients. Great. Thank you. And then another kind of technical question. What does your center do if skin fibroblasts don't grow well for adequate DNA extraction? Is there a reflex that you'd like to do instead? Very good question. So, as I mentioned before, we do not perform germline assessment. And for that reason, we didn't encounter that particular problem. However, if the decision for germline assessment will be ever done and we will bring it in-house, so we have also a cytogenetics facility with extensive experience of growing skin fibroblast cultures. And in that situation with unsuccessful skin fibroblast cultures, probably the best way of assessing for possible germline origin would be, what, looking at follow-up samples, right, and carefully correlating. If patient is in a clinical remission and still that variant is persistent, that would be a good suspicion, right, for possible germline origin. If it's gone, then it would be very suggestive that it's somatic. But again, we don't worry about failure of skin fibroblast culture yet because we're not performing that. Great. Thank you for that. We do have a participant that was interested in kind of the classification with kind of aligning with the WHO fifth edition for two of the cases. So for the first case, would you say that that is consistent with myodysplastic neoplasm with biallelic T53 activation per that WHO? Yeah. So the fifth edition, similar to previous edition, any history of therapy-related should exclude the diagnosis of other de novo MDS. So with that history, the patient fell into the category of therapy-related MDS. However, we do notice, you know, the MDS with biallelic T53 activation also has a very poor prognosis. So the prognosis for the patient, which category belong to, is probably very similar. But based on WHO classification, given that history, the patient should be classified as therapy-related MDS. Got it. And I think that kind of answers the second question as well from individual in the chat about the second case is low hypodiploid, biallel, subtype Harbers, T53 alterations in over 90% of the cases, approximately half of which are somatic and half are germline. I'm not sure. So I think they were just trying to make sure they're understanding the classification there too. But if I'm hearing, if I'm understanding it correctly, the same argument goes with this one as with the previous question that if the patient was treated with therapy previously, then this would be classified as a therapy-related MN. Is that correct? Yes. Yeah. So for lymphoid, it's a little bit different because the lymphoid cases are so rare. It's not well-established as myeloid. So WHO did not specifically create a category for therapy-related lymphoid because there's no, it's not well-established as myeloid, but the new ICC guideline, they do bring up, if it's lymphoid, we should also consider, it could be therapy-related. I think a lot of these are still developing. Right. And I have a kind of interesting question about risk stratification and guidelines right now, because currently cytogenetics kind of sets the framework for risk stratification. Do you predict that those stratification constructs will change and become more molecular focused, or do you feel like cytogenetics is really the best risk stratification system there is? I think they both complement each other. So I don't think one will exclude the other. So because it's important, like for instance, and this is a simple example. So let's say if you have a core binding factor, if you have a FLT3, you still don't target the FLT3 and a core binding factor AML at initial presentation. We still go with a geo-based approach, whether we typically do either 7 plus 3 geo-based on the alpha trial or flag geo-based on MD Anderson data. And we typically don't target the FLT3, even if it's an ITT mutation, given the very good outcomes in these patients with a geo-based regimen. So I think that's a very simple example where you have both cytogenic NGS and cytogenics in this case will trump the NGS finding. So they're both important, and I don't think one will trump the other. It's just we need to have both information at hand. Great. Thank you for that. So with that, I see we're at the top of the hour. Would you mind sharing the last slide of that slide deck again? I just have a quick announcement from ACMT. So the ACMT annual clinical genetics meeting is coming up in March of next year. And starting on October 4th, there is a call for abstracts, and the registration will open at that time. And the abstract submission deadline is on November 18th. So great. Thank you so much. So here are the essential dates for those interested in submitting abstracts. And with that, I'd like to thank our speakers once again for the fantastic presentation and for sharing your time and expertise with us today. I really enjoyed it and learned a lot. So thank you so much. Thank you. Thank you.

therapy-related neoplasms

diagnosis

clinical perspective

Northwestern University

cancer variant interpretation

chemo-resistance

targeted therapies

germline variants

genetic counseling

personalized treatment approaches