Our next speaker is Dr. Young. The title of his talk is, RNA Deconvolution Reveals Distinctive Immunogenic Landscapes in NF2 and Non-NF2 Immunogens. Good afternoon, everyone. My name is Jackie Young. I'm a fifth-year resident currently at Yale University. Thank you for the chance for me to talk about my work in using RNA information to investigate the immunological landscape in meningiomas based on their mutational subtypes. I'm going to start off with a case of a 67-year-old female with multiple medical comorbidities. We have a homogeneously enhancing meningioma that's abutting deep cerebral veins. So even though meningiomas are usually very straightforward, together with a patient picture and also some of the locations of these tumors, selected cases can be very problematic. As I said, most low-grade meningiomas can be resected, and especially with good Simpson-grade resection, the recurrence is extremely low. However, there are cases that tend to recur despite radiation therapy. There are no effective medical adjuvant therapies to prevent these recurrences or to even cure them. At Yale, we have put in tremendous effort in identifying key mutational drivers that promote the tumor genesis of the meningiomas. But as a community, we're only scratching the surface of effective medical therapies that can actually help as an adjuvant after maximal resection. Immunotherapy has gained traction, as of late, in many solid tumors. However, successful immunotherapy depends on whether there's any immune infiltration at all, and if so, what are the immune cells that comprise the infiltration. Going back to 2013, our group at Yale, we published in Science that in the NF2 subgroup, there's a high enrichment of antigen-processing genes. What this means is that it suggests a high presence of myeloid cells and antigen-presenting cells. In the same year, a European group used RNA studies and also flow cytometry to demonstrate that, indeed, tumors with chromosome 22 aberration have higher macrophage content. But what do these macrophages do, and what are the other immune cells that are present in the tumor? To pursue this question, at Yale, every excise tumor undergoes whole exome sequencing and gene expression analyses. I utilized a cohort of 188 meningiomas that have been sequenced and have their mutational profile identified and looked at a large collection of immune-related genes. Comparing NF2 with non-NF2 tumors, we see that, indeed, NF2 tumors have higher expression of genes related to macrophages, such as B7H2, which is also called PD-L2, CD68, a macrophage marker, macrophage scavenger receptor 1. Some B cell markers, such as APRIL, BAF, CD19. Interestingly, in non-NF2 tumors, we see that a gene called tryptase alpha beta 1 is higher in the non-NF2 subgroup. This gene is specifically expressed on mast cells. Now, the audience might ask, these are individual genes. What does that mean in terms of the entire immune cell composition? I utilized an algorithm called Cybersort. It is an RNA deconvolution algorithm initially developed at Stanford that looks at total RNA information and tried to investigate the immunological landscape in tumors. I want the audience to imagine using alphabet cereal to spell out your name, throw it into a bowl of milk, and try to tease out your names from that mixture. It's almost impossible, but this is what this algorithm does. As I said, I looked at a cohort of 188 tumors compared to 22 meningio biopsies as controls. At first glance, looking on the figure on the left, we see that the most dominant population of immune cells are monocytes followed by macrophages. When we compare this to normal meninges, we see that they're very similar. Looking back into the literature, it's been well documented that meningio layers actually harbors regular macrophages and mast cells as part of normal immune surveillance. This suggests that meningiomas may echo the immune composition of their embryological origin. We focus now more on myeloid cells. We see that there are a lot of monocytes and the most important finding that I'm going to talk about during this talk is the abundance of M2 macrophages. This is the first time that anyone has described the phenotype of macrophages in meningiomas. M2 macrophages are known to be immunosuppressive, provide a possible mechanism for the very common PD-L1 therapy, provide resistance to chemotherapy and radiation therapy. These, by a large margin, outweigh the M1 macrophages, which is what we learn in textbook, the inflammatory type. When we break these down by mutational subtypes, comparing them to NF2 groups, we see that the hedgehog group, the PI3K group, they tend to have lower monocyte composition. M2 macrophages, there tend to be lower composition in KLF4 and also PolR2A groups. Next, we move on to the lymphoid compartment, which comprise the adaptive arm of cellular immunity. Traditionally, in tumor immunology, we learn that type 1 immune response, where there are more CD8 cells relative to CD4 cells, is desirable. But in this case, we see that there is a lot more CD4 memory resting T cells relative to CD8 cells, suggesting that this is more of a type 2 immune response, something that we don't want. Focusing on CD8 T cells, we find that the hedgehog group, relative to the NF2 group, have cases that are highly enriched in CD8 signature. And these tend to be midline in the olfactory groove. In summary, we've used the Yale database as a robust platform to interrogate the immunological landscape in silico. And for the tumor surgeons in the audience, if there are two take-home points that I can convey, it's one, depending on the mutational subtype, the immune cell composition is different. And most importantly, the M2 macrophages, the bad macrophages, are the most dominant leukocyte population in human meningiomas. I would like to thank my mentors, Dr. Li-Ping Chen and my chairman, Marat Ganel, for giving me advice and leading on this project, the NREF for their funding, and most importantly, our patients for contributing to these studies. Thank you.

meningiomas

mutational subtypes

immune cell composition

gene expression analyses

therapies