 First of all, I would like to thank the organizers to give me this opportunity to present this analysis on behalf of the working group. A few background on adrenal cancer. So adrenocortico-casinoma is a rare cancer type within any instance of about 0.5 to 2 per million. And the overall 5-year survival with mesothetic disease is less than 20 percent, which means this malignance is pretty bad. There is no standard staging system for this disease, and mitotin has been the only drug that was approved by FDA. And the targeted therapy against this disease is quite a disappointing, mainly against the IGF-2 gene. However, this targeting of IGF-2 is only effective in 5 percent of the ACC cases. Now it's very important to mention that the adrenal cancer is the endocrine tumor, which means in the adrenal cancer patient about half of the cohort actually have this hormone access which differentiates this disease with other cancer types. Now TCGA initiated the adrenal cortico-casinoma project last year. The goal is to collect 92 cases with axome sequencing, mRNA and microRNA sequencing, DNA copper number, methylation. All of these data types, the data are almost complete, but we are expecting our protein arrays for 68 cases, and we are expecting a complete set of clinical data at this point. The TCGA working group for adrenal cancer formed last year, we have so far, we have finished the first-pass analysis, including subtype analysis on different levels of data, and we have finished the axome sequencing analysis, and we have run first-pass of the structural variations, and we have some integrated analysis that has been done. Adrenal cortico-casinoma compared to other cancer types is quite a pure cancer type. You can easily appreciate here that each dot in this figure represents one sample, and even though there are a few cases that have relatively lower purity, but overall the medium purity of adrenal cortico-casinoma is very high, suggesting this tumor type is actually a very good prototype to study tumor biology. We have a method called estimate, by which we can look at the gene expression pathway activity. We look at the email pathway activity using this method, and you can see here that adrenal cancer has the lowest email score, which probably is related to the hormone cortisol, because cortisol in adrenal cancer actually created a very hostile environment for the T cell. So looking at the genomic level, the adrenal cancer has a relatively lower mutation frequency compared to other cancer types. It is lower than ovarian cancer, lower than gluoblastoma, but it is higher than another endocrine tumor, thyroid, here. We used 91 axome sequencing. We identified six significantly mutated genes, including PP53, beta-cutinin, many one PRKAR1A, and a ribosomal protein. If you look at this mutation spectrum, mutation clustering pattern of those five significantly mutated genes, you can easily see that all of the mutations in beta-cutinin that occur in a relatively short region of this gene, which the pattern itself suggests all of this mutation could be activating mutations. In the other genes, the mutations basically distribute across the gene body, which may suggest there are inactivating mutations. I want to emphasize one gene called PRKAR1A. The story is that this protein is a regulatory subunit of the protein kinase A. This protein interacts with the catalytic domain encoded by the gene PRKACA. This gene, actually in the last two months, there are several papers that report the recurrent activating mutation in the catalytic domain of this protein. If you think about this, this could be very interesting because all of the protein kinase A catalytic domain mutation that occur, they were found in the adrenal adenoma. However, we found the PRKAR1A, the regulatory subunit mutation in the Casinoma. You would imagine this result not only highlights the importance of this protein kinase A pathway in the adrenal tumor, but also it would be interesting to speculate how the mutations in the regulatory subunit and the catalytic subunit can lead to different outcomes in the adrenal gland. Copy number analysis found focal applications and delusions. We found genes including Tert, CDK4, Tert2, and CCA2, and E1. We found a link RNA on chromosome 4. We found P16 deletions, RB1, and then RF3 gene. The Tert2 and ZNR3 have not been reported in adrenal cancer. From IGV, you can easily appreciate this focal amplification of Tert2 and focal deletion of ZNRF3. ZNRF3, this gene is a wound path for regular through gene, so I'll come back to this. With these genes, we come up with a landscape of genomic landscape of adrenocortical Casinoma. On the first panel, I show the mutation frequency, and on the left side of the hit map, you see the fraction of mutations of those genes in our cohort, and on the right side, you see the amplifications and deletions of those genes. You can see that the majority of the ZNRF3 alterations are due to the deletion. For P3, most of the alterations is mutation. Tert is amplification, and beta-catenin is mutation. However, I think a very interesting observation is that this genomic landscape clearly separates the adrenal dataset into two cohorts. One cohorts with at least one putative driver gene. With another cohort, about 30 percent of the cohort does not have single one putative driver gene alteration. This results actually confirmed the findings by our European collaborators. They published a new paper in Nature Genetics, and in their cohort, they also observed a subset of adrenal cortical Casinoma cases without a putative driver mutations. With unsupervised methylation clustering, we found three groups with a lower-level, intermediate-level, and high-level methylation. We call them normal-like SIMP and hyper-group. MicroRNA clustering found six robust subgroups, and gene expression identified two groups. Our two groups actually recapitulated the findings by the European group, which they call C1A and C1B. The C1A group has a bad outcome, and the C1B has a slightly bad outcome. So by overlaying these subtypes with the genomic landscape, we found a remarkable enrichment of these different subtypes with this group. For example, the group without putative driver mutations, they are mostly C1B group, and they have a normal-like methylation pattern. We also looked at structural variations of gene fusions in this content. We did not find a recurrent fusion transcript, however, we found recurrent pathogenes, and we found sporadic cancer genes involving in this fusion. For example, we found an m-toll gene fusion, so each box here represents one axon. You can see a clear activation of this axon expression beyond this fusion size. Very interestingly, this m-toll fusion actually retains the catalytic domain of the m-toll protein. This could be very important because we know we have m-toll inhibitor. There is another case of the fusion that involves a gene called BRE. The BRE gene has been reported in adrenal tumor genesis, and it was implicated as a driver gene, so this finding suggests this fusion might be a functional relevant fusion events in adrenal cortical carcinoma. At the pathway level, we found the one pathway to be the most frequently altered pathway with more than 45% of the patients in our cohort have at least one gene altered in the wound pathway, but most of the alterations were contributed by ZNRF3 homozygous dilution and they better continue activating mutation. With this, I would like to summarize what we have for the adrenal cortical carcinoma. We found potential new driver genes, including the NNR3, TURF and TURF, PRKA, R1A. We have subtype analysis from MRI, methylation, copper number and micro-A. Integrated analysis highlighted 30% of the ACC without any apparent driver alteration. So we're expecting the RPPA and a complete clinical data where this data might give us some insights into this subset. We have infrequent alterations such as sporadic gene fusions that may contribute to the adrenal tumor genesis and wound pathways, apparently, most out of the pathway, which we hope can bring some insights into the therapeutic development. So I would like to thank many people, but especially my chair, Dr. Tom Giordano and Dr. Robert Huck, due to the special limit, I cannot list all the names, but many people contributed to this project. Thank you. Okay, so let us have some out of the process. Before you go on, I'm really curious about this driver negative subset. As I understand it, these are also tumors with very high purity, as understood by copy number analysis. I just want to ask, what are your thoughts as to how to pursue the driver negative subset and specifically in terms of things like fusion analysis or whole genome sequencing analysis? So right now we are not going to have whole genome. Now we have two things we're thinking to approach this subset. One is we're going to look at this driver genes, and that means the known driver genes. We think they're probably just infrequent and due to our sample size that we cannot find recurrent alterations in those genes. So we're going to look at that and potentially that can have some percentage of cases with a driver. Now another thing is we should, we will look at the germline data, and now we want to see if we can find by 10 to 15 percent cases with a germline mutation. So can I just follow up on that, Matthew? So what doesn't come through in this is that those tumors are what we call low-grade adrenal cancers. Most of them do have IGF alterations though, so they have sky-high IGF2 expression, but that is the fascinating part of this project, and so we really want to focus in on those tumors. So I just want to make that comment that it's really recapitulating the classification that we do histopathologically. Thank you, Sal. So I think the easiest way to talk about this, there's no mutation, no driver mutation. You should say there's some driver pathway involved. That may be easier to understand, because if there's no mutation, but there's something driving it. Yeah, absolutely. Yeah, yeah, we hope we will have this data. Thank you. No question. So let us thank again all six fantastic, fantastic talk this morning. Yeah. So, yeah, we are going to resume the afternoon session at one o'clock. Yeah, that's a poster session at one o'clock. Yeah, okay.