 I think I closed the meeting last year and I was delighted when I saw the agenda come out that I had, although I was still in the last session, that I had moved up higher, but here I am again closing out the meeting. So maybe next time I'll be able to actually be in the middle of the meeting. Thank you all for sticking around. And I'm going to just give a context for both these presentations here, as well as a lot of things you've heard over the last couple of days. So this is the current treatment algorithm that we have for kidney cancer based on clinical trials. And as you can see, it's very full and we have a lot of choices, but there are also our limitations to all of our treatments. What we really want is Bob Piglin talked about our treatments that produce durable responses in the majority of patients, and right now we have treatments such as IL-2 that can produce durable responses in the minority of patients and treatments like VEGF inhibitors that are largely palliative and for the palliative treatments we want to give the treatments that have the best therapeutic index or treatments like MTOR inhibitors that probably help in a major way a small percentage of patients, but we really don't know who they are. So with that in mind, I'm just going to set the stage for key issues that I want to talk about, optimizing VEGF inhibitors, the role of MTOR inhibitors, understanding the biologic basis of VEGF resistance, either overcoming acquired VEGF receptor TKI resistance or treatment of the VEGF receptor TKI refractory patient and optimizing novel immunotherapies, what setting, combination, or biomarkers. So these are, I think, our key needs right now. And with regard to the VEGF inhibitors, I think we've seen in this meeting in Islantz-Cowey talked about that the potency of drugs appear to correlate with their ability to inhibit the VEGF receptor and the toxices relate to off-target effects. And Exitinib and Tivazinib appear to be cleaner drugs with higher therapeutic index than the currently approved first line agents. And unfortunately, failures of study design for these drugs mean that we may never receive, they may never receive first line approval. So one of our challenges is to say, is this the end of the story, or is there any way we can get the better, cleaner VEGF drugs available to our patients in the first line? And so people can think about how we might do that. What is the role of mTOR-targeted therapy? I think it's becoming more confusing. The Intersect trial suggests that Tensorylimus leads to worse overall survival than Seraphanib, even in TKI failures, if we've heard. The Record 3 trial, Sunitinib was better than Everlimus in practically all settings and endpoints, non-clear cell patients, maybe no worse in poor performance status patients. And the best trial that has been presented previously suggests that all arms containing Tensorylimus yielded worse outcomes, worse therapeutic index. So Torrenhibbers appear to be inferior to anti-VEGF agents as anti-angiogenic inhibitors. As angiogenic inhibitors, they are not angiogenic inhibitors. I think we can put that to rest. Kidney cancers, even in the setting of VEGF receptor, TKI resistance are not driven by mTOR. And so the hypothesis, really, that I think Martha's trial should test is that mTOR inhibitors should be limited to patients whose tumors are driven by the mTOR pathway, and that would be a much more restricted trial, potentially patients who have FOSO-S6 or FOSO-AKT staining, elevated LDH or specific mutations in the PI3 kinase pathway. And you could see whether they actually do better than VEGF receptor TKIs in that particular setting and try to validate this interesting data that's come out of Memorial Sloan Kettering, Jim Shea, and Martin Voss, showing that the patients who get long-term benefit from mTOR inhibitors are those that have some sort of mutation in the PI3 kinase, AKT mTOR pathway. So what about VEGF receptor TKI resistance? So as Bob Finglin mentioned, we have these three patterns of response to VEGF receptor TKIs, and we can't really tell up front which response pattern we're going to get, and we should be able to sort that out. We see these patients, we have their tumors, there's got to be something different about those so that we can potentially identify what the pattern is going to be, and I would suggest that if they're going to be someone who gets a long response, then these are patients who should be treated with a single agent with the best therapeutic index, while those that have a short response should be getting some sort of synergistic combo, and those that are absolutely refractory should be getting a new agent. So looking at this group here, synergistic combos, there's a lot of data that sort of looks at mechanisms of resistance to show that this is related to angiogenic escape, either decrease in anti-angiogenic factors, as was published by RuPaul Botte, or increase in pro-angiogenic factors, and here's a list of them, whenever there's a list of mechanisms by which they could escape, you know that blocking one of those is not going to have major effect because the others could come into play. So what we need to look for are things that are drivers of these multiple mechanisms of resistance, and we've heard about some of those at this meeting. We've heard Othon mention HIF-2-Alpha drive 60 factors, and so being able to block that may be able to block many of these escape mechanisms, but one that I'm particularly interested in is the trying to prevent the loss of P53 in these tumors, and there was work presented by Jim Meyer earlier yesterday in the meeting that suggests that HDM-2 antagonists could sustain P53 levels in these tumors, and this can keep, prevent the escape in several tumor models, and we're trying to move that forward into clinical trials, given that there are many companies who are developing HDM-2 antagonists. So what about this other group, and new agents? Well, we've heard a lot of potential new agents and new targets, and that might suggest what might be driving those tumors. Some of them may be driven by Pythrokinase-AKT-MTOR pathways. Some might be driven, as was mentioned yesterday, by Jack-2 or TYK-2. There also may be loss of tumor suppressors such as some of the chromatin regulatory molecules, as we heard nice talks both yesterday and today about this, either CET-D2 or BAP-1, but one of the things that we've been looking at at Georgetown is the role of the NF-2 hippo pathway, and so this is a pathway we haven't heard mentioned, and maybe Chun-Ling Yee, who's been working on this, might be invited to a future meeting to talk about this, and what we know is that NF-2 is lost in 2% of renal cancers based on TCGA data, but it's also lost in several prominent renal cancer cell lines such as SN12C and ACHN. In addition, the NF-2 knockout mice develops clear cell kidney cancer, and downstream of NF-2 is the hippo pathway, and several of the mediators of the hippo pathway are also mutated in kidney cancer, SAV-1 and LATS-1 leading to activation of YAP. So this is work, we went back to the TCGA data and looked to see whether there was loss of SAV-1 or LATS in the hippo pathway in patients with kidney cancer, and you can see that about 50% of the clear cell or VHL null tumors have loss of either SAV-1 or LATS, and about 80% of the VHL null tumors have lost these genes regulating this pathway, and of interest is where they are located. SAV-1 is located on 14Q near HIF-1 alpha, and LATS-1 is located on chromosome 6, potentially explaining why in this copy number assessment we're missing there were deletions in this particular chromosome, and what this does is up-regulate YAP, and these are some tissue micro rays with antibodies for YAP, suggesting that it's up-regulated in both clear cell and some non-clear cell tumors, and so there are potential drugs that can target YAP, and Chen Ling Li is working on ways of potentially inhibiting this pathway. Finally, I'm going to close with a novel immunotherapies, and we've heard a lot of discussion at this meeting about PD-1 pathway and PD-L-1 antibodies showing high response rates in patients with resistant renal cancer, and the responses may be associated with PD-L-1 expression. There's this phase retrial going on of NIVO versus Everlimus, and there's phase 2 data that should become available soon, so one of the questions is will this data hold up, and how do we move these treatments into the first line, because effective treatments that have low toxicity and produce durable responses are things that we would, if they're effective, would want to give in the first line, and Dave McDermott gave a nice talk about the type of things that we would want to do to try to move these agents into the first line, trying to understand whether their expression of PD-L-1 is truly a biomarker, whether it's influenced by prior VEGF therapy, how PD-L-1 expression might relate to response to other therapies, and that's led to this trial, which we're proposing in the spore, and it's sort of parallel to the type of data that Eric might be able to produce from the trial that is being proposed in the kidney cancer steering group in terms of trying to identify how these various drugs affect various biomarkers, and getting this type of data might be able to help us push companies to actually test these drugs in the first line setting as well as help us design those type of trials, and with these trials may need to have overall survival rather than a progression-free survival endpoints, so we may have to look at the results of a landmark analysis of a sequence rather than just a single therapy. So the most important question is how do we improve the efficacy of these drugs? Patient selection is certainly something that we can look at, but combinations with other checkpoint inhibitors, other immunotherapies, or anti-VEGF therapies are certainly things that could also be looked at, and based on the data with ipilimumab and nombolumab in melanoma with rapid and deep responses in patients with melanoma, that is certainly something that we would want to look at in kidney cancer as well as trying to figure out how to test these agents in the adjuvant or neo-adjuvant setting. So I'm now going to close with slides similar to what I closed with last year, because this is the only place I can present these slides, and I really like them. So the vision of the future is treatment is more personalized immunotherapy is offered first to the patients that are most likely to benefit from it. VEGF therapy is driven by therapeutic index and optimized to the extent possible by either blood pressure monitoring or PK. Tor inhibitors are restricted to patients with tumors driven by the mTOR pathway. Combinations are based on understandings of mechanisms resistant and show true synergy, and new targets and drugs are identified to treatment of refractory patients, which means we'll have potentially another algorithm sometime in 2020 that is listed here that will help us decide on first line therapy and second line therapy won't be necessary. And these rely on the future leaders of kidney cancer to help develop this algorithm. So how do we get there? It's going to require a lot of effort from all of us, translational research, teamwork, and the courage to do the studies that are really going to help our patients. And I'll put back up here in my key issues in case it'll help with discussion. Thank you very much.