 And thank you to the organizers for the kind invitation to present to you today. I wanted to give you a brief outline of where the field of vaccine sits for renal cell carcinoma and maybe where we're going to be going with this over the next several years. As a PhD cancer immunologist, I've long had an affinity for both melanoma and renal cell carcinoma. They basically are the classic examples of immunogenic cancers. There are infrequent but spontaneous regressions that occur in RCC as they do in melanoma as well, and these are associated with inflammatory cell infiltrates and or enhanced peripheral T-cell function. If you look at renal cell carcinoma as it develops over time and progresses, you can also constantly find evidence for accumulated defects in antigen processing machinery, as well as MHC expression. This is suggestive of an ongoing fight between the immune system and the cancer with the cancer cell trying to evade the immune response by down-modulating recognition elements required for immune intervention. Also, particularly IL-2 high-dose therapy, there is evidence for durable long-term survivors amongst RCC patients that are treated, not high frequencies, but when they do respond, they tend to respond for a very long period of time in association with specific T-cell responses in the periphery. We now have almost a luxury of wealth with regard to the number of antigens that we have defined in the RCC setting. This is just a partial list of these particular antigens that are expressed in greater than 50 percent incidence and somewhat more limited incidence in the RCC setting. This is more than three dozen antigens. This is not a comprehensive data set by any means. It's important to note that these typically are not mutated gene products, but they are wild-type sequences, and they tend to be overexpressed in the cancer microenvironment based on epigenetic programming, specifically because their promoter regions have hypoxia response of elements. They also are subject to demethylation events, which turn on transcriptional activation of these gene products, and in addition, many of these gene products are client proteins for HSP90, which is a chaperone molecule, which is up-regulated in many cancer cell types. The chaperone protects that particular protein by refolding it into its native confirmation and prevents it from undergoing proteolytic degradation via the proteasome in the tumor cell. Each of these different paradigms allows for these antigens to be overexpressed in the tumor cells themselves and to be serving as a differential target for immune recognition. If you have a specific antigen, whether that be an RCC peptide derived from one of these gene products that I just mentioned, it can be synthetic or natural, you can also have RCC proteins, whether these be derived from lysate material, membrane components or even Hitchcock protein complexes from these particular proteins in the cell. Genetic tumor RNA and or recombinant viruses encoding RCC antigens, as in the case of Muck 1 or the oncophetal antigen 5T4, you can also have RCC whole cell-based vaccines, which are either irradiated gene-modified to make them more immunogenic and or fusion proteins with dendritic cells. All of these have been evaluated at some level in the context of specific immunization in cancer and in particularly RCC. The intent with providing these particular vaccines into a subcutaneous site, for instance, as one vaccinates, is to access a stimulatory dendritic cell. If you're injecting a DC, obviously you can condition the DCX feeble for provision in a stimulatory format, but alternatively you need to condition the microenvironment of the vaccine to promote, and that's typically done through an adjuvant, in order to promote a co-stimulatory high phenotype, co-inhibition low phenotype in that particular DC, in a balance towards IL-12P70 production and away from IL-10. This gives you a polarizing cytokine microenvironment that's conducive to type 1 polarization in the T-cell population. You also want these DCs to turn on CCR7 expression, which is required for their migration to secondary lymph nodes, where they actually interact with the specific CD8 and CD4 T-cells that can then expand, hopefully polarized towards type 1 cytolitic function, and then circulate and be recruited into the tumor microenvironment and mediate effective anti-tumor activity over time. Rather than give you a myriad of tables, which I think would be uninterpretable, this is a summary of 41 clinical trials using various formats of antigen, whether these be whole tumor, lysates, HSB complexes, cell fusion, viral or gene modified cells, as well as peptide specific vaccinations. And what I really want you to take away from this is that based on Rhesus criteria, very modest clinical, rather complete response and PR rates in these individuals, disease stabilization maybe predictably would be the most dominant response that one sees, but these tend not to be terribly durable. And so at this particular juncture, these are not terribly effective monotherapies. Having said that, they do reach their goals in terms of specifically inducing T-cell responses against the immunizing agent. And this has been done through a hodgepodge of different immunologic monitoring assays as primitive as delayed type high sensitivity responses at the vaccine site, and then using more current cutting edge technologies, including ELISA and LE spot, cytotoxicity and MAC peptide multimers. Again, this is a hodgepodge of results based on these different methods of screening for T-cell responses, but in general, about half the patients tend to respond to these vaccines based on these immunologic outcomes. So what have we learned overall from doing all of this? Basically these vaccines are safe and well tolerated. There has been very little evidence for diminished quality of life or for pathologic autoimmune sequela. They yield a high rate of T-cell specific induction, but low rates of OCR based on Rhesus criteria. And really, I think from my perspective, there really isn't any dominant bias towards any of these different formats. I mean, whether it's a lysate or a specific peptide induction, one has a more refined approach using a specific peptide derived from an antigen, but basically they've all elicited similar levels of specific T-cell induction over time. So patients that do exhibit OCRs do tend to display or they tend to fall into the cohort of patients that exhibits the strongest reactivity to these specific activation events with regard to specific T-cell responses. In that vein, a meta-analysis of DC-based vaccines over 12 trials and 172 RCC patients supports a link between enhanced specific immunity post-vaccination and improved objective clinical response with an odds ratio of 8.4. And then while variable, some trials and specifically lysates have reported a statistically significant improvement in five-year progression-free survival or overall survival. Other lessons learned, and these are sort of based on some of the more cutting-edge screening that can be done with immune responses. Patients that exhibit objective clinical responses tend to display also coordinate reductions in T-regulatory cells and MDSCs. These are screen and peripheral blood. Improve polyfunctionality of the T-cell responses, so it's not just enough to produce interferon gamma. They may immediate acylytic function or produce alternate important cytokines that are supportive of the anti-tumor mechanism. And then finally, I think no matter what you're going to immunize the patient with, it's going to be a limited set of specificities, and overall you want this to evolve into a very broad repertoire. Tumor cells aren't stupid. They can evolve away from a specific antigen being targeted by the T-cell response, so the broader your repertoire, the greater your chance for therapeutic efficacy. This is basically a lesson learned in the context of autoimmune disease where epitope spreading occurs, and that's, I think, ideally what you want to prime with a vaccine, but ultimately you want those broader specificities to provide you protection over the long term. And maybe predictably, these vaccine trials have also suggested, wherever it's been looked at, that these are more effective in patients of earlier stage and with a lower degree of tumor burden. I think those are essentially no-brainers, but with regard to actually people documenting that, it's been fairly limited. Now are there also serum markers that are predictive for responsiveness to vaccines? And this is, again, a burgeoning field, but right now there aren't that many markers out there to look at. In a Trovax vaccine model, phase two, it was reported that normal, mean, corpuscular hemoglobin concentrations, which is associated with lower tumor burden and the lower levels of CRP in the serum, was associated with better responsiveness to the vaccine as well as outcome to vaccine. High serum concentrations of APO A1, which is a marker of reduced chronic inflammation and immune suppression, was associated in an Amatics phase two trial with better outcome and responsiveness. And then from my perspective, interestingly, two chemokines that are important for recruiting those vaccine-induced T-cells back into the tumor are upregulated in the serum of patients that do better. And then finally, keeping with Jim Finke's work and others, the Amatics phase two trial also suggests that lower levels of MDSCs, and particularly the MDSC four and five subsets were associated with improved responsiveness to these vaccines. So why is there a disconnect between cancer vaccine immunogenicity and clinical efficacy in just making more of these particular cells isn't going to be enough? And so there's going to be a need to combinate these vaccines with alternative therapeutics in order to improve the recruitment of these and functionality and survival of these T-cells once they arrive in the tumor microenvironment. We need to increase the ability of low-avidity T-cells to recognize and react against RCC. Again many of these antigens are self-antigens. If you will, non-mutated. They undergo tolerance mechanisms and the high-avidity cells are eradicated normally. We need to remove regulatory cells and function, including T-regs and MDSC, and we need to expand the repertoire through this cross-priming event that should be reiterated over time in an epitope-spreading paradigm. As Jim mentioned, we've done some preclinical work with synitinib in combination with vaccines. In this particular context, the combination of the vaccine and sous-tent was able to promote, in fact, it was the only cohort that gave us disease-free animals with about a quarter of these animals completely rejecting their tumors. This is associated with a significant infiltration in the combined therapy of the specific T-cell populations based on tetramer analysis and also an augmentation of the vaccine-draining lymph node. But again, clinically this is the more important result with regard to tumor infiltration. We've gone on now to show that synitinib isn't alone, and so basically sous-tent conditions the tumor microenvironment in a number of ways that are beneficial for vaccines in terms of recruiting those particular cells and sustaining them. It augments or increases V-cam 1 expression, which is an adhesion molecule and activated endothelial cells. This allows for those cells to bind and to be recruited into the microenvironment better. The chemokines, CXCL10, which binds to CXCR3 on type 1T cells, is increased in the stroma. This again enhances recruitment. The type 1 signal for the recruited cells, including interferon gamma and T-bet, the transcriptional factor associated with type 1 function, is also augmented. And then you have a bunch of signals associated with suppression that are down-modulated, including CXCL12, which is a chemokine that recruits MDSCs as well as mesenchymal stem cells, supportive of tumor growth, and then reduction in T-regs, MDSCs, and hypoxia-induced factors. We see a similar paradigm with regard to the changes that occur in the environment, whether this be synitinib, axitinib, or disatinib. Imatinib in our hands was a dud, and it's one of the few that we've seen so far that does not work in through these particular mechanisms. We also notice that HSP90 inhibitors and particular 17D mag does very similar things. And both disatinib and the 17D mag promote enhanced recognition of tumor cells, in part by conditionally triggering the degradation of proteins into the proteasome pathway, which is required to then be presented as peptides on the cell surface, which is there. All right, so putting this into practice, imatics in a phase three trial is now accommodating vaccine plus synitinib. This is a multi-center open label randomized phase three in 338 to positive renal cell carcinoma patients with metastatic or locally advanced disease. After one cycle of synitinib, they then get a low dose injection of cyclophosphamide to reduce MDSC content, as well as an ID injection of 10 peptides, including nine A2-presented class one peptides that allow for CD8 recognition, as well as a PAN class two antigen to facilitate help in the CD4 compartment, along with the adjuvant recombinant human GMCSF. Primary endpoint is to compare overall survival against Su-10 alone. And this particular trial is ongoing, but not recruiting patients currently with an expected completion in early 2015. Lastly, I just wanted to touch on an alternative approach, which is to not target the tumor cells themselves, which, again, are genetically unstable and undergoing immune evasion mechanisms, and to consider an alternative strategy of targeting stromal cell populations, in part, because the epigenetic programming in these particular cells is altered as a consequence of being in the tumor microenvironment, specifically targeting pericytes and vascular endothelial cells with a specific vaccination to elicit a specific T cell response. In order to do this, we basically needed a set of antigens that are differentially expressed by the tumor-associated vascular cell populations. And this is basically using real-time PCR to compare renal cell carcinoma with patient-match normal adjacent kidney, which is tumor-uninvolved tissue, with flow-sorted pericytes and vascular endothelial cells. And what we were able to see is that several of these antigens are transcriptionally up-regulated fairly profoundly in the pericyte populations, and in this particular instance for the FA2 protein and vascular endothelial cells. Mouse modeling in a system where the tumor itself cannot be recognized by the T cells because there's a genetic incompatibility with regard to the restriction of the T cells, that those particular vaccines can allow for the eradication of the tumor and for long-term disease-free status in these particular mice. And this is using a cohort of the antigens that I just mentioned. All the ones in blue gave very significant enhanced survival. TEM1 gave a somewhat modest response, but we think this can be further optimized. And then lastly, I just wanted to show you a genetic vaccine based on this DLK1, which is Delta-like I homologue. This particular gene was administered through a lentivirus vector, a single administration of the vaccine at seven days into an established RENCA-RCC model and valve C mice. And these particular vaccines, regardless of a lower or a high dose of the vaccine, were able to elicit durable protective immunity long-term in these animals as they completely regressed their tumors. Just to reiterate, the DLK1 protein is transcriptionally only found in the tumor-associated pericytes, but not the vasculinothelial cells or the tumor itself. And we subsequently showed that CDAT cells required based on depletion analysis. And then just to summarize, the tumor microvinyl changes as a consequence of the specific vaccination. This is the control tumor. This is the vaccinated tumor at 27 days. And you find there's a significant simplification in vascular arborization in these particular tumors as a consequence of vaccination. Vascular permeability is decreased, hypoxia is reduced, and you have an increase in apoptosis in areas that are distal to the remaining vessels in these particular tumors, as well as the increase in production of chemokines that are required for recruiting the T cells. So in summary, RCC-based vaccines are safe, well-tolerated, and typically immunogenic, and eliciting type 1 T cell responses against RCC. Objective clinical response rates after vaccination are modest, but the patients that benefit tend to demonstrate stronger immune reactivity, and that tends to be more durable. Early biomarker analysis suggests that chemokines, specifically CCL17 and IP10 or CXL10, may be suggestive of an ongoing response that's able to recruit those T cells into the tumor constitutively, and that can be of benefit in the context of vaccination. Combination of vaccine strategies that further enhance T cell recruitment, enhance function and survival, are also going to be important, and we think these are very simple combinations with existing drugs such as anti-CTLA4, anti-PD1, PDL1, TKI, such as synidinomaxidinib or disatinib, HSP90 inhibitors, as well as chemotherapeutic agents that reduce MDSC content and or T-REG content. And then lastly, we basically think the vaccine should be a little bit open-minded with regard to the targets themselves, so RCC is important, but the stroma is equally important with regard to supporting the growth and pretentiation of the tumor, and I think there is certainly an argument that can be made for differential expression of target antigens in that particular cohort. And I'll just say thanks to the KCA to all of you for having listened to this and to our many collaborators. And I'd like to also mention that Ron Bacowski should be on the slide because he really got us started in this about 10 years back, and I thank him very much for that. Thank you.