Kidney, Bladder and Prostate Cancer TRDG: Meetings and Minutes

Dr. Shannon Lynch spoke about “Towards Precision Prevention: A Neighborhood-wide Association Study in Prostate Cancer”. She started last November, 2015 as an Assistant Professor in the CPC Program. She is an epidemiologist with a background in molecular and cell biology. She spoke about the national precision medicine initiative and is interested in whether previous behavior, socioeconomic and epidemiologic data can be used to reduce cancer rates. She described a multi-level conceptual framework including biologic factors (GWAS, computing) and mentioned that few studies have focused on macroenvironment. She is interested in the neighborhood environment and in using census data, high dimensional computing models that may translate into precision prevention at the individual and community level. She spoke about prostate cancer and racial, social factors that impact outcomes. Income, deprivation (e.g. in education) are associated with poor outcomes. Neighborhood-wide association study (NWAS) applied to census data. She is working towards informed decision-making in at risk communities. There is support from an NCI supplement that is facilitating work at Temple/FCCC; work with Drs. Nestor Esnaola and Susan Fisher. She spoke about the PA Cancer Registry being linked to census data. She discussed the GEE model, Spatial model, principal component analysis reduced that 14,663 variables to 217 variables. She did fine mapping and reduced further to 17 variables from NWAS analysis. Poverty, income, occupation, social support, immigration status, etc. In other work can identify hotspots that could be environmental. Computing approaches lead to useful clues. Compared NWAS to previous approaches. Do NWAS studies predict clinical outcomes? Can neighborhood signatures identify areas to target intervention and screening efforts? She mentioned ACS-IRG funded work 443 men in PRAP with 71 months of follow-up, 69 with prostate cancer; looked at 17 NWAS variables. Found that NWAS variables were relevant. There was discussion about identifying areas at risk versus individuals at risk. She will be looking at machine learning approaches, applying NWAS to African Americans, and incorporating more biological factors.

Dr. El-Deiry spoke about NCI R21 funding opportunity for clinical and translational exploratory/developmental studies. Dr. Alex Kutikov spoke about the state of surgery for localized kidney cancer. This is an exciting space with strengths at Fox Chase Cancer Center. RCC is heterogeneous, ~15000 deaths per year. Some metastasize rapidly and others are more indolent. The incidence of RCC appears to be rising despite more success with therapy, “a treatment disconnect”

The mortality over incidence ratio has been decreasing. Dr. Kutikov described four key questions from a surgical perspective: If to cut, how to cut, what to cut, what to do after the cut. He spoke about individualized decision making dependent on the particular patient. RCC tumors that are <3 cm with slow kinetics get active surveillance. Spoke about treatment decisions made in context of competing risks of death from other causes. He has an interest in risk stratification. Renal biopsy holds promise. Dr. Kutikov described the idea that a benign tumor may coexist with a more malignant tumor, so called “collision tumors”. There is a 25% chance from a benign biopsy of such a coexisting tumor. He looked at resected solitary tumors had 147 with benign component, only 4 had hybrid with chromophobe in background of oncocytoma. None were high grade in the paper but then found a patient with high grade. The hybrid tumors have different genetic alterations. Sarcomatoid tumors that are rapid progressors, chromophobes are overrepresented. The heterogeneity of RCC tumors was discussed including some controversies regarding how common this is. There was discussion about some recommendations to biopsy that are based on an unusually high rate of high grade biology. There was discussion about imaging, sestamibi, CA19 as approaches that have been used but little that is clinically actionable in this setting. Biopsy is useful if benign at 4 cm in setting of higher surgical risk. Risk of progression is a question that may be helped by molecular analysis beyond biopsy. How can you assure the patient they don’t have high-grade disease. There was discussion of liquid biopsy in patients who don’t or can’t get biopsy. If positive for high-grade tumor by epigenetic changes it may be actionable even if sensitivity not high, it may be useful. Dr. Kutikov spoke about partial versus radical nephrectomy with some efforts to do partial although treatment is still individualized. There is lower mortality after partial vs radical nephrectomy. A phase III RCT of partial versus radical nephrectomy showed some advantage to radical for all comers (no difference if the benign tumors are excluded). Dr. Kutikov felt that time is ripe for an RCT for the 4-10 cm solitary kidney masses and could incorporate molecular analysis and liquid biopsy. He discussed robotic surgery; showed video of surgery. Discussed post-operative surveillance. Promise of liquid Bx.

Dr. Roland Dunbrack spoke about “The Three-dimensional Structures of Kinases During Autophosphorylation Events: Opportunities for Drug Development and Understanding Mechanisms of Drug-resistant Mutations in Kinases”. He described a recent publication in Science Signaling. There are >500 kinases in the human kinome. Autophosphorylation can occur in cis or trans. He spoke about trans autophosphorylation. Autophosphorylation of activation loop keeps kinase active or increases activity. Autophosphorylation may impact on binding to other proteins. Dr. Dunbrack described specific motifs within the kinase active site including catalytic motif “HRD”, activation loop start “DFG” and end “APE”. The D in DFG interacts with the Thr or Tyr of the substrate. He spoke about Type I and Type II kinase inhibitors versus specificity. Type II more specific and are extended molecules. He spoke about interfaces in homodimer crystals. Looked at many in the database to unravel structural aspects of autophosphorylation (structural bioinformatics). He found 15 unique “autophosphorylation complexes”

Kit has been crystallized in the phosphorylated form in the crystal. Questions of interest that were addressed for each complex:  Substrate binding and catalysis, importance of domain:domain interface, differences in phosphosite location in substrates and non-substrates, conservation of site between kinases, and what can the analysis say about substrate specificity. Modeling of symmetric and asymmetric structures of IGF1R and others. Symmetric structures are not active. His group made mutants and found some increased or decreased autophosphorylation. Identified common structural motifs that are present and conserved among kinases. Now trying to identify activating mutations in the context of model of kinase activation. Are there examples of drugs that target autophosphorylation? There are some for gleevec resistance. But that approach may be difficult due to structure reasons/lack of a pocket

Dr. Matt Zibelman spoke about checkpoint inhibitor therapy in bladder cancer, biomarkers in context of his trial of IFN-gamma and nivolumab. He described CTLA4 and PD-1 pathways. IFN-gamma drives PD-1 expression on tumor cells. Motzer et al 2014 swimmers plot showed 0.3 mg/kg showed similar responses as 3 or 10 mg/kg nivolumab. Nivolumab vs everolimus in NEJM Motzer et al, 2015. ORR 25% with Nivolumab versus 5% with everolimus. He spoke about RCC biomarker study Snozl; PD-L1 correlated some at a 5% cut-off, in terms of response to nivolumab

CXCL9 part of IFn-gamma a T-cell chemoattractant. Dr. Zibelman discussed bladder cancer further and mentioned the Pembrolizumab KEYNOTE-012 presented by Dr. Plimack at ASCO 2015 and Atezolizumab phase 1 study Petrylak ASCO PD-L1. Dr. Zibelman’s strategy is he wants to force tumors to upregulate PD-L1 such as by IFN-gamma. He noted that IFN-gamma failed as a single agent in bladder cancer. He described his study in solid tumors. IFN is the most potent inducer per Dr. Kerry Campbell. Plans include Biopsy—1 week IFN-gamma—Nivolumab 3 mg/Kg in combination with IFN-gamma x 3 months and Nivolumab will continue after 3 months. There was discussion as to whether radiation induction of PD-1 might involve IFN-gamma. There is interest in dose effects of IFN-gamma. Dr. Zibelman is working with Dr. Raza Zaidi on a mouse model for these studies. Dr. Phil Abbosh spoke about mutation clearance as a means of measuring residual disease after chemotherapy. With muscle invasive bladder cancer there is no way to know who needs chemo or who benefits; surgery complicated, life altering. Can chemo be used with curative intent? There is 30-40% major complication from radical cystectomy; 5% 90 day mortality; 50% long term survival. Those with residual disease do worse after cystectomy, and most are not good candidates for chemotherapy. He is interested in urine biomarkers and described some future directions.

To advance translational goals, Dr. Paul Cairns spoke about genome-wide promoter methylome for diagnosis of small renal masses. He is interested in early detection as well as diagnosis and prognostication. There are few genes with aberrant methylation in RCC, and few specific to kidney. VHL in 5% of clear cells has some specificity but still quite rare. Dr. Cairns has used genomics to detect changes in gene expression in response to 5-AzaC and TSA versus mock treated cells. 262 genes were upregulated >3-fold but he mentioned that the technology is now obsolete. He turned to newer technology to study subtypes, biology and assess potential for epigenetic therapy. Infinium Beadchip technology advantages include low cost with good coverage for CpG islands only, rapid analysis. RCC has heterogeneity, is mostly sporadic, most detected incidentally and mostly small <4 cm. Candidates for active surveillance to avoid surgery early on. Size is followed. 30% don’t grow and the rest is divided between slow and fast growers. Imaging doesn’t differentiate benign tumors from more malignant ones. Clear Cell accounts of >95% of metastatic RCC. Some can metastasize while still small. Different types are believed to have different cells of origin and potential to ID based on methylation patterns. He described a series of 24 clear cell RCCs, 14 papillary, 10 chromophobe, 25 oncocytomas, 4 age-matched normal kidneys. His goal is to find differentially expressed genes in tissue, serum or urine to achieve goals and starting point are the available cancer tissues. Some patients especially younger currently get biopsies but some don’t.  He saw clustering of methylation patterns with methylation patterns; clinical outcomes may be worth looking at in order to evaluate slow or aggressive. Clear cell clustered together, papillary clustered together, normal clustered together. Chromophobe and oncocytoma were mixed together and these can be difficult to distinguish. Dr. Cairns spoke about CIMP in ccRCC as opposed to other ccRCC with much less or intermediate methylation. Question about whether vascularity has been correlated with biological behavior or whether they are all similarly vascular. He spoke about various statistical analysis methods. Settled on Wilcoxon method to avoid biases. He spoke about VHL methylation. Only 2 tumors of the 25 RCC had hypermethylation of VHL almost certainly in one allele. He did some validation experiments on methylated genes in ccRCC. He identified methylated genes that were specific to cell type for clear cell, papillary, etc. Also found organ specificity and could distinguish from bladder methylated genes. Needle biopsies are becoming used increasingly and so the goal isn’t just to clarify imaging. He made the point that body fluids may better to capture intra-tumoral heterogeneity if certain changes are rare within the tumors and may be missed in a biopsy. Dr. Cairns made the point that precision medicine needs to pay attention to epigenetics. There was discussion about miRNA and exosomes. Dr. Testa added discussion about how much of the work from the kidney keystone was done using different technologies on different tumors.

Dr. Betsy Plimack has been redefining neoadjuvant chemotherapy for bladder cancer. She mentioned a NEJM paper from 2003 that showed survival benefit of MVAC adjuvant therapy and spoke about her recent JCO paper. She described the neoadjuvant trial of AMVAC for muscle invasive bladder cancer with biopsies and the observation of 38% CR. She did another trial with dose dense gemcitabine/cisplatin that was previously tested in the metastatic setting. This regimen was toxic with a 32% response rate. Tissue was obtained at baseline and at the time of surgery. It was found that pathologic response at surgery predicts survival. Dr. Cukierman raised a point about if in some patients nothing is left why do the surgery. Dr. Plimack spoke about a collaboration with Foundation Medicine using a test of 236 cancer relevant genes and 19 rearranged genes. She mentioned there were more alterations in the group that responded. Dr. Eric Ross found that ATM, RB1, or FANCC predict pathologic response. Her group examined a validation set of tumors with dose dense gemcitabine/cisplatin and showed the signature correlated with survival. There was discussion about whether clustering of non-responders may suggest other targets and some discussion of methylation patterns, biomolecular assembly analysis, PDX models and whether creating them might help demonstrate mechanistic aspects of the drug responses after the regimens. Dr. Plimack spoke about collaboration and sequencing by Dr. David McConkey at MDACC that identified a p53-like more resistant disease. She spoke about COXEN neoadjuvant SWOG trial and bladder mapping effort. The goal is to use signature to avoid cystectomy in the future in some patients.  mTOR, FGFR are targets that have emerged from TCGA. Dr. Vladimir Kolenko spoke about LDL and TKI resistance. Lack of PTEN expression correlates with sutent resistance in renal and prostate cancer cells using one RCC 786-0 and 2 prostate lines PC-3 and LNCaP. Re-expression of PTEN restores resistance, work described in Molecular Cancer Therapeutics 11:1510-7, 2012. Dr. Kolenko mentioned some publication that suggested sutent effects on endothelial cells. He described sunitinib concentration is >10x higher in tumor versus plasma. Akt, PI3K, or mTORC1 inhibitors sensitize the resistant cells to sunitinib. There was discussion about whether screening for PTEN loss in RCC might be used as a way to avoid a non-effective treatment and discussion about whether since PTEN loss is so much more common in prostate cancer if combination of PI3K and sunitinib might be tried. Dr. Kolenko spoke about LDL treatment which stimulates Akt to mediate resistance in RCC. LDL made RCC cells resistant to various kinase inhibitors including sorafenib, pazopanib, or lapatinib. Pure cholesterol does not protect SK-45 RCC cells from sorafenib. LDL did not protect from doxorubicin or docetaxel. A PI3K inhibitor reinstated sensitivity to TKI in RCC cells despite LDL treatment. Dr. Kolenko showed results from and in vivo experiment giving a high fat, high cholesterol diet and showed this conferred resistance to sunitinib therapy and promoted tumor growth. HDL also restores proliferation and suppresses apoptosis of TKI therapy of SK-45 RCC cells. He described a number of mechanistic & translational plans to move the strategy towards the clinic.

Historically the funding of Keystone Projects helped develop projects in labs. There was discussion that translation needs bolstering and Dr. Plimack mentioned that we need the right trials here. We need a critical mass. Need the passengers in the car. There was discussion about various meeting formats and discussion of the SPORE mechanism and what disease site. Dr. Paul Cairns mentioned DOD as a good mechanism to submit to while preparing for a SPORE. Dr. Plimack discussed using the next meeting to take inventory. One slide per person that may include a project title, names of collaborators, possible aims, possible cores to be used and to give a sense of the impact. What can you do as a project.  Regarding multi-PI grants it was discussed that the quality of the science within the projects is paramount and makes all the difference. It was agreed that it is important for the clinicians to talk to the researchers. There was discussion that Sid’s project is a good example of how translation can be supported and can become successful. In addition to the inventory goal, it was discussed that the group would benefit from some review of the grant mechanisms, as well as databases and TMAs. The idea that the TRDG should be very inclusive and reach out to potential members including more efforts at scheduling through a doodle poll.