Search

Pancreatic ductal adenocarcinoma (PDAC) is one of the top five deadliest forms of cancer with very few treatment options. The 5-year survival rate for PDAC is 10% following diagnosis. Preclinical murine models have been developed that leverage key driver mutations and have significantly contributed to our understanding of PDAC. One such genetically engineered mouse model (GEMM) that has emerged as an important tool is the KPC mouse (LSL-KrasG12D/+;LSL-Trp53R172H/+; p48-Cre) that spontaneously develops pancreatic tumors at ~14-16 weeks of age. Cadherin-11 (Cdh11), a cell-to-cell adhesion molecule has been suggested to play a role in development of the desmoplastic stroma in PDAC, that leads to difficulties in drug accessibility and has been hypothesized to contribute to chemotherapeutic resistance and correlate with poor prognosis. However, the mechanisms by which Cdh11 deficiency in the stromal microenvironment of PDAC-bearing KPC mice influences tumor infiltrating immune cells, has yet to be fully understood. Single-cell RNA sequencing (scRNAseq) of the immune (CD45+) compartment of tumor bearing Cdh11 proficient (KPC/Cdh11+/+), tumor bearing Cdh11 deficient (KPC/Cdh11+/-), non-tumor bearing Cdh11 deficient (Cdh11+/-) and wildtype (Cdh11+/+) mice was performed. We observed a sharp decrease in the presence of myeloid/monocyte lineage cells (CD14+) in KPC/Cdh11+/- tumors and also an increase in T, B and plasma cells, compared to KPC/Cdh11+/+ tumors. Genes upregulated in infiltrating T- and NK cells specific to a Cdh11 deficient background include Cd8a, Nkg7, Maf. Additionally, genes found to be upregulated in B cell clusters in Cdh11 deficient mice include those related to B cell differentiation/activation such as Lgals1, Id2, Itgb1, Rgs1. The increase in B and T cell infiltration was specific to the Cdh11 deficient background, since both pancreata from KPC/Cdh11+/+and Cdh11+/- mice had elevated levels of infiltration. Immunohistochemical validation of these findings has confirmed these changes in tumor infiltrating immune cells. Additionally of note, an increase in antibody-producing plasma cells was observed specifically in a Cdh11 deficient background. Igkc, an immunoglobulin found to be enriched in plasma cells was highly expressed in this group of immune cells. We also observed that KPC/Cdh11+/-had significantly more Igkc expressing cells than KPC/Cdh11+/+. Future work is needed to clearly define the role of Cdh11 in modulating B, T and plasma cell behavior and subsequent contributions to PDAC outcome. This study received funding by LDRD 19-SI-003. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344), LLNL-ABS-820889. This work was supported by AACR-AstraZeneca Fellowship in Immunooncology Research, grant 17-40-12-PERA; The Ruesch Center for the Cure of Gastrointestinal Cancers grant award; NIH R01 CA170653; and NIH Cancer Center Support Grant P30 CA051008. Citation Format: Kelly A. Martin, Aimy Sebastian, Nicholas Hum, Ivana Peran, Stephen Byers, Elizabeth Wheeler, Matthew Coleman, Gabriela Loots. Loss of cadherin 11 in pancreatic cancer induces altered immune cell infiltration [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2516.

Lung cancer is the second most diagnosed type of cancer and is the leading cause of cancer-related deaths in both men and women. An estimated 235,760 new cases of lung cancer and 131,880 lung cancer deaths are expected this year accounting for 12% of new cancer cases and 22% of cancer deaths in the United States. Exosomes are secreted vesicles that contain selectively packaged biomaterials (protein, lipids, and RNA) and serve as a form of extracellular communication between cells in the tumor microenvironment. Previous studies have shown the effects of cancer-derived exosomes on specific cell types or gross tumor response in vivo yet, the competitive nature and dynamics of exosomal uptake in the tumor microenvironment remains largely unknown. This work seeks to identify molecular responses induced by cancer-derived exosomes on cells residing in the tumor microenvironment. In order to control tumor cell exposure to cancer-derived exosomes, heterogeneous ex vivo cultures derived from subcutaneous Lewis lung carcinoma (LLC) tumors in C57BL6 mice were exposed to fluorescently dyed LLC exosomes for up to 72 hours. Flow cytometric analysis was performed to identify preferential uptake of cancer exosomes into the stromal subpopulations of tumor cells. Increased uptake was observed throughout the duration of the exposure and preferential uptake was observed into neutrophils, macrophages, and endothelial cells while cancer cells, fibroblasts, and lymphocytes were underrepresented in the exosome uptaken population. Additionally, single cell RNA-sequencing was performed 24 hours post-exosomal exposure to identify transcriptional changes promoted by cancer exosome uptake. We found tumor-derived cancer cells to up-regulate transcripts associated with proliferation and down-regulated genes associated with interferon signaling in response to exosome uptake. These preliminary findings suggest that cancer-derived exosomes have targeted effects on specific cell types in the tumor microenvironment. Specifically in cancer cells, exosomal uptake can elicit protumor (cancer proliferation and immune suppression via decreased interferon signaling) effects that may serve as mediators of disease progression. An understanding of both cancer and host-derived exosomal response to disease progression may identify novel biomarkers and potential therapeutic targets previously undiscovered using canonical discovery methods. This study received funding from LLNL LDRD grant 21-LW-028. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). LLNL-ABS-829271 Citation Format: Nicholas R. Hum, Nicole F. Leon, Aimy Sebastian, Kelly A. Martin, Gabriela G. Loots. Molecular characterization of the effects of cancer-derived exosomes on murine lung cancer tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3136.

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal forms of cancer with very few treatment options. Less than 10 percent of patients diagnosed with PDAC survive 5 years post diagnosis. Mutations in CDKN2A, SMAD4, KRAS and P53 have been well linked to the development of PDAC. Preclinical murine models have been developed that leverage key driver mutations and have significantly contributed to our understanding of PDAC. One such genetically engineered mouse model (GEMM) that has emerged as an important tool in PDAC investigations is the KPC mouse (LSL-KrasG12D/+LSL-Trp53R172H/+Pdx-1-Cre) that spontaneously develops pancreatic tumors at ~14-16 weeks of age. Cadherin-11 (Cdh11), a cell-to-cell adhesion molecule, is highly expressed in desmoplastic stroma, a characteristic of PDAC, that leads to difficulties in drug accessibility and has been hypothesized to contribute to chemotherapeutic resistance. However, the mechanisms by which Cdh11 deficiency in the stromal microenvironment of PDAC-bearing mice (KPC) influences therapeutic outcomes, has yet to be fully understood. Single-cell RNA sequencing (scRNAseq) of both the non-immune (CD45-) and immune (CD45+) cellular compartments of tumor bearing (KPC/Cdh11+/+), tumor bearing Cdh11 deficient (KPC/Cdh11+/-), non-tumor bearing Cdh11 deficient (Cdh11-/+) and wildtype (KP) were performed. We observed changes in the abundance and types of infiltrating immune cells (T-cells, B-cells, myeloid lineage cells) of KPC/Cdh11+/- tumors when compared to tumors harvested from KPC/Cdh11+/+ mice. KPC/Cdh11+/+ pancreata had significantly more myeloid cells while KPC/Cdh11+/- tumors favored an increase in the numbers of infiltrating B- and T- cells. Genes upregulated in infiltrating T-cells specific to KPC/Cdh11+/+ mice include Spp1, Ifi30, Apoe, Ifitm3, Fn1. The increase in B and T cell infiltration was specific to the Cdh11-/+ deficient background, since both pancreata from KPC/Cdh11+/- and Cdh11-/+ mice had elevated levels of infiltration, compared to the KPC group. We also observed a decrease in the number of antigen-presenting cancer associated fibroblasts (apCAFs) in Cdh11-/+ and KPC/Cdh11+/- pancreata, denoted by the lack of CD74+ fibroblasts. Further validation of these findings will help to define the role of Cdh11-/+ in modulating B and T-cell behavior in addition to providing insight into Cdh11-/+ as a therapeutic target for PDAC through altering the tumor microenvironment. Citation Format: Kelly A. Martin, Aimy Sebastian, Nicholas Hum, Ivana Peran, Stephen Byers, Elizabeth K. Wheeler, Matthew A. Coleman, Gabriela Loots. Loss of Cadherin-11 in PDAC induces altered immune cell infiltration and remodels stromal landscape [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2741.

Pancreatic ductal adenocarcinoma (PDAC) is one of the most incurable types of cancer. Cancer that develops in the acinar cells of the pancreas is typically not diagnosed until later stages, such as stage 3 or 4. As such, this form of cancer is particularly lethal with only 9% of patients reaching 5-year survival. PDAC is known to have a particularly dense extracellular matrix composed of fibroblasts, which have been previously shown to play an important role in promoting resistance to drug therapy. Characterization of the stromal networks involved in PDAC tumor development as well as protein markers of fibroblast subpopulations within the tumor stroma are critical to developing new fibroblast-targeted therapeutic approaches as well as understanding key signaling molecules that ultimately promote tumor progression and drug resistance. Single-cell RNA sequencing (scRNAseq) was utilized to analyze cancer-associated fibroblasts (CAFs) from KPC mouse-derived, MT3 subcutaneous murine allografts along with two fibroblast lines derived from human PDAC tumors: CRC-811 and IA-1340. Single cell gene expression profiling and subsequent analysis of MT3-derived CAFs resulted in the identification of three CAF subpopulations including a myofibroblast subpopulation that expressed high levels of smooth muscle actin (Acta2), which was also observed in both human samples. Fibroblast subpopulations enriched in Acta2 expression, expressed high levels of Wnt5a along with several other secreted factors including Tgfb1, Tgfb2 and Ctgf. Wnt5a is a secreted protein that activates non-canonical Wnt signaling pathways and is known to regulate normal developmental processes, including proliferation, differentiation, migration, adhesion and polarity. However Wnt5a is not natively expressed in MT3 cancer cells derived from syngeneic tumors, but potentially in the stroma. We hypothesize it is exclusively derived from fibroblasts. Previously it has been shown that Wnt5a inhibition suppressed gastric cancer metastasis, therefore further validation of the role of myofibroblast-derived Wnt5a on PDAC disease progression is warranted. This study received funding by LDRD 19-SI-003. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). IM Release Number: LLNL-ABS-798442. Citation Format: Kelly A. Martin, Aimy Sebastian, Nicholas R. Hum, Stephen Byers, Elizabeth K. Wheeler, Matthew A. Coleman, Gabriela G. Loots. Analysis of stromal myofibroblasts identifies secreted proteins that promote pancreatic ductal adenocarcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2757.

Epithelial to mesenchymal transition (EMT) of cancer cells during tumor progression has been implicated in tumor initiation, growth, invasion, metastasis, colonization and resistance to therapy. In building a successful in vitro tumor model, it is critical to recapitulate in vivo cancer cell heterogeneity inclusive of both cell types and transition state present. This study investigates the EMT hybrid states of mouse triple negative breast cancer cells during ex vivo culturing across multiple methodologies. Mouse mammary carcinoma (4T1) tumors were harvested from syngeneic, orthotopic mice and subsequently cultured using various methods over the course of 12 days. Ex vivo cultures were first assessed for retention of tumor cellular heterogeneity using endogenous Thy1.1 (CD90.1) expression via flow cytometry to distinguish 4T1 cancer cells from stromal derived cells. Cancer cell populations rapidly became the majority of cells in culture. After 3 days of ex vivo culture, we found the cancer cell population to have significantly expanded ~2-3 fold compared to the original tumor population, while other stromal cellular subtypes decreased. Additionally, monolayer culture of ex vivo cells contained significantly more cancer cells relative to the spheroid or tumoroid (tumor fragment) culturing techniques. Cancer cells from each culturing technique were also evaluated for loss of Epcam expression and mesenchymal cell fate in reference to the initial EMT distribution at time of isolation. 4T1 cells cast into hydrogel retained high proportions of cells undergoing EMT, evidenced by fewer cells expressing Epcam. However, all other in vitro conditions favored the expansion of cancer cells in an epithelial state compared to in vivo tumors. Subsequent analysis of EMT populations for transitional hybrid states based on CD51, CD61, and CD106 expression was conducted using flow cytometric analysis. We found that in vitro culturing promoted mesenchymal character and this selection was time dependent. Additionally, we found that spheroids cultured in hydrogel for 7 days most closely resembled early hybrid EMT cell states ratios found in vivo Future research aims to optimize ex vivo culturing methodologies to best retain cancer cell characteristics and behavior of tumors obtained from human biopsies. Alterations in growth conditions and identifying critical stromal populations of interest will be critical in development of optimized preclinical ex vivo tumor culture models for drug discovery or personalized treatment. This study received funding from LLNL LDRD grant 19-SI-003. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). IM: LLNL-ABS-798441 Citation Format: Nicholas R. Hum, Kelly A. Martin, Elizabeth Wheeler, Matthew A. Coleman, Gabriela G. Loots. Epithelial-mesenchymal hybrid population changes from monolayer, spheroid, and tumoroid ex vivo culture of syngeneic murine mammary tumors [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2810.

Exosomes are endosomal secreted vesicles containing a variety of genetic and non-genetic material that can be transferred between cells (DNA, RNA, protein, lipid, etc.). Cancer derived exosomes have been implicated in a wide range of cancer mechanisms such as promotion of tumor growth, tumor angiogenesis, immune evasion, drug resistance, and metastasis. In addition to functional significance, cancer exosomes may possess novel biomarkers with potential uses for non-invasive liquid biopsies for cancer detection and monitoring of disease progression. Here we examined differences that may exist in extracellular vesicles or exosomes (EVs) shed by cancer cell lines with various metastatic potential (MDA-MB-231, highly metastatic; MCF-7, weakly metastatic; MCF-10a, normal breast epithelial cells). Morphological analysis of EVs was performed using fluorescent Nanosight Tracking Analysis and revealed decreased mean diameter of EVs derived from MCF-7 cells compared to the other two cell lines, there was also increased exosome particle yield in this cell line. Small RNAs from EVs cargo were also analyzed via sequencing (exoRNA-Seq). The majority of sequenced RNA aligned to coding regions of the human genome (~60%) across all cell lines. Non-coding RNA classification also showed little variability across cell lines examined regardless of metastatic potential, with ~38% of RNA reads corresponding to non-coding RNAs. Pairwise comparison of these cell lines demonstrated that each line packaged unique cargos into their shed exosomes. The non-coding regions corresponded to a variety of small RNAs (miRNA, snoRNA, snRNA, miscRNA, rRNA) but included other RNA features such as pseudogenes and antisense RNAs. We found 304 genes significantly up-regulated and 150 genes significantly down-regulated when comparing metastatic cancer exosomes (MCF-7 and MDA-MB-231) to normal, non-tumorigenic exosomes (MCF-10A). Among examined microRNAs we found several miRNAs associated with metastatic behavior that have previously been implicated in cancer biology, which will be prioritized for validation. Also, 21 RNAs were upregulated at high levels in exosomes shed by metastatic cells, and their expression level was directly proportional with the metastatic character, suggesting that these RNAs may be tested for their potential as biomarkers. Future research will expand upon these newly identified genetic signatures of metastatic exosomal cargo and further validate their influence in driving metastasis. This study received funding from DOD grant BC151687. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). IM number: LLNL-POST-758941 Citation Format: Kelly A. Martin, Nicholas R. Hum, Aimy Sebastian, Gabriela G. Loots. Comparison of exosomes shed by breast cancer cell lines with varying metastatic potential [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 5157.

Bladder cancer is among the top ten most common cancers diagnosed, with about ~380,000 new cases and ~150,000 deaths per year reported worldwide. Platinum-based chemotherapy in combination with gemcitabine, a nucleoside analog, is a widely used treatment option for advanced bladder cancer. It has been shown that only ~50% of the patients with advanced bladder cancer respond to platinum-based therapy. Cancer cells often become non-responsive to therapy that once proved efficacious and are now drug resistant. Drug resistance represents a significant, ongoing challenge in eradicating cancer. We have employed a patient-derived bladder cancer xenograft (PDX) platform to further investigate the molecular mechanisms that contribute to gemcitabine-induced drug resistance in advanced bladder cancer. Transcriptome profiling of passage 4 bladder cancer xenograft tumors from 2 gemcitabine sensitive PDX lines (BL0440 & BL0293) was performed using RNA sequencing (RNAseq) analysis, before and after a 21-day cisplatin/gemcitabine drug treatment regimen. Key regulatory pathways and genes contributing to drug resistant bladder cancer have been identified and validated by overexpression in a 5637 bladder cancer cell line. RNA sequencing data has indicated significant differences between the transcriptional profiles of drug-sensitive and drug-resistant tumors. PDXs retained morphology and shared 92-97% of genetic alterations of parental cancer cells. We identified 106 genes >1.5 fold up-regulated and 45 genes >1.5 fold down-regulated in the drug resistant tumors compared to their drug sensitive counterparts. Of the genes that were significantly upregulated, two methyltransferase enzymes were further validated: Nicotinamide N-methyltransferase (NNMT) and methionine methyltransferase 1A (MAT1A). These genes were found to be contributors to gemcitabine-mediated drug resistance. 5637 cells overexpressing NNMT yielded a 200-fold increase in gemcitabine resistance relative to parental strain. In conclusion, chemoresistance to gemcitabine is associated with differential expression of genes and alteration of downstream signaling pathways. Upregulation of NNMT & MAT1A can potentially be used as novel biomarkers for subpopulations of drug resistant bladder cancer for which improved therapeutics can be developed. Future direction will likely include studies to elucidate exact molecular mechanisms by which cancer cells utilize methyltransferases to no longer respond to gemcitabine therapy. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). Citation Format: Kelly A. Martin, Aimy Sebastian, Nicholas R. Hum, Deepa K. Murugesh, Chong-xian Pan, Ai-Hong Ma, Ralph W. de Vere White, Gabriela G. Loots. Overexpression of nicotinamide N-methyltransferase confers gemcitabine resistance in bladder cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 908.

Breast cancer tumors have a high rate of metastasis and certain aggressive subtypes favor the bone environment. Clinically, it is estimated that metastatic breast cancer has a prevalence of up to 70% of bone tumors. Exosomes are important in cell-cell communication, and are increasingly being recognized as key contributors for priming targeted tissues for metastasis. Despite numerous advances in exosome detection and cargo characterization, transcriptional effects of cancer exosomes on targeted cells remain difficult to characterize at physiologically relevant levels. This is largely due to a combination of limited exosome secretion/uptake and an inability to segregate affected cells from the population. Conventional approaches to circumvent these limitations include spiking cell environments with exosome quantities far in excess of physiological concentrations. We sought to analyze the transcriptional effect of cancer exosomes on cells that have uptaken cancer-derived exosomes in a dynamic co-culture environment. In order to accomplish this, we first engineered a human breast cancer cell line (MDA-MB-231) to express GFP labeled exosomes. This novel cell line expresses three GFP fusion proteins for tetraspanin exosome markers (CD9, CD63, and CD81) known to be found on exosomes derived from MDA-MB-231 cells. Stable, intrinsic, fluorescent labeling of three exosome markers allows for an improved, more sensitive method for the analysis of exosome transfer. We can now examine transfer representing a more diverse pool of exosomes requiring no manipulation post exosome isolation prior to visualization. To investigate breast cancer exosome-mediated alterations on gene expression in bone cells to mimic breast cancer bone metastasis, this GFP exosome expressing line (MDA-MB-231 exo-GFP) was co-cultured with a mouse osteoblast cell line (MC3T3) then analyzed via flow cytometry to quantify the uptake of cancer exosomes in target cell population over time. MC3T3 cells that become GFP positive, indicative of exosome mediated transfer, increased from 0.91% of the population on day 3 post co-culture to 17.3% on day 14. Subsequent cell sorting and RNA-seq of GFP positive MC3T3 populations yielded novel insights into the progressive transcriptional effect of cancer exosomes on target cells. Future research will expand upon this approach to examine the effects on bone cells non-metastatic and metastatic cancer cell lines have on the metastatic niche through exosome-mediated signaling. This approach introduces a novel method for prolonged transcriptional analysis of cancer and target cell co-culture allowing the segregation of exosome affected and unaffected subpopulations of target cells. This study was funded by DOD grant BC151687. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). Citation Format: Nicholas H. Hum, Kelly A. Martin, Gabriela G. Loots. Transcriptome analysis of osteoblasts fused with cancer-derived exosomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2136.

The extent of angiogenesis and/or vascular endothelial growth factor (VEGF) expression in neuroblastoma tumors correlates with metastases, N-myc amplification, and poor clinical outcome. Recently, we have shown that insulin-like growth factor-I and serum-derived growth factors stimulate VEGF expression in neuroblastoma cells via induction of hypoxia-inducible factor-1α (HIF-1α). Because another marker of poor prognosis in neuroblastoma tumors is high expression of brain-derived neurotrophic factor (BDNF) and its tyrosine kinase receptor, TrkB, we sought to evaluate the involvement of BDNF and TrkB in the regulation of VEGF expression. VEGF mRNA levels in neuroblastoma cells cultured in serum-free media increased after 8 to 16 hours in BDNF. BDNF induced increases in VEGF and HIF-1α protein, whereas HIF-1β levels were unaffected. BDNF induced a 2- to 4-fold increase in VEGF promoter activity, which could be abrogated if the hypoxia response element in the VEGF promoter was mutated. Transfection of HIF-1α small interfering RNA blocked BDNF-stimulated increases in VEGF promoter activity and VEGF protein expression. The BDNF-stimulated increases in HIF-1α and VEGF expression required TrkB tyrosine kinase activity and were completely blocked by inhibitors of phosphatidylinositol 3-kinase (PI3K) and mammalian target of rapamycin (mTOR) pathways. These data indicate that BDNF plays a role in regulating VEGF levels in neuroblastoma cells and that targeted therapies to BDNF/TrkB, PI3K, mTOR signal transduction pathways, and/or HIF-1α have the potential to inhibit VEGF expression and limit neuroblastoma tumor growth. (Cancer Res 2006; 66(8): 4249-55)

Background: The inability to effectively treat metastases is the main reason for the limited progress in reducing the rates of cancer morbidity and mortality. One major drawback is the lack of quantitative assays for assessing the size and tissue prevalence of tumors in newly diagnosed individuals. Current methods for quantifying tumor burden are mainly qualitative and include measuring the gross weight of the affected organ, counting tumors on the surface of the organ, or evaluating a small sample of the organ using histologic sections. These methods are crude measures of tumor burden and size distribution, and in the case of histology, they are time consuming, difficult to process an adequate sample size and non-quantitative. Methods: Animal models of metastasis have been useful in identifying genes that regulate susceptibility to the development and progression of metastasis and have helped to highlight potential novel targets for drug development. In particular several small animal imaging technologies including magnetic resonance imaging, high frequency ultrasound, and optical imaging have been recently applied to this task. Each of these methods may be useful for specific research projects, based on their unique combination of resolution, image acquisition time, animal throughput, and cost-effectiveness, yet none of these modalities adequately address the need for rapid quantification of tumors across the entire organism, nor do they assess therapeutic effectiveness in eradicating cancer in xenograft models. We have developed an Accelerator Mass Spectrometry (AMS)-based high precision quantitative method for assessing the metastatic potential of primary tumors isolated from newly diagnosed patients. Results: Our AMS-based methodology to study metastasis uses xenograft cancer cells labeled with 14C-labeled thymidine that are delivered intravenously into NSG mice and allowed to develop metastatic cancer over the course of up to 10 weeks. At the end of the experiment, all vital organs are collected; the DNA is isolated and is examined by AMS for the presence of 14C-signal. The labeling was optimized to achieve sufficient signal such that a tumor derived from a single cell could be detected by AMS, in secondary tumors, in vivo, independent of histological data. Conclusions: Using this approach we have determined that tissue colonization by tumor cells is a very rare event, where most metastatic tumors are initiated by less than 10 cells delivered into NSG mice. Further optimization of these techniques will allow us to explore the metastatic potential of primary tumors, isolated from biopsies and expanded in Avatar mice. This study was supported in part by NIH P41MI03483 and was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). IM number: LLNL-678306 Citation Format: Nicholas R. Hum, Kelly A. Martin, Michael Malfatti, Kurt Haack, Bruce A. Buchholz, Gabriela G. Loots. Tracking cancer colonization in xenografts using ultrasensitive accelerator mass spectrometry methods [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1977. doi:10.1158/1538-7445.AM2017-1977

Background: Bladder cancer is among the top ten most common cancers, with about ~380,000 new cases and ~150,000 deaths per year worldwide. Platinum-based combination chemotherapy is commonly used to treat advanced bladder cancer. It has been shown that only ~50% of the patients with advanced bladder cancer respond to platinum-based therapy. Methods: We have utilized a patient-derived bladder cancer xenograft (PDX) platform to characterize the molecular mechanisms that contribute to resistance of gemcitabine-cisplatin combination therapy in advanced bladder cancer. We have also identified key regulatory pathways in our PDX models that can be targeted to treat chemotherapy resistant bladder cancer using RNAseq analysis. Transcriptome profiling of P0 (passage 0) bladder cancer xenograft tumors from 4 PDX lines (2 gemcitabine-cisplatin resistant lines and 2 drug sensitive lines) was performed by RNA-Seq analysis, before and after a 21-day cisplatin/gemcitabine drug treatment regimen. Results: The RNA-seq data has indicated significant differences between the transcriptional profiles of drug-sensitive and drug-resistant tumors. PDXs retained morphology and shared 92-97% of genetic alterations of parental cancer cells. We identified 333 genes >2 fold up or down regulated in the drug resistant tumors compared to the drug sensitive tumors. Significantly up-regulated genes in drug resistant tumors analyzed include metabolic enzymes ALDH2, ALDH3A1, ALDH4A1 and ALDH7A1, transporter proteins ABCA1, SLC1A4, SLC2A5, SLC30A1, SLC39A6, SLC7A5 and SLC9A3, Notch ligand JAG2, Growth hormone receptor GHR and transmembrane protein GPNMB. Consistent with the change of cell surface proteins such as GHR and GPNMB, the MAPK and the PI3K-AKT pathways were upregulated when PDXs became resistant to cisplatin treatment. Additional changes in gene expression based on RNA-seq data before and after drug treatment were also found. Conclusion: Chemoresistance to gemcitabine and cisplatin is associated with altered expression of several cell surface proteins and upregulation of the downstream signaling pathways. Targeting these cell surface proteins can possibly be harnessed to overcome chemoresistance. GPNMB is a type I transmembrane protein that has previously been shown to be up-regulated in many metastatic cancers including breast cancer. Potentially targeting GPNMB with glembatumumab vedotin, an antibody-drug-conjugate has shown promising results in treating several cancers including breast cancer and osteosarcoma. Further studies will elucidate whether targeting GPNMB is an effective strategy for the treatment of chemotherapy resistant bladder cancer. This study received funding by a developmental grant from the UCDCC. This work was conducted under the auspices of the USDOE by LLNL (DE-AC52-07NA27344). IM number: LLNL-688318 Citation Format: Kelly A. Martin, Nicholas R. Hum, Aimy Sebastian, Deepa K. Murugesh, Chong-Xian Pan, Ai-Hong Ma, Ralph de Vere White, Gaby Loots. RNA sequencing of bladder cancer patient-derived xenograft models identifies genes associated with chemoresistance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 97. doi:10.1158/1538-7445.AM2017-97

Background: The inability to effectively treat metastases is the main reason for the limited progress in reducing the rates of cancer morbidity and mortality. One major drawback is the lack of quantitative assays for assessing the size and tissue prevalence of tumors in newly diagnosed individuals. Current methods for quantifying tumor burden are mainly qualitative and include measuring the gross weight of the affected organ, counting tumors on the surface of the organ, or evaluating a small sample of the organ using histologic sections. These methods are crude measures of tumor burden and size distribution, and in the case of histology, they are time consuming, difficult to process an adequate sample size and non-quantitative. Methods: Animal models of metastasis have been useful in identifying genes that regulate susceptibility to the development and progression of metastasis and helped highlight potential novel targets for drug development. In particular several small animal imaging technologies including magnetic resonance imaging, high frequency ultrasound, and optical imaging have been recently applied to this task. Each of these methods may be useful for specific research projects, based on their unique combination of resolution, image acquisition time, animal throughput, and cost-effectiveness, yet none of these modalities adequately address the need for rapid quantification of tumors across the entire organism, nor do they assess therapeutic effectiveness in eradicating cancer in xenograft models. We have developed an Accelerator Mass Spectrometry (AMS)-based high precision quantitative method for assessing the metastatic potential of primary tumors isolated from newly diagnosed patients. Results: Our Accelerator Mass Spectrometry-based methodology to study metastasis uses xenograft cancer cells labeled with 14C-labeled thymidine that are delivered intravenously into NSG mice and allowed to develop metastatic cancer over the course of up to 10 weeks. At the end of the experiment, all vital organs are collected; the DNA is isolated and is examined by AMS for the presence of 14C-signal. The labeling was optimized to achieve sufficient signal such that a tumor derived from a single cell could be detected by AMS, in secondary tumors, in vivo, independent of histological data. Conclusions: Using this novel approach we have evaluated the metastatic potential of several prostate cancer cell lines, characterized stem-cell like sublines derived from prostate cancer cell lines [PC3] and examined tissue tropism of cancer sublines derived from kidney and liver metastatic tumors. Further optimization of these techniques will allow us to explore the metastatic potential of primary tumors, isolated from biopsies and expanded in Avatar mice. Citation Format: Kelly A. Martin, Nicholas Hum, Kurt W. Haack, Bruce A. Buchholz, Gabriela G. Loots. Development of quantitative methods for assessing metastatic potential of human primary tumors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3891.

Background: Bladder cancer is among the ten most common cancers, with about ∼380,000 new cases and ∼150,000 deaths per year worldwide. Platinum-based combination chemotherapy is commonly used to treat advanced bladder cancer. It has been shown that only ∼50% of the patients with advanced bladder cancer respond to platinum-based therapy. Methods: We employed a patient-derived bladder cancer xenograft (PDX) platform to characterize the molecular mechanisms contributing to resistance to gemcitabine-cisplatin combination therapy in advanced bladder cancer and to identify novel candidates that can be targeted to treat chemotherapy resistant bladder cancer. Transcriptome profiling of P0 (passage 0) bladder cancer xenograft tumors from 4 PDX lines (2 gemcitabine-cisplatin resistant lines and 2 drug sensitive lines) was performed by RNA-Seq analysis. Results: PDXs retained the morphology fidelity and shared 92-97% of genetic alterations of parental cancer cells. The RNA-seq data suggested the presence of significant differences between the transcription profiles of drug-sensitive and drug-resistant tumors. We identified 333 genes >2 fold up or down regulated in the drug resistant tumors compared to the drug sensitive tumors. Genes down-regulated in drug resistant tumors include tight junction protein CLDN3 and regulators of G-protein signaling RGS2 and RGS3. Significantly up-regulated genes include metabolic enzymes ALDH2, ALDH3A1, ALDH4A1 and ALDH7A1, transporter proteins ABCA1, SLC1A4, SLC2A5, SLC30A1, SLC39A6, SLC7A5 and SLC9A3, Notch ligand JAG2, Growth hormone receptor GHR and transmembrane glycoprotein GPNMB. Consistent with the change of cell surface proteins such as GHR and GPNMB, the MAPK and the PI3K-AKT pathways were upregulated when PDXs became resistant to cisplatin treatment. Conclusion: Chemoresistance to gemcitabine and cisplatin is associated with altered expression of several cell surface proteins and upregulation of the downstream signaling pathways. Targeting these cell surface proteins can possibly be harnessed to overcome chemoresistance. GPNMB, a type I transmembrane protein highly up-regulated in the drug resistant tumors, has previously been shown to be over-expressed in various cancers. Targeting GPNMB with an antibody-drug-conjugate, glembatumumab vedotin, has shown promising results in treating several cancers including breast cancer and osteosarcoma. Further studies will elucidate whether targeting GPNMB is an effective strategy for the treatment of chemotherapy resistant bladder cancer. Citation Format: Aimy Sebastian, Kelly A. Martin, Chong-xian Pan, Ai-hong Ma, Ralph W. deVere White, Gabriela G. Loots. Transcriptome analysis of patient-derived bladder cancer xenografts identifies genes associated with chemoresistance. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2106.