Fibroblasts are thought to be key players in the tumor microenvironment. Means to identify and isolate fibroblasts as well as an understanding of their cancer-specific features are essential to dissect their role in tumor biology. To date, the identification of cancer-associated fibroblasts is widely based on generic markers for activated fibroblasts in combination with their origin in tumor tissue. This study was focused on a deep characterization of the cell surface marker profile of cancer-associated fibroblasts in widely used mouse tumor models and defining aberrant expression profiles by comparing them to their healthy counterparts. We established a generic workflow to isolate healthy and cancer-associated fibroblasts from solid tissues, thereby reducing bias, and background noise introduced by non-target cells. We identified CD87, CD44, CD49b, CD95, and Ly-6C as cancer-associated fibroblast cell surface markers, while CD39 was identified to mark normal fibroblasts from healthy tissues. In addition, we found a functional association of most cancer-related fibroblast markers to proliferation and a systemic upregulation of CD87, and CD49b in tumor-bearing mice, even in non-affected tissues. These novel markers will facilitate the characterization of fibroblasts and shed further light in their functions and implication in cancer progression.

Introduction Metastasis to different organs or tissues may require distinct sets of regulators which may influence the homing and growth of tumour cells to specific secondary sites. Under the hypothesis that the immune microenvironment of the different niches may play an important role in this process, we have categorised metastatic samples from different primary tumours based on their immune profile. Material and methods Gene expression data from metastatic samples with different primary tumour origin (n=342) were downloaded from open repository GEO. Samples were scored using different gene expression profiles and characterised on the basis of their immune and stromal infiltration and activation of immune response pathways (Immunophenoscore, MCPcounter, ESTIMATE; among others). Resultant scores were analysed for statistical differences with ANOVA test. Multivariate analysis was used for clustering the samples based in their immune-features. Results and discussions As expected, significant differences were found between the immune profiles of samples metastasizing in distinct organs. For instance, breast cancer metastasis in lung showed a much higher immunogenic score than breast metastasis in brain (p=5e-4), suggesting a different immune microenvironment modulation. Also in breast, significant differences have been found in cell lineages infiltration, lung metastasis being the ones with the highest T cell component (p=0.002) and liver metastasis the ones with the lowest infiltration of endothelial cells (p=0.005). Moreover, in other cancer types like melanoma, samples showed differences among different metastatic locations. Interestingly, when comparing metastatic samples originating from different primary tumour, a high concordance among secondary tumours in immune scores were found; specifically in brain metastasis. These results suggest that cells needs to share similar molecular profiles to evade the immune surveillance and growth in a specific niche, independently of their origin. Conclusion Metastases from the same primary tumour growing in different organs show differences in their immune profile. However, those samples from different primary origin but growing in the same secondary organ shared a characteristic immune profile. These results suggest that immune system plays a role in determining the organ-specific homing of metastasis. Introduction Syngeneic mouse tumour models are widely used to analyse tumour immunology due to their fully competent immune-repertoire and have paved the way for novel immunotherapy agents in multiple tumour entities. However, the amount and composition of tumour infiltrating leukocytes (TIL) is highly variable. This complicates targeted analysis, in particular for small leukocyte subpopulations that may not be analysed properly or lost in the background noise. When working with large cohort sizes, immune-phenotyping by flow cytometry is time consuming and highly work intensive. We have developed improved workflows combining automated tissue dissociation with novel TIL spe...

Cancer cell lines are used to study tumor biology and as models for testing of novel anti-cancer therapeutics. The vast majority of this work has been done using established cell lines that have been cultured for decades. Upon extensive in vitro propagation, cancer cell lines acquire multiple genetic and epigenetic alterations, lose initial heterogeneity present in the parental tumor, and tend to lack tumor initiating as well as multi-lineage differentiation capacity. Consequently, it has been shown that ‘classical' tumor cell lines only insufficiently resemble the characteristics of a tumor. In order to generate improved models for cancer research, novel cell lines are derived from primary tissues and used at low passages. However, this process is very inefficient for most tumor entities. In addition, most of the media used include largely undefined serum, such as FBS, which has been shown to drive primary tumor cell cultures to a more differentiated state when used over multiple passages. We have developed an advanced, serum-free medium for derivation and expansion of tumor cell lines from pancreatic tumors (Noll et al., 2016). This medium has been further optimized concerning formulation, stability, and usability. It allows for efficient generation of primary cell lines from both, patient and xenotransplanted tumors. Primary cell lines derived with this medium closely resembled essential characteristics of the parental tumor, including expression of subtype-specific markers, cellular heterogeneity, as well as genetic and epigenetic signatures. The derived cell lines have been used for genetic engineering and the developed medium allowed for the subsequent expansion of single-cell clones, a pre-requisite for reliable downstream assays. Importantly, culturing with our medium retained the tumorigenic potential as shown by xenotransplantation in immunodeficient mice. The resulting tumors closely and reliably resembled the initial patient tumor. This has been shown on the histomorphological as well as functional level, for example by conservation of tumor specific resistance mechanisms. Taken together, we have developed a serum-free medium for efficient derivation and expansion of tumor cell lines from primary and xenotransplanted pancreatic tumors, allowing for the establishment of easily accessible in vitro models. Moreover, the cell lines could also be used to generate corresponding xenograft models facilitating the translation of in vitro findings directly into a pre-clinical in vivo setting. Citation Format: David Agorku, Anne Langhammer, Elisa M. Noll, Christian Eisen, Silke Schult, Andreas Bosio, Martin R. Sprick, Andreas Trumpp, Olaf Hardt. Efficient derivation and expansion of tumor cell lines from primary and xenotransplanted pancreatic tumors [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 1050.

Immunotherapy against cancer has proven clinical efficacy and tremendous potential in multiple tumor entities. Syngeneic mouse tumor models represent the gold standard to analyze effects of immunotherapy due to their fully competent immune-repertoire. However, the amount and composition of tumor infiltrating leukocytes (TIL) is highly variable, complicating the targeted analysis of subpopulations. In particular, small subpopulations cannot be analyzed properly but may be lost in the background noise. When working with large cohort sizes, even the immune-phenotyping of TIL by flow cytometry is time consuming and data processing highly work intensive making pre-enrichment methods for sample debulking attractive. We have established an automated workflow combining tissue dissociation with TIL specific isolation to improve and accelerate downstream analysis. Tumor dissociation was automated using the gentleMACS™Octo Dissociator and optimized for epitope conservation to overcome bias in immune-phenotyping caused by dissociation with aggressive of impure enzymes. Next, isolation of TIL was improved by developing a new CD45-specific enrichment reagent for the magnetic cell sorting (MACS) based isolation directly from dissociated tumor tissue. The whole workflow takes only about 90 min. To validate this method on starting material showing variable frequencies of TIL infiltration, we used syngeneic mouse tumors derived by injection three independent tumor cell lines. Tumors derived from injection of B16.F10 melanoma showed TIL frequencies of 2-4%, CT26.WT colon carcinoma tumors 15-21%, and 4T1 breast carcinoma tumors 32-37% among total viable cells after dissociation. Using a manual separation system, TIL were enriched to purities above 90% at yields above 70% for CT26.WT and 4T1 tumors, and purities above 80% at high yields above 95% for B16.F10 tumors. To address the need of parallelization and automation for sample processing in large cohort sizes used in in vivo studies, a fully automated version of the MultiMACS™ Cell24 Separator was developed by integrating the instrument in a liquid handling robotic platform. This new system, the MultiMACS™ X, can process 24 cell separations in parallel with minimal hands-on time. When compared to the manual system, equal purities were achieved whereas the overall yield of target cells was increased from 70% to 90% as shown in the case of CT26.WT tumors. Importantly, while the TIL enrichment significantly reduced the time and reagent costs in immune subset analysis, the composition of infiltrating immune cells was not affected, excluding the risk of introducing a bias by this method. Taken together, we have developed an automated workflow for the isolation of TIL from mouse tumors reducing time and costs of downstream analysis while standardizing and enhancing the detection and quantification of immune cell subpopulations. Citation Format: Janina Brauner, David Agorku, Anne Langhammer, Lukas Pohlig, Jan Drewes, Thomas Dino Rockel, Oliver Schilling, Stephan Mehlhose, Wa'el Al Rawashdeh, Andreas Bosio, Olaf Hardt. Workflow automation and parallelization improves the isolation and analysis of tumor-infiltrating immune subpopulations [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 1672. doi:10.1158/1538-7445.AM2017-1672

No abstract

No abstract

Cancer cell lines are widely used as in vitro models to study tumor biology and for efficacy testing of novel anti-cancer therapeutics. In the past, most of this work has been done using established cell lines that have been cultured for decades. Extensive in vitro propagation has been shown to lead to the acquisition of multiple genetic and epigenetic alterations in cell lines, leading to decreased heterogeneity and the lack of tumor initiating as well as multi-lineage differentiation capacity. Consequently, established cell lines only insufficiently resemble the characteristics of tumors and thus have limited applicability as in vitro models for example. To overcome this hurdles, cell lines can be derived from primary cancer biopsies. However, this process is very inefficient for most tumor entities. In addition, most of the media used include largely undefined serum, such as FBS, which has been shown to drive primary tumor cell cultures to a more differentiated state when used over multiple passages. We have developed advanced, serum-free media for derivation and expansion of tumor cell lines from pancreatic, ovarian or renal tumors. Our media have been optimized concerning formulation, stability, and usability and allow for efficient generation of primary cell lines from both, patient and xenotransplanted tumors. Primary cell lines derived with our media retained their tumorigenic potential and could therefore be xenotransplanted and propagated in immunodeficient mice. Notably, the resulting tumors closely resembled the initial patient tumor as shown on the histomorphological as well as functional level. Moreover, the primary cell lines closely resembled essential characteristics of the parental tumor in vitro, including expression of subtype-specific markers, cellular heterogeneity, as well as genetic and epigenetic signatures. Taken together, we have developed serum-free medium for efficient derivation and expansion of tumor cell lines from primary and xenotransplanted pancreatic, ovarian and renal tumors, allowing for the establishment of easily accessible in vitro as well as corresponding xenografting vivo models. This facilitates the translation of in vitro findings directly into in vivo settings, allowing for more reliable pre-clinical modelling. Citation Format: Olaf Hardt, David Agorku, Anne Langhammer, Franziska Zickgraf, Felix Geist, Elisa M. Noll, Christian Eisen, Andreas Bosio, Martin R. Sprick, Andreas Trumpp. Efficient derivation and expansion of tumor cell lines from primary and xenotransplanted pancreatic, ovarian and renal tumors [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 3714.