Eileen Snowden

Immunophenotyping and Transcriptomic Outcomes in PDX-derived TNBC Tissue

Cancer tissue functions as an ecosystem of a diverse set of cells that interact in a complex tumor microenvironment. Genomic tools applied to biopsies in bulk fail to account for this tumor heterogeneity, whereas single-cell imaging methods limit the number of cells which can be assessed or are very resource intensive. The current study presents methods based on flow cytometric analysis and cell sorting using known cell surface markers (CXCR4/CD184, CD24, THY1/CD90) to identify and interrogate distinct groups of cells in triple-negative breast cancer clinical biopsy specimens from patient-derived xenograft (PDX) models. The results demonstrate that flow cytometric analysis allows a relevant subgrouping of cancer tissue and that sorting of these subgroups provides insights into cancer cell populations with unique, reproducible, and functionally divergent gene expression profiles. The discovery of a drug resistance signature implies that uncovering the functional interaction between these populations will lead to deeper understanding of cancer progression and drug response. Implications: PDX-derived human breast cancer tissue was investigated at the single-cell level, and cell subpopulations defined by surface markers were identified which suggest specific roles for distinct cellular compartments within a solid tumor. Mol Cancer Res; 15(4); 429–38. ©2016 AACR.

Abstract 1656: Comprehensive evaluation of human immune system reconstitution in NSGTMand NSGTM-SGM3 toward the development of a novel Onco-HuTMxenograft model

The recent successes of immunotherapeutic approaches to the treatment of melanoma and the promise of similar treatments in a variety of other cancers underscore the importance of the immune system in cancer. Indeed, effective therapeutic design and evaluation require a comprehensive understanding of the interplay between the immune compartment and the proliferating tumor cells that comprise the tumor microenvironment. A humanized mouse strain engrafted with cancerous tissue from a patient derived xenograft (PDX) tumor provides researchers with a highly sophisticated tool, ideally suited to facilitate the design of treatment strategies that prevent tumor evasion of immune cells and that improve cytotoxic responses. Severely combined immunodeficient mice such as NOD scid gamma (NSGTM) and triple transgenic NSG mice expressing human cytokines KITLG, CSF2, and IL-3 (NSGTM-SGM3) are proven hosts for the engraftment of human tumors and establishment of human immune system components following hematopoietic stem cell (CD34+) transplantation. The endogenous expression of cytokines that support the development of myeloid lineages and regulatory T cells potentially represents a substantial improvement over standard NSG mice. Here we employ four 14-color flow cytometry panels to perform a comprehensive and detailed analysis of the entire immune system. The four panels are designed to fully characterize specific branches of the immune system: 1) T cells 2) NK cells/dendritic cells/B cells 3) myeloid lineages, and 4) immune checkpoint markers. Blood, spleen, and bone marrow tissue from both NSG and NSG-SGM3 mice were evaluated at 9, 16, 21, and 31 weeks of age using each of the four phenotyping panels. Our results indicate that the triple transgenic NSG-SGM3 mice exhibit a more completely humanized immune system compared to NSG mice, with specific improvements in the distribution of T-cell subsets and overall representation of the myeloid lineage. NSG mice engrafted with allogeneic human tumors represent a valuable preclinical testing platform for immuno-oncology. Citation Format: Aaron J. Middlebrook, Eileen Snowden, Warren Porter, Friedrich Hahn, Mitchell Ferguson, Brian Soper, James Keck, Joan Malcolm, Shannon Dillmore, Smita Ghanekar, Rainer Blaesius. Comprehensive evaluation of human immune system reconstitution in NSGTM and NSGTM-SGM3 toward the development of a novel Onco-HuTM xenograft model [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 1656. doi:10.1158/1538-7445.AM2017-1656

Abstract 2391: Flow cytometric sorting of subpopulations followed by RNASeq reveals distinct phenotypes in PDX model of basal breast cancer

The recognition of tumor tissue as an interactive ecosystem of distinct cell types has recently emerged as a very promising basis for more successful treatment of cancer patients. While intratumor heterogeneity (ITH) as a phenomenon has been known for decades it is only recent that its functional significance can be investigated with effective tools. Starting with a correlation of cellular heterogeneity with aggressiveness, metastatic potential and drug susceptibility of a cancerous lesion the current focus on functional differences between various tumor cell compartments is revealing a number of distinct functions and interactions. The discovery of communication between various tumor cells through soluble factors as well as differences in implantation of homogeneous vs. heterogeneous cell populations into immune compromised mouse models suggest a “division of labor” among cell types within many tumor tissues. We have investigated the surface marker distribution of more than 8 different PDX tumor models by flow cytometry and detected extensive immunophenotypic heterogeneity. Using a range of markers associated with cancer stem cells, EMT and invasiveness (e.g. CD 24, 44, 133, 184, 326 (EpCAM), and CD45) we find heterogeneity with respect to many surface markers as well as individual immunophenotypic signatures for each model. Building on this characterization we chose several markers for sorting of subpopulations and performed gene expression analysis. Transcriptome analysis of a breast cancer model revealed two phenotypic signatures which suggest a strong proliferative population alongside a second population which is far less proliferative but much more active in angiogenesis, ECM organization and secretion of various soluble factors. This observation suggests a very clear example of distinct roles of multiple cell types to form a tumor tissue. Our findings could have implications for therapeutic strategies directed at more than one cell type as well as development of better diagnostic tools which take into account the presence of various phenotypically distinct cell populations. Citation Format: Warren Porter, Friedrich Hahn, Eileen Snowden, Mitchell Ferguson, Frances Tong, Shannon Dillmore, Joel S. Parker, Aaron Middlebrook, Smita Ghanekar, Rainer Blaesius. Flow cytometric sorting of subpopulations followed by RNASeq reveals distinct phenotypes in PDX model of basal breast cancer. [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 2391.

Abstract A38: Optimization of whole-genome amplification for analysis of single cells using next-generation sequencing

The accurate analysis of both genotypic variation in tumors and tumor evolution due to selective pressure is important in determining treatment options in the clinic. Next generation sequencing (NGS) of single cells from solid tumors and circulating tumor cells has the potential to dramatically increase the amount of genotypic detail obtained from these samples. For the analysis of single cells the challenge is to employ whole genome amplification (WGA) to obtain enough DNA for analysis while minimizing amplification bias and maintaining copy accuracy. We are developing tools for single cell sequencing to improve these parameters while also increasing sample throughput. Initially, three commercially available WGA methods were tested on 5-10 genome quantities of HCT-15 genomic DNA. Using the Ion Torrent AmpliSeq Cancer Hotspot Panel, the amplification accuracy, amplicon coverage and DNA yield were determined. When compared to controls significant differences were observed in the three WGA methods ability to accurately amplify DNA. The most accurate kit produced a specificity and sensitivity of 98.5% and 100% respectively. This protocol also produced a substantial DNA yield of 25 µgs and although, the average amplicon coverage indicated some bias, greater than 99% of the amplicons had at least 100-fold coverage. This WGA kit was then tested on single cells from a human breast cancer; an ER/PR/Her2 negative spindle cell metaplastic carcinoma. To obtain single cells the tumor was grown in a PDX mouse model and index sorted using a BD FACSAria II flow cytometer. The WGA produced an average DNA yield of 2.7± 0.8 µg, with 100-fold amplicon coverage of 92±7%. For the variants identified, the amplification bias differed by as much as 3,000 reads/variant. We are now taking steps to improve WGA of single cells and will report on these efforts in detail. To date when monitoring variant allele frequency and coverage depth in the triple negative tumor, KRAS and MET mutations were identified in single cells, as well as in the bulk sample. Interestingly, a TP53 mutation was identified only in single cells. This data supports the utility of WGA and single cell analysis to identify mutations, some having clinically relevant role for targeted therapy. Citation Format: Terry J. Amiss, Frances Tong, Eileen Snowden, Richard Kelly, Rainer Blaesius, Nick Herrmann, Friedrich Hahn, Warren Porter, Mitchell Ferguson, Chen Chang, Daphne Clancy, Stewart Jurgensen. Optimization of whole-genome amplification for analysis of single cells using next-generation sequencing. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Drug Sensitivity and Resistance: Improving Cancer Therapy; Jun 18-21, 2014; Orlando, FL. Philadelphia (PA): AACR; Clin Cancer Res 2015;21(4 Suppl): Abstract nr A38.

Abstract 227: Characterization of single cells from dissociated solid tumors

The heterogeneous nature of solid tumors, coupled with the relatively small sample size of available biopsies, has led to an emerging need to glean as much information as possible from these valuable specimens. Current approaches to solid tumor analysis fail to completely reveal the diverse range of cellular compartments that comprise the tumor microenvironment. A comprehensive approach to tumor interrogation requires efficient tissue dissociation to facilitate analysis at the single-cell level. In contrast to current methods, single-cell analysis of tumor derived cell suspensions by flow cytometry has the potential to provide a more complete understanding of the many subpopulations within the tumor microenvironment and the cell to cell interactions that govern this space. Here we demonstrate an efficient workflow that enables comprehensive single-cell analysis of solid tumors from breast cancers. Using tumors from clinical samples and mouse models, we evaluated different dissociation and processing techniques for their effects on cellular viability and surface marker expression. Solid tumors were dissociated into single-cell suspensions using a combination of mechanical dissociation and enzymatic digestion. Phenotypic distribution and morphology of cells within the tumor microenvironment was evaluated using flow cytometry. As this approach evolves, and a knowledge base of relevant surface markers is established, this technology has the potential to significantly impact how tumor biopsies are processed to get multiparametric information at a single cell level. Citation Format: Aaron J. Middlebrook, Shahryar Niknam, Joyce Ruitenberg, Albert J. Mach, Maria Suni, Warren Porter, Friedrich Hahn, Eileen Snowden, Rainer Blaesius, Smita Ghanekar. Characterization of single cells from dissociated solid tumors. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 227. doi:10.1158/1538-7445.AM2015-227

Abstract 2009: Flow cytometric analysis, sorting and molecular analysis of dissociated cells from human solid tumors derived from PDX mouse models

Functional and genetic heterogeneity in tumor tissue has been a well described phenomenon for many decades but only recently emerged as a potentially crucial contributor to cancer development and progression. The correlation between cellular heterogeneity and aggressiveness, metastatic potential and drug susceptibility of a cancerous lesion have led to models in which the existence of multiple clonal cell lineages is a central feature enabling a neoplastic lesion to overcome selective pressures caused by the surrounding tissues’ defensive capabilities as well as therapeutic interventions. In addition, the role of the tumor microenvironment as an integral part of tumorigenesis was recognized and infiltrating leukocytes or tumor associated fibroblasts are no longer viewed as mere contaminants of a solid tumor biopsy. The emerging picture is compared to macroscopic ecosystems and a detailed understanding of the interactions between numerous cell subgroups seems necessary for the complete understanding of cancer pathogenesis. Scarcity of appropriate tools and model systems are an obstacle to the investigation of this heterogeneity at a molecular level but advances over the last few years have led to a significant acceleration in this field. More sensitive and far cheaper methods for collection of genomic and transcriptomic data have revealed a complex picture of the evolution of individual solid tumors. To turn this deeper understanding of tumorigenesis into improved clinical outcomes, routine methods are required to separate complex tumors into subpopulations. This stratification will provide a more comprehensive characterization of the tumor and enable more detailed prediction of disease progression and resistance development. We have developed an integrated workflow for dissociation and flow cytometric analysis and sorting for multiple downstream analysis modalities. Using patient derived xenograft (PDX) mouse models derived from primary human breast cancer biopsies we have demonstrated the ability to identify distinct immunophenotypes for each model and use this analysis to isolate distinct subpopulations. Our successful optimization of a variety of well characterized surface markers (e.g. CD 24, 44, 133, 184, 326 (EpCAM), and CD45) provides a basis for effective fingerprinting of cancer cells from a variety of sources. In an effort to demonstrate the potential of FACS sorting of solid tumor derived cell populations we have interrogated sorted fractions by NGS as well as RT-PCR array analysis and show distinct genotypic as well as gene expression signatures for each subgroup. The evidence provided by our data suggests that the single cell focused approach flow cytometry has traditionally enabled in hematological cancers is accessible for solid tumors as well and may unlock valuable biological insights. Citation Format: Rainer Blaesius, Friedrich Hahn, Eileen Snowden, Warren Porter, Mitchell Ferguson, Frances Tong, Stewart Jurgensen, Chang Chen, Daphne Clancy, Jamal Sirriyah, John Alianti, Perry Haaland, Shannon Dillmore, Jeff Baker, Aaron Middlebrook, Joyce Ruitenberg, Maria Suni, Smita Ghanekar. Flow cytometric analysis, sorting and molecular analysis of dissociated cells from human solid tumors derived from PDX mouse models. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2009. doi:10.1158/1538-7445.AM2015-2009

Abstract B37: Phenotypic analysis of single cells dissociated from solid tumors

Current methods for solid tumor analysis may be inadequate for addressing both the heterogeneity of tumor cells and the components of the tumor microenvironment. Comprehensive analysis requires the ability to analyze tumor biopsy specimens at the single-cell level. A better understanding of how the tumors are behaving at a cellular level, in relation to the other cells within the tumor microenvironment, could lead to more accurate and specific treatment, and better patient prognosis. Evaluation of tumor derived single cells by flow cytometry could provide unique information that is not readily obtained from current methods. Here we demonstrate the single-cell analysis of solid tumors from breast and colorectal cancers. Since single-cell analysis of solid tumors may not occur at the collection site, it is important that tumors be preserved in order to retain their characteristics during transport. Using tumors from PDX mouse models, and human samples, different preservatives were evaluated for their effects on cellular viability and surface marker expression. Following shipment of the tumor samples in preservation solutions, the solid tumors were dissociated into single-cell suspensions using enzyme cocktails containing collagenase. Phenotypic evaluation was performed using flow cytometry after staining the single cells with monoclonal antibody panels specific for either tumor or immune cells. The results indicate that the dissociation method did not seem to adversely impact the expression of surface proteins. We demonstrate that it is feasible to analyze dissociated tumor cell populations with relevant surface markers. Analysis of the data revealed distinct phenotype patterns for single cells dissociated from breast and colorectal cancers. Further extensive evaluation of heterogeneity in these tumor types could reveal phenotypic signatures that may be clinically relevant. Citation Format: Joyce J. Ruitenberg, Aaron Middlebrook, Maria Suni, Friedrich Hahn, Eileen Snowden, Warren Porter, Mitchell Ferguson, Rainer Blaesius, Smita A. Ghanekar. Phenotypic analysis of single cells dissociated from solid tumors. [abstract]. In: Abstracts: AACR Special Conference on Cellular Heterogeneity in the Tumor Microenvironment; 2014 Feb 26-Mar 1; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(1 Suppl):Abstract nr B37. doi:10.1158/1538-7445.CHTME14-B37

Abstract A198: Flow cytometric analysis and sorting of dissociated cells from human solid tumors derived from PDX mouse models.

Functional and genetic heterogeneity in tumor tissue was first observed over 50 years ago. Today, tumor heterogeneity is frequently evoked in describing the pathway from pre-cancerous lesions to aggressive, metastatic cancer. During this progression, multiple clonal lineages are thought to arise, leading to subpopulations of the tumor showing different metastatic profiles and susceptibility to anti-cancer therapy. In addition, the role of the tumor microenvironment became recognized and infiltrating leukocytes or tumor associated fibroblasts are no longer viewed as mere contaminants of a solid tumor biopsy. The emerging picture is increasingly compared to macroscopic ecosystems and a detailed understanding of the interactions between numerous cell subgroups seems necessary for the complete understanding of cancer pathogenesis. Scarcity of appropriate tools and model systems are an obstacle to the investigation of this heterogeneity at a molecular level but advances over the last few years have led to a significant acceleration in this field. More sensitive and far cheaper methods for collection of genomic and transcriptomic data have revealed a complex picture of the evolution of individual solid tumors. To turn this deeper understanding of tumorigenesis into improved clinical outcomes, routine methods are required to separate complex tumors into subpopulations. This stratification will provide a more comprehensive characterization of the tumor and enable more detailed prediction of disease progression and resistance development. We have developed dissociation methods for solid tumor tissue which allows flow cytometric analysis as well as sorting to provide cells for multiple downstream analysis modalities. Using patient derived xenograft (PDX) mouse models derived from primary human breast, colorectal and lung cancer biopsies we have demonstrated efficient dissociation, surface marker analysis and nucleic acid purification from sorted populations. Conditions have been optimized for a range of relevant surface markers (e.g. CD 24, 44, 133, 184, 326 (EpCAM), and CD45) which are suitable to identify cells predicted to have stem cell, endothelial, epithelial or immune cell functions, respectively. Through sequencing of subpopulations identified by their phenotype we have demonstrated the compatibility of our workflow with downstream analysis methods such as Next Generation Sequencing (NGS). Our RNA stability measurements suggest that gene expression analysis is equally feasible. Our data provide a standardized basis for in depth investigation of subpopulations of cells from solid tumors with various molecular techniques. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A198. Citation Format: Rainer Blaesius, Friedrich Hahn, Eileen Snowden, Warren Porter, Mitchell Ferguson, Tina Marmura, Frances Tong, Shannon Dillmore, Aaron Middlebrook, Joyce Ruitenberg, Maria Suni, Smita Ghanekar. Flow cytometric analysis and sorting of dissociated cells from human solid tumors derived from PDX mouse models. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A198.