Patient-Derived Xenografts for Prognostication and Personalized Treatment for Head and Neck Squamous Cell Carcinoma.

Overall survival remains very poor for patients diagnosed as having head and neck squamous cell carcinoma (HNSCC). Identification of additional biomarkers and novel therapeutic strategies are important for improving patient outcomes. Patient-derived xenografts (PDXs), generated by implanting fresh tumor tissue directly from patients into immunodeficient mice, recapitulate many of the features of their corresponding clinical cancers, including histopathological and molecular profiles. Using a large collection of PDX models of HNSCC, we demonstrate that rapid engraftment into immunocompromised mice is highly prognostic and show that genomic deregulation of the G1/S checkpoint pathway correlates with engraftment. Furthermore, CCND1 and CDKN2A genomic alterations are predictive of response to the CDK4and CDK6 inhibitor abemaciclib. Overall, our study supports the pursuit of CDK4 and CDK6 inhibitors as a therapeutic strategy for a substantial proportion of HNSCC patients and demonstrates the potential of using PDX models to identify targeted therapies that will benefit patients who have the poorest outcomes.

Proteomic Analysis of Cancer-Associated Fibroblasts Reveals a Paracrine Role for MFAP5 in Human Oral Tongue Squamous Cell Carcinoma.

Bidirectional communication between cells and their microenvironment is crucial for both normal tissue homeostasis and tumor growth. During the development of oral tongue squamous cell carcinoma (OTSCC), cancer-associated fibroblasts (CAFs) create a supporting niche by maintaining a bidirectional crosstalk with cancer cells, mediated by classically secreted factors and various nanometer-sized vesicles, termed as extracellular vesicles (EVs). To better understand the role of CAFs within the tumor stroma and elucidate the mechanism by which secreted proteins contribute to OTSCC progression, we isolated and characterized patient-derived CAFs from resected tumors with matched adjacent tissue fibroblasts (AFs). Our strategy employed shotgun proteomics to comprehensively characterize the proteomes of these matched fibroblast populations. Our goals were to identify CAF-secreted factors (EVs and soluble) that can functionally modulate OTSCC cells in vitro and to identify novel CAF-associated biomarkers. Comprehensive proteomic analysis identified 4247 proteins, the most detailed description of a pro-tumorigenic stroma to date. We demonstrated functional effects of CAF secretomes (EVs and conditioned media) on OTSCC cell growth and migration. Comparative proteomics identified novel proteins associated with a CAF-like state. Specifically, MFAP5, a protein component of extracellular microfibrils, was enriched in CAF secretomes. Using in vitro assays, we demonstrated that MFAP5 activated OTSCC cell growth and migration via activation of MAPK and AKT pathways. Using a tissue microarray of richly annotated primary human OTSCCs, we demonstrated an association of MFAP5 expression with patient survival. In summary, our proteomics data of patient-derived stromal fibroblasts provide a useful resource for future mechanistic and biomarker studies.

Cell surface profiling using high-throughput flow cytometry: a platform for biomarker discovery and analysis of cellular heterogeneity.

Cell surface proteins have a wide range of biological functions, and are often used as lineage-specific markers. Antibodies that recognize cell surface antigens are widely used as research tools, diagnostic markers, and even therapeutic agents. The ability to obtain broad cell surface protein profiles would thus be of great value in a wide range of fields. There are however currently few available methods for high-throughput analysis of large numbers of cell surface proteins. We describe here a high-throughput flow cytometry (HT-FC) platform for rapid analysis of 363 cell surface antigens. Here we demonstrate that HT-FC provides reproducible results, and use the platform to identify cell surface antigens that are influenced by common cell preparation methods. We show that multiple populations within complex samples such as primary tumors can be simultaneously analyzed by co-staining of cells with lineage-specific antibodies, allowing unprecedented depth of analysis of heterogeneous cell populations. Furthermore, standard informatics methods can be used to visualize, cluster and downsample HT-FC data to reveal novel signatures and biomarkers. We show that the cell surface profile provides sufficient molecular information to classify samples from different cancers and tissue types into biologically relevant clusters using unsupervised hierarchical clustering. Finally, we describe the identification of a candidate lineage marker and its subsequent validation. In summary, HT-FC combines the advantages of a high-throughput screen with a detection method that is sensitive, quantitative, highly reproducible, and allows in-depth analysis of heterogeneous samples. The use of commercially available antibodies means that high quality reagents are immediately available for follow-up studies. HT-FC has a wide range of applications, including biomarker discovery, molecular classification of cancers, or identification of novel lineage specific or stem cell markers.

Tumor-initiating cells are rare in many human tumors.

Tumor-initiating cells (TICs) are defined by their ability to form tumors after xenotransplantation in immunodeficient mice and appear to be relatively rare in most human cancers. Recent data in melanoma indicate that the frequency of TICs increases dramatically via more permissive xenotransplantation conditions, raising the possibility that the true frequency of TICs has been greatly underestimated in most human tumors. We compared the growth of human pancreatic, non-small cell lung, and head and neck carcinomas in NOD/SCID and NSG mice. Although TIC frequency was detected up to 10-fold higher in NSG mice, it remained low (<1 in 2500 cells) in all cases. Moreover, aldehyde dehydrogenase-positive (ALDH(+)) and CD44(+)CD24(+) cells, phenotypically distinct cells enriched in TICs, were equally tumorigenic in NOD/SCID and NSG mice. Our findings demonstrate that TICs are rare in these cancers and that the identification of TICs and their frequency in other human malignancies should be validated via primary tumors and highly permissive xenotransplantation conditions.