Segmentation of cellular ultrastructures on sparsely labeled 3D electron microscopy images using deep learning

Focused ion beam-scanning electron microscopy (FIB-SEM) images can provide a detailed view of the cellular ultrastructure of tumor cells. A deeper understanding of their organization and interactions can shed light on cancer mechanisms and progression. However, the bottleneck in the analysis is the delineation of the cellular structures to enable quantitative measurements and analysis. We mitigated this limitation using deep learning to segment cells and subcellular ultrastructure in 3D FIB-SEM images of tumor biopsies obtained from patients with metastatic breast and pancreatic cancers. The ultrastructures, such as nuclei, nucleoli, mitochondria, endosomes, and lysosomes, are relatively better defined than their surroundings and can be segmented with high accuracy using a neural network trained with sparse manual labels. Cell segmentation, on the other hand, is much more challenging due to the lack of clear boundaries separating cells in the tissue. We adopted a multi-pronged approach combining detection, boundary propagation, and tracking for cell segmentation. Specifically, a neural network was employed to detect the intracellular space; optical flow was used to propagate cell boundaries across the z-stack from the nearest ground truth image in order to facilitate the separation of individual cells; finally, the filopodium-like protrusions were tracked to the main cells by calculating the intersection over union measure for all regions detected in consecutive images along z-stack and connecting regions with maximum overlap. The proposed cell segmentation methodology resulted in an average Dice score of 0.93. For nuclei, nucleoli, and mitochondria, the segmentation achieved Dice scores of 0.99, 0.98, and 0.86, respectively. The segmentation of FIB-SEM images will enable interpretative rendering and provide quantitative image features to be associated with relevant clinical variables.

Efficient semi-supervised semantic segmentation of electron microscopy cancer images with sparse annotations

Electron microscopy (EM) enables imaging at a resolution of nanometers and can shed light on how cancer evolves to develop resistance to therapy. Acquiring these images has become a routine task.However, analyzing them is now a bottleneck, as manual structure identification is very time-consuming and can take up to several months for a single sample. Deep learning approaches offer a suitable solution to speed up the analysis. In this work, we present a study of several state-of-the-art deep learning models for the task of segmenting nuclei and nucleoli in volumes from tumor biopsies. We compared previous results obtained with the ResUNet architecture to the more recent UNet++, FracTALResNet, SenFormer, and CEECNet models. In addition, we explored the utilization of unlabeled images through semi-supervised learning with Cross Pseudo Supervision. We have trained and evaluated all of the models on sparse manual labels from three fully annotated in-house datasets that we have made available on demand, demonstrating improvements in terms of 3D Dice score. From the analysis of these results, we drew conclusions on the relative gains of using more complex models, and semi-supervised learning as well as the next steps for the mitigation of the manual segmentation bottleneck.

Efficient semi-supervised semantic segmentation of electron microscopy cancer images with sparse annotations

Electron microscopy (EM) enables imaging at nanometer resolution and can shed light on how cancer evolves to develop resistance to therapy. Acquiring these images has become a routine task; however, analyzing them is now the bottleneck, as manual structure identification is very time-consuming and can take up to several months for a single sample. Deep learning approaches offer a suitable solution to speed up the analysis. In this work, we present a study of several state-of-the-art deep learning models for the task of segmenting nuclei and nucleoli in volumes from tumor biopsies. We compared previous results obtained with the ResUNet architecture to the more recent UNet++, FracTALResNet, SenFormer, and CEECNet models. In addition, we explored the utilization of unlabeled images through semi-supervised learning with Cross Pseudo Supervision. We have trained and evaluated all of the models on sparse manual labels from three fully annotated in-house datasets that we have made available on demand, demonstrating improvements in terms of 3D Dice score. From the analysis of these results, we drew conclusions on the relative gains of using more complex models, semi-supervised learning as well as next steps for the mitigation of the manual segmentation bottleneck.

Automated large volume sample preparation for vEM

New developments in electron microscopy technology, improved efficiency of detectors, and artificial intelligence applications for data analysis over the past decade have increased the use of volume electron microscopy (vEM) in the life sciences field. Moreover, sample preparation methods are continuously being modified by investigators to improve final sample quality, increase electron density, combine imaging technologies, and minimize the introduction of artifacts into specimens under study. There are a variety of conventional bench protocols that a researcher can utilize, though most of these protocols require several days. In this work, we describe the utilization of an automated specimen processor, the mPrep™ ASP-2000™, to prepare samples for vEM that are compatible with focused ion beam scanning electron microscopy (FIB-SEM), serial block face scanning electron microscopy (SBF-SEM), and array tomography (AT). The protocols described here aimed for methods that are completed in a much shorter period of time while minimizing the exposure of the operator to hazardous and toxic chemicals and improving the reproducibility of the specimens' heavy metal staining, all without compromising the quality of the data acquired using backscattered electrons during SEM imaging. As a control, we have included a widely used sample bench protocol and have utilized it as a comparator for image quality analysis, both qualitatively and using image quality analysis metrics.

SEG: Segmentation Evaluation in absence of Ground truth labels

Identifying individual cells or nuclei is often the first step in the analysis of multiplex tissue imaging (MTI) data. Recent efforts to produce plug-and-play, end-to-end MTI analysis tools such as MCMICRO1- though groundbreaking in their usability and extensibility - are often unable to provide users guidance regarding the most appropriate models for their segmentation task among an endless proliferation of novel segmentation methods. Unfortunately, evaluating segmentation results on a user's dataset without ground truth labels is either purely subjective or eventually amounts to the task of performing the original, time-intensive annotation. As a consequence, researchers rely on models pre-trained on other large datasets for their unique tasks. Here, we propose a methodological approach for evaluating MTI nuclei segmentation methods in absence of ground truth labels by scoring relatively to a larger ensemble of segmentations. To avoid potential sensitivity to collective bias from the ensemble approach, we refine the ensemble via weighted average across segmentation methods, which we derive from a systematic model ablation study. First, we demonstrate a proof-of-concept and the feasibility of the proposed approach to evaluate segmentation performance in a small dataset with ground truth annotation. To validate the ensemble and demonstrate the importance of our method-specific weighting, we compare the ensemble's detection and pixel-level predictions - derived without supervision - with the data's ground truth labels. Second, we apply the methodology to an unlabeled larger tissue microarray (TMA) dataset, which includes a diverse set of breast cancer phenotypes, and provides decision guidelines for the general user to more easily choose the most suitable segmentation methods for their own dataset by systematically evaluating the performance of individual segmentation approaches in the entire dataset.

Endoplasmic reticulum stress and lipids in health and diseases

The endoplasmic reticulum (ER) is a complex and dynamic organelle that regulates many cellular pathways, including protein synthesis, protein quality control, and lipid synthesis. When one or multiple ER roles are dysregulated and saturated, the ER enters a stress state, which, in turn, activates the highly conserved unfolded protein response (UPR). By sensing the accumulation of unfolded proteins or lipid bilayer stress (LBS) at the ER, the UPR triggers pathways to restore ER homeostasis and eventually induces apoptosis if the stress remains unresolved. In recent years, it has emerged that the UPR works intimately with other cellular pathways to maintain lipid homeostasis at the ER, and so does at cellular levels. Lipid distribution, along with lipid anabolism and catabolism, are tightly regulated, in part, by the ER. Dysfunctional and overwhelmed lipid-related pathways, independently or in combination with ER stress, can have reciprocal effects on other cellular functions, contributing to the development of diseases. In this review, we summarize the current understanding of the UPR in response to proteotoxic stress and LBS and the breadth of the functions mitigated by the UPR in different tissues and in the context of diseases.

Neuronal IRE-1 coordinates an organism-wide cold stress response by regulating fat metabolism

Cold affects many aspects of biology, medicine, agriculture, and industry. Here, we identify a conserved endoplasmic reticulum (ER) stress response, distinct from the canonical unfolded protein response, that maintains lipid homeostasis during extreme cold. We establish that the ER stress sensor IRE-1 is critical for resistance to extreme cold and activated by cold temperature. Specifically, neuronal IRE-1 signals through JNK-1 and neuropeptide signaling to regulate lipid composition within the animal. This cold-response pathway can be bypassed by dietary supplementation with unsaturated fatty acids. Altogether, our findings define an ER-centric conserved organism-wide cold stress response, consisting of molecular neuronal sensors, effectors, and signaling moieties, which control adaptation to cold conditions in the organism. Better understanding of the molecular basis of this stress response is crucial for the optimal use of cold conditions on live organisms and manipulation of lipid saturation homeostasis, which is perturbed in human pathologies.

A new combinatorial megaplasmid library assembly method designed to screen for minimal pathways by using SCRaMbLE

Human proteins expressed in yeast are common to enhance protein production while the expression of functional human pathways remain challenging. Here, we propose a simple and economical high-throughput gene assembly method to create a yeast megaplasmid library from human cDNA to screen for minimal human functional pathways. We introduced artificial promoters followed by symmetric loxP sites into the megaplasmids using Golden Gate assembly coupled with streptavidin-bead-based purification. The isolated high molecular weight, randomly assembled cDNA megaplasmid library may be useful for high-throughput directed evolution experiments and may be adapted for use in other model organisms.

Human PERK rescues unfolded protein response-deficient yeast cells

Protein folding and quality control is tightly regulated at the endoplasmic reticulum (ER), and its disruption is associated with many diseases. In eukaryotes, the accumulation of unfolded protein in the ER is sensed by the three sensors, IRE1, PERK, and ATF6 to activate the unfolded protein response (UPR) to restore ER homeostasis. However, uncoupling the sensing of each sensor and their respective downstream pathways has been challenging as the absence of one is compensated by the remaining two sensors. Here, we report a fully functional human PERK (hPERK) chimeric protein expressed in Saccharomyces cerevisiae that could be used for high throughput screen to identify new PERK inhibitory or activating compounds as well as to characterize the PERK stress sensing mechanisms.

An omic and multidimensional spatial atlas from serial biopsies of an evolving metastatic breast cancer

Mechanisms of therapeutic resistance and vulnerability evolve in metastatic cancers as tumor cells and extrinsic microenvironmental influences change during treatment. To support the development of methods for identifying these mechanisms in individual people, here we present an omic and multidimensional spatial (OMS) atlas generated from four serial biopsies of an individual with metastatic breast cancer during 3.5 years of therapy. This resource links detailed, longitudinal clinical metadata that includes treatment times and doses, anatomic imaging, and blood-based response measurements to clinical and exploratory analyses, which includes comprehensive DNA, RNA, and protein profiles; images of multiplexed immunostaining; and 2- and 3-dimensional scanning electron micrographs. These data report aspects of heterogeneity and evolution of the cancer genome, signaling pathways, immune microenvironment, cellular composition and organization, and ultrastructure. We present illustrative examples of how integrative analyses of these data reveal potential mechanisms of response and resistance and suggest novel therapeutic vulnerabilities.

A high-throughput genetic screening protocol to measure lipid bilayer stress-induced unfolded protein response in Saccharomyces cerevisiae

The endoplasmic reticulum (ER) stress is defined by the accumulation of unfolded proteins at the ER and perturbation at the ER membrane, known as lipid bilayer stress (LBS). In turn, ER stress triggers the unfolded protein response (UPR) to restore ER homeostasis. Here, we provide a modified protocol based on the synthetic genetic array analysis in Saccharomyces cerevisiae to identify genetic perturbations that induce the UPR by LBS. This method is adaptable to other canonical stress pathways.

For complete details on the use and execution of this protocol, please refer to Ho et al. (2020), Jonikas et al. (2009) and Baryshnikova et al. (2010).

•

Generation and validation of IRE1 and IRE1ΔLD query strains with a UPR reporter

•

Detailed protocol of query strains mated to the yeast deletion library using SGA

•

High-throughput measurement of reporter fluorescence levels by flow cytometry

•

Data analysis to identify gene deletions activating the UPR by lipid bilayer stress

Abstract PO-014: VISTA: VIsual Semantic Tissue Analysis for pancreatic disease quantification in murine cohorts

Objective and quantifiable assessment of tissue pathology is necessary to study mechanistic disease progression; however, current quantification methods based on tissue staining have many drawbacks including cost, time, labor, batch effects, as well as uneven staining which can result in misinterpretation and investigator bias. Here we present VISTA, an automated deep learning tool for semantic segmentation and quantification of histologic features from hematoxylin and eosin (H&E) stained pancreatic tissue sections. VISTA is trained to identify four key tissue types in developing murine PDAC samples: normal acinar, acinar-to-ductal metaplasia (ADM), dysplasia, and other normal tissue. Predicted segmentations were quantitatively evaluated against pathologist annotation with Dice Coefficients, achieving scores of 0.79, 0.70, 0.79 for normal acinar, ADM, and dysplasia, respectively. Predictions were evaluated against biological ground truth using the mean structural similarity index to immunostainings amylase and pan-keratin (0.925 and 0.920, respectively). The total area of feature prediction was also correlated to the area of immunostaining in whole tissue sections using spearman correlation (0.86, 0.97, and 0.92 for DAPI, amylase, and cytokeratins, respectively). Importantly, our tool is not only able to predict staining information, but it is able to distinguish between ADM and dysplasia, which are not reliably distinguished with common immunostaining methods, showing VISTA’s potential to expand research beyond what is capable with current standards. As a use case example of VISTA, we quantified abundance of histologic features in murine cohorts with oncogenic Kras-driven disease. We observed stromal expansion, a reduction in normal acinar, and an increase in both ADM and dysplasia as the disease progresses, which matches known biology. Since VISTA is an automated algorithm, it can accelerate histological analysis and improve the consistency of quantification between laboratories and investigators. This work has been published in Nature Scientific Reports, and the code is available on github at https://github.com/GelatinFrogs/MicePan-Segmentation.

Citation Format: Luke Ternes, Ge Huang, Christian Lanciault, Guillaume Thibault, Rachelle Riggers, Joe Gray, John Muschler, Young Hwan Chang. VISTA: VIsual Semantic Tissue Analysis for pancreatic disease quantification in murine cohorts [abstract]. In: Proceedings of the AACR Virtual Special Conference on Pancreatic Cancer; 2021 Sep 29-30. Philadelphia (PA): AACR; Cancer Res 2021;81(22 Suppl):Abstract nr PO-014.

Spatial Profiles of Intratumoral PD-1+ Helper T Cells Predict Prognosis in Head and Neck Squamous Cell Carcinoma

Background

Functional interactions between immune cells and neoplastic cells in the tumor immune microenvironment have been actively pursued for both biomarker discovery for patient stratification, as well as therapeutic anti-cancer targets to improve clinical outcomes. Although accumulating evidence indicates that intratumoral infiltration of immune cells has prognostic significance, limited information is available on the spatial infiltration patterns of immune cells within intratumoral regions. This study aimed to understand the intratumoral heterogeneity and spatial distribution of immune cell infiltrates associated with cell phenotypes and prognosis in head and neck squamous cell carcinoma (HNSCC).

Methods

A total of 88 specimens of oropharyngeal squamous cell carcinoma, categorized into discovery (n = 38) and validation cohorts (n = 51), were analyzed for immune contexture by multiplexed immunohistochemistry (IHC) and image cytometry-based quantification. Tissue segmentation was performed according to a mathematical morphological approach using neoplastic cell IHC images to dissect intratumoral regions into tumor cell nests versus intratumoral stroma.

Results

Tissue segmentation revealed heterogeneity in intratumoral T cells, varying from tumor cell nest-polarized to intratumoral stroma-polarized distributions. Leukocyte composition analysis revealed higher ratios of TH1/TH2 in tumor cell nests with higher percentages of helper T cells, B cells, and CD66b+ granulocytes within intratumoral stroma. A discovery and validation approach revealed a high density of programmed death receptor-1 (PD-1)+ helper T cells in tumor cell nests as a negative prognostic factor for short overall survival. CD163+ tumor-associated macrophages (TAM) provided the strongest correlation with PD-1+ helper T cells, and cases with a high density of PD-1+ helper T cells and CD163+ TAM had a significantly shorter overall survival than other cases.

Conclusion

This study reveals the significance of analyzing intratumoral cell nests and reports that an immune microenvironment with a high density of PD-1+ helper T cells in tumoral cell nests is a poor prognostic factor for HNSCC.

Toward reproducible, scalable, and robust data analysis across multiplex tissue imaging platforms

The emergence of megascale single-cell multiplex tissue imaging (MTI) datasets necessitates reproducible, scalable, and robust tools for cell phenotyping and spatial analysis. We developed open-source, graphics processing unit (GPU)-accelerated tools for intensity normalization, phenotyping, and microenvironment characterization. We deploy the toolkit on a human breast cancer (BC) tissue microarray stained by cyclic immunofluorescence and present the first cross-validation of breast cancer cell phenotypes derived by using two different MTI platforms. Finally, we demonstrate an integrative phenotypic and spatial analysis revealing BC subtype-specific features.

Genomic Alterations during the In Situ to Invasive Ductal Breast Carcinoma Transition Shaped by the Immune System

The drivers of ductal carcinoma in situ (DCIS) to invasive ductal carcinoma (IDC) transition are poorly understood. Here, we conducted an integrated genomic, transcriptomic, and whole-slide image analysis to evaluate changes in copy-number profiles, mutational profiles, expression, neoantigen load, and topology in 6 cases of matched pure DCIS and recurrent IDC. We demonstrate through combined copy-number and mutational analysis that recurrent IDC can be genetically related to its pure DCIS despite long latency periods and therapeutic interventions. Immune "hot" and "cold" tumors can arise as early as DCIS and are subtype-specific. Topologic analysis showed a similar degree of pan-leukocyte-tumor mixing in both DCIS and IDC but differ when assessing specific immune subpopulations such as CD4 T cells and CD68 macrophages. Tumor-specific copy-number aberrations in MHC-I presentation machinery and losses in 3p, 4q, and 5p are associated with differences in immune signaling in estrogen receptor (ER)-negative IDC. Common oncogenic hotspot mutations in genes including TP53 and PIK3CA are predicted to be neoantigens yet are paradoxically conserved during the DCIS-to-IDC transition, and are associated with differences in immune signaling. We highlight both tumor and immune-specific changes in the transition of pure DCIS to IDC, including genetic changes in tumor cells that may have a role in modulating immune function and assist in immune escape, driving the transition to IDC. IMPLICATIONS: We demonstrate that the in situ to IDC evolutionary bottleneck is shaped by both tumor and immune cells.

VISTA: VIsual Semantic Tissue Analysis for pancreatic disease quantification in murine cohorts

Mechanistic disease progression studies using animal models require objective and quantifiable assessment of tissue pathology. Currently quantification relies heavily on staining methods which can be expensive, labor/time-intensive, inconsistent across laboratories and batch, and produce uneven staining that is prone to misinterpretation and investigator bias. We developed an automated semantic segmentation tool utilizing deep learning for rapid and objective quantification of histologic features relying solely on hematoxylin and eosin stained pancreatic tissue sections. The tool segments normal acinar structures, the ductal phenotype of acinar-to-ductal metaplasia (ADM), and dysplasia with Dice coefficients of 0.79, 0.70, and 0.79, respectively. To deal with inaccurate pixelwise manual annotations, prediction accuracy was also evaluated against biological truth using immunostaining mean structural similarity indexes (SSIM) of 0.925 and 0.920 for amylase and pan-keratin respectively. Our tool's disease area quantifications were correlated to the quantifications of immunostaining markers (DAPI, amylase, and cytokeratins; Spearman correlation score = 0.86, 0.97, and 0.92) in unseen dataset (n = 25). Moreover, our tool distinguishes ADM from dysplasia, which are not reliably distinguished with immunostaining, and demonstrates generalizability across murine cohorts with pancreatic disease. We quantified the changes in histologic feature abundance for murine cohorts with oncogenic Kras-driven disease, and the predictions fit biological expectations, showing stromal expansion, a reduction of normal acinar tissue, and an increase in both ADM and dysplasia as disease progresses. Our tool promises to accelerate and improve the quantification of pancreatic disease in animal studies and become a unifying quantification tool across laboratories.

The yeast FIT2 homologs are necessary to maintain cellular proteostasis and membrane lipid homeostasis

Lipid droplets (LDs) are implicated in conditions of lipid and protein dysregulation. The fat storage-inducing transmembrane (FIT; also known as FITM) family induces LD formation. Here, we establish a model system to study the role of the Saccharomyces cerevisiae FIT homologues (ScFIT), SCS3 and YFT2, in the proteostasis and stress response pathways. While LD biogenesis and basal endoplasmic reticulum (ER) stress-induced unfolded protein response (UPR) remain unaltered in ScFIT mutants, SCS3 was found to be essential for proper stress-induced UPR activation and for viability in the absence of the sole yeast UPR transducer IRE1 Owing to not having a functional UPR, cells with mutated SCS3 exhibited an accumulation of triacylglycerol within the ER along with aberrant LD morphology, suggesting that there is a UPR-dependent compensatory mechanism that acts to mitigate lack of SCS3 Additionally, SCS3 was necessary to maintain phospholipid homeostasis. Strikingly, global protein ubiquitylation and the turnover of both ER and cytoplasmic misfolded proteins is impaired in ScFITΔ cells, while a screen for interacting partners of Scs3 identifies components of the proteostatic machinery as putative targets. Together, our data support a model where ScFITs play an important role in lipid metabolism and proteostasis beyond their defined roles in LD biogenesis.This article has an associated First Person interview with the first author of the paper.

Abstract PO-059: The genomic landscape of the in situ to invasive ductal breast carcinoma transition shaped by the immune system

Background: The transition from ductal carcinoma in situ (DCIS) to invasive ductal carcinoma (IDC) is an evolutionary bottleneck where progression occurs only in 30% of patients. Whilst the genetic drivers of this transition remain poorly understood, we have previously shown that immune escape is a key event. In this study, we profile the evolutionary trajectory of matched pure DCIS and IDC in the context of the immune microenvironment. Methods: We have evaluated changes in copy number profiles, mutational profiles, expression and neoantigen load in 6 cases of matched pure DCIS and IDC using exome and RNA sequencing. We have integrated this information with topologic assessment of H&E images and cyclic immunofluorescence. Results: We provide evidence for an evolutionary bottleneck during DCIS to IDC in matched patient samples, showing that copy number aberrations are early events, but low overlap in mutational profiles. Variation in immune composition and spatial orientation can arise as early as in DCIS and are subtype specific. Tumor-specific copy number changes including loss of MHC-I presentation machinery or changes at cytokine rich loci specifically in ER− tumors could contribute to a more immunosuppressive environment in IDC. Oncogenic hotspot mutations can present as neoantigens yet are paradoxically conserved during the DCIS-to-IDC transition. We suggest these mutations have a secondary immune-modulatory function or may be present in normal tissue, escaping immune surveillance as early as in DCIS. Conclusions: We show both genomic and microenvironmental differences in matched pure DCIS and recurrent IDC, highlighting that progression is shaped by both tumor and immune system at this evolutionary bottleneck.

Citation Format: Anne Trinh, Carlos R. Gil Del Alcazar, Sachet A. Shukla, Koei Chin, Young Hwan Chen, Guillaume Thibault, Jennifer Eng, Bojana Jovanovic, C. Marcelo Aldaz, So Yeon Park, Joon Jong, Catherine Wu, Joe Gray, Kornelia Polyak. The genomic landscape of the in situ to invasive ductal breast carcinoma transition shaped by the immune system [abstract]. In: Proceedings of the AACR Virtual Special Conference on Tumor Heterogeneity: From Single Cells to Clinical Impact; 2020 Sep 17-18. Philadelphia (PA): AACR; Cancer Res 2020;80(21 Suppl):Abstract nr PO-059.

SHIFT: speedy histological-to-immunofluorescent translation of a tumor signature enabled by deep learning

Spatially-resolved molecular profiling by immunostaining tissue sections is a key feature in cancer diagnosis, subtyping, and treatment, where it complements routine histopathological evaluation by clarifying tumor phenotypes. In this work, we present a deep learning-based method called speedy histological-to-immunofluorescent translation (SHIFT) which takes histologic images of hematoxylin and eosin (H&E)-stained tissue as input, then in near-real time returns inferred virtual immunofluorescence (IF) images that estimate the underlying distribution of the tumor cell marker pan-cytokeratin (panCK). To build a dataset suitable for learning this task, we developed a serial staining protocol which allows IF and H&E images from the same tissue to be spatially registered. We show that deep learning-extracted morphological feature representations of histological images can guide representative sample selection, which improved SHIFT generalizability in a small but heterogenous set of human pancreatic cancer samples. With validation in larger cohorts, SHIFT could serve as an efficient preliminary, auxiliary, or substitute for panCK IF by delivering virtual panCK IF images for a fraction of the cost and in a fraction of the time required by traditional IF.

A Role of Agrin in Maintaining the Stability of Vascular Endothelial Growth Factor Receptor-2 during Tumor Angiogenesis

Endothelial cell (EC) recruitment is central to the vascularization of tumors. Although several proteoglycans have been implicated in cancer and angiogenesis, their roles in EC recruitment and vascularization during tumorigenesis remain poorly understood. Here, we reveal that Agrin, which is secreted in liver cancer, promotes angiogenesis by recruiting ECs within tumors and metastatic lesions and facilitates adhesion of cancer cells to ECs. In ECs, Agrin-induced angiogenesis and adherence to cancer cells are mediated by Integrin-β1, Lrp4-MuSK pathways involving focal adhesion kinase. Mechanistically, we uncover that Agrin regulates VEGFR2 levels that sustain the angiogenic property of ECs and adherence to cancer cells. Agrin attributes an ECM stiffness-based stabilization of VEGFR2 by enhancing interactions with Integrin-β1-Lrp4 and additionally stimulates endothelial nitric-oxide synthase (e-NOS) signaling. Therefore, we propose that cross-talk between Agrin-expressing cancer and ECs favor angiogenesis by sustaining the VEGFR2 pathway.

Abstract 1870: The impact of the microenvironment on heterogeneity and trametinib response in HCC1143 triple negative breast cancer cells

Triple negative breast cancer (TNBC) lacks expression of hormone receptors (ER and PR) and HER2 and is characterized by aggressive disease with poor outcomes. Recent work suggests that TNBC also has a high degree of intratumoral heterogeneity, as measured by lineage differentiation status. This heterogeneity may impact therapeutic response, as it has been shown that treatment with PI3K/mTOR (BEZ235) or MEK (trametinib) inhibitors can drive TNBC cells into more homogeneous states, but that the surviving cells are resistant to the targeted therapy. In this study, we sought to understand how the microenvironment impacts differentiation state heterogeneity and response to targeted therapeutics in HCC1143 cells using our microenvironment microarray (MEMA) platform. Under low serum growth conditions, we found that several ligands could drive the growth of HCC1143, particularly EGF family ligands like AREG and EGF. With respect to differentiation state and heterogeneity, EGF and TGFB1 drove HCC1143 cells into a more mesenchymal like state, with increased expression of VIM and decreased expression of KRT14. In contrast, BMP2 led to higher levels of KRT14 and lower levels of VIM, leading to a more basal-like state. We also grew HCC1143 on MEMA with trametinib treatment. Here we found that combinations of collagen-based substrates and NRG1, HGF, and EGF ligands all led to higher cell counts and EdU incorporation rates compared to PBS-control treated cells. However, the levels of resistance conferred by the microenvironment was less than we had previously seen in HER2 positive MEMA, as the GR50 values (dose required to inhibit growth by 50%) only increased modestly (18 nM for untreated cells, 40 nM for NRG1, 45 nM for HGF). Interestingly, in that HER2 positive MEMA study, we identified HGF and NRG1 as potent resistance factors to lapatinib, but that they functioned in a subtype specific manner. HGF was effective in basal subtype cells and NRG1 in luminal, but not vice versa. We postulated that the modest resistance we observed was due to ligands acting on subsets of cells. We thus treated cells with a combination of NRG1 plus HGF, and found that this resulted in increased resistance (GR50= 91 nM). Imaging showed that trametinib drove HCC1143 cells to a homogenous KRT14 positive state, but surprisingly, addition of ligands reverted the cells to a more heterogeneous state that was resistant to trametinib. These data demonstrate that the microenvironment can impact the differentiation state of TNBC cells and is also capable of conferring resistance within subsets of the heterogeneous cell populations.

Citation Format: Rebecca Smith, Kaylyn Devlin, Moqing Liu, Tiera Liby, David Kilburn, Elmar Bucher, Damir Sudar, Guillaume Thibault, Mark Dane, Joe Gray, Laura Heiser, James E. Korkola. The impact of the microenvironment on heterogeneity and trametinib response in HCC1143 triple negative breast cancer cells [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 1870.

Stress sensor Ire1 deploys a divergent transcriptional program in response to lipid bilayer stress

Membrane integrity at the endoplasmic reticulum (ER) is tightly regulated, and its disturbance is implicated in metabolic diseases. Using an engineered sensor that activates the unfolded protein response (UPR) exclusively when normal ER membrane lipid composition is compromised, we identified pathways beyond lipid metabolism that are necessary to maintain ER integrity in yeast and in C. elegans. To systematically validate yeast mutants that disrupt ER membrane homeostasis, we identified a lipid bilayer stress (LBS) sensor in the UPR transducer protein Ire1, located at the interface of the amphipathic and transmembrane helices. Furthermore, transcriptome and chromatin immunoprecipitation analyses pinpoint the UPR as a broad-spectrum compensatory response wherein LBS and proteotoxic stress deploy divergent transcriptional UPR programs. Together, these findings reveal the UPR program as the sum of two independent stress responses, an insight that could be exploited for future therapeutic intervention.

A workflow for visualizing human cancer biopsies using large-format electron microscopy

Recent developments in large format electron microscopy have enabled generation of images that provide detailed ultrastructural information on normal and diseased cells and tissues. Analyses of these images increase our understanding of cellular organization and interactions and disease-related changes therein. In this manuscript, we describe a workflow for two-dimensional (2D) and three-dimensional (3D) imaging, including both optical and scanning electron microscopy (SEM) methods, that allow pathologists and cancer biology researchers to identify areas of interest from human cancer biopsies. The protocols and mounting strategies described in this workflow are compatible with 2D large format EM mapping, 3D focused ion beam-SEM and serial block face-SEM. The flexibility to use diverse imaging technologies available at most academic institutions makes this workflow useful and applicable for most life science samples. Volumetric analysis of the biopsies studied here revealed morphological, organizational and ultrastructural aspects of the tumor cells and surrounding environment that cannot be revealed by conventional 2D EM imaging. Our results indicate that although 2D EM is still an important tool in many areas of diagnostic pathology, 3D images of ultrastructural relationships between both normal and cancerous cells, in combination with their extracellular matrix, enables cancer researchers and pathologists to better understand the progression of the disease and identify potential therapeutic targets.

Quantification of fluorophore distribution and therapeutic response in matched in vivo and ex vivo pancreatic cancer model systems

Therapeutic resistance plagues cancer outcomes, challenging treatment particularly in aggressive disease. A unique method to decipher drug interactions with their targets and inform therapy is to employ fluorescence-based screening tools; however, to implement productive screening assays, adequate model systems must be developed. Patient-derived pancreatic cancer models (e.g., cell culture, patient-derived xenograft mouse models, and organoids) have been traditionally utilized to predict personalized therapeutic response. However, cost, long read out times and the inability to fully recapitulate the tumor microenvironment have rendered most models incompatible with clinical decision making for pancreatic ductal adenocarcinoma (PDAC) patients. Tumor explant cultures, where patient tissue can be kept viable for up to weeks, have garnered interest as a platform for delivering personalized therapeutic prediction on a clinically relevant timeline. To fully explore this ex vivo platform, a series of studies were completed to quantitatively compare in vivo models with tumor explants, examining gemcitabine therapeutic efficacy, small molecule uptake and drug-target engagement using a novel fluorescently-labeled gemcitabine conjugate. This initial work shows promise for patient-specific therapeutic selection, where tumor explant drug distribution and response recapitulated the in vivo behavior and could provide a valuable platform for understanding mechanisms of therapeutic response and resistance.

Early Prediction of Breast Cancer Therapy Response using Multiresolution Fractal Analysis of DCE-MRI Parametric Maps

We aimed to determine whether multiresolution fractal analysis of voxel-based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parametric maps can provide early prediction of breast cancer response to neoadjuvant chemotherapy (NACT). In total, 55 patients underwent 4 DCE-MRI examinations before, during, and after NACT. The shutter-speed model was used to analyze the DCE-MRI data and generate parametric maps within the tumor region of interest. The proposed multiresolution fractal method and the more conventional methods of single-resolution fractal, gray-level co-occurrence matrix, and run-length matrix were used to extract features from the parametric maps. Only the data obtained before and after the first NACT cycle were used to evaluate early prediction of response. With a training (N = 40) and testing (N = 15) data set, support vector machine was used to assess the predictive abilities of the features in classification of pathologic complete response versus non-pathologic complete response. Generally the multiresolution fractal features from individual maps and the concatenated features from all parametric maps showed better predictive performances than conventional features, with receiver operating curve area under the curve (AUC) values of 0.91 (all parameters) and 0.80 (K^trans), in the training and testing sets, respectively. The differences in AUC were statistically significant (P < .05) for several parametric maps. Thus, multiresolution analysis that decomposes the texture at various spatial-frequency scales may more accurately capture changes in tumor vascular heterogeneity as measured by DCE-MRI, and therefore provide better early prediction of NACT response.

Predicting primary site of secondary liver cancer with a neural estimator of metastatic origin

Purpose: Pathologists rely on relevant clinical information, visual inspection of stained tissue slide morphology, and sophisticated molecular diagnostics to accurately infer the biological origin of secondary metastatic cancer. While highly effective, this process is expensive in terms of time and clinical resources. We seek to develop and evaluate a computer vision system designed to reasonably infer metastatic origin of secondary liver cancer directly from digitized histopathological whole slide images of liver biopsy. Approach: We illustrate a two-stage deep learning approach to accomplish this task. We first train a model to identify spatially localized regions of cancerous tumor within digitized hematoxylin and eosin (H&E)-stained tissue sections of secondary liver cancer based on a pathologist's annotation of several whole slide images. Then, a second model is trained to generate predictions of the cancers' metastatic origin belonging to one of three distinct clinically relevant classes as confirmed by immunohistochemistry. Results: Our approach achieves a classification accuracy of 90.2% in determining metastatic origin of whole slide images from a held-out test set, which compares favorably to an established clinical benchmark by three board-certified pathologists whose accuracies ranged from 90.2% to 94.1% on the same prediction task. Conclusions: We illustrate the potential impact of deep learning systems to leverage morphological and structural features of H&E-stained tissue sections to guide pathological and clinical determination of the metastatic origin of secondary liver cancers.

Moving beyond the current limits of data analysis in longevity and healthy lifespan studies

Living longer with sustainable quality of life is becoming increasingly important in aging populations. Understanding associative biological mechanisms have proven daunting, because of multigenicity and population heterogeneity. Although Big Data and Artificial Intelligence (AI) could help, naïve adoption is ill advised. We hold the view that model organisms are better suited for big-data analytics but might lack relevance because they do not immediately reflect the human condition. Resolving this hurdle and bridging the human-model organism gap will require some finesse. This includes improving signal:noise ratios by appropriate contextualization of high-throughput data, establishing consistency across multiple high-throughput platforms, and adopting supporting technologies that provide useful in silico and in vivo validation strategies.

Variational Autoencoding Tissue Response to Microenvironment Perturbation

This work applies deep variational autoencoder learning architecture to study multi-cellular growth characteristics of human mammary epithelial cells in response to diverse microenvironment perturbations. Our approach introduces a novel method of visualizing learned feature spaces of trained variational autoencoding models that enables visualization of principal features in two dimensions. We find that unsupervised learned features more closely associate with expert annotation of cell colony organization than biologically-inspired hand-crafted features, demonstrating the utility of deep learning systems to meaningfully characterize features of multi-cellular growth characteristics in a fully unsupervised and data-driven manner.

Deuterated Polyunsaturated Fatty Acids Reduce Oxidative Stress and Extend the Lifespan of C. elegans

Chemically reinforced essential fatty acids (FAs) promise to fight numerous age-related diseases including Alzheimer’s, Friedreich’s ataxia and other neurological conditions. The reinforcement is achieved by substituting the atoms of hydrogen at the bis-allylic methylene of these essential FAs with the isotope deuterium. This substitution leads to a significantly slower oxidation due to the kinetic isotope effect, inhibiting membrane damage. The approach has the advantage of preventing the harmful accumulation of reactive oxygen species (ROS) by inhibiting the propagation of lipid peroxidation while antioxidants potentially neutralize beneficial oxidative species. Here, we developed a model system to mimic the human dietary requirement of omega-3 in Caenorhabditis elegans to study the role of deuterated polyunsaturated fatty acids (D-PUFAs). Deuterated trilinolenin [D-TG(54:9)] was sufficient to prevent the accumulation of lipid peroxides and to reduce the accumulation or ROS. Moreover, D-TG(54:9) significantly extended the lifespan of worms under normal and oxidative stress conditions. These findings demonstrate that D-PUFAs can be used as a food supplement to decelerate the aging process, resulting in extended lifespan.

Lipid bilayer stress and proteotoxic stress-induced unfolded protein response deploy divergent transcriptional and non-transcriptional programmes

The unfolded protein response (UPR) is activated by endoplasmic reticulum (ER) stress and is designed to restore cellular homeostasis through multiple intracellular signalling pathways. In mammals, the UPR programme regulates the expression of hundreds of genes in response to signalling from ATF6, IRE1, and PERK. These three highly conserved stress sensors are activated by the accumulation of unfolded proteins within the ER. Alternatively, IRE1 and PERK sense generalised lipid bilayer stress (LBS) at the ER while ATF6 is activated by an increase of specific sphingolipids. As a result, the UPR supports cellular robustness as a broad-spectrum compensatory pathway that is achieved by deploying a tailored transcriptional programme adapted to the source of ER stress. This review summarises the current understanding of the three ER stress transducers in sensing proteotoxic stress and LBS. The plasticity of the UPR programme in the context of different sources of ER stress will also be discussed.

Robust Cell Detection and Segmentation for Image Cytometry Reveal Th17 Cell Heterogeneity

Image cytometry enables quantitative cell characterization with preserved tissue architecture; thus, it has been highlighted in the advancement of multiplex immunohistochemistry (IHC) and digital image analysis in the context of immune-based biomarker monitoring associated with cancer immunotherapy. However, one of the challenges in the current image cytometry methodology is a technical limitation in the segmentation of nuclei and cellular components particularly in heterogeneously stained cancer tissue images. To improve the detection and specificity of single-cell segmentation in hematoxylin-stained images (which can be utilized for recently reported 12-biomarker chromogenic sequential multiplex IHC), we adapted a segmentation algorithm previously developed for hematoxlin and eosin-stained images, where morphological features are extracted based on Gabor-filtering, followed by stacking of image pixels into n-dimensional feature space and unsupervised clustering of individual pixels. Our proposed method showed improved sensitivity and specificity in comparison with standard segmentation methods. Replacing previously proposed methods with our method in multiplex IHC/image cytometry analysis, we observed higher detection of cell lineages including relatively rare T_H 17 cells, further enabling sub-population analysis into T_H 1-like and T_H 2-like phenotypes based on T-bet and GATA3 expression. Interestingly, predominance of T_H 2-like T_H 17 cells was associated with human papilloma virus (HPV)-negative status of oropharyngeal squamous cell carcinoma of head and neck, known as a poor-prognostic subtype in comparison with HPV-positive status. Furthermore, T_H 2-like T_H 17 cells in HPV-negative head and neck cancer tissues were spatiotemporally correlated with CD66b⁺ granulocytes, presumably associated with an immunosuppressive microenvironment. Our cell segmentation method for multiplex IHC/image cytometry potentially contributes to in-depth immune profiling and spatial association, leading to further tissue-based biomarker exploration. © 2019 International Society for Advancement of Cytometry.

Modeling Tumor Phenotypes In Vitro with Three-Dimensional Bioprinting

The tumor microenvironment plays a critical role in tumor growth, progression, and therapeutic resistance, but interrogating the role of specific tumor-stromal interactions on tumorigenic phenotypes is challenging within in vivo tissues. Here, we tested whether three-dimensional (3D) bioprinting could improve in vitro models by incorporating multiple cell types into scaffold-free tumor tissues with defined architecture. We generated tumor tissues from distinct subtypes of breast or pancreatic cancer in relevant microenvironments and demonstrate that this technique can model patient-specific tumors by using primary patient tissue. We assess intrinsic, extrinsic, and spatial tumorigenic phenotypes in bioprinted tissues and find that cellular proliferation, extracellular matrix deposition, and cellular migration are altered in response to extrinsic signals or therapies. Together, this work demonstrates that multi-cell-type bioprinted tissues can recapitulate aspects of in vivo neoplastic tissues and provide a manipulable system for the interrogation of multiple tumorigenic endpoints in the context of distinct tumor microenvironments.

IFNγ-activated dermal lymphatic vessels inhibit cytotoxic T cells in melanoma and inflamed skin

Mechanisms of immune suppression in peripheral tissues counteract protective immunity to prevent immunopathology and are coopted by tumors for immune evasion. While lymphatic vessels facilitate T cell priming, they also exert immune suppressive effects in lymph nodes at steady-state. Therefore, we hypothesized that peripheral lymphatic vessels acquire suppressive mechanisms to limit local effector CD8+ T cell accumulation in murine skin. We demonstrate that nonhematopoietic PD-L1 is largely expressed by lymphatic and blood endothelial cells and limits CD8+ T cell accumulation in tumor microenvironments. IFNγ produced by tissue-infiltrating, antigen-specific CD8+ T cells, which are in close proximity to tumor-associated lymphatic vessels, is sufficient to induce lymphatic vessel PD-L1 expression. Disruption of IFNγ-dependent crosstalk through lymphatic-specific loss of IFNγR boosts T cell accumulation in infected and malignant skin leading to increased viral pathology and tumor control, respectively. Consequently, we identify IFNγR as an immunological switch in lymphatic vessels that balances protective immunity and immunopathology leading to adaptive immune resistance in melanoma.

Lipid bilayer stress-activated IRE-1 modulates autophagy during endoplasmic reticulum stress

Metabolic disorders, such as non-alcoholic fatty liver disease (NAFLD), are emerging as epidemics that affect the global population. One facet of these disorders is attributed to the disturbance of membrane lipid composition. Perturbation of endoplasmic reticulum (ER) homeostasis through alteration in membrane phospholipids activates the unfolded protein response (UPR) and causes dramatic transcriptional and translational changes in the cell. To restore cellular homeostasis, the three highly conserved UPR transducers ATF6, IRE1 (also known as ERN1 in mammals) and PERK (also known as EIF2AK3 in mammals) mediate adaptive responses upon ER stress. The homeostatic UPR cascade is well characterised under conditions of proteotoxic stress, but much less so under lipid bilayer stress-induced UPR. Here, we show that disrupted phosphatidylcholine (PC) synthesis in Caenorhabditis elegans causes lipid bilayer stress, lipid droplet accumulation and ER stress induction. Transcriptional profiling of PC-deficient worms revealed a unique subset of genes regulated in a UPR-dependent manner that is independent from proteotoxic stress. Among these, we show that autophagy is modulated through the conserved IRE-1-XBP-1 axis, strongly suggesting of the importance of autophagy in maintaining cellular homeostasis during the lipid bilayer stress-induced UPR.

Tissue Factor Activity of Syncytiotrophoblast Plasma Membranes and Tumoral Trophoblast Cells in Culture

During pregnancy, important modifications of hemostasis occur resulting in mothers in hypercoagulability and the role of placental cells such as trophoblast cells has been hypothesized. In this study, we first showed that syncytiotrophoblast plasma membranes, isolated from normal human placenta, expressed a strong tissue factor (TF) activity. We then studied TF activity of two continuous trophoblast cell lines (JEG-3 and BeWo) in comparison to human umbilical vein endothelial cells (HUVEC) and transformed human endothelial cells (ECV-304). TF assays were performed on intact detached confluent cells. Unstimulated JEG-3 and BeWo cells exhibited a very high TF activity which slightly increased after 2 to 4 h TNF-α stimulation. In contrast, HUVEC and ECV-304 had a lower basal TF activity which was mainly inducible by TNF-a, with a maximum effect after 4 to 6 h stimulation. For both cell types, TF activity was decreased to basal value after 16-hour TNF-α stimulation. These results support that trophoblast cells are able to express TF but the involvement of this property in the hemostatic physiological changes observed during pregnancy, remains to be demonstrated.

Abstract 220: Intra-tumor immune heterogeneity is associated with prognosis of oropharyngeal head and neck squamous cell carcinoma

As cell-cell interactions among tumor and immune cells are known to contribute to tumor progression, in depth evaluation of tumor-immune heterogeneity will enable biomarker-guided patient stratification and improvement of treatment response. Here we examined head and neck squamous cell carcinomas (HNSCC) to reveal the prognostic significance of leukocytes in tumors with regards to their complexity, effector status and spatial characteristics via multiplex immunohistochemistry and image cytometry. To accomplish this, we examined oropharyngeal HNSCC (N = 38), where lymphoid, myeloid, and hypo-inflamed leukocyte complexity correlated with HPV-status (Tsujikawa et al. Cell Reports, 2017). In the present study, quantitative analysis of cell density and distribution of 14-distinct immune cell lineages (e.g., CD8+ T cells, regulatory T cells, Th1, Th2, Th17, Th0 lymphocytes, B cells, NK cells, CD163+ and CD163- macrophages, mature and immature dendritic cells, mast cells, granulocytes) was analyzed via immune cell density mapping and tissue segmentation algorithms. We revealed tropism of polarized Th1-type cells within tumor nests versus stroma in HPV-associated HNSCC. In addition, Cox regression analysis of cell density and distribution of the 14 immune populations revealed that CD66b+ granulocyte infiltration within tumor nests reflected a negative prognostic indicator for HNSCC outcome. Furthermore, cell-cell proximity analysis in HPV-associated HNSCC further revealed a correlation between PD-L1 positive immune cells, and micro-regionally polarized immune characteristics biased towards Th1, coincident with high density of CD8+ T cells. These results reveal intra-tumor immune heterogeneity is associated with micro-regional immune complexity profiles, and provide insight into in situ immune characteristics that may aid patient stratification for immune therapy going forward. Acknowledgement: This project was supported by the Japan Society for the Promotion of Science Grant-in-Aid for Young Scientists (Start-up, 17H07016), Oregon Clinical and Translational Research Institute (OCTRI), grant number (UL1TR000128) from the National Center for Advancing Translational Sciences (NCATS) at the National Institutes of Health (NIH), and P30 CA069533-17 OHSU Knight Cancer Institute. LMC acknowledges support from the NIH/NCI, DOD BCRP Era of Hope Scholar Expansion Award, Susan G. Komen Foundation, Stand Up To Cancer - Lustgarten Foundation Pancreatic Cancer Convergence Dream Team Translational Research Grant, Breast Cancer Research Foundation, and the Brenden-Colson Center for Pancreatic Health.

Citation Format: Takahiro Tsujikawa, Guillaume Thibault, Young Hwan Chang, Edward El Rassi, Daniel R. Clayburgh, David Sauer, Akihito Arai, Molly F. Kulesz-Martin, Motomi Mori, Shigeru Hirano, Paul W. Flint, Lisa M. Coussens. Intra-tumor immune heterogeneity is associated with prognosis of oropharyngeal head and neck squamous cell carcinoma [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 220.

Analysis of DCE-MRI for Early Prediction of Breast Cancer Therapy Response

Positive response to neoadjuvant chemotherapy (NACT) has been correlated to better long-term outcomes in breast cancer treatment. Early prediction of response to NACT can help modify the regimen for non-responding patients, sparing them of potential toxicities of ineffective therapies. It has been observed that tumor functions such as vascularization and vascular permeability change even before noticeable changes occur in the tumor size in response to the treatment. Therefore, it is essential to have reliable imaging based features to measure these changes. Texture analysis on parametric maps from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has shown to be a good predictor of breast cancer response to NACT at an early stage. But hand crafted texture features might not be able to capture the rich spatio-temporal information in the parametric maps. In this work, we studied the ability of convolutional neural networks in predicting the response to NACT at an early stage.

Deep learning based Nucleus Classification in pancreas histological images

Tumor specimens contain a variety of healthy cells as well as cancerous cells, and this heterogeneity underlies resistance to various cancer therapies. But this problem has not been thoroughly investigated until recently. Meanwhile, technological breakthroughs in imaging have led to an explosion of molecular and cellular profiling data from large numbers of samples, and modern machine learning approaches including deep learning have been shown to produce encouraging results by finding hidden structures and make accurate predictions. In this paper, we propose a Deep learning based Nucleus Classification (DeepNC) approach using paired histopathology and immunofluorescence images (for label), and demonstrate its classification prediction power. This method can solve current issue on discrepancy between genomic- or transcriptomic-based and pathology-based tumor purity estimates by improving histological evaluation. We also explain challenges in training a deep learning model for huge dataset.

From the unfolded protein response to metabolic diseases - lipids under the spotlight

The unfolded protein response (UPR) is classically viewed as a stress response pathway to maintain protein homeostasis at the endoplasmic reticulum (ER). However, it has recently emerged that the UPR can be directly activated by lipid perturbation, independently of misfolded proteins. Comprising primarily phospholipids, sphingolipids and sterols, individual membranes can contain hundreds of distinct lipids. Even with such complexity, lipid distribution in a cell is tightly regulated by mechanisms that remain incompletely understood. It is therefore unsurprising that lipid dysregulation can be a key factor in disease development. Recent advances in analysis of lipids and their regulators have revealed remarkable mechanisms and connections to other cellular pathways including the UPR. In this Review, we summarize the current understanding in UPR transducers functioning as lipid sensors and the interplay between lipid metabolism and ER homeostasis in the context of metabolic diseases. We attempt to provide a framework consisting of a few key principles to integrate the different lines of evidence and explain this rather complicated mechanism.

Quantitative Multiplex Immunohistochemistry Reveals Myeloid-Inflamed Tumor-Immune Complexity Associated with Poor Prognosis

Here, we describe a multiplexed immunohistochemical platform with computational image processing workflows, including image cytometry, enabling simultaneous evaluation of 12 biomarkers in one formalin-fixed paraffin-embedded tissue section. To validate this platform, we used tissue microarrays containing 38 archival head and neck squamous cell carcinomas and revealed differential immune profiles based on lymphoid and myeloid cell densities, correlating with human papilloma virus status and prognosis. Based on these results, we investigated 24 pancreatic ductal adenocarcinomas from patients who received neoadjuvant GVAX vaccination and revealed that response to therapy correlated with degree of mono-myelocytic cell density and percentages of CD8⁺ T cells expressing T cell exhaustion markers. These data highlight the utility of in situ immune monitoring for patient stratification and provide digital image processing pipelines to the community for examining immune complexity in precious tissue sections, where phenotype and tissue architecture are preserved to improve biomarker discovery and assessment.

Quantitative analysis of histological tissue image based on cytological profiles and spatial statistics

The cellular heterogeneity and complex tissue architecture of most tumor samples is a major obstacle in image analysis on standard hematoxylin and eosin-stained (H&E) tissue sections. A mixture of cancer and normal cells complicates the interpretation of their cytological profiles. Furthermore, spatial arrangement and architectural organization of cells are generally not reflected in cellular characteristics analysis. To address these challenges, first we describe an automatic nuclei segmentation of H&E tissue sections. In the task of deconvoluting cellular heterogeneity, we adopt Landmark based Spectral Clustering (LSC) to group individual nuclei in such a way that nuclei in the same group are more similar. We next devise spatial statistics for analyzing spatial arrangement and organization, which are not detectable by individual cellular characteristics. Our quantitative, spatial statistics analysis could benefit H&E section analysis by refining and complementing cellular characteristics analysis.

DCE-MRI Texture Features for Early Prediction of Breast Cancer Therapy Response

This study investigates the effectiveness of hundreds of texture features extracted from voxel-based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parametric maps for early prediction of breast cancer response to neoadjuvant chemotherapy (NAC). In total, 38 patients with breast cancer underwent DCE-MRI before (baseline) and after the first of the 6-8 NAC cycles. Quantitative pharmacokinetic (PK) parameters and semiquantitative metrics were estimated from DCE-MRI time-course data. The residual cancer burden (RCB) index value was computed based on pathological analysis of surgical specimens after NAC completion. In total, 1043 texture features were extracted from each of the 13 parametric maps of quantitative PK or semiquantitative metric, and their capabilities for early prediction of RCB were examined by correlating feature changes between the 2 MRI studies with RCB. There were 1069 pairs of feature-map combinations that showed effectiveness for response prediction with 4 correlation coefficients >0.7. The 3-dimensional gray-level cooccurrence matrix was the most effective feature extraction method for therapy response prediction, and, in general, the statistical features describing texture heterogeneity were the most effective features. Quantitative PK parameters, particularly those estimated with the shutter-speed model, were more likely to generate effective features for prediction response compared with the semiquantitative metrics. The best feature-map pair could predict pathologic complete response with 100% sensitivity and 100% specificity using our cohort. In conclusion, breast tumor heterogeneity in microvasculature as measured by texture features of voxel-based DCE-MRI parametric maps could be a useful biomarker for early prediction of NAC response.

Abstract NG02: Lymphatic vessels: Balancing immune priming and immune evasion in melanoma

Lymphatic vessel remodeling and lymphangiogenesis is correlated with melanoma progression and lymph node metastasis. While lymphatic vessels provide an important route for disseminating tumor cells, they are also a crucial interface between a developing malignancy and the host immune response. Rather than acting as passive conduits, lymphatic vessels directly regulate their transport function and facilitate leukocyte trafficking for efficient induction of adaptive immunity in downstream draining lymph nodes. We recently published that in the absence of dermal lymphatic vessels, the tumor microenvironment of murine melanoma remains completely uninflamed and fails to induce a robust T-cell response (1). Consistently, TCGA analysis of human cutaneous metastatic melanoma identified positive correlations between LEC gene expression and immune genes, including a T cell-inflamed signature, indicating a relationship between lymphatic vessel remodeling and local immunity. Furthermore, others have recently demonstrated that lymphatic vessel density (LVD) in combination with intratumoral T cell function stratified best overall survival in nonmetastatic and metastatic colorectal cancer (2). In contrast to this, however, many reports independently correlate peritumoral LVD with lymph node metastasis and some poor prognosis (3). Furthermore, our previous work demonstrated that vascular endothelial growth factor C (VEGFC)-driven lymphangiogenesis in the context of murine melanoma drove increased leukocyte infiltration but associated with poor CD8+ T cell priming in draining lymph nodes (4). We therefore hypothesize that lymphatic vessels are (A) required for induction of adaptive immunity but (B) acquire immunosuppressive activity as a function of the accumulation of local cytotoxic immunity. Furthermore, we predict that LVD may be a relevant biomarker of in situ immune responsiveness and response to therapy (5).

To test the first part of this hypothesis (A), we have continued our published work and used a cutaneous model of viral infection to demonstrate the requirement for lymphatic vessel drainage in cutaneous immunity. Following cutaneous vaccinia infection we demonstrate that in the absence of lymphatic vessel transport, both cellular and humoral adaptive immune responses fail to initiate, leading to enhanced cutaneous immunopathology and persistent viral load. The complete absence of primed CD8+ and CD4+ T cells in cutaneous tissue following challenge mirrors our observations in melanoma and is consistent with correlations of intratumoral lymphocytes and LVD both by our group in cutaneous metastatic melanoma as well as by others. This unequivocal requirement for a functional lymphatic vasculature in the priming of cutaneous immunity further supports the prediction that LVD may be a novel biomarker of immune reactivity within tumor parenchyma. To test this, we simultaneously evaluated immune and vascular components in human primary melanoma samples using a multiplex-immunohistochemistry-based approach. Tissue regions that include tumor/stroma borders and show high CD8+ T-cell infiltrates are selected for analysis, followed by tissue segmentation, and automated detection of cell populations within intratumoral regions and bordering stroma. Interestingly, those tumors with enhanced hematopoietic infiltrate (CD68 and CD8) also appear to demonstrate increased vasculature, both blood (CD31 and CD34) and lymphatic (D2-40 and Prox1). Preliminary data demonstrate that lymphatic vessels, blood vessels, and CD8+ T cells are significantly enriched at the tumor-stroma border in primary melanoma and positively correlate with one another, indicating that lymphatic vessels may be a dynamic component of the “T cell-inflamed” microenvironment.

While we demonstrate that lymphatic vessels are necessary for immune induction, we further hypothesized (B) that in the context of an ongoing immune response lymphatic vessels adapt their function to promote immune resolution. The adaptive resistance hypothesis proposes that upon accumulation of local cytotoxic immunity, both tumors as well as stromal components adapt and acquire therapeutically relevant immunosuppressive function. We demonstrate that peripheral, tumor-associated lymphatic endothelial cells (LEC; CD45-CD31+gp38+) acquire expression of immunoregulatory proteins, most notably programmed death receptor ligand-1 (PD-L1) and major histocompatibility complex II (MHCII), coincident with CD8+ T-cell infiltration in an interferon gamma (IFNg)-dependent manner. Adoptive transfer of activated CD8+ T cells induced higher expression of PD-L1 by LECs in B16 F10 tumors, while neutralization of IFNg reduced levels to that of naïve skin. Furthermore, conditional knockout of the IFNgR (Lyve1-Cre) prevented upregulation of PD-L1 on tumor-associated LECs. Notably, lymph node LECs constitutively express PD-L1 and this expression contributes to the attenuation of self-reactive CD8+ T-cell responses (6). Importantly, in the absence of IFNgR on peripheral LECs we observed significantly enhanced accumulation of antigen-specific CD8+ T cells in cutaneous tissue. Thus, cutaneous lymphatic vessels, while necessary for immune induction, acquire immunodulatory properties in a context-dependent manner and may participate in immune evasion within tumor microenvironments.

In conclusion, our work across multiple model systems provides strong experimental evidence to indicate that the lymphatic vasculature is an important, active component of the antitumor immune response and may represent a biomarker to stratify patient response and survival for effective clinical immunotherapy. These data indicate a need to revisit the passive paradigm of lymphatic vessel involvement in tumor progression and metastasis to a more active model whereby lymphatic vessels are both required for antitumor immunity but functionally evolve in response to accumulating cytotoxicity to drive immune evasion.

References:

1. Lund AW et al. Lymphatic vessels regulate immune microenvironments in human and murine melanoma. J Clin Invest 2016;126:3389-402. doi: 10.1172/JCI79434.

2. Mlecnik B et al. The tumor microenvironment and Immunoscore are critical determinants of dissemination to distant metastasis. Sci Transl Med 2016;8:327ra26.

3. Pasquali S et al. Lymphatic biomarkers in primary melanomas as predictors of regional lymph node metastasis and patient outcomes. Pigment Cell Melanoma Res 2013;26:326-37.

4. Lund AW et al. VEGF-C promotes immune tolerance in B16 melanomas and cross-presentation of tumor antigen by lymph node lymphatics. Cell Rep 2012;1:191-9.

5. Lund AW. Rethinking lymphatic vessels and antitumor immunity. Trends Cancer 2016;2:548-51. doi: 10.1016/j.trecan.2016.09.0056. Tewalt EF et al. Lymphatic endothelial cells induce tolerance via PD-L1 and lack of costimulation leading to high-level PD-1 expression on CD8 T cells. Blood 2012;120:4772-82.

Citation Format: Ryan S. Lane, Julia Femel, Jamie Booth, Christopher Loo, Nicholas Nelson, Takahiro Tsujikawa, Guillaume Thibault, Amanda W. Lund. Lymphatic vessels: Balancing immune priming and immune evasion in melanoma [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 NG02. doi:10.1158/1538-7445.AM2017-NG02

A fully integrated, three-dimensional fluorescence to electron microscopy correlative workflow

While fluorescence microscopy provides tools for highly specific labeling and sensitive detection, its resolution limit and lack of general contrast has hindered studies of cellular structure and protein localization. Recent advances in correlative light and electron microscopy (CLEM), including the fully integrated CLEM workflow instrument, the FEI CorrSight with MAPS, have allowed for a more reliable, reproducible, and quicker approach to correlate three-dimensional time-lapse confocal fluorescence data, with three-dimensional focused ion beam-scanning electron microscopy data. Here we demonstrate the entire integrated CLEM workflow using fluorescently tagged MCF7 breast cancer cells.

BREAST CANCER HISTOPATHOLOGY IMAGE ANALYSIS PIPELINE FOR TUMOR PURITY ESTIMATION

The translation of genomic sequencing technology to the clinic has greatly advanced personalized medicine. However, the presence of normal cells in tumors is a confounding factor in genome sequence analysis. Tumor purity, or the percentage of cancerous cells in whole tissue section, is a correction factor that can be used to improve the clinical utility of genomic sequencing. Currently, tumor purity is estimated visually by expert pathologists; however, it has been shown that there exist vast inter-observer discrepancies in tumor purity scoring. In this paper, we propose a quantitative image analysis pipeline for tumor purity estimation and provide a systematic comparison between pathologists' scores and our image-based tumor purity estimation.

Integrative Analysis on Histopathological Image for Identifying Cellular Heterogeneity

This study has brought together image processing, clustering and spatial pattern analysis to quantitatively analyze hematoxylin and eosin-stained (H&E) tissue sections. A mixture of tumor and normal cells (intratumoral heterogeneity) as well as complex tissue architectures of most samples complicate the interpretation of their cytological profiles. To address these challenges, we develop a simple but effective methodology for quantitative analysis for H&E section. We adopt comparative analyses of spatial point patterns to characterize spatial distribution of different nuclei types and complement cellular characteristics analysis. We demonstrate that tumor and normal cell regions exhibit significant differences of lymphocytes spatial distribution or lymphocyte infiltration pattern.