The single-cell pathology landscape of breast cancer.

Single-cell analyses have revealed extensive heterogeneity between and within human tumours1-4, but complex single-cell phenotypes and their spatial context are not at present reflected in the histological stratification that is the foundation of many clinical decisions. Here we use imaging mass cytometry5 to simultaneously quantify 35 biomarkers, resulting in 720 high-dimensional pathology images of tumour tissue from 352 patients with breast cancer, with long-term survival data available for 281 patients. Spatially resolved, single-cell analysis identified the phenotypes of tumour and stromal single cells, their organization and their heterogeneity, and enabled the cellular architecture of breast cancer tissue to be characterized on the basis of cellular composition and tissue organization. Our analysis reveals multicellular features of the tumour microenvironment and novel subgroups of breast cancer that are associated with distinct clinical outcomes. Thus, spatially resolved, single-cell analysis can characterize intratumour phenotypic heterogeneity in a disease-relevant manner, with the potential to inform patient-specific diagnosis.

Design Parameters for a Mass Cytometry Detectable HaloTag Ligand.

Mass cytometry permits the high dimensional analysis of complex biological samples; however, some techniques are not yet integrated into the mass cytometry workflow due to reagent availability. The use of self-labeling protein systems, such as HaloTag, are one such application. Here, we describe the design and implementation of the first mass cytometry ligands for use with HaloTag. "Click"-amenable HaloTag warheads were first conjugated onto poly(l-lysine) or poly(acrylic acid) polymers that were then functionalized with diethylenetriaminepentaacetic acid (DTPA) lutetium metal chelates. Kinetic analysis of the HaloTag labeling rates demonstrated that the structure appended to the 1-chlorohexyl warhead was key to success. A construct with a diethylene glycol spacer appended to a benzamide gave similar rates (kobs ∼ 102 M-1 s-1), regardless of the nature of the polymer. Comparison of the polymer with a small molecule chelate having rapid HaloTag labeling kinetics (kobs ∼ 104 M-1 s-1) suggests the polymers significantly reduced the HaloTag labeling rate. HEK293T cells expressing surface-exposed GFP-HaloTag fusions were labeled with the polymeric constructs and 175Lu content measured by cytometry by time-of-flight (CyTOF). Robust labeling was observed; however, significant nonspecific binding of the constructs to cells was also present. Heavily pegylated polymers demonstrated that nonspecific binding could be reduced to allow cells bearing the HaloTag protein to be distinguished from nonexpressing cells.

histoCAT: analysis of cell phenotypes and interactions in multiplex image cytometry data.

Single-cell, spatially resolved omics analysis of tissues is poised to transform biomedical research and clinical practice. We have developed an open-source, computational histology topography cytometry analysis toolbox (histoCAT) to enable interactive, quantitative, and comprehensive exploration of individual cell phenotypes, cell-cell interactions, microenvironments, and morphological structures within intact tissues. We highlight the unique abilities of histoCAT through analysis of highly multiplexed mass cytometry images of human breast cancer tissues.

Limited utility of tissue micro-arrays in detecting intra-tumoral heterogeneity in stem cell characteristics and tumor progression markers in breast cancer.

BACKGROUND: Intra-tumoral heterogeneity has been recently addressed in different types of cancer, including breast cancer. A concept describing the origin of intra-tumoral heterogeneity is the cancer stem-cell hypothesis, proposing the existence of cancer stem cells that can self-renew limitlessly and therefore lead to tumor progression. Clonal evolution in accumulated single cell genomic alterations is a further possible explanation in carcinogenesis. In this study, we addressed the question whether intra-tumoral heterogeneity can be reliably detected in tissue-micro-arrays in breast cancer by comparing expression levels of conventional predictive/prognostic tumor markers, tumor progression markers and stem cell markers between central and peripheral tumor areas. METHODS: We analyzed immunohistochemical expression and/or gene amplification status of conventional prognostic tumor markers (ER, PR, HER2, CK5/6), tumor progression markers (PTEN, PIK3CA, p53, Ki-67) and stem cell markers (mTOR, SOX2, SOX9, SOX10, SLUG, CD44, CD24, TWIST) in 372 tissue-micro-array samples from 72 breast cancer patients. Expression levels were compared between central and peripheral tumor tissue areas and were correlated to histopathological grading. 15 selected cases additionally underwent RNA sequencing for transcriptome analysis. RESULTS: No significant difference in any of the analyzed between central and peripheral tumor areas was seen with any of the analyzed methods/or results that showed difference. Except mTOR, PIK3CA and SOX9 (nuclear) protein expression, all markers correlated significantly (p < 0.05) with histopathological grading both in central and peripheral areas. CONCLUSION: Our results suggest that intra-tumoral heterogeneity of stem-cell and tumor-progression markers cannot be reliably addressed in tissue-micro-array samples in breast cancer. However, most markers correlated strongly with histopathological grading confirming prognostic information as expression profiles were independent on the site of the biopsy was taken.

Correction to: Limited utility of tissue micro-arrays in detecting intra-tumoral heterogeneity in stem cell characteristics and tumor progression markers in breast cancer.

Following publication of the original article [1], a typesetting mistake is reported. For Fig. 7b, a copy of Fig. 6b has been published. The correct Fig. 7b is given in this correction and the original article has been updated.

Progesterone induces adult mammary stem cell expansion.

Reproductive history is the strongest risk factor for breast cancer after age, genetics and breast density. Increased breast cancer risk is entwined with a greater number of ovarian hormone-dependent reproductive cycles, yet the basis for this predisposition is unknown. Mammary stem cells (MaSCs) are located within a specialized niche in the basal epithelial compartment that is under local and systemic regulation. The emerging role of MaSCs in cancer initiation warrants the study of ovarian hormones in MaSC homeostasis. Here we show that the MaSC pool increases 14-fold during maximal progesterone levels at the luteal dioestrus phase of the mouse. Stem-cell-enriched CD49fhi cells amplify at dioestrus, or with exogenous progesterone, demonstrating a key role for progesterone in propelling this expansion. In aged mice, CD49fhi cells display stasis upon cessation of the reproductive cycle. Progesterone drives a series of events where luminal cells probably provide Wnt4 and RANKL signals to basal cells which in turn respond by upregulating their cognate receptors, transcriptional targets and cell cycle markers. Our findings uncover a dynamic role for progesterone in activating adult MaSCs within the mammary stem cell niche during the reproductive cycle, where MaSCs are putative targets for cell transformation events leading to breast cancer.

Multiplex imaging of breast cancer lymph node metastases identifies prognostic single-cell populations independent of clinical classifiers.

Although breast cancer mortality is largely caused by metastasis, clinical decisions are based on analysis of the primary tumor and on lymph node involvement but not on the phenotype of disseminated cells. Here, we use multiplex imaging mass cytometry to compare single-cell phenotypes of primary breast tumors and matched lymph node metastases in 205 patients. We observe extensive phenotypic variability between primary and metastatic sites and that disseminated cell phenotypes frequently deviate from the clinical disease subtype. We identify single-cell phenotypes and spatial organizations of disseminated tumor cells that are associated with patient survival and a weaker survival association for high-risk phenotypes in the primary tumor. We show that p53 and GATA3 in lymph node metastases provide prognostic information beyond clinical classifiers and can be measured with standard methods. Molecular characterization of disseminated tumor cells is an untapped source of clinically applicable prognostic information for breast cancer.

An Automation Workflow for High-Throughput Manufacturing and Analysis of Scaffold-Supported 3D Tissue Arrays.

Patient-derived organoids have emerged as a useful tool to model tumour heterogeneity. Scaling these complex culture models while enabling stratified analysis of different cellular sub-populations, however, remains a challenge. One strategy to enable higher throughput organoid cultures is the scaffold-supported platform for organoid-based tissues (SPOT). SPOT allows the generation of flat, thin, and dimensionally-defined microtissues in both 96- and 384-well plate footprints that are compatible with longitudinal image-based readouts. SPOT is currently manufactured manually, however, limiting scalability. In this study, an automation approach to engineer tumour-mimetic 3D microtissues in SPOT using a liquid handler is optimized and comparable within- and between-sample variation to standard manual manufacturing is shown. Further, a liquid handler-supported cell extraction protocol to support single-cell-based end-point analysis using high-throughput flow cytometry and multiplexed cytometry by time of flight is developed. As a proof-of-value demonstration, 3D complex tissues containing different proportions of tumour and stromal cells are generated to probe the reciprocal impact of co-culture. It is also demonstrated that primary patient-derived organoids can be incorporated into the pipeline to capture patient-level tumour heterogeneity. It is envisioned that this automated 96/384-SPOT workflow will provide opportunities for future applications in high-throughput screening for novel personalized therapeutic targets.

An Engineered Paper-Based 3D Coculture Model of Pancreatic Cancer to Study the Impact of Tissue Architecture and Microenvironmental Gradients on Cell Phenotype.

The spatial configuration of cells in the tumor microenvironment (TME) affects both cancer and fibroblast cell phenotypes contributing to the clinical challenge of tumor heterogeneity and therapeutic resistance. This is a particular challenge in stroma-rich pancreatic ductal adenocarcinoma (PDAC). Here, a versatile system is described to study the impact of tissue architecture on cell phenotype using PDAC as a model system. This fully human system encompassing both primary pancreatic stellate cells and primary organoid cells using the TRACER platform to allow the creation of user-defined TME architectures that have been inferred from clinical PDAC samples and are analyzed by CyTOF to characterize cells extracted from the system. High dimensional characterization using CyTOF demonstrates that tissue architecture leads to distinct hypoxia and proliferation gradients. Furthermore, phenotypic markers for both cell types are also graded in ways that cannot be explained by either hypoxia or coculture alone. This demonstrates the importance of using complex models encompassing cancer cells, stromal cells, and allowing control over architecture to explore the impact of tissue architecture on cell phenotype. It is anticipated that this model will help decipher how tissue architecture and cell interactions regulate cell phenotype and hence cellular and tissue heterogeneity.

In vivo CRISPR screens reveal Serpinb9 and Adam2 as regulators of immune therapy response in lung cancer.

How the genetic landscape governs a tumor's response to immunotherapy remains poorly understood. To assess the immune-modulatory capabilities of 573 genes associated with altered cytotoxicity in human cancers, here we perform CRISPR/Cas9 screens directly in mouse lung cancer models. We recover the known immune evasion factors Stat1 and Serpinb9 and identify the cancer testis antigen Adam2 as an immune modulator, whose expression is induced by KrasG12D and further elevated by immunotherapy. Using loss- and gain-of-function experiments, we show that ADAM2 functions as an oncogene by restraining interferon and TNF cytokine signaling causing reduced presentation of tumor-associated antigens. ADAM2 also restricts expression of the immune checkpoint inhibitors PDL1, LAG3, TIGIT and TIM3 in the tumor microenvironment, which might explain why ex vivo expanded and adoptively transferred cytotoxic T-cells show enhanced cytotoxic efficacy in ADAM2 overexpressing tumors. Together, direct in vivo CRISPR/Cas9 screens can uncover genetic alterations that control responses to immunotherapies.

Genetic perturbations go spatial.

Tissue-tumor interactivity is the culmination of cell intrinsic features and their extrinsic interactions with the environment. Recently in Cell, Dhainaut and Rose et al. established a strategy to track pooled CRISPR-modified cells in vivo using protein barcodes (Pro-Codes) and measure their impact on the tumor microenvironment through multiplexed imaging and spatial transcriptomics of intact tissues.1.

Single-cell proteomics defines the cellular heterogeneity of localized prostate cancer.

Localized prostate cancer exhibits multiple genomic alterations and heterogeneity at the proteomic level. Single-cell technologies capture important cell-to-cell variability responsible for heterogeneity in biomarker expression that may be overlooked when molecular alterations are based on bulk tissue samples. This study aims to identify prognostic biomarkers and describe the heterogeneity of prostate cancer and the associated microenvironment by simultaneously quantifying 36 proteins using single-cell mass cytometry analysis of over 1.6 million cells from 58 men with localized prostate cancer. We perform this task, using a high-dimensional clustering pipeline named Franken to describe subpopulations of immune, stromal, and prostate cells, including changes occurring in tumor tissues and high-grade disease that provide insights into the coordinated progression of prostate cancer. Our results further indicate that men with localized disease already harbor rare subpopulations that typically occur in castration-resistant and metastatic disease.

Loss of Epigenetic Regulation Disrupts Lineage Integrity, Induces Aberrant Alveogenesis, and Promotes Breast Cancer.

Systematically investigating the scores of genes mutated in cancer and discerning disease drivers from inconsequential bystanders is a prerequisite for precision medicine but remains challenging. Here, we developed a somatic CRISPR/Cas9 mutagenesis screen to study 215 recurrent "long-tail" breast cancer genes, which revealed epigenetic regulation as a major tumor-suppressive mechanism. We report that components of the BAP1 and COMPASS-like complexes, including KMT2C/D, KDM6A, BAP1, and ASXL1/2 ("EpiDrivers"), cooperate with PIK3CAH1047R to transform mouse and human breast epithelial cells. Mechanistically, we find that activation of PIK3CAH1047R and concomitant EpiDriver loss triggered an alveolar-like lineage conversion of basal mammary epithelial cells and accelerated formation of luminal-like tumors, suggesting a basal origin for luminal tumors. EpiDriver mutations are found in ∼39% of human breast cancers, and ∼50% of ductal carcinoma in situ express casein, suggesting that lineage infidelity and alveogenic mimicry may significantly contribute to early steps of breast cancer etiology. SIGNIFICANCE: Infrequently mutated genes comprise most of the mutational burden in breast tumors but are poorly understood. In vivo CRISPR screening identified functional tumor suppressors that converged on epigenetic regulation. Loss of epigenetic regulators accelerated tumorigenesis and revealed lineage infidelity and aberrant expression of alveogenesis genes as potential early events in tumorigenesis. This article is highlighted in the In This Issue feature, p. 2711.

Automated assignment of cell identity from single-cell multiplexed imaging and proteomic data.

A major challenge in the analysis of highly multiplexed imaging data is the assignment of cells to a priori known cell types. Existing approaches typically solve this by clustering cells followed by manual annotation. However, these often require several subjective choices and cannot explicitly assign cells to an uncharacterized type. To help address these issues we present Astir, a probabilistic model to assign cells to cell types by integrating prior knowledge of marker proteins. Astir uses deep recognition neural networks for fast inference, allowing for annotations at the million-cell scale in the absence of a previously annotated reference. We apply Astir to over 2.4 million cells from suspension and imaging datasets and demonstrate its scalability, robustness to sample composition, and interpretable uncertainty estimates. We envision deployment of Astir either for a first broad cell type assignment or to accurately annotate cells that may serve as biomarkers in multiple disease contexts. A record of this paper's transparent peer review process is included in the supplemental information.

Imaging mass cytometry and multiplatform genomics define the phenogenomic landscape of breast cancer.

Genomic alterations shape cell phenotypes and the structure of tumor ecosystems in poorly defined ways. To investigate these relationships, we used imaging mass cytometry to quantify the expression of 37 proteins with subcellular spatial resolution in 483 tumors from the METABRIC cohort. Single-cell analysis revealed cell phenotypes spanning epithelial, stromal and immune types. Distinct combinations of cell phenotypes and cell-cell interactions were associated with genomic subtypes of breast cancer. Epithelial luminal cell phenotypes separated into those predominantly impacted by mutations and those affected by copy number aberrations. Several features of tumor ecosystems, including cellular neighborhoods, were linked to prognosis, illustrating their clinical relevance. In summary, systematic analysis of single-cell phenotypic and spatial correlates of genomic alterations in cancer revealed how genomes shape both the composition and architecture of breast tumor ecosystems and will enable greater understanding of the phenotypic impact of genomic alterations.

Simultaneous Multiplexed Imaging of mRNA and Proteins with Subcellular Resolution in Breast Cancer Tissue Samples by Mass Cytometry.

To build comprehensive models of cellular states and interactions in normal and diseased tissue, genetic and proteomic information must be extracted with single-cell and spatial resolution. Here, we extended imaging mass cytometry to enable multiplexed detection of mRNA and proteins in tissues. Three mRNA target species were detected by RNAscope-based metal in situ hybridization with simultaneous antibody detection of 16 proteins. Analysis of 70 breast cancer samples showed that HER2 and CK19 mRNA and protein levels are moderately correlated on the single-cell level, but that only HER2, and not CK19, has strong mRNA-to-protein correlation on the cell population level. The chemoattractant CXCL10 was expressed in stromal cell clusters, and the frequency of CXCL10-expressing cells correlated with T cell presence. Our flexible and expandable method will allow an increase in the information content retrieved from patient samples for biomedical purposes, enable detailed studies of tumor biology, and serve as a tool to bridge comprehensive genomic and proteomic tissue analysis.

Ubiquitin ligase RNF8 suppresses Notch signaling to regulate mammary development and tumorigenesis.

The E3 ubiquitin ligase RNF8 plays critical roles in maintaining genomic stability by promoting the repair of DNA double-strand breaks (DSBs) through ubiquitin signaling. Abnormal activation of Notch signaling and defective repair of DSBs promote breast cancer risk. Here, we found that low expression of the full-length RNF8 correlated with poor prognosis for breast cancer patients. Our data revealed that in addition to its role in the repair of DSBs, RNF8 regulated Notch1 signaling and cell-fate determination of mammary luminal progenitors. Mechanistically, RNF8 acted as a negative regulator of Notch signaling by ubiquitylating the active NOTCH1 protein (N1ICD), leading to its degradation. Consistent with abnormal activation of Notch signaling and impaired repair of DSBs in Rnf8-mutant mammary epithelial cells, we observed increased risk of mammary tumorigenesis in mouse models for RNF8 deficiency. Notably, deficiency of RNF8 sensitized breast cancer cells to combination of pharmacological inhibitors of Notch signaling and poly(ADP-ribose) polymerase (PARP), suggesting implications for treatment of breast cancer associated with impaired RNF8 expression or function.

TIMPs: versatile extracellular regulators in cancer.

A compelling long-term goal of cancer biology is to understand the crucial players during tumorigenesis in order to develop new interventions. Here, we review how the four non-redundant tissue inhibitors of metalloproteinases (TIMPs) regulate the pericellular proteolysis of a vast range of matrix and cell surface proteins, generating simultaneous effects on tumour architecture and cell signalling. Experimental studies demonstrate the contribution of TIMPs to the majority of cancer hallmarks, and human cancers invariably show TIMP deregulation in the tumour or stroma. Of the four TIMPs, TIMP1 overexpression or TIMP3 silencing is consistently associated with cancer progression or poor patient prognosis. Future efforts will align mouse model systems with changes in TIMPs in patients, will delineate protease-independent TIMP function, will pinpoint therapeutic targets within the TIMP-metalloproteinase-substrate network and will use TIMPs in liquid biopsy samples as biomarkers for cancer prognosis.

Tumor suppression by stromal TIMPs.

The tumor stroma has the capacity to drive cancer progression, although the mechanisms governing these effects are incompletely understood. Recently, we reported that deletion of tissue inhibitor of metalloproteinases (Timps) in fibroblasts unleashes the function of cancer-associated fibroblasts and identifies a novel mode of stromal-tumor communication that activates key oncogenic pathways invoving Notch and ras homolog gene family, member A (RhoA) via stromal exosomes.