Atlas of clinically distinct cell states and ecosystems across human solid tumors.

Determining how cells vary with their local signaling environment and organize into distinct cellular communities is critical for understanding processes as diverse as development, aging, and cancer. Here we introduce EcoTyper, a machine learning framework for large-scale identification and validation of cell states and multicellular communities from bulk, single-cell, and spatially resolved gene expression data. When applied to 12 major cell lineages across 16 types of human carcinoma, EcoTyper identified 69 transcriptionally defined cell states. Most states were specific to neoplastic tissue, ubiquitous across tumor types, and significantly prognostic. By analyzing cell-state co-occurrence patterns, we discovered ten clinically distinct multicellular communities with unexpectedly strong conservation, including three with myeloid and stromal elements linked to adverse survival, one enriched in normal tissue, and two associated with early cancer development. This study elucidates fundamental units of cellular organization in human carcinoma and provides a framework for large-scale profiling of cellular ecosystems in any tissue.

Landscape of innate lymphoid cells in human head and neck cancer reveals divergent NK cell states in the tumor microenvironment.

Natural killer (NK) cells comprise one subset of the innate lymphoid cell (ILC) family. Despite reported antitumor functions of NK cells, their tangible contribution to tumor control in humans remains controversial. This is due to incomplete understanding of the NK cell states within the tumor microenvironment (TME). Here, we demonstrate that peripheral circulating NK cells differentiate down two divergent pathways within the TME, resulting in different end states. One resembles intraepithelial ILC1s (ieILC1) and possesses potent in vivo antitumor activity. The other expresses genes associated with immune hyporesponsiveness and has poor antitumor functional capacity. Interleukin-15 (IL-15) and direct contact between the tumor cells and NK cells are required for the differentiation into CD49a+CD103+ cells, resembling ieILC1s. These data explain the similarity between ieILC1s and tissue-resident NK cells, provide insight into the origin of ieILC1s, and identify the ieILC1-like cell state within the TME to be the NK cell phenotype with the greatest antitumor activity. Because the proportions of the different ILC states vary between tumors, these findings provide a resource for the clinical study of innate immune responses against tumors and the design of novel therapy.

A novel stratification framework for predicting outcome in patients with prostate cancer.

BACKGROUND: Unsupervised learning methods, such as Hierarchical Cluster Analysis, are commonly used for the analysis of genomic platform data. Unfortunately, such approaches ignore the well-documented heterogeneous composition of prostate cancer samples. Our aim is to use more sophisticated analytical approaches to deconvolute the structure of prostate cancer transcriptome data, providing novel clinically actionable information for this disease. METHODS: We apply an unsupervised model called Latent Process Decomposition (LPD), which can handle heterogeneity within individual cancer samples, to genome-wide expression data from eight prostate cancer clinical series, including 1,785 malignant samples with the clinical endpoints of PSA failure and metastasis. RESULTS: We show that PSA failure is correlated with the level of an expression signature called DESNT (HR = 1.52, 95% CI = [1.36, 1.7], P = 9.0 × 10-14, Cox model), and that patients with a majority DESNT signature have an increased metastatic risk (X2 test, P = 0.0017, and P = 0.0019). In addition, we develop a stratification framework that incorporates DESNT and identifies three novel molecular subtypes of prostate cancer. CONCLUSIONS: These results highlight the importance of using more complex approaches for the analysis of genomic data, may assist drug targeting, and have allowed the construction of a nomogram combining DESNT with other clinical factors for use in clinical management.

A single cell framework identifies functionally and molecularly distinct multipotent progenitors in adult human hematopoiesis.

Hematopoietic multipotent progenitors (MPPs) regulate blood cell production to appropriately meet the biological demands of the human body. Human MPPs remain ill-defined whereas mouse MPPs have been well characterized with distinct immunophenotypes and lineage potencies. Using multiomic single cell analyses and complementary functional assays, we identified new human MPPs and oligopotent progenitor populations within Lin-CD34+CD38dim/lo adult bone marrow with distinct biomolecular and functional properties. These populations were prospectively isolated based on expression of CD69, CLL1, and CD2 in addition to classical markers like CD90 and CD45RA. We show that within the canonical Lin-CD34+CD38dim/loCD90CD45RA-MPP population, there is a CD69+ MPP with long-term engraftment and multilineage differentiation potential, a CLL1+ myeloid-biased MPP, and a CLL1-CD69-erythroid-biased MPP. We also show that the canonical Lin-CD34+CD38dim/loCD90-CD45RA+ LMPP population can be separated into a CD2+ LMPP with lymphoid and myeloid potential, a CD2-LMPP with high lymphoid potential, and a CLL1+ GMP with minimal lymphoid potential. We used these new HSPC profiles to study human and mouse bone marrow cells and observe limited cell type specific homology between humans and mice and cell type specific changes associated with aging. By identifying and functionally characterizing new adult MPP sub-populations, we provide an updated reference and framework for future studies in human hematopoiesis.

AML/T cell interactomics uncover correlates of patient outcomes and the key role of ICAM1 in T cell killing of AML.

T cells are important for the control of acute myeloid leukemia (AML), a common and often deadly malignancy. We observed that some AML patient samples are resistant to killing by human-engineered cytotoxic CD4+ T cells. Single-cell RNA-seq of primary AML samples and CD4+ T cells before and after their interaction uncovered transcriptional programs that correlate with AML sensitivity or resistance to CD4+ T cell killing. Resistance-associated AML programs were enriched in AML patients with poor survival, and killing-resistant AML cells did not engage T cells in vitro. Killing-sensitive AML potently activated T cells before being killed, and upregulated ICAM1, a key component of the immune synapse with T cells. Without ICAM1, killing-sensitive AML became resistant to killing by primary ex vivo-isolated CD8+ T cells in vitro, and engineered CD4+ T cells in vitro and in vivo. While AML heterogeneity implies that multiple factors may determine their sensitivity to T cell killing, these data show that ICAM1 acts as an immune trigger, allowing T cell killing, and could play a role in AML patient survival in vivo.

Spatially Segregated Macrophage Populations Predict Distinct Outcomes in Colon Cancer.

Tumor-associated macrophages are transcriptionally heterogeneous, but the spatial distribution and cell interactions that shape macrophage tissue roles remain poorly characterized. Here, we spatially resolve five distinct human macrophage populations in normal and malignant human breast and colon tissue and reveal their cellular associations. This spatial map reveals that distinct macrophage populations reside in spatially segregated micro-environmental niches with conserved cellular compositions that are repeated across healthy and diseased tissue. We show that IL4I1+ macrophages phagocytose dying cells in areas with high cell turnover and predict good outcome in colon cancer. In contrast, SPP1+ macrophages are enriched in hypoxic and necrotic tumor regions and portend worse outcome in colon cancer. A subset of FOLR2+ macrophages is embedded in plasma cell niches. NLRP3+ macrophages co-localize with neutrophils and activate an inflammasome in tumors. Our findings indicate that a limited number of unique human macrophage niches function as fundamental building blocks in tissue. Significance: This work broadens our understanding of the distinct roles different macrophage populations may exert on cancer growth and reveals potential predictive markers and macrophage population-specific therapy targets.

Sarcoma microenvironment cell states and ecosystems are associated with prognosis and predict response to immunotherapy.

Characterization of the diverse malignant and stromal cell states that make up soft tissue sarcomas and their correlation with patient outcomes has proven difficult using fixed clinical specimens. Here, we employed EcoTyper, a machine-learning framework, to identify the fundamental cell states and cellular ecosystems that make up sarcomas on a large scale using bulk transcriptomes with clinical annotations. We identified and validated 23 sarcoma-specific, transcriptionally defined cell states, many of which were highly prognostic of patient outcomes across independent datasets. We discovered three conserved cellular communities or ecotypes associated with underlying genomic alterations and distinct clinical outcomes. We show that one ecotype defined by tumor-associated macrophages and epithelial-like malignant cells predicts response to immune-checkpoint inhibition but not chemotherapy and validate our findings in an independent cohort. Our results may enable identification of patients with soft tissue sarcomas who could benefit from immunotherapy and help develop new therapeutic strategies.

Profiling Cellular Ecosystems at Single-Cell Resolution and at Scale with EcoTyper.

Tissues are composed of diverse cell types and cellular states that organize into distinct ecosystems with specialized functions. EcoTyper is a collection of machine learning tools for the large-scale delineation of cellular ecosystems and their constituent cell states from bulk, single-cell, and spatially resolved gene expression data. In this chapter, we provide a primer on EcoTyper and demonstrate its use for the discovery and recovery of cell states and ecosystems from healthy and diseased tissue specimens.

A spatial map of human macrophage niches reveals context-dependent macrophage functions in colon and breast cancer.

Tumor-associated macrophages (TAMs) display heterogeneous phenotypes. Yet the exact tissue cues that shape macrophage functional diversity are incompletely understood. Here we discriminate, spatially resolve and reveal the function of five distinct macrophage niches within malignant and benign breast and colon tissue. We found that SPP1 TAMs reside in hypoxic and necrotic tumor regions, and a novel subset of FOLR2 tissue resident macrophages (TRMs) supports the plasma cell tissue niche. We discover that IL4I1 macrophages populate niches with high cell turnover where they phagocytose dying cells. Significantly, IL4I1 TAMs abundance correlates with anti-PD1 treatment response in breast cancer. Furthermore, NLRP3 inflammasome activation in NLRP3 TAMs correlates with neutrophil infiltration in the tumors and is associated with poor outcome in breast cancer patients. This suggests the NLRP3 inflammasome as a novel cancer immunetherapy target. Our work uncovers context-dependent roles of macrophage subsets, and suggests novel predictive markers and macrophage subset-specific therapy targets.

AML/T cell interactomics uncover correlates of patient outcomes and the key role of ICAM1 in T cell killing of AML.

T cells are important for the control of acute myeloid leukemia (AML), a common and often deadly malignancy. We observed that some AML patient samples are resistant to killing by human engineered cytotoxic CD4 + T cells. Single-cell RNA-seq of primary AML samples and CD4 + T cells before and after their interaction uncovered transcriptional programs that correlate with AML sensitivity or resistance to CD4 + T cell killing. Resistance-associated AML programs were enriched in AML patients with poor survival, and killing-resistant AML cells did not engage T cells in vitro . Killing-sensitive AML potently activated T cells before being killed, and upregulated ICAM1 , a key component of the immune synapse with T cells. Without ICAM1, killing-sensitive AML became resistant to killing to primary ex vivo -isolated CD8 + T cells in vitro , and engineered CD4 + T cells in vitro and in vivo . Thus, ICAM1 on AML acts as an immune trigger, allowing T cell killing, and could affect AML patient survival in vivo . SIGNIFICANCE: AML is a common leukemia with sub-optimal outcomes. We show that AML transcriptional programs correlate with susceptibility to T cell killing. Killing resistance-associated AML programs are enriched in patients with poor survival. Killing-sensitive, but not resistant AML activate T cells and upregulate ICAM1 that binds to LFA-1 on T cells, allowing immune synapse formation which is critical for AML elimination.

T cell characteristics associated with toxicity to immune checkpoint blockade in patients with melanoma.

Severe immune-related adverse events (irAEs) occur in up to 60% of patients with melanoma treated with immune checkpoint inhibitors (ICIs). However, it is unknown whether a common baseline immunological state precedes irAE development. Here we applied mass cytometry by time of flight, single-cell RNA sequencing, single-cell V(D)J sequencing, bulk RNA sequencing and bulk T cell receptor (TCR) sequencing to study peripheral blood samples from patients with melanoma treated with anti-PD-1 monotherapy or anti-PD-1 and anti-CTLA-4 combination ICIs. By analyzing 93 pre- and early on-ICI blood samples and 3 patient cohorts (n = 27, 26 and 18), we found that 2 pretreatment factors in circulation-activated CD4 memory T cell abundance and TCR diversity-are associated with severe irAE development regardless of organ system involvement. We also explored on-treatment changes in TCR clonality among patients receiving combination therapy and linked our findings to the severity and timing of irAE onset. These results demonstrate circulating T cell characteristics associated with ICI-induced toxicity, with implications for improved diagnostics and clinical management.

The landscape of tumor cell states and ecosystems in diffuse large B cell lymphoma.

Biological heterogeneity in diffuse large B cell lymphoma (DLBCL) is partly driven by cell-of-origin subtypes and associated genomic lesions, but also by diverse cell types and cell states in the tumor microenvironment (TME). However, dissecting these cell states and their clinical relevance at scale remains challenging. Here, we implemented EcoTyper, a machine-learning framework integrating transcriptome deconvolution and single-cell RNA sequencing, to characterize clinically relevant DLBCL cell states and ecosystems. Using this approach, we identified five cell states of malignant B cells that vary in prognostic associations and differentiation status. We also identified striking variation in cell states for 12 other lineages comprising the TME and forming cell state interactions in stereotyped ecosystems. While cell-of-origin subtypes have distinct TME composition, DLBCL ecosystems capture clinical heterogeneity within existing subtypes and extend beyond cell-of-origin and genotypic classes. These results resolve the DLBCL microenvironment at systems-level resolution and identify opportunities for therapeutic targeting (https://ecotyper.stanford.edu/lymphoma).

Convergence of Prognostic Gene Signatures Suggests Underlying Mechanisms of Human Prostate Cancer Progression.

The highly heterogeneous clinical course of human prostate cancer has prompted the development of multiple RNA biomarkers and diagnostic tools to predict outcome for individual patients. Biomarker discovery is often unstable with, for example, small changes in discovery dataset configuration resulting in large alterations in biomarker composition. Our hypothesis, which forms the basis of this current study, is that highly significant overlaps occurring between gene signatures obtained using entirely different approaches indicate genes fundamental for controlling cancer progression. For prostate cancer, we found two sets of signatures that had significant overlaps suggesting important genes (p < 10-34 for paired overlaps, hypergeometrical test). These overlapping signatures defined a core set of genes linking hormone signalling (HES6-AR), cell cycle progression (Prolaris) and a molecular subgroup of patients (PCS1) derived by Non Negative Matrix Factorization (NNMF) of control pathways, together designated as SIG-HES6. The second set (designated SIG-DESNT) consisted of the DESNT diagnostic signature and a second NNMF signature PCS3. Stratifications using SIG-HES6 (HES6, PCS1, Prolaris) and SIG-DESNT (DESNT) classifiers frequently detected the same individual high-risk cancers, indicating that the underlying mechanisms associated with SIG-HES6 and SIG-DESNT may act together to promote aggressive cancer development. We show that the use of combinations of a SIG-HES6 signature together with DESNT substantially increases the ability to predict poor outcome, and we propose a model for prostate cancer development involving co-operation between the SIG-HES6 and SIG-DESNT pathways that has implication for therapeutic design.

Determining cell type abundance and expression from bulk tissues with digital cytometry.

Single-cell RNA-sequencing has emerged as a powerful technique for characterizing cellular heterogeneity, but it is currently impractical on large sample cohorts and cannot be applied to fixed specimens collected as part of routine clinical care. We previously developed an approach for digital cytometry, called CIBERSORT, that enables estimation of cell type abundances from bulk tissue transcriptomes. We now introduce CIBERSORTx, a machine learning method that extends this framework to infer cell-type-specific gene expression profiles without physical cell isolation. By minimizing platform-specific variation, CIBERSORTx also allows the use of single-cell RNA-sequencing data for large-scale tissue dissection. We evaluated the utility of CIBERSORTx in multiple tumor types, including melanoma, where single-cell reference profiles were used to dissect bulk clinical specimens, revealing cell-type-specific phenotypic states linked to distinct driver mutations and response to immune checkpoint blockade. We anticipate that digital cytometry will augment single-cell profiling efforts, enabling cost-effective, high-throughput tissue characterization without the need for antibodies, disaggregation or viable cells.

DESNT: A Poor Prognosis Category of Human Prostate Cancer.

BACKGROUND: A critical problem in the clinical management of prostate cancer is that it is highly heterogeneous. Accurate prediction of individual cancer behaviour is therefore not achievable at the time of diagnosis leading to substantial overtreatment. It remains an enigma that, in contrast to breast cancer, unsupervised analyses of global expression profiles have not currently defined robust categories of prostate cancer with distinct clinical outcomes. OBJECTIVE: To devise a novel classification framework for human prostate cancer based on unsupervised mathematical approaches. DESIGN, SETTING, AND PARTICIPANTS: Our analyses are based on the hypothesis that previous attempts to classify prostate cancer have been unsuccessful because individual samples of prostate cancer frequently have heterogeneous compositions. To address this issue, we applied an unsupervised Bayesian procedure called Latent Process Decomposition to four independent prostate cancer transcriptome datasets obtained using samples from prostatectomy patients and containing between 78 and 182 participants. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Biochemical failure was assessed using log-rank analysis and Cox regression analysis. RESULTS AND LIMITATIONS: Application of Latent Process Decomposition identified a common process in all four independent datasets examined. Cancers assigned to this process (designated DESNT cancers) are characterized by low expression of a core set of 45 genes, many encoding proteins involved in the cytoskeleton machinery, ion transport, and cell adhesion. For the three datasets with linked prostate-specific antigen failure data following prostatectomy, patients with DESNT cancer exhibited poor outcome relative to other patients (p=2.65×10-5, p=4.28×10-5, and p=2.98×10-8). When these three datasets were combined the independent predictive value of DESNT membership was p=1.61×10-7 compared with p=1.00×10-5 for Gleason sum. A limitation of the study is that only prediction of prostate-specific antigen failure was examined. CONCLUSIONS: Our results demonstrate the existence of a novel poor prognosis category of human prostate cancer and will assist in the targeting of therapy, helping avoid treatment-associated morbidity in men with indolent disease. PATIENT SUMMARY: Prostate cancer, unlike breast cancer, does not have a robust classification framework. We propose that this failure has occurred because prostate cancer samples selected for analysis frequently have heterozygous compositions (individual samples are made up of many different parts that each have different characteristics). Applying a mathematical approach that can overcome this problem we identify a novel poor prognosis category of human prostate cancer called DESNT.