Spatial signature to predict hepatocellular carcinoma (HCC) recurrence after resection

Abstract

Provided herein, among other things, is a method for identifying a cancer patient that is at high risk for cancer recurrence. In some embodiments, the method may comprise analyzing a section of a tumor obtained from the cancer patient to quantify interactions between cancer stem cells and PDL1+ macrophages. In these embodiments, a higher level of interactions between cancer stem cells and PDL1+ macrophages in the tumor indicates that the patient is at high risk for cancer recurrence. In some embodiments, the method may comprise identifying the patient has having a high risk of cancer recurrence based on the results of this analysis and administering an adjuvant therapy to the patient.

Description

[0001] SPATIAL SIGNATURE TO PREDICT HEPATOCELLULAR CARCINOMA (HCC) RECURRENCE AFTER RESECTION

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] Pursuant to 35 U.S.C. § 119 (e), this application claims priority to the filing date of United States Provisional Patent Application Serial No. 63 / 674,648 filed July 23, 2024, the disclosure of which application is incorporated herein by reference in its entirety.

[0004] BACKGROUND

[0005] Hepatocellular carcinoma (HCC) has a dismal prognosis with a 5-year survival rate under 20%. Surgical resection is globally the most common surgery for early-stage HCC in patients with compensated liver disease, offering a hope for cure. However, this hope for cure is often undermined by high recurrence rates of 50-70%. Early recurrence, which occurs within 2 years of surgery, typically stems from invasive HCC that leaves behind residual disease after surgery. This link is supported by genetic studies confirming clonal concordance between the primary and recurrent tumor. Further evidence comes from a recent study showing that adjuvant therapy with combined blockade of VEGF and PDL1 can decrease recurrence from high-risk tumors. However, this therapy led to severe adverse events in 41% of patients. In order to select the right patients for adjuvant therapy, one needs to accurately identify patients at high risk for recurrence. However, there currently are no biomarkers to predict post-surgical HCC recurrence.

[0006] SUMMARY

[0007] Provided herein, among other things, is a method for identifying a cancer patient that is at high risk for cancer recurrence. In some embodiments, the method may comprise analyzing a section of a tumor obtained from the cancer patient to quantify interactions between cancer stem cells and PDL1+macrophages. In these embodiments, a higher level of interactions between cancer stem cells and PDL1+macrophages in the tumor indicates that the patient is at high risk for cancer recurrence. In some embodiments, the method may comprise identifying the patient as having a high risk of cancer recurrence based on the results of the analysis and administering an adjuvant therapy to the patient.

[0008] In any embodiment, the interactions are quantified by measuring the spatial proximity of two cells types, and the higher level of interactions is relative to reference samples that include samples from patients that have a low risk of cancer recurrence and high risk for cancer recurrence.

[0009] In any embodiment, the interactions between cancer stem cells and PDL1+macrophages in the tumor are quantified to produce an interaction score. In these embodiments, the method comprises: comparing the interaction score to a threshold based obtained from the analysis of reference samples that include samples from patients that have a low risk of cancer recurrence and samples from patients that have a high risk for cancer recurrence, and identifying the patient as having a high risk for cancer recurrence based on the comparison. In these embodiments, the method may further include administering an adjuvant therapy to the patient.

[0010] These and other aspects of the invention are described in greater detail below.

[0011] BRIEF DESCRIPTION OF THE FIGURES

[0012] The skilled artisan will understand that the drawings described below are for illustration purposes only. The drawings are not intended to limit the scope of the present teachings in any way.

[0013] FIGS. 1A-1D: High dimensional single-cell spatial atlas of the tumor immune microenvironment in HCC. FIG. 1A: Single-cell spatial analysis workflow of human HCC tumor cores using CODEX to determine cell frequency, cell-cell interactions, and cellular neighborhoods. FIG. IB: Heat map dendrogram of unsupervised clustering using canonical marker expression to define major tumor and immune cell subsets. Heatmap scaled by column. FIG. 1C: Sub-classification of tumor cells, macrophages and T cells using cellspecific canonical markers. Heatmap is scaled by row. FIG. ID: Graphical representation of the 20 cell types and subtypes with their absolute cell counts and proportions in the CODEX dataset.

[0014] The validity of the unsupervised clustering for each cell type was verified CODEX imaging representations of the canonical markers (data not shown).

[0015] Abbreviations: abs- antibodies, CODEX-co-detection by indexing, HCC- hepatocellular carcinoma, PanCK- pan-cytokeratin, CK19- cytokeratin 19, EPCAM- epithelial cellular adhesion molecule, PDL1- programmed death ligand 1, PD1- programmed cell death protein 1 , CD- cluster of differentiation, HLA-DR-human leukocyte antigen, DR isotype, TIM3- T cell immunoglobulin and mucin domain-containing protein 3, DC- dendritic cell, NK-natural killer, Treg- regulatory T cell, Exh- Exhausted, Eff- Effector. FIGS. 2A-2H: Immunosuppressive Cell Enrichment and Tissue Remodeling in Residual HCC. FIG. 2A: Immune and tumor cell proportions in residual HCCs from chemotherapy-treated liver explants (residual HCC, n=55) compared to primary resected primary HCC (n = 53). FIG. 2B: H&E and CODEX imaging representations of a residual and primary HCC tumor core. In each CODEX image, eight canonical markers (DAPI, CD31, CD4, CD15, CD68, PanCK, CD8, and aSMA) are overlaid onto the image in different colors. FIG. 2C: Volcano plot showing the differential enrichment of tumor and immune cell populations (with statistical significance; shown with Bonferroni correction for multiple comparisons) in residual (n=55) vs primary HCCs (n=53). FIG. 2D: Proportion of PDL1+ macrophages, PDL1+ tumor cells, mast cells, and exhausted T cells in residual (n=55) vs primary HCCs (n=53). FIG. 2E: Representative Voronoi plot of cores showing PDL1+ macrophage, PDL1+ tumor cells, mast cells, and exhausted T cells density in residual (n=55) vs primary HCCs (n=53). FIG. 2F: Comparison of normalized expression of sternness marker CD44 and CK19 marker in tumor cells from residual (n=217,105) and primary HCC (n=297,608) (both pAdj value <3.OxlO-300). FIG. 2G: Comparison of normalized expression of CD1 lb and podoplanin in PDL1+ macrophages from residual (n=47,605) and primary HCC (n=6108) (both pAdj value <3.OxlO-300). FIG. 2H: Comparison of normalized expression of PD1 and TIM3 in T cells and NK cells from residual (T cells, n= l 6,549; NK cells n= 11,701) and primary HCC (T cells, n=45,494; NK cells n=211 l)(both pAdj value <3.0x10300).

[0016] Statistical significance was assessed by unpaired, two-tailed t-test, Benjamini- Hochberg (BH) adjustment was used for p values.

[0017] Abbreviations: HCC- hepatocellular carcinoma, H&E- hematoxylin and eosin, CODEX- co-detection by indexing, DAPI- 4',6-diamidino-2-phenylindole, CD- cluster of differentiation, PanCK- pan cytokeratin, aSMA-smooth muscle alpha actin, PDL1- programmed death ligand 1, BCL2- B-cell lymphoma 2, PD1- programmed cell death protein 1, TIM3- T cell immunoglobulin and mucin domain-containing protein 3, Exh- Exhausted, NK-natural killer, Sig- significant.

[0018] FIGS. 3A-3I: Remodeling of Spatial Interactions Between Tumor and Immune Cells in Residual HCC. FIG. 3A: Alluvial plot of tumor cell subtype and immune cell interactions in residual (n=55) and primary HCCs (n=53). Height of each unit is proportional to the frequency of interaction between two cell types. Only significant interactions with pAdj<0.05 are depicted. FIG. 3B: Size-modulated circular heatmap showing mean frequency and adjusted p values of direct interactions between tumor cell subtypes and PDL1+ macrophages in residual (n=55) and primary HCCs (n=53). FIG. 3C: Size-modulated circular heatmap showing mean frequency and adjusted p values of direct interactions between tumor cell subtypes and exhausted CD8T cells in residual (n=55) and primary HCCs (n=53). FIG. 3D: Kaplan Meier plots signifying the recurrence-free survival of tumors stratified by median frequency of interaction between EPCAM+ tumor cells and either CD206+ or PDL1+ macrophages in residual HCC (n=55). CODEX representative IF images demonstrate the interaction between representative cells. Log rank test used to statistically compare the groups. FIG. 3E: CODEX imaging representations of PDL1+ macrophage interactions with exhausted and effector CD8+ T cells, shown with box plots quantifying the proportion of interactions in primary (n=53) compared to residual HCC (n=55). FIG. 3F: CODEX imaging representations of PDL1+ macrophage interactions with fibroblasts and endothelial cells, shown with bar plots quantifying the proportion of interactions in primary (n=53) compared to residual HCC (n=55). FIG. 3G: Model of direct and indirect interactions with a single central cell using the scalar variable “cell-cell distance”. RA2 values for predicting Ki67 and BCL2 marker expression in radius of increasing sizes. Volcano plots show that increasing neighborhood sizes explain more variance in expression of a central cell’s Ki67 and BCL2. RA2 values for Ki67 are higher on average for a neighborhood radius of 100pm vs a neighborhood radius of 25pm (Wilcoxon rank sums pAdj-value 8.48 x 10"6), and for BCL2 (Wilcoxon rank sums pAdj-value 1.76 x 105). FIG. 3H: Size-modulated circular heatmap showing mean frequency and adjusted p values of indirect interactions between tumor cell subtypes and PDL1+ macrophages in residual (n=55) and primary HCCs (n=53). FIG. 31: Size-modulated circular heatmap showing mean frequency and adjusted p values of frequency of indirect interactions between tumor cell subtypes and exhausted CD8T cells in residual (n=55) and primary HCCs (n=53).

[0019] Statistical significance was assessed by unpaired, two-tailed t-test, Benjamini- Hochberg (BH) adjustment was used for p values.

[0020] Abbreviations: H&E- hematoxylin and eosin, CODEX- co-detection by indexing, HCC- hepatocellular carcinoma, DAPI- 4',6-diamidino-2-phenylindole, CD- cluster of differentiation, PanCK- pan cytokeratin, CK19- cytokeratin 19, EPCAM- epithelial cell adhesion molecule, aSMA-smooth muscle alpha actin, PDL1- programmed death ligand 1, BCL2- B-cell lymphoma 2, PD1- programmed cell death protein 1, TIM3- T cell immunoglobulin and mucin domain-containing protein 3, HLA-DR- Human Leukocyte Antigen - DR isotype, Treg-regulatory T cell, NK-natural killer, Exh-exhausted, Eff- effector, NK - natural killer

[0021] FIGS. 4A-4J: Synchronized Spatial Remodeling of Neighborhood Structures in Residual HCC. FIG. 4A: Schematic showing cellular neighborhood (CN) identification based on an iterative 10-cell clustering algorithm. Color codes show hypothetical spatial structures within the tumor microenvironment. FIG. 4B: Heatmap demonstrating the cellular compositions of the nine cellular neighborhoods defined in this study. FIG. 4C: Comparison of tumor and immune cell neighborhood distributions in primary (n=53) and residual HCCs (n=55). Box plots comparing the proportion of eight cellular neighborhoods in primary (n=53) and residual HCCs (n=55). FIG. 4D: Size-modulated circular heatmap showing mean expression of markers on y-axis in tumor cells and T cells in residual HCC within the M2- macrophage CN (tumor cells, n= 17046; T cells, n=4299) compared to the pauci-immune tumor CN (tumor cells, n= 109, 391) or T cell immune CN (T cells, n= 8345) respectively, adjusted p values shown next to the plot. Size-modulated circular heatmap showing mean expression of markers on y-axis on macrophages and T cells within the EpCAM-i- tumor cell CN (macrophages=272, n= 17046; T cells, n=13) and CK19+ tumor cell CNs (macrophages, n= 1201; T cells, n=354) compared to the pauci-immune tumor CN (macrophages, n= 4401; T cells, n=239) respectively, adjusted p values shown next to the plot. FIG. 4E: Kaplain- Meier plot showing recurrence-free survival in patients with tumors stratified based on the presence (n= 12) or absence (n=43) of EPCAM+ tumor cell CN in residual HCC (n=55 ). Log rank test used to statistically compare the groups. FIG. 4F: Schematic of in vitro 3D tumoroid co-culture of control and doxorubicin-resistant Huh7 HCC cell lines with THP1 macrophages followed by single-cell RNA sequencing followed by cell clustering and quantification. FIG. 4G: Characterizing the macrophage C3 and C4 between doxorubicin- resistant (DoxR) and control samples. Gene set expression analysis shows enrichment of the M2-like macrophage signature in the C3 cluster of macrophages enriched in the DoxR samples. FIG. 4H: Volcano plot shows mean expression of key differentially expressed genes in C3 versus C4 clusters. FIG. 41: Quantification of PDL1+ macrophages by immunofluorescence in 3D heterotypic tumoroids of control or doxorubicin-resistant (DoxR) Huh7 cancer cells with THP1 macrophages. FIG. 4G: Schematic of in vitro assay using primary HCC patient-derived 3D tumoroids and monocyte-derived macrophages from patients with HCC. Calcein staining demonstrates viability of 3D patient derived tumoroids. PDL1 expression in monocyte-derived macrophages treated with conditioned media from doxorubicin resistant patient-derived tumoroid (n=4) versus control (n=4). Unpaired t-tests used to compare the proportion of PDL1+ macrophages between the two groups.

[0022] Statistical significance was assessed by unpaired, two-tailed t-test, Benjamini- Hochberg (BH) adjustment was used for p values.

[0023] Abbreviations: H&E- hematoxylin and eosin, CODEX- co-detection by indexing, HCC- hepatocellular carcinoma, DAPI- 4',6-diamidino-2-phenylindole, CN-cellular neighborhood, CD- cluster of differentiation, PanCK- pan cytokeratin, CK19- cytokeratin 19, EPCAM- epithelial cell adhesion molecule, aSMA-smooth muscle alpha actin, PDL1- programmed death ligand 1, BCL2- B-cell lymphoma 2, PD1- programmed cell death protein 1, TIM3- T cell immunoglobulin and mucin domain-containing protein 3, HLA-DR- Human Leukocyte Antigen - DR isotype, Treg-regulatory T cell, NK-natural killer, Memmemory, Exh-exhausted.

[0024] ***p<0.001.

[0025] FIGS. 5A-5J: TGFP pathway activation promotes persistence of residual tumor cells in HCC. FIG. 5A: Workflow of Nanostring spatial transcriptomics analysis. The expression of targeted transcriptomes of cancer and immune-related genes were quantified in tumor cell and macrophage areas of interest (AOI). FIG. 5B: Representative immunofluorescent regions of interest (RO I) showing tumor cell and macrophage-enriched areas of interest (AOI) based on the expression of panCK, CD45, CD68, DAPI. FIG. 5C: Principal Component Analysis (PCA) and volcano plot showing differential expression of genes in the tumor cell (n=53) and macrophage (n=52) AOIs. FIG. 5D: Molecular pathways activated in tumor cells (n=19 vs. n=l l) and macrophage AOIs (n=18 vs. n=9) of residual HCC compared to non-tumorous liver samples. FIG. 5E: Schematic showing derivation of tumor cell and macrophage signatures from the spatial transcriptome data of primary and residual HCC. FIG. 5F: Kaplan-Meier curve showing the prognostic significance of tumor cell and macrophage signatures applied to a validation cohort of human HCC (n=334). Log rank test used to compare the groups. FIG. 5G: Upstream regulators of transcriptional regulators of gene expression changes in the tumor cell and macrophage AOIs of residual HCC compared to non-tumorous liver samples. FIG. 5H: Plot comparing TGFB1 gene expression levels in tumor cells (n= 19) and macrophage AOIs (n=18) of residual HC. FIG. 51: Correlation plot showing co-expression of SMAD2 and CD274 (PDL1 ) in residual HCC (n=38) but not primary HCC (n=48). Spearman test to correlate expression. FIG. 5J: Representative fluorescent mRNA FISH images along with quantification of TGFB1 mRNA expression in primary (n=51) and residual HCC (n=46). Representative fluorescent mRNA FISH images along with quantification of TGFB1 and CD68 mRNA expression in primary (n=31) and residual HCC (n=30).

[0026] Abbreviations: DAPI- 4',6-diamidino-2-phenylindole, TMA-tumor microarray, AOI- area of interest, CODEX- co-detection by indexing, rHCC- residual hepatocellular carcinoma, TGFB 1- transcription growth factor beta 1

[0027] DEFINITIONS

[0028] Unless defined otherwise herein, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are described.

[0029] All patents and publications, including all sequences disclosed within such patents and publications, referred to herein are expressly incorporated by reference.

[0030] Numeric ranges are inclusive of the numbers defining the range. Unless otherwise indicated, nucleic acids are written left to right in 5’ to 3’ orientation; amino acid sequences are written left to right in amino to carboxy orientation, respectively.

[0031] The headings provided herein are not limitations of the various aspects or embodiments of the invention. Accordingly, the terms defined immediately below are more fully defined by reference to the specification as a whole.

[0032] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Singleton, et al., DICTIONARY OF MICROBIOLOGY AND MOLECULAR BIOLOGY, 2D ED., John Wiley and Sons, New York (1994), and Hale & Markham, THE HARPER COLLINS DICTIONARY OF BIOLOGY, Harper Perennial, N.Y. (1991) provide one of ordinary skill in the art with the general meaning of many of the terms used herein. Still, certain terms are defined below for the sake of clarity and ease of reference.

[0033] A “plurality” contains at least 2 members. In certain cases, a plurality may have at least 2, at least 5, at least 10, at least 100, at least 1000, at least 10,000, at least 100,000, at least 106, at least 107, at least 108or at least 109or more members. In certain cases, a plurality may have 2 to 100 or 5 to 100 members.

[0034] As used herein, the term “labeling” refers to a step that results in binding of a binding agent to specific sites in a sample (e.g., sites containing an epitope for the binding agent (e.g., an antibody) being used, for example) such that the presence and / or abundance of the sites can be determined by evaluating the presence and / or abundance of the binding agent. The term “labeling” refers to a method for producing a labeled sample in which any necessary steps are performed in any convenient order, as long as the required labeled sample is produced. For example, in some embodiments and as will be exemplified below, a sample can be labeled using a plurality of binding agents that are each linked to an oligonucleotide.

[0035] As used herein, the term “tissue section” refers to a piece of tissue that has been obtained from a subject, fixed, sectioned, and mounted on a planar surface, e.g., a microscope slide or coverslip. In particular embodiments, the tissue section may be a section of a tissue biopsy obtained from a patient. Biopsies of interest include both tumor and non- neoplastic biopsies of skin (melanomas, carcinomas, lymphomas, etc.), soft tissue, bone, breast, colon, liver, kidney, adrenal, gastrointestinal, pancreatic, gall bladder, salivary gland, cervical, ovary, uterus, testis, prostate, lung, thymus, thyroid, parathyroid, pituitary (adenomas, etc.), brain, spinal cord, ocular, nerve, and skeletal muscle, etc.

[0036] As used herein, the term “formalin-fixed paraffin embedded (FFPE) tissue section” refers to a piece of tissue, e.g., a biopsy sample that has been obtained from a subject, fixed in formalin, embedded in wax, cut into thin sections, and then mounted on a microscope slide.

[0037] As used herein, the term “interactions” refers to a measure of the spatial proximity of two cells types. Specifically, to quantify interactions between two cell types one would measure the spatial proximity of two cells types relative to one another.

[0038] Other definitions of terms may appear throughout the specification.

[0039] DETAILED DESCRIPTION

[0040] As noted above, this disclosure provides, among other things, a method for identifying a cancer patient that is at high risk for cancer recurrence. In some embodiments, the method may comprise analyzing a section of a tumor obtained from the cancer patient to quantify interactions between cancer stem cells and PDL1+macrophages. In these embodiments, a higher level of interactions between cancer stem cells and PDL1+macrophages indicates that the patient is at high risk for cancer recurrence. In some embodiments, the method may comprise identifying the patient as having a high risk of cancer recurrence based on the results of the analysis and administering an adjuvant therapy to the patient. For example, in some embodiments, the adjuvant therapy may comprise an immune checkpoint therapy and / or chemotherapy, e.g., (i) atezolizumab and bevacizumab, (ii) durvalumab and tremelimumab or (iii) nivolumab and ipilimumab. Further adjuvant therapies are described below. Kits for performing the method are also provided.

[0041] In any embodiment, the analysis is done by performing a multiplexed binding assay on the section to identify at least: (i) the cancer stem cells and (ii) the PDL1+ macrophages. The multiplexed binding assay may be done by detecting binding of at least 10, at least 20, at least 50, up to 100 or 300 binding agents (e.g., antibodies) to a tissue section, e.g., a crosslinked tissue section such as an FFPE section. Methods for performing multiplexed binding assay include, but are not limited to, multiplex colorimetric immunohistochemistry (mCIHC), multiplex immunofluorescence (mIF), cyclic immunofluorescence (CycIF), iterative indirect immunofluorescent imaging (4i), imaging mass cytometry (IMC), multiplexed ion beam imaging (MIB1), codetection by indexing (CODEX), and digital spatial profiling (DSP). These methods are reviewed in, e.g., Patel et al (Methods Mol Biol. 2020 2055:455-465) and Francisco-Cruz et al (Methods Mol. Biol. 2020 2055: 467- 495). The general principles of CycIF are described in Rashid et al. Sci. Data. 2019 6: 323 and Lin et al. eLife 2018 10: 31657. The general principles of 4i are described in Gut et al (Science 2018 361: 1-13). The general principles of IMC and MIBI, which are mass- spectrometry approaches for performing multiplexed tissue labeling, are described in a variety of publications including, but not limited to Angelo et al. (Nature Medicine 2014 20:436), Rost et al. (Lab. Invest. 2017 97: 992-1003) US9,766,224, US9,312,111 and US2015 / 0080233, among many others.

[0042] The general principles of CODEX are described in Goltsev et al. (Cell 2018 174: 968-981), Schiirch et al. (bioRxiv 2019 743989) and US20180030504. CODEX-based implementations of the method may involve (a) obtaining: i. a plurality of capture agents (e.g., 20-200 antibodies) that are each linked to a different oligonucleotide; and ii. a corresponding plurality of labeled nucleic acid probes, wherein each of the labeled nucleic acid probes specifically hybridizes with only one of the oligonucleotides; (b) labeling the sample with the plurality of capture agents; (c) specifically hybridizing a first sub-set (e.g., 2, 3, or 4) of the labeled nucleic acid probes of (a)(ii) with the sample, wherein the probes in the first sub-set are distinguishably labeled, to produce labeled probe / oligonucleotide duplexes; (d) reading the sample to obtain an image showing binding pattern for each of the probes hybridized in step (c); (e) inactivating or removing the labels that are associated with the sample in step (c), leaving the plurality of capture agents of (b) and their associated oligonucleotides still bound to the sample; and (f) repeating steps (c) and (d) multiple times with a different sub-set of the labeled nucleic acid probes of (a)(ii), each repeat followed by step (e) except for the final repeat, to produce a plurality of images of the sample, each image corresponding to a sub-set of labeled nucleic acid probes used in (c). In these embodiments, multiple images may be registered and superimposed. Using any method, the image(s) provide information on the amount of each antibody that is bound to the sample as well as the location of the epitope to which it binds.

[0043] Immunological markers to identify cancer cells are generally well known and available for many types of cancers (see, generally, Painter et al, Toxicol. Pathol. 2010 38: 131-141 and Bahrami et al Arch Pathol Lab Med. 2008 132:326-48, among many others). For example, the cancer cells identified in step (a) may be i. melanoma cells identified by expression of one or more of the following markers: S-100, Melan-A, SoxlO, MITF, tyrosinase, and HMB45 (e.g., S-100, Melan-A, SoxlO and HMB45); ii. carcinoma cells identified by the expression of one or more of the following markers: pan-cytokeratin (CK), CK7, CK20, CK5 / 6, CK8 / 18, napsin A, TTF-1, PSA, PSMA, CDX2, GAT A3, synaptophysin, chromogranin A, NSE, EpCAM, and MUC-1 (e.g., CK7, CK20, TTF-1, PSA, CDX2, GATA3); iii. lymphoma / leukemia cells identified by the expression of one or more of the following markers: CD45, CD3, PAX5, CD20, Myc, CyclinDl, BCL-2, BCL-6, IRF4, CD138, CD30, kappa, lambda, TdT, CD10, ALK, and lysoszyme (e.g., CD45, PAX5, CD20, Myc, CyclinDl, BCL-2, BCL-6, IRF4, CD138, and CD30); iv. sarcoma / mesothelioma cells identified by the expression of one or more of the following markers: vimentin, SMA, desmin, caldesmin, MyoDl, CD34, calretinin, podoplanin, and CD47 (e.g., vimentin, SMA, desmin, CD34); v. glioma cells / neural tumor cells identified by the expression of one or more of the following markers: GFAP, IDH-1(R132H), neurofilament, and NeuN (e.g, GFAP, IDH-1(R132H)); or vi. germ cell tumor cells identified by the expression of one or more of the following markers: beta-HCG, OCT4, SALL4, PLAP, inhibin A, HPL and AFP. Many alternative panels of antibodies can be used, depending on the cancer and other factors.

[0044] In any embodiment, the cells may be stained for DAPI (for nuclear staining). PanCK defined epithelial cells.

[0045] In some embodiments, the cancer stem cells are positive for PanCK, CK19 and EpCAM, and, optionally, CD44, although other markers may be used, e.g., PanCK and at least one of CK19, EpCAM and CD44. Likewise, the PDL1+macrophages may be positive for CD68, CD163, CD206, and PDL1. PDL1+macrophages can also be identified because they are CD68+ / PDLl+ / PanCK-ve. In some embodiments, the cancer patient may have hepatocellular carcinoma (HCC). In other embodiments, the cancer patient may have melanoma, another type of carcinoma, sarcoma cells or glioma. For example, the cancer may be melanoma, lung cancer, breast cancer, head and neck cancer, bladder cancer, Merkel cell cancer, cervical cancer, gastric cancer, cutaneous squamous cell cancer, colorectal carcinoma, pancreatic carcinoma, gastric or breast carcinoma, for which the markers are known. In some embodiments, the cancer stem cells are CD44Hi.

[0046] In any embodiment the patient may be treated with surgical resection and the method may be used to determine if the patient should receive an adjuvant therapy in addition to the surgical resection.

[0047] Interactions between the cells at the tumor edge (outermost area of tumor where it meets the normal liver, e.g., up to 200 micron of the tumor) can be measured by a variety of different methods. In some embodiments, the method may involve identifying single cells based on their nuclei. Cell types are determined through specific marker expressions: (i) PanCK, CK19, and EpCAM and optionally CD44 or (ii) PanCK and at least one of CK19, EpCAM and CD44 may be used to identify stem cells, while CD68, CD206, and PDL1 are used to identify macrophages. PDL1+macrophages can also be identified because they are CD68+ / PDLl+ / PanCK-ve. By utilizing the x and y coordinates, the distances between cells can be calculated, defining all cells within a 25-micron radius of a center cell as directly interacting. The frequency of direct interactions between stem-like cancer cells and M2-like macrophages in each sample can be calculated. Tumors are classified into high and low-risk categories based on the median frequency of these interactions. In this disclosure, interactions between two cell types can be measured by their spatial proximity. Specifically, if cells are close to one another, then they are deemed to interact with one another. The median frequency of the interactions may be provided as a numerical score, which score can be compared to a threshold obtained from the control samples that have been analyzed using similar methods.

[0048] These interactions occur throughout the tumor and their increased interactions predict tumor recurrence. However, the interactions occur more frequently at the tumor edge (i.e., the outermost area of tumor where it meets the normal tissue, e.g., up to 200 micron of the tumor at the tumor edge) and, as such, the analysis may be done on cells that are at the tumor edge.

[0049] A number of image processing tools can potentially be used in this analysis. For example, the cells in the image can be segmented, meaning that the boundaries or edges of the cells are defined. In some cases, image segmentation may be done for only the cells being analyzed. However, in other embodiments, all cells in the image may be segmented. Image segmentation may be by any a variety of techniques. In some embodiments, the cells may be segmented using a watershed algorithm (see, e.g., Al-Lofahi et al. BMC Bioinformatics. 2018 19: 365) although many other may be used. In watershed-based segmentation, the contents of each cell’s nucleus are identified by a nuclear staining, such as by DRAQ-5 or Hoechst. The watershed algorithm identifies each nucleus and draws a border around it, allowing cells to be detected and touching cells to be separated. In addition to the watershed algorithm, other algorithms include manual tracing of cells, levelset method, morphology-based segmentation, active contours model, snake algorithm, and more recently, deep learning techniques. Segmentation methods that can be used to define the edges of cells in highly multiplexed tissue images are described in a variety of publications, including Schuffler et al. (Cytometry A. 2015 87: 936-42) and Wang et al. (Am J Pathol. 2019 189:1686-1698). After segmentation, the binding pattern of the binding agents to the cells (which, in turn, reflects the presence and abundance of the epitopes to which agents bind) are used to discriminate the cells types and compute their numbers, distribution, and proximity.

[0050] After the cells are segmented and the cell types are identified, the interactions between the cells can be measured.

[0051] If used, the immune checkpoint inhibitor may be an antibody that binds to CTLA-4, PD1, PD-L1, TIM-3, VISTA, LAG-3, IDO or KIR. In these embodiments, the immune checkpoint inhibitor may be an antibody, e.g., an anti-CTLA-4 antibody, anti-PDl antibody, an anti-PD-Ll antibody, an anti-TIM-3 antibody, an anti-VISTA antibody, an anti-LAG-3 antibody, an anti-IDO antibody, or an anti-KIR antibody, although others are known, where the term “antibody” is intended to include nanobodies, phage display antibodies, single chain antibodies, bi-specifics, etc. In some embodiments, the immunotherapy may also include a co-stimulatory antibody such as an antibody against CD40, GITR, 0X40, CD137, or ICOS, for example. In some embodiments, the antibody may be an anti-PD-1 antibody, an anti-PD-Ll antibody or an anti-CTLA-4 antibody. Examples of such antibodies include, but are not limited to: Ipilimumab (CTLA-4), Nivolumab (PD-l), Pembrolizumab (PD-1), Atezolizumab (PD-L1), Avelumab (PD-L1), and Durvalumab (PD-L1). These therapies may be combined with one another and with other therapies. In some embodiments, the dose administered may be in the range of 1 mg / kg to 10 mg / kg, or in the range of 50 mg to 1.5g every few weeks (e.g., every 3 weeks), depending on the weight of the patient. In certain embodiments, the patient will be treated with the immune checkpoint inhibitor without knowing the PD1, CTLA-4, TIM-3, VISTA, LAG-3, IDO or KIR status of the tumor. However, as noted above, in some cases the tissue section may be stained for PD1, CTLA-4, TIM-3, VISTA, LAG-3, IDO and / or KIR and, as such, the immune checkpoint inhibitor may be selected based on those results. For example, the patient may be identified as having a tumor that contains cells which are positive for one or more of the markers, CTLA-4, PD1, PD-L1, TIM-3, VISTA, LAG-3, IDO or KIR. In these embodiments, if a tumor or the tumor-infiltrating immune cells, are PD-L1 positive, and the score is at or above a threshold, then the method may involve administering an anti-PDl or anti-PD-Ll antibody to the patient. The same principle can be applied to tumors in which cells are positive for other markers.

[0052] Alternative adjuvant therapy that may be administered to the patient may be a nontargeted therapy, i.e., a therapy that is not targeted to a particular sequence variation. Nontargeted therapies include radiation therapy, systemic or local chemotherapy, hormone therapy, and surgery. Examples of systemic chemotherapies include platinum-based doublet chemotherapy such as the combination of cisplatin and pemetrexed and the combination of cisplatin and gemcitabine. In other cases, the alternative therapy may be a therapy that is targeted to an actionable sequence variation, i.e., a therapy that targets the activity of the protein having a causative sequence variation, where the term “actionable sequence variation” is a sequence variation for which there is a therapy that specifically targets the activity of the protein having the variation. In many embodiments an actionable sequence variation causes an increase in an activity of the protein, thereby resulting in cells containing the variation to grow, divide and / or metastasize without check and in combination with other variations, such as in tumor suppressor genes, leading to cancer. Therapy that is targeted to an actionable sequence variation often inhibits an activity of the mutated protein. Examples of actionable sequence variations are known. For example, targeted therapies directed against these activating alterations in EGFR, ALK, ROS1 and BRAF have been approved for use in patients harboring these activating mutations and fusions, and thus, these are described as “actionable” mutations, although others are known.

[0053] Chemotherapeutic agents that can be used as adjuvant therapy include alkylating agents (for example nitrogen mustards (such as mechlorethamine, cyclophosphamide, melphalan, chlorambucil, ifosfamide and busulfan), nitrosoureas (such as N-Nitroso-N- methylurea (MNU), carmustine (BCNU), lomustine (CCNU) and semustine (MeCCNU), fotemustine and streptozotocin), tetrazines (such as dacarbazine, mitozolomide and temozolomide), aziridines (such as thiotepa, mytomycin and diaziquone), cisplatins and derivatives thereof (such as carboplatin and oxaliplatin), and non-classical alkylating agents (such as procarbazine and hexamethylmelamine)), antimetabolites (for example anti-folates (such as methotrexate and pemetrexed), fluoropyrimidines (such as fluorouracil and capecitabine), deoxynucleoside analogues (such as cytarabine, gemcitabine, decitabine, Vidaza, fludarabine, nelarabine, cladribine, clofarabine and pentostatin) and thiopurines (such as thioguanine and mercaptopurine)), anti-microtubule agents ( for example Vinca alkaloids (such as vincristine, vinblastine, vinorelbine, vindesine, and vinflunine) and taxanes (such as paclitaxel and docetaxel)), platins (such as cisplatin and carboplatin), topoisomerase inhibitors (for example irinotecan, topotecan, camptothecin, etoposide, doxorubicin, mitoxantrone, teniposide, novobiocin, merbarone, and aclarubicin), and cytotoxic antibiotics (for example anthracyclines (such as doxorubicin, daunorubicin apirubicin, idarubicin, pirarubicin, aclarubicin, mitoxantrone), bleomycins, mitomycin C, mitoxantrone, and actinomycin), and combinations thereof.

[0054] The adjuvant therapy administered to the patient may vary depending on the type of cancer. Exemplary adjuvant therapies include, but are not limit to: bladder cancer: atezolizumab (Tecentriq), avelumab (Bavencio), enfortumab vedotin-ejfv (Padcev), erdafitinib (Balversa), nivolumab (Opdivo), nogapendekin alfa inbakicept-pmln (Anktiva), pembrolizumab (Keytruda), brain cancer:, , belzutifan (Welireg), bevacizumab (Avastin), dabrafenib (Tafinlar), everolimus (Afinitor), tovorafenib (Ojemda), trametinib (Mekinist), vorasidenib (Voranigo), breast cancer:, , abemaciclib (Verzenio), ado-trastuzumab emtansine (Kadcyla), alpelisib (Piqray), anastrozole (Arimidex), capivasertib (Truqap), datopotamab deruxtecan-dlnk (Datroway), elacestrant dihydrochloride (Orserdu), everolimus (Afinitor), exemestane (Aromasin), fam-trastuzumab deruxtecan-nxki (Enhertu), fulvestrant (Faslodex), goserelin acetate (Zoladex), inavolisib (Itovebi), lapatinib ditosylate (Tykerb), letrozole (Femara), margetuximab-cmkb (Margenza), neratinib maleate (Nerlynx), olaparib (Lynparza), palbociclib (Ibrance), pembrolizumab (Keytruda), pertuzumab (Perjeta), pertuzumab, trastuzumab, and hyaluronidase-zzxf (Phesgo), ribociclib (Kisqali), ribociclib succinate and letrozole (Kisqali Femara Co-Pack), sacituzumab govitecan-hziy (Trodelvy), talazoparib tosylate (Talzenna), tamoxifen citrate (Soltamox), toremifene (Fareston), trastuzumab (Herceptin), tucatinib (Tukysa), cervical cancer:, , bevacizumab (Avastin), pembrolizumab (Keytruda), tisotumab vedotin-tftv (Tivdak), colorectal cancer:, , adagrasib (Krazati), bevacizumab (Avastin), cetuximab (Erbitux), encorafenib (Braftovi), fruquintinib (Fruzaqla), ipilimumab (Yervoy), nivolumab (Opdivo), panitumumab (Vectibix), pembrolizumab (Keytruda), ramucirumab (Cyramza), regorafenib (Stivarga), sotorasib (Lumakras), tucatinib (Tukysa), ziv-aflibercept (Zaltrap), dermatofibrosarcoma protuberans:, , imatinib mesylate (Gleevec), endocrine and neuroendocrine tumors:, , avelumab (Bavencio), everolimus (Afinitor), iobenguane I 131 (Azedra), lanreotide acetate (Somatuline Depot), lutetium Lu 177-dotatate (Lutathera), endometrial cancer:, , dostarlimab-gxly (Jemperli), durvalumab (Imfinzi), lenvatinib mesylate (Lenvima), pembrolizumab (Keytruda), esophageal cancer:, , fam-trastuzumab deruxtecan-nxki (Enhertu), ipilimumab (Yervoy), nivolumab (Opdivo), pembrolizumab (Keytruda), ramucirumab (Cyramza), tislelizumab-jsgr (Tevimbra), trastuzumab (Herceptin), zolbetuximab-clzb (Vyloy), head and neck cancer: cetuximab (Erbitux), nivolumab (Opdivo), pembrolizumab (Keytruda), toripalimab-tpzi (Loqtorzi), gastrointestinal stromal tumor: avapritinib (Ayvakit), imatinib mesylate (Gleevec), regorafenib (Stivarga), ripretinib (Qinlock), sunitinib malate (Sutent), giant cell tumor: denosumab (Xgeva), pexidartinib hydrochloride (Turalio), kidney cancer: avelumab (Bavencio), axitinib (Inlyta), belzutifan (Welireg), bevacizumab (Avastin), cabozantinib-s -malate (Cabometyx), everolimus (Afinitor), ipilimumab (Yervoy), lenvatinib mesylate (Lenvima), nivolumab (Opdivo), pazopanib hydrochloride (Votrient), pembrolizumab (Keytruda), sorafenib tosylate (Nexavar), sunitinib malate (Sutent), temsirolimus (Torisel), tivozanib hydrochloride (Fotivda), leukemia: acalabrutinib (Calquence), alemtuzumab (Campath), asciminib hydrochloride (Scemblix), avapritinib (Ayvakit), blinatumomab (Blincyto), bosutinib (Bosulif), brexucabtagene autoleucel (Tecartus), dasatinib (Sprycel), duvelisib (Copiktra), enasidenib mesylate (Idhifa), gemtuzumab ozogamicin (Mylotarg), gilteritinib fumarate (Xospata), glasdegib maleate (Daurismo), ibrutinib (Imbruvica), idelalisib (Zydelig), imatinib mesylate (Gleevec), inotuzumab ozogamicin (Besponsa), ivosidenib (Tibsovo), lisocabtagene maraleucel (Breyanzi), midostaurin (Rydapt), nilotinib (Tasigna), obecabtagene autoleucel (Aucatzyl), obinutuzumab (Gazyva), ofatumumab (Arzerra), olutasidenib (Rezlidhia), pemigatinib (Pemazyre), pirtobrutinib (Jaypirca), ponatinib hydrochloride (Iclusig), quizartinib dihydrochloride (Vanflyta), revumenib citrate (Revuforj), rituximab (Rituxan), rituximab and hyaluronidase human (Rituxan Hycela), tagraxofusp-erzs (Elzonris), tisagenlecleucel (Kymriah), tretinoin (Vesanoid), venetoclax (Venclexta), zanubrutinib (Brukinsa), liver and bile duct cancer: atezolizumab (Tecentriq), atezolizumab and hyaluronidase-tqjs (Tecentriq Hybreza), bevacizumab (Avastin), cabozantinib-s-malate (Cabometyx), durvalumab (Imfinzi), futibatinib (Lytgobi), ipilimumab (Yervoy), ivosidenib (Tibsovo), lenvatinib mesylate (Lenvima), nivolumab (Opdivo), pembrolizumab (Keytruda), pemigatinib (Pemazyre), ramucirumab (Cyramza), regorafenib (Stivarga), sorafenib tosylate (Nexavar), tremelimumab-actl (Imjudo), zanidatamab-hrii (Ziihera), lung cancer: adagrasib (Krazati), afatinib dimaleate (Gilotrif), alectinib (Alecensa), amivantamab-vmjw (Rybrevant), atezolizumab (Tecentriq), atezolizumab and hyaluronidase-tqjs (Tecentriq Hybreza), bevacizumab (Avastin), binimetinib (Mektovi), brigatinib (Alunbrig), capmatinib hydrochloride (Tabrecta), cemiplimab-rwlc (Libtayo), ceritinib (Zykadia), crizotinib (Xalkori), dabrafenib mesylate (Tafinlar), dacomitinib (Vizimpro), durvalumab (Imfinzi), encorafenib (Braftovi), ensartinib hydrochloride (Ensacove), entrectinib (Rozlytrek), erlotinib hydrochloride (Tarceva), famtrastuzumab deruxtecan-nxki (Enhertu), gefitinib (Iressa), ipilimumab (Yervoy), lazertinib mesylate hydrate (Lazcluze), lorlatinib (Lorbrena), necitumumab (Portrazza), nivolumab (Opdivo), osimertinib mesylate (Tagrisso), pembrolizumab (Keytruda), pralsetinib (Gavreto), ramucirumab (Cyramza), repotrectinib (Augtyro), selpercatinib (Retevmo), sotorasib (Lumakras), tarlatamab-dlle (Imdelltra), tepotinib hydrochloride (Tepmetko), trametinib dimethyl sulfoxide (Mekinist), tremelimumab-actl (Imjudo), zenocutuzumab-zbco (Bizengri), malignant mesothelioma: ipilimumab (Yervoy), nivolumab (Opdivo), pembrolizumab (Keytruda), myelodysplastic and myeloproliferative disorders: fedratinib hydrochloride (Inrebic), imatinib mesylate (Gleevec), imetelstat sodium (Rytelo), ivosidenib (Tibsovo), momelotinib dihydrochloride monohydrate (Ojjaara), pacritinib citrate (Vonjo), pemigatinib (Pemazyre), ruxolitinib phosphate (Jakafi), Targeted therapy approved for neuroblastoma, , dinutuximab (Unituxin), naxitamab-gqgk (Danyelza), ovarian epithelial, fallopian tube, and primary peritoneal cancers: bevacizumab (Avastin), mirvetuximab soravtansine-gynx (Elahere), niraparib tosylate monohydrate (Zejula), olaparib (Lynparza), rucaparib camsylate (Rubraca), pancreatic cancer: belzutifan (Welireg), erlotinib hydrochloride (Tarceva), everolimus (Afinitor), olaparib (Lynparza), sunitinib malate (Sutent), zenocutuzumab-zbco (Bizengri), plexiform neurofibroma: selumetinib sulfate (Koselugo), :prostate cancer: abiraterone acetate (Zytiga), apalutamide (Erleada), bicalutamide (Casodex), cabazitaxel (Jevtana), darolutamide (Nubeqa), degarelix (Firmagon), enzalutamide (Xtandi), flutamide, goserelin acetate (Zoladex), leuprolide acetate (Lupron Depot, Eligard), leuprolide mesylate (Camcevi), lutetium Lu 177 vipivotide tetraxetan (Pluvicto), nilutamide (Nilandron), niraparib tosylate monohydrate and abiraterone acetate (Akeega), olaparib (Lynparza), talazoparib tosylate (Talzenna), radium 223 dichloride (Xofigo), relugolix (Orgovyx), rucaparib camsylate (Rubraca), triptorelin pamoate (Trelstar), skin cancer: alitretinoin (Panretin), atezolizumab (Tecentriq), atezolizumab and hyaluronidase-tqjs (Tecentriq Hybreza), avelumab (Bavencio), binimetinib (Mektovi), cemiplimab-rwlc (Libtayo), cobimetinib fumarate (Cotellic), cosibelimab-ipdl (Unloxcyt), dabrafenib mesylate (Tafinlar), encorafenib (Braftovi), ipilimumab (Yervoy), nivolumab (Opdivo), nivolumab and relatlimab-rmbw (Opdualag), pembrolizumab (Keytruda), retifanlimab-dlwr (Zynyz), sonidegib (Odomzo), tebentafusp-tebn (Kimmtrak), trametinib dimethyl sulfoxide (Mekinist), vismodegib (Erivedge), vemurafenib (Zelboraf), soft tissue sarcoma: afamitresgene autoleucel (Tecelra), alitretinoin (Panretin), atezolizumab (Tecentriq), atezolizumab and hyaluronidase-tqjs (Tecentriq Hybreza), crizotinib (Xalkori), nirogacestat hydrobromide (Ogsiveo), pazopanib hydrochloride (Votrient), sirolimus protein-bound particles (Fyarro), tazemetostat hydrobromide (Tazverik), solid tumors generally: dabrafenib mesylate (Tafinlar), dostarlimab-gxly (Jemperli), entrectinib (Rozlytrek), fam-trastuzumab deruxtecan-nxki (Enhertu), larotrectinib sulfate (Vitrakvi), pembrolizumab (Keytruda), repotrectinib (Augtyro), selpercatinib (Retevmo), trametinib dimethyl sulfoxide (Mekinist), stomach (gastric) cancer: fam-trastuzumab deruxtecan-nxki (Enhertu), nivolumab (Opdivo), pembrolizumab (Keytruda), ramucirumab (Cyramza), tislelizumab-jsgr (Tevimbra), trastuzumab (Herceptin), zolbetuximab-clzb (Vyloy), thyroid cancer: cabozantinib-s-malate (Cometriq), dabrafenib mesylate (Tafinlar), lenvatinib mesylate (Lenvima), pralsetinib (Gavreto), selpercatinib (Retevmo), sorafenib tosylate (Nexavar), trametinib dimethyl sulfoxide (Mekinist), vandetanib (Caprelsa), In some embodiments, the tissue section may be analyzed at a remote location, potentially by a third party, and the treatment decision may be made upon receipt of a report produced at the remote location and forwarded to a medical professional.

[0055] In these embodiments, the method may comprise (a) receiving a report that provides a score indicating the strength and / or number of interactions between cancer stem cells and PDL1+macrophages in a tumor from a patient; and (b) identifying the patient as a candidate for adjuvant therapy if the score is at or below a threshold.

[0056] In some embodiments, the method may be for selecting a patient for treatment by an adjuvant therapy. In these embodiments, the method may comprise selecting a cancer patient for treatment by an immune checkpoint inhibitor based on the interactions between cancer stem cells and PDL1+macrophages in a tumor from the patient, particularly around the tumor edge. In these embodiments, the report may be in an electronic form, and the method comprises forwarding the report to a remote location, e.g., to a doctor or other medical professional to help identify a suitable course of action, e.g., to identify a suitable therapy for the subject. The report may be used along with other metrics to determine whether the subject has a high probability of recurrence and therefore should be recommended for adjuvant therapy. In some cases, the report may indicate a score as well as a threshold at or below which a patient should be recommend for adjuvant treatment. For example, the report may indicate a score (e.g., 0.2, 1.0 or 1.5) as well as the threshold (e.g., 0.5) at or below which the patient should be recommended for adjuvant treatment. The doctor or other medical professional can review the report and make a treatment decision after reviewing the report.

[0057] In any embodiment, a report can be forwarded to a “remote location”, where “remote location,” means a location other than the location at which the sequences are analyzed. For example, a remote location could be another location (e.g., office, lab, etc.) in the same city, another location in a different city, another location in a different state, another location in a different country, etc. As such, when one item is indicated as being "remote" from another, what is meant is that the two items can be in the same room but separated, or at least in different rooms or different buildings, and can be at least one mile, ten miles, or at least one hundred miles apart. "Communicating" information references transmitting the data representing that information as electrical signals over a suitable communication channel (e.g., a private or public network). "Forwarding" an item refers to any means of getting that item from one location to the next, whether by physically transporting that item or otherwise (where that is possible) and includes, at least in the case of data, physically transporting a medium carrying the data or communicating the data. Examples of communicating media include radio or infra-red transmission channels as well as a network connection to another computer or networked device, and the internet, including email transmissions and information recorded on websites and the like. In certain embodiments, the report may be analyzed by an MD or other qualified medical professional, and a report based on the results of the analysis of the sequences may be forwarded to the patient from which the sample was obtained.

[0058] In some embodiments, a sample may be collected from a patient at a first location, e.g., in a clinical setting such as in a hospital or at a doctor’s office, and the sample may be forwarded to a second location, e.g., a laboratory where it is processed, and the abovedescribed method is performed to generate a report. A “report” as described herein, is an electronic or tangible document which includes report elements that provide test results, including the score and optionally the threshold. Once generated, the report may be forwarded to another location (which may be the same location as the first location), where it may be interpreted by a health professional (e.g., a clinician, a laboratory technician, or a physician such as an oncologist, surgeon, pathologist or virologist), as part of a clinical decision.

[0059] The results provided by this method may be diagnostic, prognostic, theranostic and, in some cases, may be used to monitor a treatment. In the latter embodiments, ratio may be analyzed at multiple time points in the same patient. In some embodiments, a decrease in the score may indicate that a treatment is working and should therefore be continued. In some embodiments, an increase in the ratio, may indicate that a treatment is not working and should therefore be modified or stopped.

[0060] As would be readily appreciated, many steps of the method, e.g., image analysis, segmentation, cell identification, centroid identification and distance measurements can be implemented on a computer. As would be apparent, the computational steps described may be computer-implemented and, as such, instructions for performing the steps may be set forth as programing that may be recorded in a suitable physical computer readable storage medium.

[0061] The following illustrates some of the general principles of the method.

[0062] This method for calculating interactions between cells involves a sophisticated process of spatial analysis within tumor tissue, focusing on the proximity and types of cells present.

[0063] Sample Preparation and Initial Imaging:

[0064] A section of a tumor, often a formalin- fixed paraffin-embedded (FFPE) tissue section, is obtained from the cancer patient.

[0065] A multiplexed IF assay is performed on the tissue section to identify cancer stem cells and PDL1+ macrophages.

[0066] Cell Segmentation and Identification:

[0067] Single cells are identified, based on their nuclei, which are stained (e.g., by DAPI).

[0068] Image processing tools are employed to segment the cells, which means defining their boundaries or edges and separating touching cells using deep learning techniques.

[0069] After segmentation, the binding patterns of the binding agents to the cells are quantified to discriminate and identify different cell types.

[0070] For instance, cancer stem cells are identified as positive for markers such as PanCK, CK19, and EpCAM, and CD44. PDL1+ macrophages are identified as positive for CD68, CD206, and PDL1.

[0071] Spatial Coordinate Derivation:

[0072] The x and y coordinates of each identified cell are derived by taking the centroid of each segmented cell nucleus. Interaction Measurement Based on Proximity.

[0073] Interactions between cells are fundamentally measured by their spatial proximity.

[0074] A fixed radius neighbor algorithm is applied to these spatial coordinates to construct a spatial nearest-neighbor graph which represents, for each cell, its closest neighbors in 2D space within a specified fixed radius.

[0075] Direct interactions are defined as occurring between cells within a 25-micron radius of a central cell. This radius may vary and, in any embodiment, may be in the range of 10- 100 microns, e.g., a radius in the range of 20-50 microns.

[0076] The analysis focuses on interactions between cancer stem cells and PDL1+ macrophages.

[0077] Frequency and Proportion Calculation:

[0078] The frequency of direct interactions between specific cell types, stem-like cancer cells and PDL1+ M2-like macrophages, is calculated for each sample.

[0079] To normalize for varying cell densities, the proportion of each cell type engaged in interactions with other distinct cell types can be determined by dividing the number of interacting cells of a specific type by the total number of that central cell type in the sample.

[0080] Area of Analysis:

[0081] While interactions occur throughout the tumor tissue, they are noted to occur more frequently and are often analyzed at the tumor edge. This "tumor edge" refers to the outermost area of the tumor where it meets normal tissue, typically defined as up to 200 microns from this interface.

[0082] Interpretation for Risk Assessment:

[0083] A higher level or frequency of interactions between cancer stem cells and PDL1+ macrophages in the tumor indicates that the patient is at high risk for cancer recurrence. This spatial interaction signature is considered a critical driver of recurrence, more so than just the individual abundance of these cell types. Tumors are classified into high and low-risk categories based on the frequency of these interactions.

[0084] Reporting and Clinical Action: The analysis will result in a report that provides a score indicating the strength and / or number of these interactions. This individual score would then be compared to a pre-established threshold, which could be the median frequency of interactions observed in a well-defined reference cohort. If the patient's score is at or above this established median threshold, they are identified as high risk; otherwise, they are low risk. If the score is reported as high, the patient may be recommended for an adjuvant therapy, such as immune checkpoint therapy or chemotherapy.

[0085] Kits

[0086] Also provided by this disclosure are kits for practicing the present method, as described above. In some embodiments, the kit may comprise a panel of antibodies that label cancer stem cells and a panel of antibodies that label PDL1+macrophages, where the term “label” is intended to mean label them in a way that allows them to be identified relative to other cells, e.g., malignant cells, normal cells and other immune cells.

[0087] These antibodies may be in a cocktail or in different vessels. In any embodiment, the total number of antibodies in the kit may be less than 30, less than 20 or less than 15, for example. In some embodiments, the kit may comprise (a) antibodies to (i) PanCK, CK19 and EpCAM and optionally CD44 or (ii) PanCK and at least one of CK19, EpCAM and CD44 (which identify cancer stem cells) and antibodies to CD68, CD206, and PDL1 (which identify PDL1+macrophages). PDL1+macrophages can also be identified because they are CD68+ / PDLl+ / PanCK-ve.

[0088] The various antibodies in the kit may be distinguishably labeled, e.g., via distinguishable fluorophores or mass tags, or they may be linked to oligonucleotides that have different sequences (which should allow them to be used in CODEX-type assays), where the term “distinguishably labeled” means that the antibodies are labeled in a way that they can be distinguished from one another, even, when they are in the same location.

[0089] The various components of the kit may be present in separate containers or certain compatible components may be precombined into a single container, as desired. The kit may also comprise buffers, labels and instructions for performing the present method.

[0090] In addition to the above-mentioned components, the subject kit may further include instructions for using the components of the kit to practice the subject method.

[0091] EXAMPLES

[0092] In order to further illustrate some embodiments of the present invention, the following specific examples are given with the understanding that they are being offered to illustrate examples of the present invention and should not be construed in any way as limiting its scope.

[0093] Hepatocellular carcinoma (HCC) frequently recurs even after initial complete response to therapy, due to the persistence of minimal residual disease (MRD). Here, the mechanisms of persistence of residual tumor cells were identified through single-cell spatial transcriptomic and proteomic analyses of both post-chemoembolization residual human HCC and transgenic mouse models of MRD. The spatial organization of residual human HCC were defined into neighborhoods within which EpCAM+ / CK19+ stem-like tumor cells interact with PDL1+ M2-like macrophages. This was associated with PD1+ / TIM3+ CD8T cell exhaustion and poor recurrence-free survival. Then, through spatially resolved transcriptomic analysis, macrophage-mediated TGFpi pathway activation was identified as a mediator of persistence of residual tumor cells. Through a mouse model of MRD, it was demonstrated that Tgfprl+ residual stem-like tumor cells survive in Tgfpi+ / Pdll+ macrophage-rich niches which enforce CD8T cell exhaustion and enable tumor recurrence. Finally, in two mouse models: a transgenic mouse model of MRD and a syngeneic orthotopic allograft model of doxorubicin-resistant HCC, it was found that the combined blockade of Tgfp and Pdll pathways excluded immunosuppressive macrophages, thereby recruiting activated CD8T cells that eliminate residual tumor cells. Thus, through spatial mapping, the mechanistic roles of the TGFP and PDL1 pathways in persistence of residual tumor cells in human HCC were identified. Targeting these pathways holds therapeutic potential to eliminate MRD, which could reduce recurrence and improve patient outcomes.

[0094] In performing this research, a spatial signature predicting HCC recurrence was determined by performing spatial single-cell 41- plex CODEX analysis 56,60 of resected human HCC (n=53; 638,158 cells). It was shown that spatial interactions between stem- like cancer cells (PanCK+ / CK19+ / EpCAM-i- / CD44Hi) and PDL1+ pro-tumor macrophages (PDLl+ / CD206Hi / HLA-DR-) were associated with recurrence (n=53, p=0.01), but not their individual cellular abundance. To investigate the mechanisms, the analysis was expanded to conduct multi -regional sampling of aggressive HCC that recurred, to determine spatial organization at the invasive edge ( n=8, 246,107 cells). Increased interactions between stemlike cancer cells, PDL1+ macrophages and exhausted PD1+ CD8T cells (PD1+ / TIM3+ / CD44-) were observed along the invasive edge rather than the tumor core. These findings show that a spatial signature of interactions of stem-like cancer cells with PDL1+ macrophages along the invasive tumor edge predict HCC recurrence.

[0095] METHODS

[0096] Patient Cohort Selection. In this study, two cohorts of patients diagnosed with HCC who identified who met the following inclusion criteria. For cohort 1 (residual HCC)- patients with a confirmed diagnosis of HCC who received bridging therapy with transarterial chemoembolization (TACE), subsequently underwent liver transplantation, and had viable residual HCC in the explanted liver tissue were selected. Additional criteria included availability of sufficient tissue in formalin-fixed paraffin-embedded (FFPE) blocks. Nodules which had been targeted for TACE were carefully selected to ensure adequate samples for downstream analysis. Exclusion criteria included a history of other malignancies, receipt of resection prior to liver transplantation, receipt of radiation therapy prior to transplant or evidence of metastatic cancer. All patients in this cohort had received doxorubicin-based TACE- a majority of which was DEB-TACE with doxorubicin eluting beads (n=102, 89%) and a smaller proportion received conventional TACE (n=14, 11%), none had received bland TACE. A consistent team of experienced interventional radiologists at a single institution performed all TACE procedures. Patients undergoing TACE underwent follow-up CT or MR1 scans 8-12 weeks post-procedure to assess therapeutic response, which was evaluated using the modified Response Evaluation Criteria in Solid Tumors (mRECIST) system63. Decisions regarding additional TACE treatments were made during multidisciplinary tumor board meetings. For the control group (primary HCC), patients who underwent surgical resection for HCC and had not received any locoregional therapy or systemic therapy prior to resection were selected. Additionally, similar to cohort 1, these patients did not have any other form of cancer or metastatic tumors. In both groups, comprehensive clinical and pathological data were collected, including patient demographics, tumor characteristics, details of the treatments received, and outcomes following the treatment.

[0097] CODEX multiplex staining and analysis- A 41-plex custom CODEX antibody panel was developed and validated (Enable Medicine, Menlo Park, CA, USA) for ultra- highplex imaging utilizing purified, carrier-free antibodies conjugated to unique DNA oligonucleotide barcodes (Akoya Biosciences, Menlo Park, CA, USA). Image processing and analysis were performed as described before32.

[0098] Nanostring Digital Spatial Profiling- Experimental methods for the Nanostring GeoMx analysis were used19. The GeoMx Digital Spatial Profiling instrument from NanoString Technologies, Inc. (Seattle, WA) was used for immunofluorescence imaging. Differential gene expression and gene set enrichment analysis were performed in the Omics Explorer software from QluCore (Lund, Sweden). The Benjamini-Hochberg correction was used to decrease the false discovery rate. Principal component analysis was conducted. Upstream regulators of transcription were discovered using the Ingenuity Pathway Analysis software (Aarhus, Denmark). Statistical analysis- Differences between groups were analyzed using Student’s t- test or one-way analysis of variance (ANOVA). The Benjamini-Hochberg method was used for adjusting p values. Chi-square test was used to compare categorical variables. Kaplan Meier analysis with the Log Rank test was performed for survival analysis. All graphs are presented as the mean + / - SEM. An adjusted P value of less than 0.05 was considered to be significant.

[0099] Construction of Tissue Microarrays (TMA) Tumor tissue was obtained in the intra-op period for each patient, and was placed in formalin for 24 to 48 hours prior to placement in 70% ethanol for storage prior to processing. Patients were followed until death or until August 2023. Clinical variables were extracted from the electronic medical record of each patient in this study. These variables included demographic data, etiology of liver disease, comorbidities, Child-Pugh score, initial HCC staging, and number / type of locoregional therapies received, in addition to pathologic variables such as total tumor number, maximum tumor diameter, grade, micro- and macro-vascular invasion, and American Joint Committee on Cancer (AJCC) tumor staging. Clinical outcomes assessed in this study included recurrence-free survival (RFS) and overall survival (OS).

[0100] All tumor tissues were processed uniformly in the Stanford clinical pathology lab. Hematoxylin and eosin stained sections from each FFPE block were carefully reviewed by the pathologist and areas of viable tumor or non tumorous liver were selected. A total of 5 TMAs with 1.5 mm diameter cores were assembled using a TMA Grand Master automated tissue microarray er. CODEX multiplex immunostaining was performed on a HCC tumors, matched adjacent cirrhotic liver.

[0101] Statistical analysis of clinical variables Statistical Package for the Social Sciences (SPSS, IBM) was used to compare patient risk factors, demographics, and clinical outcomes. Kaplan Meier analysis was used for survival analyses, with recurrence-free survival being defined as the duration between the date of surgery and the date of recurrence or death from any cause and overall survival being defined as the duration between the date of surgery and the date of death from any cause. Multivariable Cox regression analyses were performed to investigate how the tumor microenvironment and patient characteristics affected RFS and OS. Statistically significant variables were determined to have p- values < 0.05.

[0102] CODEX panel development and staining CODEX antibodies were validated on FFPE tonsil sections, and staining patterns were confirmed via comparison with online databases (The Human Protein Atlas, Pathology Outlines) and the published literature. Tissue microarrays containing FFPE biopsies from the cohort described in this study were sectioned at 5 pm, and placed on 15 x 15 mm glass coverslips (Electron Microscopy Sciences, # 72204-01) pre-coated with poly-L-lysine (Sigma, # P8920). Coverslip staining was performed by Enable Medicine.

[0103] Briefly, FFPE tissue sections on coverslips were pretreated by heating on a slide warmer for 25 minutes at 55 degrees C. Tissue deparaffinization and hydration were next performed by incubating the FFPE tissue sections on coverslips for 5 minutes each following a solvent series (Histochoice Clearing Agent, Histochoice Clearing Agent, 100% Ethanol, 100% Ethanol, 90% Ethanol, 70% Ethanol, 50% Ethanol, 30% Ethanol, ddH20, ddH20). Antigen retrieval was performed in 0.01M Citrate Buffer at high pressure. The tissue was washed and equilibrated before staining for 3 hours at room temperature with the 41-plex CODEX antibody cocktail in a staining buffer containing blocking solution (Akoya Biosciences). Post-staining, the tissues were washed and fixed in 1.6% PFA, followed by an ice cold methanol incubation. After washing, the final tissue fix was performed using Fixative reagent (Akoya Biosciences). FFPE tissues on coverslips were stored in a 6-well plate containing the storage buffer at 4 degrees C until CODEX acquisition.

[0104] CODEX multiplexed imaging and processing Stained coverslips were mounted onto the CODEX stage plate version 2 (Akoya) and secured onto the stage of a BZ-X810 inverted fluorescence microscope (Keyence). Reporter plates were prepared by adding fluorescently labeled oligonucleotides (Atto550, Cy5, AF750) made up in a reporter stock solution of nuclease free water, lOx CODEX buffer, assay reagent and nuclear stain to a black Coming 96 well plate. Automated image acquisition of tissue regions was performed at Enable Medicine using a CFI Plan Apo X 20x / 0.75 objective (Nikon) and fluidics exchange managed via the CODEX instrument and CODEX Instrument Manager software (CIM version 1.29.3.6, Akoya Biosciences), according to the manufacturer’s instructions, with slight modifications. Raw fluorescent TIFF image files were processed, deconvolved and background subtracted utilizing the Enable Processor Pipeline (Enable Medicine), and antibody staining was visually assessed for each biomarker and tissue region using the Enable Visualizer. OME-TIFF hyperstacks were segmented based on DAPI stain, pixel intensities were quantified, and spatial fluorescence compensation was performed, which generated comma-separated value (CSV) and flow cytometry standard (FCS) files for downstream analysis.

[0105] CODEX Data analysis All analyses were run in R-4.0.5 unless otherwise indicated. R functions are specified using the following notation : “<package_name> : : <function_name>” . Cell clustering- For cell clustering and cell neighborhood analysis, data from 108 cores were analyzed. Possible batch effects were addressed by performing an inverse hyperbolic sine transform (“base::asinh”) on cell expression values for every marker, in every ROI. Next, normalized values were z-scaled across both cells and markers. To cluster cells, dimensionality reduction was first performed on scaled expression values using principal component analysis with 20 components (“stats: :prcomp”). Next, a k-nearest neighbor graph was constructed to build a similarity network between cells in principal component space (“dbscan::kNN”, k = 30). Finally, cells were clustered using the Leiden graph clustering algorithm (“igraph::cluster_leiden”, cluster_resolution = 1.0).

[0106] Cell populations were defined using iterative unsupervised clustering using subsets of the full markers-by-cells expression matrix. An initial set of coarse cell clusters was first defined by unsupervised clustering on all cells that passed QC and the major cell lineage markers in the panel: CD20 (B cell), CD15 (Neutrophil), CD68 (Macrophage), FoxP3 (Treg), CD31 (Endothelial cell), CD56 (NK cell), CD8 / CD4 / CD45 / CD3e (T cell), CD117 (Mast cell), CDllc (Dendritic cell), PanCK (Tumor), and aSMA (Fibroblast). Clustering parameters, including granularity, nearest-neighbor number, and marker subsets, were optimized by assessing clustering results visually, overlaid on images (Enable Medicine Visualizer); by manually examining the distribution of expression values in each cluster; and by quantifying cluster purity using the silhouette score. Next, cell subtypes were defined by sub clustering of major cell categories. The full markers-by-cells expression matrix was subsetted with the following criteria: Tumor cells - PanCK, EpCAM, CK19, PDL1. Macrophages - HLA-DR, PDL1, CD206. T cells - CD4, PD1, TIM3, CD44, CD45RO. Other canonical markers of these cell types were evaluated to further confirm the cell definitions. In each case, sub clustered cells were examined visually for proper expression of lineage and subtype markers.

[0107] Construction of a spatial cellular interaction graph- To perform spatial analyses on the data, a spatial nearest neighbor graph was constructed. Cell coordinates were derived by taking the centroid of each segmented cell nucleus relative to the corner of the ROI. A fixed radius neighbor algorithm was next used on these coordinates (“dbscan::fixedrad”, r = 25 or r=100). This graph thus represents, for each cell, its closest neighbors in 2D space within the specified fixed radius.

[0108] Cell neighborhood analysis- To define cellular neighborhoods (CNs), the number of neighbors of each cell type was counted, resulting in a matrix of cells by cell clusters, with each row representing a cell, each column representing a cell annotation (cell type) from the clustering above, and each value representing the count of neighbors of the given annotation. The neighbor cell proportion was computed for each row. The resulting matrix was clustered using k-means clustering (“stats: :kmeans”), where the optimal k was determined empirically by maximizing the silhouette score metric (“cluster: :silhouette”). Each cluster was defined as a CN. Thus, each cell was given both a cell type annotation, which depends only on the cell’s own marker expression, and a cell neighborhood annotation, which depends on the cell’s type and the identities of its nearest neighbors. To compare CNs between patient cohorts, the proportion of cells in each ROI belonging to each CN was determined. Proportions were transformed using the inverse hyperbolic tangent (“base::asinh”) and split by cohort. Pairwise t-tests (“stat: :t. test”) were then performed on the transformed proportions, comparing each CN between PR and PD cohorts. The resulting p-values were corrected for multiple testing by the Bonferroni method (“stat: :p. adjust”, method = “Bonferroni”).

[0109] Nanostring Digital Spatial Profiling 5 pm sections of the FFPE human HCC TMAs were used for H&E and cancer transcriptome atlas (CTA) processing. CTA involved baking slides at 60°C, deparaffinization, antigen retrieval, proteinase K digestion, hybridization to RNA probes, washing to remove off-target probes, and counterstaining with morphology markers. The morphology markers used were anti-panCK-Alexa Fluor 532, anti-CD45- Alexa Fluor 594, and anti-CD68-Alexa Fluor 647. The GeoMx Digital Spatial Profiling instrument from NanoString Technologies, Inc (Seattle, WA) was used for immunofluorescence imaging, ROI selection, AOI segmentation, and spatially-indexed barcode cleavage and collection. The tissue microarrays underwent staining with 4 markers, including panCK, CD45, CD68, and DAPI. These slides were imaged on the GeoMX platform, which functions in part as a fluorescent slide scanner. Regions of interest (ROIs; n = 12 / slide) were selected on the basis of the visualization markers, using a custom-designed web-based control program (NanoString Inc.). Both H&E (Hematoxylin and Eosin) slides and multiplex immunofluorescence (IF) stained slides were visually examined in collaboration with a pathologist. Regions of Interest (ROI) were selected based on criteria such as sufficient cellularity and the absence of artifacts. Within each ROI, two specific areas of interest (AOI) were identified: one enriched in epithelial cells (PanCK+ / CD45- / CD68-) and the other in macrophages (PanCK- / CD45+ / CD68-). After AOIs were chosen, the GeoMX platform utilized an automatically controlled UV laser to illuminate each AOI in turn, specifically cleaving oligonucleotide tags within the AOI but not in the surrounding tissue. A microcapillary collection system then collected the liberated oligonucleotides from each region and plated them into an individual well on a microtiter plate. This process was repeated in turn for each AOI. After AOI collection was complete, oligonucleotides were hybridized to complementary NanoString counting beads and counted using an nCounter analysis platform (NanoString Inc.).

[0110] Library preparation involved PCR amplification and sequencing on a NovaSeq S2 to achieve a minimum sequencing depth of 150-200 reads per square micron of illumination area. Digital counts from barcodes corresponding to gene probes were normalized using internal spike-in controls to account for system variation. Subsequently, these counts were further normalized to the area of their respective compartments. Differential gene expression and gene set enrichment analysis were performed in the Nanostring GeoMx web portal. The Benjamini-Hochberg correction was used to decrease the false discovery rate. Principal component analysis was conducted in the Omics Explorer software from QluCore (Lund, Sweden). Upstream regulators of transcription were discovered using the Ingenuity Pathway Analysis software from QIAGEN Digital Insights (Aarhus, Denmark).

[0111] Multiplex Fluorescent in situ hybridization (FISH) The RNAscope (Advanced Cell Diagnostics, Inc, Newark, CA) fluorescent in-situ hybridization technology designed primarily for use on formalin fixed paraffin sections was used to detect TGFB 1 and CD68 mRNA in human HCC tissue microarrays. The procedure was performed as described previously1. Post-fixed TMA sections were subjected to staining using the RNAscope assay protocol, employing the TGFB 1 and CD68 primary target probe. Simultaneously, positive and negative in-house control probes were incorporated within each run. The stained sections were scanned on Fluorescence Microscope (BZ-X800) from Keyence Corporation of America (Itasca, IL) and quantified using QuPath analysis software2.

[0112] References

[0113] Cosi, C., Millar, M., Beltran, M., Sherry, L. & Gatti-McArthur, S. Quantitative analysis of RNAscope staining for expression in mouse brain tissue as a measure of Neuronal Activation. MethodsX 8, 101348 (2021).

[0114] Bankhead, P. et al. QuPath: Open source software for digital pathology image analysis. Sci. Rep. 7, 16878 (2017).

[0115] Genin, M., Clement, F., Fattaccioli, A., Raes, M. & Michiels, C. Ml and M2 macrophages derived from THP-1 cells differentially modulate the response of cancer cells to etoposide. BMC Cancer 15, 577 (2015). Gambari, R., Hau, D. K.-P., Wong, W.-Y. & Chui, C.-H. Sensitization of Hep3B hepatoma cells to cisplatin and doxorubicin by corilagin. Phylolher. Res. 28, 781-783 (2014).

[0116] Hu, Q.-D. et al. NSC 74859 enhances doxorubicin cytotoxicity via inhibition of epithelial-mesenchymal transition in hepatocellular carcinoma cells. Cancer Lett. 325, 207- 213 (2012).

[0117] Buschauer, S. Hepatocellular Carcinoma Cells Surviving Doxorubicin Treatment Exhibit Increased Migratory Potential and Resistance to Doxorubicin Re-Treatment in Vitro. (2017).

[0118] Namur, J. et al. Embolization of hepatocellular carcinoma with drug-eluting beads: doxorubicin tissue concentration and distribution in patient liver explants. J. Hepatol. 55, 1332-1338 (2011).

[0119] Ulbrich, J., Lopez-Salmeron, V. & Gerrard, I. BD Rhapsody1MSingle-Cell Analysis System Workflow: From Sample to Multimodal Single-Cell Sequencing Data. Methods Mol. Biol. 2584, 29-56 (2023).

[0120] Hao, Y. et al. Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat. Biotechnol. (2023) doi:10.1038 / s41587-023-01767-y.

[0121] INielsen, M. C., Andersen, M. N. & Mpller, H. J. Monocyte isolation techniques significantly impact the phenotype of both isolated monocytes and derived macrophages in vitro. Immunology 159, 63-74 (2020).

[0122] Wu, P. V. & Nusse, R. 3D Culture of Primary Patient-Derived Hepatoblastoma Tumoroids. Methods Mol. Biol. 2544, 259-267 (2022).

[0123] Lai, I. et al. Lipid nanoparticles that deliver IL-12 messenger RNA suppress tumorigenesis in MYC oncogene-driven hepatocellular carcinoma. J Immunother Cancer 6, 125 (2018).

[0124] Dhanasekaran, R. et al. MYC ASO Impedes Tumorigenesis and Elicits Oncogene Addiction in Autochthonous Transgenic Mouse Models of HCC and RCC. Mol. Ther. Nucleic Acids 21, 850-859 (2020).

[0125] Palmer, G. M. et al. In vivo optical molecular imaging and analysis in mice using dorsal window chamber models applied to hypoxia, vasculature and fluorescent reporters. Nat. Protoc. 6, 1355-1366 (2011).

[0126] RESULTS High dimensional single-cell spatial atlas reveals tumor and immune heterogeneity in human HCC

[0127] To elucidate the spatial organization of tumor and immune cells in post chemoembolization residual HCC, a comprehensive single-cell spatial map of human HCC was constructed. Areas of viable HCC were microdissected, formalin-fixed paraffin- embedded (FFPE) tissue microarrays (TMA) were constructed, and co-detection by indexing (CODEX) was used to analyze the spatial architecture. A total of 1.07 million cells from 108 HCC samples were imaged with multiplex staining using 41 antibodies targeting tumor, immune, stromal, and functional markers (FIG. 1A). Through cellular segmentation, marker quantification, normalization, and unsupervised clustering, 11 major cell types within HCC were identified (FIG. IB). Thus, utilizing FFPE-optimized CODEX allowed for highly multiplexed single-cell marker visualization and cellular phenotyping in human HCC.

[0128] Canonical marker expression was used in CODEX to further subset three major cell types: tumor cells, macrophages and T cells (FIGS. 1C-1D). First, among the tumor cells three discrete subsets were identified. Two were stem cell-like, CK19+ (6.7%, 34436 cells) and EpCAM-i- (5.2%, 26622 cells) tumor cells (FIGS. 1C-1D). They expressed sternness, pro-survival and mesenchymal markers. A third subset of tumor cells expressed the immune checkpoint PDL1 (3.8%, 19737 cells). The heterogeneity observed among these tumor cell subsets aligned with prior reports from single marker studies22 21. Second, among the host immune cells, CD68+ tumor-associated macrophages (TAMs) were identified as the most common immune cell subset (14.5%, 154,899 cells). Three subsets of TAMs were identified: CD206+ M2-like (CD206+ / CD163+ / HLA-DR- / PDL1-), PDL1+ M2-like (PDLl+ / CDl lb+ / CD206- / HLA-DR-), and HLA-DR+ Ml-like (HLA- DR+ / S 100A4+ / CD206- / PDL1-) (FIGS. 1C-1D). The expression of other macrophage markers aligned with this classification. Third, three subsets among the tumor-infiltrating CD8T cells (17,647 cells, 0.01%) were defined: exhausted (PD1+ / TIM3+ / CD44-), effector (CD45RO+ / CD44+ / PD1- / TIM3-), and memory (CD45RO+ / CD44- / PD1-) CD8T cells (FIGS. 1C-1D). Thus, through CODEX analysis, discrete subsets of tumor cells, macrophages and CD8T cells were classified (FIG. ID).

[0129] To validate the tumor and immune cell classification, three approaches were used. First, the unsupervised cell-calling was overlaid on immunostained images and the accuracy of classification was confirmed based on marker expression (data not shown). Second, a consistent proportion of immune and stromal cell types were demonstrated across various HCC clinical subgroups, indicating the reliability of the classification, with NASH-HCC notably displaying increased CD8T cell exhaustion as previously reported25. Third, it was noted that the observed subset identification and proportion of macrophage and T-cell subsets are similar to prior reports26-28, However, a higher proportion of neutrophils was noted, possibly attributable to their underrepresentation in transcriptome-based deconvolution analyses2930. Thus, the CODEX analysis provides an accurate and quantitative single-cell spatial profile of 20 major cell types in human HCC.

[0130] Residual human HCC has enrichment of Immunosuppressive cells

[0131] Next, the above single-cell phenotyping was used to identify differences in tumor and immune cell subsets that would distinguish primary and residual HCC. To study residual HCC, patients with HCC who had undergone transarterial chemoembolization (mean of 2.42 cycles; SEM=1.32) and retained post-treatment viable HCC in the liver explant (n=55; 427,126 cells) were identified. For a comparative analysis with residual HCC, a control group of patients with treatment-naive primary HCC (n=53; 638,158 cells) was included. The two groups were matched for key variables including age, sex, race, ethnicity, comorbidities, grade, microvascular invasion, AFP, BCLC stage and AJCC stage, but not matched for etiology and cirrhosis. All 20 cell types were measured in both residual and primary HCC. Thus, primary and residual HCC were able to be distinguished through their distinct distribution of cellular subtypes (FIGS. 2A-2B).

[0132] Higher infiltration of immunosuppressive tumor and immune cell subsets was found in residual HCC compared to primary HCC. Specifically, PDL1+ macrophages (pAdj=2.2xlO-6), PDL1+ tumor cells (pAdj=3.4xlO-5), mast cells (pAdj=3.1xlO-4), exhausted CD8T cells (pAdj=6.6xlO-4) and NK cells (pAdj=1.8xlO-3) were more abundant in residual HCC (FIGS. 2C-2E). Conversely, neutrophils (pAdj=1.4xlO-4) and fibroblasts (pAdj=1.5xl0‘2) were relatively depleted in residual HCC compared to primary HCC (FIG. 2C). Despite differences in the frequency of cirrhosis between primary and residual HCC, these observations remained consistent within the subgroup of HCCs arising in a cirrhotic background (data not shown). Moreover, it was shown that M2-like macrophage enrichment was specific to the presence of therapy-resistant residual tumors post-TACE, rather than treatment effects, as evidenced by higher macrophage levels in T ACE-refractory HCC31compared to TACE-exposed peri-tumoral cirrhotic liver tissue or sites of post-TACE complete response, with results being consistent across different TACE methods. Hence, residual HCC showed a more pro-tumor redistribution of specific cell types than primary HCC. In addition to changes in cellular distribution, four differences in cellular phenotype were observed between residual and primary HCC. First, tumor cells in residual HCC showed a more stem-like (CD44hlgh, CK19hlgh) phenotype (FIG. 2F). Second, the abundant PDL1+ macrophages within the residual HCC were more likely to be monocyte-derived (CDl lbhlgh) and migratory (podoplaninhlgh) (FIG. 2G). Third, the CD8T cells infiltrating residual HCCs were more exhausted, with higher PD1 and TIM3 expression (FIG. 2H). Fourth, the NK cells present in residual HCCs were also dysfunctional, with higher levels of PD1 and TIM3 expression (FIG. 2H). Thus, residual HCC appears to be characterized by cancer sternness, enrichment of pro-tumor macrophages, and decreased immune surveillance associated with exhausted CD8T cells.

[0133] Spatial Interactions of Tumor and Immune Cells in Residual HCC Drive Immune evasion

[0134] It was investigated whether the differences observed between residual and tumor HCC in both the tumor and immune cellular subtypes were related to direct interactions between tumor cells and host immune cells. To test this, single-cell phenotype data was integrated with spatial coordinates to analyze interactions between non-homotypic cells in direct contact with each other, which was defined as within a 25pm radius. Based on the results, interactions between tumor cells, macrophages, and CD8T cells were examined.

[0135] The interactions of tumor cells in residual versus primary HCC were compared (FIG. 3A). The most significant change was seen in tumor cell interactions with PDL1+ macrophages (FIG. 3B). In residual HCC, all four types of tumor cells that were identified had more frequent direct interactions with PDL1+ macrophages compared to primary HCC (FIG. 3B). This was true, whether tumors had high or low levels of PDL1+ macrophages. In contrast, tumor cells did not differentially interact with exhausted CD8T cells, despite the increased presence of the latter in residual HCC (FIG. 3C). Among the interactions of tumor cells, those between EpCAM-i- tumor cells and either M2-like PDL1+ or CD206+ macrophages were associated with poor recurrence-free survival in residual HCC (FIG. 3D). This finding was validated in the TCGA cohort by showing that overexpression of genes representing the interaction between stem-like cancer cells (EPCAM, KRT19, CD44) and M2-like macrophages (CD68, CD274, MRC1, CD163) was indeed associated with a significantly poorer recurrence-free survival in HCC (p=5.2xl0-5). Further, by employing 3D co-culture tumoroid in vitro experiments, it was shown that co-culture of HCC cancer cells with M2-like polarized macrophages induced more cancer sternness than co-culture with Ml -like macrophages. Thus, in residual HCC, stem-like tumor cells appear to spatially interact with M2-like PDL1+ macrophages, and this interaction is associated with a worse clinical outcome.

[0136] Next, the interactions of macrophages were investigated. M2-like PDL1+ macrophages were found to more frequently directly interact with exhausted CD8T cells (p=l.l lxl0-3) (FIG. 3E) but not effector (p=0.17) or memory CD8T cells (p=0.38) in residual HCC. These interactions of PDL1+ macrophages in residual HCC appeared to occur predominantly within fibrovascular bundles. PDL1+ macrophages interacted more frequently with both fibroblasts (pAdj=2.99xlO-3), and endothelial cells (p=7.7x!0-3) in residual HCC than primary HCC (FIG. 3F). Similarly, it was found that there was closer spatial proximity between PDL1+ macrophages and fibroblasts (44.99 vs. 179.75pm; p = 0.027) or endothelial cells (36.2 vs. 68.5 pm; p = 0.004) in residual than primary HCC. This was not observed for the other two subsets of HLA-DR+ or CD206+ macrophages and fibroblasts (42.2 vs. 38.6 pm p = 0.329; 30.3 vs. 31.9, p = 0.197, respectively). Thus, in residual HCC, PDL1+ macrophages appear to directly interact with, and may result in the exhaustion of CD8T cells within fibrovascular bundles.

[0137] Next, interactions that did not occur by direct contact but could still be mediated indirectly by intermediary cells were examined. To first establish the range of indirect influence of a central cell, a regression model was developed. It was found that the variance in a given central cell's expression of Ki67 and BCL2 could be explained by evaluating interaction radius ranging from 25 to 100pm (FIG. 3G). Based on this, indirect interactions between non-homotypic cells lying between 25 pm and 100pm of a given cell were evaluated.

[0138] In residual versus primary HCC, all four tumor cell subsets more frequently indirectly interacted with PDL1+ macrophages (FIG. 3H). However, different from what was observed for direct interactions, in residual HCC, the tumor cells did indirectly interact more frequently with exhausted CD8T cells (FIG. 31). This raises the possibility that PDL1+ macrophages could serve as an intermediary facilitating indirect interactions between tumor cells and exhausted CD8T cells. Thus, residual HCC in contrast to primary HCC exhibits differences both in cellular composition, and in cellular direct and indirect interactions. This apparent remodeling of the cellular types and interactions in residual HCC amongst the stem-like tumor cells, PDL1+ macrophages, and exhausted CD8T cells could be responsible for changes in immune surveillance, as discussed below.

[0139] Spatial Neighborhoods Reprogram Macrophages and CD8T Cells in Residual HCC Next, it was investigated whether the remodeling of cell-cell interactions in residual HCC has a higher-order spatial organization. To test this, "cellular neighborhoods (CNs)” were defined, aiming to capture the intricate spatial arrangements within HCC as opposed to viewing them just as uniform sheets of cells. To determine these spatial CNs, a previously reported approach that defined neighborhoods by clustering individual cells and their neighbors was employed to identify broad patterns of spatial organization32(FIG. 4A). Nine distinct neighborhoods were identified across all the HCC tissues (FIG. 4B). Among these, four were tumor cell-dominant, four were immune cell-dominant, and one was of mixed cell population (termed “other”, of unclear significance) (FIG. 4B). Thus, through CN analysis, nine distinct neighborhoods of higher-order spatial organization were delineated within HCC.

[0140] To confirm that these spatial structures were real, the following three approaches were used. First, the neighborhoods were overlaid with H&E-stained sections and fluorescent images to confirm the accurate recapitulation of known spatial structures such as fibrovascular bundles and lymphoid infiltrates. Second, the presence of these CNs in tumors was observed across all stages, grades, and etiologies, suggesting their presence was a shared feature in the HCC microenvironment. Third, it was confirmed that the canonical marker of the dominant cell within each CN was indeed overexpressed in its respective CN. Thus, the microenvironment of HCC appears to be organized into spatial CNs.

[0141] Next, it was examined whether the distribution of the identified spatial CNs was different between residual and primary HCCs (FIG. 4C). Residual HCCs exhibited a higher prevalence of two neighborhoods, the M2-like macrophage immune CN (p=4xl0-5), and the vascular inflammatory tumor CN (p=0.02). In contrast, the CNs with abundant anti-tumor immune cells, the innate immune CN (p=8.5xl0-7), and T cell immune CN (p=0.012), were less frequent in residual HCC (FIG. 4C). These findings show that in residual versus primary HCC there is a pro-tumor restructuring of the microenvironment.

[0142] The M2-like macrophage immune CN enriched in residual HCC was examined. This CN may influence the cellular phenotype of tumor cells and T cells residing within it. The tumor cells residing in the M2-macrophage CN (17,046 cells, 7.4%) exhibited a more cancer stem-like (CK19hlgh / EpCAMhlgh / CD44hlgh), mesenchymal (vimentinhlgh), and pro-survival (BCL2hlgh, Ki67hlgh) phenotype than the tumor cells within another CN abundant in tumor cells but devoid of M2-like macrophages, the pauci-immune tumor CN (109,391 cells, 47.2%) (FIG. 4D). Additionally, T cells within the M2-macrophage CN (4299 cells, 25.9%) exhibited a more exhausted phenotype (PDlhlgh, TIM3hlgh, CD44low, CD45ROlow) than the T cells residing within another CN enriched in T-cells (8345 cells, 50.3%) (FIG. 4D). The observations at the cellular level were also true at the tissue level. The residual HCC with higher M2-like macrophage CN had greater infiltration of EpCAM-i- tumor cells and exhausted CD8T cells. Thus, M2-like macrophages appear to promote CD8T cell exhaustion within specific spatial CNs rather than across the entire tumor. This may facilitate immune evasion of stem-like tumor cells located within the M2-macrophage immune CN.

[0143] Next, the phenotype of macrophages and T cells residing within the two stem-like tumor cell CNs were investigated in residual HCC. Macrophages residing within the two stem-like tumor CNs were more likely to be M2-like (CD206hlgh, CD163hlgh), and CD8T cells more likely to be exhausted (PDlhlgh, TIM3hlgh), than in the other two non-stem-like tumor CNs (FIG. 4D). Additionally, the abundance of EpCAM+-tumor CNs was associated with poor recurrence-free survival in residual HCC (p=0.02, HR 5.0) (FIG. 4E). Taken together, these data indicate that the spatial organization into CNs is distinct in residual HCC than primary HCC. Specifically, the M2-macrophage CN and EpCAM+-tumor cell CN appear to promote spatially constrained CD8T cell exhaustion. This suggests that such spatial organization into M2-macrophage CN may serve as a mechanism by which residual tumor cells evade CD8T cell surveillance.

[0144] To experimentally test these observations from post-TACE residual human HCC, in vitro experiments were conducted to investigate how resistance to doxorubicin, the most commonly used chemotherapy agent in TACE, reprograms cancer cells and macrophages. Employing single-cell RNA sequencing, 3D heterotypic tumoroids (n=31,058 cells) composed of either doxorubicin-resistant or control HCC cells were analyzed (FIG. 4F). In the doxorubicin resistant tumoroids, cancer cells demonstrated a stem-cell-like phenotype with higher expression of sternness, chemoresistance, and cytokines which can drive M2-like macrophage polarization. Further, doxorubicin resistant tumoroids were enriched in a specific macrophage cluster (C3) which showed a distinct M2-like phenotype, in contrast to the C4 cluster, which was depleted in doxorubicin resistant tumoroids iCXCR4 / LPLP '-, TAT1 / 1 G15' '’) (FIGS. 4G-4H). Additionally, IF analysis confirmed that macrophages cocultured with doxorubicin-resistant cancer cells exhibited PDL1 overexpression compared to those co-cultured with control cells (p=1.6xl0-6) (FIG. 41). In concordance, IF and flow cytometry analysis confirmed that monocyte-derived macrophages from patients with HCC treated with the conditioned media from doxorubicin-resistant patient-derived tumoroids exhibited significantly higher PDL1 expression compared to those treated with conditioned media from control tumoroids (p=1.6xl0-6) (FIG. 4J) or with doxorubicin alone. Overall, these findings demonstrate that doxorubicin-resistant cancer cells not only adopt a more stem-cell-like phenotype but also significantly influence macrophage polarization towards an M2-like phenotype with elevated PDL1 expression.

[0145] TGFf pathway activation is a mechanism for persistence of residual tumor cells in HCC

[0146] To investigate the transcriptional changes within the tumor cell and macrophage neighborhoods identified by CODEX analysis, spatially resolved transcriptomics of residual HCC were used. The expression of 1812 tumor and immune-related genes within tumor-cell (pan CK+ve, CD45- / CD68-) and macrophage (pan CK-ve, CD45+ / CD68+) areas of interest (AOIs) were separately quantified using Nanostring GeoMx DSP (FIGS. 5A-5B). A total of 105 AOIs (tumor-cell (n=53) and macrophage (n=52) AOIs were included. The robustness of AOI classification was confirmed by canonical gene expression of each AOI (FIG. 5C). Pathways activated within the AOIs of residual HCC were compared to non-tumorous liver AOIs. Residual HCC tumor-cell AOIs displayed upregulated immunosuppressive IL10 pathway (p=4.7xl0-47) and PDL1 pathway (p= 2.79xl0-37) (FIG. 5D). On the other hand, macrophage AOIs showed upregulated angiogenesis (p= 3.75xlO10) and invasiveness (p= 8.33xl0-9) pathways (FIG. 5D). Thus, spatial transcriptomics revealed distinct yet complementary pro-tumor pathway activation within the tumor-cell and macrophage AOIs of residual HCC.

[0147] To examine if transcriptional changes within tumor-cell or macrophage AOIs were prognostic, two gene signatures enriched in the respective AOIs of residual, but not primary HCC were established (FIG. 5E). The residual HCC tumor-cell and macrophage signatures were applied to stratify the independent cohort of human TCGA HCC (n=372). Enrichment of the macrophage signature was associated with poor overall- and recurrence-free survival on multivariable analysis of TCGA HCC, adjusting for age, sex, and tumor stage (p=8.2xl0-4, HR 1.9 [1.3-3.1]) (FIG. 5F). In contrast, the tumor-cell signature did not show significant associations with overall or recurrence-free survival (p=0.628, HR 1.1 [0.8-1.6]) (FIG. 5F). This suggests that macrophages, and not residual tumor cells alone, influence the trajectory of residual HCC towards recurrence and poor prognosis.

[0148] Next, upstream regulators of the molecular pathways activated within tumor-cell and macrophage AOIs were assessed. The receptor TGFBR1 / 2 kinase pathway was the top upstream regulator of the transcriptional changes in the tumor-cell AOIs (p=3.1x!0-44, z- score 2.5), while its corresponding ligand, TGFB1 was the top upstream regulator of transcriptional changes in the macrophage AOIs of residual HCC (p=2.7xl0-9, z-score 1.2) (FIG. 5G). Moreover, TGFB1 was expressed at a higher level in the macrophage AOIs than tumor AOIs of residual HCC (FIG. 5H). TGFB 1 pathway effector gene SMAD2 positively correlated with CD274 (PDL1) within residual HCC, but not primary HCC (FIG. 51). Further, mRNA FISH showed that TGFB1 mRNA expression was indeed higher in residual than primary HCC (FIG. 5J). Additionally, multiplex FISH confirmed that the proportion of TGFB1 expressing cells which were CD68+ macrophages was also higher in residual than primary HCC (FIG. 5J). Collectively, the integration of CODEX and spatial transcriptomics analyses elucidates the central role of macrophage-mediated immunosuppressive pathways in residual tumor cell persistence. PDL1+ macrophages, identified to be enriched in residual HCC on CODEX analysis, may serve as a source for TGFpi pathway activation in residual tumor cells.

[0149] DISCUSSION

[0150] Spatial analysis of both human clinical samples and transgenic mouse models of HCC were combined to determine the mechanism of persistence of residual disease in HCC, and to suggest a possible therapeutic approach to improve clinical outcome. Interactions between stem-like tumor cells and immunosuppressive PDL1+ macrophages were identified within spatially constrained neighborhoods, in both human and mouse residual HCC, were found to be linked to CD8T cell exhaustion and immune evasion. Further, macrophage- mediated TGF[3 pathway activation within residual tumor cells was found to enable persistence of both human and mouse MRD. Finally, preclinical evidence shows that combined inhibition of PDL1+ macrophage-mediated TGFP activation eliminates MRD in mouse HCC and prevents recurrence. These results provide mechanistic insight that suggests a therapeutic strategy for eliminating residual disease to prevent cancer recurrence.

[0151] The power of multiple complementary spatial biology technologies was harnessed here to create a high-resolution spatial map of residual human HCC. Notably, residual disease that persists after therapy is a poorly characterized entity, especially in human solid tumors, due to challenges in detecting these areas. A few recent studies have employed single-cell RNA sequencing of primary HCC43-45or CODEX of treatment-naive HCC46,47, however, the spatial organization of post-TACE residual HCC is not known. Spatial mapping provides three key insights. First, spatial interactions of cancer stem cells and protumor M2-like macrophages are critical drivers of recurrence in residual HCC, rather than just the abundance of these cells, as previously suggested48-50. Second, monocyte-derived PDL1+ macrophages are recruited to residual HCC and are associated with exhaustion of CD8T cells4051,52, a finding that adds a new dimension to their known role in cancer progression53-55. Third, the spatial organization of residual HCC into M2-like macrophage enriched neighborhoods, where CD8T cells are rendered exhausted, allows stem-like tumor cells within these constrained areas to evade immune detection. Thus, spatial analysis identified tumor cell and host immune cell interactions could enable evasion of immune surveillance and eventual tumor recurrence of HCC.

[0152] This study creates a unique comprehensive single-cell spatial map of postchemoembolization residual human HCC, revealing critical interactions among stem-like tumor cells, M2-like macrophages, and exhausted CD8T cells, insights gained only by preserving spatial context. Moreover, using a transgenic mouse model for disseminated MRD, it is demonstrated that the insights gained from spatial analysis can indeed be actionable, thus providing guidance to target Tgl'f> and Pdll to eliminate MRD in HCC. Thus, these results suggest a new adjuvant therapeutic strategy for reducing recurrence in HCC and improving patient outcomes.

[0153] REFERENCES

[0154] 1. Lee, Y.-T. et al. The Mortality and Overall Survival Trends of Primary Liver Cancer in the United States. J. Natl. Cancer Inst. 113, 1531-1541 (2021).

[0155] 2. Lei, J. et al. Response to transarterial chemoembolization may serve as selection criteria for hepatocellular carcinoma liver transplantation. Oncotarget 8, 91328— 91342 (2017).

[0156] 3. Kim, D. J. et al. Recurrence of hepatocellular carcinoma: importance of mRECIST response to chemoembolization and tumor size. Am. J. Transplant 14, 1383-1390 (2014).

[0157] 4. Agopian, V. G. et al. Impact of Pretransplant Bridging Locoregional Therapy for Patients With Hepatocellular Carcinoma Within Milan Criteria Undergoing Liver Transplantation: Analysis of 3601 Patients From the US Multicenter HCC Transplant Consortium. Ann. Surg. 266, 525-535 (2017).

[0158] 5. Adeniji, N. et al. Impact of Bridging Locoregional Therapies for Hepatocellular Carcinoma on Post-transplant Clinical Outcome. Clin. Transplant. 34, el4128 (2020).

[0159] 6. Ghajar, C. M. Metastasis prevention by targeting the dormant niche. Nat. Rev. Cancer 15, 238-247 (2015). 7. Cable, J. et al. Cancer stem cells: advances in biology and clinical translation- a Keystone Symposia report. Ann. N. Y. Acad. Sci. 1506, 142-163 (2021).

[0160] 8. Cole, A. J., Fayomi, A. P., Anyaeche, V. I., Bai, S. & Buckanovich, R. J. An evolving paradigm of cancer stem cell hierarchies: therapeutic implications. Theranostics 10, 3083-3098 (2020).

[0161] 9. Lee, T. K.-W., Guan, X.-Y. & Ma, S. Cancer stem cells in hepatocellular carcinoma - from origin to clinical implications. Nat. Rev. Gastroenterol. Hepatol. 19, 26-44 (2022).

[0162] 10. Price, T. T. et al. Dormant breast cancer micrometastases reside in specific bone marrow niches that regulate their transit to and from bone. Sci. Transl. Med. 8, 340ra73 (2016).

[0163] 11. Gilbertson, R. J. & Graham, T. A. Cancer: Resolving the stem-cell debate. Nature vol. 488 462-463 (2012).

[0164] 12. Lan, L. & Behrens, A. Are There Specific Cancer Stem Cell Markers? Cancer Res. 83, 170-172 (2023).

[0165] 13. Shachaf, C. M. et al. MYC inactivation uncovers pluripotent differentiation and tumour dormancy in hepatocellular cancer. Nature 431, 1112-1117 (2004).

[0166] 14. Ding, W. et al. Clinicopathologic and prognostic significance of tumor- associated macrophages in patients with hepatocellular carcinoma: A meta-analysis. PLoS One 14, e0223971 (2019).

[0167] 15. Zhu, X.-D. et al. High expression of macrophage colony-stimulating factor in peritumoral liver tissue is associated with poor survival after curative resection of hepatocellular carcinoma. J. Clin. Oncol. 26, 2707-2716 (2008).

[0168] 16. Borriello, L. et al. Primary tumor associated macrophages activate programs of invasion and dormancy in disseminating tumor cells. Nat. Commun. 13, 626 (2022).

[0169] 17. Dhanasekaran, R. et al. MYC Overexpression Drives Immune Evasion in Hepatocellular Carcinoma That Is Reversible through Restoration of Proinflammatory Macrophages. Cancer Res. 83, 626-640 (2023).

[0170] 18. Dhanasekaran, R. et al. and cooperate to drive metastasis by eliciting crosstalk between cancer and innate immunity. Elife 9, (2020).

[0171] 19. Merritt, C. R. et al. Multiplex digital spatial profiling of proteins and RNA in fixed tissue. Nat. Biotechnol. 38, 586-599 (2020). 20. Zollinger, D. R., Lingle, S. E., Sorg, K., Beechem, J. M. & Merritt, C. R. GeoMx™ RNA Assay: High Multiplex, Digital, Spatial Analysis of RNA in FFPE Tissue. Methods Mol. Biol. 2148, 331-345 (2020).

[0172] 21. Black, S. et al. CODEX multiplexed tissue imaging with DNA-conjugated antibodies. Nat. Protoc. 16, 3802-3835 (2021).

[0173] 22. Uenishi, T. et al. Cytokeratin 19 expression in hepatocellular carcinoma predicts early postoperative recurrence. Cancer Sci. 94, 851-857 (2003).

[0174] 23. Khosla, R. et al. EpCAM+ Liver Cancer Stem-Like Cells Exhibiting Autocrine Wnt Signaling Potentially Originate in Cirrhotic Patients. Stem Cells Transl. Med. 6, 807-818 (2017).

[0175] 24. Liu, C.-Q. et al. Expression patterns of programmed death ligand 1 correlate with different microenvironments and patient prognosis in hepatocellular carcinoma. Br. J. Cancer 119, 80-88 (2018).

[0176] 25. Pfister, D. et al. NASH limits anti-tumour surveillance in immunotherapy- treated HCC. Nature 592, 450-456 (2021).

[0177] 26. Comprehensive and Integrative Genomic Characterization of Hepatocellular Carcinoma. Cell 169, 1327-1341.e23 (2017).

[0178] 27. Guo, Y. et cd. The Heterogeneity of Immune Cell Infiltration Landscape and Its Immunotherapeutic Implications in Hepatocellular Carcinoma. Front. Immunol. 13, 861525 (2022).

[0179] 28. Yu, S. et al. Tumor-infiltrating immune cells in hepatocellular carcinoma: Tregs is correlated with poor overall survival. PLoS One 15, e0231003 (2020).

[0180] 29. Veglia, F. et cd. Analysis of classical neutrophils and polymorphonuclear myeloid-derived suppressor cells in cancer patients and tumor-bearing mice. J. Exp. Mecl. 218, (2021).

[0181] 30. Salcher, S. et al. High-resolution single-cell atlas reveals diversity and plasticity of tissue-resident neutrophils in non-small cell lung cancer. Cancer Cell 40, 1503— 1520.e8 (2022).

[0182] 31. Singal, A. G. et al. AASLD Practice Guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology 78, 1922-1965 (2023).

[0183] 32. Schiirch, C. M. et al. Coordinated Cellular Neighborhoods Orchestrate Antitumoral Immunity at the Colorectal Cancer Invasive Front. Cell 183, 838 (2020). 33. Dasari, A., Grothey, A. & Kopetz, S. Circulating Tumor DNA-Defined Minimal Residual Disease in Solid Tumors: Opportunities to Accelerate the Development of Adjuvant Therapies. J. Clin. Oncol. 36, JCO2018789032 (2018).

[0184] 34. Villanueva, A. et al. Combining clinical, pathology, and gene expression data to predict recurrence of hepatocellular carcinoma. Gastroenterology 140, 1501-12.e2 (2011).

[0185] 35. Pello, O. M. et al. Role of c-MYC in alternative activation of human macrophages and tumor-associated macrophage biology. Blood 119, 411-421 (2012).

[0186] 36. Li, Z. et al. Cancer-associated fibroblasts promote PD-L1 expression in mice cancer cells via secreting CXCL5. Int. J. Cancer 145, 1946-1957 (2019).

[0187] 37. Zhang, W. et al. IL-6 promotes PD-L1 expression in monocytes and macrophages by decreasing protein tyrosine phosphatase receptor type O expression in human hepatocellular carcinoma. J Immunother Cancer 8, (2020).

[0188] 38. Xi, X. et al. Interleukin-22 promotes PD-L1 expression via STAT3 in colon cancer cells. Oncol. Lett. 22, 716 (2021).

[0189] 39. Kobori, T. et al. Interleukin- 18 Amplifies Macrophage Polarization and Morphological Alteration, Leading to Excessive Angiogenesis. Front. Immunol. 9, 334 (2018).

[0190] 40. Petty, A. J. et al. Hedgehog-induced PD-L1 on tumor-associated macrophages is critical for suppression of tumor-infiltrating CD8+ T cell function. JCI Insight 6, (2021).

[0191] 41. Peranzoni, E. et al. Macrophages impede CD8 T cells from reaching tumor cells and limit the efficacy of anti-PD-1 treatment. Proc. Natl. Acad. Sci. U. S. A. 115, E4041-E4050 (2018).

[0192] 42. Patel, M. & Pillai, A. Management of Intermediate-Stage Hepatocellular Carcinoma: Systemic Versus Locoregional Therapy. Surg. Oncol. Clin. N. Am. 33, 159-172 (2024).

[0193] 43. Lu, Y. et al. A single-cell atlas of the multicellular ecosystem of primary and metastatic hepatocellular carcinoma. Nat. Commun. 13, 4594 (2022).

[0194] 44. Ma, L. et al. Single-cell atlas of tumor cell evolution in response to therapy in hepatocellular carcinoma and intrahepatic cholangiocarcinoma. J. Hepatol. 75, 1397-1408 (2021).

[0195] 45. Ma, L. et al. Tumor Cell Biodiversity Drives Microenvironmental Reprogramming in Liver Cancer. Cancer Cell 36, 418-430.e6 (2019). 46. Ruf, B. et al. Tumor-associated macrophages trigger MAIT cell dysfunction at the HCC invasive margin. Cell 186, 3686-3705. e32 (2023).

[0196] 47. Maestri, E. et al. Spatial proximity of Tumor-Immune interactions predicts patient outcome in hepatocellular carcinoma. Hepatology (2023) doi: 10.1097 / HEP.0000000000000600.

[0197] 48. Govaere, O. et al. Keratin 19: a key role player in the invasion of human hepatocellular carcinomas. Gut 63, 674-685 (2014).

[0198] 49. Zhang, J. et al. Overexpression of Epcam and CD 133 Correlates with Poor Prognosis in Dual-phenotype Hepatocellular Carcinoma. J. Cancer 11, 3400-3406 (2020).

[0199] 50. Noh, C.-K. et al. EpCAM as a Predictive Marker of Tumor Recurrence and Survival in Patients Who Underwent Surgical Resection for Hepatocellular Carcinoma. Anticancer Res. 38, 4101-4109 (2018).

[0200] 51. Huang, Y.-K. et al. Macrophage spatial heterogeneity in gastric cancer defined by multiplex immunohistochemistry. Nat. Commun. 10, 3928 (2019).

[0201] 52. Liu, Y. et al. Immune Cell PD-L1 Colocalizes with Macrophages and Is Associated with Outcome in PD-1 Pathway Blockade Therapy. Clin. Cancer Res. 26, 970- 977 (2020).

[0202] 53. Zhou, C. et al. Disruption of SLFN11 Deficiency-Induced CCL2 Signaling and Macrophage M2 Polarization Potentiates Anti-PD- 1 Therapy Efficacy in Hepatocellular Carcinoma. Gastroenterology 164, 1261-1278 (2023).

[0203] 54. Murai, H. et al. Multiomics identifies the link between intratumor steatosis and the exhausted tumor immune microenvironment in hepatocellular carcinoma. Hepatology 77, 77-91 (2023).

[0204] 55. Li, X. et al. Targeting of tumour-infiltrating macrophages via CCL2 / CCR2 signalling as a therapeutic strategy against hepatocellular carcinoma. Gut 66, 157-167 (2017).

[0205] 56. Martin-Padura, I. et al. Residual dormant cancer stem-cell foci are responsible for tumor relapse after antiangiogenic metronomic therapy in hepatocellular carcinoma xenografts. Lab. Invest. 92, 952-966 (2012).

[0206] 57. Haraguchi, N. et al. CD 13 is a therapeutic target in human liver cancer stem cells. J. Clin. Invest. 120, 3326-3339 (2010).

[0207] 58. Schrader, J. et al. Matrix stiffness modulates proliferation, chemotherapeutic response, and dormancy in hepatocellular carcinoma cells. Hepatology 53, 1192-1205 (2011). 59. Ravi, R. et al. Bifunctional immune checkpoint-targeted antibody-ligand traps that simultaneously disable TGFP enhance the efficacy of cancer immunotherapy. Nat. Commun. 9, 741 (2018).

[0208] 60. Mariathasan, S. et al. TGFP attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature 554, 544-548 (2018).

[0209] 61. Lan, Y. et al. Enhanced preclinical antitumor activity of M7824, a bifunctional fusion protein simultaneously targeting PD-L1 and TGF-p. Sci. Transl. Med. 10, (2018).

[0210] 62. Metropulos, A. E., Munshi, H. G. & Principe, D. R. The difficulty in translating the preclinical success of combined TGFP and immune checkpoint inhibition to clinical trial. EBioMedicine 86, 104380 (2022).

[0211] 63. Lencioni, R. & Llovet, J. M. Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin. Liver Dis. 30, 52-60 (2010).

[0212] 64. Wang, S. K., Xue, Y. & Cepko, C. L. Microglia modulation by TGF-P 1 protects cones in mouse models of retinal degeneration. J. Clin. Invest. 130, 4360-4369 (2020).

[0213] 65. Kress, S. et al. p53 mutations are absent from carcinogen-induced mouse liver tumors but occur in cell lines established from these tumors. Mol. Carcinog. 6, 148- 158 (1992).

[0214] 66. Brown, Z. J., Heinrich, B. & Greten, T. F. Establishment of Orthotopic Liver Tumors by Surgical Intrahepatic Tumor Injection in Mice with Underlying Non-Alcoholic Fatty Liver Disease. Methods Protoc 1, (2018).

[0215] 67. Zabransky, D. J. et al. Profiling of syngeneic mouse HCC tumor models as a framework to understand anti-PD-1 sensitive tumor microenvironments. Hepatology 77, 1566-1579 (2023).

[0216] Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it is readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.

Claims

CLAIMSWhat is claimed is:

1. A method for identifying a cancer patient that is at high risk for cancer recurrence, comprising: analyzing a section of a tumor obtained from the cancer patient to quantify interactions between cancer stem cells and PDL1+macrophages in the tumor; wherein a higher level of interactions between cancer stem cells and PDL1+macrophages in the tumor indicates that the patient is at high risk for cancer recurrence.

2. The method of claim 1, further comprising: identifying the patient as having a high risk of cancer recurrence based on the analysis; and administering an adjuvant therapy to the patient.

3. The method of claim 2, wherein the adjuvant therapy comprises an immune checkpoint therapy and / or chemotherapy.

4. The method of claim 2 or 3, wherein the adjuvant therapy comprises (i) atezolizumab and bevacizumab, (ii) durvalumab and tremelimumab or (iii) nivolumab and ipilimumab.

5. The method of any prior claim, wherein the analysis is done by performing a multiplexed binding assay on the section to identify at least:(i) the cancer stem cells and(ii) the PDL1+ macrophages.

6. The method of claim 5, wherein the cancer stem cells are positive for: (a) PanCK, CK19 and EpCAM and, optionally, CD44, or (b) positive for PanCK and at least one of CK19, EpCAM and CD44.

7. The method of any of claims 5 and 6, wherein the PDL1+macrophages (a) are positive for CD68, CD206, and PDL1, or (b) arc CD68+ / PDLl+ / PanCK-vc.

8. The method of any prior claim, wherein the cancer patient has hepatocellular carcinoma (HCC).

9. The method of claim 8, wherein the cancer stem cells are CD44Hi10. The method of any prior claim, comprising treating the patient by surgical resection and / or administering an adjuvant therapy to the patient..

11. The method of any prior claim, wherein the method comprises analyzing the edge of the tumor to quantify interactions between cancer stem cells and PDL1+macrophages at the tumor edge.

12. A method of treatment, comprising: analyzing a section of a tumor obtained from the cancer patient to quantify interactions between cancer stem cells and PDL1+macrophages in the tumor, wherein a higher level of interactions between cancer stem cells and PDL1+macrophages in the tumor indicates that the patient is at high risk for cancer recurrence; identifying the patient has having a high risk of cancer recurrence based on the analysis; and administering an adjuvant therapy to the patient.

13. The method of claim 12, wherein the adjuvant therapy comprises an immune checkpoint therapy and / or chemotherapy.

14. The method of claim 12 or 13, wherein the adjuvant therapy comprises (i) atezolizumab and bevacizumab, (ii) durvalumab and tremelimumab or (iii) nivolumab and ipilimumab.

15. The method of any of claims 12-14, wherein the cancer stem cells are positive for: (a) PanCK, CK19 and EpCAM and, optionally, CD44, or (b) PanCK and at least one of CK19, EpCAM and CD44, and the PDL1+macrophages are (a) positive for CD68, CD206, and PDL1 or (b) CD68+ / PDLl+ / PanCK-ve.

16. The method of any of claims 12-15, wherein the patient has HCC.

17. The method of any of claims 12-16, wherein the method comprises analyzing the edge of the tumor to quantify interactions between cancer stem cells and PDL1+macrophages at the tumor edge.

18. A kit comprising: a panel of antibodies that label cancer stem cells and a panel of antibodies that label PDL1+macrophages.

19. The kit of claim 18, wherein the kit comprises antibodies that bind to(a) (i) PanCK, CK19, EpCAM and optionally CD44, or (ii) PanCK and at least one of CK19, EpCAM and CD44; and(b) (i) CD68, CD206, and PDL1 or (ii) CD68 and PDL1 and PanCK.

20. The kit of claim 18 or 19, wherein the antibodies are distinguishably labeled.

Images