Methods for identification and targeting of tumor hybrid cells

Abstract

The subject matter described herein relates to a method of treating cancer in a subject in need thereof.

Description

METHODS FOR IDENTIFICATION AND TARGETING OF TUMOR HYBRID CELLS

[0001] This application claims the benefit of and priority to US Provisional Patent Application No. 63 / 570611, filed March 27,2024, which is incorporated herein by reference in its entirety.

[0002] This patent disclosure contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves any and all copyright rights.

[0003] All patents, patent applications and publications cited herein are hereby incorporated by reference in their entirety. The disclosure of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art as known to those skilled therein as of the date of the invention described herein.BACKGROUND

[0004] Cancer represents a group of conditions characterized by abnormal cell growth. Cancerous cells have the potential to spread and invade other organs of the body. The most common symptoms of cancer include a lump, abnormal bleeding, prolonged cough, and unexplained weight loss. There are also cancers of the blood which do not form a cell mass. Over 100 types of cancers can develop within the human body and most of them are incurable.

[0005] Cancer cells can undergo fusion with other cancer cells, as well as immune cells, fibroblasts, and other mesenchymal cells and these cells can have increased metastatic potential and drug resistance. There is a need for improved methods of identifying, characterizing, and treating tumor hybrid cells.SUMMARY OF THE INVENTION

[0006] In certain aspects, the subject matter described herein provides a method of treating or preventing cancer in a subject in need thereof, the method comprisingadministering to the subject a cancer treatment, wherein a sample from the subject comprises one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

[0007] In some embodiments, the one or more hybrid cells express at least two epithelial cell lineage markers and at least two monocyte or macrophage cell lineage markers. In some embodiments, the one or more epithelial cell lineage markers are selected from EPCAM, KRT8, KRT9 (CK9), KRT19 (CK19), or CK18; and the one or more monocyte or macrophage cell lineage markers are selected from CD163, CD14, CD68, FCER1G, TYROBP, or AIF1. In some embodiments, the epithelial cell lineage marker is EPCAM and the monocyte or macrophage cell lineage markers is macrophage cell lineage marker CD 163. In some embodiments, the at least two epithelial cell lineage markers are EPCAM and KRT8, and at least two monocyte or macrophage cell lineage markers are macrophage cell lineage markers CD163 and CD14.

[0008] In some embodiments, the one or more hybrid cells further express one or more blood lineage markers. In some embodiments, the blood lineage marker is PTPRC (CD45). In some embodiments, one or more hybrid cells have a doublet score of <0.5. In some embodiments, one or more hybrid cells have a doublet score between 0.1 and 0.5.

[0009] In some embodiments, a cell type annotation of the one or more hybrid cells is a myeloid cell type. In some embodiments, the myeloid cell type is monocyte or macrophage. In some embodiments, the cell type annotation of the one or more hybrid cells is an epithelial cell type. In some embodiments, the cell type annotation is a graph-clustering and nonnegative matrix factorization (NMF) approach.

[0010] In some embodiments, a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subjects, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects. In some embodiments, a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject.

[0011] In some embodiments, the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared to signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject, or signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of one or more healthy subjects. In some embodiments, the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared to signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject.

[0012] In some embodiments, the gene expression pattern of the one or more hybrid cells comprises gene expression of one or more of the genes described in Figure 15 or Figure 16. In some embodiments, the gene expression pattern is for genes associated with epithelial-to- mesenchymal transition (EMT). In some embodiments, the one or more hybrid cells comprise a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non-cancerous cell of the subject, or a gene expression level of said genes in a non-cancerous cell of one or more healthy subjects. In some embodiments, the one or more hybrid cells comprise a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non- cancerous cell of the subject.

[0013] In some embodiments, the gene expression is expression of CD44, SPP1, or both CD44 and SPP1. In some embodiments, the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject. In some embodiments, the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of one or more healthy subjects.

[0014] In some embodiments, the cancer treatment comprises one or more bispecific molecules comprising at least two antigen binding regions, including any cancer treatment described in international patent publication No. WO202313007353. In some embodiments, the first and second antigen binding regions bind, respectively, a first and second antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, a geneexpression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subjects, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects. In some embodiments, the first and second antigen binding regions bind, respectively, a first and second antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject. In some embodiments, the antigens are selected from the antigens encoded by the genes described in Figure 15 or Figure 16. In some embodiments, the antigens are selected from EPCAM, CD163, SPP1, or CD44.

[0015] In some embodiments, the cancer treatment comprises one or more monoclonal antibodies comprising an antigen binding region. In some embodiments, the antigen binding regions binds to an antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects. In some embodiments, the antigen binding regions binds to an antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject. In some embodiments, the antigen is selected from the antigens encoded by of the genes described in Figure 15 or Figure 16. In some embodiments, the antigen is selected from EPCAM, CD163, SPP1, or CD44.

[0016] In some embodiments, the cancer treatment comprises one or more chimeric antigen receptor T cell therapy (CAR-T).

[0017] In some embodiments, the cancer treatment comprises one or more antibody-drug conjugates.

[0018] In some embodiments, the cancer is a solid cancer. In some embodiments, the solid cancer is colon, breast, ovarian, pancreatic, renal, esophageal, melanoma, lung, brain, prostate, or uveal melanoma. In some embodiments, the cancer is colorectal cancer. In someembodiments, the cancer is breast cancer. In some embodiments, the cancer is lung cancer. In some embodiments, the lung cancer is non-small cell lung cancer (NSCLC). In some embodiments, the one or more hybrid cells comprises a gene expression profile of a lung cancer cell-type and the subject is administered a lung cancer treatment. In some embodiments, the lung cancer cell-type is NSCLC, and the subject is administered a NSCLC treatment.

[0019] In some embodiments, the NSCLC treatment comprises cisplatin and another drug selected from the group of pemetrexed, gemcitabine, docetaxel, vinorelbine, or etoposide. In some embodiments, the NSCLC treatment comprises Alectinib, Osimertinib, Atezolizumab, or Pembrolizumab. In some embodiments, the NSCLC treatment comprises a wedge resection, segmentectomy, lobectomy, sleeve lobectomy, pneumonectomy, thoracotomy or a thoracoscopy. In some embodiments, the lung cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Non-Small Cell Lung Cancer, Version 3.2025 (January 14, 2025).

[0020] In some embodiments, the one or more hybrid cells comprises a gene expression profile of a colon cell-type and the subject is administered a colon cancer treatment. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a colorectal cell-type and the subject is administered a colorectal cancer treatment.

[0021] In some embodiments, the colorectal cancer treatment comprises colectomy, chemotherapy, adjuvant or neoadjuvant chemotherapy. In some embodiments, the colorectal cancer treatment comprises 5 -Fluorouracil (5-FU), Capecitabine, Irinotecan, Oxaliplatin, Bevacizumab, FOLFOXIRI, Cetuximab, Aflibercept, Ramucirumab, Panitumumab, Anti- epidermal growth factor receptor (EGFR) antibody, anti-vascular endothelial growth factor (VEGF) antibody with first-line chemotherapy, Trifluridine-tipiracil, Regorafenib, Encorafenib with cetuximab (for patients with BRAF V600E mutations), or immunotherapy targeting checkpoints (Pembrolizumab monotherapy, Nivolumab monotherapy, or Nivolumab and ipilimumab). In some embodiments, the colorectal cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Colon cancer, Version 1.2024 (January 29, 2024), the content of which is hereby incorporated by reference in its entirety.

[0022] In some embodiments, the one or more hybrid cells comprises a gene expression profile of a breast cell-type and the subject is administered a breast cancer treatment.

[0023] In some embodiments, the breast cancer treatment comprises a lumpectomy or a mastectomy. In some embodiments, the breast cancer treatment comprises chemotherapy, radiation therapy, adjuvant therapy, or hormonal therapy. In some embodiments, the breast cancer treatment comprises Tamoxifen, Aromatase inhibitor (Al), Trastuzumab, Pertuzumab, Neratinib, or Olaparib. In some embodiments, the breast cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Breast Cancer, Version 2.2024 (March 11, 2024), the content of which is hereby incorporated by reference in its entirety.

[0024] In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using scRNAseq. In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using RT-qPCR.

[0025] In certain aspects the subject matter described herein provides a method for diagnosing cancer in a subject, the method comprising: detecting in a sample one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

[0026] In certain aspects, the subject matter described herein provides a method for detecting cancer recurrence in a subject in need thereof, the method comprising: detecting in a sample one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

[0027] In some embodiments, the one or more hybrid cells express at least two epithelial cell lineage markers and at least two monocyte or macrophage cell lineage markers. In some embodiments, the one or more epithelial cell lineage markers are selected from EPCAM, KRT8, KRT9 (CK9), KRT19 (CK19), or CK18; and the one or more monocyte or macrophage cell lineage markers are selected from CD163, CD14, CD68, FCER1G, TYROBP, or AIF1. In some embodiments, the epithelial cell lineage marker is EPCAM and the monocyte or macrophage cell lineage markers is macrophage cell lineage marker CD 163. In some embodiments, the at least two epithelial cell lineage markers are EPCAM and KRT8, and atleast two monocyte or macrophage cell lineage markers are macrophage cell lineage markersCD163 and CD14.

[0028] In some embodiments, the one or more hybrid cells further express one or more blood lineage markers. In some embodiments, the blood lineage marker is PTPRC (CD45).

[0029] In some embodiments, the one or more hybrid cells have a doublet score of <0.5. In some embodiments, the one or more hybrid cells have a doublet score between 0.1 and 0.5.

[0030] In some embodiments, a cell type annotation of the one or more hybrid cells is a myeloid cell type. In some embodiments, the myeloid cell type is monocyte or macrophage. In some embodiments, a cell type annotation of the one or more hybrid cells is an epithelial cell type. In some embodiments, the cell type annotation is a graph-clustering and nonnegative matrix factorization (NMF) approach.

[0031] In some embodiments, a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject. In some embodiments, the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared to signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject. In some embodiments, the gene expression pattern of the one or more hybrid cells comprises gene expression of one or more of the genes described in Figure 15 or Figure 16.

[0032] In some embodiments, the gene expression pattern is for genes associated with EMT. In some embodiments, the one or more hybrid cells comprise a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non-cancerous cell of the subject.

[0033] In some embodiments, the gene expression is expression of CD44, SPP1, or both CD44 and SPP1. In some embodiments, the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject. In some embodiments, the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increasedcompared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of one or more healthy subjects.

[0034] In some embodiments, the cancer is a solid cancer. In some embodiments, the solid cancer is colon, breast, ovarian, pancreatic, renal, esophageal, melanoma, lung, brain, prostate, or uveal melanoma. In some embodiments, the cancer is colorectal cancer. In some embodiments, the cancer is a lung cancer. In some embodiments, the lung cancer is NSCLC. In some embodiments, the cancer is breast cancer. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a colon cell-type and the subject is administered a colon cancer treatment. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a colorectal cell-type. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a breast cell-type. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a lung celltype. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a NSCLC cell-type.

[0035] In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using scRNAseq. In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using RT-qPCR.

[0036] In certain aspects, the subject matter described herein provides a method of determining a cancer prognosis in a subject in need thereof, wherein the prognosis depends on numbers and / or size of hybrid cells as described herein in the sample. In some embodiments, the subject is determined to have a poor prognosis if the subject has a high number of hybrid cells and / or large hybrid cells.

[0037] In some embodiments, the sample is biopsy tissue or blood. In some embodiments, the subject is a human.BRIEF DESCRIPTION OF FIGURES

[0038] The patent or application file contains at least one drawing executed in color. To conform to the requirements for PCT patent applications, many of the figures presented herein are black and white representations of images originally created in color.

[0039] FIGS. 1A-J show that hybrid cells are found in tumor and normal colon biopsies, frequently associated with myeloid cluster, and show distinct gene expression profile compared to parental cells. (A) Uniform manifold approximation and projection (UMAP) plot of all cells from scRNAseq data. Hybrid cells are marked with only two markers (EPCAM and CD 163) highlighted in blue and the rest of the cells in light grey. (B) UMAP plot of all cells with hybrid cells marked with four markers (EPCAM, KRT8, CD14, CD163) highlighted in purple and rest in light grey. (C) Violin plots with box plots doublet score before (top) and after (bottom) filtering out for cells with >0.5 score. (D) UMAP plot showing expression of four marker genes (EPCAM, KRT8, CD14, CD163) and other epithelial (KRT18 and KRT19) and monocyte / macrophage (('1)68) and leukocyte (CD45) genes. (E) Clustered dotplot of the eight genes noted in D above and 19 cell types along with hybrid cells showing hybrid cells closely clustering with monocyte and macrophage cluster. (F) UMAP plot of myeloid subcluster showing 10 myeloid clusters (Left) and hybrid cells (right). The hybrid cells are found in both monocyte and macrophage cluster but predominantly in macrophage. (G) Dotplot of myeloid subcluster. (H) Volcano plot of differential gene expression between epithelial cells (marked by expression of EPCAM and KRT8) and tumor hybrid cells. (I) Volcano plot of differential gene expression between monocyte / macrophage cells (marked by expression of CD14 and CD163) and tumor hybrid cells. (J) Dotplot of top 5 genes that were differentially expressed between epithelial vs tumor hybrid cells comparison and monocyte / macrophage vs tumor hybrid cells.

[0040] FIGS. 2A-L show that hybrid cells are more frequent in colon tumors and show a distinct gene expression profile marked by upregulation of inflammatory pathways and epithelial-mesenchymal transition genes. (A) UMAP plot of all cells with hybrid cells highlighted in purple and rest in light grey in normal and tumor tissue. (B) A bar plot showing percentage of different cell types between tumor and normal tissues in the dataset. (C) UMAP plot of all cells with all myeloid cells highlighted in purple and rest in light grey in normal and tumor tissue. (D) Donut plot showing percentage of hybrid cells from each tissue type. (E) Table showing patient ID and percentage of hybrid cells in normal and tumor tissue of each patient where a matching normal tissue was available. (F) Volcano plot of differential gene expression between hybrid cells from tumor and normal tissues. (G) Dot plot of 33 top differentially expressed genes (padj < 0.05) between hybrid cells from tumor and normal tissues. (H) Dot plot showing expression of four marker genes and KRT18 in hybrid cells from normal and tumor tissues. (I) A summary gene set enrichment analysis(GSEA) plot of top 10 hallmark pathways along with normalized enrichment scores (NES), pvalue (pval) and padusted (padj) values. Sorted with the most significant pathways at the top. (J) Enrichment plot of epithelial-mesenchymal transition (EMT) pathway. (K) UMAP plot of all cells with tumor hybrid cells highlighted in purple and rest in light grey in cells subsetted based on stage of the tumor. (L) Line plot showing percentage of tumor hybrid cells within the myeloid cells in different stages of tumor.

[0041] FIGS. 3A-G show that hybrid cells are found in tumor and normal colon biopsy sections assayed using in situ high resolution spatial mapping and show distinct gene expression profiles. (A) UMAP plot of all cells with hybrid cells highlighted in purple and rest in light grey in normal tissue section of a non-cancer donor analyzed using Xenium spatial seq. (B) Spatial plot of normal section showing hybrid cells (EPCAM+, KRT8+, CD14+, and CD163+), epithelial cells (EPCAM+ and KRT8+), and monocyte / macrophage (CD 14+ and CD 163+) cells in a section from a normal colon at lx (left, entire section), lOx (middle; zoomed), and 100 (right; zoomed). The zoomed region was manually selected to show a region where all three types of cells were present. (C) UMAP plot of all cells with hybrid cells highlighted in purple and rest in light grey in tumor tissue section of colon cancer patient analyzed using Xenium spatial seq. (D) Spatial plot of tumor section showing hybrid cells (EPCAM+, KRT8+, CD14+, and CD163+), epithelial cells (EPCAM+ and KRT8+), and monocyte / macrophage (CD14+ and CD163+) cells in a section from a colon tumor at lx (left, entire section), lOx (middle; zoomed), and 100 (right; zoomed). The zoomed region was manually selected to show a region where all three types of cells were present. (E) Volcano plot of differential gene expression between hybrid cells from norma and tumor tissues shown in figure 3A and 3C respectively. (F) Dot plot showing expression of four marker genes in hybrid cells from normal and tumor tissues. (G) Spatial plot of tumor section showing expression of four marker genes.

[0042] FIGS. 4A-E show that hybrid cells are found in tumor single cells and biopsy section from breast cancer. (A) UMAP plot of all cells in scRNAseq data from breast tumor tissue. Cells are colored by their cellular identity as reported in Wu, S.Z., et al., A single-cell and spatially resolved atlas of human breast cancers. Nat Genet, 2021. 53(9): p. 1334-1347. (B) UMAP plot of all cells with hybrid cells highlighted in purple and the rest of the cells in light grey in breast tumor tissue. The myeloid cluster is marked with a dashed line. On the right, UMAP plot of all myeloid cells with monocyte (top) and macrophages (bottom)highlighted in red and blue respectively. The rest of the cells are highlighted in light grey. (C) Clustered dotplot of 30 cell types along with hybrid cells showing hybrid cells closely clustering with monocyte and macrophage cluster. (D) UMAP plot of all cells with hybrid cells highlighted in purple and rest in light grey in tumor tissue section of breast cancer patient analyzed using Xenium spatial seq. (E) Spatial plot of tumor section showing hybrid cells (EPCAM+, KRT8+, CD14+, and CD163+), epithelial cells (EPCAM+ and KRT8+), and monocyte / macrophage (CD14+ and CD163+) cells in a section from a breast tumor at lx (left, entire section), lOx (middle; zoomed), and 100 (right; zoomed). The zoomed region was manually selected to show a region where all three types of cells were present.

[0043] FIGS. 5A-E show UMAP plots of scRNAseq data of colon and matched normal tissue showing cell type identities. (A) UMAP plot of all cells from scRNAseq data from colon tumor and normal tissue. Cells are colored by 43 Seurat clusters. (B) Left: UMAP plot of all cells from scRNAseq data from colon tumor and normal tissue. Cells are colored by seven “clTopLevel” cell type identity as defined in Pelka et al. Right: UMAP plot of all cells with each cell type highlighted to appreciate the clustering of each cell type. The rest of the cells are highlighted in light grey. (C) UMAP plot of all cells from scRNAseq data from colon tumor and normal tissue. Cells are colored by 19 “clMidwayPr” cell type identity as defined in Pelka et al. (D) UMAP plot of all cells from scRNAseq data from colon tumor and normal tissue. Cells are colored by 87 “cl295vl ISubFull” cell type identity as defined in Pelka et al. (E) Left: UMAP plot of all cells within myeloid subset as defined in clTopLevel (see above). Cells are colored by 10 myeloid cell type identity as defined in Pelka et al. Right: UMAP plot of each cell type highlighted to appreciate the clustering of each cell type. The rest of the cells are highlighted in light grey.

[0044] FIGS. 6A-B show UMAP plots of scRNAseq data of colon and matched normal tissue from individual patients. (A) UMAP plot of all cells from scRNAseq data from colon tumor and normal tissue from individual patients. Each plot is labeled with patient ID followed by a letter N for normal tissue and T for tumor tissue. Hybrid cells highlighted in purple and rest in light grey. (B) A plot showing absolute number (left axis) and percentage of hybrid cells (right axis) in each patient. The percentage is based on the total number of single cells obtained from that patient.

[0045] FIGS. 7A-B show expression of CD44 and SPP1 genes in THCs. (A) UMAP plot of myeloid subcluster showing expression of CD44 in normal (N) and tumor (T) hybrid cells.The percentage of cells and expression was higher in THC compared to NHC. (B) UMAP plot of myeloid subcluster showing expression of SPP1 in normal (N) and tumor (T) hybrid cells. Expression of SPP1 was exclusive to THC within the myeloid cell subset.

[0046] FIGS. 8A-B show gene set enrichment analysis comparing THC with NHC. (A) A bar plot showing normalized enrichment score (NES) of hallmark pathways. The bars are colored and sorted from lowest to highest based on padjusted values. (B) Enrichment plots showing enrichment of 19 hallmark pathways that were significantly (padj < 0.05) in THC.

[0047] FIGS. 9A-C show Xenium high resolution in situ analysis of tumor and normal section from cancer patients. (A) Spatial plot (left) and UMAP plot (right) showing 30 clusters identified in a section of normal colon. (B) Spatial plot (left) and UMAP plot (right) showing 37 clusters identified in a section of tumor from colon cancer patient. (C) Spatial plot (left) and UMAP plot (right) showing 29 clusters identified in a section of tumor from breast cancer patient. The color scales in both spatial and UMAP plots is same and indicated on the right of UMAP plot.

[0048] FIGS. 10A-10C show Xenium high resolution in situ analysis of tumor and normal section from cancer patients. (A) Two spatial plots of a manually selected regions showing hybrid cells identified in a section of normal colon. (B) Two spatial plots of manually selected regions showing hybrid cells identified in a section of tumor from colon cancer patient. (C) Two spatial plots of manually selected regions showing hybrid cells identified in a section of tumor from breast cancer patient. The color scales (bottom right) are the same for all plots.

[0049] FIG. 11 shows UMAP plot of scRNAseq data of breast cancer tissue showing cell type identities. UMAP plot of all cells from scRNAseq data from breast tumor. Cells are colored by cell type identity as defined in Wu, S.Z., et al., A single-cell and spatially resolved atlas of human breast cancers. Nat Genet, 2021. 53(9): p. 1334-1347.

[0050] FIGS. 12A-C show expression of syncytin-1 and syncytin-2 in scRNAseq colon cancer data. (A) UMAP plot of all cells with cells showing expression of syncytin-1 (ERVW- 1) highlighted in yellow-purple gradient and rest in light grey in normal and tumor tissue. (B) UMAP plot of all myeloid cells with cells showing expression of syncytin-1 (ER VW-1) highlighted in yellow-purple gradient and rest in light grey in tumor tissue. (C) UMAP plot ofall cells with cells showing expression of syncytin-2 (ERVFRD-1) highlighted in yellowpurple gradient and rest in light grey in normal and tumor tissue.

[0051] FIG. 13 shows a list of differentially expressed genes between epithelial cells and tumor hybrid cells from colon tissue (Figure 1H).

[0052] FIG. 14 shows a list of differentially expressed genes between monocyte / macrophage cells and tumor hybrid cells from colon tissue (Figure II).

[0053] FIG. 15 shows a list of differentially expressed genes between normal hybrid cells and tumor hybrid cells from colon cancer (Figure 2F).

[0054] FIG. 16 shows a list of differentially expressed genes between normal hybrid cells and tumor hybrid cells from breast tissue (Figure 3E).

[0055] FIG. 17 shows a UMAP plot of all cells in scRNAseq data from lung tumor tissue. Cells are colored by their cellular identity as reported in De Zuani, M., et al., Singlecell and spatial transcriptomics analysis of non-small cell lung cancer. Nat Communications, 2024. 15(4388): p. 1-17.

[0056] FIG. 18 shows a UMAP plot of all cells with hybrid cells highlighted in purple and the rest of the cells in light grey in lung tumor tissue.

[0057] FIG. 19 shows a clustered dot plot of 47 cell types along with hybrid cells showing hybrid cells closely clustering with myeloid and epithelial cell clusters.DETAILED DESCRIPTION OF THE INVENTION

[0058] All patent applications, published patent applications, issued and granted patents, texts, and literature references cited in this specification are hereby incorporated herein by reference in their entirety to more fully describe the state of the art to which the present disclosed subject matter pertains.

[0059] Unless otherwise defined, 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. Methods and materials are described herein for use in the present invention; other, suitable methods and materials known in the art can also be used. The materials, methods, and examples are illustrative only and not intended to be limiting. Allpublications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

[0060] As used herein, the term “subject” refers to a vertebrate animal. In one embodiment, the subject is a mammal or a mammalian species. In one embodiment, the subject is a human. In one embodiment, the subject is a healthy human adult. In other embodiments, the subject is a non-human vertebrate animal, including, without limitation, non-human primates, laboratory animals, livestock, racehorses, domesticated animals, and non-domesticated animals. In one embodiment, the term “human subjects” means a population of healthy human adults.

[0061] The singular forms “a”, “an” and “the” include plural reference unless the context clearly dictates otherwise.

[0062] Resistance to chemotherapy and metastasis to distant organs are key characteristics of cancer cells associated with mortality. While the mechanisms that contribute to these processes are numerous, a growing body of evidence points to the involvement of tumor hybrid cells (THCs); fusions between cancer cells and stromal cells, such as monocytes and macrophages, which can impart drug resistance and help evade the immune system. THCs can be studied through advances in isolating circulating tumor cells and by analysis of lineage-specific protein markers. However, the precise transcriptomic characteristics of THCs has not been fully detailed, which limits the utility of future research into the impact of tumor hybrid cells.

[0063] In certain aspects, the subject matter described herein provides a framework for detecting tumor hybrid cells in patient samples. In some embodiments, single cell RNA- sequencing is used in combination with in situ spatial transcriptomics to identify and characterize THCs at a complete transcriptome level. In some embodiments, the methods described herein are used to identify THC-specific pathways that drive the formation and maintenance of hybrid cells. In some embodiments, the methods described herein are used to identify and develop therapies that specifically target tumor hybrid cells.Methods of Treatment

[0064] The practice of aspects of the present invention can employ, unless otherwise indicated, conventional techniques of cell biology, cell culture, molecular biology, transgenic biology, microbiology, recombinant DNA, and immunology, which are within the skill of the art. Such techniques are explained fully in the literature. See, e.g., Molecular Cloning A Laboratory Manual, 3rd Ed., ed. by Sambrook (2001), Fritsch and Maniatis (Cold Spring Harbor Laboratory Press: 1989); DNA Cloning, Volumes I and II (D. N. Glover ed., 1985); Oligonucleotide Synthesis (M. J. Gait ed., 1984); Mullis et al. U.S. Pat. No: 4,683,195; Nucleic Acid Hybridization (B. D. Hames & S. J. Higgins eds. 1984); Transcription and Translation (B. D. Hames & S. J. Higgins eds. 1984); Culture Of Animal Cells (R. I. Freshney, Alan R. Liss, Inc., 1987); Immobilized Cells and Enzymes (IRL Press, 1986); B. Perbal, A Practical Guide To Molecular Cloning (1984); the series, Methods In Enzymology (Academic Press, Inc., N.Y.), specifically, Methods In Enzymology, Vols. 154 and 155 (Wu et al. eds.); Gene Transfer Vectors For Mammalian Cells (J. H. Miller and M. P. Calos eds., 1987, Cold Spring Harbor Laboratory); Immunochemical Methods In Cell And Molecular Biology (Caner and Walker, eds., Academic Press, London, 1987); Handbook Of Experimental Immunology, Volumes I-FV (D. M. Weir and C. C. Blackwell, eds., 1986); Manipulating the Mouse Embryo, (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 1986). All patents, patent applications and references cited herein are incorporated by reference in their entireties.

[0065] In certain aspects the subject matter described herein provides a method of treating or preventing cancer in a subject in need thereof, the method comprising administering to the subject a cancer treatment, wherein a sample from the subject comprises one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers. In some embodiments, the one or more hybrid cells express at least two epithelial cell lineage markers and at least two monocyte or macrophage cell lineage markers. In some embodiments, the one or more epithelial cell lineage markers are selected from EPCAM, KRT8, KRT9 (CK9), KRT19 (CK19), or CK18; and the one or more monocyte or macrophage cell lineage markers are selected from CD163, CD14, CD68, FCER1G, TYROBP, or AIF1. In some embodiments, the epithelial cell lineage marker is EPCAM and the monocyte or macrophage cell lineage markers is macrophage cell lineage marker CD163. In some embodiments, the atleast two epithelial cell lineage markers are EPCAM and KRT8, and at least two monocyte or macrophage cell lineage markers are macrophage cell lineage markers CD163 and CD14.

[0066] In some embodiments, the one or more hybrid cells further express one or more blood lineage markers. In some embodiments, the blood lineage marker is PTPRC (CD45). In some embodiments, one or more hybrid cells have a doublet score of <0.5. In some embodiments, one or more hybrid cells have a doublet score between 0.1 and 0.5.

[0067] In some embodiments, a cell type annotation of the one or more hybrid cells is a myeloid cell type. In some embodiments, the myeloid cell type is monocyte or macrophage. In some embodiments, the cell type annotation of the one or more hybrid cells is an epithelial cell type. In some embodiments, the cell type annotation is a graph-clustering and nonnegative matrix factorization (NMF) approach. In some embodiments, the NMF approach is a correlation consensus NMF (ccNMF). In some embodiments, the ccNMF is executed or embodied by the ccNMF algorithm (https: / / github.com / matanhofree / crc-immune- hubs / tree / main / code / dimReduction / ccNMF) described in Pelka et al

[0036] ,

[0068] In some embodiments, a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects. In some embodiments, a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject,

[0069] In some embodiments, the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared to signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject, or signaling levels of KRAS, TP53, or both in a transcripome of a non-cancerous cell of one or more healthy subjects. In some embodiments, the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared to signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject.

[0070] In some embodiments, the gene expression pattern of the one or more hybrid cells comprises gene expression of one or more of the genes described in Figure 15 or Figure 16. In some embodiments, the gene expression pattern is for genes associated with epithelial-to- mesenchymal transition (EMT). Genes associated with EMT are part of the Hallmark gene set collection of Molecular Signatures Database that can be found at www.gsea- msigdb.org / gsea / msigdb / human / geneset / HALLMARK_EPITHELIAL_MESENCHYMAL_T RA.NSITION.html and as described by Liberzon et al., The Molecular Signatures Database (MSigDB) hallmark gene set collection, Cell Syst. 2015 Dec 23;l(6):417-425, the entire disclosures of which are herein incorporated by reference. In some embodiments, the one or more hybrid cells comprise a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non-cancerous cell of the subject, or a gene expression level of said genes in a non-cancerous cell of one or more healthy subjects. In some embodiments, the one or more hybrid cells comprise a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non-cancerous cell of the subject.

[0071] In various embodiments, expression of EMT gene of a subject may be increased by at least 1.5 times greater than EMT gene expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects. Expression of EMT gene of a subject may be increased by at least 3 times greater than EMT gene expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects. Expression of EMT gene of a subject may be increased by at least 5 times greater than EMT gene expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects. Expression of EMT gene of a subject may be increased by at least 10 times greater than EMT gene expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects.

[0072] In some embodiments, the gene expression is expression of CD44, SPP1, or both CD44 and SPP1. In some embodiments, the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject. Insome embodiments, the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of one or more healthy subjects.

[0073] In various embodiments, CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 1.5 times greater than a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects. CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 3 times greater than a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject, or a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of one or more healthy subjects. CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 5 times greater than a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject, or a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of one or more healthy subjects. CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 10 times greater than a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects.

[0074] In some embodiments, the cancer treatment comprises one or more bispecific molecules comprising at least two antigen binding regions. In some embodiments, the first and second antigen binding regions bind, respectively, a first and second antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subject, or a gene expression pattern ofa non-cancerous macrophage or monocyte cell of one or more healthy subjects. In some embodiments, the first and second antigen binding regions bind, respectively, a first and second antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject.

[0075] In some embodiments, the antigens are selected from the antigens encoded by the genes described in Figure 15 or Figure 16. In some embodiments, the antigens are selected from EPCAM, CD163, SPP1, or CD44.

[0076] In some embodiments, the cancer treatment comprises one or more monoclonal antibodies comprising an antigen binding region. In some embodiments, the antigen binding regions binds to an antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects. In some embodiments, the antigen binding regions bind to an antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject. In some embodiments, the antigen is selected from the antigens encoded by of the genes described in Figure 15 or Figure 16. In some embodiments, the antigen is selected from EPCAM, CD163, SPP1, or CD44.

[0077] In some embodiments, the cancer treatment comprises one or more chimeric antigen receptor T cell therapy (CAR-T).

[0078] In some embodiments, the cancer treatment comprises antibody-drug conjugates.

[0079] In some embodiments, the cancer is a solid cancer. In some embodiments, the solid cancer is colon, breast, ovarian, pancreatic, renal, esophageal, melanoma, lung, brain, prostate, oruveal melanoma. In some embodiments, the cancer is colorectal cancer. In some embodiments, the cancer is breast cancer. In some embodiments, the cancer is lung cancer. In some embodiments, the cancer is SCLC. In some embodiments, the lung cancer is NSCLC. In some embodiments, the one or more hybrid cells comprises a gene expression profile of acolon cell-type and the subject is administered a colon cancer treatment. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a colorectal cell-type and the subject is administered a colorectal cancer treatment.

[0080] In some embodiments, the colorectal cancer treatment comprises colectomy, chemotherapy, adjuvant or neoadjuvant chemotherapy. In some embodiments, the colorectal cancer treatment comprises 5 -Fluorouracil (5-FU), Capecitabine, Irinotecan, Oxaliplatin, Bevacizumab, FOLFOXIRI, Cetuximab, Aflibercept, Ramucirumab, Panitumumab, Anti- epidermal growth factor receptor (EGFR) antibody, anti-vascular endothelial growth factor (VEGF) antibody with first-line chemotherapy, Trifluridine-tipiracil, Regorafenib, Encorafenib with cetuximab (for patients with BRAF V600E mutations), or immunotherapy targeting checkpoints (Pembrolizumab monotherapy, Nivolumab monotherapy, or Nivolumab and ipilimumab). In some embodiments, the colorectal cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Colon cancer, Version 1.2024 (January 29, 2024), the content of which is hereby incorporated by reference in its entirety.

[0081] In some embodiments, the one or more hybrid cells comprises a gene expression profile of a breast cell-type and the subject is administered a breast cancer treatment.

[0082] In some embodiments, the breast cancer treatment comprises a lumpectomy or a mastectomy. In some embodiments, the breast cancer treatment comprises chemotherapy, radiation therapy, adjuvant therapy, or hormonal therapy. In some embodiments, the breast cancer treatment comprises Tamoxifen, Aromatase inhibitor (Al), Trastuzumab, Pertuzumab, Neratinib, or Olaparib. In some embodiments, the breast cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Breast Cancer, Version 2.2024 (March 11, 2024), the entire disclosure of which is incorporated herein by reference.

[0083] In some embodiments, one or more hybrid cells comprises a gene expression profile of a lung cell-type and the subject is administered a lung cancer treatment. In some embodiments, the lung cancer cell-type is a small cell lung cancer (SCLC), and the subject is administered a small cell lung cancer treatment. In some embodiments, the lung cancer celltype is non-small cell lung cancer (NSCLC), and the subject is administered an NSCLC treatment.

[0084] In some embodiments, the small lung cancer treatment comprises carboplatin and another drug selected from the group of etoposide, atezolizumab, durvalumab, and irinotecan. In some embodiments, the small lung cancer treatment comprises cisplatin and another drug selected from the group of durvalumab, etoposide, and irinotecan. In some embodiments, the small lung cancer treatment comprises topotecan, lurbinectedin, tarlatamab-dlle, paclitaxel, docetaxel, temozolomide, cyclophosphamide, doxorubicin, vincristine, vinorelbine, gemcitabine, nivolumab, pembrolizumab, and bendamustine. In some embodiments, the small lung cancer treatment comprises stereotactic radiotherapy, external beam radiation therapy, or chest radiation. In some embodiments, the small lung cancer treatment comprises a lobectomy. In some embodiments, the small lung cell cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Small Cell Lung Cancer, Version 4.2025 (January 13, 2025), the content of which is hereby incorporated by reference in its entirety.

[0085] In some embodiments, the NSCLC treatment comprises cisplatin and another drug selected from the group of pemetrexed, gemcitabine, docetaxel, vinorelbine, and etoposide. In some embodiments, the NSCLC treatment comprises carboplatin and another drug selected from the group of paclitaxel, gemcitabine, and pemetrexed. In some embodiments, the NSCLC treatment comprises Alectinib, Osimertinib, Atezolizumab, or Pembrolizumab. In some embodiments, the lung cancer treatment comprises a wedge resection, segmentectomy, lobectomy, sleeve lobectomy, pneumonectomy, thoracotomy or a thoracoscopy. In some embodiments, the NSCLC cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Non-Small Cell Lung Cancer, Version 3.2025 (January 14, 2025), the content of which is hereby incorporated by reference in its entirety.

[0086] The cancer treatment described herein can include any cancer treatment described in the International patent publication No. WO2023130073, the entire disclosure of which is herein incorporated by reference.

[0087] In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using scRNAseq. In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using RT-qPCR.

[0088] As described herein, the methods of treatment refer generally to obtaining a desired pharmacological and / or physiological effect. The effect may be prophylactic in terms of completely or partially preventing a disease or symptom thereof and / or may be therapeutic in terms of a partial or complete stabilization or cure for a disease and / or adverse effect attributable to the disease. Methods described herein covers any treatment of a disease in a subject, and includes: (a) preventing the disease or symptom from occurring in a subject which may be predisposed to the disease or symptom, may or may not be diagnosed as having it; (b) inhibiting the disease symptom, i.e., arresting its development; or (c) relieving the disease symptom, i.e., causing regression of the disease or symptom.

[0089] In some embodiments, the sample is biopsy tissue or blood. In some embodiments, the subject is a human.

[0090] A therapeutically effective amount of an agent or composition disclosed herein, for example, is one that is effective for preventing, ameliorating, treating or delaying the onset of a disease or condition.

[0091] Pharmaceutical compositions can be administered to any animal that can experience the beneficial effects of the agents of the invention. Such animals include humans and non-humans such as primates, pets and farm animals.

[0092] The present invention also comprises pharmaceutical compositions comprising the therapeutic agents described herein. Routes of administration and dosages of effective amounts of the pharmaceutical compositions comprising the agents are also disclosed. The agents of the present invention can be administered in combination with other pharmaceutical agents in a variety of protocols for effective treatment of disease.

[0093] Pharmaceutical compositions of the present invention are administered to a subject in a manner known in the art. The dosage administered will be dependent upon the age, health, and weight of the recipient, kind of concurrent treatment, if any, frequency of treatment, and the nature of the effect desired. One may administer the pharmaceutical compositions in a local rather than systemic manner, for example, via injection of directly into the desired target site, often in a depot or sustained release formulation. Furthermore, one may administer the composition in a targeted drug delivery system.

[0094] One of ordinary skill in the art will appreciate that a method of administering pharmaceutically effective amounts of pharmaceutical compositions to a patient in need thereof, can be determined empirically, or by standards currently recognized in the medical arts. The agents can be administered to a patient as pharmaceutical compositions in combination with one or more pharmaceutically acceptable excipients. It will be understood that, when administered to a human patient, the total daily usage of the agents of the pharmaceutical compositions of the present invention will be decided within the scope of sound medical judgment by the attending physician. The specific therapeutically effective dose level for any particular patient will depend upon a variety of factors: the type and degree of the cellular response to be achieved; activity of the specific agent or composition employed; the specific agents or composition employed; the age, body weight, general health, gender and diet of the patient; the time of administration, route of administration, and rate of excretion of the agent; the duration of the treatment; drugs used in combination or coincidental with the specific agent; and like factors well known in the medical arts. It is well within the skill of the art to start doses of the agents at levels lower than those required to achieve the desired therapeutic effect and to gradually increase the dosages until the desired effect is achieved.

[0095] Dosaging can also be administered in a patient-specific manner to provide a predetermined concentration of the agents in the blood, as determined by techniques accepted and routine in the art.Methods of diagnosis and prognosis

[0096] In certain aspects the subject matter described herein provides a method for diagnosing cancer in a subject, the method comprising: detecting in a sample one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

[0097] In certain aspects, the subject matter described herein provides a method for detecting cancer recurrence in a subject in need thereof, the method comprising: detecting in a sample one or more hybrid cells, wherein the one or more hybrid cells express one ore more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

[0098] In some embodiments, the one or more hybrid cells express at least two epithelial cell lineage markers and at least two monocyte or macrophage cell lineage markers. In some embodiments, the one or more epithelial cell lineage markers are selected from EPCAM, KRT8, KRT9 (CK9), KRT19 (CK19), or CK18; and the one or more monocyte or macrophage cell lineage markers are selected from CD163, CD14, CD68, FCER1G, TYROBP, or AIF1. In some embodiments, the epithelial cell lineage marker is EPCAM and the monocyte or macrophage cell lineage markers is macrophage cell lineage marker CD 163. In some embodiments, the at least two epithelial cell lineage markers are EPCAM and KRT8, and at least two monocyte or macrophage cell lineage markers are macrophage cell lineage markers CD163 and CD14.

[0099] In some embodiments, the one or more hybrid cells further express one or more blood lineage markers. In some embodiments, the blood lineage marker is PTPRC (CD45).

[0100] In some embodiments, one or more hybrid cells have a doublet score of <0.5. In some embodiments, one or more hybrid cells have a doublet score between 0.1 and 0.5.

[0101] In some embodiments, a cell type annotation of the one or more hybrid cells is a myeloid cell type. In some embodiments, the myeloid cell type is monocyte or macrophage. In some embodiments, a cell type annotation of the one or more hybrid cells is an epithelial cell type. In some embodiments, the cell type annotation is a graph-clustering and nonnegative matrix factorization (NMF) approach. In some embodiments, the NMF approach is a correlation consensus NMF (ccNMF) or NMF using Nesterov’s optimal gradient method (NeNMF). In some embodiments, the ccNMF is executed or embodied by the ccNMF algorithm (github.com / matanhofree / crc-immune-hubs / tree / main / code / dimReduction / ccNMF) described in Pelka et al

[0036] the contents of which is hereby incorporated by reference in its entirety.

[0102] In some embodiments, a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject. In some embodiments, the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared to signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject. In some embodiments, thegene expression pattern of the one or more hybrid cells comprises gene expression of one or more of the genes described in Figure 15 or Figure 16.

[0103] In some embodiments, the gene expression is expression pattern is for genes associated with EMT. The genes associated with EMT are part of the Hallmark gene set collection of Molecular Signatures Database that can be found at www.gsea- msigdb.org / gsea / msigdb / human / geneset / HALLMARK_EPITHELIAL_MESENCHYMAL_T RA.NSITION.html and as described by Liberzon et al., The Molecular Signatures Database (MSigDB) hallmark gene set collection, Cell Syst. 2015 Dec 23;l(6):417-425, the entire disclosures of which are herein incorporated by reference. In some embodiments, the one or more hybrid cells comprises a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non-cancerous cell of the subject. In various embodiments, expression level of genes associated with EMT of a subject may be increased by at least 1.5 times greater than gene expression level of said genes in a non-cancerous cell of the subject, or a gene expression level of said genes in a non- cancerous cell of one or more healthy subjects. Expression level of genes associated with EMT of a subject may be increased by at least 3 times greater than gene expression level of said genes in a non-cancerous cell of the subject, or a gene expression level of said genes in a non-cancerous cell of one or more healthy subjects. Expression level of genes associated with EMT of a subject may be increased by at least 5 times greater than gene expression level of said genes in a non-cancerous cell of the subject, or a gene expression level of said genes in a non-cancerous cell of one or more healthy subjects. Expression level of genes associated with EMT of a subject may be increased by at least 10 times greater than gene expression level of said genes in a non-cancerous cell of the subject, or a gene expression level of said genes in a non-cancerous cell of one or more healthy subjects.

[0104] In some embodiments, the gene expression is expression of (4)44, SPP1, or both CD44 and SPP1. In some embodiments, the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject. In various embodiments, CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 1.5 times greater than a CD44 expression level, SPP1 expression level, or both CD44expression level and SPP1 expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects. CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 3 times greater than a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject, or a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of one or more healthy subjects. CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 5 times greater than a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject, or a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of one or more healthy subjects. CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level of the hybrid cell may be increased by at least 10 times greater than a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non-cancerous cell of the subject, or a EMT gene expression level in a non-cancerous cell of one or more healthy subjects.

[0105] In some embodiments, the cancer is a solid cancer. In some embodiments, the solid cancer is colon, breast, ovarian, pancreatic, renal, esophageal, melanoma, lung, brain, prostate, or uveal melanoma. In some embodiments, the cancer is colorectal cancer. In some embodiments, the cancer is breast cancer. In some embodiments, the cancer is lung cancer. In some embodiments, the cancer is SCLC. In some embodiments, the cancer is NSCLC. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a colon cell-type. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a colorectal cell-type. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a breast cell-type. In some embodiments, the one or more hybrid cells comprises a gene expression profile of a lung cell-type.

[0106] In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using scRNAseq. In some embodiments, the expression level of one or more markers or genes in any of the preceding embodiments is determined using RT-qPCR.

[0107] In certain aspects, the subject matter described herein provides a method of determining a cancer prognosis in a subject in need thereof, wherein the prognosis depends on numbers and / or size of hybrid cells as described herein in the sample. In some embodiments, the subject is determined to have a poor prognosis if the subject has a high number of hybrid cells and / or large hybrid cells.

[0108] In some embodiments, the sample is biopsy tissue or blood. In some embodiments, the subject is a human.

[0109] In one embodiment, a biological sample comprises, a blood sample, serum, cells (including whole cells, cell fractions, cell extracts, and cultured cells or cell lines), tissues (including tissues obtained by biopsy), body fluids (e.g., urine, sputum, amniotic fluid, synovial fluid), or from media (from cultured cells or cell lines). In one embodiment, a biological sample comprises, liver cells. The methods of detecting or quantifying a molecule include, but are not limited to, amplification-based assays with (signal amplification) hybridization based assays and combination amplification-hybridization assays. For detecting and quantifying a molecule, an exemplary method is an immunoassay that utilizes an antibody or other binding agents that specifically bind to protein or epitope of such, for example, Western blot or ELISA assays. In some embodiments, the level of gene expression is determined using single-cell RNA, RT-qPCR or RNA-seq.EXAMPLES

[0110] Examples are provided below to facilitate a more complete understanding of the invention. The following examples illustrate the exemplary modes of making and practicing the invention. However, the scope of the invention is not limited to specific embodiments disclosed in these Examples, which are for purposes of illustration only, since alternative methods can be utilized to obtain similar results.EXAMPLE 1 — scRNAseq and high-throughput spatial analysis of tumor and normal microenvironment in solid tumors reveal a possible origin of circulating tumor hybrid cells.

[0111] This study explored a potentially significant player in cancer spread: "hybrid cells" (HCs). These cells combine features of epithelial cells (common in tissues) and immune cells (like macrophages). HCs were found in both healthy (NHC) and canceroustissues (THC), but much more so in tumors. The transcriptional states and functional pathways were distinct between NHC and THC, with THCs showing functional profiles that enables dissemination in circulation suggesting a role in cancer progression. Notably, these findings weren't limited to one type of cancer.

[0112] The study also developed a way to identify HCs in large datasets generated at single cell level. This paves the way for further research on their function and potential as targets for new cancer treatments.

[0113] Metastatic cancer is one of the leading causes of death in cancer patients worldwide. While circulating hybrid cells (CHCs) are implicated in metastatic spread, studies documenting their tissue origins remain sparse with a few limited candidate approaches using 1-2 markers. Utilizing high-throughput single-cell and spatial transcriptomics, hybrid cells (HCs) were identified in both normal and tumor biopsies, yet notably enriched within tumor tissue. These HCs co-expressed epithelial and macrophage markers, suggesting their hybrid nature. HCs closely clustered with monocytes and macrophages, while differential gene expression unveiled a distinct transcriptome. In situ spatial mapping further validated their presence across both tissue types, confirming their higher abundance in tumor. Notably, similar findings were observed in breast cancer, suggesting HCs as a generalizable phenomenon across cancer types.

[0114] This study establishes a framework for HC identification in large datasets, providing compelling evidence for their tissue residence and offering comprehensive transcriptomic characterization. Furthermore, it sheds light on potential differential functions and identifies pathways potentially governing their enhanced dissemination. The potential of THCs as novel therapeutic targets is highlighted.

[0115] Death from metastatic cancer remains one of the leading causes of mortality from cancers [1], The process of metastasis involves the tumor cells leaving the primary site and landing and establishing at distant sites [2, 3], Many different theories are proposed to explain the process of metastasis but a relatively understudied and one of the oldest theories suggests fusion of tumor cells to leucocytes including a monocyte or macrophage result in a hybrid cell (HC) that aid and abet in metastasis [2, 3], In 1908, Otto Aichel suggested that cancerous cells are able to fuse with leukocytes, creating malignant tumor hybrid cells (THCs) [4], While the concept of THC is old, the studies characterizing THC from patients are relativelynascent and are limited to a small number of investigators across the globe mainly due to limitations in THC detections technologies and isolation methods. Recent advances in circulating tumor cells (CTC) isolation and detection technologies identified a new type of cell along with CTC in circulation that is characterized by the presence of a hybrid proteome resembling tumor and immune cell lineages resulting in the resurgence of the concept proposed by Aichel [5-7], These hybrid cells were described using various nomenclature including circulating hybrid cells (CHC), tumor macrophage hybrid (TMH), tumor hybrid cells (THC), and cancer associated macrophage like (CAML, these are generally large and polyploid) cells [5-19],

[0116] Several studies have shown a significant relation with the presence, size, and or number of HCs in circulation to tumor stage, progression, and or survival [9-11, 13, 14, 19- 22], The process of hybrid cell formation between a tumor and an immune cell has been documented in cell culture and animal models [9, 19], While the source of these cells in circulation is believed to be the primary and or metastatic tumor and a few studies demonstrated their presence, mainly using immunohistochemistry, in the primary tumor using limited number of markers and candidate approaches [23-28], Due to lack of markers and a consensus in the definition of these cells, several different proteins including CD14, CD45, CD68, and or CD 163 to identify immune lineage and EPCAM and or pan-cytokeratins including CK8, CK9, and CK19, to identify epithelial lineage were used to identify and characterize these cells from circulation and in tumor biopsies [5-8, 18, 23-34], THCs were described in the circulation of patients with several different solid cancers including colon, breast, ovarian, pancreatic, renal, esophageal, and melanoma [1, 2, 4-8, 10, 12-15, 19, 20, 22, 30, 31, 33, 34], Recently, polyploid giant cells expressing CD61, Syncytin-1, telomerase, Runx2, and macrophage markers (CD1 lb, CD68, and CD163) were identified

[0035] ,

[0117] Recent advances in single cell technologies and their application in characterizing tumors and their microenvironment (TME) resulted in the explosion of datasets from various cancers that provides an opportunity for the identification and characterization of THC using these datasets. Analysis of scRNAseq data provides comprehensive information on THC including their cellular origin, frequency, immune partner, gene expression profiles, and an opportunity to correlate their presence with clinical data and outcome. In this study, using scRNAseq and spatial dataset from colon and breast cancer and normal tissues, evidence for the presence of HCs is provided and their transcriptome is comprehensively characterized.Material and Methods scRNAseq and spatial datasets

[0118] The colorectal cancer dataset

[0036] is downloaded from www.ncbi.nlm. nih.gov / geo / query / acc.cgi?acc=GSE178341 and the breast cancer dataset

[0037] is downloaded from singlecell.broadinstitute.org / single_cell / study / SCP1039 / a-single-cell- and-spatially-resolved-atlas-of-human-breast-cancers#study-download^

[0119] The colorectal dataset, initially containing 371,223 cells, was filtered following quality control procedures described in Pelka et al. (2023) to retain 370,115 cells. The breast cancer dataset consisted of 100,064 cells with provided barcodes, features, and matrix files. Both datasets included metadata for cell lineage and type.

[0120] The colorectal cancer and non-diseased colon Xenium spatial datasets were downloaded from www.10xgenomics.com / datasets / human-colon-preview-data-xenium- human-col on-gene-expression-panel- 1 -standard. These dataset were generated using the predesigned panel of 325 gene and add-on panel of 100 additional genes. The colon cancer dataset was generated on FFPE-pre served tissue obtained from a stage 2A adenocarcinoma.

[0121] The Xenium breast cancer dataset is downloaded from www.10xgenomics.com / datasets / xenium-ffpe-human-breast-with-custom-add-on-panel-l- standard. This dataset was generated using the pre-designed panel plus an add-on panel of 100 custom genes on 5 um section from an Infiltrating ductal carcinoma (IDC) tumor of a breast cancer patient. scRNAseq and spatial data processing

[0122] Both the single-cell RNA-seq and spatial in situ datasets were analyzed using Seurat v5.0.1

[0038] , For each dataset, a Seurat object was created and processed. Expression data was normalized and scaled using SCTransform (vl.6.0), which also identified highly variable features. Principal component analysis (PCA) was used to reduce dimensionality, and the top 10 principal components were fed into a graph-based clustering algorithm with resolution 0.8. For visualization, UMAP was used to project the data down to two dimensions. The scCustomize package (v2.0.1)

[0039] and Seurat’s built-in visualization tools facilitated data exploration through dimplots, ImageDimPlots, violin plots, and dotplots.Additionally, for spatial data, the cluster and group information were exported to annotate the data visualized in Xenium explorer (10X genomics).Doublet scoring analysis

[0123] To identify and remove doublets, the scds package (vl.18.0)

[0040] was leveraged. First, the Seurat object was converted to a SingleCellExperiment format using as.SingleCellExperiment. Three doublet scoring methods were then applied: co-expression based (cxds), binary classification based (beds), and a hybrid combining both approaches. Benchmarking revealed the hybrid approach outperformed cxds and beds individually, so its score was added to the Seurat object for subsequent doublet filtering.Hybrid cell identification

[0124] To mark hybrid cells, WhichCells function was used to mark cells if the expression EPCAM > 0, KRT8 > 0, CD 14 > 0, and CD 163 > 0. Additionally, the cells doublet score should be less than or equal 0.5 (mean+2 S.D.).Differential gene expression

[0125] To identify cluster-specific marker genes, differential expression analysis was employed. FindAllMarkers() was used to compare each cluster against all others, and for more focused comparisons, FindMarkers() for individual pairs of clusters or groups. Both functions were called with default parameters. For each identified marker gene, several key metrics were calculated: the p-value for significance, average fold change in expression between the cluster and the rest (avg logFC), the percentage of cells expressing the gene within the specific cluster (pct.l), the percentage expressing it in all other clusters (pct.2), and the adjusted p-value (p_val_adj). Among these, avg logFC was chosen as the primary indicator of cluster-specificity.

[0126] To visualize the differentially expressed genes, volcano plots were generated using the VolcaNoseR shiny app (github.com / JoachimGoedhart / VolcaNoseR). Before uploading the results of the FindMarkers function to VolcaNoseR, a column containing the negative log 10 of the p-values was added.Functional gene set enrichment analysis

[0127] Gene set enrichment analysis was performed using fgsea library in R. Genes were ranked according to their average log2fold change, and this ranking was used as the input to fgsea. 10 000 gene permutations were used to calculate statistical significance, and a false discovery corrected p-value of 0.05 was required for statistical significance of a gene set. The hallmark gene sets available from the Human MSigDB collection (www.gsea- msigdb.org / gsea / msigdb / human / collections.jsp) were used.Statistical analysis

[0128] Statistical analyses were performed using GraphPad Prism (vlO.1.2, GraphPad Software). Student’s t test was used for statistical analysis, and p-values < 0.05 were considered statistically significant.ResultsHybrid cells in colon cancer and normal biopsies are more frequently found in myeloid cell cluster

[0129] To identify hybrid cells (HC) in the tumor biopsies, a previously published single cell sequencing data of 370,115 high-quality cells from tumor (257,251 cells) and adjacent normal (112861 cells) tissue from patients with colon cancer was analyzed

[0036] , This dataset is obtained by sequencing single cells from 64 tumors from 62 patients and adjacent normal tissue from a subset of 36 patients. A Louvain algorithm utilizing K-nearest neighbor (KNN) graph-based distance data was used to cluster the cells resulting in 43 clusters which were then visualized and explored using non-linear dimensional reduction technique uniform manifold approximation and projection (UMAP) (Figure 5A).

[0130] To assign cell type identity to clusters, the cell type annotation provided in the original study was used as it was based on a robust two-step graph-clustering and nonnegative matrix factorization (NMF) approach. Three levels of cell annotation were provided: clTopLevel, clMidwayPr, and cl295vl 1 SubFull each with a higher degree of subclassification within the preceding level. The top-level annotation divided cells into 7 major cell types including myeloid, epithelial, T / natural killer [NK] / innate lymphoid cell [ILC] (TNKILC), plasma, B, mast, and stromal cells (Figure 5B).

[0131] To identify HC with both epithelial and macrophage characteristics, cells that express RNA for both the epithelial marker (EPCAM) and macrophage marker (CD 163) were marked. 2126 (0.57%) HC (Figure 1A) were identified. Alternatively, a stringent criterion was also used and HC were marked based on the presence of multiple tumor (EPCAM and KRT8) and macrophage (CD163 and CD14) markers and identified 778 (0.21%) HC (Figure IB). In both the cases, HC were predominantly found within the clusters 9, 12, 14, and 21 corresponding to myeloid cell types (Figure 1A, IB, 5A, and 5B). A small percentage of HC were also found in the same cluster as epithelial cells (Figure IB).Doublets analysis further refine hybrid cells identity

[0132] To identify doublets that are artifacts of droplet-based sequencing, a doublet score was calculated for all the top level clusters and removed cells with high doublet score of above 0.5 (mean+2 S.D.) (Figure 1C). As expected, the median doublet score for HC (0.29) was slightly higher compared to the rest of the top clusters but close to myeloid (0.27) and mast cell (0.28) clusters. After removing doublets, only 360,975 of the total 370,115 cells remained for further analysis. This also resulted in HC numbers reducing to 610 from 778.

[0133] For the rest of the analysis below, HC was defined as EPCAM+KRT8+CD163+CD14+ with a doublet score of <0.5. Using this definition, greater than 96% of HC were found with myeloid cells followed by around 4% with epithelial cell types. The HC represents 1.5 and 0.01 percent of myeloid and epithelial cells respectively. In addition o EPCAM, KRT8, CD 14, and CD 163, HC express RNA for other epithelial markers including cytokeratins KRT9 (CK9) and KRT19 (CK19) and blood lineage marker PTPRC (CD45) and monocyte / macrophage marker CD68 (Figure ID).Hybrid cells in colon cancer and normal biopsies are more frequently found with macrophage / monocyte cluster

[0134] Since most of the HC were found in myeloid cluster, this group of cells was focused on next to identify the exact identity of the myeloid cell types that HC clustered with. For this, ‘clMidWayPr’ and ‘cl295vl 1 subFull’ annotations provided in Pelka et al

[0036] were used. The mid-level annotation identified 19 cell types (Figure 5C) and the full annotation divided cells into 87 cell types (Figure 5D). Unsupervised clustering of 19 cell types, identified within the 7 top level cell types, using these 9 markers clustered hybrid with monocyte / macrophage cluster (Figure IE).

[0135] Cells from the myeloid cluster of tumor fraction were extracted, these cells were clustered in 10 myeloid cell types including, monocyte, macrophage-like, dendritic cells (DC) 1, DC2, DC2 C1Q+, DC IL22RA2, pDC, AS-DC, mregDC, and granulocyte (Figure 5E). The HC were clustered with monocytes and macrophages, but predominantly with macrophages (Figure IF) with expression of CD14 and CD163 very similar to macrophage (Figure 1G).Differential gene expression analysis revealed an extensive hybrid transcriptome

[0136] To identify the extent of transcriptome remodeling in HC from tumor (THC), the tumor fraction was extracted and compared the gene expression profile of THC with tumor epithelial cells (EPCAM+ & KRT8+) (Figure 1H and Figure 13) and THC with tumor associated macrophage / monocyte cells (CD 14+ & CD 163+) (Figure II and Figure 13). 3824 genes significantly (padj < 0.05) differentially expressed between THC and epithelial cells were found. Also, 5800 genes significantly (padj < 0.05) differentially expressed between THC and macrophage / monocyte cells were found (Figures 1H & II). Compared to the fusion partners the percentage and or expression of top 5 genes from each comparison was relatively low in the THC (Figure 1 J).The majority of the hybrid cells were found in tumor tissue compared to normal

[0137] To identify the proportion of HC within normal (NHC) and tumor tissue, the dataset was split into normal and tumor fractions. The percentage of HCs within tumor fraction was 0.21, which was 4.2 times more than within normal tissue (0.05) (Figure 2A). It is important to note that the myeloid cell infiltration was also more in tumor tissue compared to normal (Figure 2B & 2C). It was found that myeloid cells contribution was 14.8 percent of all cells within tumor tissue compared to only 2.4% in normal tissues (Figure 2B), but despite that, of the 610 HCs, 549 (90%) were from the tumor sections compared to only 61 (10%) from normal tissues (Figure 2D).

[0138] HCs were found in 60 of the 62 patient’s samples (Figure 6A & 2B). Their number varied between patients with some patients showing up to 1.8% of their tumor cells represented by HCs (Figure 6B). As noted above, on 36 patients, there were cells derived from both tumor and adjacent normal tissue which allowed us to compare the proportion of HC within the matching normal and tumor tissue in each patient. Except for one patient, in whom no HCs were found in both normal and tumor tissue, in the rest of the 35 patients,almost half had no HCs in normal tissue (Figure 2E). In patients where HCs were present in both normal and tumor tissues, HCs were consistently more in tumor compared to normal tissue (Figure 2E).THC shows a distinct gene expression profile

[0139] Differential gene expression analysis between NHC and THC identified 33 genes that were significantly differentially (padj < 0.05) expressed (Figure 15, Figure 2F & 2G). The percentage of HCs and fold change varied for each gene. EPCAM, KRT8, and KRT18 expressions were lower, but not significantly different, in normal cells whereas CD14 and CD163 showed similar expressions (Figure 2H). CD44 was highly expressed in THC (Figure 7 A) whereas SPP1 was uniquely present in THC and not in NHC (Figure 7B).

[0140] Gene set enrichment analysis using hallmark pathways identified 19 pathways significantly (padj <0.05) enriched in THC compared NHC (Figure 21 and Figure 8a). There was upregulation of inflammatory pathways and other notable pathways include upregulation of KRAS signaling and EMT pathways (Figure 2 J), which were among the top 10 enriched pathways (Figure 8A).THC number within tumor and their relation to clinical features

[0141] Next the study focused on THC. For this, the tumor cells were subset by subtracting all the cells including the HCs found in normal tissue. THCs were found in the tumor of all but two patients with their percentage ranging from 0.02 to 1.8% of all tumor cells. The absolute numbers range from 1 to 83 with a median of 6 cells.

[0142] Next, the relation of THC with tumor, node and metastasis (TNM) staging system, histology grade and outcome including survival was investigated. An increase in the percentage of THC with T stage with higher stages having twice as many THC compared to early stages was noted (Figure 2K and 2L). The median number of THCs were higher in higher T stages, but these differences were not statistically significant. No significant differences were observed with node status in this dataset (data not shown). The median number of THCs were slightly higher in high histologic grade, but these differences were not statistically significant (data not shown).In situ high resolution spatial mapping confirms the presence of THC in tumor sections

[0143] To confirm presence of HCs in tissue sections, publicly available datasets of normal colon and tumor (stage 2A colon adenocarcinoma) sections that were analyzed using a recently developed Xenium high resolution in situ targeted panel of 425 genes including EPCAM, KRT8, CD14 and C163 were analyzed. A KNN-graph based clustering identified 30 and 37 clusters respectively in normal and tumor tissues (Figure 9A and 9B). To identify HCs, HCs were defined as cells that co-expressed all four markers as were defined in scRNAseq analysis. 2026 (0.73%) and 6,958 (1.19%) HCs were identified in normal (Figure 3A, 3B) and tumor sections respectively (Figure 3C & 3D), a much larger proportion compared to scRNAseq (Figure 2B).

[0144] Similar to scRNAseq data, the gene expression profile of THCs was different from NHC (Figure 3E) with at least three markers low in normal compared to tumor (Figure 3F). The majority of genes that were either low or high in tumor from scRNAseq were also similarly low or high respectively in spatial tumor cells (Figure 16 and Figure 14). EPCAM and KRT8 were abundantly expressed in tumor epithelial tissue with scattered expression of CD14 and CD163 in monocyte / macrophages (Figure 3G). The HCs were scattered throughout the tumor tissue (Figure 3D & Figure 10A-C).THC are found in scRNAseq and in situ spatial data from breast cancer

[0145] To identify THCs in the tumor biopsies of other cancers, a previously published scRNAseq data of 100,064 cells from tumor tissue of patients with breast cancer

[0037] was analyzed. This dataset is obtained by sequencing single cells from 26 patients. 41 clusters that were distributed in nine major cell types (Figure 11) and 29 minor cell types (Figure 4A) including a myeloid cluster with four myeloid cell types including monocyte and macrophage (Figure 4B) were identified.

[0146] Using the definition as set above, that is expression of EPCAM, KRT8, CD14, and CD163 and doublet score of less than 0.5, 93 THCs (Figure 4B) were identified. Similar to colon cancer, THCs were predominantly found within the myeloid cell types, more closely to monocytes and macrophages (Figure 4B). In addition to EPCAM, KRT8, CD14, and CD163, THCs express RNA for other epithelial markers including cytokeratins KRT9 (CK9) andKRT19 (CK19) and blood lineage marker PTPRC (CD45) and monocyte / macrophage marker CD68 (Figure 4C).

[0147] The presence of THCs in tissue sections of breast cancer analyzed using Xenium in situ targeted panel of 380 genes was further confirmed. A KNN-graph based clustering identified 29 clusters (Figure 9C). 967 (0.16%) THCs were identified in this tumor section (Figure 4D & 4E, and Figure 10C).

[0148] In this study publicly available datasets generated using two independent techniques from multiple cancers were used to identify hybrid cells with characteristics of CHC using single cell and spatial transcriptomic datasets. HCs were found both in normal and tumor tissues of cancer patients, but a much higher number in the tumor area. Cell fusions have been described in normal tissues and particularly bone marrow derived cells are known to fuse with damaged or corrupted cells

[0041] , The critical finding here is that the transcriptomes of the two differ markedly with THC showing higher KRAS and P53 signaling likely co-opted from tumor component (Figure 8A-B). TP53 and KRAS mutations are common in colorectal cancer patients

[0042] , THCs also show a higher EMT gene expression compatible with the notion that fusion with a macrophage allows tumor cells to escape their tissue of origin. CD44 expression was higher in THC compared to NHC (Figure 7A) consistent with previous observations

[0020] while SPP1 was expressed exclusively in THCs with a subset having very high expression (Figure 7B). SPP1+ macrophages have been implicated progression and poor outcome in colorectal cancer patients

[0043] , CD44 and SPP1 are potential targets for eliminating THCs.

[0149] NHCs could be normal stressed cells that macrophages are engulfing as a part of normal cellular homeostasis and inflammatory status, that is why their gene expression profile is distinct from THC. It has been previously proposed that fusion and or phagocytic events within normal tissue could be an early “seed” of cancer origin and not just an early seed of metastasis [44-47],

[0150] While the presence of THCs is clearly demonstrated in tumors, the mechanism of origin remains unclear with the dominant hypothesis being that the cause is cell fusion [9, 15, 47-58], Previous studies suggested a role of syncytins in the fusion of tumor and immune cells. In this dataset, a small fraction of macrophage / monocytes was observed, as well as epithelial cells expressing both syncytins with syncytin-1 is expressed in more single cells butnot in hybrid cells (Figure 12). Without being bound by theory, syncytin expression may be required for fusion but either gets diluted or is too low to be detected using scRNAseq which is known to be biased towards high expressing genes or the expression is turned off once the fusion process is complete and hence not detectable in hybrids.

[0151] It is possible that fusion events are higher in monocytes compared to macrophages which immediately suggests that the origin of a macrophage-tissue cell hybrid may be a result of phagocytosis and monocyte-tissue cell hybrid the result of fusion. A recent study showed that tumor cells with low CD47 expression induce delayed phagocytosis resulting in incomplete digestion and subsequent THC formation

[0059] , As noted above, in our analysis comparing THC with NHC, increased KRAS and TP53 signaling was found in THC. It is likely that tumor cells fuse with each other and with cell types other than monocyte / macrophage lineage but their presence in circulation is not well documented hence such a fusion may not provide an escape mechanism from tumor tissues but likely provide resistance to therapies by acquiring polyploidy [60-63],

[0152] There are multiple challenges in defining, identifying, and transcriptionally characterizing HC from high-throughput datasets. The first challenge was in defining the HC and a stringent criterion was used to mark HC. Cells were marked as HC only when (any expression level) of all four transcripts from genes EPCAM, KRT8, CD14, and CD163 was detected. CK8 is highly expressed in epithelial cells including solid tumor cells and CD 163 in macrophage lineages. EPCAM and CD14 were also incldued which also mark epithelial and monocyte / macrophage lineages respectively. These markers were selected to be consistent with the markers used to characterize CHCs, but it is likely that other markers including FCER1G, TYROBP, AIF1 that are abundantly expressed in macrophage / monocyte lineages can also be used to mark these cells. The hybrid nature of the HC identified using the above four markers was further strengthened by the fact that these cells also showed expression of other epithelial (CK18, CK19), macrophage (CD68), and pan leukocyte (CD45) markers (Figure IE) even though they were not used to mark these cells. It is further shown that the gene expression pattern of THCs was different from epithelial and macrophage / monocyte cells with more than a thousand genes differentially expressed suggesting the occurrence of the four markers used to mark these cells is not a stochastic process but involve significant transcriptome remodeling.

[0153] The second challenge was to distinguish HC from doublets that are frequently observed in scRNAseq datasets. There are several different algorithms developed to remove these cells from further analysis to allow “single” cell analysis. It was reasoned that if the HCs were doublets of monocyte / macrophage and epithelial cells co-segregating together during droplet formation, then they would have either formed a distinct cluster due to doubling of the transcriptome or if they didn’t form a distinct cluster, they would have had the same likelihood of being in epithelial and myeloid cluster thus half of the HC should have clustered with monocyte / macrophage cluster and the other half with epithelial cluster but majority of the HCs found did not form a distinct cluster but were spread across myeloid and epithelial cells, with a >75% in myeloid than epithelial cluster (Figure IF). Within the myeloid cluster, the HCs clustered with macrophage than monocytes even though CD14 that predominantly marks monocytes was used. Also, a high doublet score for HC was not found and the doublet score was similar to myeloid cell cluster (Figure 1C). But a bimodal distribution in the HC score was found with a small fraction of cells with high doublet score likely representing true doublets. A highly stringent criteria was used to remove doublets: a score of above 0.5 (mean+2 S.D.). A typical droplet scRNAseq has 0.1% doublet rate, but the criteria used removed 2.5% of all cells from the analysis including 21% of the HCs detected without filtering for doublets. Thus, HCs identified herein are confidently identified as true hybrids and not artifacts. Further, using Xenium in situ spatial dataset, it was shown that the signal for the four makers were limited to single cells. The percentage of HCs identified using Xenium in situ were more than scRNAseq. It is likely that the number of HCs in scRNAseq data is underestimated due to low depth of sequencing which is typical of most scRNAseq platforms which will likely eliminate cells that express low levels of any one of the four markers used to define HC but also because the sensitivity of in situ spatial methods are high. Also, as noted above, a rigid criterion was used for marking these cells when all four markers are present, but it is likely that cells that lack expression of any one of these four markers are not marked as hybrid. As discussed above, the use of doublet estimating methods might overcompensate and likely eliminate true HCs. Indeed, the HC estimates were reduced from 710 to 668. The use of doublet detection and elimination of doublet cells is double edged sword for HC detection in scRNAseq, while not removing them may result in true doublets contaminating the analysis but removing them may also remove some of the true hybrid because HC cells by definition have a transcriptome of cells of two different lineages and majority of doublet detection programs rely on this fact. The predominant mechanism of HC formation is believed to be due fusion of monocyte / macrophages with cancer epithelial cellsor via a “failed” phagocytosis attempt, thus removing cells by scoring for doublets will remove these cells from analysis.

[0154] Several high plex in situ technologies, including Xenium In Situ (lOx Genomics), CosMx (NanoString), MERSCOPE (Vizgen), and Molecular Cartography (Resolve) that offer single cells resolution to spatial methods are currently in rapid development and several of these have recently been commercialized. A dataset was used that was generated using a preliminary version of the Xenium Human Colon Gene Expression Panel, but other versions may have more than 425 genes, increasing the number of genes that can be simultaneously analyzed. An increase in data at single cell resolution in space that will enable further HC characterization including their location within tumors with respect to tumor geography, cellcell interaction, and communication etc.

[0155] The in situ spatial technology including the software used to analyze the data is under rapid development and it is likely that improved methods may refine the HC detection.

[0156] Here, a framework for identifying hybrid cells using high throughput scRNAseq and in situ spatial data from cancer and normal tissues is described using colon cancer as a model, but it is shown that hybrid cells are present in other cancers. Simultaneous characterization of THC from tumors and circulation of the same patient can be used to gain a better understanding of their role in tumor biology and to develop effective methods for targeting THCs.

[0157] This study has established a robust framework for identifying hybrid cells in tumor biopsies analyzed using high-throughput approaches. This approach provides compelling evidence for the presence and frequency of HCs that reside within tissues and offers a comprehensive characterization of their transcription activity and functional state. Importantly, the study pinpoints potential differences in HC between normal and tumor and identifies pathways that may contribute to their increased dissemination throughout the body. These findings raise the exciting possibility of targeting HCs for novel cancer therapies.Data Availability Statement

[0158] All the data are publicly available. The colorectal cancer dataset

[0035] is downloaded from www.ncbi.nlm. nih.gov / geo / query / acc. cgi?acc=GSE178341 and the breast cancer dataset

[0036] is downloaded fromsinglecell.broadinstitute.org / single_cell / study / SCP1039 / a-single-cell-and-spatially-resolved- atlas-of-human-breast-cancers#study-download.

[0159] The colorectal cancer and non-diseased colon Xenium spatial datasets were downloaded from www.10xgenomics.com / datasets / human-colon-preview-data-xenium- human-col on-gene-expression-panel- 1 -standard. The Xenium breast cancer dataset is downloaded from www.10xgenomics.com / datasets / xenium-ffpe-human-breast-with-custom- add-on-panel - 1 - standard .

[0160] References for Example 11. Mehlen, P. and A. Puisieux, Metastasis: a question of life or death. Nat Rev Cancer, 2006. 6(6): p. 449-58.2. Castaneda, M., et al., Mechanisms of cancer metastasis. Semin Cancer Biol, 2022. 87: p. 17-31.3. Fares, J., et al., Molecular principles of metastasis: a hallmark of cancer revisited. Signal Transduct Target Ther, 2020. 5(1): p. 28.4. Aichel, O., Eine neue Hypothese liber Ursachen und Wesen bosartiger Geschwillste . 1908: Universo.5. Adams, D.L., et al., Circulating Cancer-AssociatedMacrophage-Like Cells Differentiate Malignant Breast Cancer and Benign Breast Conditions. Cancer Epidemiol Biomarkers Prev, 2016. 25(7): p. 1037-42.6. Adams, D.L., et al., Circulating giant macrophages as a potential biomarker of solid tumors. Proc Natl Acad Sci U S A, 2014. 111(9): p. 3514-9.7. Augustyn, A., et al., Giant Circulating Cancer-Associated Macrophage-Like Cells Are Associated With Disease Recurrence and Survival in Non-Small-Cell Lung Cancer Treated With Chemoradiation and Atezolizumab. Clin Lung Cancer, 2021. 22(3): p. e451-e465.8. Gironda, D.J., et al., Cancer associated macrophage-like cells and prognosis of esophageal cancer after chemoradiation therapy. J Transl Med, 2020. 18(1): p. 413.9. Gast, C.E., et al., Cell fusion potentiates tumor heterogeneity and reveals circulating hybrid cells that correlate with stage and survival. Sci Adv, 2018. 4(9): p. eaat7828.Sutton, T.L., et al., Circulating Cells with Macrophage-like Characteristics in Cancer: The Importance of Circulating Neoplastic-Immune Hybrid Cells in Cancer. Cancers (Basel), 2022. 14(16). Manjunath, Y., et al., Circulating Giant Tumor-Macrophage Fusion Cells Are Independent Prognosticators in Patients WithNSCLC. J Thorac Oncol, 2020. 15(9): p. 1460-1471. Sutton, T.L., B.S. Walker, and M.H. Wong, Circulating Hybrid Cells Join the Fray of Circulating Cellular Biomarkers. Cell Mol Gastroenterol Hepatol, 2019. 8(4): p. 595- 607. Henn, T.E., et al., Circulating hybrid cells predict presence of occult nodal metastases in oral cavity carcinoma. Head Neck, 2021. Walker, B.S., et al., Circulating Hybrid Cells: A Novel Liquid Biomarker of Treatment Response in Gastrointestinal Cancers. Ann Surg Oncol, 2021. 28(13): p. 8567-8578. Ozel, C., et al., Hybrid cells derived from breast epithelial cell / breast cancer cell fusion events show a differential RAF-AKT crosstalk. Cell Commun Signal, 2012. 10(1): p. 10. Tosun, S., et al., Hybrid Cells Derived from Human Breast Cancer Cells and Human Breast Epithelial Cells Exhibit Differential TLR4 and TLR9 Signaling. Int J Mol Sci, 2016. 17(5). Gauck, D., et al., Hybrid clone cells derived from human breast epithelial cells and human breast cancer cells exhibit properties of cancer stem / initiating cells. BMC Cancer, 2017. 17(1): p. 515. Clawson, G.A., et al., Macrophage-tumor cell fusions from peripheral blood of melanoma patients. PLoS One, 2015. 10(8): p. e0134320. Aguirre, L.A., et al., Tumor stem cells fuse with monocytes to form highly invasive tumor-hybrid cells. Oncoimmunology, 2020. 9(1): p. 1773204. Dietz, M.S., et al., Relevance of circulating hybrid cells as a non-invasive biomarker for myriad solid tumors. Sci Rep, 2021. 11(1): p. 13630.Manjunath, Y., et al., Tumor -Cell-Macrophage Fusion Cells as Liquid Biomarkers and Tumor Enhancers in Cancer. Int J Mol Sci, 2020. 21(5). Manjunath, Y., et al., PD-L1 expression with epithelial mesenchymal transition of circulating tumor cells is associated with poor survival in curatively resected nonsmall cell lung cancer. Cancers, 2019. 11(6). Shabo, I., et al., Macrophage traits in cancer cells are induced by macrophage-cancer cell fusion and cannot be explained by cellular interaction. BMC Cancer, 2015. 15: p. 922. Shabo, I., et al., Macrophage Infdtration in Tumor Stroma is Related to Tumor Cell Expression ofCD163 in Colorectal Cancer. Cancer Microenviron, 2014. 7(1-2): p. 61-9. Shabo, I., et al., Breast cancer expression of DAP 12 is associated with skeletal and liver metastases and poor survival. Clin Breast Cancer, 2013. 13(5): p. 371-7. Shabo, I., et al., Expression of the macrophage antigen CD163 in rectal cancer cells is associated with early local recurrence and reduced survival time. Int J Cancer, 2009. 125(8): p. 1826-31. Shabo, I., et al., Breast cancer expression of CD163, a macrophage scavenger receptor, is related to early distant recurrence and reduced patient survival. Int J Cancer, 2008. 123(4): p. 780-6. Shabo, I. and J. Svanvik, Expression of macrophage antigens by tumor cells. Adv Exp Med Biol, 2011. 714: p. 141-50. Clawson, G. A., et al., Circulating tumor cells in melanoma patients. PLoS ONE, 2012. 7(7). Clawson, G.A., et al., "Stealth dissemination" of macrophage-tumor cell fusions cultured from blood of patients with pancreatic ductal adenocarcinoma. PLoS ONE, 2017. 12(9). Kaifi, J.T., et al., Circulating tumor cell isolation during resection of colorectal cancer lung and liver metastases: A prospective trial with different detection techniques. Cancer Biology and Therapy, 2015. 16(5): p. 699-708.Adams, D.L., et al., Sequential Tracking of PD-L1 Expression and RAD50 Induction in Circulating Tumor and Stromal Cells of Lung Cancer Patients Undergoing Radiotherapy. Clin Cancer Res, 2017. 23(19): p. 5948-5958. Adams, D.L., et al., Cytometric characterization of Circulating Tumor Cells Captured by microfdtration and their correlation to the cellsearch® CTC test. Cytometry Part A, 2015. 87(2): p. 137-144. Tang, C.M., et al., Blood-based biopsies-clinical utility beyond circulating tumor cells. Cytometry A, 2018. 93(12): p. 1246-1250. Ali, A.M., et al., Circulating cancer giant cells with unique characteristics frequently found in patients with myelodysplastic syndromes (MDS). Med Oncol, 2023. 40(7): p. 204. Pelka, K., et al., Spatially organized multicellular immune hubs in human colorectal cancer. Cell, 2021. 184(18): p. 4734-4752 e20. Wu, S.Z., et al., A single-cell and spatially resolved atlas of human breast cancers. Nat Genet, 2021. 53(9): p. 1334-1347. Hao, Y., et al., Dictionary learning for integrative, multimodal and scalable singlecell analysis. Nat Biotechnol, 2023. Marsh, S.E., scCustomize: Custom Visualizations & Functions for Streamlined Analyses of Single Cell Sequencing. 2021. Bais, A.S. and D. Kostka, scds: computational annotation of doublets in single-cell RNA sequencing data. Bioinformatics, 2020. 36(4): p. 1150-1158. Rizvi, A.Z., et al., Bone marrow -derived cells fuse with normal and transformed intestinal stem cells. Proc Natl Acad Sci U S A, 2006. 103(16): p. 6321-5. Markowitz, S.D. and M.M. Bertagnolli, Molecular origins of cancer: Molecular basis of colorectal cancer. N Engl J Med, 2009. 361(25): p. 2449-60. Qi, J., et al., Single-cell and spatial analysis reveal interaction ofFAP(+) fibroblasts and SPP1(+) macrophages in colorectal cancer. Nat Commun, 2022. 13(1): p. 1742. Bjerkvig, R., et al., Opinion: the origin of the cancer stem cell: current controversies and new insights. Nat Rev Cancer, 2005. 5(11): p. 899-904.Dittmar, T., et al., Recurrence cancer stem cells— made by cell fusion? Med Hypotheses, 2009. 73(4): p. 542-7. Zeng, C., et al., CD34(+) Liver Cancer Stem Cells Were Formed by Fusion of Hepatobiliary Stem / Progenitor Cells with Hematopoietic Precursor-Derived Myeloid Intermediates. Stem Cells Dev, 2015. 24(21): p. 2467-78. Hass, R., J. von der Ohe, and H. Ungefroren, Potential Role ofMSC / Cancer Cell Fusion and EMT for Breast Cancer Stem Cell Formation. Cancers (Basel), 2019. n(io). Hass, R., J. von der Ohe, and T. Dittmar, Cancer Cell Fusion and Post-Hybrid Selection Process (PHSP). Cancers (Basel), 2021. 13(18). Laberge, G.S., et al., Leukocyte (-) Cancer Cell Fusion-Genesis of a Deadly Journey. Cells, 2019. 8(2). Lazova, R., A. Chakraborty, and J.M. Pawelek, Leukocyte-cancer cell fusion: initiator of the warburg effect in malignancy? Adv Exp Med Biol, 2011. 714: p. 151-72. Pawelek, J., et al., Co-opting macrophage traits in cancer progression: a consequence of tumor cell fusion? Contrib Microbiol, 2006. 13: p. 138-155. Pawelek, J.M., Tumour-cell fusion as a source of myeloid traits in cancer. Lancet Oncol, 2005. 6(12): p. 988-93. Pawelek, J.M., Cancer-cell fusion with migratory bone-marrow -derived cells as an explanation for metastasis: new therapeutic paradigms. Future Oncol, 2008. 4(4): p. 449-52. Shabo, I., et al., Roles of cell fusion, hybridization and polyploid cell formation in cancer metastasis. World J Clin Oncol, 2020. 11(3): p. 121-135. Sieler, M., J. Weiler, and T. Dittmar, Cell-Cell Fusion and the Roads to Novel Properties of Tumor Hybrid Cells. Cells, 2021. 10(6). Sutton, T.L., B.S. Walker, and M.H. Wong, Digesting the Importance of Cell Fusion in the Intestine. Cell Mol Gastroenterol Hepatol, 2021. 11(1): p. 299-302. Wang, H.F., et al., Cell fusion in cancer hallmarks: Current research status and future indications. Oncol Lett, 2021. 22(1): p. 530.58. Weiler, J. and T. Dittmar, Cell Fusion in Human Cancer: The Dark Matter Hypothesis. Cells, 2019. 8(2).59. Chou, C.W., et al., Phagocytosis-initiated tumor hybrid cells acquire a c-Myc- mediated quasi-polarization state for immunoevasion and distant dissemination. Nat Commun, 2023. 14(1): p. 6569.60. Pienta, K. J., et al., Cancer recurrence and lethality are enabled by enhanced survival and reversible cell cycle arrest of polyaneuploid cells. Proc Natl Acad Sci U S A, 2021. 118(7).61. Pienta, K.J., et al., Poly-aneuploid cancer cells promote evolvability, generating lethal cancer. Evol Appl, 2020. 13(7): p. 1626-1634.62. Amend, S.R., et al., Polyploid giant cancer cells: Unrecognized actuators of tumorigenesis, metastasis, and resistance. Prostate, 2019. 79(13): p. 1489-1497.63. Liu, J., J. Erenpreisa, and E. Sikora, Polyploid giant cancer cells: An emerging new field of cancer biology. Semin Cancer Biol, 2022. 81: p. 1-4.Example 2: scRNAsequencing of Lung Cancer Reveals Presence of Tumor Hybrid Cells

[0161] To identify THCs in the lung cancer, we analyzed a previously published scRNAseq data of 503,979 cells from tumor tissue of patients with non-small cell lung cancer (NSCLC) [1], This dataset is obtained by sequencing single cells from 25 treatment-naive patients with adenocarcinoma and squamous-cell carcinoma. These 503,979 cells were classified in 44 cell types (Figure 17) including a myeloid, lymphoid, and epithelial cells (Figure 17).

[0162] Using the definition as set above, that is expression of EPCAM, KRT8, CD14, and CD163 and doublet score of less than 0.5, we identified 3,820 THCs (Figure 18). Similar to colon cancer and breast cancers, THCs were predominantly clustered with epithelial and myeloid cell types (Figure 19). In addition to EPCAM, KRT8, CD14, and CD163, THCs express RNA for other epithelial markers including cytokeratins KRT9 (CK9) and KRT19 (CK19) and blood lineage marker PTPRC (CD45) and monocyte / macrophage marker CD68 (Figure 19).

[0163] References for Example 2

[0164] 1. De Zuani, M., et al., Single-cell and spatial transcriptomics analysis of nonsmall cell lung cancer. Nat Commun, 2024. 15(1): p. 4388.

Claims

CLAIMSWhat is claimed is:

1. A method of treating or preventing cancer in a subject in need thereof, the method comprising administering to the subject a cancer treatment, wherein a sample from the subject comprises one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

2. The method of claim 1, wherein the one or more hybrid cells express at least two epithelial cell lineage markers and at least two monocyte or macrophage cell lineage markers.

3. The method of claims 1-2, wherein the one or more epithelial cell lineage markers are selected from EPCAM, KRT8, KRT9 (CK9), KRT19 (CK19), or CK18; and the one or more monocyte or macrophage cell lineage markers are selected from CD163, CD14, CD68, FCER1G, TYROBP, or AIF1.

4. The method of claim 1, wherein the epithelial cell lineage marker is EPCAM and the monocyte or macrophage cell lineage markers is macrophage cell lineage marker CD163.

5. The method of claim 2, wherein the at least two epithelial cell lineage markers are EPCAM and KRT8, and at least two monocyte or macrophage cell lineage markers are macrophage cell lineage markers CD163 and CD14.

6. The method of claims 1-5, wherein the one or more hybrid cells further express one or moreblood lineage markers.

7. The method of claim 6, wherein the blood lineage marker is PTPRC (CD45).

8. The method of claims 1-7, wherein one or more hybrid cells have a doublet score of <0.5.

9. The method of claim 8, wherein one or more hybrid cells have a doublet score between 0.1 and 0.5.

10. The method of claims 1-9, wherein a cell type annotation of the one or more hybrid cells is a myeloid cell type.

11. The method of claim 10, wherein the myeloid cell type is monocyte or macrophage.

12. The method of claims 1-9, wherein a cell type annotation of the one or more hybrid cells is an epithelial cell type.

13. The method of claims 10-12, wherein the cell type annotation is a graph-clustering and non-negative matrix factorization (NMF) approach.

14. The method of claims 1-13, wherein a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subjects, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects.

15. The method of claims 1-14, wherein the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared to signaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject, or signaling levels of KRAS, TP53, or both in a transcriptome of a non- cancerous cell of one or more healthy subjects.

16. The method of claim 14, wherein the gene expression pattern comprises gene expression of one or more of the genes described in Figure 15 or Figure 16.

17. The method of claim 14, wherein the gene expression pattern is for genes associated with epithelial-to-mesenchymal transition (EMT).

18. The method of claim 17, wherein the genes associated with EMT are part of the Hallmark gene set collection of Molecular Signatures Database available from the Human MSigDB collection.

19. The method of claim 18, wherein the one or more hybrid cells comprise a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non-cancerous cell of the subject, or a gene expression level of said genes in a non-cancerous cell of one or more healthy subjects.

20. The method of claim 16, wherein the gene expression is expression of CD44. SPP1, or both CD44 and SPP1.

21. The method of claim 20, wherein the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non- cancerous cell of the subject.

22. The method of claims 1-21, wherein the cancer treatment comprises one or more bispecific molecules comprising at least two antigen binding regions, including any cancer treatment described in international patent publication No. WO202313007353.

23. The method of claim 22, wherein the first and second antigen binding regions bind , respectively, a first and second antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non- cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects.

24. The method of claim 23, wherein the antigens are selected from the antigens encoded by the genes described in Figure 15 or Figure 16.

25. The method of claim 22-24, wherein the antigens are selected from EPC AM, CD 163, SPP1, or CD44.

26. The method of claims 1-21, wherein the cancer treatment comprises one or more monoclonal antibodies comprising an antigen binding region.

27. The method of claim 26, wherein the antigen binding regions binds to an antigen encoded by a gene of the hybrid cell of the cancer with a different expression pattern compared to a gene expression pattern of a non-cancerous epithelial cell of the subject, a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject, a gene expression pattern of a non-cancerous epithelial cell of one or more healthy subject, or a gene expression pattern of a non-cancerous macrophage or monocyte cell of one or more healthy subjects.

28. The method of claim 26 or 27, wherein the antigen is selected from the antigens encoded by of the genes described in Figure 15 or Figure 16.

29. The method of claim 26-28, wherein the antigen is selected from EPCAM, CD163, SPP1, or CD44.

30. The method of claims 1-21, wherein the cancer treatment comprises one or more chimeric antigen receptor T cell therapy (CAR-T).

31. The method of claims 1-21, wherein the cancer treatment comprises one or more antibody-drug conjugates.

32. The method of claims 1-31, wherein the cancer is a solid cancer.

33. The method of claim 32 wherein the solid cancer is colon, breast, ovarian, pancreatic, renal, esophageal, melanoma, lung, brain, prostate, or uveal melanoma.

34. The method of claim 33, wherein the cancer is colorectal cancer.

35. The method of claim 33, wherein the cancer is breast cancer.

36. The method of claim 33, wherein the cancer is lung cancer.

37. The method of claim 36, wherein the lung cancer is non-small cell lung cancer (NSCLC).

38. The method of claims 1-21, wherein the one or more hybrid cells comprises a gene expression profile of a lung cancer cell-type and the subject is administered a lung cancer treatment.

39. The method of claim 38, wherein the lung cancer cell-type is NSCLC, and the subject is administered a NSCLC treatment.

40. The method of claim 39 wherein the NSCLC treatment comprises cisplatin and another drug selected from the group of pemetrexed, gemcitabine, docetaxel, vinorelbine, or etoposide.

41. The method of claim 39 wherein the NSCLC treatment comprises carboplatin and another drug selected from the group of paclitaxel, gemcitabine, or pemetrexed.

42. The method of claim 39 wherein the NSCLC treatment comprises Alectinib, Osimertinib, Atezolizumab, or Pembrolizumab.

43. The methods of claims 38, wherein the lung cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Non-Small Cell Lung Cancer, Version 3.2025 (January 14, 2025).

44. The method of claim 39, wherein the NSCLC treatment comprises a wedge resection, segmentectomy, lobectomy, sleeve lobectomy, pneumonectomy, thoracotomy or a thoracoscopy.

45. The method of claims 1-21, wherein the one or more hybrid cells comprises a gene expression profile of a colon cell-type and the subject is administered a colon cancer treatment.

46. The method of claims 1-21, wherein the one or more hybrid cells comprises a gene expression profile of a colorectal cell-type and the subject is administered a colorectal cancer treatment.

47. The method of claims 34, 45-46, wherein the colorectal cancer treatment comprises colectomy, chemotherapy, adjuvant or neoadjuvant chemotherapy.

48. The method of claims 34, 45-46, wherein the colorectal cancer treatment comprises 5- Fluorouracil (5-FU), Capecitabine, Irinotecan, Oxaliplatin, Bevacizumab, FOLFOXIRI, Cetuximab, Aflibercept, Ramucirumab, Panitumumab, Anti-epidermalgrowth factor receptor (EGFR) antibody, anti-vascular endothelial growth factor (VEGF) antibody with first-line chemotherapy, Trifluridine-tipiracil, Regorafenib, Encorafenib with cetuximab (for patients with BRAF V600E mutations), or immunotherapy targeting checkpoints (Pembrolizumab monotherapy, Nivolumab monotherapy, or Nivolumab and ipilimumab).

49. The method of claims 34, 45-46, wherein the colorectal cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Colon cancer, Version 1.2024 (January 29, 2024).

50. The method of claims 1-21, wherein the one or more hybrid cells comprises a gene expression profile of a breast cell-type and the subject is administered a breast cancer treatment.

51. The method of claim 50, wherein the breast cancer treatment comprises a lumpectomy or a mastectomy.

52. The method of claim 50, wherein the breast cancer treatment comprises chemotherapy, radiation therapy, adjuvant therapy, or hormonal therapy.

53. The method of claim 50, wherein the breast cancer treatment comprises Tamoxifen, Aromatase inhibitor (Al), Trastuzumab, Pertuzumab, Neratinib, or Olaparib.

54. The method of claim 50, wherein the breast cancer treatment comprises one or more of the treatments described in the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Breast Cancer, Version 2.2024 (March 11, 2024).

55. The method of claims 1-54, wherein the expression level of one or more markers or genes in any of the preceding claims is determined using scRNAseq.

56. A method for diagnosing cancer in a subject, the method comprising: a. detecting in a sample one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

57. A method for detecting cancer recurrence in a subject in need thereof, the method comprising: a. detecting in a sample one or more hybrid cells, wherein the one or more hybrid cells express one or more epithelial cell lineage markers and one or more monocyte or macrophage cell lineage markers.

58. The method of claim 56 or 57, wherein the one or more hybrid cells express at least two epithelial cell lineage markers and at least two monocyte or macrophage cell lineage markers.

59. The method of any of claims 56-58, wherein the one or more epithelial cell lineage markers are selected from EPCAM, KRT8, KRT9 (CK9), KRT19 (CK19), or CK18; and the one or more monocyte or macrophage cell lineage markers are selected from CD 163, CD 14, CD68, FCER1G, TYROBP, or AIF1.

60. The method of claim 59, wherein the epithelial cell lineage marker is EPCAM and the monocyte or macrophage cell lineage markers is macrophage cell lineage marker CD163.

61. The method of claim 60, wherein the at least two epithelial cell lineage markers are EPCAM and KRT8, and at least two monocyte or macrophage cell lineage markers are macrophage cell lineage markers CD163 and CD14.

62. The method of claims 56-61, wherein the one or more hybrid cells further express one or more blood lineage markers.

63. The method of claim 62, wherein the blood lineage marker is PTPRC (CD45).

64. The method of claims 56-63, wherein the one or more hybrid cells have a doublet score of <0.5.

65. The method of claim 64, wherein the one or more hybrid cells have a doublet score between 0.1 and 0.5.

66. The method of claims 56-65, wherein a cell type annotation of the one or more hybrid cells is a myeloid cell type.

67. The method of claim 66, wherein the myeloid cell type is monocyte or macrophage.

68. The method of claims 56-65, wherein a cell type annotation of the one or more hybrid cells is an epithelial cell type.

69. The method of claims 66-68, wherein the cell type annotation is a graph-clustering and non-negative matrix factorization (NMF) approach.

70. The method of claims 56-69, wherein a gene expression pattern of the one or more hybrid cells is different from a gene expression pattern of a non-cancerous epithelial cell of the subject or a gene expression pattern of a non-cancerous macrophage or monocyte cell of the subject.

71. The method of claims 56-70, wherein the one or more hybrid cells comprise a transcriptome with higher signaling levels of KRAS, TP53, or both compared tosignaling levels of KRAS, TP53, or both in a transcriptome of a non-cancerous cell of the subject.

72. The method of claim 71, wherein the gene expression pattern comprises gene expression of one or more of the genes described in Figure 15 or Figure 16.

73. The method of claims 72, wherein the gene expression pattern is expression for genes associated with EMT.

74. The method of claim 73, wherein the one or more hybrid cells comprise a gene expression level that is increased for genes associated with EMT compared to a gene expression level of said genes in a non-cancerous cell of the subject.

75. The method of claim 74, wherein the gene expression is expression of CD44. SPP1, or both CD44 and SPP1.

76. The method of claim 74, wherein the one or more hybrid cells comprise a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level that is increased compared to a CD44 expression level, SPP1 expression level, or both CD44 expression level and SPP1 expression level in a non- cancerous cell of the subject.

77. The method of claims 56-76, wherein the cancer is a solid cancer.

78. The method of claim 77 wherein the solid cancer is colon, breast, ovarian, pancreatic, renal, esophageal, melanoma, lung, brain, prostate, or uveal melanoma.

79. The method of claim 78, wherein the cancer is colorectal cancer.

80. The method of claim 78, wherein the cancer is lung cancer.

81. The method of claim 80, wherein the lung cancer is NSCLC.

82. The method of claim 78, wherein the cancer is breast cancer.

83. The method of claims 56-79, wherein the one or more hybrid cells comprises a gene expression profile of a colon cell-type and the subject is administered a colon cancer treatment.

84. The method of claims 83, wherein the one or more hybrid cells comprises a gene expression profile of a colorectal cell-type.

85. The method of claims 56-78, 82, wherein the one or more hybrid cells comprises a gene expression profile of a breast cell-type.

86. The method of claims 56-78, 80, wherein the one or more hybrid cells comprises a gene expression profile of a lung cell-type.

87. The method of claims 56-78, 81, wherein the one or more hybrid cells comprises a gene expression profile of a NSCLC cell-type.

88. The method of claims 56-87, wherein the expression level of one or more markers or genes in any of the preceding claims is determined using scRNAseq.

89. The method of claims 56-87, the expression level of one or more markers or genes in any of the preceding embodiments is determined using RT-qPCR.

90. A method of determining a cancer prognosis in a subject in need thereof, wherein the prognosis depends on numbers and / or size of hybrid cells of claims 56-87 in the sample.

91. The method of claim 90, wherein the subject is determined to have a poor prognosis if the subject has a high number of hybrid cells and / or large hybrid cells.

92. The method of claims 1-91 wherein the sample is biopsy tissue or blood.

93. The method of claims 1-92, wherein the subject is a human.

Images