Enabling genomic testing using broad-spectrum nanoparticles

Abstract

The present invention relates to methods, compositions and devices for early detection and prevention of various diseases such as cancer, infectious diseases and autoimmune diseases using broad-spectrum enrichment technology and single-cell sequencing technology. The present invention also provides an example of a method for detecting circulating tumor cells (CTCs) using platelet membrane-coated nanoparticles (PNP) and single-cell RNA sequencing (scRNA-seq) technology.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of U.S. Provisional Application No. 62 / 841,009, filed on April 30, 2019, which is incorporated by reference in this application and becomes a part of this application. Technical Field

[0003] This patent document relates to the use of broad-spectrum enrichment technology and single-cell sequencing technology to effectively and accurately detect various diseases such as cancer, infectious diseases and autoimmune diseases. In particular, the technology disclosed in the present invention can be used to detect and identify circulating tumor cells (CTCs). Background Art

[0004] Since CTCs are extremely rare (only 1-10 circulating tumor cells exist in 1 million blood cells), they are extremely difficult to capture, which is a major obstacle to the clinical application of CTCs. A variety of CTC enrichment methods have been developed, which can be mainly divided into two categories. The first category of methods utilizes the biophysical properties of CTCs such as deformability, density, cell surface charge and size. However, due to the variability of different cells in the blood, the specificity of the methods is generally low. The second category of methods utilizes positive or negative immunoaffinity selection strategies for CTC surface markers. Positive selection strategies rely on specific surface proteins, such as EpCAM used in CellSearch (an FDA-approved CTC enrichment technology), which requires prior knowledge of the cancer type. For example, during the epithelial-mesenchymal transition (EMT) process of CTCs and when epithelial markers are lost, positive selection methods also encounter problems with cancer cell heterogeneity, which can lead to false-negative results. Negative selection strategies aim to remove a large number of red blood cells, white blood cells and platelets. Negative selection methods do not produce bias, but may retain non-CTC circulating cells such as epithelial cells in the bloodstream, which can lead to false positives. Summary of the invention

[0005] This patent document provides a new method for effectively and sensitively detecting CTCs and other biological objects using nanoparticle enrichment technology, which overcomes the limitations of existing enrichment methods.

[0006] On the one hand, this patent document provides a broad, unbiased enrichment method for a target object in a biological sample, comprising contacting the biological sample with nanoparticles, the nanoparticles being coated with a cell membrane having a binding affinity with the target object. In some embodiments, the target object is CTC, and the nanoparticles are platelet membrane-coated nanoparticles (PNP).

[0007] On the other hand, this patent document provides a method for extensive enrichment and gene identification of target cells, including enrichment of target cells using cell membrane-coated nanoparticles, and single-cell sequencing of the enriched target cells. In some embodiments, the target object is PNP-enriched CTC. In other embodiments, the single-cell sequencing is single-cell RNA sequencing (scRNA-seq).

[0008] On the other hand, this patent document provides a method for identifying CTC cancer types or subtypes in biological samples, including single-cell RNA sequencing (scRNA-seq) of PNP-enriched CTCs, and comparing the single-cell RNA sequencing (scRNA-seq) results and multiple genes whose expression profiles are associated with different cancer types or subtypes. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figures 1A-1F The preparation and characterization of PNPs are shown.

[0010] Figure 2 The adhesion of PNP on a broad spectrum of cancer cells is shown.

[0011] Figure 3 Enrichment of cancer cells by PNP from an in vitro mixture of leukocytes is shown.

[0012] Figure 4 Enrichment of cancer cells from tumor xenografts by PNP is shown.

[0013] Figures 5A-5C Diagnostic and prognostic analysis using triple-confirmed genes is shown.

[0014] Figures 6A-6H The Heracles framework is shown.

[0015] Figure 7A-7B Shows the enrichment of cancer cells in blood samples from cancer patients.

[0016] Figures 8A-8B The spike analysis is shown. DETAILED DESCRIPTION

[0017] Biologically, in order to evade attack by immune cells and reduce the shear stress experienced by tumor cell apoptosis and circulation, CTCs adsorb platelets on their surface. The interaction between CTCs and platelets is essential for the survival of CTCs in the bloodstream. Therefore, the platelet membrane is embedded with receptors that can recognize non-specific CTCs of any cancer type, making the platelet membrane a natural universal capture probe for CTCs of all cancer types.

[0018] PNP is a nanoparticle coated with the natural cell membrane of platelets. Due to the natural interaction between CTCs and platelets, these biomimetic nanoparticles can effectively enrich CTCs in blood samples. PNP can capture live CTCs without prior information about the cancer type, so it has a unique advantage in cancer detection in the early stages of non-clinical diagnosis. Importantly, because platelets interact with CTCs of all cancer types, PNP-supported CTC capture does not rely on antibodies that recognize specific antigens on CTCs, such as epithelial markers. The ability of this technology has been demonstrated by successfully capturing small amounts of cancer cells mixed with large amounts of white blood cells or added to human blood. The technology disclosed in the present invention is a broad-spectrum technology that can enrich live cancer cells for genomic characterization.

[0019] PNP can enrich CTCs of any cancer type, which provides a universal probe for detecting CTCs. This feature is particularly suitable for early detection and treatment of cancer without the need for prior information on cancer type or subtype. At the same time, this also poses a challenge to determining the cancer type or subtype in which CTCs are enriched. In order to fully realize the advantages of PNP capture, accurate and robust identification of captured CTCs is required. Considering the small number of CTCs in the circulation, single-cell RNA sequencing (scRNA-seq) analysis is a good choice to obtain the gene expression profile of PNP-enriched CTCs and characterize their cancer type or subtype based on this gene expression profile.

[0020] However, because cancer cells are heterogeneous and single-cell measurements are noisy, it remains a challenge to determine cancer types or subtypes based on single-cell RNA sequencing (scRNA-seq). In fact, single-cell data from tumor samples can be used as a benchmark for capturing CTC annotations, but these data are very limited. To overcome these challenges, the present invention further provides a novel bioinformatics strategy, a benchmark dataset for each cancer type or subtype, and a robust algorithm to accurately classify single CTCs. The present invention provides genes that can distinguish between cancer samples and normal samples and different cancer types studied by TCGA. These so-called "triple proof genes" are characterized by differential gene expression and DNA methylation and are associated with somatic mutations. Utilizing this discovery, the present invention provides a new bioinformatics framework and computational strategy to identify a subset of these genes (which can be robustly detected by single-cell RNA sequencing (scRNA-seq), which can be used to determine the cancer origin of captured CTCs).

[0021] The technology disclosed in the present invention provides at least the following innovative contents. First, the novel broad-spectrum method can capture CTCs from any type of cancer without the use of prior information. In an embodiment, the nanoparticles are coated with a platelet membrane having a series of surface proteins to form PNP. Although it is not clear which receptor is important for CTC-platelet interactions, coating the entire membrane can avoid any bias in the selection of specific surface markers. Importantly, since the interaction between CTCs and platelets is crucial to the survival of CTCs of any cancer type, the present invention is universally used and does not require prior information on the cancer type. In addition, since CTCs are released during early tumor development, PNP can be used to capture CTCs without prior information, which is particularly suitable for early detection of cancer and metastasis.

[0022] Secondly, the present invention uses a novel computational framework to accurately identify cancer types or subtypes based on single-cell RNA sequencing (scRNA-seq) data. The innovative method overcomes two major obstacles to identifying CTCs based on single-cell RNA sequencing (scRNA-seq) data: (i) due to the noise of single-cell RNA sequencing (scRNA-seq) data, there is still a lack of robust bioinformatics methods that can accurately predict cancer types based on single-cell RNA sequencing (scRNA-seq) data; and (ii) in order to reduce the impact of tumor heterogeneity, single-cell RNA sequencing (scRNA-seq) of cancer cells obtained from tumor samples is an optimal way to discover cancer transcriptome features, but it is currently not available for every major cancer. In the study described in this article, cancer transcriptome features were obtained using bulk RNA-seq data from multiple cancers and hundreds or thousands of patients. Using this feature, a new computational strategy has been developed that can determine the origin of each CTC based on single-cell RNA sequencing (scRNA-seq) data, which is the first strategy of its kind.

[0023] Third, a conceptually novel and practical approach is provided by combining nanoengineering with genomics technologies. Broad-spectrum capture of CTCs does not distinguish the cancer type or subtype from which the CTCs originate. This limitation is overcome by performing transcriptome analysis on captured CTCs. Effective enrichment of CTCs using PNPs facilitates single-cell analysis because it can recover more CTCs and improve detection capabilities. This type of synergy makes the method a powerful liquid biopsy tool. In addition, captured live CTCs can also undergo additional single-cell analysis, such as DNA sequencing, ATAC-seq, and splicing analysis, to better understand cancer biology and characterize tumor progression.

[0024] The present invention provides a method for enriching a target object in a biological sample. In an embodiment, the present invention provides a method, comprising contacting the biological sample with one or more nanoparticles, the nanoparticles being coated with a cell membrane having a binding affinity with the target object. In an embodiment, the present invention provides a method, comprising coating one or more nanoparticles with a cell membrane having a binding affinity with the target object, and contacting the nanoparticles with the biological sample. In an embodiment, the cell membrane of the present invention is a platelet membrane.

[0025] In an embodiment, the target object of the present invention is a circulating tumor cell (CTC). In an embodiment, the CTC of the present invention is a circulating tumor cell of breast cancer, lung cancer, ovarian cancer or gastric cancer. In an embodiment, the biological sample of the present invention is derived from a mammal, including a human patient. In an embodiment, the nanoparticle of the present invention comprises a polylactic-co-glycolic acid (PLGA) core. In an embodiment, the method of the present invention further comprises at least partially separating the nanoparticle bound to the target object from the biological sample.

[0026] The present invention provides a method for enriching and gene identifying target cells in a biological sample. In an embodiment, the method includes enriching target cells by contacting the biological sample with nanoparticles, wherein the nanoparticles are coated with a cell membrane having binding affinity with the target cells; and performing single-cell nucleic acid sequencing on the enriched target cells. In an embodiment, triple proof genes are used to distinguish the cancer sample from non-cancerous cells. In an embodiment, the single-cell nucleic acid sequencing is single-cell RNA sequencing (scRNA-seq). In an embodiment, the method further includes at least partially separating the nanoparticles bound to the target object from the biological sample before performing single-cell nucleic acid sequencing.

[0027] In an embodiment, the cell membrane of the present invention is a platelet membrane. In an embodiment, the target object of the present invention is a circulating tumor cell (CTC). In an embodiment, the CTC of the present invention is a circulating tumor cell of breast cancer, lung cancer, ovarian cancer or gastric cancer. In an embodiment, the biological sample of the present invention is derived from a mammal, including a human patient. In an embodiment, the nanoparticle of the present invention comprises a polylactic-co-glycolic acid (PLGA) core. In an embodiment, the method of the present invention further comprises at least partially separating the nanoparticles bound to the target object from the biological sample.

[0028] The present invention provides a method for identifying a CTC cancer type or subtype in a biological sample, comprising: contacting a biological sample with PNP to enrich CTCs; performing single-cell RNA sequencing (scRNA-seq) on the enriched CTCs; and determining the CTC cancer type or subtype based on a comparison of single-cell RNA sequencing (scRNA-seq) and multiple genes (whose expression profiles are associated with different cancer types or subtypes). In an embodiment, the cancer type of the present invention is breast cancer, lung cancer, ovarian cancer, or gastric cancer. In an embodiment, the biological sample of the present invention is derived from a mammal, including a human patient. The present invention further provides a method for treating a patient, comprising administering an effective amount of treatment (drug, surgery, or radiation) to a patient in need to inhibit the cancer identified in the patient's biological sample.

[0029] The present invention provides a CTC-enriched composition comprising CTCs at least partially isolated from a patient biological sample, wherein the CTCs are combined with platelet membrane-coated nanoparticles. In an embodiment, the CTCs are circulating tumor cells of breast cancer, lung cancer, ovarian cancer or gastric cancer. In an embodiment, the biological sample of the present invention is derived from a mammal, including a human patient.

[0030] Unless otherwise indicated, in the practice of the present invention, conventional nanotechnology, nanoengineering, molecular biology (including recombinant techniques), microbiology, cell biology, biochemistry, immunology, and pharmacology techniques within the skill of the art will be employed. Such techniques are fully explained in the literature, such as the Molecular Cloning: A Laboratory Manual (2nd Edition) (Sambrook et al., 1989); Oligonucleotide Synthesis (Editor: MJ Gait, 1984); Animal cell culture ” (Editor: R.I. Freshney, 1987); Enzymatic Methods (Academic Press); Modern Molecular Biology Experiment Guide (Editor: F.M. Ausubel et al., 1987, and regularly updated); PCR: polymerase chain reaction (Editor: Mullis et al., 1994); and Remington: The Science and Practice of Pharmacy 20th Edition (Lippincott Williams & Wilkins, 2003).

[0031] To facilitate understanding of the content of the invention, some terms and abbreviations used in this document are defined as follows:

[0032] When introducing elements of the present invention or the preferred embodiments thereof, the articles "a," "an," "the," and "said" mean that there are one or more of the elements. The terms "comprising," "including," and "having" are inclusive and mean that there may be additional elements other than the listed elements.

[0033] The term "and / or" when used when listing two or more items means that any one of the listed items can be used alone or in combination with any one or more of the listed items. For example, the expression "A and / or B" means A or B, or A and B, that is, A alone, B alone, or A and B in combination. The expression "A, B, and / or C" means A alone, B alone, C alone, A and B in combination, A and C in combination, B and C in combination, or A, B and C in combination.

[0034] As used herein, the term "cell membrane" refers to a biological membrane encapsulation or separation structure used as a selective barrier in or around a cell or an emerging virus particle. In an embodiment, the cell membrane is a platelet membrane. The cell membrane is selectively permeable to ions and organic molecules and controls the entry and exit of substances into and out of the cell. The cell membrane comprises a phospholipid monolayer or bilayer and may also comprise associated proteins and carbohydrates. As used herein, the cell membrane refers to a membrane obtained from or derived from a biological membrane naturally present in a cell or organelle. As used herein, the term "naturally present" refers to a substance present in nature. As used herein, the term "derived from" refers to any subsequent modification of a natural membrane, such as separating a cell membrane, forming a portion or fragment of a membrane, removing certain components from the membrane of a cell or organelle and / or adding certain components thereto, such as lipids, proteins or carbohydrates. The membrane may be derived from a naturally present membrane by any suitable method. For example, a membrane may be prepared or isolated from a cell or virus, and the prepared or isolated membrane may be combined with other substances or materials to form a derived membrane. In another example, a cell or virus can be recombinantly engineered to produce a "non-natural" substance that associates with a cell or virus membrane in vivo, which is then prepared or isolated from the cell or virus to form a derived membrane.

[0035] In various embodiments, the cell membrane covering the monolayer or multilayer nanoparticles can be further modified, saturated or unsaturated with other lipid components such as cholesterol, free fatty acids and phospholipids, and the cell membrane can also include endogenous or added proteins and carbohydrates such as cell surface antigens. In this case, excess amounts of other lipid components can be added to the membrane wall, which will fall off until the concentration in the membrane wall reaches equilibrium, depending on the environment of the nanoparticles. The membrane can also include other agents that may or may not increase the activity of the nanoparticles. In other examples, functional groups such as antibodies and aptamers can be added to the outer surface of the membrane to enhance site targeting, such as adding functional groups such as antibodies and aptamers to cell surface epitopes found in cancer cells. The nanoparticle membrane can also include biodegradable particles, cationic nanoparticles, including but not limited to gold, silver and synthetic nanoparticles.

[0036] The term "synthetic membrane" or "artificial membrane" refers to an artificial membrane produced using organic materials such as polymers and liquids, as well as inorganic materials. Various synthetic membranes are well known in the art.

[0037] As used herein, the term "nanoparticle" refers to a nanostructure, particle, vesicle, or fragment thereof having at least one dimension (such as height, length, width, or diameter) of about 1 nm to about 10 μm. For systemic application, an average diameter of about 50 nm to about 500 nm or 100 nm to 250 nm may be preferred. The term "nanostructure" includes, but is not limited to, particles and engineered features. For example, particles and engineered features may have regular or irregular shapes. The particles are also referred to as nanoparticles. The nanoparticles may be composed of organic materials or other materials and may be implemented using porous particles. A nanoparticle layer may be implemented using a single layer of nanoparticles or a nanoparticle agglomerate layer. In some embodiments, the nanoparticle comprises an inner core covered by an outer surface, and the outer surface comprises a membrane as described herein. The present invention relates to any nanoparticle currently known and later developed that may be coated with the membrane as described herein.

[0038] As used herein, the term "phospholipid" refers to any lipid comprising a diglyceride, a phosphate group, and a simple organic molecule such as choline. Examples of phospholipids include, but are not limited to, phosphatidic acid (phosphatidate) (PA), phosphatidylethanolamine (cephalin) (PE), phosphatidylcholine (lecithin) (PC), phosphatidylserine (PS), and phosphoinositides, including but not limited to phosphatidylinositol (PI), phosphatidylinositol phosphate (PIP), phosphatidylinositol diphosphate (PIP2), and phosphatidylinositol triphosphate (PIP3). Other examples of PC include DDPC, DLPC, DMPC, DPPC, DSPC, DOPC, POPC, DRPC, and DEPC as defined in the art.

[0039] It is to be understood that various aspects and embodiments of the present invention include "comprising" and / or "essentially comprising" various aspects and embodiments.

[0040] In the present disclosure, the various aspects of the present invention are presented in range format. It should be understood that the description in range format is only for convenience and brevity and should not be regarded as a fixed limitation on the scope of the present invention. Therefore, the range description should be regarded as having clearly disclosed all possible sub-ranges and each numerical value within this range. For example, the description of the range 1-6 should be regarded as having clearly disclosed sub-ranges such as 1-3, 1-4, 1-5, 2-4, 2-6, 3-6, and each number within this range, such as 1, 2, 3, 4, 5 and 6. This applies regardless of the breadth of the range.

[0041] The present method can be used to detect CTC or other targeted objects using membrane-coated nanoparticles in any applicable patient biological sample. In some embodiments, the biological sample is obtained from a mammal. In some embodiments, the mammal is a human. In other embodiments, the mammal is a non-human mammal, including pets, farm animals, economic animals, sports animals, and experimental animals, such as cats, dogs, horses, cows, bulls, pigs, donkeys, sheep, lambs, goats, mice, rabbits, primates (including monkeys and chimpanzees).

[0042] The nanoparticles used in the present method may comprise any applicable core. For example, the nanoparticle core may comprise a polymer particle core, a silica particle core, or a metal (e.g., gold) particle core. Any appropriate polymer particle core may be used. In some embodiments, the polymer particle core may have optical shift properties. In other embodiments, the polymer particle core may comprise a metal, such as gold, iron oxide, or quantum dots. In other embodiments, the nanoparticle core may comprise a biocompatible or synthetic material, such as polylactic-co-glycolic acid (PLGA), polylactic acid (PLA), polyglycolic acid (PGA), polycaprolactone (PCL), polylysine, and polyglutamic acid. In other embodiments, the nanoparticle core supports the outer surface.

[0043] The nanoparticles used in the present method may comprise any suitable cell membrane derived from platelets or other cell types. For example, the nanoparticles used in the present method may comprise a cytoplasmic membrane derived from platelets. In another example, the nanoparticles used in the present method may comprise an intracellular membrane derived from platelets.

[0044] The nanoparticle can comprise the releasable content of any suitable location.For example, the releasable content can be located in the kernel or on the kernel, between the kernel and the outer surface or in the outer surface or on the outer surface. The release operation of the releasable content can be triggered by any suitable mechanism.For example, the nanoparticle can be contacted with an object or an object cell, or the physical parameter of the nanoparticle can be changed, thereby triggering the release operation of the releasable content. The nanoparticle can comprise the releasable content of any suitable type.For example, the releasable content can be a metal particle, a polymer particle, a dendrimer particle or an inorganic particle.

[0045] The nanoparticles used in the present method can adopt any size.For example, the diameter of the nanoparticles can be about 10nm to about 10μm.In certain embodiments, the diameter of the nanoparticles is about 10nm, 20nm, 30nm, 40nm, 50nm, 60nm, 70nm, 80nm, 90nm, 100nm, 110nm, 120nm, 130nm, 140nm, 150nm, 200nm, 300nm, 400nm, 500nm, 600nm, 700nm, 800nm, 900nm, 1μm, 2μm, 3μm, 4μm, 5μm, 6μm, 7μm, 8μm, 9μm and 10μm, or any sub-range in the range of about 10nm to about 10μm, such as any range between any two sizes mentioned above.

[0046] The nanoparticles used in the present method can have any suitable shape, including but not limited to spheres, squares, rectangles, triangles, discs, cube-like shapes, cubes, cuboids, cones, cylinders, prisms, pyramids, right-angled cylinders, and other regular or irregular shapes.

[0047] In some embodiments, the nanoparticles used in the present method substantially lack components of platelet-derived cell membranes. For example, the nanoparticles may lack 10%, 20%, 30%, 40%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% of the components of platelet-derived cell membranes.

[0048] In some embodiments, the nanoparticles described herein can be used to isolate the targeted objects (e.g., CTCs) used in the methods and compositions from a biological sample such that they are substantially devoid of components of the biological sample derived targeted objects. For example, the targeted objects may lack 10%, 20%, 30%, 40%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% of the components of the biological sample derived targeted objects.

[0049] In some embodiments, the nanoparticles used in the present method substantially maintain the natural structural integrity or activity of the platelet-derived cell membrane or platelet-derived cell membrane components. For example, the nanoparticles can maintain 10%, 20%, 30%, 40%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% of the natural structural integrity or activity to target the damaged or leaky vasculature of the subject or to target platelet-targeted microorganisms.

[0050] In some embodiments, the nanoparticles used in the present method are biocompatible or biodegradable. For example, the nanoparticle core comprises PLGA, and the nanomaterial outer surface comprises a platelet-derived cytoplasmic membrane.

[0051] The nanoparticle used in the present method can have any suitable half-life in vivo. For example, the half-life of the nanoparticle in the blood circulation in vivo may be from about 30 minutes to about 10 hours, such as about 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours.

[0052] The outer surface of the nanoparticles used in the present method may further include a synthetic membrane. In certain embodiments, the nanoparticles used in the present method include a mixture, and the mixture is composed of nanoparticles whose outer surface contains platelet cell membranes, and nanoparticles whose outer surface contains synthetic membranes. The nanoparticles whose outer surface contains synthetic membranes can or cannot target the CTCs of patients. In certain embodiments, the nanoparticles whose outer surface contains platelet cell membranes, and the nanoparticles whose outer surface contains synthetic membranes can both target the CTCs of patients. In other embodiments, the nanoparticles whose outer surface contains platelet cell membranes can target the CTCs of patients, but the nanoparticles whose outer surface contains synthetic membranes cannot target the CTCs of patients.

[0053] The composition used in the method may contain nanoparticles having platelet cell membranes on their outer surfaces and nanoparticles having synthetic membranes on their outer surfaces in any proportion. In some embodiments, the present composition may comprise at least about 1% (w / w), 2% (w / w), 3% (w / w), 4% (w / w), 5% (w / w), 6% (w / w), 7% (w / w), 8% (w / w), 9% (w / w), 10% (w / w), 20% (w / w), 30% (w / w), 40% (w / w), 50% (w / w), 60% (w / w), 70% (w / w), 80% (w / w), 90% (w / w), 91% (w / w), 92% (w / w), 93% (w / w), 94% (w / w), 95% (w / w), 96% (w / w), 97% (w / w), 98% (w / w), 99% (w / w) or more nanoparticles having platelet cell membrane on their outer surface. In other embodiments, the composition may comprise at least about 1% (w / w), 2% (w / w), 3% (w / w), 4% (w / w), 5% (w / w), 6% (w / w), 7% (w / w), 8% (w / w), 9% (w / w), 10% (w / w), 20% (w / w), 30% (w / w), 40% (w / w), 50% (w / w), 60% (w / w), 70% (w / w), 80% (w / w), 90% (w / w), 91% (w / w), 92% (w / w), 93% (w / w), 94% (w / w), 95% (w / w), 96% (w / w), 97% (w / w), 98% (w / w), 99% (w / w) or more of nanoparticles having a synthetic film on their outer surfaces. For example, the composition used in the present method may contain about 1-10% (w / w) nanoparticles containing platelet cell membranes on the outer surface and about 90-99% (w / w) nanoparticles containing synthetic membranes on the outer surface, about 11-25% (w / w) nanoparticles containing platelet cell membranes on the outer surface and about 75-89% (w / w) nanoparticles containing synthetic membranes on the outer surface, about 50% (w / w) nanoparticles containing platelet cell membranes on the outer surface and about 50% (w / w) nanoparticles containing synthetic membranes on the outer surface, about 51-75% (w / w) nanoparticles containing platelet cell membranes on the outer surface and about 49-25% (w / w) nanoparticles containing synthetic membranes on the outer surface, or about 90-100% (w / w) nanoparticles containing platelet cell membranes on the outer surface and about 0-10% (w / w) nanoparticles containing synthetic membranes on the outer surface.

[0054] The outer surface of the nanoparticle may include a hybrid membrane comprising a platelet-derived cell membrane and a synthetic membrane. In some embodiments, the outer surface of the nanoparticle may include a hybrid membrane comprising at least about 5% (w / w), 6% (w / w), 7% (w / w), 8% (w / w), 9% (w / w), 10% (w / w), 20% (w / w), 30% (w / w), 40% (w / w), 50% (w / w), 60% (w / w), 70% (w / w), 80% (w / w), 90% (w / w), 91% (w / w), 92% (w / w), 93% (w / w), 94% (w / w), 95% (w / w), 96% (w / w), 97% (w / w), 98% (w / w), 99% (w / w) of platelet-derived cell membrane. In some embodiments, the outer surface of the nanoparticle may comprise a hybrid film, comprising at least about 1% (w / w), 2% (w / w), 3% (w / w), 4% (w / w), 5% (w / w), 6% (w / w), 7% (w / w), 8% (w / w), 9% (w / w), 10% (w / w), 20% (w / w), 30% (w / w), 40% (w / w), 50% (w / w), 60% (w / w), 70% (w / w), 80% (w / w), 90% (w / w), 91% (w / w), 92% (w / w), 93% (w / w), 94% (w / w), 95% (w / w) of a synthetic film. For example, the outer surface of the nanoparticle may comprise a hybrid membrane comprising about 5-10% (w / w) platelet-derived cell membrane and about 90-95% (w / w) synthetic membrane, about 11-25% (w / w) platelet-derived cell membrane and about 75-89% (w / w) synthetic membrane, about 50% (w / w) platelet-derived cell membrane and about 50% (w / w) synthetic membrane, about 51-75% (w / w) platelet-derived cell membrane and about 49-25% (w / w) synthetic membrane, or about 90-99% (w / w) platelet-derived cell membrane and about 1-10% (w / w) synthetic membrane.

[0055] In some embodiments, the nanoparticles used in the present method are substantially lacking in immunogenicity of the subject (e.g., a mammal). For example, the cell membrane may be derived from platelets of the same subject. In another example, the subject is a human and the cell membrane is derived from human platelets. In some embodiments, the cell membrane may be derived from platelets of a mammal to be treated. For example, the cell membrane may be derived from platelets of a human to be treated.

[0056] One aspect of the present invention relates to a method for preparing polymer nanoparticles encapsulated in human platelet cytoplasmic membranes, wherein the polymer nanoparticles are a unique cell fragment population adhered to various disease matrices. The obtained nanoparticles have a monolayer membrane covering with the right side facing outward, which functions through immunomodulation and adhesion antigens associated with platelets. Compared with uncoated particles, platelet membrane-coated nanoparticles can reduce cellular uptake by macrophage-like cells and lack particle-induced complement activation in human autologous plasma. The coated nanoparticles also have platelet biomimetic properties and enhance binding to platelet-adherent pathogens. In a mouse model of systemic bacterial infection, docetaxel and vancomycin were both enhanced in efficacy when delivered via platelet biomimetic nanoparticles. The multilayer biological interface achieved by the platelet membrane coating method provides a new method for developing functional nanoparticles for disease-targeted delivery.

[0057] Example

[0058] Example 1: PNP preparation and characterization. PNPs were prepared according to previously published protocols. 23 Briefly, human platelet membranes were collected and fused to poly(lactic-co-glycolic acid) (PLGA) cores with a diameter of 100 ± 3.1 nm. The PNPs obtained had a diameter of 115 ± 2.3 nm ( Figure 1A ). Adding cell membrane to the PLGA core increased the diameter by about 15 nm. After coating, the surface charge increased from -54.3±7.1 mV (core) to -30.4±6.6 mV (PNP), probably due to the charge shielding effect of the platelet membrane. Transmission electron microscopy (TEM) images showed that the formation of a unique core-shell structure was consistent with the membrane shell coated with the PLGA core ( Figure 1B At the same time, immunogold staining and antibodies targeting the extracellular domain of CD47 confirmed that PNP was oriented to the right side of the outer membrane ( Figure 1C ). In addition, PNPs expressed higher levels of P-selectin, an activation marker critical for CTC binding, compared to platelets (Figure 1D). Activation likely occurred during the freeze-thaw cycles of the membrane derivatization process. It was also observed that the colloidal stability of PNPs was increased compared to the bare NP core due to the stabilizing effect of hydrophilic surface glycans of the platelet membrane (Figure 1E). In addition, platelet membrane protein content was demonstrated to be translocated to the nanoparticles by western blotting ( Figure 1F ). These results together demonstrate that PNPs have been successfully formed.

[0059] Specifically, Figures 1A-1F The preparation and characterization of PNPs are shown. Figure 1A Shown are the sizes and surface zeta potentials of PLGA cores and PNPs measured by dynamic light scattering (n=3). Figure 1BTEM images of bare cores and PNPs are shown. Scale bar = 100 nm. Figure 1C TEM images of PNPs stained with an extracellular domain-specific anti-CD47 antibody and secondary immunogold staining are shown. Scale bar = 40 nm. FIG. 1D shows a comparison of PNPs and platelet P-selectin expression levels containing equal amounts of membranes. FIG. 1E shows particle diameters of bare NP cores and PNPs in water and 1X PBS (n=3). Figure 1F Representative protein bands resolved by western blotting are shown.

[0060] Example 2: PNP adheres to a broad spectrum of cancer cells. To study the adhesion of PNP, cancer cell lines were selected, including AGS (gastric adenocarcinoma), SK-OV-3 (ovarian cystadenocarcinoma), H1299 (non-small cell lung cancer), and HCT 116 (colorectal cancer). 6 PNPs (0.1 mg / mL in 1X PBS) labeled with DiD dye (excitation / emission = 644 / 665 nm) were added to 100 μg / mL cells / mL. RBC membrane-coated nanoparticles (RBC-NPs) were used as negative controls. The mixture was incubated at 4°C for 30 minutes and then washed to remove unbound nanoparticles. PNP adhesion was measured by flow cytometry. Regardless of the potential binding mechanism, PNP adhered to all tested cancer cell lines ( Figure 2 ).

[0061] Specifically, Figure 2 The adhesion of PNP to a broad spectrum of cancer cells is shown. The flow cytometric histograms show an increase in fluorescence intensity for various cancer cell lines cultured with PNP compared to cancer cell lines cultured with RBC-NP or without treatment. Untreated cells appear light grey, RBC-NP treated cells appear dark grey, and PNP treated cells appear medium grey and shaded below the curve.

[0062] PNPs targeting different breast cancer subtypes were also tested. Although their metastatic potential varied, breast cancer cell lines including MDA-MB-453, SK-BR-3, MCF-7, and MDA-MB-231 are known to activate and bind to platelets. 16，24，25 Preliminary results demonstrated that PNP inherited the properties of platelets and could bind to all breast cancer cell lines tested. This ability demonstrated the potential of PNP as a broad-spectrum CTC enrichment tool.

[0063] Example 3: PNP enrichment of cancer cells from in vitro cell mixtures. Using a cancer cell-leukocyte mixture, the enrichment efficiency was analyzed by flow cytometry. 6 (High) and 10:5x106 MDA-MB-453 cells and leukocytes were mixed at a ratio of 1:1 (low). PNP was added to the cell mixture at 3.9 to 500 μg / mL. The samples were incubated at 4°C for 30 minutes and washed with PBS to remove unbound PNP. All enriched cells (blue, PNP + CD45 – ) and all enriched cancer cells (red, PNP + CD45 – Hoechst + ) for counting. The percentage of cancer cells is plotted against the concentration of PNP (purple). PNP dramatically increases the percentage of cancer cells above 20% (high ratio) and 1% (low ratio), which corresponds to an enrichment factor (i.e., a doubling of the concentration of cancer cells in the cell mixture) of 10 5 and 5×10 3 ( Figure 3 ). This study demonstrated the ability of PNP to enrich CTCs in a cell mixture and that the enrichment efficiency depended on the total number of cells, the CTC to leukocyte ratio, and the PNP concentration.

[0064] Specifically, Figure 3 PNP enrichment of cancer cells from a mixture of leukocytes in vitro is shown. MDA-MB-453 cells (stained with Hoechst 33342) and leukocytes (stained with PE-anti-CD45 antibody) were mixed at different ratios. 3.9 to 500 μg / mL of PNP (labeled with DiD) were added to the mixture. The total enriched cells (PNP + CD45 – ) and enriched cancer cells (PNP + CD45 – Hoechst + ) to count. Calculate the percentage of cancer cells in the enriched cell sample.

[0065] Example 4: PNP enriches cancer cells from in vivo tumor xenografts. To further demonstrate the CTC enrichment potential of PNP, we extracted cell suspensions from MDA-MB-453 mouse xenograft tumors. 6 (high) and 100:5×10 6 (Low) Tumor cells and leukocytes were mixed at two ratios. Leukocytes were added to the cells and stained with PNP, anti-CD45 antibody, and EpCAM. In flow cytometry analysis, all enriched cells (PNP + CD45 – ) were counted, and EpCAM + Count the cells. Draw EpCAM +The relationship between the percentage and the PNP concentration. + The percentage of cells increased significantly by more than 4% (high ratio) and 0.2% (low ratio), which corresponded to an enrichment factor of 10 4 and 2.5×10 3 Corresponding ( Figure 4 ). Specifically, Figure 4 Figure 2 shows the enrichment of cancer cells by PNP from tumor xenografts. MDA-MB-453 xenograft cells (stained with anti-EpCAM antibody) and leukocytes (stained with PE-anti-CD45 antibody) were mixed at different ratios. 3.9 to 500 μg / mL of PNP (labeled with DiD) were added to the mixture. The total enriched cells (left, PNP + CD45 – ) and enriched cancer cells (middle, PNP + CD45 – EpCAM + ) to count. The percentage of cancer cells in the enriched cell sample was calculated (right).

[0066] Overall, the CTC enrichment ability of PNP was further demonstrated using tumor xenograft cells. The optimization strategy aimed at further improving the enrichment efficiency of single-cell RNA sequencing (scRNA-seq) is described below.

[0067] Example 5: Identification of genes that differentiate between various cancers. The ability to integrate multiple omics data to distinguish disease from normal patients has been demonstrated previously. 57 In particular, a new method was developed to significantly expand the coverage of DNA methylation data metrics in patient samples using the Illumina 450K array (18-fold) 21，58 By expanding the methylome data, we were able to successfully identify individual genes that best characterized the core cancer transcriptome. We identified 4,123 triple-validated genes that showed (1) differential gene expression; (2) DNA methylation; and (3) association with somatic mutations in a pan-cancer analysis using TCGA data. 20 These genes can robustly distinguish cancer samples from normal samples and differentiate different cancers using bulk RNA-seq data ( Figures 5A-5C ).

[0068] Specifically, Figures 5A-5C Diagnostic and prognostic analysis using triple-confirmed genes is shown. Figure 5A Diagnostic analysis using triple proven genes to distinguish cancer samples from normal samples in 10-fold cross validation (t-test) is shown. SE: sensitivity; SP: specificity; ACC: accuracy; AUC: area under the ROC curve. Figure 5BConsistency box plots (repeated 100 times) using expression data alone, extended methylation data alone, expression data and extended methylation data together, or expression data and COAD raw 450K data together are shown. Hierarchical clustering heat maps (not shown) were created using the selected features (gene expression and methylation sites) in the prognostic analysis. Tumor samples and the features were clustered, and gene expression values ​​log2 (RPKM) were normalized to [0,1]. Figure 5C Kaplan-Meier survival plots for the two cluster samples are shown.

[0069] To determine whether a single cell captured by PNP is a cancer cell or a normal cell, the Heracles framework was developed ( Figure 5A ), and if it is a cancer cell, the cancer type is determined. This workflow starts with SMART-seq on PNP-captured cells 18，53，54 The experiment begins. Processed and normalized single-cell RNA sequencing (scRNA-seq) data is input into a hierarchical classification model. First, at the top level "Cancer Cell vs. Normal Cell", the cell is classified as normal or cancer cell. If it is a cancer cell, the cell is classified as a specific cancer type in the next layer "Cancer Cell vs. Cancer Cell". A new strategy was developed to convert the identified transcriptomic features from bulk data to single-cell RNA-seq data. Despite the problem of tumor heterogeneity, the bulk data is also very useful for selecting classification gene features. The ability to integrate multiple omics data to distinguish between disease and normal patients has been demonstrated previously. 55 In particular, a new method was developed to significantly expand the coverage of DNA methylation data metrics in patient samples using the Illumina 450K array. 22，56 Inspired by the success of these studies, we identified 4123 triple-validated genes that showed (1) differential gene expression; (2) DNA methylation; and (3) association with somatic mutations in a pan-cancer analysis using TCGA data. 20 Using bulk RNA-seq data, these genes can robustly distinguish cancer samples from normal samples and differentiate between different cancers 20 .

[0070] Some of the above feature genes may not be robustly covered in single-cell experiments due to biological or technical noise, so they may be removed. Importantly, the expression profiles of the selected feature genes should be the same in bulk and single-cell measurements. Using this strategy, feature genes for each layer of the classification model are selected.

[0071] Example 6: “Cancer cells vs. normal cells” and “Cancer cells vs. cancer cells” classification.

[0072] Cancer cell and normal cell classification. Model training was performed using batch RNA-seq data from matched tissues, and normal cell data was taken from GTEx 57 , cancer cell data are taken from TCGA 58 Database. The following tissues were considered: bladder, breast, colon, colorectum, esophagus, brain, kidney, ovary, prostate, lung, stomach, pancreas, thyroid, skin, blood, testis, and cervix. To identify transcriptome features suitable for analysis by single-cell RNA sequencing (scRNA-seq), high-quality Drop-seq was obtained for 82 cells, including breast normal cells (MCF-10a cell line) and cancer cells (MDA-MB-453, MCF7, SK-BR-3). 59 Data. These single-cell data were normalized, and 1002 of the 4123 triple-proven genes were selected as classification features. These 1002 genes are the triple-proven genes with the greatest variability in single cells, and their average expression value differences between normal and cancer samples are the same in both bulk and single-cell data. They have rich functions and are widely involved in cell proliferation and tumorigenesis-related pathways, such as the PI3K / AKT / mTOR pathway; importantly, these pathways are not specific to breast cancer or tissue. In general, matching normal and cancer tissues should remove or at least reduce tissue specificity, so these 1002 genes can represent transcriptome features that distinguish normal samples from cancer samples.

[0073] “Cancer cell vs. cancer cell” classification. If a single cell is classified as a cancer cell, its cancer type needs to be determined next. Initial lesions include breast cancer, gastric cancer, ovarian cancer, and lung cancer. Triple proven genes specific to each of the above cancers were selected from TCGA RNA-seq data. Drop-seq for these four cancers was generated using cancer cell lines (breast cancer cell MCF7, gastric cancer cell AGS, ovarian cancer cell SK-OV-3, and lung cancer cell H1299). 59 Data. Triple-proven genes were selected based on the following criteria: (1) they were the most variable in the four cancers; and (2) their mean expression value differences between the considered cancer and other cancers were identical in both bulk and single-cell data.

[0074] Classification models were built. In both classification tasks, 100 random forest models were trained and regularized using batch RNA-seq data, i.e., 7200 normal samples from GTEx and 9330 cancer samples from TCGA in the “Cancer Cells vs. Normal Cells” classification and various cancers in the “Cancer Cells vs. Cancer Cells” classification were combined. The probability that each single cell is a cancer cell or a specific cancer type was calculated individually by each of the 100 random forest models. The median of the 100 probabilities was taken as the group voting score. In “Cancer Cells vs. Normal Cells”, single cells with a group voting score >0.6 were considered “cancer cells”, single cells with a group voting score <0.4 were considered “normal cells”, and single cells between the two were considered “undetermined”. In “Cancer Cells vs. Cancer Cells”, the group voting scores for each cancer type were calculated. A Poisson test was performed on all scores of single cells. A p-value less than the cutoff value of 10 was considered as negative. -3 Single cells with p-values ​​of 0 are assigned as cancer cells. If the scores are all significant and indicate the possibility of metastasis or multiple cancer origins, single cells can be assigned to multiple cancer types. If there are no significant p-values, single cells are labeled as "undetermined", indicating that the cancer type was not included in the training set.

[0075] Performance evaluation. Drop-seq was performed on four cancer cell lines, gastric cancer (86 AGS cells), breast cancer (68 MCF7 cells), ovarian cancer (76 SK-OV-3 cells), and lung cancer (45 H1299 cells), as well as PBMCs (2200 cells published by 10X Genomics and obtained from their website) 59 The data was validated by Heracles. Please note that these data were generated separately and cannot be used for model training.

[0076] “Cancer cell vs. normal cell” assessment. Cancer cells and PBMCs were mixed at 8 different ratios by random sampling, with replacement and enrichment levels ranging from 0.01% (10 / 100000) to 30% (300 / 1000). Each cancer type was repeated 100 times (Figures 6B-6E). Similarly, to simulate the presence of multiple cancer types in circulation, all cancer cells were merged and then randomly mixed with PBMCs 100 times (Figure 6B-6E). Figure 6F-6G ). Overall, the performance of this model is very good. The higher the enrichment level, the smaller the change in classification. These simulation data show that the current model can reliably detect CTCs at enrichment levels exceeding 0.01%, which is achievable using PNPs.

[0077] “Cancer Cells vs. Cancer Cells” evaluation. For cells classified as cancer cells (80 / 86AGS, 67 / 68MCF7, 70 / 76SK-OV-3, 41 / 45H1299), nested 5-fold cross validation was used to further classify them into specific cancer types ( Figure 6H ). The overall performance is very satisfactory.

[0078] Specifically, Figures 6A-6H The Heracles framework is shown. Fig. 6A In the , a given single cell is classified as a cancer cell or a normal cell, and if it is a cancer cell, the specific cancer type is determined (Figure 6B-6E). Consider the sensitivity of the four cancer types in the current model. The overall sensitivity of the “cancer cell vs. normal cell” classification ( Fig. 6F ) and specificity ( Figure 6G ). Figure 6H The sensitivity of the “cancer cell vs. cancer cell” classification is shown.

[0079] Example 7: Processing and enrichment of patient blood samples. Upon receipt of a blood sample (5 mL) from a patient diagnosed with stage IIA triple-negative breast cancer, the sample was immediately processed by adding 25 mL of suspension buffer (1XPBS containing 2 mM EDTA, 25 mM HEPES, and 1% heat-inactivated FBS) and repeatedly washing with centrifuge (300 g, 5 minutes) until a pure white buffy coat was observed above the RBC layer. The buffy coat was carefully transferred to the RBC lysis buffer to remove residual RBCs. The remaining cells were resuspended in 100 μL of solution and stained with PNP, PE-labeled anti-CD45 antibody, Pacific Blue-labeled anti-EpCAM antibody, and Calcein-AM. During the sorting process, a total of 5×10 6 cells (enrichment factor: 10 3 ) collected 5036 PNP + CD45 - cells, including 335 living cells (Calcein AM + ). In Calcein AM + In cells, 25 EpCAMs were identified + Cells and 310 EpCAM - cell( Figure 7A-7B ).

[0080] Specifically, Figure 7A-7B The enrichment of cancer cells in a blood sample from a cancer patient is shown. The blood sample was processed to remove all RBCs and then stained with PNP, anti-CD45 antibody, anti-EpCAM antibody, and Calcein-AM. Fig. 7AIn the , PNP positively selects cells and CD45 depletes leukocytes. Figure 7B , select Fig. 7A Sequencing was performed on living cells (Calcein-AM+) in the culture medium.

[0081] Example 8: Spike analysis. About 30 MDA-MB-231 breast cancer cells were added to 5 ml of patient blood, and 7 cells were recovered using PNP for SMART-seq analysis. In the "Cancer Cells vs. Normal Cells" layer, all cells were classified as cancer cells ( Fig. 8A ). In the “Cancer Cells vs. Cancer Cells” layer, we observe that significant p-values ​​are obtained from the Poisson test for breast cancer and lung cancer, and insignificant p-values ​​are obtained from the Poisson test for ovarian cancer and gastric cancer ( Figure 8B MDA-MB-231 is highly invasive and metastatic, and its transcriptome profile is highly similar to lung cancer 70，71 This result indicates that the method can detect both primary and metastatic sites.

[0082] Specifically, Figures 8A-8B The spike analysis is shown. Fig. 8A The probabilities of “cancer cells and normal cells” are shown. Figure 8B The probability of "cancer cell vs. cancer cell" is shown.

[0083] In summary, the combination of unique PNP enrichment technology and novel single cell analysis technology has been shown to be effective. We have demonstrated that our method has high sensitivity and specificity in recovering cancer cells mixed with white blood cells at very low percentages and in recovering cancer cells in the blood of cancer patients. Embodiments of the technology disclosed in the present invention can promote the development of new and robust methods, that is, CTCs can be detected without the help of specific surface markers, and the origin of cancer can be determined without prior information.

[0084] Although PNP has been used for CTC enrichment and detection, the same concept can be easily extended to other cell types. Using nanoparticles coated with the appropriate type of cell membrane, targeted objects with specific binding affinity to the cell membrane can be enriched and detected in a similar manner, thereby enabling subsequent single-cell analysis. For example, using cancer cell membranes to coat nanoparticles can enrich immune cells in the bloodstream that have binding affinity to cancer cells. Cancer can then be predicted by analyzing the enriched immune cells. In another example, using bacterial membranes to coat nanoparticles can identify immune cells or human host cells that have specific binding affinity to pathogens. Therefore, the technical embodiments disclosed in the present invention can also promote the development of early detection and prevention or treatment technologies for cancer and other human diseases.

[0085] Although this patent document contains many details, these details should not be viewed as limitations on the scope of any invention or its claims, but rather as descriptions of features that may be specific to a particular embodiment of a particular invention. Certain features described in the context of separate embodiments of this patent document may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented separately in multiple embodiments, or in any suitable subcombination. In addition, although features are described above as working in certain combinations, or even initially claimed as such, in some cases, one or more features in the claimed combination may be deleted from the combination, and the claimed combination may refer to a subcombination or a variation of a subcombination.

[0086] Similarly, although the operations are described in a particular order in the drawings, this should not be understood as requiring that the operations be performed in the particular order shown or in sequence, or that all the operations shown be performed to achieve the desired results. In addition, the separation of the system components in the embodiments described in this patent document should not be understood as requiring the separation in all embodiments.

[0087] Only some implementations and examples are described herein, and other implementations, enhancements, and variations may be used as described and illustrated in this patent document.

[0088] References:

[0089] 1. Pantel, K. and Speicher, M. R. Biology of circulating tumor cells. Oncogene 35, 1216-1224 (2016).

[0090] 2. Micalizzi, D.S., Maheswaran, S., and Haber, D.A. The metastatic duct: the biology of circulating tumor cells. Genes & Development 31, 1827-1840 (2017).

[0091] 3. Lin, E., Cao, T., Nagrath, S., and King, M. R. Circulating tumor cells: diagnostic and therapeutic applications. Annual Review of Biomedical Engineering 20, 329-352 (2018).

[0092] 4. Rack, B. et al. Circulating tumor cells predict survival in patients with early-stage, intermediate- and high-risk breast cancer. Journal of the National Cancer Institute 106 (2014).

[0093] 5. Ilie, M. et al. “Sentinel” circulating tumor cells allow early lung cancer diagnosis in patients with chronic obstructive pulmonary disease. PLoS One 9, e111597 (2014).

[0094] 6. Shaw Bagnall, J. et al. Variability of tumor cells and blood cells. Scientific Reports 5, 18542 (2015).

[0095] 7. Liu, HY et al. Improved method to increase sensitivity to circulating hepatoma cells. World Journal of Gastroenterology 21, 2918-2925 (2015).

[0096] 8. Reid, A. L., Freeman, J. B., Millward, M., Ziman, M., and Gray, E. S. Detection of BRAF-V600E and V600K in circulating tumor cells of melanoma by droplet digital PCR. Clinical Biochemistry 48, 999-1002 (2015).

[0097] 9. Hou, H. W. et al. Isolation and recovery of circulating tumor cells using centrifugal force. Scientific Reports 3, 1259 (2013).

[0098] 10. Allard, WJ et al. Tumor cells circulate in the peripheral blood of all major cancers but not in healthy subjects or patients with non-malignant diseases. Clinical Cancer Research 10, 6897-6904 (2004).

[0099] 11. Ozkumur, E. et al. Inertial focusing of tumor antigen-dependent and -independent classification of rare circulating tumor cells. Science Translational Medicine 5, 179ra147 (2013).

[0100] 12. Karabacak, N.M. et al. Isolation of circulating tumor cells from blood samples in a microfluidic, marker-free format. Handbook of Experiments in Nature 9, 694-710 (2014).

[0101] 13. Yu, M. et al. Circulating breast tumor cells undergo dynamic changes in epithelial and mesenchymal composition. Science 339, 580-584 (2013).

[0102] 14. Nieswandt, B., Hafner, M., Echtenacher, B., and Mannel, DN. Tumor cell lysis by natural killer cells in mice is impeded by platelets. Cancer Research 59, 1295-1300 (1999).

[0103] 15. Velez, J. et al. Platelets promote mitochondrial uncoupling and anti-apoptosis in leukemic cells: a new paradigm for the bone marrow microenvironment. Cancer Microenvironment 7, 79-90 (2014).

[0104] 16. Franco, A. T., Corken, A., and Ware, J. Platelets at the interface of thrombosis, inflammation, and cancer. Blood 126, 582-588 (2015).

[0105] 17. Powell, AA et al. Single-cell analysis of circulating tumor cells: transcriptional heterogeneity and diversity in breast cancer cell lines. PLoS One 7, e33788 (2012).

[0106] 18. Ramskold, D. et al., RNA single-cell level and full-length mRNA-Seq of individual circulating tumor cells. Nature Biotechnology 30, 777-782 (2012).

[0107] 19. Miyamoto, DT, et al., RNA-Seq of single prostate CTCs implicates non-canonical Wnt signaling in anti-androgen resistance. Science 349, 1351-1356 (2015).

[0108] 20. Fan, S. et al. Integrative analysis using expanded DNA methylation data reveals key regulators and pathways common in cancer. Genomic Medicine 4, 2 (2019).

[0109] 21. Whitaker, J. W., et al. Comprehensive omics analysis of rheumatoid arthritis identifies nonobvious therapeutic targets. PLoS One 10, e0124254 (2015).

[0110] 22. Fan, S., Li, C., Ai, R., Firestein, G. S., and Wang, W. Computationally extending InfiniumHumanMethylation450 BeadChip array data to reveal distinct DNA methylation patterns in rheumatoid arthritis. Bioinformatics (2016).

[0111] 23. Hu, C.-MJ et al., Nanoparticle biointerfaces achieved through platelet membrane coating. Nature 526, 118-121 (2015).

[0112] 24. Felding-Habermann, B. et al., Integrin activation controls metastasis in human breast cancer. Proceedings of the National Academy of Sciences of the United States of America, 98, 1853-1858 (2001).

[0113] 25. Zara, M. et al. Molecular mechanisms of breast cancer cell-induced platelet activation and aggregation. Cell Signaling 48, 45-53 (2018).

[0114] 26. Labelle, M., Begum, S., and Hynes, R. O. Direct signaling between platelets and cancer cells induces epithelial-mesenchymal transition and promotes metastasis. Cancer Cell 20, 576-590 (2011).

[0115] 27. Bambace, NM and Holmes, CE. The role of platelets in cancer progression. Journal of Thrombosis and Hemostasis 9, 237-249 (2011).

[0116] 28. Nobili, M., Sheriff, J., Morbiducci, U., Redaelli, A., and Bluestein, D. Platelet activation due to hemodynamic shear stress: an injury accumulation model and comparison with in vitro measurements. Journal of the American Society of Artificial Organs 54, 64-72 (2008).

[0117] 29. Plantureux, L. et al. Effects of cancer on platelet production, activation and education and mechanisms of cancer-related thrombosis. Cancer (Basel), 10, 441 (2018).

[0118] 30. Stefanini, L. and Bergmeier, W. Negative regulators of platelet activation and adhesion. Journal of Thrombosis and Hemostasis 16, 220-230 (2018).

[0119] 31. Estevez, B. and Du, X. P. New concepts and mechanisms of platelet activation signaling. Physiology 32, 162-177 (2017).

[0120] 32. Shi, L. et al. Predictable resistance to gemcitabine / cisplatin and overall survival by platelet activation markers in non-small cell lung cancer. Med Sci Maxim 24, 8655-8668 (2018).

[0121] 33. Yang, J., Furie, BC, and Furie, B. The biology of P-selectin glycoprotein ligand 1: its role as a selectin counter-receptor in leukocyte-endothelial and leukocyte-platelet interactions. Thrombosis and Hemostasis 81, 1-7 (1999).

[0122] 34. Burdorf, L., Riner, A., and Rybak, E. Platelet sequestration and activation during perfusion of GalTKO.hCD46 pig lungs with human blood is mediated by GPIb, GPIIb / IIIa, and von Willebrand factor (Vol. 23, p. 222, 2016). Xenotransplantation 23, 423-423 (2016).

[0123] 35. Bruserud, O. Agkistrodon acutus venom proteins - a clinically important toxin and a useful experimental tool for the study of C-type lectin-like receptor 2 (CLEC-2). Toxins (Basel), 5, 665-674 (2013).

[0124] 36. Luk, BT, et al. Interfacial interactions between native RBC membranes and synthetic polymer nanoparticles. Nanoscale 6, 2730-2737 (2014).

[0125] 37. Hu, C. M. J. et al. Self-marker functionalization of nanoscale particles via a top-down cell membrane coating approach. Nanoscale, 5, 2664-2668 (2013).

[0126] 38. Rosenberg, R. et al. Comparison of two density gradient centrifugation systems for enrichment of disseminated tumor cells in blood. Cytometry 49, 150-158 (2002).

[0127] 39. Lagoudianakis, EE et al. Detection of epithelial cells by RT-PCR targeting CEA, CK20, and TEM-8 in patients with colorectal cancer using the OncoQuick density gradient centrifugation system. Journal of Surgical Research 155, 183-190 (2009).

[0128] 40. Sharma, S. et al. Circulating tumor cell isolation, culture, and downstream molecular analysis. Advances in Biotechnology 36, 1063-1078 (2018).

[0129] 41. Colella, R., Jackson, T., and Goodwyn, E. Matrigel (R) invasion and cathepsin L+B activity of prostate cancer cell lines PC3 and DU145. Biotechnology and Histochemistry 79, 121-127 (2004).

[0130] 42. Kobayashi, M. et al. Ovarian cancer cell invasiveness is associated with discordant exosome sequestration of Let-7 miRNA and miR-200. Journal of Translational Medicine 12, 4 (2014).

[0131] 43. Francia, G., Cruz-Munoz, W., Man, S., Xu, P., and Kerbel, R. S. A mouse model of advanced spontaneous metastasis for experimental treatment. Nature Reviews Cancer 11, 135-141 (2011).

[0132] 44. Tripathi, S., Belkacemi, L., Cheung, MS, and Bose, RN. Correlations between genetic variants, signaling pathways, and efficacy of chemotherapeutic agents in colon cancer. Cancer Information 15, 1-13 (2016).

[0133] 45. Park, KC et al. Genetic changes in the transforming growth factor beta (TGF-β) type II receptor gene in human gastric cancer cells: correlation with sensitivity to growth inhibition by TGF-β. Proceedings of the National Academy of Sciences of the United States of America, 91, 8772-8776 (1994).

[0134] 46. ​​Kennecke, H. et al. Metastatic behavior of breast cancer subtypes. Journal of Clinical Oncology 28, 3271-3277 (2010).

[0135] 47. Feldman, BJ and Feldman, D. Development of androgen-independent prostate cancer. Nature Reviews Cancer 1, 34-45 (2001).

[0136] 48. Sharpless, NE and DePinho, RA. Model organisms - the rat: a genetically engineered mouse model for anticancer drug development. Nature Reviews Drug Discovery 5, 741-754 (2006).

[0137] 49. Esteva-Font, C., Jin, B.J., and Verkman, A.S. Genetic deletion of aquaporin 1 slows mammary tumor growth and lung metastasis in tumorigenic MMTV-PyVT mice. FASEB Journal 28, 1446-1453 (2014).

[0138] 50. Moser, AR, Pitot, HC and Dove, WF. Dominant mutations that predispose to multiple intestinal tumors in mice. Science 247, 322-324 (1990).

[0139] 51. Minn, AJ, et al. Lung metastasis genes are associated with breast tumor size and metastatic spread. Proceedings of the National Academy of Sciences of the United States of America, 104, 6740-6745 (2007).

[0140] 52. Grabowska, MM, et al. Mouse models of prostate cancer: choosing the best model for the problem. Cancer and Metastasis Reviews 33, 377-397 (2014).

[0141] 53. Picelli, S. et al. Single-cell sensitive full-length transcriptome analysis using Smart-seq2. Nature Methods 10, 1096-1098 (2013).

[0142] 54. Picelli, S. et al. Single-cell full-length RNA-seq using Smart-seq2. Nature Experiment Manual 9, 171-181 (2014).

[0143] 55. Ai, R. et al. Comprehensive epigenetic landscape of rheumatoid arthritis fibroblast-like synoviocytes. Nature Communications 9, 1921 (2018).

[0144] 56. Fan, S., Huang, K., Ai, R., Wang, M., and Wang, W. Prediction of CpG methylation levels by integrating InfiniumHumanMethylation450 BeadChip array data. Genomics (2016).

[0145] 57. Consortium, TG. Genotype-Tissue Expression (GTEx) Project. Nature Genetics 45, 580-585 (2013).

[0146] 58. Grossman, R.L. et al. A shared vision for cancer genomic data. New England Journal of Medicine 375, 1109–1112 (2016).

[0147] 59. Macosko, E. Z. et al. Highly parallel genome-wide expression profiling of single cells using nanoliter droplets. Cell 161, 1202-1214 (2015).

[0148] 60. Lake, B.B., et al. Neuronal subtypes and diversity revealed by single-nucleus RNA sequencing of the human brain. Science 352, 1586–1590 (2016).

[0149] 61. Sarwar, S. S., Ankit, A., and Roy, K. Incremental learning in deep convolutional neural networks via partial network sharing. arXiv, 1712.02719 (2017).

[0150] 62. Roy, D., Panda, P., and Roy, K. Tree-CNN: Hierarchical Deep Convolutional Neural Networks for Incremental Learning. arXiv, 1802.05800 (2018).

[0151] 63. Liu, Y. et al. Comparative molecular analysis of gastrointestinal adenocarcinoma. Cancer Cell 33, 721–735e728 (2018).

[0152] 64. Berger, A.C. et al. Comprehensive pan-cancer profiling of gynecologic and breast cancers. Cancer Cell 33, 690–705e699 (2018).

[0153] 65. Ricketts, C. J. et al. The Cancer Genome Atlas: An Integrated Molecular Characterization of Renal Cell Carcinoma. Cell Reports 23, 3698 (2018).

[0154] 66. Campbell, JD et al. Genomic, pathway network, and immunological signatures that distinguish squamous cell carcinoma. Cell Reports 23, 194–212e196 (2018).

[0155] 67. Ding, B. et al. Normalization and noise reduction of single-cell RNA-seq experiments. Bioinformatics 31, 2225-2227 (2015).

[0156] 68. Ding, B., Zheng, L. and Wang, W. Evaluation of normalization methods for single-cell RNA-Seq. G3 (Bethesda) 7, 2039-2045 (2017).

[0157] 69. Liu, C. and Wang, W. Contextual regression: an accurate and easily interpretable nonlinear model for discovery in scientific data mining. arXiv, 1710.10728 [q-bio.QM] (2017).

[0158] 70. Minn, AJ et al. Genes mediating lung metastasis of breast cancer. Nature 436, 518-524 (2005).

[0159] 71. Yang, S., Zhang, J. J., and Huang, X. Y. Mouse model of tumor metastasis. Methods in Molecular Biology, 928, 221-228 (2012).

Claims

1. Use of one or more nanoparticles in the preparation of a composition for enriching circulating tumor cells in a human biological sample, wherein the nanoparticles are coated with platelet membranes from humans, The enrichment comprises contacting the biological sample with the one or more nanoparticles, wherein the nanoparticles comprise an inner core supporting an outer surface of the nanoparticles coated with the platelet membrane, and wherein the biological sample is blood.

2. The use according to claim 1, wherein The nanoparticles do not contain antibodies that recognize specific antigens on circulating tumor cells.

3. The use according to claim 1, wherein The circulating tumor cells are breast cancer, lung cancer, ovarian cancer or gastric cancer circulating tumor cells.

4. The use according to claim 1, wherein The nanoparticles comprise a poly(lactic-co-glycolic acid) core.

5. The use according to claim 1, wherein the enrichment further comprises at least partially separating the nanoparticles bound to the target object from the biological sample.

6. Use of nanoparticles in the preparation of a composition for enriching and genetically identifying circulating tumor cells in a human biological sample, wherein the nanoparticles are coated with a platelet membrane from a human, The enrichment and gene identification include: enriching the circulating tumor cells by contacting the biological sample with the nanoparticles; and performing single-cell nucleic acid sequencing on the circulating tumor cells, The nanoparticle comprises an inner core, and the inner core supports an outer surface of the nanoparticle coated with the platelet membrane, and the biological sample is blood.

7. The use according to claim 6, wherein The nanoparticles do not contain antibodies that recognize specific antigens on circulating tumor cells.

8. The use according to claim 6, wherein The circulating tumor cells are breast cancer, lung cancer, ovarian cancer or gastric cancer circulating tumor cells.

9. The use according to claim 6, wherein The nanoparticles are platelet membrane-coated nanoparticles.

10. The use according to claim 6, wherein: Using triplet of proven genes to differentiate cancer samples from non-cancerous cells.

11. The use according to claim 6, wherein: The single-cell nucleic acid sequencing is single-cell RNA sequencing.

12. The use according to claim 6, wherein the enrichment and gene identification further comprises at least partially separating the nanoparticles bound to the target object from the biological sample before performing single-cell nucleic acid sequencing.

13. Use of platelet membrane-coated nanoparticles in the preparation of a composition for identifying cancer types or subtypes of circulating tumor cells in human biological samples, The identification includes: enriching circulating tumor cells by contacting the biological sample with the platelet membrane-coated nanoparticles; Single-cell RNA sequencing of enriched circulating tumor cells; as well as determining the cancer type or subtype of circulating tumor cells based on single cell RNA sequencing and comparison of multiple genes whose expression profiles are associated with different cancer types or subtypes, The nanoparticle comprises an inner core, and the inner core supports an outer surface of the nanoparticle coated with the platelet membrane, and the platelet membrane is from a human and the biological sample is blood.

14. The use according to claim 13, wherein The cancer type is breast cancer, lung cancer, ovarian cancer or gastric cancer.

15. The use according to claim 13, wherein The nanoparticles do not contain antibodies that recognize specific antigens on circulating tumor cells.

16. A composition enriched for circulating tumor cells, comprising circulating tumor cells at least partially isolated from a biological sample of a human patient, wherein: The circulating tumor cells are bound to platelet membrane-coated nanoparticles, wherein the nanoparticles comprise an inner core supporting an outer surface of the nanoparticle coated with the platelet membrane, and wherein the platelet membrane is from a human and the biological sample is blood.

17. The composition according to claim 16, wherein The circulating tumor cells are breast cancer, lung cancer, ovarian cancer or gastric cancer circulating tumor cells.

18. The composition according to claim 16, wherein The nanoparticles do not contain antibodies that recognize specific antigens on circulating tumor cells.

Images