Microvessel-based oral mucosa canceration process prediction method, device and program product

Abstract

The invention relates to the field of intelligent medical treatment, in particular to an oral mucosa canceration process prediction method and device based on microvessels and a program product. Comprising the following steps: S1, acquiring a pathological image of oral mucosa tissue of a to-be-tested person; s2, obtaining morphological characteristics of microvessels based on the pathological image; the morphological characteristics comprise a fusion type, a budding type and a finger-shaped propelling type; s3, the oral mucosa canceration process is judged based on the morphological characteristics, and when the morphological characteristics of the microvessels are fusion type and budding type, the oral mucosa canceration process is oral leukoplakia accompanied by abnormal epithelial hyperplasia or highly differentiated oral squamous cell carcinoma; when the morphological characteristics of the microvessels are fusion type, budding type and finger-shaped propelling type, the canceration process of the oral mucosa is middle-differentiated oral squamous cell carcinoma; when the morphological characteristic of the microvessels is a finger-shaped propulsion type, the cancerization process of the oral mucosa is low-differentiation oral squamous cell carcinoma. The application has good clinical value.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical care, and specifically to a method, device, program product and computer-readable storage medium for predicting the progression of oral mucosal carcinogenesis based on microvessels. Background Art

[0002] In the early stages of oral mucosal carcinogenesis, cells begin to proliferate abnormally. At this time, microvascular mimicry (VM) may have begun to undergo some subtle changes. Normal oral mucosa has its own vascular supply system, and in the precancerous stage, such as oral leukoplakia or the early stage of lichen planus malignancy, tumor cells may induce changes in endothelial cells and surrounding stromal cells by secreting certain angiogenic factors. These changes may promote the initial formation of vascular mimicry and provide early nutritional support for tumor cells. As oral mucosal carcinogenesis progresses, from carcinoma in situ to invasive cancer, the formation of microvascular mimicry becomes more obvious. In order to meet the needs of their rapid growth and invasion, tumor cells will further transform the surrounding vascular environment. At this time, microvascular mimicry not only increases in number, but also becomes more complex in structure. Tumor cells can arrange themselves into vascular-like channels, which can transport blood, provide oxygen and nutrients for tumor growth, and also help tumor cells excrete metabolic waste. This vessel-like channel may be similar to normal blood vessels in morphology, but there are differences in function and structure. It does not have a complete vascular endothelial cell layer, but is composed of tumor cells or tumor cells and endothelial cells. However, the current imaging technology used to observe microvascular mimicry has certain limitations. For example, although traditional angiography technology can display the general morphology of blood vessels, its resolution may not be high enough for complex and delicate vascular structures such as microvascular mimicry. In relatively small and complex tissues such as oral mucosa, it is difficult to clearly distinguish normal blood vessels and microvascular mimicry, and it is impossible to accurately observe the dynamic changes of microvascular mimicry in the process of carcinogenesis. Summary of the invention

[0003] In view of the above problems, the present invention provides a method for predicting the progression of oral mucosal carcinogenesis based on microvessels, which specifically includes: S1 obtains the pathological image of the oral mucosal tissue of the subject; S2 obtains the morphological characteristics of the microvessels based on the pathological image; the morphological characteristics include fusion type, sprouting type, and finger-like propulsion type; the fusion type is that the cross-section of the circular microvessel lumen is partially fused, the sprouting type is that the microvessel lumen sprouts to form a vascular branch, and the finger-like propulsion type is that the longitudinal section of the microvessel lumen is clustered and arranged in parallel; S3 judges the progression of oral mucosal carcinoma based on the morphological characteristics. When the microvascular morphological characteristics are fusion type and budding type, the progression of oral mucosal carcinoma is oral leukoplakia with epithelial abnormal proliferation or well-differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are fusion type, budding type and finger-like advancement type, the progression of oral mucosal carcinoma is moderately differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are finger-like advancement type, the progression of oral mucosal carcinoma is poorly differentiated oral squamous cell carcinoma.

[0004] The step S1 is replaced by: obtaining oral mucosal tissue of a subject to be tested; the step S2 is replaced by: detecting the expression level of the ANGPTL2 gene based on the oral mucosal tissue; the step S3 is replaced by: when the expression level of the ANGPTL2 gene is greater than a first preset threshold, determining the patient to be a well-differentiated oral squamous cell carcinoma; Optionally, the first preset threshold is the ANGPTL2 gene expression level in normal mucosal tissue; Optionally, S2 is replaced by: detecting the CK15 gene expression level based on the oral mucosal tissue; S3 is replaced by: when the CK15 gene expression level is less than a second preset threshold, determining it as well-differentiated oral squamous cell carcinoma, moderately differentiated oral squamous cell carcinoma, or poorly differentiated oral squamous cell carcinoma; Optionally, the second preset threshold is the CK15 gene expression level of normal mucosal tissue; Optionally, S2 is replaced by: detecting the expression level of EEF1E1 gene based on the oral mucosal tissue; S3 is replaced by: when the expression level of EEF1E1 gene is greater than a third preset threshold, determining it as moderately differentiated oral squamous cell carcinoma; Optionally, the third preset threshold is the EEF1E1 gene expression level of normal mucosal tissue; Optionally, S2 is replaced by: detecting the expression of ANGPTL2 and CK15 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression of ANGPTL2 and CK15 genes increases, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression of ANGPTL2 gene decreases and the expression of CK15 gene increases, it is determined to be moderately differentiated oral squamous cell carcinoma or poorly differentiated oral squamous cell carcinoma; Optionally, S2 is replaced by: detecting the expression of ANGPTL2 and EEF1E1 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression of ANGPTL2 and CK15 genes increases, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression of ANGPTL2 gene decreases and the expression of EEF1E1 gene increases, it is determined to be moderately differentiated oral squamous cell carcinoma; Optionally, S2 is replaced by: detecting the expression of CK15 and EEF1E1 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression of CK15 and EEF1E1 genes increases, it is determined to be well-differentiated oral squamous cell carcinoma or moderately differentiated oral squamous cell carcinoma; when the expression of CK15 gene increases and the expression of EEF1E1 gene decreases, it is determined to be poorly differentiated oral squamous cell carcinoma; Optionally, S2 is replaced by: detecting the expression levels of ANGPTL2, CK15, and EEF1E1 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression levels of ANGPTL2, CK15, and EEF1E1 genes increase, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression level of ANGPTL2 gene decreases and the expression levels of CK15 and EEF1E1 genes increase, it is determined to be moderately differentiated oral squamous cell carcinoma; when the expression levels of ANGPTL2 and EEF1E1 genes decrease and the expression level of CK15 gene increases, it is determined to be poorly differentiated oral squamous cell carcinoma.

[0005] The S2 is replaced by: inputting the pathological image into the prediction model to obtain the prediction result of the oral mucosal carcinogenesis process, and removing S3; wherein the prediction model is obtained by training the pathological images of the oral mucosal carcinogenesis process at different stages through a neural network; optionally, the neural network includes one or more of the following: convolutional neural network, residual network, Transformer.

[0006] The embodiment of the present invention provides a method for predicting the progression of oral mucosal carcinogenesis based on MAPH, comprising: Obtaining the oral mucosal microvascular data of the test subject, including microvascular morphological characteristics, structural characteristics, and spatial characteristics; The microvascular data is input into the MAPH model to obtain the process of oral mucosal carcinogenesis; the MAPH model is composed of the microvascular morphological characteristics, structural characteristics, and spatial characteristics of different stages of oral mucosal carcinogenesis, and the process includes oral leukoplakia with epithelial abnormal proliferation, well-differentiated oral squamous cell carcinoma, moderately differentiated oral squamous cell carcinoma, and poorly differentiated oral squamous cell carcinoma.

[0007] The microvascular morphological characteristics of the MAPH model include: oral leukoplakia with epithelial dysplasia are fusion type and budding type, well-differentiated oral squamous cell carcinoma are fusion type and budding type, moderately differentiated oral squamous cell carcinoma are fusion type, budding type and finger-like advancement type, and poorly differentiated oral squamous cell carcinoma is finger-like advancement type; Optionally, the structural features include blood vessels mimicking tumor cell linings; Optionally, the spatial feature is that the microvessels are mainly distributed at the tumor invasion front and advance in a finger-like manner.

[0008] The object of the present invention is to provide a computer program product, which includes a computer program or instructions, and the computer program or instructions are executed by a processor to implement the above-mentioned microvascular-based oral mucosal cancer progression prediction method or MAPH-based oral mucosal cancer progression prediction method.

[0009] The object of the present invention is to provide a computer device, which includes a memory, a processor and a computer program or instructions stored in the memory, wherein the computer program or instructions are executed by the processor to implement the above-mentioned microvascular-based oral mucosal cancer progression prediction method or MAPH-based oral mucosal cancer progression prediction method.

[0010] The object of the present invention is to provide a computer-readable storage medium having a computer program or instruction stored thereon, wherein the computer program or instruction is executed by a processor to implement the above-mentioned microvascular-based oral mucosal cancer progression prediction method or MAPH-based oral mucosal cancer progression prediction method.

[0011] The object of the present invention is to provide a TSCs-targeting substance for use in the preparation of a drug for anti-angiogenic treatment of oral mucosal carcinogenesis; optionally, the TSCs-targeting substance acts on the recruitment of TSCs.

[0012] Use of an agent for inhibiting genes or overexpressing genes in the preparation of a drug for anti-angiogenic treatment of oral mucosal carcinogenesis, wherein the gene includes ANGPTL2; optionally, the ANGPTL2 is replaced by CK15; optionally, the ANGPTL2 is replaced by EEF1E1; optionally, the gene also includes one or more of the following: CK15, EEF1E1, TIMP1; optionally, the ANGPTL2 is replaced by TIMP1; optionally, the drug is used to regulate the mediation effect between TIMP1 and MMPs; the MMPs include one or more of the following: MMP1, MMP2, MMP3, MMP9, MMP10, MMP14.

[0013] Advantages of the present invention: 1. Predict different stages of oral mucosal carcinogenesis based on the morphological characteristics of microvessels, including fusion type, budding type, and finger-like propulsion type. Based on different morphological characteristics, doctors can evaluate the stage of oral mucosal carcinogenesis based on the morphological characteristics of microvessels, providing basic conditions for the later treatment of oral mucosal carcinogenesis, which has great clinical value.

[0014] 2. A MAPH related to the progression of oral mucosal carcinogenesis is proposed. The progression of oral mucosal carcinogenesis is predicted by the microvascular morphological characteristics, microvascular structural characteristics, and microvascular spatial characteristics in MAPH, so that doctors can conduct targeted later diagnosis and treatment.

[0015] 3. ANGPTL2, CK15, EEF1E1, and TIMP1 show different gene expression levels at different stages of oral mucosal carcinogenesis. Regulating the gene expression levels of ANGPTL2, CK15, EEF1E1, and TIMP1 can help diagnose the development stage of oral mucosal cancer, control the development of oral mucosal cancer, and reduce adverse prognostic reactions.

[0016] 4. TSCs interfere with the process of oral mucosal carcinogenesis and induce characteristic changes in microvascular morphological heterogeneity, promoting tumor growth. By targeting TSCs, the development of oral mucosal cancer can be inhibited and the treatment of oral mucosal cancer can be assisted. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic diagram of a process flow of a method for predicting the progression of oral mucosal carcinogenesis based on microvessels provided in an embodiment of the present invention; Figure 2 A schematic diagram of a microvascular-based oral mucosal cancer progression prediction system provided in an embodiment of the present invention; Figure 3 A schematic diagram of a microvascular-based oral mucosal cancer progression prediction device provided in an embodiment of the present invention; Figure 4 The microvascular classification diagram of the microvascular architecture morphology in the process of oral mucosal carcinogenesis provided by the embodiment of the present invention; (A1) budding type; (B1) fusion type; (C1) terminal expansion type; (D1) finger-like propulsion type; (E1) crisscross type. (HE, ×200) (A2) budding type; (B2) fusion type; (C2) terminal expansion type; (D2) finger-like propulsion type; (E2) crisscross type; Figure 5 The following are the analysis diagrams of the characteristic microvascular architecture morphology at each stage of the oral mucosal carcinogenesis process provided by the embodiments of the present invention; (A) There is no characteristic microvascular architecture morphology in NOM; (B) Fusion type and budding type are characteristic microvascular architecture morphologies in OLK ED; (C) Fusion type and budding type are characteristic microvascular architecture morphologies in OSCC HD; (D) Fusion type, budding type and finger-like propulsion type are characteristic microvascular architecture morphologies in OSCC MD; (E) Finger-like propulsion type is a characteristic microvascular architecture morphology in OSCC PD; Figure 6The microvascular architecture morphology classification diagram of the oral mucosal carcinomatous tumor-bearing tissue provided in the embodiment of the present invention; (A) budding type; (B) fusion type; (C) terminal expansion type; (D) finger-like propulsion type; (E) crisscross type; Figure 7 The microvascular architecture morphology analysis diagram of the tumor-bearing tissue in the process of oral mucosal carcinogenesis provided by the embodiment of the present invention; (A) There is no characteristic microvascular architecture morphology in NOM; (B) Fusion type and budding type are characteristic microvascular architecture morphologies in PT; (C) Fusion type and budding type are characteristic microvascular architecture morphologies in SCC HD; (D) Fusion type, budding type and finger-like advancement type are characteristic microvascular architecture morphologies in SCC MD; (E) Finger-like advancement type is a characteristic microvascular architecture morphology in SCC PD; Figure 8 A microvessel distribution map of the tumor-bearing tissue in the process of oral mucosal carcinogenesis provided by an embodiment of the present invention; Fig. 9 TIMP1 immunohistochemical staining of various stages of oral mucosal carcinogenesis provided by the embodiment of the present invention; Fig.10 Differential analysis of TIMP1 expression at various stages of oral mucosal carcinogenesis provided by the embodiments of the present invention; Fig.11 The expression difference diagram of MMPs in the process of oral mucosal carcinogenesis provided by the embodiment of the present invention; Fig.12 Analysis of differential expression of MMPs in the process of oral mucosal carcinogenesis provided by the embodiments of the present invention; Fig.13 The immunoprecipitation results provided in the embodiments of the present invention are shown.

[0019] Fig.14 The VM in the process of oral mucosal carcinogenesis provided by the embodiment of the present invention, (A) tumor cells invade into microvessels, the integrity of vascular endothelium is destroyed, and tumor cells around blood vessels tend to migrate into blood vessels; (B) tumor cells surround and invade microvessels, the structure of luminal red blood cells is destroyed, and a pink homogeneous structure appears; (C) tumor cells recruit and invade microvessels, the vascular endothelium disappears, and the morphology changes; (D) the normal morphology of microvessels disappears, and red blood cells appear as a pink homogeneous structure, with unclear boundaries with surrounding tumor tissues. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0021] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as S101, S102, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0022] Figure 1 The schematic diagram of the method for predicting the progression of oral mucosal carcinogenesis based on microvessels provided in the embodiment of the present invention specifically includes: S1: Obtain pathological images of the oral mucosal tissue of the subject; In a specific embodiment, human gingival epithelial cells (HGECs) are derived from healthy human gingival tissue, cultured and amplified by tissue block method, and after identification, the 3rd to 6th generation HGECs in good condition are used for subsequent experiments. Human oral squamous cell carcinoma (OSCC) cell line HSC44 was purchased from Shenzhen Haodi Huatuo Biotechnology Co., Ltd.

[0023] In a specific embodiment, the normal oral mucosa (NOM) tissue, oral leukoplakia with epithelial dysplasia (OLK ED) tissue, highly differentiated oral squamous cell carcinoma (OSCC HD) tissue, moderately differentiated oral squamous cell carcinoma (OSCC MD) tissue, and poorly differentiated oral squamous cell carcinoma (OSCC PD) fresh tissue are the remaining tissues after pathological examination of surgically removed lesions from outpatients and inpatients of the Affiliated Stomatological Hospital of Southwest Medical University.

[0024] In a specific embodiment, clinical samples of oral mucosal carcinogenesis were grouped and identified: hematoxylin-eosin (HE) staining: paraffin sections were placed in an oven for 30 min until the paraffin on the sections melted, and the following steps were performed in sequence: xylene (I) for 12 min, xylene (II) for 12 min, 100% ethanol for 2 min, 95% ethanol for 2 min, 80% ethanol for 2 min, tap water rinse, hematoxylin staining for 5 min, tap water rinse, 1% hydrochloric acid ethanol for 1 s, tap water rinse, 1% ammonia water for 1 min, tap water rinse, eosin staining for 1.5 min, tap water rinse, 80% ethanol for 2 s, 95% ethanol for 2 s, 100% ethanol for 10 min, xylene for 10 min, and neutral gum sealing. After HE staining, experienced pathologists performed Physicians performed pathological diagnosis on the tissues and divided the embedded wax blocks into NOM group, OLKED group, OSCC HD group, OSCC MD group, and OSCC PD group according to the results of pathological diagnosis.

[0025] In a specific embodiment, the construction and material collection of the oral mucosal carcinogenesis disease animal model: 4-w-old SPF-grade BALB / C male nude mice were purchased and kept in the SPF-grade animal room of the Experimental Animal Center of Southwest Medical University. After being sent to the animal room, the state of the nude mice was observed and the vitality was good. 80 male nude mice were randomly divided into 5 groups, 16 in each group. After 1 w of adaptive feeding, logarithmic phase HSC4 cells and HGECs cells were taken, digested and centrifuged to make single cell suspensions, and cell counts and viability were measured, and the cell concentration was adjusted to 8×106 / mL. The nude mice in the normal oral mucosa (NOM) group were subcutaneously injected with HGECs cell suspension in the right upper limb, 0.2 mL per mouse. In the precancerous tissue (PT) group, HGECs cell suspension was subcutaneously injected into the right upper limb of nude mice, 0.2 mL per mouse. After 2 weeks, a stable epithelial-like structure was formed, and then HSC4 cells were injected into the epithelial-like structure, 0.2 mL per mouse. In the highly differentiated squamous cell carcinoma (SCC HD) group, moderately differentiated squamous cell carcinoma (SCC MD) group, and poorly differentiated squamous cell carcinoma (SCC PD) group, HSC4 cell suspension was subcutaneously injected into the right upper limb of nude mice, 0.2 mL per mouse. The nude mice in the NOM and PT groups were anesthetized 14 days after subcutaneous injection, and the nude mice in the SCC HD, SCC MD, and SCC PD groups were anesthetized 28, 42, and 56 days after subcutaneous injection, and the subcutaneous tumor-bearing tissue was removed. The subcutaneous tissue of the right upper limb was removed in the NOM group. The fresh tissue was cut into two parts with the largest section, one part was numbered and stored in a -120℃ ultra-low temperature refrigerator, and the other part was immediately fixed in 10% neutral formalin, fixed and embedded, paraffin sectioned, and HE stained. Two experienced pathological clinical workers were independently identified under the condition of blinding the construction process to determine the pathological stage of the oral mucosal carcinogenesis disease animal model. (The normal anesthetic dose in the experiment was 40 mg / kg).

[0026] In a specific embodiment, microvascular typing analysis of an animal model of oral mucosal carcinogenesis disease: tumor-bearing sections of an animal model of oral mucosal carcinogenesis disease are sliced ​​and HE-stained to observe microvascular morphological characteristics, determine the classic vascular morphology, and calculate the vascular morphology composition ratio (n=10).

[0027] Observation of microvascular structure in animal model of oral mucosal carcinogenesis: Animal model of oral mucosal carcinogenesis was established, with 3 animals in each group. After sodium pentobarbital anesthesia, the chest cavity was exposed, the right atrial appendage was cut, and sodium heparin solution was perfused through the left ventricle until colorless liquid flowed out of the right atrial appendage. 3 mL of ink gelatin aqueous solution at 37°C was quickly perfused through the left ventricle until the tumor was uniformly dark black and black liquid flowed out of the right atrial appendage. The perfusion was stopped and the tumor was transferred to ice. After 20 minutes, the tumor was peeled off and fixed in 10% neutral formalin solution. After 48 hours, the tissue blocks were rinsed with running water, dehydrated with gradient alcohol, and transparentized with wintergreen oil. After the tissue was completely transparent, the microvascular branching and microvascular diameter changes of the NOM group, PT group, SCC HD group, SCC MD group, and SCC PD group were observed under a stereomicroscope.

[0028] In a specific embodiment, the results of the classification of microvascular architecture in the process of oral mucosal carcinogenesis: after identification, the clinical sample meets NOM, OLK ED, OSCC HD, OSCC MD, and OSCC PD. There are 5 characteristic microvascular architecture morphologies in the process of oral mucosal carcinogenesis, namely, fusion type: the cross-section of the circular microvascular lumen is partially fused; budding type: the microvascular lumen sprouts to form vascular branches; finger-like propulsion type: the longitudinal section of the microvascular lumen is parallel and clustered; crisscross type: the longitudinal section of the microvascular lumen is closely arranged with the cross-section; terminal expansion type: the longitudinal section of the microvascular lumen is partially expanded, such as Figure 4 shown.

[0029] Statistical analysis results of microvascular architecture morphology in the process of oral mucosal carcinogenesis: The characteristic microvascular architecture morphology of each pathological stage of NOM, OLK ED, OSCC HD, OSCC MD, and OSCC PD was obtained by one-way analysis of variance. Among them, the NOM group had no characteristic microvascular architecture morphology; the characteristic microvascular architecture morphology of the OLK ED group and the OSCC HD group were fusion type and budding type; the characteristic microvascular architecture morphology of the OSCC MD group was fusion type, budding type and finger-like advancement type; the characteristic microvascular architecture morphology of the OSCC PD group was finger-like advancement type, such as Figure 5 shown.

[0030] Construction and identification results of the animal model of oral mucosal carcinogenesis: This experiment successfully constructed an animal model of oral mucosal carcinogenesis, and obtained tumor-bearing tissues in each group. Microscopically, the tumor-bearing tissues in the NOM group had flaky epithelial structures. In the PT group, tumor cell masses did not involve flaky epithelial structures, and epithelial cells had abnormal nuclear divisions. In the SCC HD group, tumor cell masses involved flaky epithelial structures and formed round-shaped cancer nests with capsules around them. The central blood vessels of the cancer nests were sparse, the tumor cells were closely arranged, and microvessels were seen surrounding the cancer nests. In the SCC MD group, the cancer nests involved epithelial structures, and the capsules were incomplete. The number of microvessels in the central part of the cancer nests increased, and the microvascular walls were intact. In the SCC PD group, the cancer nests involved epithelial structures, and no capsules were seen. The microvascular walls in the central part of the cancer nests were partially missing, and there were phenomena such as luminal fibrosis, tumor cells invading blood vessels, and "VM".

[0031] Results of microvascular typing analysis of oral mucosal carcinogenesis animal model: HE staining slides at various stages of oral mucosal carcinogenesis showed that the main microvascular morphologies in tumor-bearing cells were fusion type, budding type, finger-like propulsion type, terminal expansion type, and crisscross type (as shown in Figure 6). The NOM group had no characteristic microvascular morphology; the characteristic microvascular morphologies of the PT group and the SCC HD group were fusion type and budding type; the characteristic microvascular morphologies of the SCC MD group were fusion type, budding type, and finger-like propulsion type; the characteristic microvascular morphologies of the SCC PD group were finger-like propulsion type (as shown in Figure 7).

[0032] Results of observation and analysis of the microvascular ultrastructure in the animal model of oral mucosal carcinogenesis: The results of ultrastructure observation of tumor-bearing microvessels showed that the diameter of the reticular microvessels in the NOM group was relatively uniform, the spatial distribution was relatively uniform, the microvascular surface was smooth, the vascular classification of each level was natural, and the morphology was relatively consistent. In the PT group, there were more microvascular branches, and the microvascular walls of some branches were irregular in shape, and the vascular classification was not natural. The microvascular architecture in the tumor changed significantly, the capillary network was irregular, there were abnormal anastomoses between adjacent blood vessels, the microvascular walls were twisted, the thickness was different, and blind-end blood vessel formation was visible (as shown in Figure 8).

[0033] In a specific embodiment, statistical analysis selected 2215 proteins with different expression levels. When FC>1.2 and p≤0.05, up-regulated proteins were screened, and when FC<0.83 and p≤0.05, down-regulated proteins were screened. Compared with the carcinogenesis control group, the canceration process experimental group had 1025 up-regulated proteins and 1190 down-regulated proteins. Excluding genes with poor correlation and meaningless genes between samples, egpl vs cgpl, eghdscc vs cghdscc, egmdscc vs chmdscc, egpdscc vs cgpdscc had a total of 5 differentially expressed up-regulated proteins (KYNU, PSMD11, CCNY, SEC61G, RPS27L), and 1 down-regulated protein (DERL1); cgpl, cghdscc, cgmdscc, and cgpdscc had a total of 4 up-regulated proteins (CK15, APCS, CLSTN1, USO1), and 17 down-regulated proteins (DERL1). There were 2 up-regulated proteins (SNAP47, IGHM) and 2 down-regulated proteins (ME1, ANGPTL2) among egpl, eghdscc, egmdscc, and egpdscc. Among them, SEC61G, CK15, MT3, DNPEP, EEF1E1, MYCBP, NAXE, and ANGPTL2 were closely related to the angiogenesis mechanism.

[0034] Western blot analysis was performed to analyze the expression of each protein in the normal group, precancerous lesion tissue group, well-differentiated squamous cell carcinoma group, moderately differentiated squamous cell carcinoma group, and poorly differentiated squamous cell carcinoma group. ANGPTL2, CK15, and EEF1E1 were detected in all five groups. The internal reference GADPH was used for normalization, and the gray value difference analysis was used to understand the expression of each protein in each group. In each stage of oral mucosal carcinoma in each group, the expression of CK15 was gradually increased, the expression of ANGPTL2 was gradually increased from normal to well-differentiated squamous cell carcinoma stage, and gradually decreased from well-differentiated squamous cell carcinoma to poorly differentiated squamous cell carcinoma stage, with the highest expression in the well-differentiated squamous cell carcinoma stage, and the expression of EEF1E1 was gradually increased from normal to moderately differentiated squamous cell carcinoma stage, and gradually decreased from moderately differentiated squamous cell carcinoma to poorly differentiated squamous cell carcinoma stage, with the highest expression in the moderately differentiated squamous cell carcinoma stage.

[0035] Immunofluorescence staining: ANGPTL2, CK15, and EEF1E1 were used to perform immunohistofluorescence staining on oral mucosal carcinogenesis tissues. The results showed that in each stage of oral mucosal carcinogenesis, ANGPTL2, CK15, and EEF1E1 could be selectively expressed in microvessels with heterogeneous microvascular morphology at each stage of oral mucosal carcinogenesis. The vascular lumen was small and irregular, the structure was incomplete, and the boundary with the surrounding tissue was unclear.

[0036] In a specific embodiment, during the process of oral mucosal carcinogenesis, TSCs may enhance their stemness and multidifferentiation ability by upregulating the expression of CK15 and downregulating the expression of ANGPTL2 and EEF1E1, promote the formation of VM, further induce characteristic changes in MAPH, promote tumor growth, and reduce the degree of tumor differentiation.

[0037] S2: based on the pathological image, the morphological characteristics of the microvessels are obtained; the morphological characteristics include fusion type, sprouting type, and finger-like propulsion type; the fusion type is that the cross-section of the circular microvessel lumen is partially fused, the sprouting type is that the microvessel lumen sprouts to form a vascular branch, and the finger-like propulsion type is that the longitudinal section of the microvessel lumen is parallel and aggregated; In one embodiment, the method also includes normal morphology judgment, performing feature extraction on the pathological image, judging whether the morphological features have microvascular morphological features, and judging as normal oral mucosa when there are no microvascular morphological features or when the morphological features are elliptical; when there are microvascular morphological features or when they are not elliptical, further judging the type of microvascular morphological features, when the microvascular morphological features are fusion type and budding type, the process of oral mucosal canceration is oral leukoplakia with epithelial abnormal proliferation or well-differentiated oral squamous cell carcinoma; when the microvascular morphological features are fusion type, budding type and finger-like advancement type, the process of oral mucosal canceration is moderately differentiated oral squamous cell carcinoma; when the microvascular morphological features are finger-like advancement type, the process of oral mucosal canceration is poorly differentiated oral squamous cell carcinoma.

[0038] S3: Based on the morphological characteristics, the progression of oral mucosal carcinoma is judged. When the microvascular morphological characteristics are fusion type and budding type, the progression of oral mucosal carcinoma is oral leukoplakia with epithelial abnormal proliferation or well-differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are fusion type, budding type and finger-like advancement type, the progression of oral mucosal carcinoma is moderately differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are finger-like advancement type, the progression of oral mucosal carcinoma is poorly differentiated oral squamous cell carcinoma.

[0039] In one embodiment, the vascular mimicry (structural features) of the microvessels is detected. When the cell lining of the vascular mimicry is an endothelial cell lining, it is a normal oral mucosa; when the cell lining of the vascular mimicry is a tumor cell lining, it is the oral mucosa canceration stage.

[0040] In one embodiment, the method further includes a spatial structure to detect the distribution of microvessels. When the microvessels are distributed at the front of tumor invasion and advance in a finger-like manner, the progression of oral mucosal carcinogenesis progresses from one stage to another.

[0041] In one embodiment, S1 is replaced by: obtaining oral mucosal tissue of a subject to be tested; S2 is replaced by: detecting the expression level of ANGPTL2 gene based on the oral mucosal tissue; S3 is replaced by: when the expression level of ANGPTL2 gene is greater than a first preset threshold, determining it as well-differentiated oral squamous cell carcinoma.

[0042] In one embodiment, the first preset threshold is the ANGPTL2 gene expression level in normal mucosal tissue.

[0043] In another embodiment, S2 is replaced by: detecting the CK15 gene expression level based on the oral mucosal tissue; S3 is replaced by: when the CK15 gene expression level is less than a second preset threshold, determining it as well-differentiated oral squamous cell carcinoma, moderately differentiated oral squamous cell carcinoma, or poorly differentiated oral squamous cell carcinoma.

[0044] Optionally, the second preset threshold is the CK15 gene expression level in normal mucosal tissue.

[0045] In another embodiment, S2 is replaced by: detecting the expression level of EEF1E1 gene based on the oral mucosal tissue; S3 is replaced by: when the expression level of EEF1E1 gene is greater than a third preset threshold, determining it as moderately differentiated oral squamous cell carcinoma.

[0046] Optionally, the third preset threshold is the EEF1E1 gene expression level in normal mucosal tissue.

[0047] In another embodiment, S2 is replaced by: detecting the expression levels of ANGPTL2 and CK15 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression levels of ANGPTL2 and CK15 genes increase, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression level of ANGPTL2 gene decreases and the expression level of CK15 gene increases, it is determined to be moderately differentiated oral squamous cell carcinoma or poorly differentiated oral squamous cell carcinoma.

[0048] In another embodiment, S2 is replaced by: detecting the expression levels of ANGPTL2 and EEF1E1 genes based on the oral mucosal tissue; S3 is replaced by: when the expression levels of ANGPTL2 and CK15 genes increase, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression level of ANGPTL2 gene decreases and the expression level of EEF1E1 gene increases, it is determined to be moderately differentiated oral squamous cell carcinoma.

[0049] In another embodiment, S2 is replaced by: detecting the expression levels of CK15 and EEF1E1 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression levels of CK15 and EEF1E1 genes increase, it is determined to be well-differentiated oral squamous cell carcinoma or moderately differentiated oral squamous cell carcinoma; when the expression level of CK15 gene increases and the expression level of EEF1E1 gene decreases, it is determined to be poorly differentiated oral squamous cell carcinoma.

[0050] In another embodiment, S2 is replaced by: detecting the expression levels of ANGPTL2, CK15, and EEF1E1 genes based on the oral mucosal tissue; S3 is replaced by: when the expression levels of ANGPTL2, CK15, and EEF1E1 genes increase, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression level of ANGPTL2 gene decreases and the expression levels of CK15 and EEF1E1 genes increase, it is determined to be moderately differentiated oral squamous cell carcinoma; when the expression levels of ANGPTL2 and EEF1E1 genes decrease and the expression level of CK15 gene increases, it is determined to be poorly differentiated oral squamous cell carcinoma.

[0051] In one embodiment, S2 is replaced by: inputting the pathological image into a prediction model to obtain a prediction result of the oral mucosal cancer progression, and removing S3; wherein the prediction model is obtained by training the pathological images of the oral mucosal cancer progression at different stages through a neural network.

[0052] Optionally, the neural network includes one or more of the following: convolutional neural network, residual network, Transformer.

[0053] The disclosed embodiments of the present invention further provide a computer program product or system, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for predicting the progression of oral mucosal carcinogenesis based on microvessels.

[0054] Figure 2 The schematic diagram of the microvascular-based oral mucosal cancer progression prediction system provided in the embodiment of the present invention specifically includes: Acquisition unit: acquiring pathological images of oral mucosal tissue of the subject to be tested; Characteristic unit: based on the pathological image, the morphological characteristics of the microvessels are obtained; the morphological characteristics include fusion type, sprouting type, and finger-like propulsion type; the fusion type is that the cross-section of the circular microvessel lumen is partially fused, the sprouting type is that the microvessel lumen sprouts to form a blood vessel branch, and the finger-like propulsion type is that the longitudinal section of the microvessel lumen is parallelly clustered and arranged; Prediction unit: judging the progression of oral mucosal carcinogenesis based on the morphological characteristics, when the microvascular morphological characteristics are fusion type and budding type, the progression of oral mucosal carcinogenesis is oral leukoplakia with epithelial abnormal proliferation or well-differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are fusion type, budding type and finger-like advancement type, the progression of oral mucosal carcinogenesis is moderately differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are finger-like advancement type, the progression of oral mucosal carcinogenesis is poorly differentiated oral squamous cell carcinoma.

[0055] Figure 3 The schematic diagram of the microvascular-based oral mucosal cancer progression prediction device provided in the embodiment of the present invention specifically includes: A memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, any one of the above-mentioned microvascular-based oral mucosal cancer progression prediction methods is executed.

[0056] The disclosed embodiments of the present invention further provide a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, is any one of the above-mentioned microvascular-based oral mucosal carcinogenesis progression prediction methods.

[0057] The embodiment of the present invention provides a method for predicting the progression of oral mucosal carcinogenesis based on MAPH, comprising: Obtaining the oral mucosal microvascular data of the test subject, including microvascular morphological characteristics, structural characteristics, and spatial characteristics; The microvascular data is input into the MAPH model to obtain the process of oral mucosal carcinogenesis; the MAPH model is composed of the microvascular morphological characteristics, structural characteristics, and spatial characteristics of different stages of oral mucosal carcinogenesis, and the process includes oral leukoplakia with epithelial abnormal proliferation, well-differentiated oral squamous cell carcinoma, moderately differentiated oral squamous cell carcinoma, and poorly differentiated oral squamous cell carcinoma.

[0058] In one embodiment, the microvascular morphological characteristics of the MAPH model include: oral leukoplakia with epithelial abnormal proliferation is fusion type and budding type, well-differentiated oral squamous cell carcinoma is fusion type and budding type, moderately differentiated oral squamous cell carcinoma is fusion type, budding type and finger-like advancement type, and poorly differentiated oral squamous cell carcinoma is finger-like advancement type.

[0059] Optionally, the structural feature includes blood vessels mimicking the lining of tumor cells. Optionally, the spatial feature is that microvessels are mainly distributed at the front of tumor invasion and advance in a finger-like manner.

[0060] In a specific embodiment, in terms of morphology: as the oral mucosal carcinogenesis progresses, the characteristic microvascular morphology changes from mainly fusion and budding to mainly finger-like propulsion; in the process of oral mucosal carcinogenesis, the microvascular architecture is structurally heterogeneous: normal oral mucosal microvascular vessels are lined by endothelial cells, and some microvascular vessels in the oral squamous cell carcinoma (OSCC) stage are lined by tumor cells, forming "VM". In the process of oral mucosal carcinogenesis, the microvascular architecture is spatially heterogeneous: normal oral mucosal microvascular vessels are evenly distributed, and as the oral mucosal carcinogenesis progresses, microvascular vessels are mainly distributed at the front of tumor invasion, propelling in a finger-like manner.

[0061] In a specific embodiment, the possible existence form of "microvascular spatial distribution heterogeneity" in the process of oral mucosal carcinogenesis is: in the process of oral mucosal carcinogenesis, the tumor microvessels appear unevenly distributed in space, the tumor invades the blood vessels in the central part of the large area of ​​tumor parenchyma, the vascular endothelial cells disappear, and the blood vessels degenerate. At the front of tumor progression, blood vessels grow densely to provide nutrition for the rapidly proliferating tumor tissue. As shown in the microvascular typing results of this experiment, the microvessels in the PD stage of OSCC are unevenly distributed, concentrated in the front of tumor invasion and arranged in parallel, advancing and growing toward the tumor-normal tissue junction area, and microvessels are rare in the central part of the tumor parenchyma. The spatial distribution heterogeneity of microvascular architecture is the adaptive result of tumor microvessels to meet the needs of tumor proliferation and invasion. The traditional research method of evaluating the process of tumor deterioration by selecting several vascular-rich areas to measure the positive staining of tumor microvascular endothelial cells by immunohistochemical staining and determining the microvessel density (MVD) has limitations. By observing the transparent specimens of the animal model of oral mucosal carcinogenesis in this experiment, it was found that the process of oral mucosal carcinogenesis is accompanied by the spatial distribution heterogeneity of microvascular architecture. In the OSCC MD stage, blood vessels branched to form a densely distributed microvascular network in some areas of the tumor, while microvessels were rare in other parts. This may be because microvessels mainly proliferate densely in the parts where tumors proliferate and invade actively. In summary, there is heterogeneity in the spatial distribution of microvascular architecture in the process of oral mucosal carcinogenesis.

[0062] In a specific embodiment, this experiment observed that the more comprehensive microvascular endothelial cell markers did not have a good immunohistochemical positive labeling for finger-like propulsion microvessels. This may be because finger-like propulsion vessels often appear at the front of tumor progression, are related to tumor invasion, and are mostly formed by tumor cell mutation to form tumor lining cells, which do not express endothelial cell markers.

[0063] The embodiment of the present invention provides a system for predicting the progression of oral mucosal canceration of MAPH. The system includes a computer program or instructions. The computer program or instructions are executed by a processor to implement the method for predicting the progression of oral mucosal canceration of MAPH.

[0064] The schematic diagram of the device for predicting the progression of oral mucosal carcinogenesis of MAPH provided in the embodiment of the present invention specifically includes: A memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, any one of the above-mentioned methods for predicting the progression of oral mucosal canceration of MAPH is obtained.

[0065] The disclosed embodiment of the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, is any one of the above-mentioned methods for predicting the progression of oral mucosal carcinogenesis of MAPH.

[0066] The present invention provides an application of a substance targeting TSCs in the preparation of a drug for anti-angiogenic treatment of oral mucosal carcinogenesis. In one embodiment, the substance targeting TSCs acts on the recruitment process of TSCs, specifically on the process of TSCs recruiting microangiogenesis.

[0067] The embodiment of the present invention provides a use of a reagent for inhibiting a gene or overexpressing a gene in the preparation of a drug for anti-angiogenic treatment of oral mucosal carcinogenesis, wherein the gene includes ANGPTL2.

[0068] In one embodiment, the gene further comprises one or more of the following: CK15, EEF1E1, TIMP1. In one embodiment, the ANGPTL2 is replaced by CK15. In one embodiment, the ANGPTL2 is replaced by EEF1E1. In one embodiment, the ANGPTL2 is replaced by TIMP1. In one embodiment, the drug is used to regulate the mediation between TIMP1 and MMPs. In one embodiment, the MMPs comprise one or more of the following: MMP1, MMP2, MMP3, MMP9, MMP10, MMP14.

[0069] In one embodiment, the prediction of different processes of oral mucosal canceration includes one or more of the following: prediction of different processes of oral mucosal canceration by microvascular morphological characteristics and ANGPTL2; The morphological characteristics of microvessels and ANGPTL2 are used to predict the different processes of oral mucosal carcinogenesis; The morphological characteristics of microvessels and CK15 were used to predict the different processes of oral mucosal carcinogenesis; The morphological characteristics of microvessels and EEF1E1 are used to predict the different processes of oral mucosal carcinogenesis; The morphological characteristics of microvessels and ANGPTL2 and CK15 were used to predict the different processes of oral mucosal carcinogenesis; The morphological characteristics of microvessels and ANGPTL2 and EEF1E1 were used to predict the different processes of oral mucosal carcinogenesis; The morphological characteristics of microvessels and CK15 and EEF1E1 were used to predict the different processes of oral mucosal carcinogenesis; The morphological characteristics of microvessels and ANGPTL2, CK15, and EEF1E1 were used to predict the different processes of oral mucosal carcinogenesis; The morphological characteristics of microvessels and ANGPTL2, CK15, EEF1E1, and TIMP1 were used to predict the different processes of oral mucosal carcinogenesis; Microvascular MAPH, ANGPTL2, and CK15 were used to predict the different processes of oral mucosal carcinogenesis; Microvascular MAPH and ANGPTL2, CK15, and EEF1E1 were used to predict the different processes of oral mucosal carcinogenesis; Microvascular MAPH, CK15 and EEF1E1 are used to predict the different processes of oral mucosal carcinogenesis.

[0070] In a specific embodiment, laser capture microdissection of characteristic microvessels: fresh tissues of the corresponding numbered NOM group, OLK ED group, OSCC HD group, OSCC MD group, and OSCC PD group were taken out from a -120°C refrigerator, frozen and embedded, sliced ​​at a thickness of 10 μm, and loaded with MMI DAB staining kit Beijing BiossBCA protein assay kit Beijing Solebow Membrane Slide to carry the sample. After the OCT embedding agent was completely evaporated, the Membrane Slide was immersed in hematoxylin for 5 min, rinsed with tap water, 1% hydrochloric acid ethanol for 1 s, rinsed with tap water, ammonia water for 1 min, rinsed with tap water, eosin staining solution for 1.5 min, rinsed with tap water, 80% ethanol for 2 s, 95% ethanol for 2 s, 100% ethanol for 10 min, and xylene for 10 min. The Membrane Slide was inverted on an adhesive slide to form a sandwich structure, which was placed on a laser capture microdissection instrument. Use the mouse to select the microvascular morphology to be cut on the monitor, use laser capture microdissection to cut the target microvascular morphology, and collect the separated microvascular morphology through the adhesive cap. After clarifying the microvascular morphology to be cut in each group, perform continuous sectioning on the fresh tissue, laser capture microdissection of the target microvascular morphology that has not been stained under bright field, and collect it into a new adhesive cap for subsequent experiments.

[0071] Single-cell sequencing data analysis: Data were obtained from the Supplementary file of the GEO database (https: / / www.ncbi.nlm.nih.gov / geo / ). The Seurat package was used to generate objects and filter out cells of poor quality. Standard data preprocessing procedures were performed, and the percentages of gene number, cell number, and mitochondrial content were calculated. The filtering criteria were genes detected in less than 3 cells and cells with less than 200 genes detected. Preprocessing: Genes detected in at least 15,313 cells were retained, and cells with less than 200 or more than 4,000 genes detected and cells with high mitochondrial content (>10%) were filtered out. After discarding poor-quality cells, a total of 2,000 cells were retained for downstream analysis. To standardize each cell, the UMI counts were scaled using scale.factor = 10,000. After logarithmic transformation of the data, the ScaleData function in Seurat was used. Cluster identification: The corrected normalized data were applied to standard analysis. The top 20 variable genes were extracted for principal component analysis. The first 14 principal components were retained for UMAP visualization and clustering. Cell clustering was performed using the FindClusters function (resolution = 0.5) implemented in the Seurat R package.

[0072] Results of single-cell sequencing data analysis: After quality control, a total of 2,000 cells were obtained. After dimensionality reduction clustering, a total of 6 cell clusters were obtained, and 5 types of cell populations were defined, including tumor cells, tumor-associated fibroblasts, endothelial cells, NKT cells, and fibrochondrocytes. Among them, tumor cells were mainly composed of cluster 0, tumor-associated fibroblasts were mainly composed of clusters 4 and 1, endothelial cells were mainly composed of cluster 3, NKT cells were mainly composed of cluster 6, and fibrochondrocytes were mainly composed of cluster 2. The expression heat map of the TIMP1 gene indicated that TIMP1 was expressed in oral tumor tissues, and the expression abundance in tumor cells and endothelial cells was high.

[0073] Analysis results of TIMP1 expression level: Immunohistochemical staining of TIMP1 in NOM, OLK ED, OSCC HD, OSCC MD, and OSCC PD showed that scattered interstitial cell cytoplasm staining was observed in the NOM group. With the progression of oral mucosal carcinogenesis, the number of positive cells increased, and cytoplasmic staining was obvious in cancer cells (as shown in Figure 9). By one-way ANOVA, the expression level of OSCC PD group was higher than that of NOM group and OLK ED group, and the difference was statistically significant (as shown in Figure 10).

[0074] This experiment verified that the expression of TIMP1 markers increased during the process of oral mucosal carcinogenesis, suggesting that high expression of TIMP1 is associated with the process of oral mucosal carcinogenesis. Through cell clustering identification and dimensionality reduction clustering analysis of the OSCC single-cell sequencing data set in the GEO database, it was found that TIMP1 was highly expressed in tumor cells and endothelial cells, suggesting that TIMP1 may affect the process of oral mucosal carcinogenesis by affecting tumor cells and endothelial cells.

[0075] In a specific embodiment, this experiment uses LCM to capture characteristic microvessels for protein chip testing to find potential diagnostic markers and therapeutic targets for oral mucosal carcinogenesis, and clearly defines the site of action of TIMP1 as tumor microvessels. TIMP1 is not affected by MAPH in the process of oral mucosal carcinogenesis, and is expressed in microvessels lined by vascular endothelial cells and "VM" lined by tumor cells. In addition, TIMP1 has different expression levels in different pathological stages of oral mucosal carcinogenesis.

[0076] In one embodiment, analysis of MMPs marker expression levels: paraffin blocks of NOM, OLK ED, OSCC HD, OSCC MD, and OSCC PD groups were sliced ​​at 5 μm. The sections were placed in an oven for 30 min until the paraffin on the sections melted, and the following steps were performed in sequence: immersion in xylene (I) for 12 min, xylene (II) for 12 min, 100% ethanol for 2 min, 95% ethanol for 2 min, 80% ethanol for 2 min, deionized water rinse for 3×3 min, immersion in sodium citrate solution for high-temperature heat repair for 40 min, natural cooling to room temperature, rinsing with PBS for 3×3 min, adding 3% hydrogen peroxide 50-200 μL / slice, incubation at room temperature for 10 min, rinsing with PBS for 3×3 min, adding serum blocking for 15 min, absorbing excess liquid and adding immunohistochemical markers, incubating in a light-proof 4℃ wet box for 12 h, rinsing with PBS for 3×3 min, adding goat anti-mouse / rabbit IgG polymer for incubation at room temperature for 15 min, rinsing with PBS for 3×3 min, incubating with horseradish enzyme marker for 15 min, adding DAB colorimetric solution 50-200 μL / slice, and incubating at room temperature for 10 min. μL / slice, observed under a microscope, and after the cell cytoplasm was completely colored, rinsed with deionized water to stop coloring, immersed in hematoxylin for 5 min, rinsed with tap water, immersed in 1% hydrochloric acid alcohol for 1 s, rinsed with tap water, immersed in 1% ammonia water for 1 min, rinsed with tap water, immersed in 80% ethanol for 2 s, immersed in 95% ethanol for 2 s, immersed in 100% ethanol for 10 min, immersed in xylene for 10 min, sealed with neutral gum, and observed under a light microscope. A negative control was set up (PBS solution was used instead of immunohistochemical markers). Immunohistochemical markers were diluted with deionized water and used immediately. The dilution ratios were: MMP1 (1:100), MMP2 (1:100), MMP3 (1:100), MMP9 (1:100), MMP10 (1:100), and MMP14 (1:100), with n=5 in each group. Five positive expression-rich fields were randomly selected from each section for image collection, and the percentage of immunohistochemical staining-positive area in the images was determined using IMAGE J. One-way ANOVA was used to analyze the differences in the percentage of immunohistochemical staining-positive area among the NOM group, OLK ED group, OSCC HD group, OSCC MD group, and OSCC PD group.

[0077] Correlation analysis: TCGA-HNSC RNAseq data and clinical data were downloaded and sorted from the TCGA database, and data without clinical information were discarded. In the clinical information, samples belonging to oral cancer sites were retained, and samples from non-oral cancer sites were removed. Spearman correlation analysis was performed on the variables in the data using R (4.2.1), and the analysis results were visualized as heat maps using ggplot2[3.3.6]. In this experiment, immunohistochemical staining was performed on the NOM group, OLK ED group, OSCC HD group, OSCC MD group, and OSCC PD group using MMP1, MMP2, MMP3, MMP9, MMP10, and MMP14 as markers to analyze whether TIMP1 was correlated with MMP1, MMP2, MMP3, MMP9, MMP10, and MMP14.

[0078] Immunoprecipitation to verify TIMP1 / MMPs interaction: Fresh samples of NOM, PT, SCCHD, SCC MD, and SCC PD groups were taken out from the ultra-low temperature refrigerator (see Section 1, 4.5 for details), lysed to obtain total protein of each group, and immunoprecipitated with the Beyotime Immunoprecipitation Kit (Protein A+G agarose gel method) to obtain TIMP1 protein complex. 10% separation gel and 5% concentration gel were prepared, and the protein samples were separated by 80V constant voltage electrophoresis. Western blotting was performed by wet transfer method, and the samples were blocked overnight with 5% skim milk powder. The primary antibody was incubated at 4°C overnight (MMP3, MMP9, MMP14, the working solution dilution ratio was 1:1000), rinsed with TBST 3 times for 30 min in total, incubated with goat anti-rabbit IgG (H+L) HRP (working solution dilution ratio was 1:100000) at room temperature for 1 h, rinsed with TBST 3 times for 30 min in total, and used ECL reagent to react in the dark for 1 min. The samples were put on the machine and exposed to light for development.

[0079] PPI analysis showed that protein interactions existed between TIMP1 and MMP1, MMP2, MMP3, MMP9, MMP10, and MMP14.

[0080] Analysis of MMPs expression levels: Immunohistochemical staining was performed on NOM, OLK ED, OSCC HD, OSCC MD, and OSCC PD using MMP1, MMP2, MMP3, MMP9, MMP10, and MMP14 as markers. MMP1, MMP2, MMP3, MMP9, MMP10, and MMP14 markers were found scattered in the cytoplasm of interstitial cells in the NOM group. With the progression of oral mucosal carcinogenesis, the number of positive cells increased, and the cytoplasm of cancer cells was obviously stained ( Fig.11). One-way ANOVA showed that the expression level of MMP1 in OSCC was higher than that in NOM and OLK ED groups, and the expression level of MMP1 was upregulated with tumor progression; the expression levels of MMP2 in OSCC MD and OSCC PD groups were higher than those in NOM group; the expression levels of MMP3 in OSCC MD and OSCC PD groups were higher than those in NOM group; the expression level of MMP3 in OSCC MD group was higher than that in NOM group; the expression level of MMP10 in OSCC PD group was higher than that in NOM group; the expression levels of MMP14 in OSCC MD and OSCC PD groups were higher than those in NOM group ( Fig.12 as shown).

[0081] Results of correlation analysis: After analysis, TIMP1 was correlated with MMP1, MMP2, MMP3, MMP9, MMP10, and MMP14. In the process of oral mucosal carcinogenesis: TIMP1 was positively correlated with MMP1 in the NOM group and OSCC MD group; TIMP1 was positively correlated with MMP2 in the OLKED group and OSCC HD group; TIMP1 was positively correlated with MMP3 in the OSCC HD group; TIMP1 was positively correlated with MMP9 in the NOM group, OLK ED group, OSCC HD group, OSCC MD group, and OSCC PD group; TIMP1 was positively correlated with MMP10 in the OSCC HD group and OSCC MD group; TIMP1 was positively correlated with MMP14 in the NOM group, OLK ED group, OSCC HD group, and OSCC MD group.

[0082] In the process of oral mucosal carcinogenesis, there are protein interactions among TIMP1 / MMP3, TIMP1 / MMP9, and TIMP1 / MMP14, such as Fig.13 shown.

[0083] In the process of oral mucosal carcinogenesis, TIMP1 may affect the process of oral mucosal carcinogenesis by mediating the regulation of MAPH by MMP1, MMP2, MMP3, MMP9, MMP10, and MMP14.

[0084] In a specific embodiment, the recruitment and integration of MAPH by tumor stem cells in the process of oral mucosal carcinogenesis: tissue Western blot, extraction of tissue protein, taking out tissue samples of each stage of oral mucosal carcinogenesis in the experimental group and the control group from the -80°C refrigerator, numbering the experimental normal group egn, the experimental precancerous lesion tissue group egp1, the experimental well-differentiated squamous cell carcinoma group eghdscc, the experimental moderately differentiated squamous cell carcinoma group egmdscc, the experimental poorly differentiated squamous cell carcinoma group egpdscc, the control normal group cgn, the control precancerous lesion tissue group cgp1, the control well-differentiated squamous cell carcinoma group cghdscc, the control moderately differentiated squamous cell carcinoma group cgmdscc, and the control poorly differentiated squamous cell carcinoma group cgpdscc, weighing, cutting, adding liquid nitrogen to grind the sample into powder; adding 150 μL RIPA working solution (RIPA: PMSF: 10× phosphatase inhibitor = 100:1:10) to each 20 mg tissue sample, mixing well, and lysing on ice for 20 min; set the ultrasonic intensity of the ultrasonic disruptor to 30%, the working time to 3 s, the interval time to 6 s, and ultrasonically lyse the sample on ice for a total of 20 s; transfer the sample to a low-temperature high-speed centrifuge and centrifuge for 10 min, set the temperature to 4°C, and the speed to 12000 rpm; transfer the supernatant to a new EP tube, mark it, and store it at -80°C.

[0085] Tissue protein concentration determination: according to the number of samples, BCA working solution was prepared at a ratio of BCA:Cu=50:1 and allowed to stand at room temperature for use; 0, 2, 4, 6, 8, 12, 16, 20 μL of standard (the standard was diluted to a concentration of 0.5 mg / mL with PBS) was added to a 96-well plate, PBS was added to 20 μL, the sample to be tested was diluted 10 times and 20 μL was added to a 96-well plate, and 3 replicate wells were made for the standard and sample; 200 μL of BCA working solution was added to each well, mixed, and incubated at 37°C in the dark for 30 min; the absorbance of each well at a wavelength of 562 nm was measured using an enzyme reader, and the protein concentration of each sample after 10-fold dilution was calculated using the standard curve.

[0086] Denaturation of tissue proteins: After protein quantification, add 1 / 4 volume of 5× loading buffer to the sample, boil at 100℃ for 10 min, cool to room temperature, and store at -80℃ for later use.

[0087] Western blot: (1) Preparation of gel and loading of samples: According to the molecular weight of the target protein, prepare a separation gel of appropriate concentration, slowly add the separation gel solution along one side of the glass plate wall, and add a layer of isopropanol solution on the gel; after the separation gel solidifies, discard the isopropanol solution on the gel; prepare a 5% stacking gel, fill the remaining space with the stacking gel, and then select a 10-hole loading comb and insert it into the stacking gel; after the stacking gel solidifies, place the gel plate in the electrophoresis tank, add 1× electrophoresis buffer to the inner tank, pull out the loading comb vertically, and load the sample.

[0088] (2) Electrophoresis: Add an appropriate amount of 1× electrophoresis buffer to the inner and outer tanks, turn on the power, adjust the constant voltage to 80 V, and perform electrophoresis for 20 min. When the protein enters the separation gel, adjust the constant voltage to 120 V and perform electrophoresis for 30-60 min. Stop electrophoresis when the marker is properly separated.

[0089] (3) Gel cutting: Gel cutting was performed according to the marker level corresponding to the molecular weight of the target protein, ANGPTL2 (55 kDa), CK15 (49 kDa), and EEF1E1 (21 kDa).

[0090] (4) Transfer: Place three layers of filter paper on the sponge pads on both sides of the transfer clamp, and soak them thoroughly with 1× transfer solution; gently transfer the cut gel to the sandwich clamp, and rinse with 1× transfer solution to keep the gel moist; cut a PVDF membrane that matches the size of the gel, immerse it in methanol for activation for 30 seconds, and then cover the PVDF membrane on the surface of the gel to avoid generating bubbles; cover the filter paper and sponge pad on the opposite side, fasten the transfer clamp and install it in the transfer tank, pour in an appropriate amount of 1× transfer solution, turn on the power, adjust the constant current to 200 mA, and transfer at low temperature for 30-60 minutes.

[0091] (5) Blocking: After transfer, immerse the PVDF membrane in 5% skim milk blocking solution and block at room temperature for 1 h. After blocking, rinse with 1× TBST three times, 5 min each time.

[0092] (6) Primary antibody incubation: Immerse the PVDF membrane in ANGPTL2, CK15 and EEF1E1 antibodies (1:1000) diluted in 1× TBST and incubate overnight at 4°C; rinse three times with 1× TBST, 10 min each time.

[0093] (7) Secondary antibody incubation: Immerse the PVDF membrane in secondary antibody diluted in 1× TBST (1:10,000) and incubate on a shaker for 1 h; rinse three times with 1× TBST, 10 min each time.

[0094] (8) Development: Immerse the PVDF membrane in ECL developer prepared at a ratio of 1:1, react for 1 min in the dark, and develop on a development machine to obtain and save the bands.

[0095] (9) Result analysis: The bands were analyzed using Image J image analysis software.

[0096] Immunofluorescence staining: Immunofluorescence staining was performed on the tissues of oral mucosa at different stages of carcinogenesis in the experimental group and the normal group. (1) The specific steps of paraffin section preparation and dewaxing are the same as those in Section 2.7; (2) 3% H2O2 treatment for 10 min to block endogenous peroxidase, and then washing with PBS three times, 5 min each time; (3) Primary antibody incubation (rabbit anti-human ANGPTL2, CK15, and EEF1E1 antibodies were diluted with PBS solution at a certain ratio for use), incubated in a refrigerator at 4°C overnight, and then washed with PBS three times, 5 min each time; (4) Secondary antibody incubation (PBS solution diluted with secondary antibody goat anti-rabbit Rhodamine at a ratio of 1:100), incubated at 37°C in the dark for 1 h, and then washed with PBS three times, 5 min each time; (5) DAPI solution was diluted to the working concentration (1:1000), incubated at room temperature in the dark for 30 min, and then washed with PBS three times, 5 min each time; (6) 5% glycerol was added for sealing, and images were collected and analyzed under an inverted fluorescence microscope.

[0097] In a specific embodiment, when the oral mucosal carcinoma progresses to the stage of moderately differentiated and poorly differentiated squamous cell carcinoma, VM can be observed under the microscope in some oral mucosal carcinoma tissues. Tumor cells are recruited to and invade the microvessels, the microvessel basement membrane is loose, the endothelial cells are arranged irregularly or even disappear, the blood vessels may be lined by tumor cells instead of endothelial cells, and the red blood cells in the lumen are pink homogeneous structures with unclear boundaries, such as Fig.14 shown.

[0098] In a specific embodiment, EdU Apollo 567 is used to trace and label tumor tissues and microvessels of various organ tissues in vivo: TSCs labeled with EdU Apollo 567 are injected into the nude mice of the experimental group through the tail vein and distributed in various parts of the nude mice. As the carcinogenesis progresses, TSCs tend to migrate to the perivascular cavity of the tumor tissue to grow and proliferate, and express red fluorescence; normal tissues and diseased tissues express blue fluorescence. At the same time, in addition to egn, TSCs can also be seen in the various organ tissues (heart, liver, spleen, lung, and kidney) of nude mice, expressing red fluorescence. Combined with the observation of HE staining of nude mouse organ tissues, TSCs tend to migrate through the vascular system and reside around the tissue microvessels, and TSCs can be seen invading the microvessels, resulting in heterogeneity in the morphological structure of the microvessels.

[0099] In a specific embodiment, this study used microvascular morphological heterogeneity to study the heterogeneity of newly formed microvessels at multiple levels in tumors, and comprehensively described the heterogeneity of tumor newly formed blood vessels in terms of density, morphology, and structure, as well as the characteristic changes of microvascular morphological heterogeneity in the process of oral mucosal carcinogenesis when TSCs interfered. Based on the characteristics of microvascular morphological heterogeneity, this study quantitatively analyzed the relationship between each development stage of oral mucosal carcinogenesis and related vascular morphological architecture. The results showed that in different development stages of oral mucosal carcinogenesis, tumor vascular morphology heterogeneity was shown, that is, the characteristics of microvascular morphological heterogeneity were different between oral mucosal cancer tissues at different development stages, between different growth sites at the same development stage, and even between different distribution areas of the same vascular segment. In the absence of TSCs interference, compared with the normal oral mucosa stage, the vascular heterogeneity in the precancerous lesion stage and squamous cell carcinoma stage was different, and the vascular morphological heterogeneity of different differentiation degrees in the squamous cell carcinoma stage was also different. Among them, the dominant microvascular morphology in the precancerous lesion stage is fusion type and budding type, and the dominant microvascular morphology in the early stage of squamous cell carcinoma is still fusion type and budding type. As the degree of differentiation decreases, the finger-like advancement type gradually becomes the dominant microvascular morphology of moderately and poorly differentiated squamous cell carcinoma. The appearance of the finger-like advancement type of microvascular morphology seems to indicate the acceleration of tumor deterioration. However, by injecting TSCs into the tail vein at various stages of oral mucosal carcinogenesis, we found that except for the heterogeneity of microvascular morphology in the normal stage of oral mucosa, the characteristics did not change, and the heterogeneity of microvascular morphology in other stages changed under the interference of TSCs. The main type of microvascular morphology in the precancerous lesion stage of TSCs interference changed from fusion type and budding type to fusion type, the main type of microvascular morphology in the well-differentiated squamous cell carcinoma stage changed from fusion type and budding type to fusion type, the main type of microvascular morphology in the moderately differentiated squamous cell carcinoma stage changed from fusion type, budding type and finger-like advancement type to budding type, and the main type of microvascular morphology in the poorly differentiated squamous cell carcinoma stage was still finger-like advancement type. The above results indicate that the heterogeneity of microvascular morphology can reflect the carcinogenesis process of oral mucosa to a certain extent. TSCs interfere with the characteristic changes of microvascular morphology in the carcinogenesis process of oral mucosa, and the morphology of microvessels changes with the transformation of fusion type, budding type and finger-like propulsion type, and the degree of tumor differentiation gradually decreases. In addition, we observed that compared with the uninterrupted moderately and poorly differentiated squamous cell carcinoma tissues, the microvascular lumen area in the moderately and poorly differentiated squamous cell carcinoma tissues after TSCs interference was large, the blood perfusion was high, but they were often in a state of hypoxia and necrosis. This may be due to the fact that VM is not mature in structure and function in the process of oral mucosal carcinogenesis mediated by TSCs, which is manifested as the new microvessels of tumors being squeezed by tumor cells, the vascular diameter is uneven, the arrangement of vascular walls is abnormal, there is no hierarchical branching of normal blood vessels, and the blood flow is disordered, resulting in limited tumor perfusion and reduced oxygen supply in the tumor.

[0100] Studies have shown that there are spatial differences in the distribution of TSCs. As the carcinogenesis progresses, TSCs invade more microvessels and transform into endothelial cells. TSCs' recruitment of blood vessels and their own plasticity may be one of the mechanisms for the formation of heterogeneity in microvascular morphology. TSCs can participate in tumor microvascular angiogenesis through different pathways and methods, playing an important role in the heterogeneity of microvascular morphology. The cross-talk between TSCs and tumor microvascular angiogenesis can synergistically promote tumor growth.

Claims

1. A method for predicting the progression of oral mucosal carcinogenesis based on microvessels, characterized in that: include: S1 obtains the pathological image of the oral mucosal tissue of the subject; S2 obtains morphological characteristics of microvessels based on the pathological image; The morphological characteristics include fusion type, sprouting type, and finger-like propulsion type; the fusion type is that the cross-section of the circular microvascular lumen is partially fused, the sprouting type is that the microvascular lumen sprouts to form vascular branches, and the finger-like propulsion type is that the longitudinal sections of the microvascular lumen are clustered and arranged in parallel; S3 judges the progression of oral mucosal carcinoma based on the morphological characteristics. When the microvascular morphological characteristics are fusion type and budding type, the progression of oral mucosal carcinoma is oral leukoplakia with epithelial abnormal proliferation or well-differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are fusion type, budding type and finger-like advancement type, the progression of oral mucosal carcinoma is moderately differentiated oral squamous cell carcinoma; when the microvascular morphological characteristics are finger-like advancement type, the progression of oral mucosal carcinoma is poorly differentiated oral squamous cell carcinoma.

2. The method for predicting the progression of oral mucosal carcinogenesis based on microvessels according to claim 1, characterized in that: The step S1 is replaced by: obtaining oral mucosal tissue of a subject to be tested; the step S2 is replaced by: detecting the expression level of the ANGPTL2 gene based on the oral mucosal tissue; the step S3 is replaced by: when the expression level of the ANGPTL2 gene is greater than a first preset threshold, determining the patient to be a well-differentiated oral squamous cell carcinoma; Optionally, the first preset threshold is the ANGPTL2 gene expression level in normal mucosal tissue; Optionally, S2 is replaced by: detecting the CK15 gene expression level based on the oral mucosal tissue; S3 is replaced by: when the CK15 gene expression level is less than a second preset threshold, determining it as well-differentiated oral squamous cell carcinoma, moderately differentiated oral squamous cell carcinoma, or poorly differentiated oral squamous cell carcinoma; Optionally, the second preset threshold is the CK15 gene expression level of normal mucosal tissue; Optionally, S2 is replaced by: detecting the expression level of EEF1E1 gene based on the oral mucosal tissue; S3 is replaced by: when the expression level of EEF1E1 gene is greater than a third preset threshold, determining it as moderately differentiated oral squamous cell carcinoma; Optionally, the third preset threshold is the EEF1E1 gene expression level of normal mucosal tissue; Optionally, S2 is replaced by: detecting the expression of ANGPTL2 and CK15 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression of ANGPTL2 and CK15 genes increases, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression of ANGPTL2 gene decreases and the expression of CK15 gene increases, it is determined to be moderately differentiated oral squamous cell carcinoma or poorly differentiated oral squamous cell carcinoma; Optionally, S2 is replaced by: detecting the expression of ANGPTL2 and EEF1E1 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression of ANGPTL2 and CK15 genes increases, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression of ANGPTL2 gene decreases and the expression of EEF1E1 gene increases, it is determined to be moderately differentiated oral squamous cell carcinoma; Optionally, S2 is replaced by: detecting the expression of CK15 and EEF1E1 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression of CK15 and EEF1E1 genes increases, it is determined to be well-differentiated oral squamous cell carcinoma or moderately differentiated oral squamous cell carcinoma; when the expression of CK15 gene increases and the expression of EEF1E1 gene decreases, it is determined to be poorly differentiated oral squamous cell carcinoma; Optionally, S2 is replaced by: detecting the expression levels of ANGPTL2, CK15, and EEF1E1 genes based on the oral mucosal tissue, and S3 is replaced by: when the expression levels of ANGPTL2, CK15, and EEF1E1 genes increase, it is determined to be well-differentiated oral squamous cell carcinoma; when the expression level of ANGPTL2 gene decreases and the expression levels of CK15 and EEF1E1 genes increase, it is determined to be moderately differentiated oral squamous cell carcinoma; when the expression levels of ANGPTL2 and EEF1E1 genes decrease and the expression level of CK15 gene increases, it is determined to be poorly differentiated oral squamous cell carcinoma.

3. The method for predicting the progression of oral mucosal carcinogenesis based on microvessels according to claim 1, characterized in that: S2 is replaced by: inputting the pathological image into a prediction model to predict the progression of oral mucosal canceration and obtaining a prediction result, and removing S3; wherein the prediction model is obtained by training the pathological images of the progression of oral mucosal canceration at different stages through a neural network; Optionally, the neural network includes one or more of the following: convolutional neural network, residual network, Transformer.

4. A method for predicting the progression of oral mucosal carcinogenesis based on MAPH, characterized in that: include: Acquire the oral mucosal microvascular data of the subject to be tested, including microvascular morphological characteristics, structural characteristics, and spatial characteristics; the morphological characteristics are extracted from the method for predicting the progression of oral mucosal carcinogenesis based on microvessels as described in claim 1; The microvascular data is input into the MAPH model to obtain the process of oral mucosal carcinogenesis; the MAPH model is composed of the microvascular morphological characteristics, structural characteristics, and spatial characteristics of different stages of oral mucosal carcinogenesis, and the process includes oral leukoplakia with epithelial abnormal proliferation, well-differentiated oral squamous cell carcinoma, moderately differentiated oral squamous cell carcinoma, and poorly differentiated oral squamous cell carcinoma.

5. The method for predicting the progression of oral mucosal carcinogenesis based on MAPH according to claim 4, characterized in that: The microvascular morphological characteristics of the MAPH model include: oral leukoplakia with epithelial dysplasia are fusion type and budding type, well-differentiated oral squamous cell carcinoma are fusion type and budding type, moderately differentiated oral squamous cell carcinoma are fusion type, budding type and finger-like advancement type, and poorly differentiated oral squamous cell carcinoma is finger-like advancement type; Optionally, the structural features include blood vessels mimicking tumor cell linings; Optionally, the spatial feature is that the microvessels are mainly distributed at the tumor invasion front and advance in a finger-like manner.

6. A computer program product comprising a computer program or instructions, characterized in that: The computer program or instructions are executed by a processor to implement the method for predicting the progression of oral mucosal canceration based on microvessels as described in any one of claims 1 to 3, or to implement the method for predicting the progression of oral mucosal canceration based on MAPH as described in claims 4 to 5.

7. A computer device comprising a memory, a processor and a computer program or instruction stored in the memory, characterized in that: The computer program or instructions are executed by a processor to implement the method for predicting the progression of oral mucosal canceration based on microvessels as described in any one of claims 1 to 3, or to implement the method for predicting the progression of oral mucosal canceration based on MAPH as described in claims 4 to 5.

8. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: The computer program or instructions are executed by a processor to implement the method for predicting the progression of oral mucosal canceration based on microvessels as described in any one of claims 1 to 3, or to implement the method for predicting the progression of oral mucosal canceration based on MAPH as described in claims 4 to 5.

9. Application of substances targeting TSCs in the preparation of drugs for anti-angiogenic therapy of oral mucosal carcinogenesis; Optionally, the substance targeting TSCs acts on the recruitment of TSCs.

10. Use of an agent for inhibiting or overexpressing a gene in the preparation of a drug for anti-angiogenic treatment of oral mucosal carcinogenesis, wherein the gene comprises ANGPTL2; Optionally, the ANGPTL2 is replaced by CK15; Optionally, the ANGPTL2 is replaced by EEF1E1; Optionally, the ANGPTL2 is replaced by TIMP1; Optionally, the drug is used to regulate the mediation between TIMP1 and MMPs; The MMPs include one or more of the following: MMP1, MMP2, MMP3, MMP9, MMP10, MMP14; Optionally, the genes also include one or more of the following: CK15, EEF1E1, TIMP1.

Images