Tumor low-resistance blood vessel recognition and positioning method and application of anti-tumor blood vessel drug benefit patient

Abstract

The invention provides a tumor low-resistance blood vessel recognition and positioning method and application of a patient benefitting an anti-tumor blood vessel drug. According to the application, the perfusion CT imaging method and the pathological tissue staining method are adopted to jointly identify and position the tumor low-resistance blood vessels of the antitumor vascular drug benefit patient, and when the tumor perfusion CT imaging mode is displayed as a fast-forward slow-out mode, the relative perfusion peak reaching time is short, the average blood vessel passing time is short, and the tumor low-resistance blood vessels can be accurately positioned. And when the lumen structure of the blood vessel on the dyed tissue section is closer to the endothelial structure of the normal blood vessel and the lumen is relatively thicker, the tumor blood vessel is identified and positioned as the tumor low-resistance blood vessel. According to the application, the characteristic of low resistance of tumor blood vessels beneficial to anti-angiogenesis drugs is put forward and demonstrated for the first time, and the existence of the tumor low-resistance blood vessels is put forward as the marker for predicting the positive curative effect of the anti-tumor angiogenesis drugs.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, in particular to the field of perfusion CT imaging technology. Background Art

[0002] Since Dr John Hunter first described angiogenesis in 1787, some German pathologists have observed the high vascularization of some human tumors since 1800, suggesting that new blood vessels may play an important pathogenic role in tumor progression. Then in 1971, Judah Folkman made the landmark proposal that tumor growth is dependent on angiogenesis. Angiogenesis refers to the process of generating new blood vessels from the existing vascular system. The vascular structure formed by endothelial cells is achieved through budding and growth. This process is regulated by stimulation or interaction between angiogenic factors. Angiogenesis is a necessary condition for the malignant progression of solid tumors. It is currently believed that inducing tumor angiogenesis is one of the fourteen major characteristics of tumors. Angiogenesis is also the most basic factor in tumor growth and metastasis. When the tumor first metastasizes, it has no blood vessels and only obtains nutrients through diffusion, and its volume does not exceed 2mm. 3 , which is in a dormant stage, which can also explain why anti-angiogenic drugs have no clinical application value in currently curable solid tumors. 3When the tumor cells begin to secrete a large amount of VEGF, which promotes the formation of tumor blood vessels. At this time, the tumor with blood supply grows rapidly and invades and metastasizes. Therefore, the tumor microenvironment (vascular environment) plays a very important role in the growth and metastasis of the tumor. The formation of tumor blood vessels is affected by many factors. The most critical factor known at present is vascular endothelial growth factor (VEGF), also known as VEGF-A, which binds to VEGFR-2 and stimulates the growth of endothelial cells. High levels of VEGF can lead to increased vascular permeability and increased interstitial pressure, and the formed tumor blood vessels are immature. Therefore, the main target of subsequent anti-angiogenic drugs is VEGF or its receptor. It can degenerate tumor blood vessels, cut off the nutrient supply of tumor cells, normalize surviving tumor blood vessels, reduce vascular permeability, reduce interstitial pressure, and enable drugs to better enter tumor tissues to play a role in killing tumors. At present, the mechanism of action of anti-angiogenic drugs mainly works by reducing the free concentration of active VEGF or destroying the VEGF receptor signaling system. Preparation types include anti-VEGF antibodies, VEGFR antibodies, soluble VEGFR and small molecule TKIs. The main targets include VEGF, VEGFR-2, VEGF, PIGF, VEGF-B, VEGFR-1, VEGFR-2, PDGFR-β, c-kit, and Flt-3. For example, bevacizumab, the first anti-angiogenic drug launched in 2004, is a recombinant humanized monoclonal antibody that has been used clinically for nearly 20 years. With the development and application of anti-angiogenic targeted drugs, new problems have also arisen, namely, the toxic reactions of anti-angiogenic drugs cannot be ignored. Hypertension, proteinuria, bleeding, perforation, thrombotic events, hand-foot syndrome, oral mucosal reactions, and severe fatigue are all unavoidable adverse reactions in the use of anti-angiogenic drugs, especially in some elderly people, who have hypertension, diabetic nephropathy, coronary heart disease after heart stent implantation, a history of cerebral infarction, and combined use of anticoagulants and antiplatelets. These patients are intolerant to conventional cytotoxic drugs, and immune drugs also have certain limitations. Anti-angiogenic drugs are often the first choice for such patients to receive anti-tumor treatment. However, due to the adverse reactions of anti-angiogenic drugs, it is necessary to continuously evaluate and balance the risks and benefits. However, due to the fact that anti-angiogenic drugs have no potential efficacy prediction targets, it is difficult to predict when conducting risk-benefit assessments. The strategy often adopted is to start with a small dose and climb up, but the problem that needs to be faced is that when the tumor progresses, it is difficult to judge whether the drug dose has not reached the therapeutic concentration, resulting in ineffective treatment or initial drug resistance. However, since most of the new blood vessels in tumors come from vascular endothelial cells, unlike the tumor itself due to the presence of some new mutations and new antigens, tumor vascular endothelial cells are not a "new organism" to the body.Although some angiogenesis stimulating factors, such as VEGF mentioned above, cannot be used as ideal biomolecular targets due to their low specificity, this has always been a bottleneck in predicting the efficacy of anti-angiogenic drugs. Therefore, it is imperative to find possible predictive markers, screen out potential effective beneficiaries, avoid meaningless drug trials and toxic exposure, and avoid delays in the disease caused by ineffective drug treatment. Summary of the invention

[0003] In order to solve the above technical problems, the first aspect of the present invention provides a method for identifying and locating tumor low-resistance blood vessels in patients who benefit from anti-tumor vascular drugs. The method uses a perfusion CT imaging method and a pathological tissue staining method to jointly locate the tumor low-resistance blood vessels in patients who benefit from anti-tumor vascular drugs. When the tumor perfusion CT imaging mode shows a fast-in-slow-out mode, a short relative perfusion peak time (5.87s VS 11.28s), and a short average blood vessel transit time (1.9s VS 3.35s), and is further verified by a tissue section staining method (the vascular lumen structure on the tissue section staining is closer to the normal vascular endothelial structure, and the lumen is relatively thicker (for example, 6-8um)), the tumor blood vessel is identified and located as a tumor low-resistance blood vessel.

[0004] Furthermore, the perfusion CT imaging method comprises the following steps:

[0005] (1) Contrast agent injection: During CT scanning, contrast agent is injected intravenously and distributed throughout the body through the bloodstream;

[0006] (2) Continuous scanning: The CT machine performs continuous, rapid, dynamic scanning of selected slices to capture the circulation of contrast agents in the body;

[0007] (3) Data collection: A large amount of data is acquired through scanning, recording the dynamic changes of the contrast agent in the scanned area and forming a perfusion image sequence;

[0008] (4) Image processing: Post-process the perfusion image to establish a time density curve, and calculate the perfusion parameters and color function diagram through a mathematical model. The perfusion parameters include mean blood flow (BF), blood volume (BV), mean transit time (MTT), peak time (TTP), and vascular permeability value / flow extraction product (FMD).

[0009] Furthermore, the mathematical model is a deconvolution algorithm, which uses an impulse residual function to calculate perfusion parameters and a color function diagram. The impulse residual function (IRF) is:

[0010]

[0011] Among them, IRF stands for impulse response function (human system function), BF is the mean blood flow, MTT is the mean transit time, FE is the vascular permeability, e is the uptake constant, Ve is the extravascular volume, HK is the outflow rate, and τ is the internal diffusion time constant.

[0012] Furthermore, the conditions for perfusion CT imaging are as follows: the equipment is a 64-row multi-angle CT scanner, the patient is in a supine position, and axial same-layer continuous dynamic scanning is performed in cine mode, with a dynamic scanning range radius of 22.5 cm, an injection rate of 5 ml / s, a delayed scan time of 6 s, 40 ml of iodine agent of 320 mg iodine / 100 ml is injected, the scanning layer thickness is 5 mm, and the scanning and reconstruction time is 1.5 s.

[0013] Furthermore, the pathological tissue staining method is HE staining combined with immunohistochemical staining or multicolor immunofluorescence staining technology.

[0014] Furthermore, the identification and positioning method further includes the construction of tumor blood vessel transparency: using a tissue transparency method to remove lipid molecules that form optical scattering in the entire tumor tissue, thereby obtaining an overall tumor tissue with relatively uniform optical properties and close to transparency; at the same time, using fluorescent microsphere perfusion and vascular wall immunofluorescence labeling methods to mark all blood vessels inside the tumor, and using a light sheet microscope to perform complete three-dimensional fluorescent imaging of the entire tumor sample, thereby obtaining three-dimensional in situ information of all marked blood vessels in the entire tumor tissue sample, and on this basis, combining an artificial intelligence image processing algorithm to obtain three-dimensional spatial quantitative information of all blood vessels inside the tumor, thereby accurately locating low-resistance blood vessels.

[0015] Furthermore, the patients who benefit from anti-tumor vascular drugs are patients with PFS>135 days.

[0016] The second aspect of the present application provides the use of tumor low-resistance vessels as a marker for predicting the positive efficacy of anti-tumor angiogenesis drugs. When tumor low-resistance vessels exist, the positive efficacy of anti-angiogenesis drugs is predicted, and when tumor vessels lose the low-resistance vessel characteristics, the resistance to anti-angiogenesis drugs is predicted.

[0017] Furthermore, the anti-angiogenesis drug is a TKI of an anti-angiogenesis drug. Specifically, the anti-angiogenesis drug is apatinib, but is not limited to this drug, and other commonly used clinical anti-angiogenesis small molecule drugs such as anlotinib, fruquintinib, regorafenib, lenvatinib, pazopanib, sunitinib, etc.

[0018] Furthermore, the tumor includes but is not limited to lung cancer, esophageal cancer, gastric cancer, colorectal cancer, liver cancer, kidney cancer, lung cancer, bile duct cancer, osteosarcoma, pancreatic cancer, endometrial cancer, cervical cancer, small intestinal malignancies, etc.

[0019] Compared with the prior art, the above technical solution has the following beneficial effects:

[0020] This application first proposes and demonstrates that tumor blood vessels that benefit from anti-vascular drugs have low resistance characteristics, and identifies and locates tumor low-resistance blood vessels through perfusion CT and conventional pathological tissue staining methods, and uses the presence of tumor low-resistance blood vessels as a marker to predict the positive efficacy of anti-tumor angiogenesis drugs. Different from the prior art that blindly pursues the search for highly specific biological labels in the blood, this application combines the characteristics of the tumor vascular endothelium itself. This application innovatively analyzes the characteristics of the tumor through perfusion imaging and assists in monitoring the efficacy of anti-vascular therapeutic drugs, providing new ideas for future new drug development (such as vasodilator drugs), and by focusing on the tumor vascular endothelial structure, it will promote the functionalization of structural characteristics and the predictability of functional characteristics in clinical transformation applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Schematic diagram showing the difference between perfusion CT using the cradle bed mode and traditional perfusion CT using the stationary examination bed mode;

[0022] Figure 2 This is an example of a perfusion curve for a good responder;

[0023] Figure 3 An example of a perfusion curve for a poor responder;

[0024] Figure 4 This is a simulation diagram of the perfusion mode;

[0025] Figure 5 This is a comparison chart of perfusion parameters between good responders and poor responders;

[0026] Figure 6 This is CD31 staining of vascular endothelial cells in liver metastases of patients with poor response (brown part);

[0027] Figure 7 This is CD31 staining of vascular endothelial cells in liver metastases of good responders (brown part). DETAILED DESCRIPTION

[0028] The advantages of the present invention are further described below in conjunction with the accompanying drawings and specific embodiments. Those skilled in the art should understand that the following specific description is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present invention.

[0029] Compared with normal blood vessels, tumor blood vessels have the characteristics of poor maturity, narrow lumen, disordered structure, disordered arrangement, poor blood flow and oxygen perfusion, and high permeability, but these characteristics are not conducive to the sustainable growth of tumors. In a prospective single-arm exploratory study (registration number: ChiCTR1900021799), it was found that the tumor perfusion CT imaging mode of patients who benefited from anti-angiogenic therapy showed a fast-in and slow-out mode, with a short relative perfusion peak time and a short average vascular transit time, which suggests that these tumor blood vessels that benefit from anti-vascular drugs have low resistance characteristics. Therefore, this application proposes for the first time: There are certain low-resistance blood vessels in solid tumors. The existence of these blood vessels results in a fast-in and slow-out mode on perfusion CT, with a short relative perfusion peak time and a short average vascular transit time.

[0030] In order to confirm the above conclusion, it is necessary to identify and locate the tumor low-resistance blood vessels of patients who benefit from anti-tumor vascular drugs. Therefore, this embodiment provides a method for identifying and locating tumor low-resistance blood vessels of patients who benefit from anti-tumor vascular drugs, which uses a perfusion CT imaging method and a pathological tissue staining method to jointly identify and locate the tumor low-resistance blood vessels of patients who benefit from anti-tumor vascular drugs. The tumors in this application include but are not limited to lung cancer, esophageal cancer, gastric cancer, colorectal cancer, liver cancer, kidney cancer and other tumors. Perfusion CT imaging has a shorter scanning time than functional magnetic resonance imaging and can scan multiple organs at one time, which is more in line with the clinical needs of overall efficacy judgment for tumor patients; compared with the high cost of PET-CT and certain radionuclide exposure, perfusion CT is inexpensive, has no radionuclide exposure, and is easier to implement and promote in clinical practice. The results of perfusion CT studies depend on the selection of acquisition parameters and mathematical perfusion models. It can study microvascular changes in angiogenesis that reflect tumor perfusion in vivo. However, in the prior art, perfusion CT has not been included in the routine clinical practice of tumor patients, mainly because it is a difficult problem to define the best CT scheme and perfusion algorithm.

[0031] When the tumor perfusion CT imaging mode shows a fast-in-slow-out mode, a short relative perfusion peak time, a short average vascular transit time, and is verified by conventional tissue staining methods, and when the vascular lumen structure on tissue section staining is closer to the normal vascular endothelial structure and the lumen is relatively thicker, the tumor vessel is identified and located as a tumor low-resistance vessel.

[0032] The specific process is as follows:

[0033] 1. First, the clinical characteristics of the population enrolled in the ChiCTR1900021799 registered clinical study were analyzed, and the characteristics related to endothelial function were collected, including blood pressure, blood lipids, blood sugar, echocardiography, blood cell count, coagulation function, right brachial artery FMD, etc. All subjects who met the inclusion and exclusion criteria received anti-tumor treatment with apatinib alone, and at the baseline of enrollment, 1 month, 3 months, and 5 months of apatinib treatment (perfusion CT examination was performed every 2 months thereafter). According to the tumor treatment effect evaluation standard RECIST1.1 standard, patients who were judged to have effective drug treatment (complete response (CR), partial response (PR), and stable disease (SD) continued to receive apatinib treatment. Those who were evaluated for disease progression or intolerable toxicity stopped apatinib treatment and were clinically judged to be out of the group. They could receive other anti-tumor treatment or best supportive care in the future. The subsequent survival time of the patients was followed up.

[0034] The anti-angiogenic drug used in the examples of this application is apatinib mesylate tablets. It is a small molecule VEGFR tyrosine kinase inhibitor developed by Jiangsu Hengrui Medicine Co., Ltd. with independent intellectual property rights. It is a new derivative of Vatalanib (PTK787), with a chemical name of methanesulfonic acid N-[4-(cyanocyclopentyl)phenyl][2[(4-pyridylmethyl)amino](3-pyridine)]formamide, molecular formula C25 H27 N5O3S, and a molecular weight of 493.58 (mesylate). Apatinib can effectively inhibit VEGFR at very low concentrations, and higher concentrations can also inhibit kinases such as platelet-derived growth factor receptor (PDGFR), c-Kit and c-Src. Activity detection found that its binding ability to VEGFR is more than 10 times stronger than PTK787. The site of action of apatinib is the intracellular ATP binding site of the tyrosine kinase receptor. Pharmacodynamic studies have shown that apatinib can inhibit VEGFR tyrosine kinase activity, block signal transduction after VEGF binding, and strongly inhibit tumor angiogenesis.

[0035] 2. According to the actual clinical efficacy of the subjects, since the clinical efficacy of most patients with late-line tumors is around SD, it is impossible to distinguish between effective and ineffective populations based on the spatial indicators for measuring the objective response rate. From the current efficacy data, the median PFS (Progression-Free-Survival, representing the time from the start of treatment to the progression of the disease or death of cancer patients) is between 2-4 months, and the median OS (Overall Survival) is basically around half a year. Which patients can really benefit from anti-angiogenic drugs, which patients are initially resistant to anti-angiogenic drugs, and how to distinguish between the two groups of people is also a difficulty in existing research. This application uses the length of PFS to distinguish between people with good and bad efficacy. According to the average PFS time of late-line treatment reported in the literature of more than 10 tumor types collected by this research group, the longest reported mPFS is 4.47 months. Therefore, this application defines patients with PFS>135 days as the anti-angiogenic drug benefit population, and patients with PFS<135 days as the non-benefit population.

[0036] 3. Sample measurement: This study is a single-sample diagnostic test. This application can predict the clinical efficacy of apatinib based on the results of the previous phase IB study combined with the current MTT value of perfusion CT (see Table 1). The estimated sensitivity is 80%, the specificity is 80%, and the allowable error is 0.2. Two-sided test, α is 0.1. The sample size N=27 was calculated using PASS15 software, and this study intends to include 27 cases as research subjects. At present, with the development trend of overall new drug or new device research, blind data collection by human sea tactics is no longer encouraged, but more inclined to more sophisticated and efficient research design and multi-dimensional verification, which saves manpower and material resources while promoting the high-level development of clinical research.

[0037] Table 1 Relationship between MTT value of perfusion CT and tumor control

[0038]

[0039] Sensitivity SEN = a / (a+c)

[0040] Specificity SPE = d / (b+d)

[0041] Accuracy ACC = (a+d) / (a+b+c+d)

[0042] Positive predictive value + PV = a / (a+b)

[0043] Negative predictive value - PV = d / (c+d)

[0044] In order to increase the statistical power, in this part of the statistical analysis, we used the number of perfused lesions of the enrolled patients as the statistical analysis object, and based on the preliminary results of the analysis of 23 lesions of 10 patients enrolled in the early stage, we considered that the MTT value has the value of predicting the clinical efficacy of apatinib. Therefore, the sensitivity is expected to be 80%, the specificity is 80%, and the allowable error is 0.1. For the two-sided test, α is 0.05. The sample size N = 34 was calculated using PASS15 software. If the sensitivity is increased to 90%, the specificity is 90%, and the allowable error is 0.1. For the two-sided test, α is 0.05. The required sample size is N = 44, and the actual statistical sample size of this study is N = 42 cases.

[0045] 4. Compare the perfusion CT characteristics of the two groups of people. The perfusion CT imaging equipment is a 64-row multi-angle CT scanner (Siemens, Germany). The patient is in a supine position, and the axial same-layer continuous dynamic scanning (cine mode). The dynamic scanning range radius is 22.5cm. The injection rate is 5ml / s, the delayed scanning time is 6s, the iodine injection is 40ml (320mg iodine / 100ml), the scanning layer thickness is 5mm, and the scanning and reconstruction time is 1.5s. The image post-processing uses Siemens' own perfusion CT post-processing platform Syngo.via. The deconvolution mathematical method is used. In medical imaging technology, the deconvolution model is a mathematical model of computer tomography perfusion. It comprehensively considers the blood flowing into the artery and outflowing from the vein according to the actual situation for mathematical calculation processing. The application of this model can help doctors better understand the changes of contrast agents in tissues and organs after contrast agent injection, thereby providing a more accurate basis for diagnosis and treatment. The non-deconvolution method uses the Fick principle, that is, the speed of contrast agent accumulation in tissues and organs is equal to the arterial inflow speed minus the venous outflow speed. It is divided into the instantaneous method and the maximum slope method. The maximum slope method (max slope method) was proposed by Peters in 1987. He believed that when the time is less than the shortest transit time, all injected contrast agents remain in the cerebral blood vessels. The premise is that there is no venous outflow from the contrast agent from the beginning of the inflow of the artery to the shortest transit time, that is, CV(t) = 0, then cerebral blood flow CBF = Q(t) maximum initial slope / Ca(t) peak height. The deconvolution method was proposed by Cenic et al. in 1999 based on the concepts of the above two non-deconvolution methods. Since the non-deconvolution method assumes that the injection rate of contrast agent is instantaneous, which is inconsistent with the actual situation, in order to obtain quantitative results of blood flow and mean transit time, the algorithm must take into account the actual injection rate of contrast agent and convert the time course data of each pixel position into the corresponding impulse residual function (IRF), or impulse response function (IRF). The deconvolution method uses the impulse residual function to calculate the venous outflow of contrast agent, comprehensively considers the inflow artery and outflow vein of perfusion, and does not require assumptions about the underlying vascular system when calculating BF, BV and MTT. It is close to the actual hemodynamics, and the calculated perfusion parameters and function diagrams can better reflect the actual situation inside the lesion. Deconvolution is the preferred mathematical model for perfusion analysis except for liver perfusion. The mathematical model selected for perfusion CT post-processing in this study is the deconvolution algorithm. The realization of perfusion CT is mainly based on the acquisition of time-density curves, and the acquisition of accurate time-density curves mainly depends on the following two points: one is that the position of voxels remains unchanged during the perfusion examination, and the other is the relatively dense dynamic acquisition of the same voxel.

[0046] Traditional perfusion is performed without the examination bed moving, and the maximum scanning width is the width of the detector. Therefore, the coverage of traditional 64-row CT perfusion is generally only about 4 cm. There are generally two strategies to expand the range of observation that can be perfused: one is to increase the width of the detector. Currently, the widest detector width can be maintained at 16 cm, which can basically cover the heart and the entire brain. The other is to use a cradle bed mode with variable pitch (adaptive 4D spiral). This mode uses the examination bed to move back and forth at a uniform speed throughout the perfusion process, thereby expanding the scanning range to areas beyond the width of the detector. See Figure 1 .

[0047] This application uses a periodic 4D spiral scanning mode (cradle mode) with variable pitch and 1.5 second sampling rate to quantitatively determine tissue flow. Their performance is equivalent to equidistant sampling using a standard dynamic scanning mode. The 100mm and 148mm ranges studied allow coverage of the entire brain or entire organ for perfusion imaging. Tumor perfusion is obtained by manually outlining the region of interest on different continuous dynamic scanning levels to obtain the corresponding parameters (VOA). Observation parameters include mean blood flow (meanBF), blood volume (BV), mean transit time (MTT), peak time (TPP), vascular permeability value / flow extraction product (FMD).

[0048] The push-residual function used in the syngo.via calculation formula is:

[0049]

[0050] Among them, BF is the mean blood flow and MTT is the mean transit time.

[0051] The perfusion results are shown in Table 2 and Figure 2-Figure 5 The bold data in Table 2 represent data with significant statistical differences.

[0052] Table 2 Perfusion parameter data of good and poor responders

[0053] Good answer Poor Respondent p.overall N＝23 N＝19 BF 60.92[22.25；211.91] 44.55[15.4；128.79] 0.211 BV 4.80[1.4；25.06] 3.35[1.71；7.48] 0.232 MTT 1.90[0.19；2.72] 3.35[1.49；3.41] 0.0003 TTD 3.83[0.28；6.35] 3.25[2.2；5.74] 0.251 FED 16.98[4.6；98.37] 16.88[5.97；28.02] 0.612 Delta_BF -18.37 -11.44 0.8925 Delta_BV -1.32 1.04 <0.0001 Delta_MTT 0.01000 -0.1800 0.0706 Delta_TTD "0.6900,n＝23" "1.080,n＝17" 0.6601 Delta_FED "-5.240,n＝23" "-5.980,n＝17" 0.6504

[0054] 5. According to Table 2 and Figure 2-Figure 5 It can be seen that the mean blood flow transit time (MTT) of patients with good response in the two groups on baseline perfusion CT was shorter (1.9s vs 3.35s), with a p value of <0.01. This suggests that patients with good anti-angiogenesis response have lower tumor vascular resistance, shorter mean blood flow transit time, and tumor blood vessels may be more similar to normal vascular structure and function.

[0055] 2. Pathological tissue staining method

[0056] according to Figure 6 and Figure 7 CD31 immunohistochemical staining of liver metastasis specimens from good responders and poor responders showed different endoscopic luminal structures.

[0057] from Figure 6 and Figure 7 The vascular endothelial structure after staining of vascular endothelial cells in tumors derived from liver metastases. Figure 6 The vascular structure is thin and occluded, with no obvious lumen structure. The surrounding tumor tissue is poorly differentiated and has obvious fibrous hyperplasia. Figure 7 The vascular structure is clear, with small tubular structures present, mostly in an open state, and the surrounding tumor tissue is arranged in a glandular duct pattern, with no obvious fibrosis. It can be seen that the vascular structure can reflect the state of the blood vessels themselves and the state of the matrix of the tumor itself (all liver metastases, magnification 200 times).

[0058] Optionally, in another preferred embodiment, the method for identifying and locating low-resistance blood vessels of tumors further includes: construction of transparent tumor blood vessels: using a tissue transparent method to remove components (such as lipid molecules, etc.) that form optical scattering in the entire tumor tissue through steps such as defatting and decolorization, thereby reducing light absorption and light scattering, and obtaining a tumor tissue with relatively uniform optical properties and close to transparency, so as to obtain better optical imaging effects. The tissue transparent step includes: ① sample fixation: in order to avoid deformation of the tissue and loss of the detection target, the sample must be fixed before transparent, but the degree of fixation needs to be controlled. If the fixation is too weak, the tissue will become soft, and if the fixation is too much, it will hinder immune labeling; generally, paraformaldehyde (PFA) and glutaraldehyde (GA) are used for tissue fixation; ② sample permeabilization (degreasing, decalcification, decolorization, dehydration or hydration are selected according to the characteristics of the tissue), for example, organic solvent type transparent method, water solvent type transparent method and hydrogel type transparent method; ③ refractive index matching: replacing tissue liquid with a high refractive index substance for refractive index matching to achieve tissue transparency. At the same time, fluorescent microsphere perfusion and vascular wall immunofluorescence labeling methods are used to label almost all blood vessels (including large blood vessels, capillaries, etc.) inside the tumor, and light sheet microscopy is used to perform complete three-dimensional fluorescence imaging of the entire tumor sample, thereby obtaining three-dimensional in situ information on the shape, size (including diameter, length, etc.), direction and distribution of all labeled blood vessels in the entire tumor tissue sample. On this basis, combined with the artificial intelligence image processing algorithm commonly used in the field of tissue transparency technology, the spatial quantitative information of almost all blood vessels inside the tumor is obtained, and low-resistance blood vessels are found and located, providing a basis for further research on the spatial distribution law of low-resistance blood vessels, omics characteristics, and the relationship with tumor occurrence and treatment effects, thereby accurately locating low-resistance blood vessels.

[0059] The content of this embodiment can demonstrate the existence of low-resistance blood vessels, so tumor low-resistance blood vessels can be used as markers to predict the positive efficacy of anti-tumor angiogenesis drugs. The characteristics of low-resistance blood vessels are related to relatively benign tumor biological behavior. The presence of low-resistance blood vessels can positively predict the efficacy of anti-angiogenesis drugs. When tumor blood vessels lose their low-resistance blood vessel characteristics, they often indicate drug resistance. Therefore, low-resistance blood vessels can be used as efficacy-related predictive markers. It can not only provide new ideas for future new drug development (such as vasodilator drugs), but also promote the functionalization of structural characteristics and the predictability of functional characteristics through focusing on the tumor vascular endothelial structure. Clinical transformation and application.

[0060] It should be noted that the embodiments of the present invention have better practicability and do not impose any form of limitation on the present invention. Any technician familiar with the field may use the technical content disclosed above to change or modify it into an equivalent effective embodiment. However, any modification or equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for identifying and locating low-resistance blood vessels in tumors of patients who benefit from anti-tumor vascular drugs, characterized in that: Perfusion CT imaging and pathological tissue staining methods are used together to identify and locate the tumor low-resistance vessels in patients who benefit from anti-tumor vascular drugs. When the tumor perfusion CT imaging mode shows a fast-in and slow-out mode, a short relative perfusion peak time, and a short average vascular transit time, and is further verified by the tissue section staining method, the tumor vessel is identified and located as a tumor low-resistance vessel.

2. The method for identifying and locating low-resistance tumor blood vessels in patients who benefit from anti-tumor blood vessel drugs according to claim 1, characterized in that: The perfusion CT imaging method comprises the following steps: (1) Contrast agent injection: During CT scanning, contrast agent is injected intravenously and distributed throughout the body through the bloodstream; (2) Continuous scanning: The CT machine performs continuous, rapid, dynamic scanning of selected slices to capture the circulation of contrast agents in the body; (3) Data collection: A large amount of data is acquired through scanning, recording the dynamic changes of the contrast agent in the scanned area and forming a perfusion image sequence; (4) Image processing: Post-process the perfusion image to establish a time density curve, and calculate the perfusion parameters and color function diagram through a mathematical model. The perfusion parameters include mean blood flow (BF), blood volume (BV), mean transit time (MTT), peak time (TTP), and vascular permeability value / flow extraction product (FMD).

3. The method for identifying and locating low-resistance tumor blood vessels in patients who benefit from anti-tumor blood vessel drugs according to claim 2, characterized in that: The mathematical model is a deconvolution algorithm, which uses a push residual function to calculate perfusion parameters and a color function diagram. The push residual function is: Where IRF stands for impulse response function, BF is the mean blood flow, MTT is the mean transit time, FE is the vascular permeability, e is the uptake constant, Ve is the extravascular volume, HK is the outflow rate, and τ is the internal diffusion time constant.

4. The method for identifying and locating low-resistance blood vessels of tumors in patients who benefit from anti-tumor blood vessel drugs according to claim 2, characterized in that: The conditions for perfusion CT imaging are as follows: the equipment is a 64-row multi-angle CT scanner, the patient is in a supine position, and axial same-layer continuous dynamic scanning is performed in cine mode. The dynamic scanning range radius is 22.5 cm, the injection rate is 5 ml / s, the delayed scanning time is 6 s, 40 ml of iodine agent of 320 mg iodine / 100 ml is injected, the scanning layer thickness is 5 mm, and the scanning and reconstruction time is 1.5 s.

5. The method for identifying and locating low-resistance blood vessels of tumors in patients who benefit from anti-tumor blood vessel drugs according to claim 1, characterized in that: The pathological tissue staining method is HE staining combined with immunohistochemical staining or multicolor immunofluorescence staining technology.

6. The method for identifying and locating low-resistance blood vessels of tumors in patients who benefit from anti-tumor blood vessel drugs according to claim 1, characterized in that: The identification and positioning method further includes a tumor blood vessel transparency method: using a tissue transparency method to remove components that form optical scattering in the entire tumor tissue, thereby obtaining an overall tumor tissue with relatively uniform optical properties and close to transparency; at the same time, using fluorescent microsphere perfusion and vascular wall immunofluorescence labeling methods to label all blood vessels inside the tumor, and using a light sheet microscope to perform complete three-dimensional fluorescent imaging of the entire tumor sample, thereby obtaining three-dimensional in situ information of all labeled blood vessels in the entire tumor tissue sample, and on this basis, combining an artificial intelligence image processing algorithm to obtain three-dimensional spatial quantitative information of all blood vessels inside the tumor, thereby accurately locating low-resistance blood vessels.

7. The method for identifying and locating low-resistance blood vessels of tumors in patients who benefit from anti-tumor blood vessel drugs according to any one of claims 1 to 6, characterized in that: The patients who benefited from anti-tumor vascular drugs were those with PFS>135 days.

8. Application of tumor low-resistance blood vessels as a predictive marker for the positive efficacy of anti-tumor angiogenesis drugs.

9. The use according to claim 8, characterized in that The anti-tumor angiogenesis drugs are apatinib, anlotinib, fruquintinib, regorafenib, lenvatinib, pazopanib, and sunitinib.

10. The use according to claim 8, characterized in that The tumors are lung cancer, esophageal cancer, gastric cancer, colorectal cancer, liver cancer, kidney cancer, lung cancer, bile duct cancer, osteosarcoma, pancreatic cancer, endometrial cancer, cervical cancer, and small intestinal malignancies.

Images