Breast cancer molecular typing detection method based on array sensing
Through an array-sensing method, the HOFs array sensing unit is used to interact with exosomes to generate an exosome fingerprint map, which solves the problem of time-consuming and high cost of breast cancer molecular typing detection in existing technologies, realizes rapid and accurate breast cancer molecular typing detection, and supports precision treatment.
Patent Information
- Application Number
- CN202511006537.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-22
- Publication Date
- 2025-09-23
AI Technical Summary
Existing molecular typing methods for breast cancer are time-consuming, costly, and highly invasive, resulting in low diagnostic efficiency, affecting treatment decisions and patient trust, and making it impossible to achieve immediate diagnosis.
An array-based sensing method was used to synthesize HOFs array sensing units, construct an exosome fingerprint map and train a detection model. Four HOFs@enzyme sensing units were used to interact with exosomes to generate a specific exosome fingerprint map, thereby achieving rapid and accurate detection of breast cancer molecular typing.
It has achieved rapid, non-invasive, and low-cost detection of breast cancer molecular typing with an accuracy rate of 90.63%, providing a basis for clinical treatment plans and promoting the development of precision medicine.
Smart Images
Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of disease detection, and particularly relates to a breast cancer molecular typing detection method based on array sensing. Background Art
[0002] In current clinical practice, breast cancer can be divided into four molecular subtypes based on the results of IHC (immunohistochemistry) of tissue specimens, namely luminal A, luminal B, HER2 (human epidermal growth factor receptor 2) and TNBC (triple-negative breast cancer).
[0003] In clinical practice, IHC directly affects the differential diagnosis of molecular typing due to factors such as time-consuming, high cost, invasive operation and tumor heterogeneity. At the same time, under China's current clinical diagnosis and treatment system, when patients face a long wait for puncture pathology results, in addition to increased psychological pressure, treatment delays and waste of medical resources, it will also cause a decline in patient trust and increase communication pressure, ultimately affecting treatment decisions. Therefore, the development of rapid and accurate molecular typing identification methods to achieve "sample in - result out" instant diagnosis has become an urgent need for the development of precision medicine. Its clinical application will directly shorten the diagnosis-treatment interval and is of great value in improving patient prognosis. Summary of the Invention
[0004] The purpose of the present invention is to provide a method for molecular typing of breast cancer based on array sensing in order to address the defects of the prior art.
[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for molecular typing of breast cancer based on array sensing, comprising the following steps:
[0006] (1) Synthesize HOFs array sensing unit;
[0007] (2) Construct an exosome fingerprint map, use the exosome fingerprint map to build a detection model and perform machine training;
[0008] (3) The trained detection model is used to detect the molecular typing of breast cancer in patients.
[0009] Furthermore, the HOFs array sensing unit in step (1) includes four types of HOFs@enzyme sensing units.
[0010] Furthermore, four HOFS@enzyme sensing units were prepared using the following steps:
[0011] 4,4',4'',4'''-methanetetrabenzamide tetrahydrochloride was dissolved in H2O, tetrakis(4-carboxyphenyl)methane was first dissolved in AMH, and then H2O was added to make up the volume. The above two solutions were mixed with HRP solution to prepare sensing unit 1;
[0012] 4,4',4'',4'''-methanetetrabenzamide tetrahydrochloride was dissolved in H2O, azobenzene-4,4-dicarboxylic acid was first dissolved in AMH, and then H2O was added to make up the volume. The above two solutions were mixed with HRP solution to prepare the second sensing unit;
[0013] Tetrakis(4-carboxyphenyl)methane was dissolved in H2O to prepare sensing unit three;
[0014] H4TBAPy was dissolved in DMF, and then HRP solution and H2O were added to prepare the sensing unit four.
[0015] Furthermore, the four HOFS@enzyme sensor units were stirred and centrifuged, and the precipitates were washed with ddH2O and then dried and stored in the dark.
[0016] Furthermore, step (2) specifically adopts the following steps:
[0017] 2.1 Extraction of exosomes from serum of patients with different subtypes of known immunochemical panel results;
[0018] 2.2 The absorption spectrum of exosomes is detected by HOFs array sensor units to generate absorption spectrum data sets and establish specific exosome fingerprints;
[0019] 2.3 A detection model was constructed using specific exosome fingerprints as the training set, and patients with different types of cases with known immunochemical group results were used as the test set to verify the model detection efficacy.
[0020] Furthermore, step 2.1 exosome extraction specifically adopts the following steps: collect patient serum samples, centrifuge at 3000 g to obtain the supernatant; collect 250 μL of supernatant into a sterile EP tube, mix with 63 μL ExoQuick-TC kit, place at 4°C for at least 30 min, then centrifuge at 1500 g to discard the supernatant, continue to remove all liquid at 1500 g, and resuspend with 100-500 μL PBS for later use.
[0021] Furthermore, in step 2.2, the absorption spectrum of the exosomes is detected using the HOFs array sensor unit, specifically by the following steps: the exosomes are added to the HOFs array sensor unit and mixed evenly, after incubation, TMB color development solution is added, and the UV-vis absorption light intensity at a wavelength of 650 nm is measured on a multifunctional microplate reader.
[0022] Furthermore, in step (3), the molecular typing of the patient's breast cancer is detected using exosomes derived from the patient's serum.
[0023] The proposed breast cancer molecular typing array sensing platform utilizes multifunctional HOFs materials. Since different types of exosomes have different surface proteins, nucleic acids, and physicochemical properties (such as hydrophobicity, pH, and charge), upon binding to the HOFs@enzyme composite, the internal enzyme molecules and active sites on the HOFs framework produce different signals to the substrate, thereby enabling the identification of exosomes with different molecular typing.
[0024] Compared with the prior art, the present invention has the following advantages: a HOFS@enzyme-based sensing unit was designed, and four HOFS@enzyme sensing units were constructed using ligands with different properties. The HOF sensing unit has multifunctional surface chemical properties and can interact with exosomes. Through multiple interactions, a specific exosome fingerprint is generated, which can be used to determine the molecular typing of breast cancer without the need for immunohistochemistry and genetic testing. Compared with other methods, this machine learning-based HOFS@enzyme array sensing has the advantages of fast response, easy operation and relatively low cost. Through a multi-channel sensor array, the four sensing units can be mixed and interacted with each sample simultaneously, thereby achieving high efficiency, high stability and high accuracy. In addition, due to its high sensitivity and excellent biological system recognition ability, this sensing platform may be very suitable for more complex biological systems. This method can provide a detection platform for rapid identification of breast cancer molecular typing, provide a basis for the formulation of clinical treatment plans, and promote the development of precision medicine. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 The present invention is a flowchart of an embodiment of an array-based sensing-based exosome fingerprint for breast cancer molecular typing detection.
[0026] Figure 2 This is an example of cell feasibility verification.
[0027] Figure 3 The embodiment was applied to the entire dataset (n=80, of which 48 were used for the training dataset and 32 were used for the test dataset), and the machine learning-based identification was compared with the traditional immunohistochemistry-based determination. DETAILED DESCRIPTION
[0028] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.
[0029] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or apparatuses.
[0030] A method for detecting molecular typing of breast cancer based on array sensing comprises the following steps:
[0031] S1: synthetic HOFs array sensing unit;
[0032] S2: extract exosomes;
[0033] S3: Construction of exosome fingerprint.
[0034] S1: Synthesis of HOFs array sensing unit:
[0035] Sensing unit 1: 5 mg of 4,4',4'',4'''-methanetetrabenzamide tetrahydrochloride was dissolved in 1.8 mL of H2O. 6 mg of tetrakis(4-carboxyphenyl)methane was first dissolved in 0.02 mL of AMH and then H2O was added to make the volume 2 mL. The above solution was mixed with 0.2 mL of HRP solution.
[0036] Sensing unit 2: 5 mg of 4,4',4'',4'''-methanetetrabenzamide tetrahydrochloride was dissolved in 1.8 mL of H2O. 3 mg of azobenzene-4,4-dicarboxylic acid was first dissolved in 0.02 mL of AMH and then H2O was added to make the volume 2 mL. The above solution was mixed with 0.2 mL of HRP solution.
[0037] Sensing unit 3: 3.7 mg of tetrakis(4-carboxyphenyl)methane was dissolved in 3.8 mL of H2O, followed by the addition of 3.8 mL of H2O;
[0038] Sensing unit 4: 2 mg of H4TBAPy was completely dissolved in 0.2 mL of DMF under ultrasonication, followed by the addition of 0.2 mL of HRP solution and 3.6 mL of H2O;
[0039] The above solutions were stirred at 1200 rpm for 5 minutes at room temperature, the mixture was centrifuged at 12000 rpm for 10 minutes, and the precipitate was washed three times with ddH2O, dried at 45°C, and stored in the dark.
[0040] S2-1: Extraction of cell-derived exosomes:
[0041] After the cells reached a density of 70%, the medium was replaced with serum-free medium and cultured at 37°C for an additional 48 hours using the exosome-free serum format. The supernatant was then collected into a 100 kD ultrafiltration tube and centrifuged at 4000 g for 20 minutes. The supernatant was then mixed with the extract from the ExoQuick-TC kit in a 5:1 ratio by volume and incubated at 4°C for at least 12 hours. The extract was then centrifuged at 1500 g for 30 minutes, the supernatant discarded, and the extract was centrifuged again at 1500 g for 5 minutes. The supernatant was then removed and resuspended in 100-500 μL of PBS. The exosome solution was analyzed by NTA and TEM to determine concentration, hydrated diameter, and morphology.
[0042] S2-2: Extraction of serum-derived exosomes:
[0043] Extract serum and transfer to a fresh tube. Centrifuge at 3000 g for 15 minutes at 4°C to remove residual cell debris. Add 63 μL of the ExoQuick exosome precipitation solution kit to every 250 μL of supernatant, mix thoroughly, and incubate at 4°C for 30 minutes to 1 hour to fully precipitate the exosomes. Centrifuge at 1500 g for 30 minutes at 4°C to pellet the exosomes. Aspirate the supernatant and resuspend in 100 μL of PBS until ready for use. Assay the exosome solution by NTA to determine the concentration and hydrated diameter of the isolated exosomes.
[0044] S3 includes the following steps:
[0045] S3-1: Array-based sensing research;
[0046] Each of the four sensor units was diluted with PBS to 99 μL per tube to prepare the detection solution. Subsequently, 1 μL of exosomes was added to each tube, mixed thoroughly, and transferred to a 96-well plate. After incubation for 30 minutes, TMB colorimetric solution was added to each well, and UV-vis absorbance intensity at a wavelength of 650 nm was measured on a multifunctional microplate reader. Six replicates were performed for each well.
[0047] S3-2: Fingerprint construction;
[0048] Serum samples from patients with four different molecular typings collected through the sample library were detected through array sensing to generate a data set of absorption spectra and specific exosome fingerprints, and four groups of fingerprint libraries were established respectively.
[0049] S4: Molecular typing test:
[0050] The fingerprint map generated by the test set is used as the input of the random forest model constructed by the training set. The serum exosomes of the patients in the test set are tested, and the test results are compared with the fingerprint library to obtain the patient typing judgment. This machine learning-based test result is matched with the traditional immunohistochemistry results.
[0051] Figure 1 Flowchart of array-based exosome fingerprinting for breast cancer molecular typing. Because different types of exosomes differ in surface proteins, nucleic acids, and physicochemical properties (such as hydrophobicity, pH, and charge), upon binding to the HOFs@enzyme composite, the internal enzyme molecules and active sites on the HOF framework produce distinct signals in response to the substrate. Four different probes generate specific signals for each sample. These response patterns are then converted into specific fingerprints through linear discriminant analysis, hierarchical cluster analysis, and deep learning algorithms, enabling identification of different exosome types.
[0052] The feasibility of the sensor array was initially verified using cell-derived exosomes. First, exosomes secreted by MCF-7, T47D, BT474, ZR-75-1, SKBR3, MDA-MB-453, MDA-MB-231, and HCC1806 cell lines were extracted and added to the sensor array to generate a data matrix containing four probes, four cell types, and six replicate samples.
[0053] 80 cases with known immunohistochemistry results were used, 20 patients for each classification. The patients' serum was collected, pre-treated by centrifuge and exosomes were extracted, and then detected by array sensing to generate a data set of absorption spectra. The data set was randomly divided into training and test sets for further analysis. For each patient's exosomes, an exosome fingerprint heat map was generated to show the response of the sensor unit to each exosome. The fingerprint map generated by the test set was used as the input of the random forest model constructed by the training set. This machine learning-based test result was matched with the traditional immunohistochemistry results. The results showed that the accuracy of machine learning in determining the molecular classification of breast cancer reached 90.63% ( Figure 3In our study, 80 patients with known immunohistochemistry results were included, with 20 patients for each subtype. Serum was collected from the patients, pre-processed by centrifuge, and exosomes were extracted. These exosomes were then detected using array sensing to generate a dataset of absorption spectra. The dataset was randomly divided into training and test sets for further analysis. For each patient's exosomes, an exosome fingerprint heatmap was generated to display the sensor unit's response to each exosome. To improve the model's generalization capabilities, we considered three commonly used machine learning techniques: support vector machines (SVMs), RF, and K-means algorithms to model the data and further distinguish breast cancer subtypes. RF was ultimately selected to construct a model for molecular subtyping, and the final evaluation results matched the immunohistochemistry results with a 90.63% match.
[0054] Figure 3 This study demonstrates the superior performance of exosome fingerprinting in determining the molecular subtype of breast cancer. Specifically, the accuracy for identifying TNBC and HER2-positive breast cancer was 100%. Given the similarities in biological behavior and molecular targets between Luminal A and Luminal B subtypes, the recognition efficiency of our array sensor requires further optimization. However, based on current breast cancer treatment guidelines, the ability to distinguish between hormone receptor-positive (HR+), TNBC, and HER2-positive subtypes, combined with clinical examination results, allows for personalized treatment plans for early-stage patients without compromising treatment progress. This rapid, non-invasive method is of great significance for the accurate diagnosis of breast cancer. Breast cancer, as a heterogeneous disease characterized by multiple molecular subtypes, presents numerous therapeutic challenges. These results validate the ability of our multi-channel sensor array to identify exosome fingerprints and determine breast cancer molecular subtypes.
[0055] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.
Claims
1. A method for detecting molecular typing of breast cancer based on array sensing, characterized in that The following steps are involved: (1) Synthesize HOFs array sensing unit; (2) Construct an exosome fingerprint map, use the exosome fingerprint map to build a detection model and perform machine training; (3) The trained detection model is used to detect the molecular typing of breast cancer in patients.
2. The method for molecular typing of breast cancer based on array sensing according to claim 1, characterized in that: The HOFs array sensing unit in step (1) includes four types of HOFs@enzyme sensing units.
3. The method for molecular typing of breast cancer based on array sensing according to claim 2, characterized in that: The four HOFS@enzyme sensing units were prepared by the following steps: 4,4',4'',4'''-methanetetrabenzamide tetrahydrochloride was dissolved in H2O, tetrakis(4-carboxyphenyl)methane was first dissolved in AMH, and then H2O was added to make up the volume. The above two solutions were mixed with HRP solution to prepare sensing unit 1; 4,4',4'',4'''-methanetetrabenzamide tetrahydrochloride was dissolved in H2O, azobenzene-4,4-dicarboxylic acid was first dissolved in AMH, and then H2O was added to make up the volume. The above two solutions were mixed with HRP solution to prepare the second sensing unit; Tetrakis(4-carboxyphenyl)methane was dissolved in H2O to prepare sensing unit three; H4TBAPy was dissolved in DMF, and then HRP solution and H2O were added to prepare the sensing unit four.
4. The method for molecular typing of breast cancer based on array sensing according to claim 3, wherein: The four HOFS@enzyme sensor units were stirred and centrifuged, and the precipitates were washed with ddH2O and then dried and stored in the dark.
5. The method for molecular typing of breast cancer based on array sensing according to claim 1, characterized in that: The step (2) specifically adopts the following steps: 2.1 Extraction of exosomes from serum of patients with different subtypes of known immunochemical panel results; 2.2 The absorption spectrum of exosomes is detected by HOFs array sensor units to generate absorption spectrum data sets and establish specific exosome fingerprints; 2.3 A detection model was constructed using specific exosome fingerprints as the training set, and patients with different types of cases with known immunochemical group results were used as the test set to verify the model detection efficacy.
6. The method for molecular typing of breast cancer based on array sensing according to claim 5, characterized in that: The extraction of exosomes in step 2.1 is specifically performed by the following steps: collecting patient serum samples and centrifuging them at 3000 g to obtain the supernatant; collecting 250 μL of the supernatant into a sterile EP tube, mixing it with 63 μL of the ExoQuick-TC kit, and placing it at 4°C for at least 30 minutes, then centrifuging it at 1500 g to discard the supernatant, continuing to remove all liquid at 1500 g, and resuspending it with 100-500 μL of PBS for later use.
7. The method for molecular typing of breast cancer based on array sensing according to claim 5, characterized in that: In step 2.2, the absorption spectrum of the exosomes is detected using the HOFs array sensor unit, specifically by the following steps: the exosomes are added to the HOFs array sensor unit and mixed evenly, after incubation, TMB colorimetric solution is added, and the UV-vis absorption light intensity at a wavelength of 650 nm is measured on a multifunctional microplate reader.
8. The method for molecular typing of breast cancer based on array sensing according to claim 1, characterized in that: In step (3), the molecular typing of the patient's breast cancer is detected using exosomes derived from the patient's serum.