Cancer prognosis marker combination and application thereof

Abstract

The invention belongs to the technical field of medical treatment, and discloses a cancer prognosis marker combination and application thereof, and the cancer prognosis marker combination comprises the following markers: CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c and Collagen I. Compared with a model depending on high-dimensional multi-omics data, a cancer prognosis evaluation model obtained by combined training of the seven markers shows more excellent prognosis accuracy. The invention provides a mIHC (multiple immunohistochemical) cancer prognosis evaluation and detection product based on the cancer prognosis marker combination. And a satisfactory prognosis result can be obtained only by using the combination of two five markers. The method shows relatively high accuracy in the aspect of predicting the survival of the gastric cancer patient, so that the method becomes a solution which is low in cost and high in benefit and can be expanded to other cancers.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to a cancer prognosis marker combination and application thereof. Background Art

[0002] Gastric cancer (GC) is a global health concern. In 2022, it ranked fifth among the most common malignant tumors in terms of incidence and mortality. Currently, the main treatments for GC include surgery, chemotherapy, and radiotherapy, all of which have limited effectiveness in patients with advanced disease. Despite continued advancements in clinical and basic research, the five-year survival rate for patients with advanced GC remains below 40%.

[0003] Given the poor prognosis of patients with advanced gastric cancer, the tumor microenvironment (TME) has become a key factor affecting treatment efficacy and the effectiveness of treatment regimens. The TME is a complex network of cellular and non-cellular components that plays a decisive role in cancer dynamics and treatment efficacy. Macrophages play a key role in this process, significantly influencing tumor behavior and treatment response. 5 . Recent studies have revealed the complex relationship between macrophages and gastric cancer, and have highlighted new therapeutic targets and mechanisms that drive tumor progression. For example, studies have shown that under hypoxic conditions, macrophages secrete CXCL8, which activates gastric cancer invasion and proliferation through the CXCR1 / 2 and JAK / STAT1 signaling pathways, forming a feedback loop that promotes the polarization of M2 macrophages, further promoting the secretion of CXCL8, and presenting a self-enhancing tumor-promoting mechanism. In addition, a study using multiple immunohistochemistry revealed the heterogeneity of tumor-associated macrophages (TAMs) and found that their subpopulations were associated with immune signaling pathways and PD-L1 expression, which further demonstrated the prognostic significance of TAMs in gastric cancer. Existing studies have further revealed a new mechanism by which TAMs mediate tumor metastasis: M2 TAMs can also deliver ApoE through exosomes, activate the PI3K-Akt signaling pathway in recipient tumor cells, and cause metastasis of gastric cancer cells. In addition, SIGLEC10, as an immunosuppressive molecule on CD68⁺ macrophages, can inhibit T cell function, thereby bringing prognostic and therapeutic significance, making it a potential immune checkpoint.

[0004] The formation and function of tertiary lymphoid structures (TLS) in the tumor microenvironment are also of interest. These structures are similar to secondary lymphoid organs, but are located outside traditional lymphoid organs and help activate initial T / B cells, enabling them to exert tumor immunity. In studies of various malignant tumors such as colorectal cancer, lung cancer, breast cancer and malignant melanoma, the presence of TLS has been shown to be associated with a good prognosis. In gastric cancer (GC), two key studies further emphasized the important role of TLS and specific immune cell populations in enhancing the effect of cancer immunotherapy. The first study revealed that TLS and CXCL13 + CD103+ CD8 + Interactions between tissue-resident memory T cells (Trm) and T cells have been shown to enhance the efficacy of anti-PD-1 therapy. Studies have also shown that upon B cell activation, the secretion of CXCL13 and granzyme B increases through the TNFR2 axis and mTOR signaling pathway, significantly enhancing the efficacy of anti-PD-1 therapy. TLS was classified into three distinct states using machine learning. This classification demonstrates the significant distribution of TLS in gastric cancer and demonstrates that the TLS scores of different patients can be used to stratify them into different prognostic groups. Patients with higher TLS scores have significantly improved survival rates, and the TLS score remains an independent prognostic factor even after adjusting for other clinicopathological variables and tumor-infiltrating lymphocytes.

[0005] The immune microenvironment has a profound impact on the prognosis and treatment outcomes of gastric cancer. This understanding lays the foundation for the development of more personalized and effective immunotherapies, and is expected to revolutionize the management strategies of malignant tumors. However, how spatial characteristics (especially the activity of macrophages) affect the function of immune cells remains understudied. To fully analyze the complex relationship between cells and spatial composition in the tumor microenvironment (TME), advanced analytical methods with spatial resolution are needed to effectively reveal spatial heterogeneity. By further combining histology with mass cytometry technology through imaging mass cytometry (IMC), 45 metal-labeled antibodies can be detected simultaneously on a single slice, accurately presenting cell interactions and marker co-expression. Integrating multi-omics methods can overcome the resolution limitations of single-cell transcriptomes and spatial transcriptomes without sacrificing spatial information, providing a more detailed perspective on TME cell interactions, which is crucial for the development of targeted therapies. Summary of the Invention

[0006] To overcome the problems of high cost, complex technology, and difficulty in implementation in clinical settings in existing technologies, the present invention proposes a combination of cancer prognostic markers and its application. The above objectives are achieved through the implementation of the following technical solutions: A cancer prognostic marker combination comprising the following markers: CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c and Collagen I.

[0007] A use of a product for detecting the above-mentioned cancer prognosis marker combination in the preparation of a product for cancer prognosis assessment.

[0008] Optionally, the product for detecting the combination of cancer prognostic markers includes a reagent, a test strip, a kit or an instrument.

[0009] Optionally, the cancer comprises gastric cancer.

[0010] Optionally, the product for detecting a combination of cancer prognostic markers is a multiplex immunohistochemistry detection product; the multiplex immunohistochemistry detection product includes a first detection product and a second detection product; The first detection product is used to detect Group A markers, and the second detection product is used to detect Group B markers; The group A markers and group B markers respectively include 2 to 5 markers in the cancer prognosis marker combination, and each marker in the cancer prognosis marker combination is included in at least one of the group A markers or group B markers.

[0011] Optionally, the first detection product is used to detect CD68, PD-L1, PAN-CK, and HLA-DR in the sample; The second detection product is used to detect CD68, CD45, CD11c and Collagen I in the sample; The first detection product includes: CD68 primary antibody specifically binding to CD68, PD-L1 primary antibody specifically binding to PD-L1, PAN-CK primary antibody specifically binding to PAN-CK, HLA-DR primary antibody specifically binding to HLA-DR, DAPI, and horseradish peroxidase-labeled secondary antibody; The second detection product includes: Primary antibody specifically binding to CD68, primary antibody specifically binding to CD45, primary antibody specifically binding to CD11c, primary antibody specifically binding to Collagen I, DAPI, and secondary antibody labeled with horseradish peroxidase.

[0012] Optionally, the multiplex immunohistochemistry detection product further comprises an eluent and a fluorescent dye; The fluorescent dye is used to cooperate with the horseradish peroxidase-labeled secondary antibody to stain the marker; the multiple immunohistochemistry detection product contains at least 7 fluorescent dyes with different wavelengths; The eluent is used to elute the fluorescent dye bound to the sample after the sample is subjected to multiple immunohistochemical detection using the first detection product.

[0013] A method for establishing a cancer prognostic risk assessment model using a combination of cancer prognostic markers comprises the following steps: Step 1) Obtain an imaging mass spectrometry flow cytometry analysis data set, wherein the imaging mass spectrometry flow cytometry analysis data set includes the average expression intensity of each marker in the above-mentioned cancer prognosis marker combination, epithelial cell-based, HLA-DR + Macrophages, CD11c +Macrophages and Collagen I + Spatial similarity characteristics of fibroblasts and prognosis; Step 2) Analyze the data in the dataset using the imaging mass spectrometry flow cytometry technology, and obtain a cancer prognosis risk assessment model through deep neural network training.

[0014] Optionally, the epithelial cell-based, HLA-DR + Macrophages, CD11c + Macrophages and Collagen I + The spatial similarity characteristics of the four cell types of fibroblasts include the structural similarity index.

[0015] A cancer prognosis risk assessment system, comprising: The detection module is used to perform multiple immunohistochemical detection on CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c and Collagen I in the tissue sample to be tested, and obtain the multiple immunohistochemical detection results of the tissue sample to be tested; and calibrate the epithelial cells, HLA-DR in the tissue sample to be tested according to the multiple immunohistochemical detection results. + Macrophages, CD11c + Macrophages and Collagen I + The cell type of fibroblasts is used to obtain the cell type calibration result; The result evaluation module inputs the multiple immunohistochemical detection results of the tissue sample to be tested and the cell type calibration results into the cancer prognosis risk assessment model obtained above, performs cancer prognosis risk assessment, and obtains a cancer prognosis risk assessment result.

[0016] The present invention has the following beneficial effects: The present invention proposes a cancer prognostic marker combination and its application. The cancer prognostic marker combination includes the following markers: CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c, and Collagen I. A cancer prognostic assessment model trained using this seven-marker combination exhibits superior prognostic accuracy compared to models relying on high-dimensional multi-omics data.

[0017] Based on the aforementioned cancer prognostic marker combination, this paper proposes a multiplex immunohistochemistry (mIHC) cancer prognosis assessment test. Existing technologies use highly complex multi-marker approaches to predict cancer, but these methods are costly, technically complex, and difficult to implement in clinical settings. However, the model based on this paper achieves promising prognostic results using only two five-marker combinations. This method demonstrates high accuracy in predicting survival in gastric cancer patients, making it a cost-effective solution that is scalable to other cancers. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is the model building process in the embodiment; Figure 2 This is the process of acquiring data from multiplex immunohistochemistry (mIHC) testing in the embodiment; Figure 3 This is the multiplex immunohistochemistry (mIHC) test data processing process in the embodiment; Figure 4 It is the result graph of survival analysis during model validation; Figure 5 This is a comparison chart of the evaluation results of the model in the embodiment and the traditional survival prediction model based on all IMC markers; Figure 6 3 is a comparison chart of the evaluation results of the model in the embodiment and the existing prognosis evaluation model. DETAILED DESCRIPTION

[0020] Various exemplary embodiments of the present invention are now described in detail. This detailed description should not be considered as a limitation of the present invention, but should be understood as a more detailed description of certain aspects, features, and embodiments of the present invention. It should be understood that the terms used in the present invention are only for describing specific embodiments and are not intended to limit the present invention.

[0021] In addition, for numerical ranges in the present invention, it is understood that each intervening value between the upper and lower limits of the range is also specifically disclosed. Each smaller range between any stated value or stated range, and any other stated value or intervening value in the stated range, is also included in the present invention. The upper and lower limits of these smaller ranges may independently be included or excluded in the range.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention pertains. Although preferred methods and materials have been described herein, any methods and materials similar or equivalent to those described herein can also be used in the practice or testing of the present invention.

[0023] The words “include,” “including,” “have,” “contain,” etc. used in this document are open-ended terms, meaning including but not limited to.

[0024] To identify regional factors influencing prognosis and clinical characteristics in gastric cancer (GC) patients, a detailed spatial multi-omics analysis was performed on 205 GC patients. Imaging mass cytometry (IMC) analysis was performed on 205 tumors and 39 surrounding regions of interest (ROIs) from 205 GC patients. This IMC-based single-cell spatial proteomics technique enabled the identification of distinct regions within the GC tumor microenvironment (TME). Furthermore, this technique revealed other spatial features, including cell-cell interactions, such as tertiary lymphoid structures (TLS). By integrating these spatial features with clinical data, such as prognosis, the most influential spatial determinants were identified.

[0025] Imaging Mass Cytometry (IMC) Single Cell Segmentation: To perform single-cell segmentation in IMC images, the pre-trained DeepCell model was used. The DeepCell model requires two imaging channels for segmentation: a nuclear stain (typically DAPI) to identify the location of the cell nucleus; and a cell membrane or cytoplasmic region to define the cell boundaries. In this study, the membrane stain was selected as the second imaging channel for segmentation analysis.

[0026] IMC pretreatment and cell type identification: First, marker expression was transformed using the hyperbolic arcsine function, and signal values ​​for each channel were constrained to the 1% and 99% ranges to define the minimum and maximum values. Min-Max Normalization was then applied to each channel to normalize the data. To account for possible batch effects, alignment was performed using the R package Harmony (version 1.2.3). Cells were then clustered using FastPG (version 0.0.8) with a setting of 100 nearest neighbors and two rounds of clustering. In the first round of clustering, the following lineage-specific markers were used to identify the main cell populations: ECadherin, Pankeratin, CD3, CD4, CD8, CD20, CD7, CD57, CD45, CD68, CD15, CD14, CD16, CD31, αSMA, Collagen I, and Vimentin. In the second round of clustering, myeloid cell subsets were re-clustered using additional markers (e.g., FOXP3, CD16, CD69, CD4, CD8, Caspase3, B7H4, VISTA, CD7, CD103, LAG3, CD20, Granzyme B, PD-1, KI67, GATA3, CD45RA, CD3, TNFα, TL1β, CD45RO, CD57, CD25). Lymphocytes were re-clustered using markers such as IL-6, CD14, CD16, Caspase3, CD163, PD-L1, CD11B, CD11C, CD15, Ki-67, HLA-DR, TNFα, and TL1β. Epithelial cells were clustered using markers such as ECadherin, PDL1, KI67, and Pankeratin. Finally, the average expression levels of the markers in each cluster were visualized through heatmaps, and this information was used to annotate the corresponding cell types (CTs).

[0027] Identification of spatial regions / patches: First, a local spatial adjacency graph is constructed for each cell using the k-nearest neighbor (KNN) algorithm based on its spatial coordinates to calculate its 20 nearest neighbors. Next, for each cell's neighborhood, various marker expression data from neighboring cells are aggregated and clustered by cell type. Local subgraphs are then filtered to ensure that the central node of each subgraph corresponds to a specific cell type (e.g., a B cell). These subgraphs are then connected through shared nodes to form a global connectivity graph, generating spatial regions / patches. Finally, the spatial extent of an existing patch is expanded by adding n neighboring nodes to it, thereby increasing the spatial coverage of the constructed patch.

[0028] Downstream spatial analysis: Based on the spatial connectivity graph, three main regions were defined: Immune area: including "lymphocytes" and "myeloid cells"; Fibroblast area: including "myofibroblasts" and "Collagen I + fibroblasts”; Epithelial region: includes "epithelial cells".

[0029] In addition, three overlapping regions were defined: the immune-fibroblast junction, the immune-epithelial junction, and the fibroblast-epithelial junction. The number of interactions between central cells and neighboring cells was counted by calculating the average number of surrounding cells within each region of interest (ROI). Subsequently, differential analysis of cell frequency, marker expression, and cell-cell interactions within the defined regions was performed.

[0030] Multivariate survival analysis: After completing the spatial analysis, functional marker expression, cell type frequencies, and cell-cell interactions were extracted and integrated for multivariate survival analysis. Features with a proportion of zero values ​​below 50% were screened, and LASSO regression was used for cross-validation to identify nonzero coefficients. Selected variables were then incorporated into a multivariate Cox regression model, and the proportional hazards assumption (PH) was assessed using the cox.zph() function. Variables that met the PH were retained and their association with survival was assessed. Results were visualized using the ggforest() function.

[0031] Based on these analysis results, the present invention has identified a core set of spatial determinants, including interactions between macrophage subsets and fibroblasts, which have been shown to play an important role in driving tumor progression and immune evasion. These results suggest that the spatial organization of cells plays a key role in shaping tumor immunity and therapeutic response. Therefore, spatial features can be further incorporated into expression-based markers to construct survival prediction models, thereby improving the accuracy of cancer prognosis assessment. Therefore, based on the markers identified by imaging mass cytometry (IMC) analysis, the present invention proposes a more clinically promising method for predicting survival in gastric cancer (GC) patients.

[0032] Based on the markers analyzed and identified by the above-mentioned imaging mass cytometry (IMC) technology, seven markers were screened in this example, namely CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c, and Collagen I.

[0033] The expression of these seven markers can reflect four key cell types: epithelial cells, HLA-DR +Macrophages, CD11c + Macrophages and Collagen I + Interactions between fibroblasts were analyzed. Independent images were extracted from each region of interest (ROI) to generate masks corresponding to these cell types and then build a survival prediction model.

[0034] The method of building the model is as follows Figure 1 As shown, the following steps are included: From imaging mass cytometry (IMC), two types of integrated image features are designed: marker intensity features and spatial similarity features.

[0035] The marker intensity feature is the average expression intensity measurement result of seven markers in the IMC image data, including CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c, and Collagen I.

[0036] Spatial similarity features were calculated using the structural similarity index (SSIM) to evaluate the spatial similarity of epithelial cells, HLA-DR + Macrophages, CD11c + Macrophages and Collagen I + The pairwise spatial relationship of fibroblasts in IMC image data reflects the epithelial cells, HLA-DR + Macrophages, CD11c + Macrophages and Collagen I + Interactions between fibroblasts. The structural similarity index (SSIM) is a metric used to measure the similarity between two images. Unlike traditional pixel-based evaluation methods such as mean squared error (MSE) or peak signal-to-noise ratio (PSNR), SSIM takes into account image brightness, contrast, and structural information, and therefore more accurately reflects the human visual system's perception of image quality.

[0037] These features and the prognosis and survival status of patients in the dataset are input into a deep neural network, and the trained model can be used for clinical prediction of the prognosis of gastric cancer patients. Combining the marker expression intensity characteristics and spatial similarity characteristics can achieve obvious advantages. The above validation data were analyzed using existing prognostic analysis models, including RF Random Forest, GCN Graph Convolutional Network and DeepSurv Deep Survival Model. The results showed that the model of the present invention showed significant performance improvement in comparison with existing studies ( Figure 6 ).

[0038] Given that multiplex immunohistochemistry (mIHC) is a cost-effective and scalable approach for cancer survival prediction, but its ability to simultaneously detect multiple markers is limited, we further validated the generalization ability of the identified key cell types in an independent mIHC cohort ( Figure 2 ).

[0039] This embodiment proposes a product for detecting a combination of cancer prognostic markers, which is a multiplex immunohistochemistry detection product; the multiplex immunohistochemistry detection product includes a first detection product and a second detection product; The first detection product is used to detect Group A markers, and the second detection product is used to detect Group B markers; The group A markers and group B markers respectively include 2 to 5 markers in the cancer prognosis marker combination, and each marker in the cancer prognosis marker combination is included in at least one of the group A markers or group B markers.

[0040] In a specific embodiment, two five-marker panels were designed: Panel A includes CD68, PD-L1, PAN-CK, HLA-D, and DAPI, and Panel B includes CD68, CD45, CD11c, Collagen I, and DAPI. These two panels integrate key spatial features identified by multi-omics analysis.

[0041] The first detection product is used to detect CD68, PD-L1, PAN-CK and HLA-DR in the sample; the second detection product is used to detect CD68, CD45, CD11c and Collagen I in the sample.

[0042] In a specific embodiment: the first detection product includes: a CD68 primary antibody that specifically binds to CD68, a PD-L1 primary antibody that specifically binds to PD-L1, a PAN-CK primary antibody that specifically binds to PAN-CK, an HLA-DR primary antibody that specifically binds to HLA-DR, DAPI, and a secondary antibody labeled with horseradish peroxidase; The second detection product includes: a CD68 primary antibody that specifically binds to CD68, a CD45 primary antibody that specifically binds to CD45, a CD11c primary antibody that specifically binds to CD11c, a Collagen I primary antibody that specifically binds to Collagen I, DAPI, and a secondary antibody labeled with horseradish peroxidase.

[0043] The multiple immunohistochemistry detection product also includes an eluent and a fluorescent dye; The fluorescent dye is used to cooperate with the horseradish peroxidase-labeled secondary antibody to stain the marker; the multiple immunohistochemistry detection product contains at least 7 fluorescent dyes with different wavelengths; The eluent is used to elute the fluorescent dye bound to the sample after the sample is subjected to multiple immunohistochemical detection using the first detection product.

[0044] To verify the accuracy of the model based on the above 7 markers, tissue samples from 27 patients were used to validate the above markers and models.

[0045] First, imaging mass cytometry (IMC) image results of tissue samples from the 27 patients were obtained and input into the traditional survival prediction model based on all IMC markers and the cancer prognosis risk assessment model based on the training of the seven key markers in each space. The experimental results ( Figure 4 and Figure 5 ) showed that our method outperformed the traditional survival prediction model based on all IMC markers in terms of C-index (CI) and stratification performance, demonstrating the effectiveness of the identified markers in predicting GC progression.

[0046] Then, the model was validated based on the results of multiple immunohistochemistry assays.

[0047] Experiments were performed using two multiplex immunohistochemistry (mIHC) panels: Panel A: CD68, PD-L1, PAN-CK, HLA-D, and DAPI; Panel B: CD68, CD45, CD11c, Collagen I, and DAPI.

[0048] Multiplex Immunohistochemistry: In this example, multiplex immunohistochemistry (mIHC) staining was performed using the Opal Polaris™ 7-Color Manual Immunohistochemistry Kit. The steps are summarized as follows: tissue sections were deparaffinized and blocked to reduce nonspecific binding. Subsequently, sections were incubated with specific primary antibodies, followed by corresponding secondary antibodies and fluorescently labeled secondary antibody reagents for signal amplification. DAPI staining was used to counterstain the tissue and visualize labeled structures. Finally, the stained sections were analyzed using the Vectra Polaris Quantitative Pathology Imaging System (Akoya Biosciences, USA) to obtain high-resolution multiplex imaging data.

[0049] The primary antibodies used in this multicolor immunohistochemistry experiment include: Pan-CK (Biolegend, catalog number: 914204 ) HLA-DR (Abcam, catalog number: ab92511 ) CD68 (Biolegend, catalog number: 916104 ) PD-L1 (Proteintech, catalog number: 66248-1-Ig ) CD45 (CST, catalog number: 13917SF ) CD11c (Abcam, catalog number: ab52632 ) Collagen I (Abcam, catalog number: ab138492 ) Multiplex immunohistochemistry specifically includes the following steps. First, the four markers in panel A are detected: Step 1: Tissue slide preparation Prepare tissue or cells for examination using standard fixation and embedding techniques using the Opal kit. (It is recommended to run isotype control slides, substituting the corresponding isotype control for the primary antibody in each experiment.) Bake each slide in an oven at 68°C for 1.5 hours.

[0050] Dewaxing: Dewax in xylene for 20 minutes → Rehydrate in a graded series of ethanol solutions: 95% ethanol for 10 minutes → 85% ethanol for 10 minutes → 75% ethanol for 10 minutes. (This step ensures that the tissue is completely immersed in xylene and ethanol. If ethanol is insufficient, prepare additional ethanol immediately.)

[0051] Place the slides in a staining jar filled with ultrapure water and wash on a shaker three times, 5 minutes each time.

[0052] Place the tissue slide in a humidified chamber, wipe dry, and add 4% paraformaldehyde fixative (stored at -20°C) to completely cover the tissue. Fix for 30 minutes. Place the slide in a staining jar filled with ultrapure water and wash three times with ultrapure water on a shaker, 2 minutes each time.

[0053] Step 2: Citric Acid Repair Place a plastic staining jar filled with sufficient 1X citric acid repair solution in a boiling heater and preheat it for 5 minutes. Place the tissue slides in the preheated 1X citric acid repair solution.

[0054] Cover with lid, boil for 20 min, keep warm for 10 min, and cool to room temperature for 30 min. Do not let the slides dry out.

[0055] Take out the slices one by one and place them in a humidified box. Immediately use a pipette to suck 1X AR6 Buffer (citric acid repair solution) and drop it on the slices to prevent them from drying out. After the slices have cooled, rinse them with IXTBST, rinse once and then wash for 2 minutes.

[0056] Step 3: Closing Wipe dry the 1X TBST wash buffer and circle the tissue sections on the slide using a histochemical pen.

[0057] Apply blocking solution (Super Blocking, stored at 4°C) to the tissue, cover completely, and block at room temperature for 10 minutes. This protocol uses PerkinElmer Antibody Diluent / Block for blocking. Other options should be independently validated.

[0058] Step 4: Primary Antibody Incubation Drain the blocking solution, add primary antibody (needs to be prepared in advance, note that the primary antibody is different for each round) to completely cover the tissue section (usually 50uL per slide), and incubate at room temperature in the dark for one hour or at 4° overnight.

[0059] Rinse the slides with 1X TBST, wash three times with 1X TBST, 2 min each time.

[0060] Step 5: Secondary Antibody Incubation Wipe dry the 1X TBST washing buffer, add secondary antibody (ready-made, no need to prepare, stored in a 4°C refrigerator), completely cover the tissue sections, and incubate at room temperature in the dark for 10 minutes.

[0061] Rinse the slides with 1X TBST, wash three times with 1X TBST, 2 min each time.

[0062] Note: Opal polymer HRP Ms + Rb is recommended for experiments using human tissue and mouse or rabbit primary antibodies. Other options should be independently validated.

[0063] Step 6: Opal signal generation (fluorescence incubation) Wipe dry 1X TBST washing buffer, add fluorescence (needs to be prepared, stored in a -20℃ refrigerator, Opal Polaris480 for the first round of fluorescence), and incubate at room temperature for 10 minutes.

[0064] Rinse the slides with 1X TBST, wash three times with 1X TBST, 2 min each time.

[0065] Rinse slides with 1X citric acid repair solution.

[0066] Step 7: Citric Acid Repair Place a plastic staining jar containing sufficient 1X citric acid repair solution in a boiling electric heater and preheat for 5 minutes. Place the tissue slides in the preheated 1X citric acid repair solution.

[0067] Cover with lid, boil for 20 min, keep warm for 10 min, and cool to room temperature for 30 min. Do not let the slides dry out.

[0068] Take out the slices one by one and place them in a humidified box. Immediately use a pipette to suck 1XAR6 Buffer (citric acid repair solution) and drop it on the slices to prevent them from drying out. After the slices cool down, rinse them with IXTBST (rinse once and then wash for 2 minutes).

[0069] This microwave step strips the primary antibody-secondary antibody-HRP complex, allowing the introduction of the next primary antibody.

[0070] Repeat steps 3-7: Until all 4 targets are marked (the first round of fluorescence is with Opal Polaris480, the second round of fluorescence is with Opal520, the third round of fluorescence is with Opal570, and the fourth round of fluorescence is with Opal620) Step 8: Nuclear staining and sealing Drain off excess wash buffer, add DAPI (prepared by three drops of DAPI in 1 ml of 1xTBST) onto the tissue slide, and incubate at room temperature for 5 min.

[0071] Wash in 1X TBST buffer for 2 min, then wash in water for 2 min.

[0072] Drain the sections. Mount the sections with anti-fluorescence quenching mounting solution, taking care to avoid air bubbles. After staining with panel A, analyze the stained sections using the Vectra Polaris Quantitative Pathology Imaging System (Akoya Biosciences, USA) to obtain high-resolution multiplex imaging data for panel A.

[0073] After scanning, wash the slides: Soak the slides vertically in PBS for 5-10 minutes to allow the coverslip to fall off naturally. Wash the slides three times with PBS, each for 5 minutes. Pour the dye wash solution (Milead MXT-5611 wash solution) into a transparent container and place the slides in the container, ensuring that the liquid completely covers the slides. Place the container between two LED light panels and incubate at room temperature for 1 hour to wash out the fluorescent dye. Wash the slides four times with PBS, each for 5 minutes. After sectioning, perform steps 2-8 of the above method for the four markers in combination B. High-resolution multiplex imaging data for combination B is obtained.

[0074] like Figure 3As shown in the figure, to process the mIHC marker data from two panels, automated non-rigid tissue image registration was applied to the data sets. Keypoints were first detected in the detection result images of panels A and B, and then the DAPI marker keypoints of panels A and B were aligned. Using DAPI as the reference channel, spatial registration was performed using Elastix, a robust image registration tool based on the ITK library. After registration, cell segmentation, dimensionality reduction, and clustering were performed. For single-cell segmentation in the DAPI channel, a deep learning-based segmentation method was used to quantify marker expression at the single-cell level and export the data in matrix format. Cell types were also annotated using the same preprocessing and cell type identification annotation methods as those used for the IMC image data. Relevant features were extracted for validation of the deep learning model.

[0075] After selecting regions of interest (ROIs) and performing batch correction, 262,114 cells were collected from 88 ROIs (500×500 pixels) of the 27 patients used for validation, and the same four key cell types were identified. After inputting the image results into the above-mentioned survival prediction model, prognostic results were seen on the mIHC dataset ( Figure 4 and Figure 5 Notably, this method demonstrated high accuracy in predicting gastric cancer patient survival, making it a cost-effective and scalable solution. The experimental results demonstrate the potential of this method in enhancing the accuracy of gastric cancer prognostic assessment.

[0076] Based on the above method, the present invention also proposes a cancer prognosis risk assessment system, comprising: The detection module is used to perform multiple immunohistochemical detection on CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c and Collagen I in the tissue sample to be tested, and obtain the multiple immunohistochemical detection results of the tissue sample to be tested; and calibrate the epithelial cells, HLA-DR in the tissue sample to be tested according to the multiple immunohistochemical detection results. + Macrophages, CD11c + Macrophages and Collagen I + The cell type of fibroblasts is used to obtain the cell type calibration result; The result evaluation module inputs the multiple immunohistochemical detection results of the tissue sample to be tested and the cell type calibration results into the cancer prognosis risk assessment model obtained above, performs cancer prognosis risk assessment, and obtains a cancer prognosis risk assessment result.

[0077] The above system can be run on a computer device that includes one or more processors, memory, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface).

[0078] The processor may be a central processing unit, a network processor, or a combination thereof. The processor may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0079] The memory stores instructions that can be executed by at least one processor, so that the at least one processor executes the method shown in the above embodiment.

[0080] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory may include high-speed random access memory and may also include non-transient memory, such as at least one disk storage device, flash memory device, or other non-transient solid-state memory device. In some optional embodiments, the memory may include a memory remotely located relative to the processor, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0081] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A cancer prognosis marker combination, characterized in that: Includes the following markers: CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c and Collagen I.

2. Use of a product for detecting the cancer prognosis marker combination according to claim 1 in the preparation of a product for cancer prognosis assessment.

3. The use according to claim 2, characterized in that Products for detecting combinations of cancer prognostic markers include reagents, test strips, kits, or instruments.

4. The use according to claim 2, characterized in that The cancer includes gastric cancer.

5. The use according to claim 3, characterized in that The product for detecting a combination of cancer prognostic markers is a multiplex immunohistochemistry detection product; the multiplex immunohistochemistry detection product includes a first detection product and a second detection product; The first detection product is used to detect Group A markers, and the second detection product is used to detect Group B markers; The group A markers and group B markers respectively include 2 to 5 markers in the cancer prognosis marker combination, and each marker in the cancer prognosis marker combination is included in at least one of the group A markers or group B markers.

6. The use according to claim 5, characterized in that The first test product is used to detect CD68, PD-L1, PAN-CK, and HLA-DR in samples; The second detection product is used to detect CD68, CD45, CD11c and Collagen I in the sample; The first detection product includes: CD68 primary antibody specifically binding to CD68, PD-L1 primary antibody specifically binding to PD-L1, PAN-CK primary antibody specifically binding to PAN-CK, HLA-DR primary antibody specifically binding to HLA-DR, DAPI, and horseradish peroxidase-labeled secondary antibody; The second detection product includes: Primary antibody specifically binding to CD68, primary antibody specifically binding to CD45, primary antibody specifically binding to CD11c, primary antibody specifically binding to Collagen I, DAPI, and secondary antibody labeled with horseradish peroxidase.

7. The use according to claim 6, characterized in that The multiple immunohistochemistry detection product also includes an eluent and a fluorescent dye; The fluorescent dye is used to cooperate with the horseradish peroxidase-labeled secondary antibody to stain the marker; the multiple immunohistochemistry detection product contains at least 7 fluorescent dyes with different wavelengths; The eluent is used to elute the fluorescent dye bound to the sample after the sample is subjected to multiple immunohistochemical detection using the first detection product.

8. The method for establishing a cancer prognosis risk assessment model using a combination of cancer prognostic markers according to claim 1, wherein: The steps include: Step 1) Obtain an imaging mass spectrometry flow cytometry analysis data set, wherein the imaging mass spectrometry flow cytometry analysis data set includes the average expression intensity of each marker in the cancer prognosis marker combination according to claim 1, epithelial cell-based, HLA-DR + Macrophages, CD11c + Macrophages and Collagen I + Spatial similarity characteristics of fibroblasts and prognosis; Step 2) Analyze the data in the dataset using the imaging mass spectrometry flow cytometry technology, and obtain a cancer prognosis risk assessment model through deep neural network training.

9. The method for establishing a cancer prognosis risk assessment model according to claim 8, wherein: The epithelial cells, HLA-DR + Macrophages, CD11c + Macrophages and Collagen I + The spatial similarity characteristics of the four cell types of fibroblasts include the structural similarity index.

10. A cancer prognosis risk assessment system, characterized in that: include: The detection module is used to perform multiple immunohistochemical detection on CD68, PD-L1, PAN-CK, HLA-DR, CD45, CD11c and Collagen I in the tissue sample to be tested, and obtain the multiple immunohistochemical detection results of the tissue sample to be tested; and calibrate the epithelial cells, HLA-DR in the tissue sample to be tested according to the multiple immunohistochemical detection results. + Macrophages, CD11c + Macrophages and Collagen I + The cell type of fibroblasts is used to obtain the cell type calibration result; The result evaluation module inputs the multiple immunohistochemical detection results of the tissue sample to be tested and the cell type calibration results into the cancer prognosis risk assessment model obtained in claim 8 or 9 to perform cancer prognosis risk assessment and obtain a cancer prognosis risk assessment result.

Images