Use of immune cell subpopulations in the preparation of a test product for assessing the efficacy of neoadjuvant therapy and the risk of relapse and metastasis in breast cancer

Abstract

This application relates to the field of biotechnology, specifically disclosing the application of immune cell subsets in the preparation of detection products for assessing the efficacy of neoadjuvant therapy and the risk of recurrence and metastasis in breast cancer. The detection method described in this application is non-invasive and convenient, requiring only peripheral blood, and is easy to repeat sampling and dynamic monitoring. This application also provides a predictive model for the efficacy of neoadjuvant therapy in breast cancer, and a predictive model for the risk of recurrence and metastasis in breast cancer treated with neoadjuvant therapy. The constructed predictive model has a high AUC value (0.82-0.87), and good sensitivity and specificity. Furthermore, through regular postoperative monitoring, abnormal dynamic changes in immune cell subsets can be detected, potentially indicating signs of recurrence earlier than imaging and traditional tumor markers. This helps identify advantageous populations before treatment and high-risk populations after treatment, enabling optimization of treatment plans and individualized adjustment of follow-up density.

Description

Technical Field

[0001] This application relates to the field of biotechnology, and in particular to the application of immune cell subsets in the preparation of diagnostic products for assessing the efficacy of neoadjuvant therapy for breast cancer and the risk of recurrence and metastasis. Background Technology

[0002] Breast cancer is the most common malignant tumor among women worldwide. Neoadjuvant therapy has become one of the standard treatment modalities for locally advanced breast cancer. Its primary goal is to downstage the tumor to improve the surgical resection rate and the chance of breast-conserving surgery. A deeper pursuit is to improve the long-term survival prognosis of patients by inducing pathological complete response (pCR).

[0003] Currently, clinical methods for assessing the efficacy of tumor resection (NAT) and predicting prognosis mainly rely on the following aspects, but all have significant limitations. First, while routine imaging examinations (such as ultrasound and MRI) can monitor changes in tumor size, they are difficult to accurately distinguish between active tumors and post-treatment changes, resulting in assessment lag, and their correlation with the final pCR is not ideal. Second, although postoperative pathology is the "gold standard" for assessing pCR, it is a retrospective judgment and cannot be used for pre-treatment prediction or dynamic adjustments during treatment. Pre-treatment biopsy can provide static information such as molecular subtyping, but it cannot reflect the dynamic interaction between the tumor and the immune system under treatment stress. Third, widely used serum tumor markers (such as CEA and CA15-3) have insufficient sensitivity in early / non-metastatic breast cancer, and their levels usually rise later than radiographic recurrence, failing to provide early warning.

[0004] In recent years, numerous studies have attempted to identify biomarkers for predicting the efficacy of tumor-associated disease (NAT) treatment, including tumor-infiltrating lymphocytes (TILs) and certain gene expression profiles. However, most of these indicators are based on single pre-treatment tests (such as biopsy tissue) and represent static assessments. The interaction between tumors and the immune system is a highly dynamic process, especially under the pressure of systemic therapy, where the host's overall immune status undergoes dramatic and complex changes. A pre-treatment "static snapshot" is insufficient to fully capture the crucial prognostic information inherent in this dynamic evolution. Furthermore, repeated tissue biopsies are invasive, patient compliance is poor, and continuous, dynamic monitoring is difficult to achieve.

[0005] In summary, there is an urgent clinical need for a biological monitoring system that can non-invasively, dynamically, and in real time reflect the host's overall anti-tumor immune status and is closely related to treatment efficacy and long-term prognosis. Summary of the Invention

[0006] The purpose of this application is to overcome the shortcomings of the prior art and provide an application of immune cell subsets in the preparation of a detection product for assessing the efficacy of neoadjuvant therapy and the risk of recurrence and metastasis in breast cancer. This application achieves dynamic prediction of the efficacy of neoadjuvant therapy (NAT) in breast cancer and accurate assessment of prognostic risk by systematically and quantitatively detecting immune cell subsets in peripheral blood at multiple time points and utilizing the mathematical model constructed therefrom.

[0007] To achieve the above objectives, the technical solution adopted in this application is as follows: This application provides the application of immune cell subsets in the preparation of a detection product for assessing the efficacy of neoadjuvant therapy and the risk of recurrence and metastasis in breast cancer, wherein the immune cell subsets include at least one of CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD8+HLA-DR+T / CD3+T, and Treg / CD4+T.

[0008] Extensive experiments have shown that the aforementioned immune cell subsets can serve as biomarkers for the efficacy of neoadjuvant therapy in breast cancer and the risk of recurrence and metastasis.

[0009] As a preferred embodiment of the application described in this application, the immune cell subset includes any one of (1) to (3): (1) Combinations of CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T; (2) Combinations of Treg / CD4+T and CD3+HLA-DR+T / CD3+T; (3) CD3+CD8+HLA-DR+T / CD3+T.

[0010] Among them, the increased proportions of CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T in peripheral blood before treatment (baseline) were independent predictors of patients achieving pathological complete remission (pCR). At the two key time points of neoadjuvant therapy (preoperative) and surgery (within 1 year), the dynamic changes of the following indicators were significantly associated with patient survival outcomes: increased proportions of CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD8+CD69+T / CD3+T, and CD3+CD8+HLA-DR+T / CD3+T predicted better disease-free survival (DFS) and overall survival (OS).

[0011] Furthermore, an elevated Treg / CD4+ T (regulatory T cell) ratio predicts worse disease-free survival (DFS) and overall survival (OS).

[0012] As a preferred embodiment of the application described in this application, the time points for the neoadjuvant therapy include the baseline period before the start of neoadjuvant therapy, the period before surgery after the completion of neoadjuvant therapy, and the period within one year after surgery after the completion of neoadjuvant therapy.

[0013] In a preferred embodiment of the application described in this application, the source of the immune cell subset includes peripheral blood.

[0014] In some specific implementations, only 50 μL of anticoagulated whole blood is required, which greatly reduces the amount of blood drawn for cancer patients, and is especially important for patients who need regular check-ups.

[0015] Peripheral blood, as an easily reproducible sample source, contains circulating immune cells, particularly T lymphocyte subsets with diverse functions, considered a "liquid biopsy" window reflecting the body's immune status and the activity of anti-tumor immune responses. Therefore, this application systematically analyzes the dynamic changes of specific T lymphocyte subsets in peripheral blood of breast cancer patients before, during, and after neoadjuvant therapy. It identifies a combination of immune biomarkers with excellent predictive and prognostic value and constructs corresponding mathematical prediction models, thus providing a novel solution for the precise implementation and comprehensive management of neoadjuvant therapy for breast cancer.

[0016] This application also provides the application of reagents for detecting the content level of immune cell subsets in the preparation of products for assessing the efficacy of neoadjuvant therapy and the risk of recurrence and metastasis in breast cancer, wherein the immune cell subsets include any one of (1) to (3): (1) Combinations of CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T; (2) Combinations of Treg / CD4+T and CD3+HLA-DR+T / CD3+T; (3) CD3+CD8+HLA-DR+T / CD3+T.

[0017] This application also provides a predictive model for the efficacy of neoadjuvant therapy for breast cancer. The predictive model is based on baseline CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T levels prior to neoadjuvant therapy, and constructs a logistic regression model. P=1 / (1+e^(-(0.814×CD3+CD69+T / CD3+T+0.092×CD3+HLA-DR+T / CD3+T -5.378))).

[0018] This application uses the above-mentioned predictive model for the efficacy of neoadjuvant therapy for breast cancer. The results are: AUC=0.87, sensitivity 81.8%, specificity 76.2%, and threshold P=0.21.

[0019] This application also provides a predictive model for the risk of recurrence and metastasis after neoadjuvant therapy plus radical surgery for breast cancer, the predictive model including a predictive model before surgery after completion of neoadjuvant therapy and a predictive model within 1 year after surgery; Preoperative prediction models after neoadjuvant therapy are based on Treg / CD4+T and CD3+HLA-DR+T / CD3+T levels, constructing a logistic regression model: P=1 / (1+e^(-(0.540×Treg / CD4+T-0.130×CD3+HLA-DR+T / CD3+T-1.843))); or, The predictive model within one year post-surgery is based on CD3+CD8+HLA-DR+T / CD3+T and CA125 levels, and a logistic regression model is constructed: P=1 / (1+e^(-(-0.086×CD3+CD8+HLA-DR+T / CD3+T+0.254×CA125 -2.750))).

[0020] The results of the predictive model used in this application for neoadjuvant therapy before surgery were: AUC=0.82, sensitivity 87.5%, specificity 85.0%, and threshold P=0.23.

[0021] The results of the prediction model used in this application within 1 year after surgery are: AUC=0.87, sensitivity 91.7%, specificity 84.7%, and threshold P=0.23.

[0022] This application also provides a system for assessing the efficacy of neoadjuvant therapy for breast cancer and the risk of recurrence and metastasis. The system includes a sample collection module, a detection and analysis module, a data input and calculation module, and a result output module. The sample collection module is used to collect peripheral blood samples from breast cancer patients during the baseline period before starting neoadjuvant therapy, before surgery after completing neoadjuvant therapy, and within 1 year after surgery after completing neoadjuvant therapy. The detection and analysis module is used to detect the levels of immune cell subsets and serum tumor markers. The data input and calculation module is used to input the proportion of immune cell subsets and the level of serum tumor markers into the predictive model of the efficacy of neoadjuvant therapy for breast cancer or the predictive model of the risk of recurrence and metastasis of neoadjuvant therapy for breast cancer, and to calculate the probability of pathological complete remission P or the probability of recurrence / metastasis P. The result output module is used to comprehensively evaluate the efficacy of neoadjuvant therapy and the long-term risk of recurrence and metastasis of the patient based on the calculated pathological complete remission probability P or recurrence / metastasis risk probability P, combined with clinical information. The immune cell subsets include at least one of CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD8+HLA-DR+T / CD3+T, and Treg / CD4+T; The serum tumor markers include CA125.

[0023] As a preferred embodiment of the assessment system described in this application, if the probability of pathological complete remission P is higher than 0.21, it is predicted that the patient has a high probability of achieving pathological complete remission; if the probability of relapse / metastasis risk P is higher than 0.23, it is predicted that the patient has a high probability of relapse / metastasis.

[0024] The system provided in this application for assessing the efficacy and recurrence / metastasis risk of neoadjuvant therapy for breast cancer detects the proportion of specific T lymphocyte subsets in peripheral blood samples at at least one predetermined time point during the neoadjuvant therapy process for breast cancer patients. The results are then input into a predetermined assessment model to output assessment information regarding the probability of pathological complete remission (pCR) and / or the risk of recurrence / metastasis. The core of this assessment system lies in identifying and applying a series of immunobiological biomarkers with clear predictive and early warning value at different stages of treatment. By constructing a multivariate statistical model, it achieves early prediction of individualized treatment response in breast cancer patients and accurate assessment of long-term recurrence / metastasis risk.

[0025] This application also relates to a single-tube 9-color flow cytometry assay for the precise differentiation and identification of activated T cell subsets. The core advantage of this assay is that it requires only a single flow cytometer run to simultaneously detect nine differentiation antigens on the T cell membrane surface, providing a comprehensive and accurate analysis of T cell subsets and their activation status. The procedure is simple and rapid, taking only 30 minutes from sample preparation to flow cytometry analysis. Specifically, it includes instrument quality control, sample processing and staining, flow cytometry analysis, and data analysis and gating strategies.

[0026] Compared with the prior art, this application has the following beneficial effects: This application provides an application of immune cell subsets in the preparation of a detection product for assessing the efficacy of neoadjuvant therapy and the risk of recurrence and metastasis in breast cancer. The detection method of this application is non-invasive and convenient, requiring only peripheral blood, and is easy to repeat sampling and dynamic monitoring. This application also provides a predictive model for the efficacy of neoadjuvant therapy in breast cancer and a predictive model for the risk of recurrence and metastasis in breast cancer treated with neoadjuvant therapy. The constructed predictive model has a high AUC value (0.82-0.87) and good sensitivity and specificity. Furthermore, through regular postoperative monitoring, abnormal dynamic changes in immune cell subsets can be detected, which may indicate signs of recurrence earlier than imaging and traditional tumor markers. This helps to identify the dominant population before treatment and the high-risk population after treatment, enabling the optimization of treatment plans and individualized adjustment of follow-up density. Attached Figure Description

[0027] Figure 1 The graph shows the prediction model results for pathological complete remission (pCR) after neoadjuvant therapy; Figure 2 The results of the predictive model for recurrence and metastasis after NAT surgery and before surgery (A and B) and after surgery (C and D) are shown in the figure. Figure 3 Internal validation and decision curve analysis (DCA) plot for the predictive model; Figure 4 This is a distribution map of sample collection time points within the first year after surgery; Figure 5 Postoperative T lymphocyte subset disease-free survival curve analysis diagram Figure 6 A graph showing the results of a longitudinal analysis of immune cell subsets in patients with relapse / metastasis and those without relapse / metastasis during a 3-year follow-up period. Figure 7 This is a flowchart for predicting the efficacy of NAT treatment and 3-year prognosis in breast cancer based on T-cell activation analysis. Detailed Implementation

[0028] To better illustrate the purpose, technical solution, and advantages of this application, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0029] In the following examples and comparative examples, unless otherwise specified, the experimental methods used are conventional methods, and the materials and reagents used are commercially available unless otherwise specified. Furthermore, the raw materials used in each parallel experiment are the same.

[0030] The embodiments of this application follow a logically clear closed-loop process: Step 1: Establishing the research cohort and developing a sample strategy: Identify the target population and conduct peripheral blood sampling at three fixed time points with clear clinical significance.

[0031] Step 2: Quantitative detection of key biomarkers: Using standardized multicolor flow cytometry, key T-cell immune subsets in blood samples are precisely quantified.

[0032] Step 3: Data integration and modeling analysis: Integrate the test data with clinical pathology information and apply pre-established statistical models for calculation.

[0033] Step 4: Results Output and Clinical Interpretation: Generate quantitative prediction and risk assessment reports to provide direct evidence for clinical decision-making.

[0034] Step 5: Dynamic monitoring and early warning: Continuous monitoring of high-risk or high-risk patients, and early risk warning by observing the dynamic changes in immune subsets.

[0035] To ensure the feasibility of the scheme and the quality of the data, this application has strictly defined the implementation targets and sampling timing.

[0036] 1. Selection criteria: Based on a single-center, retrospective observational cohort study design. Enrolled patients were women aged 18-75 years with pathologically confirmed primary invasive breast cancer, all of whom had completed standard NAT and subsequent radical surgery at our institution.

[0037] To exclude confounding factors, the exclusion criteria were clearly defined as follows: non-ductal breast cancer; extensive distant metastasis; any preoperative treatment other than neoadjuvant chemotherapy; other primary malignancies or a history of cancer within the past 3 years; comorbidities such as severe infection or active autoimmune disease that may significantly interfere with immune assessment; patients who are pregnant or lactating; and patients with missing key clinical data.

[0038] 2. Key Sampling Time Points: Peripheral blood sample collection is strictly controlled within three time windows that are crucial for NAT treatment decisions and assessments: 1) T1 (baseline period): Before the initiation of any neoadjuvant therapy (chemotherapy, targeted therapy, etc.). Data at this time point reflects the patient's baseline immune status before treatment and is used to predict initial efficacy.

[0039] 2) T2 (Postoperative): After completing all planned cycles of NAT and before radical surgery. Data at this time point reflects the patient's systemic immune response to NAT and is crucial for assessing NAT treatment response and short-term prognosis.

[0040] 3) T3 (Postoperative Monitoring Period): During regular follow-up examinations within one year after successful surgery. Data at this time point reflects immune homeostasis during the postoperative recovery period and is of core value in predicting the long-term risk of recurrence.

[0041] All samples were collected using EDTA anticoagulant tubes and processed within a specified time after collection to ensure cell viability.

[0042] Example 1 This embodiment employs a rigorously validated and standardized single-tube 9-color flow cytometry assay for the precise differentiation and identification of activated T cell subsets. The core advantage of this assay is that it requires only a single flow cytometer run to simultaneously detect nine differentiation antigens on the T cell membrane surface, providing a comprehensive and accurate analysis of T cell subsets and their activation status. The procedure is simple and rapid, taking only 30 minutes from sample preparation to flow cytometry analysis. Specifically, it includes instrument quality control, sample processing and staining, flow cytometry analysis, and data analysis and gating strategies.

[0043] 1. Quality control procedures: 1.1 Routine performance quality control of the instrument (calibration microsphere method): To ensure that the flow cytometer maintains optimal and consistent performance throughout the study, standardized quality control procedures are performed before each experiment: • Optical and Flow Path Calibration: Before daily operation, use Flow-Check Pro fluorescent microspheres (Beckman Coulter, Cat. No. A63493) for calibration. The software automatically detects the laser delay time and coefficient of variation (CV) of the microsphere population in each channel. The CV value of each channel should typically be <2-3%. The microsphere signal position is compared with the instrument's historical baseline record to ensure the stability of the optical and flow path systems. Any significant deviation triggers instrument maintenance and recalibration.

[0044] • Fluorescence channel standardization: Flow-Set Pro fluorescent microspheres (Beckman Coulter, Cat. No. A63492) were used for detection. These microspheres have preset target average fluorescence intensity values ​​in each fluorescence channel. Based on the measured MFI values ​​of the microspheres, the photomultiplier tube voltage was adjusted to match the target values ​​for each channel, ensuring batch-to-batch consistency of fluorescence intensity.

[0045] 1.2 Experimental Procedure Quality Control (Commercial Cell Quality Control Method): To comprehensively monitor the stability and sensitivity of the staining and experimental procedures, two levels of commercially available multi-level cell quality control samples were processed in parallel for each batch of experiments: • Process stability quality control: Using IMMUNO-TROL cells (Beckman Coulter), after resuspending them according to the instructions, they were stained and detected in the same batch as patient samples using the same antibody cocktail, incubation conditions, hemolysis and washing steps, and instrument acquisition parameters to monitor the stability of the entire experimental process.

[0046] • Sensitivity control: IMMUNO-TROL Low Cells (Beckman Coulter) were used, with the same processing method as above, to evaluate the sensitivity of the detection method for low-expression cell subpopulations.

[0047] 2. Sample processing and antibody staining: Take 50 μL of anticoagulated whole blood, add pre-prepared fluorescently labeled monoclonal antibody cocktails, and incubate at room temperature in the dark for 15 minutes. Then, add 1 mL of OptiLyse C hemolysin (Beckman Coulter Inc.), and incubate at room temperature in the dark for 10 minutes to lyse red blood cells. After lysis, wash once with 1 mL of phosphate buffer, centrifuge at 300×g for 5 minutes, and discard the supernatant. Finally, resuspend in 500 μL of PBS and immediately analyze.

[0048] The fluorescently labeled monoclonal antibodies used in this protocol are as follows: anti-human HLA-DR-FITC (clone L243), anti-human CD25-APC (clone 2A3), and anti-human CD38-V450 (clone HB7) were purchased from BD Biosciences; anti-human CD3-ECD (clone UCHT1), anti-human CD127-PC7 (clone R34.34), anti-human CD4-PC5.5 (clone 13B8.2), and anti-human CD45-KO (clone J33) were purchased from Beckman Coulter Inc.; and anti-human CD69-PE (clone FN50) and anti-human CD8-APC-CY7 (clone SK1) were purchased from Zhejiang Zhengxi Biotechnology Co., Ltd.

[0049] 3. Flow cytometry analysis: All stained samples were analyzed using a 10-color flow cytometer (Navios, Beckman Coulter Inc.), with data acquisition performed using the calibrated and validated instrument parameters described above.

[0050] 4. Data Analysis and Gating Strategy: Data analysis was performed using Kaluza 2.1.1 software (Beckman Coulter Inc.), employing the following hierarchical gating strategy to accurately distinguish T cell subsets and their activation states: • Total nucleated cell identification: Based on the physical properties of cells, gating is performed on the scatter plot of side-scattered light (SSC) versus front-scattered light area (FSC-A) to exclude cell debris and very small particles, while retaining all nucleated cells (including lymphocytes, monocytes and granulocytes).

[0051] • Leukocyte enrichment: In a population of nucleated cells, a scatter plot of CD45 fluorescence intensity is used to gate the cells and remove CD45-negative non-target cells (such as platelets and non-hematopoietic cells), thereby enriching the population of CD45-positive leukocytes.

[0052] • Lymphocyte definition: Within the CD45+ leukocyte population, the lymphocyte phylum (Lym) is defined based on the lymphocyte-specific characteristics of low lateral angle scattering (low SSC) and high CD45 expression.

[0053] • Total T cell identification: Using lymphocytes as the parent cell line, CD3+ T cells were identified by scatter plot of CD3 fluorescence intensity using SSC.

[0054] • T cell subset sorting and activation analysis: All subsequent analyses were performed based on CD3+ T cells. Specifically, this included: CD4+ and CD8+ T cell subsets: further distinguishing helper / inducing T cells (CD4+) and cytotoxic / suppressive T cells (CD8+) from CD3+ T cells.

[0055] Activation biomarker analysis: The expression of various activation-related biomarkers, including CD25, CD38, CD69, and HLA-DR, was detected in the CD3+ T cell phylum, CD4+ T cell phylum, and CD8+ T cell phylum.

[0056] • Precise identification of regulatory T cells: Within the CD4+ T cell phylum, regulatory T cells are further precisely identified through a combination of markers, with their phenotype defined as CD4+CD25+CD127-.

[0057] This test yields fine T-subgroup indices, which can be used to assess the efficacy of neoadjuvant therapy and the risk of recurrence and metastasis in breast cancer.

[0058] Definition of detection markers: 1) Total T cells and subsets: CD3+ (total T cells), CD3+CD4+ (helper T cells), CD3+CD8+ (cytotoxic T cells).

[0059] 2) Early activation marker: CD69, which represents an early activation indicator of T cells, meaning that T cells can be detected as early as 6 hours after activation.

[0060] 3) Mid-term activation marker: CD25, the α chain of interleukin-2 receptor (IL-2R), represents T cells entering the mid-term activation state and is also a Treg marker.

[0061] 4) Late activation / effect marker: HLA-DR, which indicates that T cells are fully activated and have entered the effector state.

[0062] 5) Persistent activation / depletion marker: CD38, high expression may be associated with T cell functional depletion, apoptosis or immune escape.

[0063] 6) Regulatory T cell marker: CD4+CD25+Foxp3+, defined as classic regulatory T cells (Tregs), which have immunosuppressive functions.

[0064] Example 2 1. Efficacy prediction model (for pCR): ① Data Preparation: Baseline (T1) immune subset proportions and tumor marker data were used as candidate predictive variables. Initial screening immune markers included: T cells, B cells, NK cells, CD4+ T cells, CD8+ T cells, CD4+ / CD8+ ratio, Treg / CD4+ T, and activation-related markers (CD3+CD25+T / CD3+T, CD3+CD38+T / CD3+T, CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD4+CD25+T / CD3+T, CD3+CD4+CD3+). 8+T / CD3+T, CD3+CD4+CD69+T / CD3+T, CD3+CD4+HLA-DR+T / CD3+T, CD3+CD8+CD25+T / CD3+T, CD3+CD8+CD3 8+T / CD3+T, CD3+CD8+CD69+T / CD3+T, CD3+CD8+HLA-DR+T / CD3+T); tumor markers CEA, CA125, CA15-3, CYFRA21-1.

[0065] ② Variable screening: First, non-parametric tests (such as the Mann-Whitney U test) were used to compare the differences of each indicator between the pCR group and the non-pCR group. Indicators with statistically significant differences (P<0.05) between the two groups were screened, including: CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, and CD3+CD8+HLA-DR+T / CD3+T.

[0066] ③ Model construction: Univariate logistic regression analysis was performed on the above differential indicators. Variables with statistical significance (P<0.05) in the univariate analysis, as well as clinical confounding factors (including age, T stage, TNM stage, molecular subtype, number of lesions, pathological grade, and NAT protocol type) were included in the multivariate logistic regression model.

[0067] Independent predictors were selected using the stepwise regression method (Foward LR). The final model was constructed as follows: Logit(P_pCR)=β0 + β1 * X1 + β2 * X2 + …, where X1, X2, etc., are the selected independent predictors (CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T).

[0068] The final model is: Logit(P_pCR) = -5.3781 + 0.814×CD3+CD69+T / CD3+T + 0.092×CD3+HLA-DR+T / CD3+T; converting this to probabilistic form, we get P(pCR) = 1 / (1+e^(-(0.814×CD3+CD69+T / CD3+T +0.092×CD3+HLA-DR+T / CD3+T -5.378))).

[0069] ④ Model performance evaluation and validation: 1) Discrimination evaluation: The discriminative power of the model was evaluated by calculating the area under the receiver operating characteristic (ROC) curve. An optimal probability cutoff value (0.21) was determined based on the ROC curve to convert continuous probability into a binary judgment of "high probability pCR" or "low probability pCR." Specifically, when P(pCR) ≥ 0.21, the patient was predicted to achieve pathological complete remission; when P(pCR) < 0.21, the patient was predicted to achieve non-pathological complete remission.

[0070] Based on the baseline levels of CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T, a logistic regression model was constructed: P(pCR) = 1 / (1+e^(-(0.814×CD3+CD69+T / CD3+T +0.092×CD3+HLA-DR+T / CD3+T -5.378))), with AUC = 0.87, sensitivity 81.8%, and specificity 76.2%.

[0071] 2) Internal validation: Internal validation was performed using the Bootstrap resampling method (re-sampling 1000 times). The AUC of the corrected model was 0.85, which is close to the AUC of the original model, indicating that the model has good stability and low fitting risk.

[0072] 3) Clinical efficacy evaluation: The clinical net benefit of the model was evaluated using the decision curve analysis (DCA) method. The maximum net benefit of the model was 0.29, which shows that it has clear clinical application value.

[0073] Among them, the predictive model for pathological complete remission (pCR) after neoadjuvant therapy is as follows: Figure 1 As shown, Figure 1 The forest plot in the middle A region confirms that CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T are independent predictors of pCR. Figure 1 The ROC curve in B shows that the model has good predictive performance.

[0074] Figure 3For internal validation and DCA analysis of the model, Figure 3 Figure A represents the internal validation of the prediction model, which demonstrates good stability and low fitting risk. Figure 3 B in the model is a DCA analysis, and the model has clear clinical application value.

[0075] 2. Prognostic risk assessment model (for recurrence / metastasis): ① Stratified analysis: Modeling was performed separately for T2 (postoperative period) and T3 (postoperative monitoring period) data.

[0076] ② Modeling process: The process is similar to the pCR model. First, univariate analysis is performed to screen variables related to recurrence / metastasis, and then multivariate logistic regression is incorporated to construct a risk prediction model.

[0077] The final model at time point T2 includes Treg / CD4+T and CD3+HLA-DR+ / CD3+T; the final model at time point T3 includes two variables: CD3+CD8+HLA-DR+T% and CA125.

[0078] Post-NAT (Pre-Surgery) Model: A model for P (recurrence / metastasis) was constructed based on Treg / CD4+T and CD3+HLA-DR+T / CD3+T levels. =1 / (1+e^(-(0.540×Treg / CD4+T-0.130×CD3+HLA-DR+T / CD3+T-1.843))) (AUC=0.82), sensitivity 87.5%, specificity 85.0%.

[0079] Postoperative (within 1 year) model: The model was constructed based on CD3+CD8+HLA-DR+T / CD3+T and CA125 levels. P(recurrence / metastasis) = 1 / (1+e^(-(-0.086×CD3+CD8+HLA-DR+T / CD3+T+0.254×CA125-2.750))) (AUC=0.87), with a sensitivity of 91.7% and a specificity of 84.7%.

[0080] Predictive models for recurrence and metastasis after NAT surgery – preoperative (A and B) and postoperative (C and D) are as follows: Figure 2 As shown, Figure 2 China A and Figure 2 In the middle section, B represents the forest plot of independent predictors and the ROC curve of the model at the time points from post-NAT to pre-NAT, respectively. Figure 2 C and Figure 2 In the middle, D represents the forest plot of independent predictors and the ROC curve of the model at the postoperative (within 1 year) time point. Figure 3 In the middle, C represents the internal validation of the model after NAT (before surgery), with a corrected AUC of 0.78. Figure 3 D represents the DCA analysis of this model, with a maximum net benefit of 0.29; Figure 3 The mean squared error (E) represents the internal validation of the model after surgery (within 1 year), with a corrected AUC of 0.86. Figure 3 F represents the DCA analysis of this model, with a maximum net benefit of 0.21.

[0081] It should be noted that although the postoperative data collection period in this study spanned within one year post-surgery, the sample time distribution was relatively concentrated, with a median collection time of 4 months post-surgery (interquartile range: 3-6 months), ensuring the timeliness and consistency of the model evaluation. See the detailed postoperative detection time distribution chart below. Figure 4 .

[0082] 3. Survival analysis and identification of independent prognostic factors: ① Group comparison: Based on the median values ​​of each immune marker (Treg / CD4+T, CD3+CD25+T / CD3+T, CD3+CD38+T / CD3+T, CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD4+CD25+T / CD3+T, CD3+CD4+CD38+T / CD3+T, CD3+CD4+CD69+T / CD3+T, CD3+CD4+HLA-DR+T / CD3+T, CD3+CD8+CD25+T / CD3+T, CD3+CD8+CD38+T / CD3+T, CD3+CD8+CD69+T / CD3+T, CD3+CD8+HLA-DR+T / CD3+T) within 1 year after surgery, patients were divided into a "high-level group" and a "low-level group".

[0083] ② Survival curve plotting: The Kaplan-Meier method was used to plot the disease-free survival curves for each group, and the Log-rank test was used to compare the differences between groups. 1) Patients in the high Treg / CD4+T group had significantly worse DFS than those in the low expression group (P<0.05). 2) Patients with high levels of CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD8+CD69+T / CD3+T, and CD3+CD8+HLA-DR+T / CD3+T all had significantly better disease-free survival (DFS) than those with low levels (P<0.05).

[0084] Representative survival curves are shown below. Figure 5 As shown, Figure 5 China A to Figure 5The E in the middle are Treg / CD4+T, CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD8+CD69+T / CD3+T, CD3+CD8+HLA-DR+T / CD3+T.

[0085] ③ Multivariate analysis: Variables significant in the univariate analysis, along with clinical confounding factors (including age, T stage, TNM stage, molecular subtype, number of lesions, pathological grade, and NAT protocol type), were included in a Cox proportional hazards regression model to identify independent prognostic factors affecting disease-free survival (DFS) and to calculate hazard ratios and their confidence intervals. Treg / CD4+T was an independent risk factor for DFS (HR=1.34, 95%CI: 1.13-1.59, P=0.011), and CD3+CD8+HLA-DR+T / CD3+T was an independent protective factor against DFS (HR=0.92, 95%CI: 0.87-0.98, P=0.027).

[0086] 4. Dynamic monitoring and trend early warning: Long-term longitudinal monitoring clearly revealed distinct dynamic evolution patterns in key peripheral blood immune subsets between patients in the relapse / metastasis group and those without relapse / metastasis, and these patterns were compared using a linear mixed-effects model. These patterned differences provide direct scientific evidence and actionable early warning rules for translating immune surveillance into clinical applications.

[0087] result: 1. Regulatory T cells: The Treg / CD4+T% ratio in the relapse-free group remained stable at a low level throughout the follow-up period. In stark contrast, the relapse / metastasis group showed a persistently elevated Treg / CD4+T% ratio from baseline, and this high level was maintained for several years after surgery, indicating the continued presence of their immunosuppressive microenvironment.

[0088] 2. T-cell activation status: In patients without recurrence, activation markers (such as CD3+CD69+T) often show a transient peak after treatment, followed by a gradual decline, exhibiting a self-limiting physiological activation pattern. Conversely, in patients with recurrence / metastasis, these activation markers show a chronic and persistent increase, with no downward trend during long-term follow-up after surgery, and may even progressively increase, suggesting the presence of abnormal and unresolved persistent activation of the immune system, which is often related to the continuous stimulation of subclinical lesions.

[0089] 3. Effector / memory phenotype: For CD3+HLA-DR+T and CD3+CD8+HLA-DR+T cells, the relapse-free group reached a peak within one year after surgery and then tended to stabilize or slowly decline, while the relapse group showed a continuous linear upward trend.

[0090] A longitudinal analysis of immune cell subsets in patients with relapse / metastasis and those without relapse / metastasis was performed during a 3-year follow-up period. The results are as follows: Figure 6 As shown. Figure 6 The dynamic changes of Treg / CD4+T in the middle group were significantly different between the relapse-free group and the relapse group (P<0.05). The relapse-free / metastasis group maintained a stable low level, while the relapse / metastasis group showed a continuous increase. Figure 6 B, Figure 6 C and Figure 6 China F to Figure 6 The values ​​of H represent the dynamic changes of CD3+CD25+T, CD3+CD38+T, CD3+CD8+CD25+T, CD3+CD8+CD38+T, and CD3+CD8+CD69+T, respectively. There were no significant differences between the relapse / metastasis group and the non-relapse / metastasis group (P>0.05). Figure 6 D, Figure 6 China E and Figure 6 The values ​​of I and II represent the dynamic changes of CD3+CD69+T, CD3+HLA-DR+T, and CD3+CD8+HLA-DR+T, respectively. Patients in the non-recurrence / metastasis group showed a transient peak after treatment, while patients in the recurrence / metastasis group showed a sustained increase after treatment (P<0.05).

[0091] During routine follow-up for several years post-surgery, blood samples were continuously collected at the T3 time point, and the aforementioned core immune subsets were tested. A flowchart based on T-cell activation analysis to predict the efficacy of NAT treatment and 3-year prognosis in breast cancer is shown below. Figure 7 As shown. Figure 7 Figure A shows a schematic diagram of sample processing. This method requires only 50 μl of whole blood. After incubation, erythrocyte sedimentation, and a single wash, it can be used for 9-color flow cytometry analysis. With minimal sample volume and convenient operation, a single test can comprehensively analyze T cell subsets and their activation status. Figure 7 China B to Figure 7 The C-axis is a gating strategy for flow cytometry detection and result analysis. It uses CD45 gating combined with SSC features to accurately delineate lymphocyte populations, and then identifies T cells through CD3 and SSC scatter plots. It further sorts activated T cell subsets by combining activation markers (CD69, CD38, etc.). Figure 7 The formula in D is a model formula for predicting the efficacy of NAT treatment and the 3-year prognosis. The obtained proportion of corresponding activated T cell subsets and CA125 levels are incorporated into the NAT efficacy prediction model and the 3-year prognosis prediction model to calculate the corresponding probability values. When the probability value P(pCR) of the NAT efficacy model is >0.21, it is predicted that the patient may achieve pathological complete remission. When the probability value P(recurrence / metastasis) of the 3-year prognosis model is >0.23, it is predicted that the patient has a high risk of postoperative recurrence and metastasis.

[0092] By plotting line graphs showing the changes of key immune markers (especially Treg / CD4+T% and CD3+CD38+T / CD3+T%) over time in individual patients, an early warning rule was established: when these markers show a persistent and progressively increasing trend, especially exceeding the patient's baseline level or the normal fluctuation range of the population, the system will issue an early warning signal of increased early recurrence risk, suggesting that clinicians should strengthen imaging examinations or consider intervention, even if traditional tumor markers may still be within the normal range at this time.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit the scope of protection of this application. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this application without departing from the substance and scope of the technical solutions of this application.

Claims

1. The use of an immune cell subpopulation for the preparation of a test product for the evaluation of the efficacy of neoadjuvant therapy and the risk of relapse and metastasis in breast cancer, characterized in that, The immune cell subsets include at least one of CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD8+HLA-DR+T / CD3+T, and Treg / CD4+T.

2. Use according to claim 1, wherein The immune cell subsets include any one of (1) to (3): (1) Combinations of CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T; (2) Combinations of Treg / CD4+T and CD3+HLA-DR+T / CD3+T; (3) CD3+CD8+HLA-DR+T / CD3+T.

3. The use according to claim 1, wherein The timeframes for neoadjuvant therapy include the baseline period before starting neoadjuvant therapy, the period before surgery after completing neoadjuvant therapy, and the period within one year after surgery after completing neoadjuvant therapy.

4. The use according to claim 1, wherein The immune cell subsets are derived from peripheral blood.

5. Use of a reagent for detecting the level of a subpopulation of immune cells in the manufacture of a product for assessing the efficacy of neoadjuvant therapy and the risk of relapse and metastasis in breast cancer, characterized in that, The immune cell subsets include any one of (1) to (3): (1) Combinations of CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T; (2) Combinations of Treg / CD4+T and CD3+HLA-DR+T / CD3+T; (3) CD3+CD8+HLA-DR+T / CD3+T.

6. A predictive model for the efficacy of neoadjuvant therapy for breast cancer, characterized in that, The predictive model is based on baseline CD3+CD69+T / CD3+T and CD3+HLA-DR+T / CD3+T levels prior to neoadjuvant therapy, and a logistic regression model is constructed as follows: P=1 / (1+e^(-(0.814×CD3+CD69+T / CD3+T+0.092×CD3+HLA-DR+T / CD3+T -5.378))).

7. A predictive model for the risk of recurrence and metastasis in neoadjuvant therapy for breast cancer, characterized in that, The predictive models include a predictive model before surgery after neoadjuvant therapy and a predictive model within one year after surgery. Preoperative prediction models after neoadjuvant therapy are based on Treg / CD4+T and CD3+HLA-DR+T / CD3+T levels, constructing a logistic regression model: P=1 / (1+e^(-(0.540×Treg / CD4+T-0.130×CD3+HLA-DR+T / CD3+T-1.843))); or, The predictive model within one year post-surgery is based on CD3+CD8+HLA-DR+T / CD3+T and CA125 levels, and a logistic regression model is constructed: P=1 / (1+e^(-(-0.086×CD3+CD8+HLA-DR+T / CD3+T+0.254×CA125 -2.750))).

8. A system for assessing the efficacy and risk of recurrence and metastasis of neoadjuvant therapy for breast cancer, characterized in that, The system includes a sample acquisition module, a detection and analysis module, a data input and calculation module, and a result output module; The sample collection module is used to collect peripheral blood samples from breast cancer patients during the baseline period before starting neoadjuvant therapy, before surgery after completing neoadjuvant therapy, and within 1 year after surgery after completing neoadjuvant therapy. The detection and analysis module is used to detect the levels of immune cell subsets and serum tumor markers. The data input and calculation module is used to input the proportion of immune cell subsets and the level of serum tumor markers into the predictive model of the efficacy of neoadjuvant therapy for breast cancer or the predictive model of the risk of recurrence and metastasis after neoadjuvant therapy plus radical surgery for breast cancer, and to calculate the probability of pathological complete remission P or the probability of recurrence / metastasis P. The result output module is used to compare the calculated pathological complete remission probability P or recurrence / metastasis risk probability P with the corresponding model threshold: if the obtained probability P value is higher than the corresponding threshold, it indicates that the patient has the possibility of achieving pathological complete remission or experiencing recurrence / metastasis. Combined with clinical information, the efficacy of neoadjuvant therapy and the long-term risk of recurrence and metastasis can be further evaluated. The immune cell subsets include at least one of CD3+CD69+T / CD3+T, CD3+HLA-DR+T / CD3+T, CD3+CD8+HLA-DR+T / CD3+T, and Treg / CD4+T.

9. The evaluation system as described in claim 8, characterized in that, If the probability of pathological complete remission (P) is higher than 0.21, the patient is predicted to have a high probability of achieving pathological complete remission; if the probability of relapse / metastasis (P) is higher than 0.23, the patient is predicted to have a high probability of relapse / metastasis.

Images