Prediction model for evaluating curative effect of NAC or NACIT in HER2 low-early breast cancer based on sTILs

Abstract

The embodiment of the invention provides a prediction model for evaluating the curative effect of NAC or NACIT in HER2 low-early-stage breast cancer based on sTILs, and belongs to the field of cancer prognosis. The first aspect of the invention provides a column diagram for evaluating the curative effect of HER2 low-expression early breast cancer patients on neoadjuvant chemotherapy, wherein the column diagram comprises variable lines which respectively represent lymph nodes, stromal tumor infiltrating lymphocytes, Ki67 and hormone receptors. The prediction performance of the sTILs-based comprehensive prediction model provided by the invention is remarkably improved compared with that of a conventional clinical pathological prediction model, and the sTILs-based comprehensive prediction model can be used for predicting the NAC / NACIT curative effect of the HER2 low-expression early-stage breast cancer patient, so that a reference is provided for a better individualized treatment scheme, and clinical diagnosis and treatment activities are guided.

Description

Technical Field

[0001] This application relates to the technical field of cancer prognosis, and particularly to a prediction model for evaluating the efficacy of NAC or NACIT in HER2-low early breast cancer based on sTILs. Background Art

[0002] The interaction between the tumor microenvironment (TME) and chemotherapy efficacy is a key point in tumor research. The TME consists of tumor cells, stromal cells, extracellular matrix, and vascular networks, and affects tumor growth, invasion, and treatment response. Stromal tumor-infiltrating lymphocytes (sTILs) are immune cells infiltrating tumors and are closely related to the immune microenvironment. Recent studies have shown that the sTIL density in tumors is significantly correlated with chemotherapy response and patient survival. In solid tumors such as breast cancer (BC), ovarian cancer (OC), and lung cancer (LC), a high sTIL density usually predicts better chemotherapy outcomes and longer progression-free survival (PFS) and overall survival (OS). In addition, the function of sTILs is also crucial. Active sTILs can enhance chemotherapy, while exhausted or immunosuppressive sTILs will reduce its effectiveness.

[0003] In neoadjuvant chemotherapy (NAC), the presence of sTILs is usually used as a biomarker to evaluate whether patients can benefit from the treatment. In the field of breast cancer, research on this biomarker mainly focuses on early triple-negative breast cancer (TNBC) and human epidermal growth factor receptor 2 (HER2)-positive early breast cancer (EBC) treated with neoadjuvant chemotherapy plus immunotherapy (NACIT), and has obvious predictive value. However, approximately 80% of breast cancer patients are HER2-negative and can be further divided into two subtypes: HER2-low expression (HER2-low) and HER2-zero expression (HER2-0). Compared with HER2-0 BC, HER2-low BC has a larger primary tumor volume, increased lymph node involvement, and higher luminal-related gene expression. Although HER2-low BC is also more common, it still faces challenges in achieving the pathological complete remission (pCR) rate of neoadjuvant chemotherapy, usually remaining at a relatively low level of 16.3 - 29.2%.

[0004] Existing pCR prediction models and biomarkers for HER2-negative EBC patients are not specifically for HER2-low EBC and usually involve complex variables that are difficult to obtain. So far, only one messenger RNA-based multi-omics prediction model has been reported, which performs well in predicting the pCR of NAC in HER2-low EBC. However, this model is complex and difficult to implement. Therefore, there is an urgent need to develop prediction methods based on easily obtainable measurements to identify HER2-low EBC patients who can most benefit from NAC. Summary of the Invention

[0005] The main purpose of the embodiments of the present application is to propose a prediction model for evaluating the efficacy of NAC or NACIT in HER2-low early breast cancer based on sTILs, aiming to improve the accuracy of efficacy prediction.

[0006] To achieve the above object, a first aspect of the present application proposes a nomogram for evaluating the efficacy of neoadjuvant chemotherapy in patients with HER2-low early breast cancer. The nomogram includes variable lines respectively representing lymph nodes, stromal tumor-infiltrating lymphocytes, Ki67, and hormone receptors.

[0007] Among them, the hormone receptor (HR) includes estrogen receptor (ER) and progesterone receptor (PR). The results of the hormone receptor include two types: positive and negative. As long as one of ER and PR is positive, it is hormone receptor positive (HR+), that is, HR+ includes three situations: ER+PR+, ER+PR-, and ER-PR+. ER+ or PR+ can be obtained by immunohistochemical staining, and the positive standard is defined as the percentage of ER / PR-positive cells in the set area observed under the microscope being greater than or equal to 1%.

[0008] Ki67 is the Ki67 index. In some embodiments, the Ki67 index is obtained by immunohistochemical staining and is defined as the percentage of Ki67-positive cells in the set area observed under the microscope. It includes two situations: high Ki67 and low Ki67. High Ki67 is the Ki67 index > 20%, and low Ki67 is the Ki67 index ≤ 20%.

[0009] Stromal tumor-infiltrating lymphocytes (sTILs) is the sTILs level, and the evaluation criteria can at least adopt the guidelines of the International sTIL Working Group. In some embodiments, two professional pathologists with senior titles and many years of breast cancer diagnosis experience evaluate the sTILs level of H&E-stained specimens without knowledge of clinical and experimental data. Lymphocytes in direct contact with tumor cells are identified as intratumoral TILs (iTILs), while lymphocytes in the stroma (within the boundary of invasive carcinoma or in the surrounding area) are identified as stromal TILs (sTILs). The sTILs level is the average percentage of lymphocyte infiltration in each tumor and adjacent stroma. In some embodiments, the sTILs level includes high sTILs level (≥40%) and low sTILs (<40%) level.

[0010] Lymph node (abbreviated as N) indicates the presence of regional lymph node metastasis and the degree of lymph node involvement. According to the TNM staging standard, lymph nodes can be divided into N0 to N3. In some embodiments, lymph nodes are divided into lymph node positive (N1 to N3) and lymph node negative (N0).

[0011] In some embodiments of the present application, the prediction probability lines of the nomogram include the probability line for predicting pathological complete remission (pCR).

[0012] In some embodiments of the present application, the prediction probability lines of the nomogram further include probability lines for predicting at least one of major pathological remission (MPR), response rate, disease free survival (DFS), event free survival (EFS), overall survival (OS), minimal residual disease (MRD), etc.

[0013] In some embodiments of the present application, the nomogram further includes single-item score lines and a total score line. The score range of the single-item score lines is 0 - 100, and the score range of the total score line is 0 - 350. In some embodiments, the variable line representing lymph nodes takes a value of 0 when positive and 49.6 when negative. In some embodiments, the variable line representing stromal tumor-infiltrating lymphocytes takes a value of 0 when less than 40% and 100 when greater than or equal to 40%. In some embodiments of the present application, the variable line representing Ki67 takes a value of 0 when less than or equal to 20% and 93.1 when greater than 20%. In some embodiments of the present application, the variable line representing hormone receptor takes a value of 0 when positive and 52.2 when negative. In some embodiments of the present application, the probability line for predicting pathological complete remission takes values of 0.01 and 0.8 when the score of the total score line is 60 and 300 respectively.

[0014] In some embodiments of the present application, the nomogram is a graphical result of a logistic regression equation. In some embodiments, the logistic regression equation is Logit(P) = -3.576 - 1.309×hormone receptor + 2.337×Ki67 + 2.511×stromal tumor-infiltrating lymphocytes - 1.238×lymph nodes.

[0015] In the second aspect of the present application, a model for evaluating the efficacy of neoadjuvant chemotherapy in patients with HER2-low-expressing early breast cancer is proposed, and the model includes the aforementioned nomogram.

[0016] Therefore, in some embodiments of the present application, it further includes the application of the above-mentioned nomogram or the above-mentioned model in the preparation of a product for evaluating the efficacy of neoadjuvant chemotherapy in patients with HER2-low-expressing early breast cancer. In some embodiments of the present application, it further includes the application of the above-mentioned nomogram or the above-mentioned model in evaluating the efficacy of neoadjuvant chemotherapy in patients with HER2-low-expressing early breast cancer.

[0017] In a third aspect of the present application, a method for evaluating the efficacy of neoadjuvant chemotherapy in patients with HER2-low-expressing early breast cancer is proposed. The method includes the following steps:

[0018] Obtain relevant information on lymph nodes, stromal tumor-infiltrating lymphocytes, Ki67, estrogen receptor, and progesterone receptor in patients with HER2-low-expressing early breast cancer;

[0019] Evaluate the efficacy of neoadjuvant chemotherapy in patients with HER2-low-expressing early breast cancer through the aforementioned model.

[0020] In some embodiments of the present application, the neoadjuvant chemotherapy includes any one of chemotherapy (NAC) and immunotherapy combined with chemotherapy (NACIT).

[0021] In some embodiments of the present application, the chemotherapeutic drugs used in NAC include at least one of taxanes and anthracyclines. In some embodiments, NAC includes a combination of taxane drugs and platinum drugs. In some embodiments, NAC includes sequential administration of anthracycline drugs and taxane drugs.

[0022] In some embodiments of the present application, the chemotherapeutic drugs used in NACIT include a combination of taxanes and anthracyclines, and the immunotherapeutic drug includes pembrolizumab.

[0023] In some embodiments of the present application, the evaluation result includes the probability of pathological complete remission (pCR). In some embodiments, pCR means no invasive carcinoma in the primary breast lesion and negative regional lymph nodes (i.e., ypT0 / Tis ypN0).

[0024] In some embodiments of the present application, early breast cancer is breast cancer with a tumor smaller than 2 cm, no palpable metastatic lymph nodes in the axilla, and no distant metastases. In some embodiments, it includes non-invasive carcinoma.

[0025] Therefore, in some embodiments of the present application, a treatment method for patients with HER2-low-expressing early breast cancer is further provided, including evaluating patients with HER2-low-expressing early breast cancer using the method for evaluating the efficacy of neoadjuvant chemotherapy in patients with HER2-low-expressing early breast cancer, and determining whether to use neoadjuvant chemotherapy for treatment according to the evaluation result.

[0026] The method provided by the embodiments of the present application for evaluating the therapeutic efficacy of patients with HER2-low-expressing early breast cancer can be applied to a terminal, a server, or software running on a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server can be configured as an independent physical server, or as a server cluster or a distributed system composed of multiple physical servers, or as a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application for implementing the method for evaluating the therapeutic efficacy of patients with HER2-low-expressing early breast cancer, etc., but is not limited to the above forms.

[0027] The method provided by the embodiments of the present application can be used in many general-purpose or special-purpose computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0028] In a fourth aspect of the present application, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the foregoing method.

[0029] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory can include high-speed random access memory, and can also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory optionally includes a memory remotely disposed relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0030] In a fifth aspect of the present application, there is provided an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the foregoing method is implemented.

[0031] In some embodiments, the electronic device may be any intelligent terminal including a tablet computer, an in-vehicle computer, etc.

[0032] In some embodiments, the electronic device includes:

[0033] A processor, which can be implemented in ways such as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present application;

[0034] A memory, which can be implemented in forms such as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory can store an operating system and other application programs. When implementing the technical solutions provided by the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory and are called by the processor to execute the method for evaluating the therapeutic efficacy of HER2-low-expression early breast cancer patients in the embodiments of the present application;

[0035] An input / output interface, which is used to implement information input and output;

[0036] A communication interface, which is used to implement communication interaction between this device and other devices, and can achieve communication through a wired method (such as USB, network cable, etc.) or through a wireless method (such as a mobile network, WIFI, Bluetooth, etc.);

[0037] A bus, which transmits information between various components of the device (such as the processor, the memory, the input / output interface, and the communication interface);

[0038] Among them, the processor, the memory, the input / output interface, and the communication interface achieve communication connections with each other inside the device through the bus.

[0039] In a sixth aspect of the present application, there is provided a device, which includes:

[0040] A data acquisition module, which is used to acquire relevant information of lymph nodes, stromal tumor-infiltrating lymphocytes, Ki67, estrogen receptor, and progesterone receptor of HER2-low-expression early breast cancer patients;

[0041] A data processing module for evaluating the efficacy of neoadjuvant chemotherapy for HER2-low-expression early breast cancer patients through a model, the model including a nomogram, and the nomogram including variable lines respectively characterizing lymph nodes, stromal tumor-infiltrating lymphocytes, Ki67, and hormone receptors.

[0042] In some embodiments of the present application, the prediction probability line of the nomogram includes a probability line for predicting pathologic complete remission (pCR).

[0043] In some embodiments of the present application, the prediction probability line of the nomogram further includes a probability line for predicting at least one of major pathologic remission (MPR), response rate, disease free survival (DFS), event free survival (EFS), overall survival (OS), minimal residual disease (MRD), etc.

[0044] In some embodiments of the present application, the nomogram further includes single-item score lines and a total score line. The score range of the single-item score lines is 0 to 100, and the score range of the total score line is 0 to 350. In some embodiments, the variable line characterizing lymph nodes takes a value of 0 when positive and 49.6 when negative; the variable line characterizing stromal tumor-infiltrating lymphocytes takes a value of 0 when less than 40% and 100 when greater than or equal to 40%; the variable line characterizing Ki67 takes a value of 0 when less than or equal to 20% and 93.1 when greater than 20%; the variable line characterizing hormone receptors takes a value of 0 when positive and 52.2 when negative. In some embodiments of the present application, the probability line for predicting pathologic complete remission takes values of 0.01 and 0.8 when the score of the total score line is 60 and 300 respectively.

[0045] In some embodiments of the present application, the nomogram is a graphical result of a logistic regression equation. In some embodiments, the logistic regression equation is Logit(P)=-3.576 - 1.309×hormone receptor + 2.337×Ki67 + 2.511×stromal tumor-infiltrating lymphocytes - 1.238×lymph nodes.

[0046] In some embodiments of the present application, the neoadjuvant chemotherapy includes any one of chemotherapy (NAC) and immunotherapy combined with chemotherapy (NACIT).

[0047] In some embodiments of the present application, the chemotherapeutic drugs used in NAC include at least one of taxanes and anthracyclines. In some embodiments, NAC includes a combination of taxane drugs and platinum drugs. In some embodiments, NAC includes sequential administration of anthracycline drugs and taxane drugs.

[0048] In some embodiments of the present application, the chemotherapeutic drugs used in NACIT include a combination of taxanes and anthracyclines, and the immunotherapeutic drug includes pembrolizumab.

[0049] In some embodiments of the present application, the evaluation results include the probability of pathologic complete remission (pCR). In some embodiments, pCR means no invasive carcinoma in the primary breast lesion and negative regional lymph nodes (i.e., ypT0 / Tis ypN0).

[0050] In some embodiments of the present application, early breast cancer is breast cancer with a tumor smaller than 2 cm, no palpable metastatic lymph nodes in the axilla, and no distant metastasis. In some embodiments, it includes non-invasive carcinoma.

[0051] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0052] In several embodiments provided by the present application, it should be understood that the disclosed device and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the above-mentioned division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be in electrical, mechanical or other forms.

[0053] The units described as separate components above may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment. The functional units in the embodiments can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0054] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes: various media that can store programs such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0055] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and their appropriate combinations.

[0056] The beneficial effects of the present invention are:

[0057] By detecting sTILs in HER2-low-expressing breast cancer, the present invention discovers the positive correlation between sTILs or immune-promoting cells and the pCR status. It is proved that the higher the sTILs, the higher the proportion of immune-promoting cells, and the more likely HER2-low EBC patients are to obtain pCR after receiving anthracycline and taxane NAC. And through experiments, a prediction model for the efficacy of NAC in HER2-low-expressing early breast cancer patients based on sTILs is constructed. Through this prediction model, the efficacy of breast cancer patients receiving anthracycline and taxane NAC can be effectively predicted. The area under the curve (AUC) in the receiver operating characteristic curve (ROC) analysis is 0.884, and the prediction performance of this model is relatively high, indicating that it has a relatively high prediction value. It is also found that this model also has good prediction efficacy for the efficacy of patients receiving neoadjuvant chemotherapy combined with immunotherapy. In the decision curve analysis (DCA), it has a higher clinical net benefit in predicting the probability of pCR within a very wide range of pCR rate thresholds compared with other models.

[0058] Different from evaluating the efficacy by detecting the changes of the primary tumor during neoadjuvant chemotherapy through imaging, the present invention can predict the efficacy of HER2-low-expressing non-metastatic breast cancer before neoadjuvant chemotherapy, provide a reference for personalized treatment, avoid over-treatment or under-treatment of drug therapy, is convenient to operate and easy to use clinically, and improves the clinical outcomes of HER2-low EBC patients. Description of the Drawings

[0059] Figure 1This is the overall research flow chart for Example 1 and Example 2.

[0060] Figure 2 This shows the relationship between sTILs and pCR status in HER2-low-expressing early breast cancer patients in Example 1.

[0061] Figure 3 This is a nomogram for predicting the pCR probability of HER2-low-expressing early breast cancer patients using different prediction models based on the training set in Example 1.

[0062] Figure 4 This is an ROC curve for evaluating the pCR status prediction of HER2-low-expressing early breast cancer patients receiving NAC treatment using different prediction models. Among them, A is the result of the training set and B is the result of the validation set.

[0063] Figure 5 This is a calibration curve for the discrimination accuracy of the comprehensive prediction model based on sTILs for predicting the pCR status. Among them, A is the result of the training set and B is the result of the validation set.

[0064] Figure 6 This shows the results of the DCA curve analysis of different prediction models. Among them, A is the result of the training set and B is the result of the validation set.

[0065] Figure 7 This is an ROC curve for evaluating the pCR status prediction of HER2-low-expressing / HR-negative early breast cancer patients receiving NACIT treatment using different prediction models. Detailed implementation manners

[0066] In order to make the objectives, technical solutions and advantages of this application clearer, the following further elaborates on this application in combination with the accompanying drawings and examples. It should be understood that the specific examples described herein are only used to explain this application and are not intended to limit this application. Terms such as "first", "second", etc. are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. "At least one" means one or more, and "multiple" means two or more. Terms such as "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0067] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0068] The embodiments described in the embodiments of the present application are for more clearly explaining the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. As those skilled in the art know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0069] The following further details the content of the present invention through specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention.

[0070] Example 1

[0071] In this embodiment, a prediction model for the NAC efficacy of HER2-low-expressing early breast cancer patients based on sTILs was constructed and verified based on this prediction model. The process is as follows:

[0072] The overall process refers to Figure 1 , and the baseline data of HER2-low-expressing non-metastatic breast cancer patients collected are from the prospective data collected in ongoing clinical trials (NCT01503905; ClinicalTrials.gov) and a multicenter prospective cohort (ChiCTR-DDD-17013651; ChiCTR.org.cn) in China. Among them, eligible HER2-low-expressing early breast cancer patients from Sun Yat-sen Memorial Hospital, Sun Yat-sen University from March 1, 2012 to August 31, 2018 were included in this study.

[0073] Inclusion criteria:

[0074] (1) Female, age > 18 years, patients with newly diagnosed, untreated, histologically confirmed invasive breast cancer;

[0075] (2) Pathological diagnosis of HER2 low expression (IHC 2+ and FISH negative, or IHC 1+) before neoadjuvant chemotherapy;

[0076] (3) sTILs can be evaluated in the core needle biopsy specimens before neoadjuvant chemotherapy and the pathological and follow-up information is complete.

[0077] Exclusion criteria:

[0078] (1) HER2 positive or HER2-0;

[0079] (2) Primary metastasis or other specific types of breast cancer;

[0080] (3) Lack of sufficient pathological information or follow-up information;

[0081] (4) HR positive but not receiving adjuvant endocrine therapy.

[0082] Thus, 137 HER2-positive patients, 41 HER2-0 patients, 15 patients with primary metastasis or other specific types of breast cancer, 30 patients lacking pathological or follow-up information, and 4 HR-positive patients who did not receive endocrine therapy were excluded from 520 patients who all met the indications for neoadjuvant chemotherapy and had not received any other anti-tumor treatment before, and the personalized neoadjuvant chemotherapy regimens for the patients were formulated by chief physicians at Sun Yat-sen Memorial Hospital of Sun Yat-sen University according to the National Comprehensive Cancer Network (NCCN) guidelines and appropriate surgeries were performed on them. Finally, 293 eligible patients with HER2-low-expression early breast cancer were included in this study. The neoadjuvant chemotherapy methods used for these included patients with HER2-low-expression early breast cancer were anthracycline drugs, or taxane drugs, or a combination of anthracycline drugs and taxane drugs. According to the ratio of 7:3, the above patients were randomly divided into a training set and a test set, with 205 patients in the validation set and 88 patients in the test set.

[0083] Preoperative core biopsy hematoxylin-eosin staining sections of the above patients were screened from the pathological specimen bank and handed over to qualified pathologists who were unaware of this study to independently complete the sTILs assessment. The assessment criteria adopted the guidelines of the International sTIL Working Group (www.tilsinbreastcancer.org). Lymphocytes in direct contact with tumor cells were identified as intratumoral TILs (iTILs), while lymphocytes in the stroma (within the boundary of invasive carcinoma or in the surrounding area) were identified as stromal TILs (sTILs). The sTILs level was defined as the average percentage of lymphocyte infiltration in each tumor and adjacent stroma.

[0084] Statistical analysis was performed using R language, and the steps were as follows:

[0085] (1) Box plots were drawn to determine the relationship between sTILs and the pCR status. The results were as Figure 2 , and the sTILs level in the pCR group was significantly higher than that in the non-pCR group, p < 0.001. According to the maximum Youden index, the sTILs level was divided into high sTILs (≥40%) and low sTILs (<40%) with a cut-off value of 40%.

[0086] (2) Pearson chi-square test or Fisher's exact test was used to compare the clinicopathological differences between the pCR group and the non-pCR group. A two-sided p < 0.05 was considered to indicate a statistically significant difference.

[0087] (3) For the data in the training set, first perform logistic univariate regression analysis to screen out the features significantly associated with patients achieving pCR, and then use stepwise regression for logistic multivariate analysis to obtain a comprehensive prediction model (nomogram) based on sTILs, a prediction model without sTILs for comparison (nomogram without sTILs), and a sTILs univariate prediction model.

[0088] Among them, the results of multivariate analysis showed that features such as positive lymph nodes (N1 - N3, OR 0.29, p = 0.034), high sTILs (OR 12.32, p < 0.001), high Ki67 (i.e., > 20%, OR 10.35, p = 0.035), and HR positive (OR 0.27, p = 0.036) could serve as independent predictors of the efficacy of NAC in patients with HER2-low-expressing early breast cancer. Based on these independent predictors, a comprehensive prediction model based on sTILs was constructed.

[0089] The formula for the final comprehensive prediction model is:

[0090] Logit(P) = -3.576 - 1.309×HR + 2.337×Ki67 + 2.511×sTILs - 1.238×cN, and its nomogram is as shown in Figure 3 Figure A. This nomogram includes a single-item score line (Points), a variable line representing sTILs, a variable line representing lymph node cN, a variable line representing Ki67, a variable line representing hormone receptor HR, a total score line (TotalPoints), and a probability line for predicting pathologic complete remission (pCR) from top to bottom. Among them, the score range of the single-item score line is 0 - 100, and the score range of the total score line is 0 - 350; the variable line representing cN takes a value of 0 when positive and 49.6 when negative; the variable line representing sTIL takes a value of 0 when less than 40% and 100 when greater than or equal to 40%; the variable line representing Ki67 takes a value of 0 when less than or equal to 20% and 93.1 when greater than 20%; the variable line representing HR takes a value of 0 when positive and 52.2 when negative; the sum of the values mapped from each variable line of the patient to the single-item score line is the total score of the patient, and the projection of the total score onto the probability line is the corresponding probability value of predicting pCR. The probability line for predicting pathologic complete remission takes 0.01 and 0.8 when the score of the total score line is 60 and 300, respectively.

[0091] The formula for the prediction model without sTILs is:

[0092] Logit(P) = -2.604 - 1.390×cN - 1.030×HR + 2.325×Ki67, and its nomogram is as shown in Figure 3 Figure B.

[0093] The formula of the single-factor prediction model with only sTILs is:

[0094] Logit(P) = -3.270 + 2.571×sTILs, and its nomogram is as shown in Figure 3 Figure C.

[0095] (4) The ROC curve was plotted, and the predictive performance of the prediction model was evaluated by the area under the curve (AUC). P < 0.05 was considered statistically significant. The results are as shown in Figure 4 Figure. Figure A shows the ROC curves of the comprehensive prediction model based on sTILs, the prediction model without sTILs, and the single-factor prediction model of sTILs in the training set. The AUC of the comprehensive prediction model based on sTILs (0.884; 95% CI: 0.823 - 0.945) was significantly higher than that of the prediction model without sTILs (0.811; 95% CI: 0.731 - 0.891) or the single-factor prediction model of sTILs (0.774; 95% CI: 0.671 - 0.878). Figure B shows the ROC curves of the comprehensive prediction model based on sTILs, the prediction model without sTILs, and the single-factor prediction model of sTILs in the validation set. The AUC of the comprehensive prediction model based on sTILs (0.891; 95% CI: 0.807 - 0.975) was also significantly higher than that of the prediction model without sTILs (0.785; 95% CI: 0.678 - 0.893) or the single-factor prediction model of sTILs (0.792; 95% CI: 0.669 - 0.916).

[0096] (5) The comprehensive prediction model based on sTILs was internally validated by the Bootstrap repeated resampling method. The calibration curve was obtained after 100 repeated Bootstrap resamplings. The results are as shown in Figure 5 Figure. Figure A shows the calibration curve of the prediction model in the training set, and Figure B shows the calibration curve of the prediction model in the validation set. It can be seen from the figure that there is no obvious deviation in the calibration curves of the training set and the validation set.

[0097] (6) Decision curve analysis (DCA) was performed on the comprehensive prediction model based on sTILs, the prediction model without sTILs, and the single-factor prediction model of sTILs. The results are as shown in Figure 6As shown, A is the decision curve of each prediction model for the training set, and B is the decision curve of each prediction model for the validation set. It can be seen from the figure that the comprehensive prediction model based on sTILs has a higher clinical net benefit in predicting the probability of pCR within a very wide range of pCR rate thresholds compared to other models.

[0098] Example 2

[0099] This example evaluates the predictive ability of the validated comprehensive prediction model based on sTILs constructed in Example 1 in HER2-low / HR-negative EBC patients receiving NACIT.

[0100] The overall process reference Figure 1 , eligible HER2-low / HR-negative early breast cancer patients from Sun Yat-sen Memorial Hospital, Sun Yat-sen University, from January 1, 2022 to December 31, 2023 were included in this study.

[0101] Inclusion criteria:

[0102] (1) Female, age > 18 years, patients with newly diagnosed, untreated, histologically confirmed invasive breast cancer;

[0103] (2) Pathological diagnosis of HER2-low expression (IHC 2+ and FISH negative, or IHC 1+) before neoadjuvant chemotherapy, and both ER and PR are negative;

[0104] (3) sTILs can be evaluated in the preoperative biopsy specimens before neoadjuvant chemotherapy and the pathological and follow-up information is complete.

[0105] Exclusion criteria:

[0106] (1) HER2-0;

[0107] (2) Primary metastasis or other specific types of breast cancer;

[0108] (3) Lack of sufficient pathological information or follow-up information.

[0109] Therefore, 45 cases all met the indications for neoadjuvant chemotherapy combined with immunotherapy and had not received any other anti-tumor treatment before. And for all these cases, the chief physicians of Sun Yat-sen Memorial Hospital, Sun Yat-sen University developed personalized neoadjuvant chemotherapy combined with immunotherapy regimens for the patients according to the National Comprehensive Cancer Network (NCCN) guidelines and performed appropriate surgeries. Among them, 10 HER2-0 patients, 3 patients with primary metastasis or other specific types of breast cancer, and 2 patients lacking pathological or follow-up information were excluded. Finally, 30 eligible patients with HER2-low expression / HR-negative early breast cancer were included in the study. The neoadjuvant chemotherapy combined with immunotherapy used for these included patients with HER2-low expression / HR-negative early breast cancer was anthracycline combined with taxane combined with pembrolizumab.

[0110] The comprehensive prediction model based on sTILs constructed in Example 1 was used to predict the pCR of NACIT patients, and a prediction model without sTILs was used as a control. The results were as Figure 7 shown. The comprehensive prediction model based on sTILs showed a significantly higher AUC (0.902; 95% CI: 0.801 - 1.000) when predicting the pCR of NACIT patients, which was significantly higher than the AUC of the prediction model without sTILs (0.647; 95% CI: 0.466 - 0.828).

[0111] In summary, it can be seen that the prediction performance of the comprehensive prediction model based on sTILs provided in this application is significantly improved compared with the conventional clinical pathological prediction model. It can be used to predict the efficacy of NAC / NACIT in patients with HER2-low expression early breast cancer, so as to provide a reference for a better individualized treatment plan and guide clinical diagnosis and treatment activities.

[0112] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited by the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A nomogram for evaluating the efficacy of neoadjuvant chemotherapy in patients with early breast cancer with low HER2 expression, characterized in that: The nomogram includes variable lines characterizing lymph nodes, stromal tumor infiltrating lymphocytes, Ki67, and hormone receptors, respectively.

2. The nomogram according to claim 1, characterized in that: The prediction probability line of the nomogram includes a probability line for predicting pathological complete remission.

3. The nomogram according to claim 1, characterized in that: The nomogram also includes a single score line and a total score line, the score range of the single score line is 0 to 100, and the score range of the total score line is 0 to 350; the variable line representing the lymph nodes takes a score of 0 when positive and a score of 49.6 when negative; the variable line representing the matrix tumor infiltrating lymphocytes takes a score of 0 when less than 40%, and a score of 100 when greater than or equal to 40%; the variable line representing Ki67 takes a score of 0 when less than or equal to 20%, and a score of 93.1 when greater than 20%; the variable line representing the hormone receptor takes a score of 0 when positive and a score of 52.2 when negative; the probability line for predicting complete pathological remission takes 0.01 and 0.8 when the total score line is 60 and 300, respectively.

4. The nomogram according to claim 1, characterized in that: The nomogram is a graphical result of the logistic regression equation, and the logistic regression equation is Logit (P) = -3.576-1.309×hormone receptor+2.337×Ki67+2.511×stromal tumor infiltrating lymphocytes-1.238×lymph nodes.

5. A model for evaluating the efficacy of neoadjuvant chemotherapy in patients with early breast cancer with low HER2 expression, characterized in that: A nomogram comprising any one of claims 1 to 4.

6. A method for evaluating the efficacy of neoadjuvant chemotherapy in patients with early breast cancer with low HER2 expression, characterized in that: The method comprises the following steps: To obtain information on lymph nodes, stromal tumor-infiltrating lymphocytes, Ki67, estrogen receptor, and progesterone receptor in patients with early breast cancer with low HER2 expression; The model described in claim 5 is used to evaluate the efficacy of neoadjuvant chemotherapy in patients with early breast cancer with low HER2 expression.

7. The method according to claim 6, characterized in that The neoadjuvant chemotherapy includes any one of chemotherapy and immunotherapy combined with chemotherapy.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 6 to 7 is implemented.

9. An electronic device, characterized in that The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 6 to 7 when executing the computer program.

10. The device, characterized in that The device comprises: Data acquisition module, used to obtain information about lymph nodes, stromal tumor-infiltrating lymphocytes, Ki67, estrogen receptors, and progesterone receptors in patients with early breast cancer with low HER2 expression; A data processing module is used to evaluate the efficacy of neoadjuvant chemotherapy in patients with early breast cancer with low HER2 expression through a model, wherein the model includes a nomogram, and the nomogram includes variable lines that respectively characterize lymph nodes, stromal tumor-infiltrating lymphocytes, Ki67 and hormone receptors.

Images