A combination of indicators for identifying diffuse large b-cell lymphoma and use thereof

Abstract

The application discloses an index combination for identifying diffuse large B-cell lymphoma and application thereof. The index combination is composed of gender, age and eight oligosaccharide chains, namely NGA2F, NG2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb and NA4Fb. The application also provides a prediction model, a prediction system, a computer readable storage medium and a kit constructed based on the index combination. Experimental results show that the prediction model has a sensitivity of 88.06%, a specificity of 92.98% and an AUC of 0.949 in identifying diffuse large B-cell lymphoma in a verification set. The application realizes, for the first time, non-invasive, rapid and high-accuracy identification of diffuse large B-cell lymphoma based on blood oligosaccharide chains combined with clinical indexes, and provides a brand-new tool for auxiliary differential diagnosis of patients with lymph node enlargement.

Description

Technical Field

[0001] This invention belongs to the field of molecular biomedicine and in vitro diagnostic technology, specifically relating to a combination of indicators for identifying diffuse large B-cell lymphoma and its application. Background Technology

[0002] Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin's lymphoma in adults, accounting for approximately 30%–40% of all cases. This disease exhibits high clinical and molecular heterogeneity, with significant differences in prognosis and treatment response among different molecular subtypes. Accurate diagnosis is crucial for selecting individualized treatment strategies for DLBCL.

[0003] Currently, the "gold standard" for diagnosing DLBCL remains tissue biopsy of lymph nodes or extranodal affected sites. Clinically, when patients present with unexplained lymph node enlargement, invasive surgery (such as complete or partial lymph node resection, or needle biopsy) is usually required to obtain pathological tissue. Subsequently, pathologists use morphological observation (such as the presence of diffuse large B-cell proliferation), immunohistochemical staining (such as detection of markers like CD20, CD3, BCL6, and MUM1), and necessary molecular biological tests (such as FISH detection of MYC, BCL2, and BCL6 rearrangements) to ultimately confirm the diagnosis of DLBCL.

[0004] However, this diagnostic model relying on tissue biopsy has the following limitations: (1) Invasiveness and operational risks: For lesions in deep lymph nodes (such as retroperitoneal, mediastinal or pelvic lymph nodes) or adjacent to important organs, puncture biopsy is technically difficult and may cause complications such as pneumothorax, bleeding, infection, damage to surrounding tissues or even tumor needle tract implantation. Some patients are not suitable for biopsy due to their physical condition or the location of the lesion. (2) Sampling errors and heterogeneity: DLBCL has significant spatial heterogeneity. The characteristics of tumor cells in different lesions or different parts of the same lesion may differ. Biopsy can only obtain a very small part of the tissue. If the sampling site cannot represent the whole picture of the tumor, it may lead to false negative diagnosis or fail to accurately reflect its invasiveness and molecular subtype. (3) Long diagnostic cycle: From sample collection, tissue fixation, embedding, sectioning, staining to final pathological interpretation and molecular detection, it usually takes several days to several weeks, which to some extent delays the patient's early and accurate treatment. (4) False negative rate and clinical dilemma: A significant number of patients presenting with superficial or deep lymph node enlargement have lymph node biopsy pathology results of "reactive hyperplasia" or "necrotizing lymphadenitis," but their clinical presentation still strongly suggests lymphoma. Due to the limitations of biopsy, these patients often require prolonged clinical observation, imaging re-examination, and even repeated biopsies to finally diagnose or rule out DLBCL, placing enormous psychological and economic burden on them. In view of the above limitations, there is an urgent clinical need to develop a non-invasive, repeatable, rapid, and highly accurate auxiliary diagnostic method for risk assessment and early auxiliary diagnosis of DLBCL.

[0005] In recent years, the role of protein glycosylation modification in tumorigenesis and development has received increasing attention. Abnormal glycosylation has been proven to be closely related to the occurrence, progression, and prognosis of various malignant tumors. Regarding DLBCL, multiple studies have shown that specific glycan structures can serve as potential biomarkers for disease diagnosis and prognostic assessment. However, current reports on glycan chains in DLBCL research mainly focus on the tissue or cell line level, and research on the auxiliary diagnosis of DLBCL using oligosaccharide chains on glycoproteins in blood samples (serum / plasma) is still lacking. Compared with tissue biopsy, blood samples have advantages such as convenient collection, minimal trauma, real-time dynamic monitoring, and repeatable collection. Therefore, developing a DLBCL auxiliary diagnostic method based on blood oligosaccharide chains has significant clinical application value for achieving non-invasive auxiliary screening of patients with lymphadenopathy. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a combination of indicators for identifying diffuse large B-cell lymphoma and its application. It reveals for the first time an indicator combination consisting of gender, age, and eight specific oligosaccharide chains. The predictive model constructed based on this indicator combination can identify diffuse large B-cell lymphoma non-invasively, with high sensitivity and high specificity through blood samples, significantly improving the accuracy of differential diagnosis.

[0007] This invention is achieved through the following technical solution:

[0008] The application of an indicator combination in the preparation of a product for identifying diffuse large B-cell lymphoma, the indicator combination comprising sex, age and eight oligosaccharide chains: NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb.

[0009] Preferably, the product is a reagent kit, a prediction system, or a computer-readable storage medium.

[0010] A predictive model for identifying diffuse large B-cell lymphoma, wherein the predictive model calculates the DLBCL-GPS score using the following formula:

[0011] DLBCL-GPS score = exp(-0.018 × NGA2F level - 0.696 × NGA2FB level - 0.655 × NG1A2F-1 level + 0.231 × NG1A2F-2 level - 0.072 × NA2F level - 0.113 × NA2FB level + 0.051 × NA3Fb level + 1.248 × NA4Fb level - 0.156 × gender + 0.085 × age - 1.7) 29) / [1 + exp(-0.018 × NGA2F level - 0.696 × NGA2FB level - 0.655 × NG1A2F-1 level + 0.231 × NG1A2F-2 level - 0.072 × NA2F level - 0.113 × NA2FB level + 0.051 × NA3Fb level + 1.248 × NA4Fb level - 0.156 × gender + 0.085 × age - 1.729)];

[0012] In this context, the value for male is 1, and the value for female is 0;

[0013] The prediction model has a preset discrimination threshold of 0.42. When the DLBCL-GPS score is ≥0.42, it is determined to be diffuse large B-cell lymphoma, and when the DLBCL-GPS score is <0.42, it is determined to be benign lymph node enlargement.

[0014] Preferably, the NGA2F level, NGA2FB level, NG1A2F-1 level, NG1A2F-2 level, NA2F level, NA2FB level, NA3Fb level, and NA4Fb level are relative abundance values ​​obtained by detecting oligosaccharide chains in biological samples.

[0015] Preferably, the biological sample is blood, serum, or plasma from the venous or peripheral blood of the subject.

[0016] A predictive system for identifying diffuse large B-cell lymphoma, comprising:

[0017] The data input module is used to acquire the indicator combination information in the sample to be tested. The indicator combination information consists of the gender and age information of the subject and the abundance information of 8 oligosaccharide chains. The 8 oligosaccharide chains are: NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb.

[0018] The prediction module stores the aforementioned prediction model.

[0019] The results output module is used to output the prediction results of diffuse large B-cell lymphoma calculated by the prediction model based on the oligosaccharide chain abundance information, gender, and age.

[0020] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the functions of one or more modules in the aforementioned prediction system.

[0021] A kit for identifying diffuse large B-cell lymphoma, comprising:

[0022] (a) A reagent for detecting the relative levels of eight oligosaccharide chains in biological samples;

[0023] The eight oligosaccharide chains are: NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb.

[0024] (b) Instructions for use, which state that the relative level data of the eight oligosaccharide chains obtained from the test, together with the gender and age information of the test subject, are substituted into the above-mentioned prediction model for calculation, and the risk of diffuse large B-cell lymphoma is determined based on the calculation results.

[0025] Preferably, the biological sample is blood, serum, or plasma from the venous or peripheral blood of the subject.

[0026] The beneficial effects of this invention are as follows:

[0027] (1) This invention is the first to discover and verify that a combination of indicators consisting of eight specific oligosaccharide chains (NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, NA4Fb) along with gender and age shows significant differences between patients with benign lymphadenopathy and patients with diffuse large B-cell lymphoma. The predictive model constructed based on this discovery can effectively capture the glycosylation differences between DLBCL patients and patients with benign lymphadenopathy, providing a novel combination of bioindicators for the non-invasive auxiliary diagnosis of DLBCL.

[0028] (2) This invention uses blood samples (serum / plasma) as the detection object, and only routine venous blood collection is required to complete the test. Compared with traditional tissue biopsy, it has significant advantages such as less trauma, lower risk, and higher patient compliance. For patients with enlarged lymph nodes found on imaging but with atypical clinical features, or deep lymph nodes that are difficult to puncture and biopsy, this invention can serve as an effective non-invasive auxiliary screening method to help clinicians identify high-risk groups, avoid unnecessary invasive biopsies for patients with benign lymph node enlargement, and reduce the waste of medical resources and patient suffering.

[0029] (3) This invention combines blood oligosaccharide profiling with the patient's gender and age information to construct a multi-dimensional diagnostic model. Experimental results show that the area under the curve (AUC) of the predictive model reaches 0.943 in the training set and 0.949 in the validation set; in the validation set, the sensitivity for distinguishing diffuse large B-cell lymphoma reaches 88.06%, the specificity for distinguishing benign lymph node enlargement reaches 92.98%, and the overall accuracy reaches 90.32%. Further calibration evaluation shows that the Brier scores for the training set and the validation set are 0.0927 and 0.0800, respectively, and the calibration curve is basically consistent with the diagonal, indicating that the model's predicted probability is highly consistent with the actual results and no overfitting phenomenon occurs. This model can accurately and stably distinguish DLBCL patients from the lymph node enlargement population, providing objective and reliable auxiliary basis for clinical decision-making.

[0030] (4) This invention is based on the mature capillary electrophoresis method for oligosaccharide chain detection. The sample pretreatment process is standardized, the detection cycle is short, and batch detection is possible with controllable cost. Compared with tissue biopsy that relies on the experience of pathologists, this invention has the characteristics of simple operation, objective results, good repeatability, and easy standardization. It is suitable for promotion and use in medical institutions at all levels, especially primary hospitals, and helps to improve the early diagnosis rate of DLBCL.

[0031] (5) Current research on glycans in DLBCL mainly focuses on the tissue or cell line level, while research on the auxiliary diagnosis of DLBCL using oligosaccharide chains in blood samples is still lacking. This invention is the first to achieve highly sensitive non-invasive detection of DLBCL based on blood oligosaccharide chains, effectively filling the gap in non-invasive diagnostic technology for this disease, and has important clinical application value and broad market prospects. Attached Figure Description

[0032] Figure 1 Blood oligosaccharide chain maps of patients with benign lymphadenopathy (A) and diffuse large B-cell lymphoma (B) in Example 1;

[0033] Figure 2 ROC curves of the prediction model DLBCL-GPS in the training and validation sets of Example 1, distinguishing between benign lymphadenopathy and diffuse large B-cell lymphoma.

[0034] Figure 3 This is a calibration curve of the DLBCL-GPS prediction model in the training and validation sets of Example 1, used to distinguish between benign lymph node enlargement and diffuse large B-cell lymphoma. Detailed Implementation

[0035] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0036] Unless otherwise specified, the technical means used in the following embodiments are all conventional means well known to those skilled in the art, and the experimental methods without specific conditions are all conventional methods in the art.

[0037] Unless otherwise specified, all materials and reagents used in the following examples are commercially available.

[0038] Example 1: Construction and Validation of a Predictive Model for Diffuse Large B-Cell Lymphoma

[0039] 1. Test Sample

[0040] This embodiment collected blood samples from 416 patients as a dataset, including 191 cases of benign lymphadenopathy and 225 cases of diffuse large B-cell lymphoma. All samples were obtained from Nanjing University Medical School Affiliated Gulou Hospital, ethics number 2023-248-01.

[0041] The dataset was randomly divided into a training set and a validation set, with 292 cases in the training set and 124 cases in the validation set. In the training set, there were 134 cases of benign lymphadenopathy and 158 cases of diffuse large B-cell lymphoma; in the validation set, there were 57 cases of benign lymphadenopathy and 67 cases of diffuse large B-cell lymphoma.

[0042] 2. Detection of oligosaccharide chains in training set blood samples

[0043] (1) Instruments and equipment

[0044] Capillary gel electrophoresis analyzer (ABI 3500 sequencer), PCR instrument, centrifuge.

[0045] (2) Test reagents

[0046] Reagent A: Add 1% SDS solution to 5 mM NH4HCO3;

[0047] Reagent B: Add 2 U / μL of exoglycoside exonuclease solution to 1% NP-40;

[0048] Reagent C: Add 2 U / μL sialidase solution to 100 mM NH4AC at pH=5;

[0049] Reagent D: ddH2O;

[0050] Reagent E: A solution prepared by mixing 5 mM fluorescent labeling solution (trisodium 8-aminopyrene-1,3,6-trisulfonic acid) with DMSO solution (organic reducing agent NaBH3CN concentration of 1 M).

[0051] (3) Oligosaccharide chain map detection and collection

[0052] ① Release of oligosaccharide chains

[0053] Add 3 μL of reagent A to 5 μL of sample, heat at 95℃ for 5 min to denature, cool to room temperature, add 3 μL of reagent B and 4 μL of reagent C, react at 37℃ for 4 h, and add 80 μL of reagent D to terminate the reaction.

[0054] ② Marking of oligosaccharide chains

[0055] Take 10 μL of the sample solution from step ①, dry it at 70℃ for 30 min, then add 3 μL of reagent E, react at 90℃ for 2 h, and finally add 80 μL of reagent D to terminate the reaction.

[0056] ③ Detection and spectral acquisition of oligosaccharide chains

[0057] Take 10 μL of the oligosaccharide chain sample prepared in step ②, place it in an ABI 96-well plate, and detect it using an ABI 3500 sequencer to obtain the oligosaccharide chain map.

[0058] like Figure 1As shown, after protein denaturation, glycosidase treatment to release oligosaccharide chains, fluorescent labeling, and spectral detection, the blood samples yielded the detection results for 11 oligosaccharide chains in each sample. These 11 oligosaccharide chains are: NGA2F (galactosyl α-1,6-core fucosylated biantennary oligosaccharide chain), NGA2FB (galactosyl α-1,6-core fucosylated biantennary oligosaccharide chain), NG1A2F-1 (monopeptide galactosyl α-1,6-core fucosylated biantennary oligosaccharide chain), NG1A2F-2 (monopeptide galactosyl α-1,6-core fucosylated biantennary oligosaccharide chain), NA2 (galactosylated biantennary oligosaccharide chain), and NGA2F-1 (monopeptide galactosyl α-1,6-core fucosylated biantennary oligosaccharide chain). The oligosaccharide chains included NA2F (galactosylated α-1,6-core fucosylated biantennary oligosaccharide chain), NA2FB (galactosylated α-1,6-core fucosylated biantennary oligosaccharide chain), NA3 (galactosylated triantennary oligosaccharide chain), NA3Fb (galactosylated α-1,3-branched fucosylated triantennary oligosaccharide chain), NA4 (galactosylated tetraantennary oligosaccharide chain), and NA4Fb (galactosylated α-1,3-branched fucosylated tetraantennary oligosaccharide chain). NG1A2F-1 and NG1A2F-2 are isomers. The relative content of each oligosaccharide chain was calculated using a normalization method.

[0059] 3. Data Analysis and Model Building

[0060] (1) Screening of characteristic oligosaccharide chains

[0061] Depend on Figure 1 It can be seen that the oligosaccharide profiles of patients with benign lymphadenopathy are significantly different from those of patients with diffuse large B-cell lymphoma, indicating that oligosaccharide chains have the potential to identify diffuse large B-cell lymphoma.

[0062] In the training set of 292 cases, oligosaccharide chain data, gender, and age information of patients with benign lymphadenopathy and diffuse large B-cell lymphoma were analyzed. Logistic regression was used to screen candidate biomarkers for the predictive model. Based on statistical significance criteria, indicators with a p-value less than 0.05 were selected as the final feature variables included in the predictive model. The screening results are shown in Table 1 below.

[0063] Table 1. Comparative analysis of various indicators in patients with benign lymphadenopathy and diffuse large B-cell lymphoma during the training session.

[0064]

[0065] Based on the results in Table 1, 10 statistically significant indicators (P < 0.05) were finally selected and included in the prediction model, including: gender, age, NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb.

[0066] (2) Constructing a prediction model

[0067] Based on the selected 10 indicators, a classification model named DLBCL-GPS was constructed using logistic regression in the training set. During model construction, clinical diagnostic labels were used as the dependent variable, and gender, age, and data on eight oligosaccharide chains were used as independent variables, with gender represented by a value of 1 for males and 0 for females. The resulting regression equation is as follows:

[0068] DLBCL-GPS score = exp(-0.018 × NGA2F level - 0.696 × NGA2FB level - 0.655 × NG1A2F-1 level + 0.231 × NG1A2F-2 level - 0.072 × NA2F level - 0.113 × NA2FB level + 0.051 × NA3Fb level + 1.248 × NA4Fb level - 0.156 × gender + 0.085 × age - 1.7) 29) / [1+exp(-0.018×NGA2F level-0.696×NGA2FB level-0.655×NG1A2F-1 level+0.231×NG1A2F-2 level-0.072×NA2F level-0.113×NA2FB level+0.051×NA3Fb level+1.248×NA4Fb level-0.156×sex+0.085×age-1.729)].

[0069] (3) Model performance evaluation

[0070] The ROC curves of the DLBCL-GPS model in the training and validation sets for distinguishing between benign lymphadenopathy and diffuse large B-cell lymphoma are shown below. Figure 2 As shown, their receiver operating characteristic (AUC) areas under the curves were 0.943 and 0.949, respectively.

[0071] Using the score output by the DLBCL-GPS model as the predicted value, the optimal threshold for distinguishing between benign lymphadenopathy and diffuse large B-cell lymphoma was determined to be 0.42 using the maximum Yoden index. The judgment rule is: a predicted value ≥ 0.42 indicates diffuse large B-cell lymphoma, and a predicted value < 0.42 indicates benign lymphadenopathy.

[0072] The specificity, sensitivity, and accuracy of the DLBCL-GPS model in the training and validation sets are shown in Table 2 below.

[0073] Table 2. Specificity, sensitivity, and accuracy of the DLBCL-GPS model in the training and validation sets.

[0074]

[0075] To further evaluate the consistency between the predicted probabilities and actual results of the DLBCL-GPS model, this invention uses the Brier score and calibration curve to analyze the model's calibration accuracy. The Brier score measures the mean square error between the predicted probabilities and actual results, with a value ranging from 0 to 0.25; a smaller value indicates better calibration accuracy. Figure 3 As shown, the Brier score for the training set is 0.0927, and the Brier score for the validation set is 0.0800, both within the ideal range, indicating that the model has good calibration performance. Notably, the Brier score for the validation set is slightly better than that for the training set, suggesting that the model has not overfitted. Furthermore, the mean absolute errors for the training and validation sets are 0.0578 and 0.1012, respectively, and the calibration curves closely match the diagonal, further validating the model's excellent probabilistic calibration capabilities.

[0076] 4. Conclusion

[0077] The experimental results of this embodiment demonstrate that the DLBCL-GPS predictive model for diffuse large B-cell lymphoma, constructed based on gender, age, and eight specific oligosaccharide chains (NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb), exhibits excellent generalization ability and good stability. On the validation set, the model achieved a sensitivity of 88.06% and a specificity of 92.98% for identifying diffuse large B-cell lymphoma, with an overall accuracy of 90.32%. Further calibration evaluation showed that the Brier scores for the training and validation sets were 0.0927 and 0.0800, respectively, and the calibration curves closely matched the diagonal curves, indicating that the model's predicted probabilities were highly consistent with the actual results, and no overfitting was observed. This invention is the first to achieve highly sensitive non-invasive detection of diffuse large B-cell lymphoma based on blood oligosaccharide chains, effectively filling the gap in non-invasive diagnostic technology for this disease and providing a reliable auxiliary diagnostic method for clinical practice.

[0078] Example 2: Kit for identifying diffuse large B-cell lymphoma

[0079] This embodiment provides a kit for identifying diffuse large B-cell lymphoma, the kit comprising:

[0080] (1) A reagent for detecting the relative levels of an oligosaccharide biomarker combination in a blood sample, wherein the oligosaccharide biomarker combination consists of eight oligosaccharide chains: NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb. Specific reagents are as follows:

[0081] Reagent A: Add 1% SDS solution to 5 mM NH4HCO3;

[0082] Reagent B: Add 2 U / μL of exoglycoside exonuclease solution to 1% NP-40;

[0083] Reagent C: Add 2 U / μL sialidase solution to 100 mM NH4AC at pH=5;

[0084] Reagent D: ddH2O;

[0085] Reagent E: A solution prepared by mixing 5 mM fluorescent labeling solution (trisodium 8-aminopyrene-1,3,6-trisulfonic acid) with DMSO solution (organic reducing agent NaBH3CN concentration of 1 M).

[0086] (2) Instruction manual, which contains the following contents:

[0087] The relative abundance data of eight oligosaccharide chains were obtained from the blood sample (serum or plasma) of the test subject using the detection method described in Example 1. The oligosaccharide chain abundance data, along with the test subject's gender (1 for male, 0 for female) and age information, were substituted into the prediction model (DLBCL-GPS) described in Example 1 to calculate the DLBCL-GPS score. The score was compared with a threshold of 0.42: if the score was ≥0.42, the test subject was determined to be at high risk of diffuse large B-cell lymphoma, and further tissue biopsy was recommended for confirmation; if the score was <0.42, the test subject was determined to be likely to have benign lymphadenopathy.

[0088] The kit described in this embodiment is easy to operate and provides rapid detection, making it suitable for non-invasive auxiliary screening of patients with enlarged lymph nodes in clinical practice.

[0089] The embodiments described above are only some, not all, of the embodiments of the present invention. The detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments. The scope of protection of the present invention is determined by the scope claimed in the claims. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. The application of a combination of indicators in the preparation of a product for identifying diffuse large B-cell lymphoma, characterized in that, The indicator combination consists of gender, age, and eight oligosaccharide chains, namely: NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb.

2. The application according to claim 1, characterized in that, The product is a reagent kit, a prediction system, or a computer-readable storage medium.

3. A predictive model for identifying diffuse large B-cell lymphoma, characterized in that, The prediction model calculates the DLBCL-GPS score using the following formula: DLBCL-GPS score = exp(-0.018 × NGA2F level - 0.696 × NGA2FB level - 0.655 × NG1A2F-1 level + 0.231 × NG1A2F-2 level - 0.072 × NA2F level - 0.113 × NA2FB level + 0.051 × NA3Fb level + 1.248 × NA4Fb level - 0.156 × gender + 0.085 × age - 1.7) 29) / [1 + exp(-0.018 × NGA2F level - 0.696 × NGA2FB level - 0.655 × NG1A2F-1 level + 0.231 × NG1A2F-2 level - 0.072 × NA2F level - 0.113 × NA2FB level + 0.051 × NA3Fb level + 1.248 × NA4Fb level - 0.156 × gender + 0.085 × age - 1.729)]; In this context, the value for male is 1, and the value for female is 0; The prediction model has a preset discrimination threshold of 0.

42. When the DLBCL-GPS score is ≥0.42, it is determined to be diffuse large B-cell lymphoma, and when the DLBCL-GPS score is <0.42, it is determined to be benign lymph node enlargement.

4. A predictive model for identifying diffuse large B-cell lymphoma according to claim 3, characterized in that, The NGA2F level, NGA2FB level, NG1A2F-1 level, NG1A2F-2 level, NA2F level, NA2FB level, NA3Fb level, and NA4Fb level are relative abundance values ​​obtained by detecting oligosaccharide chains in biological samples.

5. A predictive model for identifying diffuse large B-cell lymphoma according to claim 4, characterized in that, The biological samples are blood, serum, and plasma from the venous or peripheral blood of the test subject.

6. A predictive system for identifying diffuse large B-cell lymphoma, characterized in that, include: The data input module is used to acquire the indicator combination information in the sample to be tested. The indicator combination information consists of the gender and age information of the subject and the abundance information of 8 oligosaccharide chains. The 8 oligosaccharide chains are: NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb. The prediction module stores the prediction model as described in claim 3; The results output module is used to output the prediction results of diffuse large B-cell lymphoma calculated by the prediction model based on the oligosaccharide chain abundance information, gender, and age.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the functionality of one or more modules in the prediction system as described in claim 6.

8. A kit for identifying diffuse large B-cell lymphoma, characterized in that, include: (a) A reagent for detecting the relative levels of eight oligosaccharide chains in biological samples; The eight oligosaccharide chains are: NGA2F, NGA2FB, NG1A2F-1, NG1A2F-2, NA2F, NA2FB, NA3Fb, and NA4Fb. (b) Instructions for use, which describe: substituting the relative level data of the 8 oligosaccharide chains obtained from the test along with the gender and age information of the test subject into the prediction model as described in claim 3 for calculation, and determining the risk of diffuse large B-cell lymphoma based on the calculation results.

9. A kit for identifying diffuse large B-cell lymphoma according to claim 8, characterized in that, The biological samples are blood, serum, and plasma from the venous or peripheral blood of the test subject.

Images