A biomarker for differential diagnosis of malignant pleural effusion and application thereof

Abstract

The present application belongs to the technical field of biological medicine, and particularly relates to a biomarker and application thereof. Three risk predictors, i.e. age, pleural effusion ADA and pleural effusion CCL2, screened by the present application are used to construct a nomogram model. The nomogram can quantitatively show the relative contribution of each variable and generate a total diagnostic score, and the higher the score, the greater the possibility of MPE, which is highly consistent with pathological diagnosis and shows excellent diagnostic performance. By combining the excellent NPV and sensitivity of CCL2 with the complementary value of age and ADA, the nomogram overcomes the shortcomings of single marker diagnosis and provides a minimally invasive and accurate tool for the diagnosis of MPE in clinical practice.

Description

Technical Field

[0001] This invention belongs to the field of biomedical technology, specifically relating to a biomarker and its application. Background Technology

[0002] Under normal circumstances, the production and absorption of fluid in the pleural cavity are in dynamic equilibrium. Its production is driven by the difference in hydrostatic and osmotic pressure between the systemic circulation, pulmonary circulation, and the pleural cavity. The lymphatic vessels located in the parietal pleura are mainly responsible for the reabsorption of fluid. Excessive production of pleural effusion (PE) represents a disruption of this balance. Increased hydrostatic pressure in pleural capillaries and decreased colloid osmotic pressure in pleural capillaries can lead to transudative effusion. Inflammation and malignant processes can promote local pleural vascular permeability, increased pleural permeability, and lymphatic obstruction, resulting in the accumulation of exudative pleural effusion (Saguil A, Wyrick K, Hallgren J. Diagnostic approach to pleural effusion[J].Am FamPhysician, 2014, 90(2): 99-104.). For transudates, the focus is usually only on the mechanism causing the effusion and treating the underlying disease, such as potential heart failure. However, for exudates, precise differential diagnosis is needed to determine the specific cause and guide the specific treatment (Beaudoin S, Gonzalez AV. Evaluation of the patient with pleural effusion[J]. CMAJ, 2018,190(10): E291-E295.). Benign effusions, most commonly tuberculous and parapneumonic, as well as malignant effusions caused by various types of tumors, are common types of exudative pleural effusions. Clinically, these common causes should be considered in all undiagnosed patients.

[0003] In recent years, the incidence of tumors has shown a trend of increasing year by year. Malignant pleural effusion (MPE) mainly originates from pleural seeding metastasis of primary lung tumors, mesothelioma, malignant pleural diseases or other malignant diseases (Asciak R, Rahman NM. Malignant Pleural Effusion: From Diagnostics to Therapeutics[J]. Clin Chest Med, 2018, 39(1): 181-193.). Up to 50% of patients with metastatic malignant tumors develop pleural effusion at the time of diagnosis or during disease progression (Egan AM, McPhillips D, Sarkar S, et al. Malignant pleural effusion[J]. QJM, 2014, 107(3):179-184; Kapp CM, LeeHJ. Malignant Pleural Effusions[J]. Clin Chest Med, 2021, 42(4): 687-696.). The mechanism of MPE is complex. Tumor growth compresses or obstructs parietal pleural vessels and lymphatic vessels, leading to drainage obstruction, which is considered to be the main factor. Furthermore, it has been discovered that tumor cells drive the formation of effusion through a cascade of interactions between the host's vascular system and immune system. Inflammatory cells, mesothelial cells, and endothelial cells in the pleural microenvironment interact with primary or metastatic tumor cells, leading to excessive effusion production through a highly permeable pleural vascular network. Multiple cells and molecules that regulate pleural inflammation, stimulate tumor angiogenesis, and affect vascular permeability are involved in this process.

[0004] Pleural effusion cytology and pleural biopsy remain the gold standard for diagnosing MPE (Porcel JM, Azzopardi M, Koegelenberg CF, et al. The diagnosis of pleural effusions. Expert Rev Respir Med. 2015;9(6): 801-15.), but both have limitations. Cytological examination has near 100% specificity but low sensitivity (Psallidas I, Kalomenidis I, Porcel JM, et al. Malignant pleural effusion: from bench to bedside. Eur Respir Rev 2016;25:189-98; Loveland P, Christie M, Hammerschlag G, et al. Diagnostic yield of pleural fluid cytology in malignant effusions: an Australian tertiary centre experience. Intern Med J 2018;48:1318-24.). Pleural biopsy, while highly specific, is an invasive procedure with associated risks (Wang XJ, Yang Y, Wang Z, et al. Efficacy and safety of diagnostic thoracoscopy in undiagnosed pleural effusions. Respiration 2015;90:251-5.). In clinical practice, the origin of pleural effusion is unclear in 20-30% of cases. These patients often require repeated thoracentesis due to the lack of a definitive diagnosis, frequently resulting in years of undiagnosed cases. This places a significant burden on patients' physical and mental well-being, and presents a substantial challenge to both diagnosis and treatment. Given these challenges, the development of non-invasive and convenient diagnostic tools for pleural effusion (MPE) holds great promise. Compared to cytological examination and biopsy, biomarker-based detection offers advantages such as being minimally invasive, more convenient, and cost-effective.Carcinoembryonic antigen (CEA), as one of the most widely studied biomarkers for malignant pleural effusion (MPE), has high specificity (0.94; 95% confidence interval: 0.93-0.95), but low sensitivity (0.54; 95% confidence interval: 0.52-0.55), limiting its application in the differential diagnosis of MPE (Shi HZ, Liang QL, Jiang J, et al. Diagnostic value of carcinoembryonic antigen in malignant pleural effusion: a meta-analysis. Respirology 2008;13:518-27.). Therefore, there is still an urgent clinical need for simple, minimally invasive, and highly accurate diagnostic biomarkers to effectively differentiate MPE from benign pleural effusion (BPE).

[0005] CC motif chemokine ligand 2 (CCL2), also known as monocyte chemoattractant protein-1 (MCP-1), is a key member of the CC chemokine family (Murphy PM, Baggiolini M, Charo IF, et al. International union of pharmacology. XXII. Nomenclature for chemokine receptors. Pharmacol Rev 2000;52:145-76.). As an important inflammatory mediator, CCL2 can be secreted by various cells, including endothelial cells, smooth muscle cells, and tumor cells. CCL2, by binding to its receptor CCR2, can effectively recruit and activate monocytes and macrophages (Antony VB, Godbey SW, Kunkel SL, et al. Recruitment of inflammatory cells to the pleural space. Chemotactic cytokines, IL-8, and monocyte chemotactic peptide-1 in human pleural fluids. J Immunol 1993;151:7216-23.), thereby playing a role in inflammatory responses, immune regulation, angiogenesis, and disease progression. Studies have found elevated CCL2 expression in various pathological states, including cardiovascular diseases, autoimmune diseases, and malignant tumors, making CCL2 a potential disease biomarker. Previous studies have shown that CCL2 expression is elevated in various inflammatory diseases, but its specificity is limited. In addition, individual differences and disease stages affect CCL2 expression, making it difficult to use as an independent diagnostic tool. Summary of the Invention

[0006] This invention utilizes three risk predictors—age, pleural effusion ADA, and pleural effusion CCL2—to construct a nomogram model. By combining the superior NPV and sensitivity of CCL2 with the complementary value of age and ADA, this nomogram overcomes the limitations of single-marker diagnosis, providing a minimally invasive and precise tool for identifying MPE in clinical practice. Based on this, this invention was completed.

[0007] In a first aspect, the present invention provides a set of biomarkers for the differential diagnosis of malignant pleural effusion, wherein the biomarkers are a combination of ADA and CCL2; the combination of biomarkers, in conjunction with the age of the subject, is used to differentially diagnose malignant pleural effusion.

[0008] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it was indicated that the subject had malignant pleural effusion.

[0009] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L and CCL2 ≥246.4 pg / mL, the older the age, the lower the ADA level, the higher the CCL2 level, the higher the total score of the nomogram, and the higher the corresponding diagnostic efficacy of MPE.

[0010] Furthermore, the total score of the nomogram is obtained by summing the corresponding scores of each biomarker and age.

[0011] Secondly, the present invention provides a differential diagnostic model for malignant pleural effusion. The model uses the relative expression levels of biomarkers in the biological samples of the subjects as training samples, and calculates the sensitivity and specificity of the ROC curve based on pathological diagnosis. The biomarker is a combination of ADA and CCL2.

[0012] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it was indicated that the subject had malignant pleural effusion.

[0013] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L and CCL2 ≥246.4 pg / mL, the older the age, the lower the ADA level, the higher the CCL2 level, the higher the total score of the nomogram, and the higher the corresponding diagnostic efficacy of MPE.

[0014] Furthermore, the total score of the nomogram is obtained by summing the corresponding scores of each biomarker and age.

[0015] Furthermore, the biological sample is selected from pleural effusion, peripheral blood, or diseased tissue.

[0016] Preferably, the biological sample is pleural effusion.

[0017] Thirdly, the present invention provides the application of the biomarker described in the first aspect in the preparation of a reagent for the differential diagnosis of malignant pleural effusion, wherein the biomarker is a combination of ADA and CCL2, and the reagent is a reagent for detecting the relative expression levels of ADA and CCL2 in the biological sample of a subject.

[0018] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it was indicated that the subject had malignant pleural effusion.

[0019] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L and CCL2 ≥246.4 pg / mL, the older the age, the lower the ADA level, the higher the CCL2 level, the higher the total score of the nomogram, and the higher the corresponding diagnostic efficacy of MPE.

[0020] Furthermore, the total score of the nomogram is obtained by summing the corresponding scores of each biomarker and age.

[0021] Furthermore, the biological sample is selected from pleural effusion, peripheral blood, or diseased tissue.

[0022] Preferably, the biological sample is pleural effusion.

[0023] Fourthly, the present invention provides a kit for the differential diagnosis of malignant pleural effusion, the kit containing reagents for detecting the relative expression levels of ADA and CCL2 in biological samples of subjects.

[0024] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it was indicated that the subject had malignant pleural effusion.

[0025] Furthermore, when the subject's age was ≥65 years, ADA <22.5 U / L and CCL2 ≥246.4 pg / mL, the older the age, the lower the ADA level, the higher the CCL2 level, the higher the total score of the nomogram, and the higher the corresponding diagnostic efficacy of MPE.

[0026] Furthermore, the total score of the nomogram is obtained by summing the corresponding scores of each biomarker and age.

[0027] Furthermore, the biological sample is selected from pleural effusion, peripheral blood, or diseased tissue.

[0028] Preferably, the biological sample is pleural effusion.

[0029] Furthermore, the kit includes an ELISA assay kit and / or an enzyme activity assay kit.

[0030] Beneficial effects This invention found that the CCL2 level in the pleural effusion of patients with myeloproliferative disorders (MPE) was significantly higher than that of non-MPE patients. In the screening group, the diagnostic cutoff value for CCL2 was 246.4 pg / mL, with an AUC of 0.706 and a sensitivity of 98.25%; in the validation group, the diagnostic cutoff value for CCL2 was 216.4 pg / mL, with an AUC of 0.635 and a sensitivity of 100%. Compared with other biochemical indicators in pleural effusion, CCL2 in pleural effusion has strong diagnostic efficacy, especially in excluding diagnoses (negative predictive value = 97.6%).

[0031] Given the limitations of single biomarkers, this invention further screened three risk predictors—age, pleural effusion ADA, and pleural effusion CCL2—and constructed a nomogram model. This nomogram can intuitively quantify the relative contribution of each variable and generate a total diagnostic score; a higher score indicates a greater likelihood of MPE. The nomogram exhibits excellent diagnostic performance: in the training group, its sensitivity was 0.956, specificity was 0.948, and AUC was 0.974, outperforming previously reported models. Calibration curves confirmed a high degree of consistency between the model's diagnostic value and pathological diagnosis; robust performance in the validation group further confirmed its reliability. By combining the excellent NPV and sensitivity of CCL2 with the complementary value of age and ADA, this nomogram overcomes the shortcomings of single biomarker diagnosis, providing a minimally invasive and precise tool for MPE diagnosis in clinical practice. Attached Figure Description

[0032] Figure 1 This invention provides the patient inclusion and exclusion criteria and procedures.

[0033] Figure 2 The concentration of the chemokine CC motif ligand 2 (CCL2) in pleural effusion in the training group and the diagnostic receiver operating characteristic (ROC) curve. Note: (A) Comparison of CCL2 concentrations in malignant pleural effusion (MPE) and non-malignant pleural effusion (non-MPE) in the training group. (B) ROC curve showing the diagnostic efficacy of CCL2 for malignant pleural effusion (MPE) in the training group.

[0034] Figure 3 The levels of the chemokine CC motif ligand 2 (CCL2) in pleural effusions stratified according to clinicopathological features. Note: (AH) are logarithmic values ​​stratified according to the following clinicopathological features. 2Post-conversion CCL2 levels: sex (female vs. male), age (<65 years vs. ≥65 years), total protein (TP, <49.85 g / L vs. ≥49.85 g / L), adenosine deaminase (ADA, <22.5 U / L vs. ≥22.5 U / L), glucose (Glu, <5.55 mmol / L vs. ≥5.55 mmol / L), chloride (Cl, <109.3 mmol / L vs. ≥109.3 mmol / L), lactate dehydrogenase (LDH, <435 U / L vs. ≥435 U / L), carcinoembryonic antigen (CEA, <2.63 ng / mL vs. ≥2.63 ng / mL). **** P <0.0001 (highly significant difference), *** P <0.001 (highly significant difference),** P <0.01 (significant difference), ns = no statistical difference. Gender, Female, Male, Age, years.

[0035] Figure 4 The present figures show the receiver operating characteristic (ROC) curves for subgroup CC motif ligand 2 (CCL2) in training groups stratified by adenosine deaminase (ADA), chloride (Cl), and carcinoembryonic antigen (CEA) levels. Note: (A, D) Diagnostic ROC curves for CCL2 in training groups stratified by adenosine deaminase (ADA, <22.5 U / L and ≥22.5 U / L); (B, E) Diagnostic ROC curves for CCL2 in training groups stratified by total protein (TP, <49.85 mmol / L and ≥49.85 mmol / L); (C, F) Diagnostic ROC curves for CCL2 in training groups stratified by carcinoembryonic antigen (CEA, <2.63 ng / mL and ≥2.63 ng / mL).

[0036] Figure 5 Comparison of receiver operating characteristic (ROC) curves for the diagnosis of malignant pleural effusion using chemokine CC motif ligand 2 (CCL2), carcinoembryonic antigen (CEA), and lactate dehydrogenase (LDH). Note: These ROC curves demonstrate the efficacy of CCL2, CEA, and LDH in diagnosing malignant pleural effusion. Wherein, CCL2 is chemokine CC motif ligand 2, CEA is carcinoembryonic antigen, and LDH is lactate dehydrogenase.

[0037] Figure 6To validate the concentration of the chemokine CC motif ligand 2 (CCL2) in pleural effusion and to obtain diagnostic receiver operating characteristic (ROC) curves. Note: (A) Comparison of CCL2 concentrations in malignant pleural effusion (MPE) and non-malignant pleural effusion (non-MPE) in the validation group. (B) ROC curves show the diagnostic efficacy of CCL2 for malignant pleural effusion (MPE) in the validation group.

[0038] Figure 7 The selection frequencies of the variables obtained using the Bootstrap LASSO method are shown. Note: The dashed lines represent the frequency thresholds of 50% (red dashed line) and 70% (blue dashed line), respectively. The bar chart represents the selection frequency (%) of each variable (adenosine deaminase (ADA), age, chemokine CC motif ligand 2 (CCL2), carcinoembryonic antigen (CEA), and total protein (TP)). Wherein, LASSO is the minimum absolute contraction and selection operator; ADA is adenosine deaminase; CCL2 is chemokine CC motif ligand 2; CEA is carcinoembryonic antigen; and TP is total protein.

[0039] Figure 8 This is a nomogram constructed based on the chemokine CC motif ligand 2 (CCL2) in the training group. Note: ADA is adenosine deaminase, and CCL2 is chemokine CC motif ligand 2.

[0040] Figure 9 The receiver operating characteristic (ROC) curves and calibration curves for the nomogram based on chemokine CC motif ligand 2 (CCL2) are shown in the training group. Note: (A) The ROC curve demonstrates the diagnostic power of the CCL2-based nomogram; (B) The calibration curve reflects the consistency between the predicted risk and the actual risk of the nomogram. Figure 10 The receiver operating characteristic (ROC) curves and calibration curves for the nomogram based on chemokine CC motif ligand 2 (CCL2) are shown in the validation group. Note: (A) The ROC curve demonstrates the diagnostic power of the CCL2-based nomogram; (B) The calibration curve reflects the consistency between the predicted risk and the actual risk of the nomogram. Detailed Implementation

[0041] The specific embodiments of the present invention will be further described below. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the embodiments described below can be combined with each other as long as they do not conflict with each other.

[0042] Unless otherwise specified, the experimental methods used in the following embodiments are conventional methods, and the experimental materials used in the following embodiments are all available through conventional commercial channels.

[0043] Example 1: Discovery Group 1. Method 1.1 Sample Source This invention collects clinical data of patients with exudative pleural effusion who were hospitalized at Beijing Chest Hospital between June 2019 and April 2022.

[0044] 1.2 Diagnosis Malignant pleural effusion (MPE): diagnosed by positive pleural effusion cytology or pleural tissue biopsy.

[0045] Tuberculous pleural effusion (TPE): diagnosed by microbiological evidence from pleural effusion (positive acid-fast staining, positive culture of Mycobacterium tuberculosis, or positive nucleic acid amplification test) and / or pleural biopsy showing caseous necrotizing granuloma (Shaw JA, Irusen EM, Diacon AH, et al. Pleuraltuberculosis: A concise clinical review. Clin Respir J 2018;12:1779-86.).

[0046] Parapneumonic pleural effusion (PPE): Diagnosed based on clear clinical and radiological evidence of pneumonia, with pleural effusion associated with pneumonia and improvement after antibiotic treatment (Davies HE, Davies RJ, Davies CW. Management of pleural infection in adults: British Thoracic Society Pleural Disease Guideline 2010. Thorax 2010;65 Suppl 2:ii41-53.).

[0047] All final diagnoses were confirmed by the attending physician based on complete clinical, laboratory, and imaging data.

[0048] 1.3 Discharge Standards Inclusion criteria: (1) diagnosed with exudative pleural effusion according to Light's criteria; (2) signed informed consent form to participate in this invention.

[0049] Exclusion criteria: (1) transudative pleural effusion; (2) having received antitumor or antituberculosis treatment within the past 5 years; (3) having diabetes or immunodeficiency diseases; (4) being under 18 years of age; (5) being pregnant; (6) pleural effusion secondary to trauma or surgery; (7) pleural effusion newly developed during hospitalization.

[0050] After screening, a total of 202 patients with exudative pleural effusion were included. The concentration of CCL2 in the pleural effusion was then measured using enzyme-linked immunosorbent assay (ELISA), excluding patients whose ELISA results for two undiluted samples were both negative. Finally, eligible patients were randomly assigned to a training group and a validation group in a 7:3 ratio using random sampling, with a fixed random seed to ensure sampling reproducibility. The training group was used to construct a diagnostic nomogram to differentiate MPE from other types of exudative pleural effusion, while the validation group was used to evaluate the diagnostic efficacy of this nomogram.

[0051] This invention has been approved by the Ethics Committee of Beijing Chest Hospital, Capital Medical University (Ethics Approval No.: (2025) Clinical Review-Research-No. (31)).

[0052] 1.4 Data Collection All members of the research team received standardized training. They collected data by reviewing patient medical records and verified the accuracy of the data using a double-entry and verification method. Clinical data were obtained within one week of the diagnosis of pleural effusion. The variables collected included patient name, gender, age, routine biochemical indicators of pleural effusion, and carcinoembryonic antigen (CEA) levels.

[0053] Pleural effusion samples were collected via diagnostic thoracentesis or medical thoracoscopy. The supernatant was obtained after centrifugation and stored for testing.

[0054] 1.5 Detection Indicators The concentration of CCL2 in pleural effusion was determined using an enzyme-linked immunosorbent assay (ELISA) kit (human MCP-1 / CCL-2 ELISA kit, catalog number: LEH831-2S, manufacturer: LAIZEE, China). The procedure was strictly performed in accordance with the kit instructions, and the results are expressed in pg / mL.

[0055] The levels of adenosine deaminase (ADA) were determined using a peroxidase assay kit (B1036, purchased from Sinocare Biotechnology Co., Ltd., China). The procedure was strictly followed according to the instrument and kit instructions, and the results are expressed as U / L.

[0056] 1.6 Statistical Analysis Data description: Continuous variables are represented by "median (interquartile range, IQR)" and categorical variables are represented by "number of cases (percentage)".

[0057] Between-group comparisons: Categorical variables were analyzed using the chi-square test or Fisher's exact test; continuous variables were compared using the t-test.

[0058] ROC curve analysis: Receiver operating characteristic (ROC) curves were plotted, and the optimal cutoff value for each indicator was determined using the Youden's Index. Based on the cutoff values ​​for each indicator, the subjects were divided into two groups, and the differences in the area under the ROC curve (AUC) of CCL2 between the different subgroups were compared.

[0059] Predictive model construction: First, perform univariate logistic regression analysis on each variable; then, in the univariate analysis... P Variables with a value <0.05 underwent multicollinearity testing. Variables without multicollinearity were included in the Bootstrap LASSO model and selected for use in constructing the diagnostic model. A nomogram was plotted based on the selected predictors. The total patient score was obtained by summing the scores of each variable, and then the total score was mapped to the diagnostic value of MPE.

[0060] Model validation: The performance of the diagnostic nomogram was validated in both the training and validation groups. The model’s discriminative power and calibration were evaluated using ROC curves, the c-index, and the calibration plot.

[0061] All statistical analyses were performed using R statistical software (version 4.5.1) and IBM SPSS 25.0 software. P A value <0.05 is considered statistically significant.

[0062] 2. Results 2.1 Clinical Features This invention included 142 patients with exudative pleural effusion in the training group. In the training group, 45 patients (31.7%) were ≥65 years old; 57 patients (40.1%) were diagnosed with MPE, and 85 patients (59.9%) had tuberculous pleural effusion (TPE) or pneumonia-like pleural effusion (PPE). There were 83 males (58.5%) in this group, with a mean age of 50 years (range: 29–67 years). Compared with the non-MPE group, the MPE group showed significantly lower levels of total protein and lactate dehydrogenase (LDH), slightly higher glucose levels, and significantly higher CCL2 levels; there were no significant differences between the two groups in chloride ion concentration, monocyte percentage, and polymorphonuclear cell percentage (Table 1).

[0063] Table 1. Clinical characteristics of patients with pleural effusion in the training group Note: For continuous variables, data are expressed as median (interquartile range, IQR); for categorical variables, data are expressed as percentage. MPE: Malignant pleural effusion; ADA: Adenosine deaminase; LDH: Lactate dehydrogenase; WBC: White blood cells; PFMCs: Monocytes in pleural effusion; PMNs: Polymorphonuclear leukocytes; CEA: Carcinoembryonic antigen; CCL2: CC chemokine ligand 2.

[0064] 2.2 CCL2 and ADA Levels The CCL2 concentration in the MPE group was significantly higher than that in the non-MPE group, while the ADA level was significantly lower in the MPE group (Table 1). The diagnostic efficacy assessment of CCL2 showed an area under the ROC curve (AUC) of 0.706, a specificity of 38.82%, a sensitivity of 98.25%, and an optimal cutoff value of 246.4 pg / mL (Figure 2).

[0065] Based on this cutoff value, patients were divided into a low CCL2 group (<246.4 pg / mL, n=34) and a high CCL2 group (≥246.4 pg / mL, n=108). High CCL2 levels were significantly associated with older patients, lower ADA levels, higher total cell counts, and lower percentage of polymorphonuclear cells (Table 2). Table 2 Comparison of clinical characteristics between the low CCL2 and high CCL2 groups. Note: For continuous variables, data are expressed as median (interquartile range, IQR); for categorical variables, data are expressed as percentage. ADA: adenosine deaminase; LDH: lactate dehydrogenase; WBC: white blood cells; PFMCs: pleural effusion mononuclear cells; PMNs: polymorphonuclear leukocytes; CEA: carcinoembryonic antigen; CCL2: chemokine (CC motif) ligand 2.

[0066] Further analysis of the median distribution of CCL2 levels across different clinicopathological features: Total protein, ADA, glucose, chloride, LDH, and CEA in pleural effusion were grouped and analyzed according to their respective optimal cutoff values. Results showed that patients aged ≥65 years, with TP < 49.85 g / L, ADA < 22.5 U / L, or CEA ≥ 2.63 ng / mL had significantly higher CCL2 levels (all...). P <0.05; Figure 3).

[0067] 2.3 Subgroup ROC curve analysis To evaluate the diagnostic efficacy of CCL2 in different subgroups, this invention plotted ROC curves for each subgroup and compared AUC values: In age-stratified subgroups (further stratified by ADA level), the AUC of CCL2 in the ADA ≥ 22.5 U / L subgroup (0.852) was higher than that in the ADA < 22.5 U / L subgroup (0.520). In the subgroups stratified by chloride ion concentration, the AUC of CCL2 in the subgroup with TP ≥ 49.85 g / L (0.620) was lower than that in the subgroup with TP < 49.85 g / L (0.732).

[0068] In the subgroups stratified by CEA level, the AUC of CCL2 in the subgroup with CEA ≥ 2.63 ng / mL (0.620) was higher than that in the subgroup with CEA < 2.63 ng / mL (0.716).

[0069] The results above indicate that CCL2 has better diagnostic efficacy in patients with ADA ≥ 22.5 U / L compared with patients with CI ≥ 109.25 mmol / L or CEA ≥ 2.62 ng / mL (Figure 4).

[0070] 2.4 Comparison of other biomarkers for pleural effusion To compare the value of CCL2 with other established biomarkers for pleural effusion in the diagnosis of mesothelial effusion (MPE), this invention plotted ROC curves for each biomarker and calculated the corresponding AUC values. The results showed that the AUC of CCL2 (0.754) was higher than that of lactate dehydrogenase (LDH, 0.619), but slightly lower than that of carcinoembryonic antigen (CEA, 0.799), suggesting that CCL2 has moderate but clinically significant value in the diagnosis of MPE. Figure 5 ).

[0071] 2.5 Diagnostic Consistency Analysis Table 3 summarizes the diagnostic consistency and performance of various biochemical indicators for pleural effusion (MPE). Although the concordance between CCL2 and the gold standard for diagnosis is only moderate (Kappa=0.359), its extremely high sensitivity (98.3%) and negative predictive value (NPV=97.6%) indicate that this indicator is outstanding in identifying true positive cases and reliably excluding false negative cases. Therefore, CCL2 is very suitable for exclusionary diagnosis—a conclusion also supported by its low negative likelihood ratio (NLR=0.044).

[0072] In contrast, the Kappa coefficients for both total protein and LDH were negative (-0.148 and -0.067, respectively), suggesting a negative correlation between them and the gold standard diagnostic results. Although these two indicators have relatively high specificity (e.g., 61.2% for total protein), their extremely low positive predictive values ​​(PPVs, 0.400 and 0.526, respectively) indicate that they are not suitable as independent diagnostic indicators.

[0073] Glucose showed weak concordance with the gold standard (Kappa=0.042), with a specificity of 75.7% and an NPV of 64.1%, indicating inferior diagnostic efficacy compared to CCL2. Chloride showed very weak concordance with the gold standard (Kappa=0.254), but maintained high specificity (91.8%), thus playing a role in identifying non-MPE cases.

[0074] It is worth noting that among all the biomarkers tested, CEA showed high specificity (89.0%), the highest PPV (0.769), and positive likelihood ratio (PLR=6.345), making it the most reliable indicator for diagnosing MPE in pleural effusion.

[0075] Table 3. Diagnostic consistency and efficacy of biochemical indicators in pleural effusion for malignant pleural effusion. Note: MPE, malignant pleural effusion; CCL2, chemokine (CC motif) ligand 2; LDH, lactate dehydrogenase; CEA, carcinoembryonic antigen; PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio.

[0076] Example 2 Verification Group 1. Method 1.1 Sample Source This invention collected clinical data from patients with exudative pleural effusion who were hospitalized at Beijing Chest Hospital between June 2019 and April 2022. A total of 60 patients with exudative pleural effusion were included in the validation group.

[0077] 1.2 Diagnosis Malignant pleural effusion (MPE): diagnosed by positive pleural effusion cytology or pleural tissue biopsy.

[0078] Tuberculous pleural effusion (TPE): Diagnosed by microbiological evidence from pleural effusion (positive acid-fast staining, positive culture of Mycobacterium tuberculosis, or positive nucleic acid amplification test) and / or pleural biopsy showing caseous necrotizing granuloma.

[0079] Parapneumonic pleural effusion (PPE): Diagnosed based on clear clinical and radiological evidence of pneumonia, with pleural effusion related to pneumonia and improvement after antibiotic treatment.

[0080] All final diagnoses were confirmed by the attending physician based on complete clinical, laboratory, and imaging data.

[0081] 1.3 Discharge Standards Inclusion criteria: (1) diagnosed with exudative pleural effusion according to Light's criteria; (2) signed informed consent form to participate in this invention.

[0082] Exclusion criteria: (1) transudative pleural effusion; (2) having received antitumor or antituberculosis treatment within the past 5 years; (3) having diabetes or immunodeficiency diseases; (4) being under 18 years of age; (5) being pregnant; (6) pleural effusion secondary to trauma or surgery; (7) pleural effusion newly developed during hospitalization.

[0083] After screening, a total of 202 patients with exudative pleural effusion were included. The concentration of CCL2 in the pleural effusion was then measured using enzyme-linked immunosorbent assay (ELISA), excluding patients whose ELISA results for two undiluted samples were both negative. Finally, eligible patients were randomly assigned to a training group and a validation group in a 7:3 ratio using random sampling, with a fixed random seed to ensure sampling reproducibility. The training group was used to construct a diagnostic nomogram to differentiate MPE from other types of exudative pleural effusion, while the validation group was used to evaluate the diagnostic efficacy of this nomogram.

[0084] This invention has been approved by the Ethics Committee of Beijing Chest Hospital, Capital Medical University (Ethics Approval No.: (2025) Clinical Review-Research-No. (31)).

[0085] 1.4 Data Collection All members of the research team received standardized training. They collected data by reviewing patient medical records and verified the accuracy of the data using a double-entry and verification method. Clinical data were obtained within one week of the diagnosis of pleural effusion. The variables collected included patient name, gender, age, routine biochemical indicators of pleural effusion, and carcinoembryonic antigen (CEA) levels.

[0086] Pleural effusion samples were collected via diagnostic thoracentesis or medical thoracoscopy. The supernatant was obtained after centrifugation and stored for testing.

[0087] 1.5 CCL2 and ADA Detection Refer to Example 1.

[0088] 1.6 Statistical Analysis Data description: Continuous variables are represented by "median (interquartile range, IQR)" and categorical variables are represented by "number of cases (percentage)".

[0089] Between-group comparisons: Categorical variables were analyzed using the chi-square test or Fisher's exact test; continuous variables were compared using the t-test.

[0090] ROC curve analysis: Receiver operating characteristic (ROC) curves were plotted, and the optimal cutoff value for each indicator was determined using the Youden's Index. Based on the cutoff values ​​for each indicator, the subjects were divided into two groups, and the differences in the area under the ROC curve (AUC) of CCL2 between the different subgroups were compared.

[0091] Predictive model construction: First, perform univariate logistic regression analysis on each variable; then, in the univariate analysis... P Variables with a criterion <0.05 underwent multicollinearity testing. Variables without multicollinearity were included in the Bootstrap LASSO model to screen predictors for constructing a diagnostic nomogram. A visual nomogram was then created based on the selected predictors. The total patient score was obtained by summing the scores of each variable, and the total score was then mapped to the predicted probability of MPE.

[0092] Model validation: The performance of the diagnostic nomogram was validated in both the training and validation groups. The model's discriminative power and calibration were evaluated using ROC curves, the c-index, and the calibration plot.

[0093] All statistical analyses were performed using R statistical software (version 4.5.1) and IBM SPSS 25.0 software. P A value <0.05 is considered statistically significant.

[0094] 2. Results 2.1 Clinicopathological features This invention included 60 patients with exudative pleural effusion in the validation group. There were no statistically significant differences in clinicopathological characteristics between the training group and the validation group (Table 4).

[0095] Table 4 Comparison of clinicopathological characteristics between the training group and the validation group Note: For continuous variables, data are expressed as median (interquartile range, IQR); for categorical variables, data are expressed as number of cases (percentage, %). Wherein, ADA: adenosine deaminase; LDH: lactate dehydrogenase; WBC: white blood cells; PFMCs: pleural effusion mononuclear cells; PMNs: polymorphonuclear leukocytes; CEA: carcinoembryonic antigen; CCL2: chemokine (CC motif) ligand 2.

[0096] 2.2 CCL2 and ADA Levels The CCL2 concentration in the MPE group was significantly higher than that in the non-MPE group, while the ADA level was significantly lower in the MPE group. Diagnostic efficacy assessment of CCL2 showed an area under the ROC curve (AUC) of 0.635, a specificity of 27.78%, and a sensitivity of 100% (Figure 6).

[0097] 2.3 Construction and Validation of Nomogram Based on CCL2 Univariate logistic regression analysis showed that age, total protein, ADA, CEA, and CCL2 were significantly associated with MPE. After removing variables with multicollinearity, Bootstrap LASSO regression was used to further screen variables. Figure 7 Ultimately, age, binary ADA (indicator), and binary CCL2 (indicator) were identified as independent predictors and incorporated into the MPE diagnostic nomogram (Table 5; Figure 8).

[0098] The total score of the nomogram is calculated by summing the corresponding scores of each predictor. When the subject's age is ≥65 years, ADA <22.5 U / L and CCL2 ≥246.2 pg / mL, the older the age, the lower the ADA level, and the higher the CCL2 level, the higher the total score and the higher the corresponding MPE diagnostic efficiency.

[0099] The nomogram demonstrates excellent discriminative power: an AUC of 0.974, a specificity of 94.8%, and a sensitivity of 95.6% (Figure 9A); the c-index is 0.974, indicating a very high degree of consistency between the predicted results and the actual outcomes. The calibration curve shows that the model's diagnostic value is highly consistent with the pathological diagnosis (Figure 9B).

[0100] The validation results in the validation group further confirmed the stability of the nomogram: its AUC value still exceeded 0.9. Figure 10 The calibration plots of the validation group further verified the close correspondence between the model's diagnostic value and the pathological diagnostic results, confirming that the model has good stability and reliability.

[0101] Table 5. Univariate Logistic Regression and Bootstrap LASSO Variable Screening for Indicators Related to the Diagnosis of Malignant Pleural Effusion Note: ADA, adenosine deaminase; LDH, lactate dehydrogenase; CEA, carcinoembryonic antigen; CCL2, chemokine (CC motif) ligand 2.

[0102] 3. Conclusion The results of this invention indicate that CC chemokine ligand 2 (CCL2) in pleural effusion can serve as a potential biomarker for diagnosing malignant pleural effusion (MPE), but its specificity when used alone remains poor. Other single biomarkers, such as total protein and lactate dehydrogenase (LDH), show an inverse correlation with the gold standard and have extremely low PPVs, thus they are not suitable as independent confirmatory indicators. The concordance between glucose and chloride and the gold standard is only weak to very weak. In summary, these results demonstrate that relying solely on a single biomarker for the diagnosis of MPE has significant limitations.

[0103] Given the limitations of single biomarkers, this invention employs a nomogram model to construct three risk predictors—age, pleural effusion ADA, and pleural effusion CCL2. This nomogram visually quantifies the relative contribution of each variable and generates a total diagnostic score; a higher score indicates a greater likelihood of MPE, demonstrating excellent diagnostic performance. By combining the superior NPV and sensitivity of CCL2 with the complementary value of age and ADA, this nomogram overcomes the shortcomings of single-marker diagnosis, providing a minimally invasive and precise tool for diagnosing MPE in clinical practice.

Claims

1. A set of biomarkers for the differential diagnosis of malignant pleural effusion, characterized in that, The biomarker is a combination of ADA and CCL2; the combination of biomarkers, in conjunction with the subject's age, is used to differentiate and diagnose malignant pleural effusion.

2. The biomarker as described in claim 1, characterized in that, When a subject is ≥65 years old, has ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it is considered that the subject has malignant pleural effusion.

3. The biomarker as described in claim 1, characterized in that, When the subject's age is ≥65 years, ADA <22.5 U / L and CCL2 ≥246.4 pg / mL, the older the age, the lower the ADA level, and the higher the CCL2 level, the higher the total score of the nomogram, and the higher the diagnostic efficacy of MPE. The total score of the nomogram is obtained by summing the corresponding scores of each biomarker and age.

4. A differential diagnostic model for malignant pleural effusion, wherein the differential diagnostic model uses the relative expression levels of biomarkers in the biological samples of subjects as training samples, and calculates the sensitivity and specificity of ROC curves based on pathological diagnosis; wherein the biomarker is a combination of ADA and CCL2.

5. The differential diagnosis model as described in claim 4, characterized in that, When a subject is ≥65 years old, has ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it is considered that the subject has malignant pleural effusion.

6. The application of the biomarker as described in claim 1 in the preparation of a reagent for the differential diagnosis of malignant pleural effusion, wherein the biomarker is a combination of ADA and CCL2, and the reagent is a reagent for detecting the relative expression levels of ADA and CCL2 in the biological sample of the subject.

7. The application as described in claim 6, characterized in that, When a subject is ≥65 years old, has ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it is considered that the subject has malignant pleural effusion.

8. The application as described in claim 6, characterized in that, The biological samples were selected from pleural effusion, peripheral blood, or diseased tissue.

9. A kit for the differential diagnosis of malignant pleural effusion, the kit containing reagents for detecting the relative expression levels of ADA and CCL2 in a biological sample of a subject.

10. The application as described in claim 9, characterized in that, When a subject is ≥65 years old, has ADA <22.5 U / L, and CCL2 ≥246.4 pg / mL, it is considered that the subject has malignant pleural effusion.

Images