Development and validation of a nomogram model of lung metastasis in breast cancer based on machine learning algorithm and cytokines

Development and validation of a nomogram model of lung metastasis in breast cancer based on machine learning algorithm and cytokines | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Development and validation of a nomogram model of lung metastasis in breast cancer based on machine learning algorithm and cytokines Zhaoyi Li, Hao Miao, Wei Bao, Lansheng Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5841908/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Apr, 2025 Read the published version in BMC Cancer → Version 1 posted 8 You are reading this latest preprint version Abstract Background The relationship between cytokines and lung metastasis (LM) in breast cancer (BC) remains unclear and current clinical methods for identifying breast cancer lung metastasis (BCLM) lack precision, thus underscoring the need for an accurate risk prediction model. This study aimed to apply machine learning algorithms for identifying the key risk factors for BCLM before developing a reliable prediction model centered on cytokines. Methods This population-based retrospective study included 326 BC patients admitted to the Second Affiliated Hospital of Xuzhou Medical University between September 2018 and September 2023. After randomly assigning the patients to a training cohort (70%; n = 228) or a validation cohort (30%; n = 98) the risk factors for BCLM were identified using Least Absolute Shrinkage and Selection Operator (LASSO), Extreme Gradient Boosting (XGBoost) and Random Forest (RF) models. Significant risk factors were visualized with a Venn diagram and incorporated into a nomogram model, the performance of which was then evaluated according to three criteria, namely discrimination, calibration and clinical utility using calibration plots, receiver operating characteristic (ROC) curves and decision curve analysis (DCA). Results Among the cohort, 70 patients developed LM. A nomogram was then developed to predict the 5-year and 10-year BCLM risk by incorporating five key variables, namely endocrine therapy, hsCRP, IL6, IFN-ɑ and TNF-ɑ. For the 5-year prediction model, the training and validation cohorts had AUC values of 0.786 (95% CI: 0.691–0.881) and 0.627 (95% CI: 0.441–0.813), respectively, while for the 10-year prediction model, the corresponding AUC values were 0.687 (95% CI: 0.528–0.847) and 0.797 (95% CI: 0.605–0.988), respectively. ROC analysis further confirmed the model’s strong discriminative ability, while calibration plots indicated that the predicted and observed outcomes were in good agreement in both cohorts. Finally, DCA demonstrated the model’s effectiveness in clinical practice. Conclusion Using machine learning algorithms, this study developed aa nomogram that could effectively identify BC patients who were at a higher risk of developing LM, thus providing a valuable tool for decision-making in clinical settings. Breast cancer metastasis cytokine machine learning nomogram Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Breast cancer (BC), the most frequently diagnosed malignant tumor in women worldwide, remains the major cause of cancer-related mortality among the female population[1]. It often begins as a localized disease but can subsequently spread to lymph nodes and distant organs, thereby posing significant challenges to effective treatment[2]. Approximately 10–15% of breast cancer patients develop invasive diseases, with distant metastases occurring within three years of the initial diagnosis. However, metastases may also emerge at least a decade after initial detection of the cancer. Furthermore, the heterogeneity of breast cancer metastasis complicates not only the determination of effective treatment strategies but also the assessment of metastasis risk factors[3]. Distant metastases of BC commonly affect organs, such as the brain, lungs, liver and bone, where they exhibit organ-specific patterns, and hence, each site is often associated with distinct symptoms, prognosis and treatment[4]. In nearly 25% of patients with metastatic BC, the lungs are the first and sometimes the only site of metastasis[5]. Additionally, the lungs represent the second most frequent site of BC metastasis, with a 5-year overall survival rate of only 16.8%[6]. Due to the typically asymptomatic nature of lung metastasis in BC, many patients remain undiagnosed until the disease becomes incurable, thus underscoring the need for timely intervention and proper care[7]. Early detection of lung metastases and accurate prognostic evaluation are crucial for improving outcomes of BC patients in clinical practice, especially to enable better clinical management and potentially achieve long-term survival. However, in Asian populations, the clinicopathological characteristics and risk factors associated with breast cancer lung metastases (BCLM) remain underexplored. This highlights the urgent need for predictive models to identify patients at the highest risk of developing lung metastases, thereby enabling physicians to tailor treatments according to patient needs. In recent years, machine learning (ML) has emerged as a powerful tool for big data analysis, particularly for predicting the early stages of cancer[8–10]. ML enables the development of algorithms that can learn from data, predict outcomes and improve over time, thereby facilitating precise decision-making[11–13]. As such, its potential in exploring risk factors for disease progression and predict patient prognosis is significant[14]. Some predictive models, such as Extreme Gradient Boosting (XGBoost)[15], Least Absolute Shrinkage and Selection Operator (LASSO) [16] and Random Forest (RF) [17], have demonstrated superior generalization capabilities compared with traditional statistical models, especially since they excel at simulating and predicting complex relationships between variables and outcomes. However, despite these advances, few studies have explored the application of ML models for predicting the risk of lung metastases in BCLM. Interleukins (ILs), a family of low-molecular-weight cytokines secreted by immune active cells, exhibit both pro-inflammatory and anti-inflammatory properties[18–21]. ILs are involved in cancer-related inflammation, influencing tumor progression through anti-tumor immune responses or by promoting a tumor-supporting microenvironment[22]. Previous research has identified specific ILs as risk factors in breast cancer subtypes. For instance, IL-5, IL-7 and IL-16 were found to be associated with HER2-positive BC, while IL-10 levels correlated with HER2-negative cases[23]. However, the role of various cytokines in BCLM patients remains unclear, especially with regards to those associated with enhanced risks of developing lung metastases in BC. Therefore, incorporating cytokines into predictive models of lung metastasis in BC is essential. In this context, the current study aimed to identify cytokine-based risk factors for lung metastases in breast cancer and establish a predictive risk model that could guide personalized treatment strategies and improve outcomes for BC patients. 2. Methods and materials 2.1 Study design and selection of patients Approval for this study was obtained from the Ethics Committee of the Second Affiliated Hospital of Xuzhou Medical University (Ethics Approval Number: 120601). Using the inpatients’ electronic medical record system of the Second Affiliated Hospital of Xuzhou Medical University, the current research retrospectively analyzed BC patients who were admitted to the hospital between September 2018 and September 2023. The following inclusion criteria was then applied: 1) a histologically-confirmed diagnosis of BC as the only primary malignant tumor; 2) Sufficient information about survival time and follow-up. In addition, BC patients were excluded if: 1) they were male; 2) the time interval from diagnosis to follow-up was less than one year; 3) rheumatic diseases and infections were present; 4) results for cytokine testing were incomplete; 5) they were unmarried; 6) multiple primary tumors were present. Overall, 326 patients met the above criteria, with 70 of them also presenting lung metastases which were diagnosed using radiological scans, biopsy or surgical resection specimens of metastatic lesions. The selected patients were then randomly assigned to two groups: a training group with 228 patients (accounting for 70%) and a validation one consisting of 98 patients (accounting for 30%). The above process of patient selection is visually represented in Fig. 1 . 2.2 Data collection and processing The study collected comprehensive baseline demographic and clinicopathological data for each participant, including age at diagnosis, TNM staging (I–IV), and the number of extrathoracic metastatic sites prior to pulmonary involvement. Treatment history before lung metastasis was documented, encompassing radiotherapy, chemotherapy, endocrine therapy, targeted agents, and immunotherapy. Surgical approach (breast-conserving or radical resection), axillary lymph node involvement, maximal tumor dimension, and immunohistochemical profiles (ER, PR, HER2, Ki67) were recorded. Additional variables included histologic grade (I–III), molecular subtype classification (Luminal A/B, HER2-enriched, triple-negative), tumor laterality (unilateral/bilateral), and histopathological categorization (ductal, lobular, or other). Anthropometric (BMI), menopausal status, serum tumor markers, and novel biomarkers (adenosine kinase 1, high-sensitivity C-reactive protein) were analyzed alongside hematologic parameters, albumin-fibrinogen ratios, and a multiplex cytokine panel (IFN-α/γ, IL-1β/2/4/5/6/8/10/12p70/17A, TNF-α). 2.3 Endpoint of study This study’s primary endpoint was the occurrence of the first lung metastasis in BC patients. The follow-up deadline was defined as the time from the initial diagnosis of BC to the development of lung metastasis or the date of the last follow-up. 2.4 Machine learning Machine learning algorithms—including XGBoost, random forest (RF), and least absolute shrinkage and selection operator (LASSO)—were employed to systematically assess risk factors associated with lung metastasis in breast cancer (BC). LASSO, a regression-based method, facilitates feature selection and regularization by identifying the most predictive variables while minimizing overfitting[24, 25]. RF is an ensemble learning method that combines multiple predictions or classifications to improve overall accuracy of prediction. It is highly versatile, capable of handling both categorical and continuous data, while also demonstrating strong noise resistance which effectively prevents overfitting, a key consideration when analyzing complex datasets[26]. Finally, XGBoost is an ML algorithm based on the gradient boosting framework and the CART decision tree algorithm. It offers high efficiency, flexibility and portability, resulting in superior prediction accuracy[27]. 2.5 Nomogram model construction Patients were randomly allocated to training and validation cohorts. The training cohort data were used to develop predictive models (RF, XGBoost, and LASSO), while internal validation was performed using the validation cohort[28, 29]. Key risk factors for breast cancer lung metastasis (BCLM) were identified through Venn diagram analysis and incorporated into nomogram models predicting 5- and 10-year BCLM risk. Lung metastasis-free survival was assessed via Kaplan-Meier analysis, with between-group differences evaluated using log-rank tests. Model performance was evaluated based on discrimination (AUC, equivalent to the C-index), calibration (calibration plots and Hosmer-Lemeshow test), and clinical utility (decision curve analysis, DCA). 2.6 Statistical analysis Metric data with a normal distribution were presented as mean ± SD, while non-normally distributed ones were expressed as median (P25, P75). Additionally, categorical data were described as counts (percentages). Two-group comparisons were then performed using the Mann-Whitney U test and the independent sample t-test for non-parametric and parametric values, respectively. In the case of categorical variables, results were compared using the chi-square test. All statistical tests, performed using SPSS version 23.0 (SPSS Italy, Florence, Italy) and statistical software package R (version 4.0.0, R statistical calculation project), were two-tailed, with differences considered to be significant at P < 0.05. 3. Results 3.1 Patients’ baseline characteristics This study included 326 breast cancer patients who visited the Second Affiliated Hospital of Xuzhou Medical University between September 2018 and September 2023, and among these, 70 patients also presented lung metastases. The cohort’s median age was 52 years, and the majority had undergone modified radical surgery (85.28%). Postoperative histopathological analysis revealed that invasive ductal carcinoma was the predominant histological type (74.85%), with Luminal B being the most common molecular subtype (73.62%). Regarding tumor staging, 44.79% of patients were classified as T2 stage, and 34.97% had no lymph node metastases. Additionally, most patients (70.55%) did not experience metastases to other organs before developing lung metastases. In terms of treatment, chemotherapy (92.64%) was the most commonly administered one, followed by radiotherapy (64.42%), endocrine therapy (43.56%), targeted therapy (34.97%) and immunotherapy (3.07%). The non-BCLM and BCLM groups were also significantly different (P < 0.05) in terms of several parameters, including CA125, CA153, hsCRP, absolute monocyte count, TNF-α, IL-8, IL-6, IL-2, IL-1 β and IL-12p70. The clinical characteristics of the 326 BC patients and baseline comparisons between the BCLM and non-BCLM groups are summarized in Table 1 . Furthermore, demographic and clinicopathological characteristics did not differ significantly between patients of the training and validation groups (Table 2 ). Table 1 Baseline characteristics of patients with lung metastases from breast cancer (BCLM) Characteristic Overall, N = 326 1 No BCLM, N = 256 1 BCLM, N = 70 1 p-value 2 Age, years 52 (45, 58) 52 (45, 58) 50 (42, 58) 0.197 BMI,kg/m 2 22.97 (21.48, 24.61) 22.97 (21.48, 24.46) 22.74 (19.81, 25.01) 0.169 β2MG, ug/ml 1.80 (1.53, 2.24) 1.79 (1.52, 2.23) 1.93 (1.59, 2.44) 0.171 CEA2, ng/ml 2 (2, 4) 2 (2, 3) 3 (2, 7) 0.148 TSGF, U/ml 53 (46, 62) 53 (46, 61) 53 (45, 62) 0.981 SCC,ug/ml 0.58 (0.47, 0.78) 0.59 (0.48, 0.78) 0.56 (0.39, 0.78) 0.182 CA125, U/ml 14 (10, 22) 14 (10, 20) 22 (12, 41) < 0.001 CA153, U/ml 10 (8, 20) 10 (7, 19) 18 (10, 44) < 0.001 CA50, IU/ml 6 (4, 11) 6 (4, 10) 7 (4, 12) 0.078 SF,ng/ml 110 (58, 219) 104 (59, 191) 175 (58, 330) 0.053 hsCRP,mg/L 2 (1, 3) 2 (1, 3) 3 (1, 9) 0.006 Neutrophil count,10⁹/L 3.05 (2.34, 4.41) 3.08 (2.30, 4.39) 3.05 (2.44, 4.99) 0.777 Lymphocyte count,10⁹/L 1.36 (1.12, 1.72) 1.36 (1.13, 1.76) 1.41 (1.09, 1.63) 0.725 Hb,g/L 125 (114, 132) 126 (114, 132) 123 (113, 131) 0.357 PLT,10⁹/L 236 (179, 290) 239 (182, 292) 231 (176, 283) 0.454 Monocyte count,10⁹/L 0.38 (0.31, 0.52) 0.36 (0.31, 0.49) 0.46 (0.35, 0.55) 0.004 Albumin,g/L 43.6 (40.7, 46.5) 43.5 (40.7, 46.5) 44.9 (41.6, 46.6) 0.215 Fibrinogen,g/L 3.86 (3.27, 4.21) 3.81 (3.27, 4.20) 3.96 (3.57, 4.26) 0.155 IFNα, pg/ml 2.14 (1.46, 3.67) 2.30 (1.46, 3.94) 1.99 (1.46, 3.00) 0.337 IFNγ,pg/ml 2.8 (1.9, 4.8) 3.0 (2.0, 5.0) 2.4 (1.7, 3.9) 0.119 IL12p70, pg/ml 1.94 (1.14, 3.04) 1.98 (1.21, 3.27) 1.71 (0.95, 2.34) 0.019 IL17A, pg/ml 4 (2, 10) 4 (2, 11) 4 (2, 8) 0.844 IL1β, pg/ml 1.75 (1.07, 2.86) 1.92 (1.15, 3.21) 1.62 (0.74, 2.72) 0.042 IL2, pg/ml 1.75 (0.98, 3.17) 1.84 (1.03, 3.37) 1.38 (0.87, 2.26) 0.005 IL4, pg/ml 2.06 (1.36, 3.35) 2.08 (1.37, 3.40) 1.94 (1.25, 2.99) 0.281 IL5, pg/ml 1.11 (0.68, 1.54) 1.15 (0.74, 1.59) 1.05 (0.63, 1.31) 0.178 IL6, pg/ml 5 (3, 9) 5 (3, 8) 8 (4, 21) 0.001 IL8, pg/ml 9 (6, 14) 9 (6, 13) 12 (6, 22) 0.005 IL10, pg/ml 3.35 (2.07, 4.48) 3.53 (2.03, 4.63) 3.10 (2.15, 4.15) 0.452 TNFα, pg/ml 1.95 (1.23, 3.45) 2.04 (1.28, 3.56) 1.59 (1.17, 2.48) 0.010 Organ transfer 0.682 No, n (%) 230 (70.55%) 182 (71.09%) 48 (68.57%) Yes, n (%) 96 (29.45%) 74 (28.91%) 22 (31.43%) Endocrine therapy 0.004 No, n (%) 70(21.47%) 22(31.43%) 48(18.75%) Yes, n (%) 256 (78.53%) 48(68.57%) 208(81.25%) Targeted therapy 0.067 No, n (%) 212 (65.03%) 160 (62.50%) 52 (74.29%) Yes, n (%) 114 (34.97%) 96 (37.50%) 18 (25.71%) Immunotherapy 0.231 No, n (%) 316 (96.93%) 250 (97.66%) 66 (94.29%) Yes, n (%) 10 (3.07%) 6 (2.34%) 4 (5.71%) Radiotherapy 0.046 No, n (%) 116 (35.58%) 84 (32.81%) 32 (45.71%) Yes, n (%) 210 (64.42%) 172 (67.19%) 38 (54.29%) Chemotherapy 0.142 No, n (%) 24 (7.36%) 16 (6.25%) 8 (11.43%) Yes, n (%) 302 (92.64%) 240 (93.75%) 62 (88.57%) AJCC-T 0.010 T1, n (%) 90 (27.61%) 78 (30.47%) 12 (17.14%) T2, n (%) 146 (44.79%) 116 (45.31%) 30 (42.86%) T3, n (%) 24 (7.36%) 20 (7.81%) 4 (5.71%) T4, n (%) 24 (7.36%) 14 (5.47%) 10 (14.29%) Unknown, n (%) 42 (12.88%) 28 (10.94%) 14 (20.00%) AJCC-N 0.099 N0, n (%) 114 (34.97%) 96 (37.50%) 18 (25.71%) N1, n (%) 82 (25.15%) 68 (26.56%) 14 (20.00%) N2, n (%) 62 (19.02%) 44 (17.19%) 18 (25.71%) N3, n (%) 42 (12.88%) 30 (11.72%) 12 (17.14%) Unknown, n (%) 26 (7.98%) 18 (7.03%) 8 (11.43%) AJCC-M 0.682 M0, n (%) 318 (97.55%) 250 (97.66%) 68 (97.14%) M1, n (%) 8 (2.45%) 6 (2.34%) 2 (2.86%) Surgery 0.038 Modified radical mastectomy, n (%) 278 (85.28%) 212 (82.81%) 66 (94.29%) Breast conserving surgery, n (%) 32 (9.82%) 30 (11.72%) 2 (2.86%) No surgery, n (%) 16 (4.91%) 14 (5.47%) 2 (2.86%) Pathological grading 0.191 I, n (%) 14 (4.29%) 12 (4.69%) 2 (2.86%) II, n (%) 106 (32.52%) 90 (35.16%) 16 (22.86%) III, n (%) 60 (18.40%) 46 (17.97%) 14 (20.00%) IV, n (%) 4 (1.23%) 4 (1.56%) 0 (0.00%) Unknown, n (%) 142 (43.56%) 104 (40.63%) 38 (54.29%) ER+ 0.257 Yes, n (%) 192 (58.90%) 156 (60.94%) 36 (51.43%) No, n (%) 104 (31.90%) 76 (29.69%) 28 (40.00%) Unknow, n (%) 30 (9.20%) 24 (9.38%) 6 (8.57%) PR+ < 0.001 Yes, n (%) 176 (53.99%) 152 (59.38%) 24 (34.29%) No, n (%) 120 (36.81%) 80 (31.25%) 40 (57.14%) Unknow, n (%) 30 (9.20%) 24 (9.38%) 6 (8.57%) HER2+ 0.970 Yes, n (%) 212 (65.03%) 166 (64.84%) 46 (65.71%) No, n (%) 78 (23.93%) 62 (24.22%) 16 (22.86%) Unknow, n (%) 36 (11.04%) 28 (10.94%) 8 (11.43%) ki67>14% 0.001 Yes, n (%) 202 (61.96%) 160 (62.50%) 42 (60.00%) No, n (%) 54 (16.56%) 50 (19.53%) 4 (5.71%) Unknow, n (%) 70 (21.47%) 46 (17.97%) 24 (34.29%) Subtype 0.109 Luminal A, n (%) 16 (4.91%) 14 (5.47%) 2 (2.86%) Luminal B, n (%) 240 (73.62%) 194 (75.78%) 46 (65.71%) Triple-negative, n (%) 42 (12.88%) 30 (11.72%) 12 (17.14%) HER2, n (%) 28 (8.59%) 18 (7.03%) 10 (14.29%) Laterality 0.338 Left, n (%) 184 (56.44%) 146 (57.03%) 38 (54.29%) Right, n (%) 138 (42.33%) 108 (42.19%) 30 (42.86%) Bilateral, n (%) 4 (1.23%) 2 (0.78%) 2 (2.86%) Pathological type 0.121 Ductal carcinoma, n (%) 244 (74.85%) 188 (73.44%) 56 (80.00%) Lobular carcinoma, n (%) 6 (1.84%) 6 (2.34%) 0 (0.00%) Other types, n (%) 36 (11.04%) 26 (10.16%) 10 (14.29%) Unknown, n (%) 40 (12.27%) 36 (14.06%) 4 (5.71%) Menopausal 0.801 No, n (%) 154 (47.24%) 120 (46.88%) 34 (48.57%) Yes, n (%) 172 (52.76%) 136 (53.13%) 36 (51.43%) 1 Median (IQR); n (%) 2 Wilcoxon rank sum test; Pearson's Chi-squared test; Fisher's exact test Note:BMI = Body Mass Index,SF = Ferritin, hs-CRP = Hypersensitive C-reactive protein,Hb = Haemoglobin,PLT = Platelet,ER = Estrogen receptors,PR = Progesterone receptors,Her2 = Human epidermal growth factor receptor-2 Table 2 Comparison of baseline features between training group and validation group Characteristic Overall, N = 3261 Training group, N = 2281 Validation groups, N = 981 p-value 2 Age, years 52 (45, 58) 52 (44, 57) 52 (46, 59) 0.685 BMI,kg/m 2 22.97 (21.48, 24.61) 22.89 (21.48, 24.61) 23.44 (21.48, 24.61) 0.973 β2MG, ug/ml 1.80 (1.53, 2.24) 1.82 (1.50, 2.23) 1.78 (1.54, 2.29) 0.717 CEA2, ng/ml 2 (2, 4) 2 (2, 4) 2 (2, 3) 0.631 TSGF, U/ml 53 (46, 62) 52 (46, 61) 53 (47, 62) 0.217 SCC,ug/ml 0.58 (0.47, 0.78) 0.59 (0.48, 0.79) 0.56 (0.45, 0.73) 0.146 CA125, U/ml 14 (10, 22) 15 (10, 23) 14 (10, 22) 0.822 CA153, U/ml 10 (8, 20) 11 (8, 20) 10 (8, 19) 0.934 CA50, IU/ml 6 (4, 11) 6 (4, 11) 7 (5, 10) 0.904 SF,ng/ml 110 (58, 219) 108 (58, 220) 111 (60, 213) 0.935 hsCRP,mg/L 2 (1, 3) 2 (1, 3) 1 (1, 3) 0.132 Neutrophil count,10⁹/L 3.05 (2.34, 4.41) 3.05 (2.42, 4.39) 3.17 (2.30, 4.45) 0.411 Lymphocyte count,10⁹/L 1.36 (1.12, 1.72) 1.40 (1.12, 1.74) 1.35 (1.12, 1.68) 0.622 Hb,g/L 125 (114, 132) 125 (115, 132) 125 (111, 132) 0.405 PLT,10⁹/L 236 (179, 290) 233 (180, 284) 242 (181, 295) 0.363 Monocyte count,10⁹/L 0.38 (0.31, 0.52) 0.38 (0.31, 0.50) 0.40 (0.30, 0.54) 0.904 Albumin,g/L 43.6 (40.7, 46.5) 43.9 (40.9, 46.5) 43.0 (40.4, 46.1) 0.301 Fibrinogen,g/L 3.86 (3.27, 4.21) 3.88 (3.27, 4.21) 3.82 (3.33, 4.21) 0.869 IFNα, pg/ml 2.14 (1.46, 3.67) 2.15 (1.40, 3.64) 2.14 (1.49, 4.14) 0.866 IFNγ,pg/ml 2.8 (1.9, 4.8) 2.8 (1.9, 4.9) 2.7 (1.9, 4.0) 0.507 IL12p70, pg/ml 1.94 (1.14, 3.04) 1.96 (1.13, 3.05) 1.93 (1.22, 3.01) 0.809 IL17A, pg/ml 4 (2, 10) 4 (2, 11) 3 (2, 8) 0.059 IL1β, pg/ml 1.75 (1.07, 2.86) 1.81 (1.16, 2.86) 1.64 (0.95, 2.81) 0.121 IL2, pg/ml 1.75 (0.98, 3.17) 1.76 (1.01, 2.92) 1.70 (0.96, 3.17) 0.713 IL4, pg/ml 2.06 (1.36, 3.35) 2.04 (1.39, 3.33) 2.09 (1.28, 3.35) 0.717 IL5, pg/ml 1.11 (0.68, 1.54) 1.10 (0.67, 1.55) 1.16 (0.75, 1.40) 0.437 IL6, pg/ml 5 (3, 9) 5 (3, 9) 5 (3, 9) 0.506 IL8, pg/ml 9 (6, 14) 9 (6, 14) 8 (4, 16) 0.118 IL10, pg/ml 3.35 (2.07, 4.48) 3.46 (2.23, 4.68) 3.05 (1.83, 4.28) 0.130 TNFα, pg/ml 1.95 (1.23, 3.45) 1.95 (1.26, 3.52) 1.92 (1.17, 3.22) 0.739 Organ transfer 0.015 No, n (%) 230 (70.55%) 170 (74.56%) 60 (61.22%) Yes, n (%) 96 (29.45%) 58 (25.44%) 38 (38.78%) Endocrine therapy 0.867 No, n (%) 70 (21.47%) 51 (22.37%) 19 (19.39%) Yes, n (%) 256 (78.53%) 177 (77.63%) 79 (80.61%) Targeted therapy 0.014 No, n (%) 212 (65.03%) 158 (69.30%) 54 (55.10%) Yes, n (%) 114 (34.97%) 70 (30.70%) 44 (44.90%) Immunotherapy 0.292 No, n (%) 316 (96.93%) 219 (96.05%) 97 (98.98%) Yes, n (%) 10 (3.07%) 9 (3.95%) 1 (1.02%) Radiotherapy 0.826 No, n (%) 116 (35.58%) 82 (35.96%) 34 (34.69%) Yes, n (%) 210 (64.42%) 146 (64.04%) 64 (65.31%) Chemotherapy 0.716 No, n (%) 24 (7.36%) 16 (7.02%) 8 (8.16%) Yes, n (%) 302 (92.64%) 212 (92.98%) 90 (91.84%) AJCC-T 0.608 T1, n (%) 90 (27.61%) 63 (27.63%) 27 (27.55%) T2, n (%) 146 (44.79%) 107 (46.93%) 39 (39.80%) T3, n (%) 24 (7.36%) 15 (6.58%) 9 (9.18%) T4, n (%) 42 (12.88%) 26 (11.40%) 16 (16.33%) Unknown, n (%) 24 (7.36%) 17 (7.46%) 7 (7.14%) AJCC-N 0.128 N0, n (%) 114 (34.97%) 82 (35.96%) 32 (32.65%) N1, n (%) 82 (25.15%) 53 (23.25%) 29 (29.59%) N2, n (%) 62 (19.02%) 45 (19.74%) 17 (17.35%) N3, n (%) 42 (12.88%) 34 (14.91%) 8 (8.16%) Unknown, n (%) 26 (7.98%) 14 (6.14%) 12 (12.24%) AJCC-M > 0.999 M0, n (%) 318 (97.55%) 222 (97.37%) 96 (97.96%) M1, n (%) 8 (2.45%) 6 (2.63%) 2 (2.04%) Surgery 0.306 Modified radical mastectomy, n (%) 278 (85.28%) 194 (85.09%) 84 (85.71%) Breast conserving surgery, n (%) 32 (9.82%) 25 (10.96%) 7 (7.14%) No surgery, n (%) 16 (4.91%) 9 (3.95%) 7 (7.14%) Pathological grading 0.417 I, n (%) 14 (4.29%) 9 (3.95%) 5 (5.10%) II, n (%) 106 (32.52%) 76 (33.33%) 30 (30.61%) III, n (%) 60 (18.40%) 47 (20.61%) 13 (13.27%) IV, n (%) 4 (1.23%) 3 (1.32%) 1 (1.02%) Unknown, n (%) 142 (43.56%) 93 (40.79%) 49 (50.00%) ER+ 0.022 Yes, n (%) 176 (53.99%) 134 (58.77%) 42 (42.86%) No, n (%) 120 (36.81%) 77 (33.77%) 43 (43.88%) Unknow, n (%) 30 (9.20%) 17 (7.46%) 13 (13.27%) PR+ 0.013 Yes, n (%) 192 (58.90%) 146 (64.04%) 46 (46.94%) No, n (%) 104 (31.90%) 65 (28.51%) 39 (39.80%) Unknow, n (%) 30 (9.20%) 17 (7.46%) 13 (13.27%) HER2+ 0.043 Yes, n (%) 212 (65.03%) 143 (62.72%) 69 (70.41%) No, n (%) 78 (23.93%) 63 (27.63%) 15 (15.31%) Unknow, n (%) 36 (11.04%) 22 (9.65%) 14 (14.29%) ki67>14% 0.264 Yes, n (%) 202 (61.96%) 143 (62.72%) 59 (60.20%) No, n (%) 54 (16.56%) 41 (17.98%) 13 (13.27%) Unknow, n (%) 70 (21.47%) 44 (19.30%) 26 (26.53%) Subtype 0.158 Luminal A, n (%) 16 (4.91%) 13 (5.70%) 3 (3.06%) Luminal B, n (%) 240 (73.62%) 164 (71.93%) 76 (77.55%) Triple-negative, n (%) 42 (12.88%) 27 (11.84%) 15 (15.31%) HER2, n (%) 28 (8.59%) 24 (10.53%) 4 (4.08%) Laterality 0.252 Left, n (%) 184 (56.44%) 135 (59.21%) 49 (50.00%) Right, n (%) 138 (42.33%) 90 (39.47%) 48 (48.98%) Bilateral, n (%) 4 (1.23%) 3 (1.32%) 1 (1.02%) Pathological type 0.655 Ductal carcinoma, n (%) 244 (74.85%) 170 (74.56%) 74 (75.51%) Lobular carcinoma, n (%) 6 (1.84%) 4 (1.75%) 2 (2.04%) Other types, n (%) 36 (11.04%) 28 (12.28%) 8 (8.16%) Unknown, n (%) 40 (12.27%) 26 (11.40%) 14 (14.29%) Menopausal 0.299 No, n (%) 154 (47.24%) 112 (49.12%) 42 (42.86%) Yes, n (%) 172 (52.76%) 116 (50.88%) 56 (57.14%) Lung Metastasis 0.234 No BCLM, n (%) 256 (78.53%) 175 (76.75%) 81 (82.65%) BCLM, n (%) 70 (21.47%) 53 (23.25%) 17 (17.35%) 1 Median (IQR); n (%) 2Wilcoxon rank sum test; Pearson's Chi-squared test; Fisher's exact test Note:BMI = Body Mass Index,SF = Ferritin, hs-CRP = Hypersensitive C-reactive protein,Hb = Haemoglobin,PLT = Platelet,ER = Estrogen receptors,PR = Progesterone receptors,Her2 = Human epidermal growth factor receptor-2. 3.2 Identification of BCLM risk factors To identify the risk factors for BCLM, the LASSO algorithm was employed, with Supplementary Figure S1-A showing the binomial deviation curve plotted against the logarithm of the tuning hyperparameter (λ). In this case, the solid vertical line indicates the binomial deviation ± standard error (SE), while the optimal λ value was determined using the minimum standard and 1-SE standard through 10-fold cross validation. Furthermore, a coefficient profile was generated from the log (λ) sequence, with 49 clinical parameters integrated into the LASSO model to enable effective penalization of non-essential features. Following model training and the 10-fold cross validation, 12 non-zero coefficients were identified as being significantly associated with lung metastasis (Supplementary S1B). According to the Lasso model’s feature importance ranking (Fig. 2 A), the relative importance of predictors from highest to lowest was as follows: other organ metastasis, endocrine therapy, PR status, absolute lymphocyte count, targeted therapy, IL-2, INF-α, TNF-α, CEA, CA125, hsCRP, and IL-6. 3.3 Identification of risk factors for BCLM The RF machine learning algorithm was used to further refine the selection of risk factors. This algorithm works by randomly extracting subsets of features from the training data, with each subset subsequently utilized to construct a decision tree. For each node within these decision trees, the optimal feature was chosen from a random subset of features for node partitioning. The decision tree was then recursively built based on the selected features until a predefined stopping condition was met. For classification problems, the final class was determined through a majority voting mechanism, while in the case of regression problems, the average of the predicted values from all trees served as the final prediction. The algorithm eventually combined the outputs from all constructed decision trees to calculate the average error rates separately for node-positive and node-negative groups. The importance of clinical features was subsequently assessed before visualizing their rankings (Fig. 2 B). Overall, 24 clinically relevant features were identified as risk factors for BCLM, and they included IL-6, PR, BMI, SCC, FIB, ER, endocrine therapy, IL-8, ALB, IL-17A, Hb, IFN-γ, hsCRP, TNF-α, IL-12p70, CA125, TSGF, PLT, INF-α, N, SF, B2M, CA50 and IL10. The XGBoost model, based on a gradient boosting framework, is another ensemble method that uses decision trees to enhance predictive accuracy. Gradient boosting is a specific implementation of the Boosting technique which iteratively minimize the objective function by fitting each new tree with the negative gradient of the previous round’s error. In this study, the XGBoost model identified 15 non-zero coefficients that were significantly correlated with lung metastasis. These features were then ranked by relative importance (Fig. 2 C) as follows: endocrine therapy, IL-6, hsCRP, IL-17A, PLT, IL-8, SCC, SF, INF-α, CEA, ALB, Hb, IL2, M, TNF-α. The RF, LASSO and XGBoost algorithms were used to independently identify BCLM-related risk factors, with overlapping variables among the three ML models subsequently selected as significant ones. The intersection of these factors was visualized using a Venn diagram (Fig. 2 D), and the results highlighted five key variables for subsequent nomogram analysis: endocrine therapy, hsCRP, IL6, IFN-α and TNF-α. To enhance clinical applicability, the Maxstat method was then used to assess the optimal risk cut-off points for the five variables (Supplementary Figure S2). Using these cut-off values (hsCRP (16.8), IL6 (16.19), IFN-ɑ (2.36) and TNF-ɑ (1.35)), the biomarkers were reclassified into high- and low-risk groups prior to analysis using Kaplan-Meier curves (Supplementary Figure S3) to determine survival outcomes. Additionally, Supplementary Figure S4 shows the lung metastasis rates across different molecular subtypes. Compared with Luminal A patients, those with the Luminal B subtype demonstrated a lower risk of lung metastasis (unadjusted HR: 1.787; P = 0.435), while HER2+ (unadjusted HR: 3.571; P = 0.094) exhibited a higher risk. In particular, patients with TNBC faced the highest risk of lung metastasis (unadjusted HR: 6.487; P = 0.018). 3.4 Establishment and validation of BCLM diagnostic nomogram A nomogram model was constructed based on five key variables: endocrine therapy, hsCRP, IL6, IFN-ɑ, and TNF-ɑ. Each variable was assigned a point value ranging from 0 to 100 (Fig. 3 ). The cumulative score, obtained by summing these points, allowed estimation of the 5- and 10-year lung metastasis probability in breast cancer (BC) patients, aiding clinical decision-making. Risk prediction involved drawing a vertical line from the total score to the probability axis (ranging from 0.1 to 0.95), though not all probabilities aligned precisely with marked values. The model's performance was evaluated based on discrimination, calibration, and clinical utility in both training and validation cohorts, with results visualized using ROC curves, calibration plots, and decision curve analysis. In the training set, the AUC values for 5- and 10-year metastasis prediction were 0.786 (95% CI: 0.691–0.881) and 0.787 (95% CI: 0.749–0.824), respectively. The validation set yielded AUCs of 0.627 (95% CI: 0.441–0.813) for 5-year and 0.797 (95% CI: 0.605–0.988) for 10-year prediction. These findings indicate robust predictive accuracy across both datasets (Fig. 4 A, B). 3.5 Calibration curve and DCA analysis Calibration curves were generated for evaluating the nomogram’s performance. Following internal validation with 1000 bootstrap iterations, the calibration curves for both the training and validation sets (Fig. 5 A, B) closely aligned with the diagonal line, indicating that the predicted and actual probabilities of lung metastasis were in strong agreement. The nomogram’s clinical utility was assessed using DCA (Fig. 6 A, B), in which the horizontal line represented the assumption of no lung metastasis where the net benefit was zero and the diagonal line represented the scenario where all patients were assumed to have BCLM. Overall, the decision curves demonstrated that the range of high threshold probabilities was broad and applicable to both the training and validation sets. Compared with individual variables, the nomogram exhibited a higher net benefit for both datasets, thus underscoring its superior predictive ability. This indicates that the nomogram can effectively predict the 5-year and 10-year risk of lung metastasis in BC patients. 4. Discussion In this study, multiple ML algorithms were applied to determine the risk factors for BCLM, with the following five significant predictors subsequently identified: endocrine therapy, hsCRP, IL6, IFN-ɑ, and TNF-ɑ. These variables were then integrated into a nomogram model. The findings provided a framework for identifying BC patients who were at a higher risk of lung metastasis, thereby improving prognostic evaluation and clinical management while offering new insights for developing more effective treatments. Additionally, this study was also the first one to construct a lung metastasis prediction model for BC patients based on cytokines. The model demonstrated high accuracy in predicting survival outcomes for BCLM patients, and in practice, the nomogram, which integrated predictions from RF, LASSO and XGBoost algorithms, exhibited robust performance across both the training and validation groups. A Mendelian randomization analysis involving 420,964 cancer-free patients from the UK Biobank cohort showed that elevated serum c-reactive protein (CRP) levels were linked to higher risks of breast cancer, colorectal cancer, head and neck as well as other malignancies over a 7.1-year follow-up period [30]. Similarly, a meta-analysis examining 119 inflammatory markers (with CRP as the primary focus) across 26 cancer types reached comparable conclusions [31]. These pan-cancer studies identified individuals with CRP levels above 3 mg/L as having a high risk of inflammation but this threshold may not apply specifically to BC[30, 32]. The findings further corroborated the link between elevated hypersensitive CRP and an increased risk of BCLM. This suggests that inflammation in BC patients may contribute to tumor proliferation and metastasis, including lung metastasis. The above analyses also determined the optimal hsCRP cut-off value for predicting BCLM to be 16.8 mg/L, hence providing a potential reference point for individualized breast cancer treatment. Interestingly, CRP was consistently identified as a key risk factor for BCLM across all three ML models used in this study. While prior meta-analyses have highlighted the limited predictive value of CRP in non-metastatic BC, the association between elevated CRP levels and poor prognosis is well documented in metastatic cases [33, 34]. For instance, in vitro studies revealed that CRP could promote the adhesion of MCF10A human breast epithelial cells through activation of the integrin α 2 signaling pathway and Fcγ receptor I (FcγRI), with the process subsequently activating paxillin, FAK and ERKs to drive autocrine effects [35]. Furthermore, using an invasion model of MDA-MB-231 TNBC cells and mouse tumor models, CRP was shown to be involved in tumor growth. Additional animal experiments further demonstrated that CRP impaired immune surveillance by inhibiting the activation of pulmonary macrophages, induced by symbiotic bacteria through an FcγR2B dependent mechanism, thereby fostering the formation of pre-metastatic niches in the lungs of tumor-bearing mice [36]. Altogether, these findings highlight the significant role of CRP in lung metastasis, thus supporting this study’s results. This study underscores the potential of endocrine therapy to reduce the risk of BCLM, with this lower risk being particularly evident among hormone receptor-positive patients who constituted over half of the total study population. Of these patients, 80% received endocrine therapy, including options such as tamoxifen and steroidal (exemestane) or nonsteroidal (letrozole or anastrozole) aromatase inhibitors. Tamoxifen is known to improve disease-free and overall survival in postmenopausal women with ER-positive tumors[37, 38]. However, the BIG trial demonstrated that first-line treatment with aromatase inhibitors lowered the absolute risk of 10-year recurrence by 3.6%, increased overall survival by 2.1% and outperformed tamoxifen monotherapy [39]. Furthermore, post hoc analyses of the SOFT and TEXT trials revealed that combining ovarian suppression with tamoxifen significantly improved 8-year disease-free and overall survival rates in comparison with tamoxifen alone [40]. Despite the success of endocrine therapy in reducing BC recurrence and mortality, both intrinsic and acquired drug resistance remain a challenge. In this context, recent advances in understanding the drivers and mechanisms underlying endocrine therapy resistance in estrogen receptor-positive BC has led to the development of targeted drugs, such as mTOR inhibitors and cyclin dependent kinase 4/6 inhibitors can markedly extend progression-free survival [41, 42]. When lung metastasis rates were further analyzed by molecular subtypes, it was found that the risk of lung metastasis was significantly lower in hormone receptor-positive patients compared with HER2 + ones, with TNBC patients exhibiting the highest risk. These results underscore the importance of endocrine therapy in mitigating the risk of BCLM. The TME comprises both cellular elements, including adipocytes, immune cells[43, 44], endothelial cells and cancer-associated fibroblasts, as well as non-cellular components[45–48], such as cytokines and the ECM. It promotes tumor progression and invasion through the secretion of growth factors and pro-inflammatory mediators as well as through intercellular interactions and metabolic crosstalk with tumor cells[49, 50]. In this study, cytokines were innovatively incorporated into a lung metastasis model, and three key inflammatory factors (IL6, IFN-ɑ and TNF-ɑ) associated with lung metastasis were then identified using RF, LASSO and XGBoost ML algorithms. These cytokines have been extensively studied in the context of BC metastasis mechanisms. For instance, early research has demonstrated that IL-6-174 promoter polymorphism was linked to clinical outcomes in a group of lymph node-positive BC patients undergoing high-dose adjuvant therapy [51]. Additionally, Adam et al. reported that fibroblasts isolated from common sites of breast cancer metastasis enhanced the growth and invasiveness of cancer cells in an IL-6-dependent manner [52]. Similarly, Laura et al. found that p53 inactivation triggered a methylation-dependent autocrine IL-6 loop that led to epigenetic reprogramming and the development of basal/stem cell-like gene expression profiles in BC cells [53]. HER2 overexpression has also been shown to induce IL-6 secretion, activate STAT3, alter gene expression and reinforce the autocrine IL-6/STAT3 loop [54]. In one study, Luca et al. reported that the combination of VEGF and IL-6 synergistically and durably activated intracellular signaling pathways, such as MAPK, AKT and p38MAPK, in BC cells [55], while Rasmus et al. demonstrated that, in ER + breast cancer, the IL6/STAT3 signaling pathway could drive metastasis independently of the estrogen receptor. Although STAT3 and ER share enhancers, the former can hijack a subset of ER enhancers to induce unique transcriptional programs. This decoupling of ER and IL6/STAT3 oncogenic pathways underscores the therapeutic potential of targeting IL6/STAT3 in ER + breast cancer [56]. In contrast to IL-6, IFN-α and TNF-α inhibit breast cancer growth and invasion through distinct mechanisms. Specifically, IFN-α is involved in tumor immune surveillance by activating CD8 α + dendritic cells (DCs) and enhancing CD8 + T cell recognition of tumor antigens[57]. Thus, a deficiency in IFN-α can disrupt this process, leading to the expansion of regulatory T cells (Tregs) which suppresses plasma cell-like DCs and facilitate BC metastasis [58]. On the other hand, TNF-α can restrict the migration of triple negative, mesenchymal-like BC cells with high TNFR1 expression, while inhibiting the migration of epithelioid cells with low TNFR1 expression [59]. 5. Conclusions This study applied three ML methods to systematically analyze clinical information and surgical pathology results, integrated treatment exposure and inflammatory markers, and established a predictive model for BCLM. This model exhibits strong discriminative ability in both training and validation queues. In fact, through this nomogram, doctors can estimate the likelihood of lung metastasis in BC patients based on the cumulative score of each risk factor. Therefore, this tool can achieve personalized risk assessment by regularly reviewing inflammation indicators for high-risk patients, and immediately initiating imaging screening for patients with improved scores. In addition, the findings highlighted the contrasting roles of cytokines in BC, with IL-6 promoting BCLM, while IFN-α and TNF-α inhibited tumor metastasis. These insights deepen current understanding of the interplay between cytokines and BCLM, thus underscoring the importance of detecting and managing inflammation associated with BC. Future works should validate the current findings through large, prospective, multi-center trials. Abbreviations BMI Body Mass Index SF Ferritin hs-CRP Hypersensitive C-reactive protein Hb Haemoglobin, PLT Platelet ER Estrogen receptors PR Progesterone receptors Her2 Human epidermal growth factor receptor-2 Declarations Ethics approval and consent to participate This study was approved by the Human Research Ethics Committee, the Second Affiliated Hospital of Xuzhou Medical University (120601). The Ethics Committee of the Second Affiliated Hospital of Xuzhou Medical University agreed to waive informed consent because this retrospective observational study uses medical recordsfrom previous clinical diagnosis, the risk of which was not greater than the minimum risk.And the research process was in accordance with the content of the Declaration of Helsinki. Consent for publication Not applicable. Availability of data and materials The datasets generated and analysed during the current study are not publicly available due to privacy or ethical restrictions but are available from the corresponding author on reasonable request. Competing Interests The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Funding This work was supported by the Project of Jiangsu Health Commission(M2020072). Authors' contributions ZL conceived of the study. ZL and HM designed the study and collected data. WB contributed to data analysis. ZL, HM, LZ contributed to the writing and revision of the paper.All authors read and approved the fnal manuscript. Acknowledgements Not applicable. References Q L, C X, H L, X Y, F Y, M C, S Z, Y T, S H, M C, W C: - Disparities in 36 cancers across 185 countries: secondary analysis of global. - Front Med 2024 Oct;18(5):911-920 doi: 101007/s11684-024-1058-6 Epub 2024 Aug : - 911-920. Park M, Kim D, Ko S, Kim A, Mo K, Yoon H: Breast Cancer Metastasis: Mechanisms and Therapeutic Implications. Int J Mol Sci 2022, 23 . Weigelt B, Peterse JL, van 't Veer LJ: Breast cancer metastasis: markers and models. Nat Rev Cancer 2005, 5: 591-602. Wang C, Xu K, Wang R, Han X, Tang J, Guan X: Heterogeneity of BCSCs contributes to the metastatic organotropism of breast cancer. J Exp Clin Cancer Res 2021, 40: 370. Pillar N, Polsky AL, Weissglas-Volkov D, Shomron N: Comparison of breast cancer metastasis models reveals a possible mechanism of tumor aggressiveness. Cell Death Dis 2018, 9: 1040. Liang Y, Zhang H, Song X, Yang Q: Metastatic heterogeneity of breast cancer: Molecular mechanism and potential therapeutic targets. Semin Cancer Biol 2020, 60: 14-27. Lin S, Mo H, Li Y, Guan X, Chen Y, Wang Z, Xu B: Clinicopathological characteristics and survival outcomes in patients with synchronous lung metastases upon initial metastatic breast cancer diagnosis in Han population. BMC Cancer 2021, 21: 1330. Jin W, Yang Q, Chi H, Wei K, Zhang P, Zhao G, Chen S, Xia Z, Li X: Ensemble deep learning enhanced with self-attention for predicting immunotherapeutic responses to cancers. Front Immunol 2022, 13: 1025330. Ren Q, Zhang P, Lin H, Feng Y, Chi H, Zhang X, Xia Z, Cai H, Yu Y: A novel signature predicts prognosis and immunotherapy in lung adenocarcinoma based on cancer-associated fibroblasts. Front Immunol 2023, 14: 1201573. Zhang S, Jiang C, Jiang L, Chen H, Huang J, Gao X, Xia Z, Tran LJ, Zhang J, Chi H, et al: Construction of a diagnostic model for hepatitis B-related hepatocellular carcinoma using machine learning and artificial neural networks and revealing the correlation by immunoassay. Tumour Virus Res 2023, 16: 200271. Zhao S, Zhang X, Gao F, Chi H, Zhang J, Xia Z, Cheng C, Liu J: Identification of copper metabolism-related subtypes and establishment of the prognostic model in ovarian cancer. Front Endocrinol (Lausanne) 2023, 14: 1145797. Chi H, Yang J, Peng G, Zhang J, Song G, Xie X, Xia Z, Liu J, Tian G: Circadian rhythm-related genes index: A predictor for HNSCC prognosis, immunotherapy efficacy, and chemosensitivity. Front Immunol 2023, 14: 1091218. Zhao S, Chi H, Yang Q, Chen S, Wu C, Lai G, Xu K, Su K, Luo H, Peng G, et al: Identification and validation of neurotrophic factor-related gene signatures in glioblastoma and Parkinson's disease. Front Immunol 2023, 14: 1090040. Chi H, Jiang P, Xu K, Zhao Y, Song B, Peng G, He B, Liu X, Xia Z, Tian G: A novel anoikis-related gene signature predicts prognosis in patients with head and neck squamous cell carcinoma and reveals immune infiltration. Front Genet 2022, 13: 984273. Kang J, Choi YJ, Kim IK, Lee HS, Kim H, Baik SH, Kim NK, Lee KY: LASSO-Based Machine Learning Algorithm for Prediction of Lymph Node Metastasis in T1 Colorectal Cancer. Cancer Res Treat 2021, 53: 773-783. Chi H, Xie X, Yan Y, Peng G, Strohmer DF, Lai G, Zhao S, Xia Z, Tian G: Natural killer cell-related prognosis signature characterizes immune landscape and predicts prognosis of HNSCC. Front Immunol 2022, 13: 1018685. Li J, Shi Z, Liu F, Fang X, Cao K, Meng Y, Zhang H, Yu J, Feng X, Li Q, et al: XGBoost Classifier Based on Computed Tomography Radiomics for Prediction of Tumor-Infiltrating CD8(+) T-Cells in Patients With Pancreatic Ductal Adenocarcinoma. Front Oncol 2021, 11: 671333. Zhai X, Zhang H, Xia Z, Liu M, Du G, Jiang Z, Zhou H, Luo D, Dou D, Li J, et al: Oxytocin alleviates liver fibrosis via hepatic macrophages. JHEP Rep 2024, 6: 101032. Xiao J, Lin H, Liu B, Xia Z, Zhang J, Jin J: Decreased S1P and SPHK2 are involved in pancreatic acinar cell injury. Biomark Med 2019, 13: 627-637. Xiao J, Huang K, Lin H, Xia Z, Zhang J, Li D, Jin J: Mogroside II(E) Inhibits Digestive Enzymes via Suppression of Interleukin 9/Interleukin 9 Receptor Signalling in Acute Pancreatitis. Front Pharmacol 2020, 11: 859. Zhang H, Xia T, Xia Z, Zhou H, Li Z, Wang W, Zhai X, Jin B: KIF18A inactivates hepatic stellate cells and alleviates liver fibrosis through the TTC3/Akt/mTOR pathway. Cell Mol Life Sci 2024, 81: 96. Habanjar O, Bingula R, Decombat C, Diab-Assaf M, Caldefie-Chezet F, Delort L: Crosstalk of Inflammatory Cytokines within the Breast Tumor Microenvironment. Int J Mol Sci 2023, 24 . Zhou H, Cai Z, Yang Q, Yang X, Chen J, Huang T: Inflammatory cytokines and two subtypes of breast cancer: A two-sample mendelian randomization study. PLoS One 2023, 18: e0293230. Zhang X, Zhuge J, Liu J, Xia Z, Wang H, Gao Q, Jiang H, Qu Y, Fan L, Ma J, et al: Prognostic signatures of sphingolipids: Understanding the immune landscape and predictive role in immunotherapy response and outcomes of hepatocellular carcinoma. Front Immunol 2023, 14: 1153423. Wang X, Zhao Y, Strohmer DF, Yang W, Xia Z, Yu C: The prognostic value of MicroRNAs associated with fatty acid metabolism in head and neck squamous cell carcinoma. Front Genet 2022, 13: 983672. Yu Y, He Z, Ouyang J, Tan Y, Chen Y, Gu Y, Mao L, Ren W, Wang J, Lin L, et al: Magnetic resonance imaging radiomics predicts preoperative axillary lymph node metastasis to support surgical decisions and is associated with tumor microenvironment in invasive breast cancer: A machine learning, multicenter study. EBioMedicine 2021, 69: 103460. Zhang H, Lin F, Zheng T, Gao J, Wang Z, Zhang K, Zhang X, Xu C, Zhao F, Xie H, et al: Artificial intelligence-based classification of breast lesion from contrast enhanced mammography: a multicenter study. Int J Surg 2024, 110: 2593-2603. Liu J, Zhang P, Yang F, Jiang K, Sun S, Xia Z, Yao G, Tang J: Integrating single-cell analysis and machine learning to create glycosylation-based gene signature for prognostic prediction of uveal melanoma. Front Endocrinol (Lausanne) 2023, 14: 1163046. Chi H, Gao X, Xia Z, Yu W, Yin X, Pan Y, Peng G, Mao X, Teichmann AT, Zhang J, et al: FAM family gene prediction model reveals heterogeneity, stemness and immune microenvironment of UCEC. Front Mol Biosci 2023, 10: 1200335. Zhu M, Ma Z, Zhang X, Hang D, Yin R, Feng J, Xu L, Shen H: C-reactive protein and cancer risk: a pan-cancer study of prospective cohort and Mendelian randomization analysis. BMC Med 2022, 20: 301. Michels N, van Aart C, Morisse J, Mullee A, Huybrechts I: Chronic inflammation towards cancer incidence: A systematic review and meta-analysis of epidemiological studies. Crit Rev Oncol Hematol 2021, 157: 103177. Siemes C, Visser LE, Coebergh JW, Splinter TA, Witteman JC, Uitterlinden AG, Hofman A, Pols HA, Stricker BH: C-reactive protein levels, variation in the C-reactive protein gene, and cancer risk: the Rotterdam Study. J Clin Oncol 2006, 24: 5216-5222. Han Y, Mao F, Wu Y, Fu X, Zhu X, Zhou S, Zhang W, Sun Q, Zhao Y: Prognostic role of C-reactive protein in breast cancer: a systematic review and meta-analysis. Int J Biol Markers 2011, 26: 209-215. Mikkelsen MK, Lindblom NAF, Dyhl-Polk A, Juhl CB, Johansen JS, Nielsen D: Systematic review and meta-analysis of C-reactive protein as a biomarker in breast cancer. Crit Rev Clin Lab Sci 2022, 59: 480-500. Kim ES, Kim SY, Koh M, Lee HM, Kim K, Jung J, Kim HS, Moon WK, Hwang S, Moon A: C-reactive protein binds to integrin α2 and Fcγ receptor I, leading to breast cell adhesion and breast cancer progression. Oncogene 2018, 37: 28-38. Feng JR, Li X, Han C, Chang Y, Fu Y, Feng GC, Lei Y, Li HY, Tang PM, Ji SR, et al: C-Reactive Protein Induces Immunosuppression by Activating FcγR2B in Pulmonary Macrophages to Promote Lung Metastasis. Cancer Res 2024, 84: 4184-4198. Jaiyesimi IA, Buzdar AU, Decker DA, Hortobagyi GN: Use of tamoxifen for breast cancer: twenty-eight years later. J Clin Oncol 1995, 13: 513-529. Legha SS: Tamoxifen in the treatment of breast cancer. Ann Intern Med 1988, 109: 219-228. Ruhstaller T, Giobbie-Hurder A, Colleoni M, Jensen MB, Ejlertsen B, de Azambuja E, Neven P, Láng I, Jakobsen EH, Gladieff L, et al: Adjuvant Letrozole and Tamoxifen Alone or Sequentially for Postmenopausal Women With Hormone Receptor-Positive Breast Cancer: Long-Term Follow-Up of the BIG 1-98 Trial. J Clin Oncol 2019, 37: 105-114. Francis PA, Pagani O, Fleming GF, Walley BA, Colleoni M, Láng I, Gómez HL, Tondini C, Ciruelos E, Burstein HJ, et al: Tailoring Adjuvant Endocrine Therapy for Premenopausal Breast Cancer. N Engl J Med 2018, 379: 122-137. Hanker AB, Sudhan DR, Arteaga CL: Overcoming Endocrine Resistance in Breast Cancer. Cancer Cell 2020, 37: 496-513. Turner NC, Neven P, Loibl S, Andre F: Advances in the treatment of advanced oestrogen-receptor-positive breast cancer. Lancet 2017, 389: 2403-2414. Xie H, Xi X, Lei T, Liu H, Xia Z: CD8(+) T cell exhaustion in the tumor microenvironment of breast cancer. Front Immunol 2024, 15: 1507283. Deng Y, Shi M, Yi L, Naveed Khan M, Xia Z, Li X: Eliminating a barrier: Aiming at VISTA, reversing MDSC-mediated T cell suppression in the tumor microenvironment. Heliyon 2024, 10: e37060. Zhang X, Zhang P, Cong A, Feng Y, Chi H, Xia Z, Tang H: Unraveling molecular networks in thymic epithelial tumors: deciphering the unique signatures. Front Immunol 2023, 14: 1264325. Xiong J, Chi H, Yang G, Zhao S, Zhang J, Tran LJ, Xia Z, Yang F, Tian G: Revolutionizing anti-tumor therapy: unleashing the potential of B cell-derived exosomes. Front Immunol 2023, 14: 1188760. Zhao Y, Wei K, Chi H, Xia Z, Li X: IL-7: A promising adjuvant ensuring effective T cell responses and memory in combination with cancer vaccines? Front Immunol 2022, 13: 1022808. Gong X, Chi H, Strohmer DF, Teichmann AT, Xia Z, Wang Q: Exosomes: A potential tool for immunotherapy of ovarian cancer. Front Immunol 2022, 13: 1089410. Xia Z, Chen S, He M, Li B, Deng Y, Yi L, Li X: Editorial: Targeting metabolism to activate T cells and enhance the efficacy of checkpoint blockade immunotherapy in solid tumors. Front Immunol 2023, 14: 1247178. Zhang P, Pei S, Wu L, Xia Z, Wang Q, Huang X, Li Z, Xie J, Du M, Lin H: Integrating multiple machine learning methods to construct glutamine metabolism-related signatures in lung adenocarcinoma. Front Endocrinol (Lausanne) 2023, 14: 1196372. DeMichele A, Martin AM, Mick R, Gor P, Wray L, Klein-Cabral M, Athanasiadis G, Colligan T, Stadtmauer E, Weber B: Interleukin-6 -174G-->C polymorphism is associated with improved outcome in high-risk breast cancer. Cancer Res 2003, 63: 8051-8056. Studebaker AW, Storci G, Werbeck JL, Sansone P, Sasser AK, Tavolari S, Huang T, Chan MW, Marini FC, Rosol TJ, et al: Fibroblasts isolated from common sites of breast cancer metastasis enhance cancer cell growth rates and invasiveness in an interleukin-6-dependent manner. Cancer Res 2008, 68: 9087-9095. D'Anello L, Sansone P, Storci G, Mitrugno V, D'Uva G, Chieco P, Bonafé M: Epigenetic control of the basal-like gene expression profile via Interleukin-6 in breast cancer cells. Mol Cancer 2010, 9: 300. Hartman ZC, Yang XY, Glass O, Lei G, Osada T, Dave SS, Morse MA, Clay TM, Lyerly HK: HER2 overexpression elicits a proinflammatory IL-6 autocrine signaling loop that is critical for tumorigenesis. Cancer Res 2011, 71: 4380-4391. De Luca A, Lamura L, Gallo M, Maffia V, Normanno N: Mesenchymal stem cell-derived interleukin-6 and vascular endothelial growth factor promote breast cancer cell migration. J Cell Biochem 2012, 113: 3363-3370. Siersbæk R, Scabia V, Nagarajan S, Chernukhin I, Papachristou EK, Broome R, Johnston SJ, Joosten SEP, Green AR, Kumar S, et al: IL6/STAT3 Signaling Hijacks Estrogen Receptor α Enhancers to Drive Breast Cancer Metastasis. Cancer Cell 2020, 38: 412-423.e419. Zhang J, Peng G, Chi H, Yang J, Xie X, Song G, Tran LJ, Xia Z, Tian G: CD8 + T-cell marker genes reveal different immune subtypes of oral lichen planus by integrating single-cell RNA-seq and bulk RNA-sequencing. BMC Oral Health 2023, 23: 464. Sisirak V, Vey N, Goutagny N, Renaudineau S, Malfroy M, Thys S, Treilleux I, Labidi-Galy SI, Bachelot T, Dezutter-Dambuyant C, et al: Breast cancer-derived transforming growth factor-β and tumor necrosis factor-α compromise interferon-α production by tumor-associated plasmacytoid dendritic cells. Int J Cancer 2013, 133: 771-778. Cruceriu D, Balacescu L, Baldasici O, Gaal OI, Balacescu O, Russom A, Irimia D, Tudoran O: Gene expression-phenotype association study reveals the dual role of TNF-α/TNFR1 signaling axis in confined breast cancer cell migration. Life Sci 2024, 354: 122982. Additional Declarations No competing interests reported. Supplementary Files Supplementalfigures.docx Cite Share Download PDF Status: Published Journal Publication published 14 Apr, 2025 Read the published version in BMC Cancer → Version 1 posted Editorial decision: Accepted 07 Apr, 2025 Reviews received at journal 04 Apr, 2025 Reviews received at journal 03 Apr, 2025 Reviewers agreed at journal 02 Apr, 2025 Reviewers agreed at journal 02 Apr, 2025 Reviewers invited by journal 02 Apr, 2025 Submission checks completed at journal 01 Apr, 2025 First submitted to journal 30 Mar, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5841908","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":437434328,"identity":"e7a4a3be-96d4-4a56-b904-cc293c952462","order_by":0,"name":"Zhaoyi Li","email":"","orcid":"","institution":"The Second Affiliated Hospital of Xuzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhaoyi","middleName":"","lastName":"Li","suffix":""},{"id":437434330,"identity":"f53bce00-78d4-4113-ad30-8b3f0c83e3ca","order_by":1,"name":"Hao Miao","email":"","orcid":"","institution":"Xuzhou Medical College","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Miao","suffix":""},{"id":437434332,"identity":"e799e107-d59b-4eae-88ac-d23a1be93097","order_by":2,"name":"Wei Bao","email":"","orcid":"","institution":"Tongji University","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Bao","suffix":""},{"id":437434333,"identity":"f7c4c22f-eaa6-49fe-b535-91a56194c0e0","order_by":3,"name":"Lansheng Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYBACPiBmhrIZHyRU1BDWwoakhdngwZljpGlhk3zYwoxPMVQZ+9nDnwsq7thtOH72WEViAxsDf3t3An4tPHlp0jPOPEvecCYv7UbiDhkGiTNnNxBwWI4ZM2/b4WSDAzlmNxLPsDEYSOQS0ML/xvgz7z+glvNvzAoS25iJ0CKRYyDN23DYzuBGjhkDkVremEnzHDucIHnjjbFEwpljPAT9ws+fY/yZp+awPd/5HMOPPypq5Pjbe/FrgYHEBQcgDB6ilIOAvXwD0WpHwSgYBaNgpAEA9uVIa9DYg4EAAAAASUVORK5CYII=","orcid":"","institution":"The Second Affiliated Hospital of Xuzhou Medical University","correspondingAuthor":true,"prefix":"","firstName":"Lansheng","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-01-16 12:38:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5841908/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5841908/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12885-025-14101-3","type":"published","date":"2025-04-14T15:56:56+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":79907634,"identity":"06d54677-38aa-40d3-b9cb-c307c002c6b8","added_by":"auto","created_at":"2025-04-04 11:12:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":377537,"visible":true,"origin":"","legend":"\u003cp\u003eThe fowchart described the process of conducting the study and statistical analysis.\u003c/p\u003e\n\u003cp\u003eNote:LASSO=Least Absolute Shrinkage and Selection Operator,RF=Random Forest, BCLM=Breast cancer lung metastases,XGBoost=Extreme Gradient Boosting;DCA=Decision Curve Analysis,ROC=Receiver Operating Characteristic Curve.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/d777b78660e6b1e21e18af39.png"},{"id":79907633,"identity":"4cc7de2e-e2f7-4e51-8032-d6d06397369d","added_by":"auto","created_at":"2025-04-04 11:12:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":125727,"visible":true,"origin":"","legend":"\u003cp\u003eFeatures selection using Lasso algorithm. (A)The importance of 12 features was ranked using Lasso algorithm. (B)Identification of risk factors for BCLM using RF.(C)Ranking of relative importance of features of XGBoost model. (D)Five common risk factors for BCLM were visualized using a Venn diagram.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/aa06f47b39f8d477f6e471ab.png"},{"id":79907635,"identity":"1deb3f8e-a366-4bdf-bd4c-98ad2ce63f91","added_by":"auto","created_at":"2025-04-04 11:12:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":105071,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram for the prediction of LM in breast cancer\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/1d49948d78132dfed6fb28a1.png"},{"id":79908539,"identity":"739c0ded-05aa-4233-bd0f-d21912c19686","added_by":"auto","created_at":"2025-04-04 11:20:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":293373,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of the ability of nomogram to predict the risk of lung metastasis in breast cancer patients within 5 years and 10 years. (A) ROC curve in the training cohort; (B) ROC curve in the validation cohort.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/70c50c00b8d06a444c89e6d1.png"},{"id":79907637,"identity":"0486cc54-8e68-4318-9274-5cc39032fa25","added_by":"auto","created_at":"2025-04-04 11:12:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":437435,"visible":true,"origin":"","legend":"\u003cp\u003eCalibration curves in the training set (A) and validation set (B). The x-axis represents the predicted probability of the nomogram plot, and the y-axis represents the actual probability of lung metastasis in breast cancer patients.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/dcba999bb856b37e2b0e8e1a.png"},{"id":79910075,"identity":"fb3081f3-f564-4c61-a780-dfe28273b4af","added_by":"auto","created_at":"2025-04-04 11:28:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":132370,"visible":true,"origin":"","legend":"\u003cp\u003eDecision curve analysis for the training set (A) and validation set (B). The horizontal line indicates that all samples are negative and untreated, with zero net benefit. A slash indicates that all samples are positive. Net income has a negative slope.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/9c8d981124cbf9ea297f745b.png"},{"id":81050737,"identity":"4eb70135-bec0-4110-adbd-615d3b123180","added_by":"auto","created_at":"2025-04-21 16:02:45","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5805055,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/b15fc39e-3541-466a-9a69-320f44744d1d.pdf"},{"id":79908543,"identity":"66aa3442-d43f-4b21-b1bf-9c459185b1a3","added_by":"auto","created_at":"2025-04-04 11:20:51","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":7323718,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementalfigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-5841908/v1/ff917efbfd4e3be3eb58fc01.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development and validation of a nomogram model of lung metastasis in breast cancer based on machine learning algorithm and cytokines","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eBreast cancer (BC), the most frequently diagnosed malignant tumor in women worldwide, remains the major cause of cancer-related mortality among the female population[1]. It often begins as a localized disease but can subsequently spread to lymph nodes and distant organs, thereby posing significant challenges to effective treatment[2]. Approximately 10\u0026ndash;15% of breast cancer patients develop invasive diseases, with distant metastases occurring within three years of the initial diagnosis. However, metastases may also emerge at least a decade after initial detection of the cancer. Furthermore, the heterogeneity of breast cancer metastasis complicates not only the determination of effective treatment strategies but also the assessment of metastasis risk factors[3]. Distant metastases of BC commonly affect organs, such as the brain, lungs, liver and bone, where they exhibit organ-specific patterns, and hence, each site is often associated with distinct symptoms, prognosis and treatment[4]. In nearly 25% of patients with metastatic BC, the lungs are the first and sometimes the only site of metastasis[5]. Additionally, the lungs represent the second most frequent site of BC metastasis, with a 5-year overall survival rate of only 16.8%[6]. Due to the typically asymptomatic nature of lung metastasis in BC, many patients remain undiagnosed until the disease becomes incurable, thus underscoring the need for timely intervention and proper care[7]. Early detection of lung metastases and accurate prognostic evaluation are crucial for improving outcomes of BC patients in clinical practice, especially to enable better clinical management and potentially achieve long-term survival. However, in Asian populations, the clinicopathological characteristics and risk factors associated with breast cancer lung metastases (BCLM) remain underexplored. This highlights the urgent need for predictive models to identify patients at the highest risk of developing lung metastases, thereby enabling physicians to tailor treatments according to patient needs.\u003c/p\u003e \u003cp\u003eIn recent years, machine learning (ML) has emerged as a powerful tool for big data analysis, particularly for predicting the early stages of cancer[8\u0026ndash;10]. ML enables the development of algorithms that can learn from data, predict outcomes and improve over time, thereby facilitating precise decision-making[11\u0026ndash;13]. As such, its potential in exploring risk factors for disease progression and predict patient prognosis is significant[14]. Some predictive models, such as Extreme Gradient Boosting (XGBoost)[15], Least Absolute Shrinkage and Selection Operator (LASSO) [16] and Random Forest (RF) [17], have demonstrated superior generalization capabilities compared with traditional statistical models, especially since they excel at simulating and predicting complex relationships between variables and outcomes. However, despite these advances, few studies have explored the application of ML models for predicting the risk of lung metastases in BCLM.\u003c/p\u003e \u003cp\u003eInterleukins (ILs), a family of low-molecular-weight cytokines secreted by immune active cells, exhibit both pro-inflammatory and anti-inflammatory properties[18\u0026ndash;21]. ILs are involved in cancer-related inflammation, influencing tumor progression through anti-tumor immune responses or by promoting a tumor-supporting microenvironment[22]. Previous research has identified specific ILs as risk factors in breast cancer subtypes. For instance, IL-5, IL-7 and IL-16 were found to be associated with HER2-positive BC, while IL-10 levels correlated with HER2-negative cases[23]. However, the role of various cytokines in BCLM patients remains unclear, especially with regards to those associated with enhanced risks of developing lung metastases in BC. Therefore, incorporating cytokines into predictive models of lung metastasis in BC is essential. In this context, the current study aimed to identify cytokine-based risk factors for lung metastases in breast cancer and establish a predictive risk model that could guide personalized treatment strategies and improve outcomes for BC patients.\u003c/p\u003e"},{"header":"2. Methods and materials","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study design and selection of patients\u003c/h2\u003e \u003cp\u003e Approval for this study was obtained from the Ethics Committee of the Second Affiliated Hospital of Xuzhou Medical University (Ethics Approval Number: 120601). Using the inpatients\u0026rsquo; electronic medical record system of the Second Affiliated Hospital of Xuzhou Medical University, the current research retrospectively analyzed BC patients who were admitted to the hospital between September 2018 and September 2023. The following inclusion criteria was then applied: 1) a histologically-confirmed diagnosis of BC as the only primary malignant tumor; 2) Sufficient information about survival time and follow-up. In addition, BC patients were excluded if: 1) they were male; 2) the time interval from diagnosis to follow-up was less than one year; 3) rheumatic diseases and infections were present; 4) results for cytokine testing were incomplete; 5) they were unmarried; 6) multiple primary tumors were present. Overall, 326 patients met the above criteria, with 70 of them also presenting lung metastases which were diagnosed using radiological scans, biopsy or surgical resection specimens of metastatic lesions. The selected patients were then randomly assigned to two groups: a training group with 228 patients (accounting for 70%) and a validation one consisting of 98 patients (accounting for 30%). The above process of patient selection is visually represented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Data collection and processing\u003c/h2\u003e \u003cp\u003eThe study collected comprehensive baseline demographic and clinicopathological data for each participant, including age at diagnosis, TNM staging (I\u0026ndash;IV), and the number of extrathoracic metastatic sites prior to pulmonary involvement. Treatment history before lung metastasis was documented, encompassing radiotherapy, chemotherapy, endocrine therapy, targeted agents, and immunotherapy. Surgical approach (breast-conserving or radical resection), axillary lymph node involvement, maximal tumor dimension, and immunohistochemical profiles (ER, PR, HER2, Ki67) were recorded. Additional variables included histologic grade (I\u0026ndash;III), molecular subtype classification (Luminal A/B, HER2-enriched, triple-negative), tumor laterality (unilateral/bilateral), and histopathological categorization (ductal, lobular, or other). Anthropometric (BMI), menopausal status, serum tumor markers, and novel biomarkers (adenosine kinase 1, high-sensitivity C-reactive protein) were analyzed alongside hematologic parameters, albumin-fibrinogen ratios, and a multiplex cytokine panel (IFN-α/γ, IL-1β/2/4/5/6/8/10/12p70/17A, TNF-α).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Endpoint of study\u003c/h2\u003e \u003cp\u003eThis study\u0026rsquo;s primary endpoint was the occurrence of the first lung metastasis in BC patients. The follow-up deadline was defined as the time from the initial diagnosis of BC to the development of lung metastasis or the date of the last follow-up.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Machine learning\u003c/h2\u003e \u003cp\u003eMachine learning algorithms\u0026mdash;including XGBoost, random forest (RF), and least absolute shrinkage and selection operator (LASSO)\u0026mdash;were employed to systematically assess risk factors associated with lung metastasis in breast cancer (BC). LASSO, a regression-based method, facilitates feature selection and regularization by identifying the most predictive variables while minimizing overfitting[24, 25]. RF is an ensemble learning method that combines multiple predictions or classifications to improve overall accuracy of prediction. It is highly versatile, capable of handling both categorical and continuous data, while also demonstrating strong noise resistance which effectively prevents overfitting, a key consideration when analyzing complex datasets[26]. Finally, XGBoost is an ML algorithm based on the gradient boosting framework and the CART decision tree algorithm. It offers high efficiency, flexibility and portability, resulting in superior prediction accuracy[27].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Nomogram model construction\u003c/h2\u003e \u003cp\u003ePatients were randomly allocated to training and validation cohorts. The training cohort data were used to develop predictive models (RF, XGBoost, and LASSO), while internal validation was performed using the validation cohort[28, 29]. Key risk factors for breast cancer lung metastasis (BCLM) were identified through Venn diagram analysis and incorporated into nomogram models predicting 5- and 10-year BCLM risk. Lung metastasis-free survival was assessed via Kaplan-Meier analysis, with between-group differences evaluated using log-rank tests. Model performance was evaluated based on discrimination (AUC, equivalent to the C-index), calibration (calibration plots and Hosmer-Lemeshow test), and clinical utility (decision curve analysis, DCA).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Statistical analysis\u003c/h2\u003e \u003cp\u003eMetric data with a normal distribution were presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, while non-normally distributed ones were expressed as median (P25, P75). Additionally, categorical data were described as counts (percentages). Two-group comparisons were then performed using the Mann-Whitney U test and the independent sample t-test for non-parametric and parametric values, respectively. In the case of categorical variables, results were compared using the chi-square test. All statistical tests, performed using SPSS version 23.0 (SPSS Italy, Florence, Italy) and statistical software package R (version 4.0.0, R statistical calculation project), were two-tailed, with differences considered to be significant at P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Patients\u0026rsquo; baseline characteristics\u003c/h2\u003e \u003cp\u003eThis study included 326 breast cancer patients who visited the Second Affiliated Hospital of Xuzhou Medical University between September 2018 and September 2023, and among these, 70 patients also presented lung metastases. The cohort\u0026rsquo;s median age was 52 years, and the majority had undergone modified radical surgery (85.28%). Postoperative histopathological analysis revealed that invasive ductal carcinoma was the predominant histological type (74.85%), with Luminal B being the most common molecular subtype (73.62%). Regarding tumor staging, 44.79% of patients were classified as T2 stage, and 34.97% had no lymph node metastases. Additionally, most patients (70.55%) did not experience metastases to other organs before developing lung metastases. In terms of treatment, chemotherapy (92.64%) was the most commonly administered one, followed by radiotherapy (64.42%), endocrine therapy (43.56%), targeted therapy (34.97%) and immunotherapy (3.07%). The non-BCLM and BCLM groups were also significantly different (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) in terms of several parameters, including CA125, CA153, hsCRP, absolute monocyte count, TNF-α, IL-8, IL-6, IL-2, IL-1 β and IL-12p70. The clinical characteristics of the 326 BC patients and baseline comparisons between the BCLM and non-BCLM groups are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Furthermore, demographic and clinicopathological characteristics did not differ significantly between patients of the training and validation groups (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics of patients with lung metastases from breast cancer (BCLM)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCharacteristic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOverall, N\u0026thinsp;=\u0026thinsp;326\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo BCLM, N\u0026thinsp;=\u0026thinsp;256\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBCLM, N\u0026thinsp;=\u0026thinsp;70\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e52 (45, 58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e52 (45, 58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e50 (42, 58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.197\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI,kg/m\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22.97 (21.48, 24.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.97 (21.48, 24.46)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22.74 (19.81, 25.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.169\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eβ2MG, ug/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.80 (1.53, 2.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.79 (1.52, 2.23)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.93 (1.59, 2.44)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.171\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCEA2, ng/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (2, 4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (2, 3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (2, 7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.148\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTSGF, U/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e53 (46, 62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e53 (46, 61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e53 (45, 62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.981\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSCC,ug/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.58 (0.47, 0.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.59 (0.48, 0.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.56 (0.39, 0.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.182\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCA125, U/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14 (10, 22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (10, 20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22 (12, 41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCA153, U/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10 (8, 20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (7, 19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18 (10, 44)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCA50, IU/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6 (4, 11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (4, 10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (4, 12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.078\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSF,ng/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e110 (58, 219)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e104 (59, 191)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e175 (58, 330)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehsCRP,mg/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (1, 3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (1, 3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (1, 9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeutrophil count,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.05 (2.34, 4.41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.08 (2.30, 4.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.05 (2.44, 4.99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.777\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphocyte count,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.36 (1.12, 1.72)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.36 (1.13, 1.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.41 (1.09, 1.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.725\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHb,g/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e125 (114, 132)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e126 (114, 132)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e123 (113, 131)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.357\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePLT,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e236 (179, 290)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e239 (182, 292)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e231 (176, 283)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.454\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonocyte count,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.38 (0.31, 0.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.36 (0.31, 0.49)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.46 (0.35, 0.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin,g/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e43.6 (40.7, 46.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e43.5 (40.7, 46.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44.9 (41.6, 46.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.215\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFibrinogen,g/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.86 (3.27, 4.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.81 (3.27, 4.20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.96 (3.57, 4.26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.155\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIFNα, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.14 (1.46, 3.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.30 (1.46, 3.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.99 (1.46, 3.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.337\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIFNγ,pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.8 (1.9, 4.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.0 (2.0, 5.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.4 (1.7, 3.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.119\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL12p70, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.94 (1.14, 3.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.98 (1.21, 3.27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.71 (0.95, 2.34)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL17A, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (2, 10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (2, 11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (2, 8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.844\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL1β, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.75 (1.07, 2.86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.92 (1.15, 3.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.62 (0.74, 2.72)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL2, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.75 (0.98, 3.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.84 (1.03, 3.37)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.38 (0.87, 2.26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL4, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.06 (1.36, 3.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.08 (1.37, 3.40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.94 (1.25, 2.99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.281\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL5, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.11 (0.68, 1.54)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.15 (0.74, 1.59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.05 (0.63, 1.31)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.178\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL6, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (3, 9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5 (3, 8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (4, 21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL8, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9 (6, 14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (6, 13)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (6, 22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL10, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.35 (2.07, 4.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.53 (2.03, 4.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.10 (2.15, 4.15)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.452\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTNFα, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.95 (1.23, 3.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.04 (1.28, 3.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.59 (1.17, 2.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOrgan transfer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.682\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e230 (70.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e182 (71.09%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48 (68.57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96 (29.45%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e74 (28.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22 (31.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEndocrine therapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70(21.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22(31.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48(18.75%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e256 (78.53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e48(68.57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e208(81.25%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTargeted therapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.067\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e212 (65.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e160 (62.50%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e52 (74.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e114 (34.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96 (37.50%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18 (25.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eImmunotherapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.231\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e316 (96.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e250 (97.66%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66 (94.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10 (3.07%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (2.34%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (5.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRadiotherapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.046\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e116 (35.58%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84 (32.81%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32 (45.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e210 (64.42%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e172 (67.19%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (54.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChemotherapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.142\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (7.36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (6.25%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (11.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e302 (92.64%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e240 (93.75%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62 (88.57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAJCC-T\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT1, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e90 (27.61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78 (30.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (17.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT2, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e146 (44.79%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e116 (45.31%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (42.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT3, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (7.36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20 (7.81%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (5.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT4, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (7.36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (5.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (14.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (12.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28 (10.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (20.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAJCC-N\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.099\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN0, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e114 (34.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96 (37.50%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18 (25.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN1, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e82 (25.15%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e68 (26.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (20.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN2, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e62 (19.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44 (17.19%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18 (25.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN3, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (12.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30 (11.72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (17.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26 (7.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (7.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (11.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAJCC-M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.682\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM0, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e318 (97.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e250 (97.66%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68 (97.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM1, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8 (2.45%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (2.34%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSurgery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModified radical mastectomy, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e278 (85.28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e212 (82.81%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66 (94.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBreast conserving surgery, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32 (9.82%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30 (11.72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo surgery, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16 (4.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (5.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePathological grading\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.191\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14 (4.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12 (4.69%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eII, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e106 (32.52%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90 (35.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16 (22.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e60 (18.40%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46 (17.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (20.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (1.23%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (1.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e142 (43.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e104 (40.63%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (54.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eER+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.257\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e192 (58.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e156 (60.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36 (51.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e104 (31.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76 (29.69%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28 (40.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (9.20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24 (9.38%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6 (8.57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePR+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e176 (53.99%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e152 (59.38%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (34.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e120 (36.81%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80 (31.25%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40 (57.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (9.20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24 (9.38%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6 (8.57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHER2+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.970\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e212 (65.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e166 (64.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46 (65.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e78 (23.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e62 (24.22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16 (22.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36 (11.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28 (10.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (11.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eki67\u0026gt;14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e202 (61.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e160 (62.50%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42 (60.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e54 (16.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e50 (19.53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (5.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70 (21.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46 (17.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (34.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSubtype\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.109\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLuminal A, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16 (4.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (5.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLuminal B, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e240 (73.62%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e194 (75.78%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46 (65.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTriple-negative, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (12.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30 (11.72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (17.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHER2, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28 (8.59%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (7.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (14.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLaterality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.338\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeft, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e184 (56.44%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e146 (57.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (54.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e138 (42.33%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e108 (42.19%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (42.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBilateral, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (1.23%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (0.78%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePathological type\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.121\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDuctal carcinoma, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e244 (74.85%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e188 (73.44%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e56 (80.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLobular carcinoma, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6 (1.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (2.34%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther types, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36 (11.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26 (10.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (14.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e40 (12.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e36 (14.06%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (5.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMenopausal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.801\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e154 (47.24%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e120 (46.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34 (48.57%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e172 (52.76%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e136 (53.13%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36 (51.43%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003csup\u003e1\u003c/sup\u003eMedian (IQR); n (%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003csup\u003e2\u003c/sup\u003eWilcoxon rank sum test; Pearson's Chi-squared test; Fisher's exact test\u003c/p\u003e \u003cp\u003eNote:BMI\u0026thinsp;=\u0026thinsp;Body Mass Index,SF\u0026thinsp;=\u0026thinsp;Ferritin, hs-CRP\u0026thinsp;=\u0026thinsp;Hypersensitive C-reactive protein,Hb\u0026thinsp;=\u0026thinsp;Haemoglobin,PLT\u0026thinsp;=\u0026thinsp;Platelet,ER\u0026thinsp;=\u0026thinsp;Estrogen receptors,PR\u0026thinsp;=\u0026thinsp;Progesterone receptors,Her2\u0026thinsp;=\u0026thinsp;Human epidermal growth factor receptor-2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of baseline features between training group and validation group\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCharacteristic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOverall, N\u0026thinsp;=\u0026thinsp;3261\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining group, N\u0026thinsp;=\u0026thinsp;2281\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eValidation groups, N\u0026thinsp;=\u0026thinsp;981\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e52 (45, 58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e52 (44, 57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e52 (46, 59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.685\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI,kg/m\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22.97 (21.48, 24.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.89 (21.48, 24.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e23.44 (21.48, 24.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.973\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eβ2MG, ug/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.80 (1.53, 2.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.82 (1.50, 2.23)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.78 (1.54, 2.29)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.717\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCEA2, ng/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (2, 4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (2, 4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2, 3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.631\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTSGF, U/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e53 (46, 62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e52 (46, 61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e53 (47, 62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.217\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSCC,ug/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.58 (0.47, 0.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.59 (0.48, 0.79)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.56 (0.45, 0.73)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.146\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCA125, U/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14 (10, 22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (10, 23)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (10, 22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.822\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCA153, U/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10 (8, 20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11 (8, 20)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (8, 19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.934\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCA50, IU/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6 (4, 11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (4, 11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (5, 10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.904\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSF,ng/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e110 (58, 219)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e108 (58, 220)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e111 (60, 213)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.935\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ehsCRP,mg/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (1, 3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (1, 3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (1, 3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.132\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeutrophil count,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.05 (2.34, 4.41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.05 (2.42, 4.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.17 (2.30, 4.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.411\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphocyte count,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.36 (1.12, 1.72)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.40 (1.12, 1.74)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.35 (1.12, 1.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.622\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHb,g/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e125 (114, 132)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e125 (115, 132)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e125 (111, 132)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.405\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePLT,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e236 (179, 290)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e233 (180, 284)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e242 (181, 295)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.363\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonocyte count,10⁹/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.38 (0.31, 0.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.38 (0.31, 0.50)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.40 (0.30, 0.54)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.904\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin,g/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e43.6 (40.7, 46.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e43.9 (40.9, 46.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e43.0 (40.4, 46.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.301\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFibrinogen,g/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.86 (3.27, 4.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.88 (3.27, 4.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.82 (3.33, 4.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.869\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIFNα, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.14 (1.46, 3.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.15 (1.40, 3.64)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.14 (1.49, 4.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.866\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIFNγ,pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.8 (1.9, 4.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.8 (1.9, 4.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.7 (1.9, 4.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.507\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL12p70, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.94 (1.14, 3.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.96 (1.13, 3.05)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.93 (1.22, 3.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.809\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL17A, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (2, 10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (2, 11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (2, 8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.059\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL1β, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.75 (1.07, 2.86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.81 (1.16, 2.86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.64 (0.95, 2.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.121\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL2, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.75 (0.98, 3.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.76 (1.01, 2.92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.70 (0.96, 3.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.713\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL4, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.06 (1.36, 3.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.04 (1.39, 3.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.09 (1.28, 3.35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.717\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL5, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.11 (0.68, 1.54)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.10 (0.67, 1.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.16 (0.75, 1.40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.437\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL6, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (3, 9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5 (3, 9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5 (3, 9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.506\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL8, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9 (6, 14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (6, 14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (4, 16)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.118\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIL10, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.35 (2.07, 4.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.46 (2.23, 4.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.05 (1.83, 4.28)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.130\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTNFα, pg/ml\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.95 (1.23, 3.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.95 (1.26, 3.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.92 (1.17, 3.22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.739\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOrgan transfer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e230 (70.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e170 (74.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e60 (61.22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96 (29.45%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58 (25.44%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e38 (38.78%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEndocrine therapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.867\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70 (21.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e51 (22.37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19 (19.39%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e256 (78.53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e177 (77.63%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e79 (80.61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTargeted therapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e212 (65.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e158 (69.30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e54 (55.10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e114 (34.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e70 (30.70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44 (44.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eImmunotherapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.292\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e316 (96.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e219 (96.05%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e97 (98.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10 (3.07%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (3.95%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (1.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRadiotherapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.826\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e116 (35.58%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e82 (35.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34 (34.69%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e210 (64.42%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e146 (64.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e64 (65.31%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChemotherapy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (7.36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (7.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (8.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e302 (92.64%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e212 (92.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e90 (91.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAJCC-T\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.608\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT1, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e90 (27.61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63 (27.63%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e27 (27.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT2, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e146 (44.79%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e107 (46.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39 (39.80%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT3, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (7.36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (6.58%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9 (9.18%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT4, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (12.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26 (11.40%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16 (16.33%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (7.36%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17 (7.46%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (7.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAJCC-N\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.128\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN0, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e114 (34.97%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e82 (35.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32 (32.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN1, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e82 (25.15%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e53 (23.25%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29 (29.59%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN2, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e62 (19.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e45 (19.74%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (17.35%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN3, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (12.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e34 (14.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (8.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26 (7.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (6.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (12.24%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAJCC-M\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;0.999\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM0, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e318 (97.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e222 (97.37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e96 (97.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM1, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8 (2.45%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (2.63%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSurgery\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.306\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModified radical mastectomy, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e278 (85.28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e194 (85.09%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e84 (85.71%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBreast conserving surgery, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32 (9.82%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25 (10.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (7.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo surgery, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16 (4.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (3.95%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (7.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePathological grading\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.417\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14 (4.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (3.95%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5 (5.10%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eII, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e106 (32.52%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76 (33.33%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (30.61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e60 (18.40%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e47 (20.61%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (13.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (1.23%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (1.32%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (1.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e142 (43.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e93 (40.79%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e49 (50.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eER+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.022\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e176 (53.99%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e134 (58.77%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42 (42.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e120 (36.81%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77 (33.77%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e43 (43.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (9.20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17 (7.46%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (13.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePR+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e192 (58.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e146 (64.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46 (46.94%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e104 (31.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65 (28.51%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39 (39.80%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (9.20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17 (7.46%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (13.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHER2+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e212 (65.03%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e143 (62.72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e69 (70.41%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e78 (23.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63 (27.63%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15 (15.31%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36 (11.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22 (9.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (14.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eki67\u0026gt;14%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.264\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e202 (61.96%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e143 (62.72%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e59 (60.20%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e54 (16.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e41 (17.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (13.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknow, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70 (21.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44 (19.30%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e26 (26.53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSubtype\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.158\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLuminal A, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16 (4.91%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (5.70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (3.06%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLuminal B, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e240 (73.62%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e164 (71.93%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e76 (77.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTriple-negative, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (12.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (11.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15 (15.31%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHER2, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28 (8.59%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24 (10.53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (4.08%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLaterality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.252\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeft, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e184 (56.44%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e135 (59.21%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e49 (50.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e138 (42.33%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90 (39.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48 (48.98%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBilateral, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (1.23%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (1.32%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (1.02%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePathological type\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.655\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDuctal carcinoma, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e244 (74.85%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e170 (74.56%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e74 (75.51%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLobular carcinoma, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6 (1.84%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (1.75%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (2.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther types, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36 (11.04%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28 (12.28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (8.16%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnknown, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e40 (12.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26 (11.40%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (14.29%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMenopausal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.299\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e154 (47.24%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e112 (49.12%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42 (42.86%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e172 (52.76%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e116 (50.88%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e56 (57.14%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLung Metastasis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.234\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo BCLM, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e256 (78.53%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e175 (76.75%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e81 (82.65%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBCLM, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70 (21.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e53 (23.25%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (17.35%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e\u003csup\u003e1\u003c/sup\u003eMedian (IQR); n (%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003e2Wilcoxon rank sum test; Pearson's Chi-squared test; Fisher's exact test\u003c/p\u003e \u003cp\u003eNote:BMI\u0026thinsp;=\u0026thinsp;Body Mass Index,SF\u0026thinsp;=\u0026thinsp;Ferritin, hs-CRP\u0026thinsp;=\u0026thinsp;Hypersensitive C-reactive protein,Hb\u0026thinsp;=\u0026thinsp;Haemoglobin,PLT\u0026thinsp;=\u0026thinsp;Platelet,ER\u0026thinsp;=\u0026thinsp;Estrogen receptors,PR\u0026thinsp;=\u0026thinsp;Progesterone receptors,Her2\u0026thinsp;=\u0026thinsp;Human epidermal growth factor receptor-2.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Identification of BCLM risk factors\u003c/h2\u003e \u003cp\u003eTo identify the risk factors for BCLM, the LASSO algorithm was employed, with Supplementary Figure S1-A showing the binomial deviation curve plotted against the logarithm of the tuning hyperparameter (λ). In this case, the solid vertical line indicates the binomial deviation\u0026thinsp;\u0026plusmn;\u0026thinsp;standard error (SE), while the optimal λ value was determined using the minimum standard and 1-SE standard through 10-fold cross validation. Furthermore, a coefficient profile was generated from the log (λ) sequence, with 49 clinical parameters integrated into the LASSO model to enable effective penalization of non-essential features. Following model training and the 10-fold cross validation, 12 non-zero coefficients were identified as being significantly associated with lung metastasis (Supplementary S1B). According to the Lasso model\u0026rsquo;s feature importance ranking (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), the relative importance of predictors from highest to lowest was as follows: other organ metastasis, endocrine therapy, PR status, absolute lymphocyte count, targeted therapy, IL-2, INF-α, TNF-α, CEA, CA125, hsCRP, and IL-6.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Identification of risk factors for BCLM\u003c/h2\u003e \u003cp\u003eThe RF machine learning algorithm was used to further refine the selection of risk factors. This algorithm works by randomly extracting subsets of features from the training data, with each subset subsequently utilized to construct a decision tree. For each node within these decision trees, the optimal feature was chosen from a random subset of features for node partitioning. The decision tree was then recursively built based on the selected features until a predefined stopping condition was met. For classification problems, the final class was determined through a majority voting mechanism, while in the case of regression problems, the average of the predicted values from all trees served as the final prediction. The algorithm eventually combined the outputs from all constructed decision trees to calculate the average error rates separately for node-positive and node-negative groups. The importance of clinical features was subsequently assessed before visualizing their rankings (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Overall, 24 clinically relevant features were identified as risk factors for BCLM, and they included IL-6, PR, BMI, SCC, FIB, ER, endocrine therapy, IL-8, ALB, IL-17A, Hb, IFN-γ, hsCRP, TNF-α, IL-12p70, CA125, TSGF, PLT, INF-α, N, SF, B2M, CA50 and IL10.\u003c/p\u003e \u003cp\u003eThe XGBoost model, based on a gradient boosting framework, is another ensemble method that uses decision trees to enhance predictive accuracy. Gradient boosting is a specific implementation of the Boosting technique which iteratively minimize the objective function by fitting each new tree with the negative gradient of the previous round\u0026rsquo;s error. In this study, the XGBoost model identified 15 non-zero coefficients that were significantly correlated with lung metastasis. These features were then ranked by relative importance (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eC) as follows: endocrine therapy, IL-6, hsCRP, IL-17A, PLT, IL-8, SCC, SF, INF-α, CEA, ALB, Hb, IL2, M, TNF-α.\u003c/p\u003e \u003cp\u003eThe RF, LASSO and XGBoost algorithms were used to independently identify BCLM-related risk factors, with overlapping variables among the three ML models subsequently selected as significant ones. The intersection of these factors was visualized using a Venn diagram (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2\u003c/span\u003eD), and the results highlighted five key variables for subsequent nomogram analysis: endocrine therapy, hsCRP, IL6, IFN-α and TNF-α. To enhance clinical applicability, the Maxstat method was then used to assess the optimal risk cut-off points for the five variables (Supplementary Figure S2). Using these cut-off values (hsCRP (16.8), IL6 (16.19), IFN-ɑ (2.36) and TNF-ɑ (1.35)), the biomarkers were reclassified into high- and low-risk groups prior to analysis using Kaplan-Meier curves (Supplementary Figure S3) to determine survival outcomes. Additionally, Supplementary Figure S4 shows the lung metastasis rates across different molecular subtypes. Compared with Luminal A patients, those with the Luminal B subtype demonstrated a lower risk of lung metastasis (unadjusted HR: 1.787; P\u0026thinsp;=\u0026thinsp;0.435), while HER2+ (unadjusted HR: 3.571; P\u0026thinsp;=\u0026thinsp;0.094) exhibited a higher risk. In particular, patients with TNBC faced the highest risk of lung metastasis (unadjusted HR: 6.487; P\u0026thinsp;=\u0026thinsp;0.018).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Establishment and validation of BCLM diagnostic nomogram\u003c/h2\u003e \u003cp\u003eA nomogram model was constructed based on five key variables: endocrine therapy, hsCRP, IL6, IFN-ɑ, and TNF-ɑ. Each variable was assigned a point value ranging from 0 to 100 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The cumulative score, obtained by summing these points, allowed estimation of the 5- and 10-year lung metastasis probability in breast cancer (BC) patients, aiding clinical decision-making. Risk prediction involved drawing a vertical line from the total score to the probability axis (ranging from 0.1 to 0.95), though not all probabilities aligned precisely with marked values. The model's performance was evaluated based on discrimination, calibration, and clinical utility in both training and validation cohorts, with results visualized using ROC curves, calibration plots, and decision curve analysis. In the training set, the AUC values for 5- and 10-year metastasis prediction were 0.786 (95% CI: 0.691\u0026ndash;0.881) and 0.787 (95% CI: 0.749\u0026ndash;0.824), respectively. The validation set yielded AUCs of 0.627 (95% CI: 0.441\u0026ndash;0.813) for 5-year and 0.797 (95% CI: 0.605\u0026ndash;0.988) for 10-year prediction. These findings indicate robust predictive accuracy across both datasets (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003eA, B).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Calibration curve and DCA analysis\u003c/h2\u003e \u003cp\u003eCalibration curves were generated for evaluating the nomogram\u0026rsquo;s performance. Following internal validation with 1000 bootstrap iterations, the calibration curves for both the training and validation sets (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, B) closely aligned with the diagonal line, indicating that the predicted and actual probabilities of lung metastasis were in strong agreement. The nomogram\u0026rsquo;s clinical utility was assessed using DCA (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e6\u003c/span\u003eA, B), in which the horizontal line represented the assumption of no lung metastasis where the net benefit was zero and the diagonal line represented the scenario where all patients were assumed to have BCLM. Overall, the decision curves demonstrated that the range of high threshold probabilities was broad and applicable to both the training and validation sets. Compared with individual variables, the nomogram exhibited a higher net benefit for both datasets, thus underscoring its superior predictive ability. This indicates that the nomogram can effectively predict the 5-year and 10-year risk of lung metastasis in BC patients.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eIn this study, multiple ML algorithms were applied to determine the risk factors for BCLM, with the following five significant predictors subsequently identified: endocrine therapy, hsCRP, IL6, IFN-ɑ, and TNF-ɑ. These variables were then integrated into a nomogram model. The findings provided a framework for identifying BC patients who were at a higher risk of lung metastasis, thereby improving prognostic evaluation and clinical management while offering new insights for developing more effective treatments. Additionally, this study was also the first one to construct a lung metastasis prediction model for BC patients based on cytokines. The model demonstrated high accuracy in predicting survival outcomes for BCLM patients, and in practice, the nomogram, which integrated predictions from RF, LASSO and XGBoost algorithms, exhibited robust performance across both the training and validation groups.\u003c/p\u003e \u003cp\u003eA Mendelian randomization analysis involving 420,964 cancer-free patients from the UK Biobank cohort showed that elevated serum c-reactive protein (CRP) levels were linked to higher risks of breast cancer, colorectal cancer, head and neck as well as other malignancies over a 7.1-year follow-up period [30]. Similarly, a meta-analysis examining 119 inflammatory markers (with CRP as the primary focus) across 26 cancer types reached comparable conclusions [31]. These pan-cancer studies identified individuals with CRP levels above 3 mg/L as having a high risk of inflammation but this threshold may not apply specifically to BC[30, 32]. The findings further corroborated the link between elevated hypersensitive CRP and an increased risk of BCLM. This suggests that inflammation in BC patients may contribute to tumor proliferation and metastasis, including lung metastasis. The above analyses also determined the optimal hsCRP cut-off value for predicting BCLM to be 16.8 mg/L, hence providing a potential reference point for individualized breast cancer treatment. Interestingly, CRP was consistently identified as a key risk factor for BCLM across all three ML models used in this study. While prior meta-analyses have highlighted the limited predictive value of CRP in non-metastatic BC, the association between elevated CRP levels and poor prognosis is well documented in metastatic cases [33, 34]. For instance, in vitro studies revealed that CRP could promote the adhesion of MCF10A human breast epithelial cells through activation of the integrin α 2 signaling pathway and Fcγ receptor I (FcγRI), with the process subsequently activating paxillin, FAK and ERKs to drive autocrine effects [35]. Furthermore, using an invasion model of MDA-MB-231 TNBC cells and mouse tumor models, CRP was shown to be involved in tumor growth. Additional animal experiments further demonstrated that CRP impaired immune surveillance by inhibiting the activation of pulmonary macrophages, induced by symbiotic bacteria through an FcγR2B dependent mechanism, thereby fostering the formation of pre-metastatic niches in the lungs of tumor-bearing mice [36]. Altogether, these findings highlight the significant role of CRP in lung metastasis, thus supporting this study\u0026rsquo;s results.\u003c/p\u003e \u003cp\u003eThis study underscores the potential of endocrine therapy to reduce the risk of BCLM, with this lower risk being particularly evident among hormone receptor-positive patients who constituted over half of the total study population. Of these patients, 80% received endocrine therapy, including options such as tamoxifen and steroidal (exemestane) or nonsteroidal (letrozole or anastrozole) aromatase inhibitors. Tamoxifen is known to improve disease-free and overall survival in postmenopausal women with ER-positive tumors[37, 38]. However, the BIG trial demonstrated that first-line treatment with aromatase inhibitors lowered the absolute risk of 10-year recurrence by 3.6%, increased overall survival by 2.1% and outperformed tamoxifen monotherapy [39]. Furthermore, post hoc analyses of the SOFT and TEXT trials revealed that combining ovarian suppression with tamoxifen significantly improved 8-year disease-free and overall survival rates in comparison with tamoxifen alone [40]. Despite the success of endocrine therapy in reducing BC recurrence and mortality, both intrinsic and acquired drug resistance remain a challenge. In this context, recent advances in understanding the drivers and mechanisms underlying endocrine therapy resistance in estrogen receptor-positive BC has led to the development of targeted drugs, such as mTOR inhibitors and cyclin dependent kinase 4/6 inhibitors can markedly extend progression-free survival [41, 42]. When lung metastasis rates were further analyzed by molecular subtypes, it was found that the risk of lung metastasis was significantly lower in hormone receptor-positive patients compared with HER2\u0026thinsp;+\u0026thinsp;ones, with TNBC patients exhibiting the highest risk. These results underscore the importance of endocrine therapy in mitigating the risk of BCLM.\u003c/p\u003e \u003cp\u003eThe TME comprises both cellular elements, including adipocytes, immune cells[43, 44], endothelial cells and cancer-associated fibroblasts, as well as non-cellular components[45\u0026ndash;48], such as cytokines and the ECM. It promotes tumor progression and invasion through the secretion of growth factors and pro-inflammatory mediators as well as through intercellular interactions and metabolic crosstalk with tumor cells[49, 50]. In this study, cytokines were innovatively incorporated into a lung metastasis model, and three key inflammatory factors (IL6, IFN-ɑ and TNF-ɑ) associated with lung metastasis were then identified using RF, LASSO and XGBoost ML algorithms. These cytokines have been extensively studied in the context of BC metastasis mechanisms. For instance, early research has demonstrated that IL-6-174 promoter polymorphism was linked to clinical outcomes in a group of lymph node-positive BC patients undergoing high-dose adjuvant therapy [51]. Additionally, Adam et al. reported that fibroblasts isolated from common sites of breast cancer metastasis enhanced the growth and invasiveness of cancer cells in an IL-6-dependent manner [52]. Similarly, Laura et al. found that p53 inactivation triggered a methylation-dependent autocrine IL-6 loop that led to epigenetic reprogramming and the development of basal/stem cell-like gene expression profiles in BC cells [53]. HER2 overexpression has also been shown to induce IL-6 secretion, activate STAT3, alter gene expression and reinforce the autocrine IL-6/STAT3 loop [54]. In one study, Luca et al. reported that the combination of VEGF and IL-6 synergistically and durably activated intracellular signaling pathways, such as MAPK, AKT and p38MAPK, in BC cells [55], while Rasmus et al. demonstrated that, in ER\u0026thinsp;+\u0026thinsp;breast cancer, the IL6/STAT3 signaling pathway could drive metastasis independently of the estrogen receptor. Although STAT3 and ER share enhancers, the former can hijack a subset of ER enhancers to induce unique transcriptional programs. This decoupling of ER and IL6/STAT3 oncogenic pathways underscores the therapeutic potential of targeting IL6/STAT3 in ER\u0026thinsp;+\u0026thinsp;breast cancer [56]. In contrast to IL-6, IFN-α and TNF-α inhibit breast cancer growth and invasion through distinct mechanisms. Specifically, IFN-α is involved in tumor immune surveillance by activating CD8 α\u0026thinsp;+\u0026thinsp;dendritic cells (DCs) and enhancing CD8\u0026thinsp;+\u0026thinsp;T cell recognition of tumor antigens[57]. Thus, a deficiency in IFN-α can disrupt this process, leading to the expansion of regulatory T cells (Tregs) which suppresses plasma cell-like DCs and facilitate BC metastasis [58]. On the other hand, TNF-α can restrict the migration of triple negative, mesenchymal-like BC cells with high TNFR1 expression, while inhibiting the migration of epithelioid cells with low TNFR1 expression [59].\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eThis study applied three ML methods to systematically analyze clinical information and surgical pathology results, integrated treatment exposure and inflammatory markers, and established a predictive model for BCLM. This model exhibits strong discriminative ability in both training and validation queues. In fact, through this nomogram, doctors can estimate the likelihood of lung metastasis in BC patients based on the cumulative score of each risk factor. Therefore, this tool can achieve personalized risk assessment by regularly reviewing inflammation indicators for high-risk patients, and immediately initiating imaging screening for patients with improved scores. In addition, the findings highlighted the contrasting roles of cytokines in BC, with IL-6 promoting BCLM, while IFN-α and TNF-α inhibited tumor metastasis. These insights deepen current understanding of the interplay between cytokines and BCLM, thus underscoring the importance of detecting and managing inflammation associated with BC. Future works should validate the current findings through large, prospective, multi-center trials.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eBMI \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Body Mass Index\u003c/p\u003e\n\u003cp\u003eSF \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Ferritin\u003c/p\u003e\n\u003cp\u003ehs-CRP \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Hypersensitive C-reactive protein\u003c/p\u003e\n\u003cp\u003eHb \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Haemoglobin,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePLT \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Platelet\u003c/p\u003e\n\u003cp\u003eER \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Estrogen receptors\u003c/p\u003e\n\u003cp\u003ePR \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Progesterone receptors\u003c/p\u003e\n\u003cp\u003eHer2 \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Human epidermal growth factor receptor-2\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Human Research Ethics Committee, the Second Affiliated Hospital of Xuzhou Medical University (120601). The Ethics Committee of the Second Affiliated Hospital of Xuzhou Medical University agreed to waive informed consent because this retrospective observational study uses medical recordsfrom previous clinical diagnosis, the risk of which was not greater than the minimum risk.And the research process was in accordance with the content of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and analysed during the current study are not publicly available due to privacy or ethical restrictions but are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Project of Jiangsu Health Commission(M2020072).\u003cstrong\u003e\u0026nbsp; \u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eZL conceived of the study. ZL and HM designed the study and collected data. WB contributed to data analysis. ZL, HM, LZ contributed to the writing and revision of the paper.All authors read and approved the fnal manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;Acknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eQ L, C X, H L, X Y, F Y, M C, S Z, Y T, S H, M C, W C: \u003cstrong\u003e- Disparities in 36 cancers across 185 countries: secondary analysis of global.\u003c/strong\u003e \u003cem\u003e- Front Med 2024 Oct;18(5):911-920 doi: 101007/s11684-024-1058-6 Epub 2024 Aug\u003c/em\u003e\u003cstrong\u003e:\u003c/strong\u003e- 911-920.\u003c/li\u003e\n\u003cli\u003ePark M, Kim D, Ko S, Kim A, Mo K, Yoon H: \u003cstrong\u003eBreast Cancer Metastasis: Mechanisms and Therapeutic Implications.\u003c/strong\u003e \u003cem\u003eInt J Mol Sci \u003c/em\u003e2022, \u003cstrong\u003e23\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eWeigelt B, Peterse JL, van \u0026apos;t Veer LJ: \u003cstrong\u003eBreast cancer metastasis: markers and models.\u003c/strong\u003e \u003cem\u003eNat Rev Cancer \u003c/em\u003e2005, \u003cstrong\u003e5:\u003c/strong\u003e591-602.\u003c/li\u003e\n\u003cli\u003eWang C, Xu K, Wang R, Han X, Tang J, Guan X: \u003cstrong\u003eHeterogeneity of BCSCs contributes to the metastatic organotropism of breast cancer.\u003c/strong\u003e \u003cem\u003eJ Exp Clin Cancer Res \u003c/em\u003e2021, \u003cstrong\u003e40:\u003c/strong\u003e370.\u003c/li\u003e\n\u003cli\u003ePillar N, Polsky AL, Weissglas-Volkov D, Shomron N: \u003cstrong\u003eComparison of breast cancer metastasis models reveals a possible mechanism of tumor aggressiveness.\u003c/strong\u003e \u003cem\u003eCell Death Dis \u003c/em\u003e2018, \u003cstrong\u003e9:\u003c/strong\u003e1040.\u003c/li\u003e\n\u003cli\u003eLiang Y, Zhang H, Song X, Yang Q: \u003cstrong\u003eMetastatic heterogeneity of breast cancer: Molecular mechanism and potential therapeutic targets.\u003c/strong\u003e \u003cem\u003eSemin Cancer Biol \u003c/em\u003e2020, \u003cstrong\u003e60:\u003c/strong\u003e14-27.\u003c/li\u003e\n\u003cli\u003eLin S, Mo H, Li Y, Guan X, Chen Y, Wang Z, Xu B: \u003cstrong\u003eClinicopathological characteristics and survival outcomes in patients with synchronous lung metastases upon initial metastatic breast cancer diagnosis in Han population.\u003c/strong\u003e \u003cem\u003eBMC Cancer \u003c/em\u003e2021, \u003cstrong\u003e21:\u003c/strong\u003e1330.\u003c/li\u003e\n\u003cli\u003eJin W, Yang Q, Chi H, Wei K, Zhang P, Zhao G, Chen S, Xia Z, Li X: \u003cstrong\u003eEnsemble deep learning enhanced with self-attention for predicting immunotherapeutic responses to cancers.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2022, \u003cstrong\u003e13:\u003c/strong\u003e1025330.\u003c/li\u003e\n\u003cli\u003eRen Q, Zhang P, Lin H, Feng Y, Chi H, Zhang X, Xia Z, Cai H, Yu Y: \u003cstrong\u003eA novel signature predicts prognosis and immunotherapy in lung adenocarcinoma based on cancer-associated fibroblasts.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1201573.\u003c/li\u003e\n\u003cli\u003eZhang S, Jiang C, Jiang L, Chen H, Huang J, Gao X, Xia Z, Tran LJ, Zhang J, Chi H, et al: \u003cstrong\u003eConstruction of a diagnostic model for hepatitis B-related hepatocellular carcinoma using machine learning and artificial neural networks and revealing the correlation by immunoassay.\u003c/strong\u003e \u003cem\u003eTumour Virus Res \u003c/em\u003e2023, \u003cstrong\u003e16:\u003c/strong\u003e200271.\u003c/li\u003e\n\u003cli\u003eZhao S, Zhang X, Gao F, Chi H, Zhang J, Xia Z, Cheng C, Liu J: \u003cstrong\u003eIdentification of copper metabolism-related subtypes and establishment of the prognostic model in ovarian cancer.\u003c/strong\u003e \u003cem\u003eFront Endocrinol (Lausanne) \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1145797.\u003c/li\u003e\n\u003cli\u003eChi H, Yang J, Peng G, Zhang J, Song G, Xie X, Xia Z, Liu J, Tian G: \u003cstrong\u003eCircadian rhythm-related genes index: A predictor for HNSCC prognosis, immunotherapy efficacy, and chemosensitivity.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1091218.\u003c/li\u003e\n\u003cli\u003eZhao S, Chi H, Yang Q, Chen S, Wu C, Lai G, Xu K, Su K, Luo H, Peng G, et al: \u003cstrong\u003eIdentification and validation of neurotrophic factor-related gene signatures in glioblastoma and Parkinson\u0026apos;s disease.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1090040.\u003c/li\u003e\n\u003cli\u003eChi H, Jiang P, Xu K, Zhao Y, Song B, Peng G, He B, Liu X, Xia Z, Tian G: \u003cstrong\u003eA novel anoikis-related gene signature predicts prognosis in patients with head and neck squamous cell carcinoma and reveals immune infiltration.\u003c/strong\u003e \u003cem\u003eFront Genet \u003c/em\u003e2022, \u003cstrong\u003e13:\u003c/strong\u003e984273.\u003c/li\u003e\n\u003cli\u003eKang J, Choi YJ, Kim IK, Lee HS, Kim H, Baik SH, Kim NK, Lee KY: \u003cstrong\u003eLASSO-Based Machine Learning Algorithm for Prediction of Lymph Node Metastasis in T1 Colorectal Cancer.\u003c/strong\u003e \u003cem\u003eCancer Res Treat \u003c/em\u003e2021, \u003cstrong\u003e53:\u003c/strong\u003e773-783.\u003c/li\u003e\n\u003cli\u003eChi H, Xie X, Yan Y, Peng G, Strohmer DF, Lai G, Zhao S, Xia Z, Tian G: \u003cstrong\u003eNatural killer cell-related prognosis signature characterizes immune landscape and predicts prognosis of HNSCC.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2022, \u003cstrong\u003e13:\u003c/strong\u003e1018685.\u003c/li\u003e\n\u003cli\u003eLi J, Shi Z, Liu F, Fang X, Cao K, Meng Y, Zhang H, Yu J, Feng X, Li Q, et al: \u003cstrong\u003eXGBoost Classifier Based on Computed Tomography Radiomics for Prediction of Tumor-Infiltrating CD8(+) T-Cells in Patients With Pancreatic Ductal Adenocarcinoma.\u003c/strong\u003e \u003cem\u003eFront Oncol \u003c/em\u003e2021, \u003cstrong\u003e11:\u003c/strong\u003e671333.\u003c/li\u003e\n\u003cli\u003eZhai X, Zhang H, Xia Z, Liu M, Du G, Jiang Z, Zhou H, Luo D, Dou D, Li J, et al: \u003cstrong\u003eOxytocin alleviates liver fibrosis via hepatic macrophages.\u003c/strong\u003e \u003cem\u003eJHEP Rep \u003c/em\u003e2024, \u003cstrong\u003e6:\u003c/strong\u003e101032.\u003c/li\u003e\n\u003cli\u003eXiao J, Lin H, Liu B, Xia Z, Zhang J, Jin J: \u003cstrong\u003eDecreased S1P and SPHK2 are involved in pancreatic acinar cell injury.\u003c/strong\u003e \u003cem\u003eBiomark Med \u003c/em\u003e2019, \u003cstrong\u003e13:\u003c/strong\u003e627-637.\u003c/li\u003e\n\u003cli\u003eXiao J, Huang K, Lin H, Xia Z, Zhang J, Li D, Jin J: \u003cstrong\u003eMogroside II(E) Inhibits Digestive Enzymes via Suppression of Interleukin 9/Interleukin 9 Receptor Signalling in Acute Pancreatitis.\u003c/strong\u003e \u003cem\u003eFront Pharmacol \u003c/em\u003e2020, \u003cstrong\u003e11:\u003c/strong\u003e859.\u003c/li\u003e\n\u003cli\u003eZhang H, Xia T, Xia Z, Zhou H, Li Z, Wang W, Zhai X, Jin B: \u003cstrong\u003eKIF18A inactivates hepatic stellate cells and alleviates liver fibrosis through the TTC3/Akt/mTOR pathway.\u003c/strong\u003e \u003cem\u003eCell Mol Life Sci \u003c/em\u003e2024, \u003cstrong\u003e81:\u003c/strong\u003e96.\u003c/li\u003e\n\u003cli\u003eHabanjar O, Bingula R, Decombat C, Diab-Assaf M, Caldefie-Chezet F, Delort L: \u003cstrong\u003eCrosstalk of Inflammatory Cytokines within the Breast Tumor Microenvironment.\u003c/strong\u003e \u003cem\u003eInt J Mol Sci \u003c/em\u003e2023, \u003cstrong\u003e24\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eZhou H, Cai Z, Yang Q, Yang X, Chen J, Huang T: \u003cstrong\u003eInflammatory cytokines and two subtypes of breast cancer: A two-sample mendelian randomization study.\u003c/strong\u003e \u003cem\u003ePLoS One \u003c/em\u003e2023, \u003cstrong\u003e18:\u003c/strong\u003ee0293230.\u003c/li\u003e\n\u003cli\u003eZhang X, Zhuge J, Liu J, Xia Z, Wang H, Gao Q, Jiang H, Qu Y, Fan L, Ma J, et al: \u003cstrong\u003ePrognostic signatures of sphingolipids: Understanding the immune landscape and predictive role in immunotherapy response and outcomes of hepatocellular carcinoma.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1153423.\u003c/li\u003e\n\u003cli\u003eWang X, Zhao Y, Strohmer DF, Yang W, Xia Z, Yu C: \u003cstrong\u003eThe prognostic value of MicroRNAs associated with fatty acid metabolism in head and neck squamous cell carcinoma.\u003c/strong\u003e \u003cem\u003eFront Genet \u003c/em\u003e2022, \u003cstrong\u003e13:\u003c/strong\u003e983672.\u003c/li\u003e\n\u003cli\u003eYu Y, He Z, Ouyang J, Tan Y, Chen Y, Gu Y, Mao L, Ren W, Wang J, Lin L, et al: \u003cstrong\u003eMagnetic resonance imaging radiomics predicts preoperative axillary lymph node metastasis to support surgical decisions and is associated with tumor microenvironment in invasive breast cancer: A machine learning, multicenter study.\u003c/strong\u003e \u003cem\u003eEBioMedicine \u003c/em\u003e2021, \u003cstrong\u003e69:\u003c/strong\u003e103460.\u003c/li\u003e\n\u003cli\u003eZhang H, Lin F, Zheng T, Gao J, Wang Z, Zhang K, Zhang X, Xu C, Zhao F, Xie H, et al: \u003cstrong\u003eArtificial intelligence-based classification of breast lesion from contrast enhanced mammography: a multicenter study.\u003c/strong\u003e \u003cem\u003eInt J Surg \u003c/em\u003e2024, \u003cstrong\u003e110:\u003c/strong\u003e2593-2603.\u003c/li\u003e\n\u003cli\u003eLiu J, Zhang P, Yang F, Jiang K, Sun S, Xia Z, Yao G, Tang J: \u003cstrong\u003eIntegrating single-cell analysis and machine learning to create glycosylation-based gene signature for prognostic prediction of uveal melanoma.\u003c/strong\u003e \u003cem\u003eFront Endocrinol (Lausanne) \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1163046.\u003c/li\u003e\n\u003cli\u003eChi H, Gao X, Xia Z, Yu W, Yin X, Pan Y, Peng G, Mao X, Teichmann AT, Zhang J, et al: \u003cstrong\u003eFAM family gene prediction model reveals heterogeneity, stemness and immune microenvironment of UCEC.\u003c/strong\u003e \u003cem\u003eFront Mol Biosci \u003c/em\u003e2023, \u003cstrong\u003e10:\u003c/strong\u003e1200335.\u003c/li\u003e\n\u003cli\u003eZhu M, Ma Z, Zhang X, Hang D, Yin R, Feng J, Xu L, Shen H: \u003cstrong\u003eC-reactive protein and cancer risk: a pan-cancer study of prospective cohort and Mendelian randomization analysis.\u003c/strong\u003e \u003cem\u003eBMC Med \u003c/em\u003e2022, \u003cstrong\u003e20:\u003c/strong\u003e301.\u003c/li\u003e\n\u003cli\u003eMichels N, van Aart C, Morisse J, Mullee A, Huybrechts I: \u003cstrong\u003eChronic inflammation towards cancer incidence: A systematic review and meta-analysis of epidemiological studies.\u003c/strong\u003e \u003cem\u003eCrit Rev Oncol Hematol \u003c/em\u003e2021, \u003cstrong\u003e157:\u003c/strong\u003e103177.\u003c/li\u003e\n\u003cli\u003eSiemes C, Visser LE, Coebergh JW, Splinter TA, Witteman JC, Uitterlinden AG, Hofman A, Pols HA, Stricker BH: \u003cstrong\u003eC-reactive protein levels, variation in the C-reactive protein gene, and cancer risk: the Rotterdam Study.\u003c/strong\u003e \u003cem\u003eJ Clin Oncol \u003c/em\u003e2006, \u003cstrong\u003e24:\u003c/strong\u003e5216-5222.\u003c/li\u003e\n\u003cli\u003eHan Y, Mao F, Wu Y, Fu X, Zhu X, Zhou S, Zhang W, Sun Q, Zhao Y: \u003cstrong\u003ePrognostic role of C-reactive protein in breast cancer: a systematic review and meta-analysis.\u003c/strong\u003e \u003cem\u003eInt J Biol Markers \u003c/em\u003e2011, \u003cstrong\u003e26:\u003c/strong\u003e209-215.\u003c/li\u003e\n\u003cli\u003eMikkelsen MK, Lindblom NAF, Dyhl-Polk A, Juhl CB, Johansen JS, Nielsen D: \u003cstrong\u003eSystematic review and meta-analysis of C-reactive protein as a biomarker in breast cancer.\u003c/strong\u003e \u003cem\u003eCrit Rev Clin Lab Sci \u003c/em\u003e2022, \u003cstrong\u003e59:\u003c/strong\u003e480-500.\u003c/li\u003e\n\u003cli\u003eKim ES, Kim SY, Koh M, Lee HM, Kim K, Jung J, Kim HS, Moon WK, Hwang S, Moon A: \u003cstrong\u003eC-reactive protein binds to integrin \u0026alpha;2 and Fc\u0026gamma; receptor I, leading to breast cell adhesion and breast cancer progression.\u003c/strong\u003e \u003cem\u003eOncogene \u003c/em\u003e2018, \u003cstrong\u003e37:\u003c/strong\u003e28-38.\u003c/li\u003e\n\u003cli\u003eFeng JR, Li X, Han C, Chang Y, Fu Y, Feng GC, Lei Y, Li HY, Tang PM, Ji SR, et al: \u003cstrong\u003eC-Reactive Protein Induces Immunosuppression by Activating Fc\u0026gamma;R2B in Pulmonary Macrophages to Promote Lung Metastasis.\u003c/strong\u003e \u003cem\u003eCancer Res \u003c/em\u003e2024, \u003cstrong\u003e84:\u003c/strong\u003e4184-4198.\u003c/li\u003e\n\u003cli\u003eJaiyesimi IA, Buzdar AU, Decker DA, Hortobagyi GN: \u003cstrong\u003eUse of tamoxifen for breast cancer: twenty-eight years later.\u003c/strong\u003e \u003cem\u003eJ Clin Oncol \u003c/em\u003e1995, \u003cstrong\u003e13:\u003c/strong\u003e513-529.\u003c/li\u003e\n\u003cli\u003eLegha SS: \u003cstrong\u003eTamoxifen in the treatment of breast cancer.\u003c/strong\u003e \u003cem\u003eAnn Intern Med \u003c/em\u003e1988, \u003cstrong\u003e109:\u003c/strong\u003e219-228.\u003c/li\u003e\n\u003cli\u003eRuhstaller T, Giobbie-Hurder A, Colleoni M, Jensen MB, Ejlertsen B, de Azambuja E, Neven P, L\u0026aacute;ng I, Jakobsen EH, Gladieff L, et al: \u003cstrong\u003eAdjuvant Letrozole and Tamoxifen Alone or Sequentially for Postmenopausal Women With Hormone Receptor-Positive Breast Cancer: Long-Term Follow-Up of the BIG 1-98 Trial.\u003c/strong\u003e \u003cem\u003eJ Clin Oncol \u003c/em\u003e2019, \u003cstrong\u003e37:\u003c/strong\u003e105-114.\u003c/li\u003e\n\u003cli\u003eFrancis PA, Pagani O, Fleming GF, Walley BA, Colleoni M, L\u0026aacute;ng I, G\u0026oacute;mez HL, Tondini C, Ciruelos E, Burstein HJ, et al: \u003cstrong\u003eTailoring Adjuvant Endocrine Therapy for Premenopausal Breast Cancer.\u003c/strong\u003e \u003cem\u003eN Engl J Med \u003c/em\u003e2018, \u003cstrong\u003e379:\u003c/strong\u003e122-137.\u003c/li\u003e\n\u003cli\u003eHanker AB, Sudhan DR, Arteaga CL: \u003cstrong\u003eOvercoming Endocrine Resistance in Breast Cancer.\u003c/strong\u003e \u003cem\u003eCancer Cell \u003c/em\u003e2020, \u003cstrong\u003e37:\u003c/strong\u003e496-513.\u003c/li\u003e\n\u003cli\u003eTurner NC, Neven P, Loibl S, Andre F: \u003cstrong\u003eAdvances in the treatment of advanced oestrogen-receptor-positive breast cancer.\u003c/strong\u003e \u003cem\u003eLancet \u003c/em\u003e2017, \u003cstrong\u003e389:\u003c/strong\u003e2403-2414.\u003c/li\u003e\n\u003cli\u003eXie H, Xi X, Lei T, Liu H, Xia Z: \u003cstrong\u003eCD8(+) T cell exhaustion in the tumor microenvironment of breast cancer.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2024, \u003cstrong\u003e15:\u003c/strong\u003e1507283.\u003c/li\u003e\n\u003cli\u003eDeng Y, Shi M, Yi L, Naveed Khan M, Xia Z, Li X: \u003cstrong\u003eEliminating a barrier: Aiming at VISTA, reversing MDSC-mediated T cell suppression in the tumor microenvironment.\u003c/strong\u003e \u003cem\u003eHeliyon \u003c/em\u003e2024, \u003cstrong\u003e10:\u003c/strong\u003ee37060.\u003c/li\u003e\n\u003cli\u003eZhang X, Zhang P, Cong A, Feng Y, Chi H, Xia Z, Tang H: \u003cstrong\u003eUnraveling molecular networks in thymic epithelial tumors: deciphering the unique signatures.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1264325.\u003c/li\u003e\n\u003cli\u003eXiong J, Chi H, Yang G, Zhao S, Zhang J, Tran LJ, Xia Z, Yang F, Tian G: \u003cstrong\u003eRevolutionizing anti-tumor therapy: unleashing the potential of B cell-derived exosomes.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1188760.\u003c/li\u003e\n\u003cli\u003eZhao Y, Wei K, Chi H, Xia Z, Li X: \u003cstrong\u003eIL-7: A promising adjuvant ensuring effective T cell responses and memory in combination with cancer vaccines?\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2022, \u003cstrong\u003e13:\u003c/strong\u003e1022808.\u003c/li\u003e\n\u003cli\u003eGong X, Chi H, Strohmer DF, Teichmann AT, Xia Z, Wang Q: \u003cstrong\u003eExosomes: A potential tool for immunotherapy of ovarian cancer.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2022, \u003cstrong\u003e13:\u003c/strong\u003e1089410.\u003c/li\u003e\n\u003cli\u003eXia Z, Chen S, He M, Li B, Deng Y, Yi L, Li X: \u003cstrong\u003eEditorial: Targeting metabolism to activate T cells and enhance the efficacy of checkpoint blockade immunotherapy in solid tumors.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1247178.\u003c/li\u003e\n\u003cli\u003eZhang P, Pei S, Wu L, Xia Z, Wang Q, Huang X, Li Z, Xie J, Du M, Lin H: \u003cstrong\u003eIntegrating multiple machine learning methods to construct glutamine metabolism-related signatures in lung adenocarcinoma.\u003c/strong\u003e \u003cem\u003eFront Endocrinol (Lausanne) \u003c/em\u003e2023, \u003cstrong\u003e14:\u003c/strong\u003e1196372.\u003c/li\u003e\n\u003cli\u003eDeMichele A, Martin AM, Mick R, Gor P, Wray L, Klein-Cabral M, Athanasiadis G, Colligan T, Stadtmauer E, Weber B: \u003cstrong\u003eInterleukin-6 -174G--\u0026gt;C polymorphism is associated with improved outcome in high-risk breast cancer.\u003c/strong\u003e \u003cem\u003eCancer Res \u003c/em\u003e2003, \u003cstrong\u003e63:\u003c/strong\u003e8051-8056.\u003c/li\u003e\n\u003cli\u003eStudebaker AW, Storci G, Werbeck JL, Sansone P, Sasser AK, Tavolari S, Huang T, Chan MW, Marini FC, Rosol TJ, et al: \u003cstrong\u003eFibroblasts isolated from common sites of breast cancer metastasis enhance cancer cell growth rates and invasiveness in an interleukin-6-dependent manner.\u003c/strong\u003e \u003cem\u003eCancer Res \u003c/em\u003e2008, \u003cstrong\u003e68:\u003c/strong\u003e9087-9095.\u003c/li\u003e\n\u003cli\u003eD\u0026apos;Anello L, Sansone P, Storci G, Mitrugno V, D\u0026apos;Uva G, Chieco P, Bonaf\u0026eacute; M: \u003cstrong\u003eEpigenetic control of the basal-like gene expression profile via Interleukin-6 in breast cancer cells.\u003c/strong\u003e \u003cem\u003eMol Cancer \u003c/em\u003e2010, \u003cstrong\u003e9:\u003c/strong\u003e300.\u003c/li\u003e\n\u003cli\u003eHartman ZC, Yang XY, Glass O, Lei G, Osada T, Dave SS, Morse MA, Clay TM, Lyerly HK: \u003cstrong\u003eHER2 overexpression elicits a proinflammatory IL-6 autocrine signaling loop that is critical for tumorigenesis.\u003c/strong\u003e \u003cem\u003eCancer Res \u003c/em\u003e2011, \u003cstrong\u003e71:\u003c/strong\u003e4380-4391.\u003c/li\u003e\n\u003cli\u003eDe Luca A, Lamura L, Gallo M, Maffia V, Normanno N: \u003cstrong\u003eMesenchymal stem cell-derived interleukin-6 and vascular endothelial growth factor promote breast cancer cell migration.\u003c/strong\u003e \u003cem\u003eJ Cell Biochem \u003c/em\u003e2012, \u003cstrong\u003e113:\u003c/strong\u003e3363-3370.\u003c/li\u003e\n\u003cli\u003eSiersb\u0026aelig;k R, Scabia V, Nagarajan S, Chernukhin I, Papachristou EK, Broome R, Johnston SJ, Joosten SEP, Green AR, Kumar S, et al: \u003cstrong\u003eIL6/STAT3 Signaling Hijacks Estrogen Receptor \u0026alpha; Enhancers to Drive Breast Cancer Metastasis.\u003c/strong\u003e \u003cem\u003eCancer Cell \u003c/em\u003e2020, \u003cstrong\u003e38:\u003c/strong\u003e412-423.e419.\u003c/li\u003e\n\u003cli\u003eZhang J, Peng G, Chi H, Yang J, Xie X, Song G, Tran LJ, Xia Z, Tian G: \u003cstrong\u003eCD8 + T-cell marker genes reveal different immune subtypes of oral lichen planus by integrating single-cell RNA-seq and bulk RNA-sequencing.\u003c/strong\u003e \u003cem\u003eBMC Oral Health \u003c/em\u003e2023, \u003cstrong\u003e23:\u003c/strong\u003e464.\u003c/li\u003e\n\u003cli\u003eSisirak V, Vey N, Goutagny N, Renaudineau S, Malfroy M, Thys S, Treilleux I, Labidi-Galy SI, Bachelot T, Dezutter-Dambuyant C, et al: \u003cstrong\u003eBreast cancer-derived transforming growth factor-\u0026beta; and tumor necrosis factor-\u0026alpha; compromise interferon-\u0026alpha; production by tumor-associated plasmacytoid dendritic cells.\u003c/strong\u003e \u003cem\u003eInt J Cancer \u003c/em\u003e2013, \u003cstrong\u003e133:\u003c/strong\u003e771-778.\u003c/li\u003e\n\u003cli\u003eCruceriu D, Balacescu L, Baldasici O, Gaal OI, Balacescu O, Russom A, Irimia D, Tudoran O: \u003cstrong\u003eGene expression-phenotype association study reveals the dual role of TNF-\u0026alpha;/TNFR1 signaling axis in confined breast cancer cell migration.\u003c/strong\u003e \u003cem\u003eLife Sci \u003c/em\u003e2024, \u003cstrong\u003e354:\u003c/strong\u003e122982.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Breast cancer, metastasis, cytokine, machine learning, nomogram","lastPublishedDoi":"10.21203/rs.3.rs-5841908/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5841908/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe relationship between cytokines and lung metastasis (LM) in breast cancer (BC) remains unclear and current clinical methods for identifying breast cancer lung metastasis (BCLM) lack precision, thus underscoring the need for an accurate risk prediction model. This study aimed to apply machine learning algorithms for identifying the key risk factors for BCLM before developing a reliable prediction model centered on cytokines.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis population-based retrospective study included 326 BC patients admitted to the Second Affiliated Hospital of Xuzhou Medical University between September 2018 and September 2023. After randomly assigning the patients to a training cohort (70%; n\u0026thinsp;=\u0026thinsp;228) or a validation cohort (30%; n\u0026thinsp;=\u0026thinsp;98) the risk factors for BCLM were identified using Least Absolute Shrinkage and Selection Operator (LASSO), Extreme Gradient Boosting (XGBoost) and Random Forest (RF) models. Significant risk factors were visualized with a Venn diagram and incorporated into a nomogram model, the performance of which was then evaluated according to three criteria, namely discrimination, calibration and clinical utility using calibration plots, receiver operating characteristic (ROC) curves and decision curve analysis (DCA).\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAmong the cohort, 70 patients developed LM. A nomogram was then developed to predict the 5-year and 10-year BCLM risk by incorporating five key variables, namely endocrine therapy, hsCRP, IL6, IFN-ɑ and TNF-ɑ. For the 5-year prediction model, the training and validation cohorts had AUC values of 0.786 (95% CI: 0.691\u0026ndash;0.881) and 0.627 (95% CI: 0.441\u0026ndash;0.813), respectively, while for the 10-year prediction model, the corresponding AUC values were 0.687 (95% CI: 0.528\u0026ndash;0.847) and 0.797 (95% CI: 0.605\u0026ndash;0.988), respectively. ROC analysis further confirmed the model\u0026rsquo;s strong discriminative ability, while calibration plots indicated that the predicted and observed outcomes were in good agreement in both cohorts. Finally, DCA demonstrated the model\u0026rsquo;s effectiveness in clinical practice.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eUsing machine learning algorithms, this study developed aa nomogram that could effectively identify BC patients who were at a higher risk of developing LM, thus providing a valuable tool for decision-making in clinical settings.\u003c/p\u003e","manuscriptTitle":"Development and validation of a nomogram model of lung metastasis in breast cancer based on machine learning algorithm and cytokines","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-04 11:12:46","doi":"10.21203/rs.3.rs-5841908/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Accepted","date":"2025-04-07T09:55:25+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-04T12:26:42+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-03T13:22:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"168599491070940996864801075553590217499","date":"2025-04-02T12:29:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"245760584886745977674611791392757101892","date":"2025-04-02T11:39:35+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-02T10:33:13+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-01T14:04:25+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Cancer","date":"2025-03-30T14:54:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"77e37c0c-eb23-40cf-b529-d720c88fa9a5","owner":[],"postedDate":"April 4th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-04-21T15:58:30+00:00","versionOfRecord":{"articleIdentity":"rs-5841908","link":"https://doi.org/10.1186/s12885-025-14101-3","journal":{"identity":"bmc-cancer","isVorOnly":false,"title":"BMC Cancer"},"publishedOn":"2025-04-14 15:56:56","publishedOnDateReadable":"April 14th, 2025"},"versionCreatedAt":"2025-04-04 11:12:46","video":"","vorDoi":"10.1186/s12885-025-14101-3","vorDoiUrl":"https://doi.org/10.1186/s12885-025-14101-3","workflowStages":[]},"version":"v1","identity":"rs-5841908","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5841908","identity":"rs-5841908","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}