Predict the prognosis of patients with non-small cell lung cancer based on CT radiomics and clinical pathological factors

Predict the prognosis of patients with non-small cell lung cancer based on CT radiomics and clinical pathological factors | 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 Predict the prognosis of patients with non-small cell lung cancer based on CT radiomics and clinical pathological factors Yubo Wang, Zefei Peng, Hao Hu, Zhiyong Ding, Xiandou Li, Yang Fu, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7907062/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background This study aims to investigate the use of computed tomography (CT) radiomics features combined with clinicopathological factors to establish and validate a radiomics nomogram for predicting overall survival (OS) in patients with non-small cell lung cancer (NSCLC). Methods This study included 177 patients with NSCLC, from whom CT images and clinicopathological data were collected (124 patients in the training set and 53 in the validation set). A total of 1,688 radiomics features were extracted from the volume of interest (VOI) of the tumors. Spearman correlation analysis and univariate Cox analysis were used for preliminary screening, followed by LASSO-COX regression combined with ten-fold cross-validation to further identify key radiomics features. Meanwhile, independent clinical risk factors were identified through Cox regression analysis. A nomogram was constructed based on the radiomics score (Radscore) combined with the independent clinical risk factors. The predictive performance of the model was evaluated using the C-index and calibration curves. Results Among the 177 patients with NSCLC, there were 107 males (60.45%)and 70 females (39.55%). In total, 16 key radiomics features were identified, and an OS nomogram was established based on the Radscore and clinical independent risk factors. The area under the curve(AUC) of the training and validation sets were 0.892 and 0.838, respectively. The calibration curve showed that the predicted OS values demonstrated good consistency with the actual values. Conclusion The construction of a nomogram based on CT radiomics features combined with clinicopathological factors demonstrates good efficacy in predicting OS in patients with NSCLC and can provide valuable guidance for individualized treatment strategies. Computed tomography Non-small cell lung cancer Radiomics Features Nomogram model Prognosis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 BACKGROUND Lung cancer is one of the main causes of cancer-related deaths worldwide, with a global incidence rate of about 11.4%, making it the second most common malignant tumor. Its mortality rate is about 18.0%, ranking it first among all malignant tumors [ 1 – 2 ] . Despite significant progress in the treatment of lung cancer in recent years, including a shift from traditional surgical and cytotoxic drug therapies to targeted therapies and immunotherapies, patients still face considerable challenges in prognosis, with a five-year overall survival rate of only 15% for lung cancer [ 3 – 4 ] . Therefore, identifying effective prognostic indicators for non-small cell lung cancer (NSCLC) is of great clinical significance for the early assessment of patients’ responses to treatment and for the prediction of overall survival (OS), eventually guiding personalized treatment strategies. At present, research on the prognosis of lung cancer involves multiple aspects. Clinical factors such as age, gender, smoking history, alcohol consumption history, family history, and tumor-node-metastasis (TNM) staging significantly impact lung cancer prognosis [ 5 – 8 ] . Although the TNM staging system is currently the most commonly used tumor staging system worldwide [ 9 ] , it still has certain limitations when used alone to comprehensively evaluate the prognosis of patients with NSCLC. This is primarily due to the significant differences in the prognosis of tumors at the same stage [ 10 ] . Blood biochemical indicators, such as C-reactive protein (CRP), platelets, D-dimer, fibrin degradation products (FDP), neuron-specific enolase (NSE), carcinoembryonic antigen (CEA), cancer antigen 125 (CA125), and cancer antigen 153 (CA153), have also received attention as prognostic factors for lung cancer [ 11 – 12 ] . However, the predictive ability of any single indicator for prognosis remains limited. Some scholars have proposed that treatment methods significantly impact the prognosis of lung cancer. Studies have also shown that surgical treatment is mainly focused on the early stages of lung cancer [ 13 ] , while targeted therapy and immunotherapy are mainly applied to patients with advanced lung cancer [ 14 – 16 ] . Various diagnostic and therapeutic methods have shown good results in the treatment of lung cancer, but their roles in prognosis can vary. CT imaging features of lung cancer, such as spiculation, lobulation, air bronchogram sign, and tumor density, have been proven to be independent risk factors for predicting prognosis. These features assist in providing prognostic information and personalized risk assessment [ 17 – 19 ] . However, the extraction of traditional image features is heavily influenced by subjective factors related to the operator, resulting in low data processing capability that no longer meets the demands of modern medicine [ 20 ] . Radiomics refers to the extraction and analysis of a large number of advanced quantitative imaging features from medical imaging images such as CT, PET, or magnetic resonance imaging (MRI) [ 21 – 23 ] . Compared to traditional imaging features, radiomics provides higher levels of automation, a greater number of feature dimensions, and enhanced data processing capabilities. These advantages have been widely used in the prognosis research for lung cancer. Yan et al. [ 24 ] found that machine learning models based on radiomics can accurately predict the OS of inpatients with stage IA solid NSCLC before surgery, thereby improving risk stratification. Dercle et al. [ 25 ] demonstrated that radiomics features can provide early assessments of expected OS in patients with NSCLC treated with nivolumab or chemotherapy, further supporting risk stratification in clinical trials. Yao et al. [ 26 ] proposed that radiomics features derived from 18 F-FDG PET/CT can effectively distinguish the Ki-67 expression and have strong predictive abilities for survival rates. Chen et al. [ 27 ] showed that a radiomics nomogram based on CT can predict preoperative lymphatic invasion (LVI) and OS in patients with NSCLC, aiding in the development of personalized treatment strategies before surgery. However, the limitations of the above research lie in the low AUC values obtained solely from CT radiomics features. Therefore, there is an urgent need to develop a new predictive model to predict the prognosis of patients with NSCLC. Qi et al. [ 28 ] constructed and validated a nomogram based on clinical features and MRI radiomics to predict short-term efficacy and intracranial progression-free survival (iPFS) in patients with lung adenocarcinoma brain metastases. The results showed that in the short-term efficacy model, the combined C-index of the training and validation sets (C + R) nomogram was 0.867 and 0.803, respectively, outperforming the individual clinical (C) and radiomics (R) nomograms. Wu et al. [ 29 ] randomly divided 200 patients with locally limited small-cell lung cancer (SCLC) into a training group (n = 140) and a testing group (n = 60). They extracted and screened imaging features from simulated localization CT images and constructed radiomics models, clinical models, and a combined model that included both clinical factors and radscore. Their research showed that the combined model of radscore and clinical factors could effectively predict the prognosis of patients with limited-stage SCLC, exhibiting better performance than imaging and clinical models alone. In summary, it has been found that the efficacy and accuracy of predicting the prognosis of patients with NSCLC based solely on clinicopathological features, blood biochemical indicators, imaging features, and radiomics are not satisfactory. In contrast, a comprehensive predictive model that combines radiomics and clinical features shows higher AUC values and has better predictive performance. Based on the above research, we hypothesize that a nomogram model combining CT radiomics with clinical and pathological factors may have good performance in classification and prognosis prediction. This nomogram integrates important influencing factors related to tumor prognosis through multiple regression analysis and constructs intuitive graphs using statistical prediction models to provide numerical probabilities of clinical events [ 30 – 32 ] . Therefore, the main objective of this study was to construct a predictive model for OS in the form of a nomogram based on CT radiomics features combined with clinicopathological factors. This model aims to predict prognosis and perform risk stratification, thereby improving the assessment of prognosis for patients with NSCLC and providing guidance for advancing personalized treatment and precision medicine. METHODS Patients and clinicopathological data This retrospective study included a total of 216 patients who underwent chest CT scans at the First People's Hospital of Kunming from February 2019 to September 2023 and were diagnosed with NSCLC through pathology from the PACS system. This study was approved by the Ethics Committee of the First People's Hospital of Kunming(YLS2023-75), and patients' informed consent was waived. The inclusion criteria were as follows: (1) Diagnosis of NSCLC confirmed through puncture pathology; (2) Completion of a chest CT scan within one month before treatment, with complete imaging data available; (3) No prior needle therapy received for anti-tumor treatment before the CT examination. The exclusion criteria were: (1) Receipt of anti-tumor treatment before chest CT examination; (2) Incomplete clinical data or missing/poor-quality images; (3) Presence of concurrent malignant tumors in other organs; (4) Patients with missing follow-up data. Based on the inclusion and exclusion criteria, a total of 39 patients who did not meet the requirements were excluded from this study. The final cohort comprised 177 patients, including 107 males and 70 females. Among these, there were 139 survivors and 38 patients who had died. There were 52 cases of squamous cell carcinoma and 125 cases of adenocarcinoma; 80 patients received surgical treatment, 99 patients received chemotherapy, and 49 patients received targeted therapy; 32 cases received radiotherapy and 19 cases received immunotherapy. The patients were randomly divided into a training set (n = 124) and a validation set (n = 53) in a 7:3 ratio. Figure 1 shows the standard flow chart for patient inclusion and exclusion. Image acquisition and reconstruction parameters All patients underwent multi-slice spiral CT scans (SOMATOM Definition AS 128, Siemens Healthineers; Perform chest CT plain scan using Brilliance iCT 256, Philips Healthcare. Before scanning, patients are required to remove any metal objects from their chest, such as metal zippers on clothing and any metallic objects on their undergarments. During the examination, the patient lies supine on the examination bed, with both arms raised, and is calm throughout the process. As the scan begins, the patient's head is advanced, and they are instructed to hold their breath after taking a deep breath in accordance with the machine’s instructions. The scanning range extends from the chest entrance to the level of the lung base. The scanning parameters were as follows: tube voltage of 120 kVp, tube current of 110/113mAs, FOV of 350 mm × 400 mm, matrix of 512 × 512, layer thickness of 1 mm, layer spacing of 1 mm, collimator width of 128 × 0.6 mm, and tube rotation time of 0.5 s. On completion of the scan, the original images are reconstructed using standard techniques. The data are uploaded to the PACS system through a 1 mm thin layer and multi-plane reconstruction, and the patient's CT images are downloaded and saved in DICOM format. Image preprocessing and segmentation The linear interpolation algorithm was used to resample all images to isotropic voxels with a size of 1 mm × 1 mm × 1 mm, using the B-spline interpolation algorithm for the resampling process. Using the Huiyi Huiying Data Artificial Intelligence Research Platform( https://mics.huiyihuiying.com/#/ ), a radiologist with three years of experience, Zefei Peng, manually delineated the region of interest (ROI) of the tumor on a lung window with a thickness of 1 mm (window width: 1600HU, window position: -550HU). The entire tumor lesion area was outlined layer by layer. Once the delineation was completed, a radiology attending physician with 10 years of experience, Yang Fu, reviewed and corrected the delineated ROI. Feature extraction and selection As shown in Fig. 2 , in the process of feature extraction, we extracted histogram features, morphological features, high-order texture features, clinical features, etc, respectively. A total of 1,688 radiomics features were extracted from the volume of interest of tumors in CT images, mainly divided into the following categories: (1) First-order statistical features: This category includes 324 local binary patterns (LBP) and wavelet filtering features that reflect the symmetry, uniformity, and local intensity distribution changes of the measured voxels. Key metrics include median, mean, minimum, maximum, standard deviation, skewness, and kurtosis. Kurtosis is used to quantify the steepness of pixel intensity distribution in tumor images, reflecting the heterogeneity and microstructure complexity of tumor tissues.(2) Shape features: A total of 14 morphological features quantitatively describe the geometric characteristics of the region of interest, such as the surface area, volume, surface area-to-volume ratio, sphericity, compactness, and three-dimensional diameter of the tumor. These features can describe the three-dimensional size and morphological information of the tumor. (3) Gray level co-occurrence matrix (GLCM) features: This category includes 432 features derived from LBP and wavelet, serving as a statistical tool to describe image texture. Key indicators of GLCM features, such as contrast, correlation, energy (or inverse entropy), homogeneity (or contrast), and entropy, provide different information about image texture and can be used for applications such as image classification, retrieval, and segmentation. The contrast reflects the sharp edge of the tumor, and high contrast indicates irregular boundary or invasive growth of the tumor; correlation is used to measure the consistency of tumor internal texture. Low correlation may correspond to high malignancy (such as GLCM characteristics of poorly differentiated squamous cell carcinoma); entropy represents the disorder of tumor tissue, and high entropy is common in necrotic or highly heterogeneous regions (the entropy of metastatic lymph nodes is significantly higher than that of benign lesions). (4) Neighborhood Gray Tone Difference Matrix (NGTDM) Features: A total of 90 NGTDM features are included, mainly used to quantify the grayscale differences between pixels or voxels and their adjacent counterparts within a predefined distance. The main characteristics of NGTDM include roughness, busyness, contrast, complexity, and texture intensity. (5) Gray Level Run Length Matrix (GLRLM) Features: This category has a total of 288 features, which serves as a statistical measure to describe the number of consecutive occurrences of gray levels in a specified direction in an image. These features assist in analyzing the texture and structural information of the image. (6) Gray Level Size Zone Matrix (GLSZM): This matrix consists of 288 units and is a statistical tool used to describe the size of texture regions in an image. (7) Gray Level Dependence Matrix (GLDM) Features: A total of 252 features are included, which describe the dependency relationship between grayscale values in images, and can aid in identifying and quantifying heterogeneity within tissues. Using the intra-class correlation coefficient (ICC) features with an ICC > 0.9 were selected, while those that are highly correlated with other features were removed. Before performing feature selection, all radiomics features were standardized by removing the mean and dividing by standard deviation, resulting in a mean of zero and a variance of one for each set of feature values. Owing to the large number of radiomics features and the aim to mitigate issues related to model overfitting and multicollinearity, two methods were used to reduce dimensionality: Firstly, Spearman correlation analysis was used to eliminate features with a correlation of 0.9 or higher. Secondly, single factor Cox analysis was used to retain features with p < 0.05, leaving 137 features. Finally, COX-LASSO and ten fold cross validation were used to further determine key radiomics features, select the indicators most relevant to the research objectives, and obtain the weights of these indicators. The Radscore for each patient was calculated using the following formula: (radscore = 0.230 * Kurtosis_firstorder_original + 0.068 * Mean_firstorder_original + 0.154 * JointAverage_glcm_original+-0.022 * DifferentiatedEntropy_glcm_original + 0.200 * ClusterProminence_glcm_original + 0.130 * DifferentiatedAverage_glcm_original + 0.310 * LargeDependenciesHighGrayLevelEmphasis_gldm_original + 0.199 * LongRunLowGrayLevelEmphasis_glrlm_original + 0.178 * RootMeanSquare_firstorder_logarithm + 0.021 * JoineEntropy_glcm_logarithm + 0.300 * JointeEnergy_glcm_logarithm+-0.008 * Idm_glcm_logarithm + 0.016 * DifferentiatedAverage_glcm_logarithm-0.025 * LargeDependenceEmphasis_gldm_logarithm-0.065 * LargeDependenceLowGrayLevelEmphasis_gldm_logarithm+-0.332 * SmallDependenceLowGrayLevelEmphasis_gldm_logarithm). The Radscore serves as a comprehensive reflection of the radiomics features and will be included in the subsequent model construction. Establishment of predictive models For the clinical pathological factors of patients in the training set, significant variables were first selected through univariate analysis. Subsequently, multivariate analysis was conducted to identify independent risk factors that are significantly correlated with prognosis, thereby establishing a clinical model. Simultaneously, radiomic features were screened, and key radiomics features were extracted to establish radiomics labels. Based on these Radscore, a radiomics model was developed. By combining the Radscore with the clinically independent risk factors, a joint prediction model was established using multiple regression analysis. A nomogram was constructed based on multiple logistic regression, and the receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of the prediction model. The AUC value was used as a quantitative indicator for evaluating the predictive performance of the model. According to the model’s formula, scores were calculated for each patient, allowing for their classification into high-risk and low-risk groups based on these scores. C-index was used to evaluate the predictive performance of the model, while the Kaplan-Meier (KM) method was used to estimate the median survival and survival rate of patients. Figure 2 shows the workflow of the radiomics analysis. Establishment and verification of radiomics nomogram A nomogram was established based on multivariate logistic regression analysis by combining clinically independent risk factors and Radscore (Fig. 3 ). The accuracy of the model’s predictions was evaluated using calibration curves, where the horizontal axis represents the predicted probability and the vertical axis represents the actual probability. The reference curve, which represents an ideal scenario where the predicted and actual probabilities are equal, serves as a benchmark; therefore, the closer the calibration curve is to the reference curve, the better the model’s consistency. Decision curve analysis (DCA) is used to evaluate the clinical practicability of the prediction model, also known as the evaluation of net benefit. DCA can integrate the preferences of decision makers and patients into the analysis. The vertical axis is the value of net benefit and the horizontal axis is the threshold probability. Under the same threshold probability, the higher the net benefit, the better the clinical efficacy of the model. Statistical analysis Statistical analysis was conducted using R language software (version 4.3.1; https://www.r-project.org ). Radiomics features were subjected to statistical analysis, with continuous data that conform to a normal distribution typically represented as mean ± standard deviation (mean ± SD). For skewed data distributions, the median and interquartile range (IQR) were used for description. Categorical variables were represented by the number of cases and the percentage [n(%)]. Depending on the distribution of data, either the t-test or the Mann-Whitney U test was used for inferential comparison. The statistical significance of categorical variables was assessed using the χ 2 test or Fisher's exact test, with p < 0.05 indicating statistically significant differences. The performance of each prediction model was quantified using the ROC curve and the AUC. Proportional Hazards (PH) is the core assumption of the Cox proportional hazards model in survival analysis, which refers to the risk ratios at different groups or covariate levels remaining constant throughout the entire follow-up period and not changing over time. We confirmed the validity of the PH hypothesis through Schoenfeld residual test (p > 0.05), and the model results were robust. A Cox regression model was established to analyze survival time, using hazard ratio (HR) values to establish the relationship between survival time and various variables. Meanwhile, C-index was used to evaluate the accuracy of the nomogram model. The KM method was used to estimate median survival and to calculate survival curves, with comparisons of survival curves (survival rates) conducted using a log-rank test. RESULTS Clinicopathological characteristics This study included a total of 177 patients with NSCLC, among whom 38 (21.47%) had died and 139 (78.53%) survived. The cohort comprised 107 males (60.45%) and 70 females (39.55%) (Table 1 ). All variables were included in the Cox regression analysis, and univariate analysis was performed to select the following factors for multivariate analysis: tumor size, lymph node metastasis, distant metastasis, tumor status, T stage, N stage, M stage, platelets, D-dimer, FDP, FIB, CA125, CRP, CA153, Radscore, and surgical treatment (Table 2.1 ). The results of the multivariate analysis showed that CRP (HR = 1.010, 95% CI: 1.004–1.019), Radscore (HR = 2.960, 95% CI: 1.679–5.225), and surgical treatment (HR = 0.020, 95% CI: 0.054–0.733) were independent risk factors significantly affecting the prognosis of NSCLC (P < 0.05) (Table 2.2 ). Both CRP and the Radscore, along with surgical treatment, showed significant correlations with OS in patients with NSCLC (P < 0.05). Table 1 Clinical, pathological, and imaging characteristics of training and validation sets of non-small cell lung cancer patients Characteristic Training set Validation set Survival Death Survival Death p -Value N = 95 N = 29 N = 44 N = 9 Brain-metastases 0.028 0 75 (78.9%) 14 (48.3%) 33 (75.0%) 7 (77.8%) 1 20 (21.1%) 15 (51.7%) 11 (25.0%) 2 (22.2%) Gender: 0.804 Male 33 (34.7%) 13 (44.8%) 20 (45.5%) 4 (44.4%) Female 62 (65.3%) 16 (55.2%) 24 (54.5%) 5 (55.6%) Age 60.0 [51.0;67.0] 62.0 [58.0;69.0] 58.0 [49.8;63.0] 57.0 [54.0;68.0] 0.122 Smoking: 0.768 0 55 (57.9%) 17 (58.6%) 26 (59.1%) 4 (44.4%) 1 40 (42.1%) 12 (41.4%) 18 (40.9%) 5 (55.6%) Drinking: 0.879 0 69 (72.6%) 22 (75.9%) 30 (68.2%) 6 (66.7%) 1 26 (27.4%) 7 (24.1%) 14 (31.8%) 3 (33.3%) Family: 0.974 0 91 (95.8%) 28 (96.6%) 42 (95.5%) 9 (100%) 1 4 (4.21%) 1 (3.45%) 2 (4.55%) 0 (0.00%) Location: 0.94 1 31 (32.6%) 7 (24.1%) 20 (45.5%) 1 (11.1%) 2 9 (9.47%) 3 (10.3%) 1 (2.27%) 0 (0.00%) 3 19 (20.0%) 7 (24.1%) 9 (20.5%) 3 (33.3%) 4 21 (22.1%) 7 (24.1%) 12 (27.3%) 3 (33.3%) 5 15 (15.8%) 5 (17.2%) 2 (4.55%) 2 (22.2%) Empty: 0.127 0 88 (92.6%) 25 (86.2%) 40 (90.9%) 9 (100%) 1 7 (7.37%) 4 (13.8%) 4 (9.09%) 0 (0.00%) Lobation: 0.586 0 65 (68.4%) 18 (62.1%) 31 (70.5%) 6 (66.7%) 1 30 (31.6%) 11 (37.9%) 13 (29.5%) 3 (33.3%) Burr : 0.645 0 68 (71.6%) 21 (72.4%) 29 (65.9%) 6 (66.7%) 1 27 (28.4%) 8 (27.6%) 15 (34.1%) 3 (33.3%) Pleural.traction: 0.119 0 59 (62.1%) 22 (75.9%) 28 (63.6%) 7 (77.8%) 1 36 (37.9%) 7 (24.1%) 16 (36.4%) 2 (22.2%) LNM: 0.004 0 47 (49.5%) 6 (20.7%) 20 (45.5%) 4 (44.4%) 1 48 (50.5%) 23 (79.3%) 24 (54.5%) 5 (55.6%) Metastasis: < 0.001 0 78 (82.1%) 15 (51.7%) 32 (72.7%) 5 (55.6%) 1 17 (17.9%) 14 (48.3%) 12 (27.3%) 4 (44.4%) Type: 0.212 Squamous cell carcinoma 28 (29.5%) 10 (34.5%) 11 (25.0%) 3 (33.3%) adenocarcinoma 67 (70.5%) 19 (65.5%) 33 (75.0%) 6 (66.7%) Stage: < 0.001 1 38 (40.0%) 1 (3.45%) 14 (31.8%) 1 (11.1%) 2 7 (7.37%) 4 (13.8%) 7 (15.9%) 0 (0.00%) 3 30 (31.6%) 7 (24.1%) 10 (22.7%) 4 (44.4%) 4 20 (21.1%) 17 (58.6%) 13 (29.5%) 4 (44.4%) T: 0.034 1 37 (38.9%) 4 (13.8%) 20 (45.5%) 0 (0.00%) 2 22 (23.2%) 10 (34.5%) 10 (22.7%) 4 (44.4%) 3 12 (12.6%) 6 (20.7%) 5 (11.4%) 2 (22.2%) 4 24 (25.3%) 9 (31.0%) 9 (20.5%) 3 (33.3%) N: 0.009 0 45 (47.4%) 5 (17.2%) 17 (38.6%) 2 (22.2%) 1 5 (5.26%) 3 (10.3%) 2 (4.55%) 0 (0.00%) 2 27 (28.4%) 13 (44.8%) 12 (27.3%) 3 (33.3%) 3 17 (17.9%) 8 (27.6%) 13 (29.5%) 4 (44.4%) 4 1 (1.05%) 0 (0.00%) M: < 0.001 0 76 (80.0%) 13 (44.8%) 32 (72.7%) 5 (55.6%) 1 19 (20.0%) 16 (55.2%) 12 (27.3%) 4 (44.4%) Operation < 0.001 0 42 (44.2%) 25 (86.2%) 24 (54.5%) 7 (77.8%) 1 53 (55.8%) 4 (13.8%) 20 (45.5%) 2 (22.2%) Surgery.for.brain.metastases: 0.435 0 93 (97.9%) 26 (89.7%) 43 (97.7%) 9 (100%) 1 2 (2.11%) 3 (10.3%) 1 (2.27%) 0 (0.00%) Chemotherapy: 0.963 0 43 (45.3%) 12 (41.4%) 19 (43.2%) 2 (22.2%) 1 52 (54.7%) 17 (58.6%) 25 (56.8%) 7 (77.8%) Targeted.therapy : 0.526 0 72 (75.8%) 19 (65.5%) 30 (68.2%) 7 (77.8%) 1 23 (24.2%) 10 (34.5%) 14 (31.8%) 2 (22.2%) Radiotherapy: 0.535 0 76 (80.0%) 23 (79.3%) 37 (84.1%) 9 (100%) 1 19 (20.0%) 6 (20.7%) 7 (15.9%) 0 (0.00%) Immunotherapy: 0.447 0 84 (88.4%) 27 (93.1%) 39 (88.6%) 8 (88.9%) 1 11 (11.6%) 2 (6.90%) 5 (11.4%) 1 (11.1%) *0 represents none, 1 represents present. Table 2.1 Univariate analysis of survival related risk factors in non-small cell lung cancer patients Characteristic HR-value 95%C-index P -value Brain-metastases* Gender Age Smoking Drinking Family Number Size Location Lung.cancer.size* Empty Lobation Burr Pleural.traction LNM* Metastasis* Type* Stage T* N* M* PLT* Leukocyte Neutrophils Monocyte Lymphocyte C.reactive.protein* D.Dimer* FDP* FIB* NSE CEA CA199 CA125* CA153* SCC.Ag CYFRA21.1 ProGRP Operation* Surgery.for.brain.metastases Chemotherapy Targeted.therapy Radiotherapy Immunotherapy Radscore* 2.245 0.906 1.027 1.113 0.934 1.045 1.048 1.211 1.066 1.297 2.250 1.228 0.822 0.513 3.449 4.936 0.613 2.474 1.351 1.648 4.744 1.004 1.104 0.997 0.907 0.970 1.014 1.138 1.030 1.366 1.0005 1.00006 0.991 1.001 1.016 0.688 0.995 1.008 0.124 1.613 0.977 1.284 0.745 0.572 5.045 (1.075, 4.687) (0.432, 1.900) (0.992, 1.062) (0.529, 2.338) (0.398, 2.190) (0.141, 7.728) (0.826, 1.330) (0.926,1.585) (0.831, 1.366) (1.119, 1.504) (0.773, 6.552) (0.580, 2.602) (0.363,1.858) (0.219, 1.204) (1.394, 8.531) (2.260, 10.779) (0.281,1.335) (1.590, 3.848) (1.004, 1.819) (1.187, 2.287) (2.213, 10.172) (1.0008,1.009) (0.998,1.221) (0.970,1.024) (0.795,1.035) (0.940,1.002) (1.006,1.021) (1.025,1.263) (1.009,1.051) (1.053,1.772) (0.967,1.034) (0.999,1.0006) (0.973,1.010) (1.0004,1.002) (1.007,1.025) (0.373,1.268) (0.962,1.029) (0.986,1.031) (0.042,0.362) (0.472,5.507) (0.461,2.068) (0.596,2.768) (0.298,1.862) (0.134,2.435) (3.109,8.187) 0.032 0.795 0.113 0.777 0.874 0.965 0.697 0.194 0.612 <0.001 0.175 0.593 0.633 0.106 0.003 <0.001 <0.001 0.121 0.046 0.0018 <0.001 0.025 0.075 0.850 0.156 0.066 <0.001 0.039 0.032 0.023 0.976 0.829 0.246 0.032 0.001 0.111 0.778 0.481 <0.001 0.471 0.952 0.527 0.517 0.414 <0.001 *LNM=lymph node metastasis;PLT=platelet;FDP=Fibrin Degradation Products；FIB=Fibrinogen；NSE=Neuron-Specific Enolase；CEA=Carcinoembryonic antigen Table 2.2 Multivariate analysis of risk factors related to overall survival in non-small cell lung cancer Characteristic HR-value 95%C-index P -value reactive.protein Operation Radscore 1.010 0.200 2.960 (1.004, 1.019) (0.054, 0.733) (1.679, 5.225) 0.002 0.015 <0.001 In the high-risk group, the median survival time was 11 months (range: 8–25 months), with postoperative survival rates at 1, 2, and 3 years of 74.87%, 57.55%, and 40.10%, respectively. In contrast, the low-risk group had a median survival period of 30 months (range: 10–38 months), with postoperative survival rates at 1, 2, and 3 years of 97.70%, 95.23%, and 92.59%, respectively (Table 3 ). Table 3 Survival Probability and Median Survival Time Grouping One year survival probability Two year survival probability Three year survival probability median survival High-risk group Low-risk group 0.748(0.645–0.868) 0.977(0.933-1.000) 0.575(0.443–0.746) 0.952(0.889-1.000) 0.401(0.232–0.691) 0.925(0.848-1.000) 11(8-25.2) 30(10–38) Screening results of radiomics features A total of 1,688 radiomics features were extracted from the VOI of tumors in the CT images. Spearman correlation analysis was used to remove features with a correlation coefficient above 0.9. Following univariate Cox analysis features with p < 0.05 were retained, resulting in 137 features. Thereafter, COX-LASSO regression and tenfold cross-validation were used for feature selection, leading to the identification of 16 key radiomics features (Table 4 ). Table 4 Key radiomics features and their correlation coefficients Radiomics features Coefficient Kurtosis_firstorder_original 0.23 Mean_firstorder_original 0.068 JointAverage_glcm_original 0.154 DifferenceEntropy_glcm_original -0.022 ClusterProminence_glcm_original 0.2 DifferenceAverage_glcm_original 0.13 LargeDependenceHighGrayLevelEmphasis_gldm_original 0.31 LongRunLowGrayLevelEmphasis_glrlm_original 0.199 RootMeanSquared_firstorder_logarithm 0.178 JointEntropy_glcm_logarithm 0.021 JointEnergy_glcm_logarithm 0.3 Idm_glcm_logarithm -0.008 DifferenceAverage_glcm_logarithm 0.016 LargeDependenceEmphasis_gldm_logarithm -0.025 LargeDependenceLowGrayLevelEmphasis_gldm_logarithm -0.065 SmallDependenceLowGrayLevelEmphasis_gldm_logarithm -0.332 Predictive model efficiency KM curves were generated based on three independent risk factors (CRP, Radscore and surgical treatment) (Fig. 4 ). The KM curve graph showed that, regardless of whether in the training set or the validation set, the survival rates of patients in the low-risk group at 12, 24, and 36 months after surgery were significantly higher than those in the high-risk group (p < 0.05). Notably, the survival rates of patients with low risk at 12, 24,and 36-months post-surgery were close to 1. Over time, the number of surviving patients in both the high-risk and low-risk groups decreased. Based on multivariate Cox regression analysis, clinical models, radiomics models and combined models were constructed separately. The AUC of the clinical model in the training set and validation set were 0.858 (95% CI: 0.789–0.927) and 0.775 (95% CI: 0.614–0.935), respectively. The AUC values of the radiomics model in the training and validation sets were 0.859 (95% CI: 0.805–0.914) and 0.778 (95% CI: 0.67–0.887), respectively. The AUC values of the combined model in the training and validation sets were 0.892 (95% CI: 0.848–0.936) and 0.838 (95% CI: 0.71–0.965), respectively. The AUC values of both the radiomics model and the combined model were higher than those of the clinical model, with the combined model demonstrating the highest AUC values in both sets. Indicating that the predictive performance of the combined model is significantly superior to that of individual clinical and radiomics models. As shown in Figs. 5 A and B, the calibration curve demonstrates a good fit, indicating good consistency between the predicted and actual values of the prediction model, as well as high predictive accuracy. The final research results showed that in both the training and validation sets, the comprehensive prediction models exhibited high AUC values for predicting patients' 1, 2, and 3 year survival rates, demonstrating good predictive performance, as shown in Figs. 6 A and B. Since this is a small data set, we use bootstrapping to test the robustness and stability of features. Through 200 bootstrapping samples, the average C-index is 0.869/0.705 (the effectiveness of model prediction at 36 months, training/validation set). Evaluation of clinical utility The clinical utility of radiomics model, clinical model and combined model was evaluated using a DCA, and the decision curve analysis results of these models in the training and validation sets are provided in Figs. 7 A and 7 B, respectively. DCA revealed that all three prediction models were clinically useful in both sets and the combined model had high overall net benefits within a reasonable threshold probability range. DISCUSSION The aim of this study is to establish a nomogram model based on CT radiomics combined with clinicopathological factors to predict the survival rates of patients with NSCLC at 1, 2, and 3 years. Additionally, the study evaluates the survival rate of high-risk and low-risk groups using KM curves. By combining CT radiomics features with clinicopathological factors, univariate and multivariate Cox regression analyses were conducted on each variable to identify independent risk factors that affect patient survival. First, Cox regression was used to establish a clinical model based on surgical treatment and CRP. Radiomics features were then incorporated into this clinical model to establish a combined model. According to the median value of the Radscore, patients were divided into high-risk and low-risk groups. The column chart model established through three independent risk factors effectively visualized the prediction results and independent risk factors. The Cox regression analysis of the comprehensive model showed a C-index of 0.892 (range: 0.848–0.936) for the training set and 0.838 (range 0.710–0.965) for the validation set, demonstrating the good predictive performance of our model. The median survival time of high-risk patients with NSCLC was 11 months (range: 8–25), with 1, 2, and 3 year survival rates of 74.87%, 57.55%, and 40.10%, respectively. In contrast, the median survival period of low-risk patients was 30 months (range: 10–38), with 1, 2, and 3 year survival rates of 97.70%, 95.23%, and 92.59%, respectively. These findings indicate that the 1, 2, and 3year survival rates of low-risk patients are significantly higher. The KM curve analysis further supports this, showing that in both the training and validation sets, the survival rates of patients with NSCLC low-risk group at 1, 2, and 3 years are close to 100%. The predictive model established in this study provides valuable guidance for risk stratification and personalized diagnosis and treatment of patients with NSCLC, providing a basis for additional treatment options and a close follow-up basis for patients at high risk. In addition, this study conducted univariate and multivariate Cox regression analyses on clinicopathological factors, and the results showed that CRP (HR = 1.010, 95% CI: 1.004–1.019), Radscore (HR = 2.960, 95% CI: 1.679–5.225), and surgical treatment (HR = 0.020, 95% CI: 0.054–0.733) were independent risk factors affecting the prognosis of NSCLC. Previous studies have also shown that multiple factors can affect the survival of patients with NSCLC [ 33 – 39 ] . For example, Guo et al. [ 33 ] found that monitoring D-Dimer levels and addressing blood hypercoagulability in patients with NSCLC across different TNM stages can aid in early prevention and treatment, thereby enhancing patient survival rates. CRP is a common acute inflammatory response protein synthesized by liver cells and released into the bloodstream. Its relationship with malignant tumors is not fully understood [ 34 ] . Buresova et al. [ 35 ] found that CRP may affect the level of circulating tumor DNA (ctDNA) in NSCLC, thereby affecting the prognosis of patients through an investigation of the potential association between tumor markers, laboratory parameters, ctDNA, and survival rate. Wang et al. [ 36 ] conducted a meta-analysis on the correlation between serum CRP levels and NSCLC prognosis and discovered that patients with elevated serum CRP levels had lower five-year survival rates. Frey et al. [ 37 ] studied 52 patients with primary NSCLC (UICC stage III) and found that a high baseline CRP to albumin ratio was associated with shorter progression-free survival (PFS) (p = 0.038) and OS (p = 0.022). This is consistent with the findings of Nassar [ 38 ] , who found that the CRP concentration during the third treatment cycle and the difference in CRP concentration between the third and second treatment cycles were the strongest predictors of PFS and OS. Measuring longitudinal CRP can monitor inflammation levels and serve as a promising prognostic marker, particularly as inflammation levels decrease during treatment cycles. This finding is consistent with Zheng’s study [ 39 ] , which noted that an increase in baseline CRP levels is significantly correlated with decreased PFS, especially in patients receiving chemotherapy combined with immune checkpoint inhibitors (ICI) treatment. While the predictive value of CRP is prominent, it may vary depending on the treatment regimen. Our research supports the potential of CRP as a broad prognostic biomarker, providing new insights into the complex interactions between inflammatory markers and cancer treatment responses and emphasizing the potential of CRP as a predictive and prognostic biomarker for NSCLC. In addition, these findings further highlight the necessity of exploring other variables that affect OS and validating these results in a broader patient population. Radiomics can be used to comprehensively and quantitatively evaluate the spatiotemporal heterogeneity of tumors, and when combined with clinicopathological factors, its predictive performance for prognosis may be improved [ 40 – 41 ] . Zheng et al. [ 42 ] showed that CT-based deep learning models exhibit good performance in predicting OS in patients with NSCLC. In terms of treatment, radiomics can non-invasively evaluate EGFR mutation status [ 43 ] , support targeted therapy, and predict microvascular invasion in patients with stage I NSCLC before surgery [ 44 ] , aiding in surgical method selection and individualized treatment planning. The study by Yang et al. [ 10 ] confirmed that a line chart based on 18 F-FDG PET/CT Radscore and clinicopathological factors has good predictive power for prognosis, effectively guiding individualized treatment for patients with NSCLC. This finding is highly consistent with our results. The study by González et al. [ 45 ] showed that patients who underwent surgery had significantly longer survival times (p < 0.001), also consistent with our findings. In the training set, 53 out of 95 patients who did not undergo surgery died, with a mortality rate of 55.8%; among the 29 patients who underwent surgery, four died, with a mortality rate of 13.8%. In the validation set, 20 out of 44 patients who did not undergo surgery died, with a mortality rate of 45.5%; among the nine patients who underwent surgery, two died, with a mortality rate of 22.2%. Therefore, surgical treatment is considered an independent protective factor for patients with NSCLC. Nakwan et al. [ 46 ] conducted a study on the survival factors among patients with cancer in a non-university hospital in Thailand from 2012 to 2021, confirming that surgical resection is an important predictor of patients with cancer survival. At present, ICI treatment has greatly changed the treatment prospects for NSCLC [ 47 ] . The US Food and Drug Administration (FDA) approved nivolumab and pembrolizumab for the treatment of lung squamous cell carcinoma in 2015. In October 2016, the FDA authorized pembrolizumab as a first-line treatment for patients with PD-L1 overexpression (≥ 50%); atezolizumab has also been approved for patients with advanced NSCLC who have progressed after chemotherapy [ 48 ] . Although immunotherapy (HR = 0.572, 95% CI: 0.134–2.435, p = 0.414) did not show statistical significance in our study, it does not mean that immunotherapy lacks value in NSCLC prognosis. At present, the application of immunotherapy in NSCLC primarily focuses on patients at the late stage and has shown good results. Lv et al. [ 49 ] have shown that immunotherapy can reshape tumor microenvironment to inhibit tumor cells, block CTLA-4 and PD-1/PD-L1 immune checkpoints to alleviate T cell functional inhibition, and facilitate tumor clearance through the body's immune system. Its application in patients with late-stage NSCLC has broad clinical prospects. Finally, we constructed a nomogram by combining the Radscore with clinicopathological factors. The calibration curve showed that the predicted survival probability closely aligns with the actual survival time of the patients, thereby validating the accuracy of the model. The validation of the OS nomogram demonstrated that as follow-up time increases, the AUC for predicting prognosis also gradually increases, further demonstrating the good predictive performance of the OS nomogram. Additionally, we used KM analysis to evaluate the reliability of the OS nomogram in predicting patient survival rates. The results of the KM analysis showed that the OS nomogram effectively distinguishes between patients at high and low risk, indicating its efficacy in predicting patients at high and low risk. Therefore, this nomogram is considered a robust and reliable model that can serve as strong evidence for additional treatment and close follow-up for patients with poor prognoses. A nomogram model based on CT radiomics characteristics and clinical pathological factors was constructed to achieve accurate risk stratification of NSCLC patients by integrating multi-dimensional data such as imaging score, surgical treatment and C-reactive protein. The model can guide the formulation of individualized treatment strategies: the high-risk group (such as stage Ⅲ with lymph node metastasis) is recommended to receive intensive treatment (neoadjuvant immunotherapy combined with chemotherapy), and the low-risk group (such as stage Ⅰ a well differentiated adenocarcinoma) is suitable for step-down treatment (surgery or targeted therapy). Through the accurate prediction of the 1, 2, and 3 year survival rates of patients with non-small cell lung cancer, dynamic monitoring of curative effect is achieved to adjust the optimization scheme, and its visual scoring system significantly improves the treatment accuracy and clinical net benefit [ 50 ] . Compared with previous studies, this study has the following advantages. First, this study extracts radiomics features based on CT and combines them with clinical pathological factors. Multimodal receipts help to improve the accuracy of the prediction model. Secondly, the nomogram model constructed by integrating multi-dimensional variables achieved a high AUC value in the prognosis prediction of NSCLC (AUC = 0.892 in the training set and AUC = 0.838 in the validation set), and achieved individualized risk stratification through the visual scoring system, which has the advantages of high accuracy, interpretability and clinical practicability [ 51 – 52 ] . Meanwhile, our study has several limitations. First, it is a retrospective study, and there may be selection bias. Second, the follow-up period is relatively short, and some outcome events have not yet occurred, which may result in loss-to-follow-up bias. Third, as a single-center study, the generalizability of the results may be limited. In future research, we will further expand the sample size and conduct multicenter and prospective studies to explore the application of artificial intelligence in predicting the prognosis of NSCLC. This approach will help provide more robust support for improving the quality of life of patients with NSCLC. CONCLUSION The nomogram constructed by combining CT radiomics features with clinicopathological factors shows good predictive efficacy and application value in predicting the prognosis of patients with NSCLC. This model can effectively evaluate the prognosis of patients and provide important guidance for improving the quality of life of patients with NSCLC. In addition, the significantly higher survival rates at 1, 2, and 3 years post-surgery underscore the positive impact of surgical treatment on patient prognosis. Abbreviations CT：Computed tomography；NSCLC：non-small-cell lung cancer；VOI：volume of interest；OS：Overall survival；CRP： C-reactive protein； ICI ： immune-checkpoint-inhibitor ； PFS：progression-free survival；TNM：Tumor-Node-Metastasis；MTV：Metabolic Tumor Volume；TLG：Total Lesion Glycolysis；LVI：lymphovascular invasion；AUC：area under curve；iPFS：intracranial progression free survival；KM：Kaplan-Meier；HR：Hazard ratio；ICC：Intra-class Correlation Coefficient；Radscore：radiomics score; TNM: tumor-node-metastasis; FDP: degradation products; NSE:neuron-specific enolase; CEA:carcinoembryonic antigen; CA125:cancer antigen 125; CA153:cancer antigen 153;MRI：magnetic resonance imaging；GLCM：Gray level co-occurrence matrix；NGTDM：Neighborhood Gray Tone Difference Matrix； GLRLM：Gray Level Run Length Matrix； GLSZM：Gray Level Size Zone Matrix ；GLDM：Gray Level Dependence Matrix；DCA：Decision curve analysis；ROC：receiver operating characteristic；ROI：region of interest；PH：Proportional Hazards；ctDNA：circulating tumor DNA Declarations Ethics approval and consent to participate This study was approved by the Ethics Committee of the First People's Hospital of Kunming(YLS2023-75), and patients' informed consent was waived. All methods were carried out in accordance with relevant guidelines and regulations. Consent for publication Patients signed informed consent regarding publishing their clinical data and images. Availability of data and materials The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This study was supported by National Natural Science Foundation of China Project (No. 82160348)；Yunnan Province Major Special Plan (No. 202302AA310018-D-8); Yunnan Province's "Xingdian Talent Support Program" Youth Talent Project (No. XDYC-QNRC-2022-0608)；Beijing Medical Award Foundation Ruiying Fund（No. YXJL-2022-0665-0216）;Yunnan Provincial Department of Education Science Research Fund Project(No. 2025Y0383). Author contributions Wang Yubo conceived the idea for this research. Wang Yubo, Peng Zefei, and Hu Hao were responsible for data collection. Ding Zhiyong, Li Xiandou，Li Jiageng and Fu Yang performed the image analysis. Wang Yubo wrote the manuscript, while Zhang Yisong、Xie bosen and Kong Mengxue conducted statistical analysis. Dr. Yang Bin edited and reviewed the manuscript. All authors discussed the results and provided comments on the manuscript. All authors contributed to this article and approved the final submitted version. References Li C, Lei S, Ding L, et al. Global burden and trends of lung cancer incidence and mortality. Chin Med J. 2023;136:1583–90. 10.1097/CM9.0000000000002529 . Ettinger DS, Wood DE, Aisner DL, et al. Non-Small Cell Lung Cancer, Version 3.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2022;20:497–530. 10.6004/jnccn.2022.0025 . Schabath MB, Cote ML. Cancer Progress and Priorities: Lung Cancer. Cancer Epidemiol Biomarkers Prev. 2019;28:1563–79. 10.1158/1055-9965.EPI-19-0221 . Jeon DS, Kim HC, Kim SH, et al. Five-year overall survival and prognostic factors in patients with lung cancer: results from the Korean Association of Lung Cancer Registry (KALC-R) 2015. CancerRes Treat. 2023;55:103–11. 10.4143/crt.2022.264 . Arriola E, Trigo JM, Sánchez-Gastaldo A, et al. Prognostic Value of Clinical Staging According to TNM in Patients With SCLC: A Real-World Surveillance Epidemiology and End-Results Database Analysis. JTO Clin Res Rep. 2021;3:100266. 10.1016/j.jtocrr.2021.100266 . Zhang P, Chen PL, Li ZH, et al. Association of smoking and polygenic risk with the incidence of lung cancer: a prospective cohort study. Br J Cancer. 2022;126:1637–46. 10.1038/s41416-022-01736-3 . Grant WB. Cancer Incidence Rates in the US in 2016–2020 with Respect to Solar UVB Doses, Diabetes and Obesity Prevalence, Lung Cancer Incidence Rates, and Alcohol Consumption: An Ecological Study. Nutrients. 2024;16:1450doi. 10.3390/nu16101450 . Ang L, Chan CPY, Yau WP, et al. Association between family history of lung cancer and lung cancer risk: a systematic review and meta-analysis. Lung Cancer. 2020;148:129–37. 10.1016/j.lungcan.2020.08.012 . Rami-Porta R, Nishimura KK, Giroux DJ, et al. The International Association for the Study of Lung Cancer Lung Cancer Staging Project: Proposals for Revision of the TNM Stage Groups in the Forthcoming (Ninth) Edition of the TNM Classification for Lung Cancer. J Thorac Oncol. 2024;19:1007–27. 10.1016/j.jtho.2024.02.011 . Yang B, Zhong J, Zhong J, et al. Development and Validation of a Radiomics Nomogram Based on 18 F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography and Clinicopathological Factors toPredict the Survival Outcomes of Patients With Non-Small Cell Lung Cancer. Front Oncol. 2020;10:1042. 10.3389/fonc.2020.01042 . Cascone T, Fradette J, Pradhan M, et al. Tumor Immunology and Immunotherapy of Non-Small-Cell Lung Cancer. Cold Spring Harb Perspect Med. 2022;12:a037895. 10.1101/cshperspect.a037895 . Ren S, Zeng G, Yi Y, et al. Combinations of plasma cfDNA concentration, integrity and tumor markers are promising biomarkers for early diagnosis of non-small cell lung cancer. Heliyon. 2023;9:e20851. 10.1016/j.heliyon.2023.e20851 . Cao X, Zheng YZ, Liao HY, et al. A clinical nomogram and heat map for assessing survival in patients with stage I non-smallcell lung cancer after complete resection. Ther Adv Med Oncol. 2020;12:1758835920970063. 10.1177/1758835920970063 . Reck M, Remon J, Hellmann MD. First-Line Immunotherapy for Non-Small-Cell Lung Cancer. J Clin Oncol. 2022;40:586–97. 10.1200/JCO.21.01497 . Patel SA, Weiss J. Advances in the Treatment of Non-Small Cell Lung Cancer: Immunotherapy. Clin Chest Med. 2020;41:237–47. 10.1016/j.ccm.2020.02.010 . Nasser NJ, Gorenberg M, Agbarya A. First line Immunotherapy for Non-Small Cell Lung Cancer. Pharmaceuticals (Basel). 2020;13:373. 10.3390/ph13110373 . Wang T, Yang Y, Liu X, et al. Primary Invasive Mucinous Adenocarcinoma of the Lung: Prognostic Value of CT Imaging Features Combined with Clinical Factors. Korean J Radiol. 2021;22:652–62. 10.3348/kjr.2020.0454 . Jain P, Khorrami M, Gupta A, et al. Novel Non-Invasive Radiomic Signature on CT Scans Predicts Response to Platinum-Based Chemotherapy and Is Prognostic of Overall Survival in Small Cell Lung Cancer. Front Oncol. 2021;11:744724. 10.3389/fonc.2021.744724 . Monaco L, Gemelli M, Gotuzzo I, et al. Metabolic Parameters as Biomarkers of Response to Immunotherapy and Prognosis in Non-Small Cell Lung Cancer (NSCLC): A Real World Experience. Cancers (Basel). 2021;13:1634. 10.3390/cancers13071634 . Tankyevych O, Trousset F, Latappy C, et al. Development of Radiomic-Based Model to Predict Clinical Outcomes in Non-Small Cell Lung Cancer Patients Treated with Immunotherapy. Cancers (Basel). 2022;14:5931. 10.3390/cancers14235931 . Saad MB, Hong L, Aminu M, et al. Predicting benefit from immune checkpoint inhibitors in patients with non-small-cell lung cancer by CT-based ensemble deep learning: a retrospective study. Lancet Digit Health. 2023;5:e404–20. 10.1016/S2589-7500(23)00082-1 . Moon JW, Yang E, Kim JH, et al. Predicting Non-Small-Cell Lung Cancer Survival after Curative Surgery via Deep Learning of Diffusion MRI. Diagnostics (Basel). 2023;13:2555. 10.3390/diagnostics13152555 . Li B, Yang L, Zhang H, et al. Outcome-Supervised Deep Learning on Pathologic Whole Slide Images for Survival Prediction of Immunotherapy in Patients with Non-Small Cell Lung Cancer. Mod Pathol. 2023;36:100208. 10.1016/j.modpat.2023.100208 . Yan H, Niimi T, Matsunaga T, et al. Preoperatively predicting survival outcome for clinical stage IA pure-solid non-small cell lung cancer by radiomics-based machine learning. J Thorac Cardiovasc Surg. 2024;S0022. 5223(24)00441-0.doi: 10.1016/j.jtcvs.2024.05.010 . Dercle L, Fronheiser M, Rizvi NA, et al. Baseline Radiomic Signature to Estimate Overall Survival in Patients With NSCLC. J Thorac Oncol. 2023;18:587–98. 10.1016/j.jtho.2022.12.019 . Yao W, Liao Y, Li X, et al. Noninvasive Method for Predicting the Expression of Ki67 and Prognosis in Non-Small-Cell Lung Cancer Patients: Radiomics. J Healthc Eng. 2022;2022:7761589. 10.1155/2022/7761589 . Chen Q, Shao J, Xue T, et al. Intratumoral and peritumoral radiomics nomograms for the preoperative prediction of lymphovascular invasion and overall survival in non-small cell lung cancer. Eur Radiol. 2023;33:947–58. 10.1007/s00330-022-09109-3 . Qi H, Hou Y, Zheng Z, et al. Clinical characteristics and MRI based radiomics nomograms can predict iPFS and short-term efficacy of third-generation EGFR-TKI in EGFR-mutated lung adenocarcinoma with brain metastases. BMC Cancer. 2024;24:362. 10.1186/s12885-024-12121-z . Wu J, Zhou Y, Xu C, et al. Effectiveness of CT radiomic features combined with clinical factors in predicting prognosis in patients with limited-stage small cell lung cancer. BMC Cancer. 2024;24:170. 10.1186/s12885-024-11862-1 . Yu G, Liu X, Li Y, et al. The nomograms for predicting overall and cancer-specific survival in elderly patients with early-stage lung cancer: A population-based study using SEER database. Front Public Health. 2022;10:946299. 10.3389/fpubh.2022.946299 . Zhu J, Shi H, Ran H, et al. Development and Validation of a Nomogram for Predicting Overall Survival in Patients with Second Primary Small Cell Lung Cancer After Non-Small Cell Lung Cancer: A SEER-Based Study. Int J Gen Med. 2022;15:3613–24. 10.2147/IJGM.S353045 . Yang B, Teng M, You H, et al. A Nomogram for Predicting Survival in Advanced Non-Small-Cell Lung Carcinoma Patients: A Population-Based Study. Cancer Invest. 2023;41:672–85. 10.1080/07357907.2023.2241547 . Guo J, Gao Y, Gong Z, et al. Plasma D-Dimer Level Correlates with Age, Metastasis, Recurrence, Tumor-Node-Metastasis Classification (TNM), and Treatment of Non-Small-Cell Lung Cancer (NSCLC) Patients. Biomed Res Int. 2021;2021:9623571doi. 10.1155/2021/9623571 . Hart PC, Rajab IM, Alebraheem M, et al. C-Reactive Protein and Cancer-Diagnostic and Therapeutic Insights. Front Immunol. 2020;11:595835. 10.3389/fimmu.2020.595835 . Buresova M, Benesova L, Minarik M, et al. Circulating Tumor DNA correlates with Lactate Dehydrogenase, CYFRA 21 – 1, and CRP levels in patients with advanced NSCLC. J Cancer. 2023;14:1–8. 10.7150/jca.78574 . Wang H, Yang R, Zhou K, et al. Association between pretreatment C-reactive protein level and survival in non-small cell lung cancer patients treated with immune checkpoint inhibitors: A meta-analysis. Int Immunopharmacol. 2023;124:110937. 10.1016/j.intimp.2023.110937 . Frey A, Martin D, D'Cruz L, et al. C-Reactive Protein to Albumin Ratio as Prognostic Marker in Locally Advanced Non-Small Cell Lung Cancer Treated with Chemoradiotherapy. Biomedicines. 2022;10:598. 10.3390/biomedicines10030598 . Nassar YM, Ojara FW, Pérez-Pitarch A, et al. C-Reactive Protein as an Early Predictor of Efficacy in Advanced Non-Small-Cell Lung Cancer Patients: A Tumor Dynamics-Biomarker Modeling Framework. Cancers (Basel). 2023;15:5429. 10.3390/cancers15225429 . Zheng X, Zhang L, Wu L, et al. Baseline C-reactive protein predicts efficacy of the first-line immune checkpoint inhibitors plus chemotherapy in advanced lung squamous cell carcinoma: a retrospective, multicenter study. BMC Cancer. 2023;23:1244. 10.1186/s12885-023-11737-x . Vanguri RS, Luo J, Aukerman AT, et al. Multimodal integration of radiology, pathology and genomics for prediction of response to PD-(L)1 blockade in patients with non-small cell lung cancer. Nat Cancer. 2022;3:1151–64. 10.1038/s43018-022-00416-8 . Zhang Y, Yang Z, Chen R, et al. Histopathology images-based deep learning prediction of prognosis and therapeutic response in small cell lung cancer. NPJ Digit Med. 2024;7:15. 10.1038/s41746-024-01003-0 . Zheng X, Liu K, Shen N, et al. Predicting overall survival and prophylactic cranial irradiation benefit in small-cell lung cancer with CT-based deep learning: A retrospective multicenter study. Radiother Oncol. 2024;195:110221. 10.1016/j.radonc.2024.110221 . Felfli M, Liu Y, Zerka F, et al. Systematic Review, Meta-Analysis and Radiomics Quality Score Assessment of CT Radiomics-Based Models Predicting Tumor EGFR Mutation Status in Patients with Non-Small-Cell Lung Cancer. Int J Mol Sci. 2023;24:11433. 10.3390/ijms241411433 . Deng L, Tang HZ, Luo YW, et al. Preoperative CT Radiomics Nomogram for Predicting Microvascular Invasion in Stage I Non-Small Cell Lung Cancer. Acad Radiol. 2024;31:46–57. 10.1016/j.acra.2023.05.015 . González M, Calvo V, Redondo I, et al. Overall survival for early and locally advanced non-small-cell lung cancer from one institution: 2000–2017. Clin Transl Oncol. 2021;23:1325–33. 10.1007/s12094-020-02521-5 . Nakwan N, Kumsuk K. Survival Analysis of Lung Cancer: A 10-Year Real-Life Experience in a Non-University-Based Hospital in Thailand (2012–2021). Asian Pac J Cancer Prev. 2023;24:3021–7. 10.31557/APJCP.2023.24.9.3021 . Chen P, Liu Y, Wen Y, et al. Non-small cell lung cancer in China. Cancer Commun (Lond). 2022;42:937–70. 10.1002/cac2.12359 . Lahiri A, Maji A, Potdar PD, et al. Lung cancer immunotherapy: progress, pitfalls, and promises. Mol Cancer. 2023;22:40. 10.1186/s12943-023-01740-y . Lv B, Wang Y, Ma D, et al. Immunotherapy: Reshape the Tumor Immune Microenvironment. Front Immunol. 2022;13:844142. 10.3389/fimmu.2022.844142 . Nishioka R, Kawahara D, Imano N, et al. A nomogram-based survival prediction model for non-small cell lung cancer patients based on clinical risk factors and multiregion radiomics features. Clin Radiol. 2025;84:106826. 10.1016/j.crad.2025.106826 . Wu L, Lou X, Kong N, et al. Can quantitative peritumoral CT radiomics features predict the prognosis of patients with non-small cell lung cancer? A systematic review. Eur Radiol. 2023;33(3):2105–17. 10.1007/s00330-022-09174-8 . Jia X, Wang Y, Zhang H, et al. Current status and quality of prognosis prediction models of non-small cell lung cancer constructed using computed tomography (CT)-based radiomics: a systematic review and radiomics quality score 2.0 assessment. Quant Imaging Med Surg. 2024;14(9):6978–89. 10.21037/qims-24-22 . Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-7907062","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":560075434,"identity":"4c92ea26-27fa-47ee-a4eb-fb6b020d6309","order_by":0,"name":"Yubo Wang","email":"","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yubo","middleName":"","lastName":"Wang","suffix":""},{"id":560075435,"identity":"be7bff5a-cd46-417c-8111-1e95e4007785","order_by":1,"name":"Zefei Peng","email":"","orcid":"","institution":"Northeast Yunnan Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zefei","middleName":"","lastName":"Peng","suffix":""},{"id":560075436,"identity":"520689c3-45c3-4e65-9c02-aa5c9313098e","order_by":2,"name":"Hao Hu","email":"","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Hu","suffix":""},{"id":560075437,"identity":"4d3c797b-c40e-4b84-a032-1b57b3177f72","order_by":3,"name":"Zhiyong Ding","email":"","orcid":"","institution":"Qujing Maternal and Child Health Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhiyong","middleName":"","lastName":"Ding","suffix":""},{"id":560075438,"identity":"d8ff0c40-846c-40c7-a143-eebdfbae4573","order_by":4,"name":"Xiandou Li","email":"","orcid":"","institution":"Dehong Vocational College","correspondingAuthor":false,"prefix":"","firstName":"Xiandou","middleName":"","lastName":"Li","suffix":""},{"id":560075439,"identity":"fb88bb32-d8de-4bda-ae01-9428a76c81bb","order_by":5,"name":"Yang Fu","email":"","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Fu","suffix":""},{"id":560075440,"identity":"d0a11255-1d68-4128-b107-01a267edc7d7","order_by":6,"name":"Jiageng Li","email":"","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jiageng","middleName":"","lastName":"Li","suffix":""},{"id":560075441,"identity":"1622884f-4478-422a-8b8a-8027b5d8b3ae","order_by":7,"name":"Yisong Zhang","email":"","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yisong","middleName":"","lastName":"Zhang","suffix":""},{"id":560075442,"identity":"282a931b-c89e-406a-b52a-13ef2ad52b41","order_by":8,"name":"Bosen Xie","email":"","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":false,"prefix":"","firstName":"Bosen","middleName":"","lastName":"Xie","suffix":""},{"id":560075443,"identity":"9d51f89f-c456-45c6-a8d6-04338e6fe651","order_by":9,"name":"Mengxue Kong","email":"","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":false,"prefix":"","firstName":"Mengxue","middleName":"","lastName":"Kong","suffix":""},{"id":560075444,"identity":"67fafd7a-5c0c-4554-80bd-91f605dfe7ed","order_by":10,"name":"Bin Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6UlEQVRIiWNgGAWjYJACZgYDMNXA8MHAxo4ULYwNjDMK0pKJ1MIA0cLM8+EQYwMh5QbHzx5+XVBwR95curHxsY3BAWYG9sNHN+DVciYvzXqGwTPDnXMONhvnGNzhY+BJS7uBV8uBHDNjHoPDjBtuJLZJ5xg8Y2aQ4DHDr+X8G7AWe6CW9t8WQL0NBLXcyDF+DNSSCLIFGHREaJG88caMeYbB4WSQXyR7DNKS2Qj5he98jvHngj+HbbdLNx/88OOPjR0/++FjeLUoHGBgkwC7UAIqwoZPOQjINzAwf0DRMgpGwSgYBaMAHQAAGjVRO/3g9UQAAAAASUVORK5CYII=","orcid":"","institution":"Kunming Medical University Affiliated Calmette Hospital","correspondingAuthor":true,"prefix":"","firstName":"Bin","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-10-20 15:23:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7907062/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7907062/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98780185,"identity":"261860a8-bf07-4020-9160-efbe11f30123","added_by":"auto","created_at":"2025-12-22 12:31:09","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1340413,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/47aa7dddc6d67e5e2a42a3fa.docx"},{"id":98777620,"identity":"f442249e-47bf-44f5-8872-4ba5433528c9","added_by":"auto","created_at":"2025-12-22 12:28:12","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11470,"visible":true,"origin":"","legend":"","description":"","filename":"86c7782dafc1410189caf1220ad323bf.json","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/95d95a134d5790e346292ab8.json"},{"id":98778246,"identity":"81505b0f-41a5-4cd4-a04d-73528db3f9a8","added_by":"auto","created_at":"2025-12-22 12:29:05","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":186856,"visible":true,"origin":"","legend":"","description":"","filename":"86c7782dafc1410189caf1220ad323bf1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/1aa961c556f676f499d0c3d9.xml"},{"id":98780192,"identity":"f46ab346-8e81-4b58-a8b1-5718317afeda","added_by":"auto","created_at":"2025-12-22 12:31:09","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117079,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/910dc4566ea7eac4b5267383.png"},{"id":98778911,"identity":"33707bdd-c3e0-4d59-9d36-8f5f8fa98841","added_by":"auto","created_at":"2025-12-22 12:29:48","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79747,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/997556d3b9692a737ed2fc6d.jpeg"},{"id":98752010,"identity":"9cade9e4-5640-4b7e-9d7f-c51e2ad162a3","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":74933,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/0bce06dd0f1cb54f88262269.jpeg"},{"id":98752016,"identity":"a1148368-1f88-493c-8750-678dbd5dc7f8","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":392430,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/5438482eff441fa6b3682bc3.png"},{"id":98777557,"identity":"b8d4852f-056a-4e29-b59d-e502d8ca8ce1","added_by":"auto","created_at":"2025-12-22 12:28:06","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117389,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/90603283c78ee3148baad534.png"},{"id":98752041,"identity":"dec9a041-4a4f-4880-8eee-2f194a6fdb29","added_by":"auto","created_at":"2025-12-22 09:13:49","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":243160,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/a8ec6a9faa0c5a9c9ec88680.jpeg"},{"id":98777763,"identity":"15c139d9-1c5b-40f1-84bc-1ecf9545ee93","added_by":"auto","created_at":"2025-12-22 12:28:25","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60474,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/b33c28c1317279e9f3a7305b.jpeg"},{"id":98752019,"identity":"9b8d0688-386a-4d2e-8fa9-cbeaa6fe2cfa","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":57069,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/48672e61491de40c9d839297.jpeg"},{"id":98752014,"identity":"926b1ea0-de99-4673-9123-dd7d3ce213df","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":53641,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/e6246a49d1f8bc39acd8d538.jpeg"},{"id":98779565,"identity":"e151f540-5d74-4979-b480-932e16f515f0","added_by":"auto","created_at":"2025-12-22 12:30:27","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":110455,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/be2e11cf7940c531c946145e.png"},{"id":98780281,"identity":"a02c2c4f-3871-4f78-8139-f3328ae2ef58","added_by":"auto","created_at":"2025-12-22 12:31:12","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124312,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/4e978b7a269322e9e2991086.png"},{"id":98777514,"identity":"916bac2f-c48c-4c31-b3c7-89a46dfcd157","added_by":"auto","created_at":"2025-12-22 12:27:52","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49396,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/36d8ff458d51e247a581debf.png"},{"id":98777667,"identity":"6ddcc57e-9a93-4595-9517-27cc4d206b3f","added_by":"auto","created_at":"2025-12-22 12:28:19","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41820,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/a2a4110668276c0db33aa46d.png"},{"id":98752031,"identity":"a43f3572-6596-48c9-b710-7db5d0384520","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38510,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/338ac76c4dca8cce8b78f885.png"},{"id":98778814,"identity":"04ea4dc4-a286-41fd-9807-46c22269a305","added_by":"auto","created_at":"2025-12-22 12:29:41","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52697,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/a1f9cf0db6f41b608ea68597.png"},{"id":98777674,"identity":"d9bdf856-2898-4122-a132-ed9bc16cc029","added_by":"auto","created_at":"2025-12-22 12:28:19","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23259,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/1fe66e407d15ff27391dd937.png"},{"id":98752028,"identity":"130d3d00-e531-452f-ace9-9b44b88ab740","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38101,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/dd92c359fa693f887eceb97e.png"},{"id":98777890,"identity":"cb029ba9-3cbd-4982-a8c3-3525086029c8","added_by":"auto","created_at":"2025-12-22 12:28:37","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17453,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/953c013333c17c4168a40091.png"},{"id":98752037,"identity":"4acbbb43-792d-41c3-8ba8-e3a4f7957419","added_by":"auto","created_at":"2025-12-22 09:13:49","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24726,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/ef50cb022d9372f429d5a616.png"},{"id":98752033,"identity":"ac742337-9255-44ca-bdf6-7427adbd2f60","added_by":"auto","created_at":"2025-12-22 09:13:49","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20955,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/e032750fa0eb855b142311c1.png"},{"id":98778818,"identity":"36043c29-189f-4190-bc48-bf5a100d6b33","added_by":"auto","created_at":"2025-12-22 12:29:41","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23273,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/e6ab55188ba002ecbc17ff30.png"},{"id":98752036,"identity":"491dbe11-1a77-44e8-9805-6e4c82afa890","added_by":"auto","created_at":"2025-12-22 09:13:49","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27195,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/06f84f913c8d412022292e9e.png"},{"id":98777539,"identity":"34215849-055a-436a-a504-5906edece224","added_by":"auto","created_at":"2025-12-22 12:27:57","extension":"xml","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":183151,"visible":true,"origin":"","legend":"","description":"","filename":"86c7782dafc1410189caf1220ad323bf1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/9806d0709cbfdf52840a72a0.xml"},{"id":98779432,"identity":"4207cbea-0e84-4b70-8217-016e73730ba7","added_by":"auto","created_at":"2025-12-22 12:30:20","extension":"html","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":199371,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/5e74bcb96dcbf1109643c049.html"},{"id":98752002,"identity":"53536a17-47f2-45d7-90ef-91747c78e32e","added_by":"auto","created_at":"2025-12-22 09:13:47","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":90186,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 1 Flow chart of patient selection\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/39719c4c3894e314a5acb866.jpg"},{"id":98752003,"identity":"67da52b3-475b-4eed-a5d6-becc7931d8ef","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":164421,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 2: The workflow of radiomics in this study consists of four steps: lung tumor VOI delineation, radiomics feature extraction, key feature selection, and prediction model construction and evaluation.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/51782d39bfebad451e29983f.jpg"},{"id":98777454,"identity":"7f6a5d96-5007-43ba-9dc4-aee5734c3f76","added_by":"auto","created_at":"2025-12-22 12:27:26","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":83404,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 3 Prediction of OS in NSCLC patients nomogram: The total score is calculated by adding the scores of each independent predictor, reflecting the accuracy of predicting OS in NSCLC patients. The higher the total score, the greater the probability of accurate prediction.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/9046bee082b18500ec117853.jpg"},{"id":98777465,"identity":"2fd632fa-1690-4d75-aaea-f479550f5c98","added_by":"auto","created_at":"2025-12-22 12:27:27","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":94921,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 4A The Kaplan Meier (KM) curve of the training set model (log-rank test for both groups, P\u0026lt;0.0001). The red curve represents the high-risk group, and the green curve represents the low-risk group. From this graph, it can be seen that the survival rate of patients in the low-risk group is higher than that of the high-risk group at 12, 24, and 36 months after surgery, and the survival rate of patients in the low-risk group at these time points is close to 1.\u003c/p\u003e","description":"","filename":"4a.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/08797874b54f68efd9ac7a47.jpg"},{"id":98777641,"identity":"ed7b9126-7707-4281-865a-695514b7af1a","added_by":"auto","created_at":"2025-12-22 12:28:15","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":91169,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 4B The Kaplan Meier (KM) curve of the validation set model (log-rank test P=0.0091 for both groups). The red curve represents the high-risk group, and the green curve represents the low-risk group. According to this chart, the survival rate of patients in the low-risk group is higher than that of the high-risk group at 12, 24, and 36 months after surgery. The survival rate of patients in the low-risk group at these time points is close to 1.\u003c/p\u003e","description":"","filename":"4b.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/c1f7a1683c9ceda4768b6a8d.jpg"},{"id":98777837,"identity":"6aa55085-3b26-4bb2-a0b7-8a3762dc6e7c","added_by":"auto","created_at":"2025-12-22 12:28:33","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":33747,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 5 A The calibration curve of the training set shows a good fit, indicating that the nomogram model has good consistency. The results of the calibration curve indicate that there is good consistency between the predicted survival rates of patients at 1, 2, and 3 years using the nomogramand the actual survival rates. This result further validates the reliability and accuracy of the model in predicting prognosis.\u003c/p\u003e","description":"","filename":"5a.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/704a20d772fac4edfb75f911.jpg"},{"id":98752008,"identity":"28beef17-f8cb-4253-a708-384cae9ee48c","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":31948,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 5B The calibration curve of the validation set shows a good fit, indicating that the nomogram model has good consistency. The results of the calibration curve indicate that there is good consistency between the predicted survival rates of patients at 1, 2, and 3 years using the nomogramand the actual survival rates. This result further validates the reliability and accuracy of the model in predicting prognosis.\u003c/p\u003e","description":"","filename":"5b.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/808d169565a7990de077e589.jpg"},{"id":98779304,"identity":"24ed83d9-2300-4d0a-9b7f-d891338a604b","added_by":"auto","created_at":"2025-12-22 12:30:12","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":108412,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 6A ROC curve for predicting the survival of NSCLC patients using a training set comprehensive prediction model. For an effective regression model (AUC\u0026gt;0.5), the closer the AUC is to 1.0, the better the model performance. The AUC values predicted by the comprehensive prediction model for the 1, 2, and 3year survival rates of patients were 0.890, 0.926, and 0.988, respectively.\u003c/p\u003e","description":"","filename":"6a.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/44e8bd2252be00bcca9df83d.png"},{"id":98780335,"identity":"d7332bf0-6eb9-431a-84d2-bcada4bc170a","added_by":"auto","created_at":"2025-12-22 12:31:13","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":127402,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 6B ROC curve for predicting the survival of NSCLC patients using a validationset comprehensive prediction model. For an effective regression model (AUC\u0026gt;0.5), the closer the AUC is to 1.0, the better the model performance. The AUC values predicted by the comprehensive prediction model for the 1, 2, and 3 year survival rates of patients were 0.890, 0.926, and 0.988, respectively.\u003c/p\u003e","description":"","filename":"6b.png","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/46eb1a937870663be9ec68a7.png"},{"id":98752023,"identity":"e403fa6a-f0c3-42ae-a260-dd8ac90fb716","added_by":"auto","created_at":"2025-12-22 09:13:48","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":128441,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 7A Decision curve analysis (DCA) of the three models in the training set.The red line indicates that the prognosis of all patients is good, and the purple line indicates that the prognosis of all patients is bad. Within a certain threshold range, the prediction effect of the three models is better, and the combined prediction model has the highest net benefit. It is suggested that the combined model has great advantages in clinical decision-making.\u003c/p\u003e","description":"","filename":"7a.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/3dccc06f6173cc10eb63d008.jpg"},{"id":98777594,"identity":"4a51d246-b8a2-4d34-8950-b75630574e4b","added_by":"auto","created_at":"2025-12-22 12:28:09","extension":"jpg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":120005,"visible":true,"origin":"","legend":"\u003cp\u003eFigure 7B Decision curve analysis (DCA) of the three models in the validation set.The red line indicates that the prognosis of all patients is good, and the purple line indicates that the prognosis of all patients is bad. Within a certain threshold range, the prediction effect of the three models is better, and the combined prediction model has the highest net benefit. It is suggested that the combined model has great advantages in clinical decision-making.\u003c/p\u003e","description":"","filename":"7b.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/58adfe2bb17d1d3302168c55.jpg"},{"id":102409599,"identity":"bcd3b8f8-cc97-4d17-8128-0876dc926fc8","added_by":"auto","created_at":"2026-02-11 11:42:48","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2360599,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7907062/v1/9164ba1b-8e0c-4929-a2de-1d3f56dcd1f1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Predict the prognosis of patients with non-small cell lung cancer based on CT radiomics and clinical pathological factors","fulltext":[{"header":"BACKGROUND","content":"\u003cp\u003eLung cancer is one of the main causes of cancer-related deaths worldwide, with a global incidence rate of about 11.4%, making it the second most common malignant tumor. Its mortality rate is about 18.0%, ranking it first among all malignant tumors\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. Despite significant progress in the treatment of lung cancer in recent years, including a shift from traditional surgical and cytotoxic drug therapies to targeted therapies and immunotherapies, patients still face considerable challenges in prognosis, with a five-year overall survival rate of only 15% for lung cancer\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e. Therefore, identifying effective prognostic indicators for non-small cell lung cancer (NSCLC) is of great clinical significance for the early assessment of patients\u0026rsquo; responses to treatment and for the prediction of overall survival (OS), eventually guiding personalized treatment strategies.\u003c/p\u003e \u003cp\u003eAt present, research on the prognosis of lung cancer involves multiple aspects. Clinical factors such as age, gender, smoking history, alcohol consumption history, family history, and tumor-node-metastasis (TNM) staging significantly impact lung cancer prognosis\u003csup\u003e[\u003cspan additionalcitationids=\"CR6 CR7\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. Although the TNM staging system is currently the most commonly used tumor staging system worldwide\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e, it still has certain limitations when used alone to comprehensively evaluate the prognosis of patients with NSCLC. This is primarily due to the significant differences in the prognosis of tumors at the same stage\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. Blood biochemical indicators, such as C-reactive protein (CRP), platelets, D-dimer, fibrin degradation products (FDP), neuron-specific enolase (NSE), carcinoembryonic antigen (CEA), cancer antigen 125 (CA125), and cancer antigen 153 (CA153), have also received attention as prognostic factors for lung cancer\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. However, the predictive ability of any single indicator for prognosis remains limited. Some scholars have proposed that treatment methods significantly impact the prognosis of lung cancer. Studies have also shown that surgical treatment is mainly focused on the early stages of lung cancer\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e, while targeted therapy and immunotherapy are mainly applied to patients with advanced lung cancer\u003csup\u003e[\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eVarious diagnostic and therapeutic methods have shown good results in the treatment of lung cancer, but their roles in prognosis can vary. CT imaging features of lung cancer, such as spiculation, lobulation, air bronchogram sign, and tumor density, have been proven to be independent risk factors for predicting prognosis. These features assist in providing prognostic information and personalized risk assessment\u003csup\u003e[\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e. However, the extraction of traditional image features is heavily influenced by subjective factors related to the operator, resulting in low data processing capability that no longer meets the demands of modern medicine\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eRadiomics refers to the extraction and analysis of a large number of advanced quantitative imaging features from medical imaging images such as CT, PET, or magnetic resonance imaging (MRI)\u003csup\u003e[\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e. Compared to traditional imaging features, radiomics provides higher levels of automation, a greater number of feature dimensions, and enhanced data processing capabilities. These advantages have been widely used in the prognosis research for lung cancer. Yan et al.\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e found that machine learning models based on radiomics can accurately predict the OS of inpatients with stage IA solid NSCLC before surgery, thereby improving risk stratification. Dercle et al.\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e demonstrated that radiomics features can provide early assessments of expected OS in patients with NSCLC treated with nivolumab or chemotherapy, further supporting risk stratification in clinical trials. Yao et al.\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e proposed that radiomics features derived from \u003csup\u003e18\u003c/sup\u003eF-FDG PET/CT can effectively distinguish the Ki-67 expression and have strong predictive abilities for survival rates. Chen et al.\u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e showed that a radiomics nomogram based on CT can predict preoperative lymphatic invasion (LVI) and OS in patients with NSCLC, aiding in the development of personalized treatment strategies before surgery. However, the limitations of the above research lie in the low AUC values obtained solely from CT radiomics features.\u003c/p\u003e \u003cp\u003eTherefore, there is an urgent need to develop a new predictive model to predict the prognosis of patients with NSCLC. Qi et al.\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e constructed and validated a nomogram based on clinical features and MRI radiomics to predict short-term efficacy and intracranial progression-free survival (iPFS) in patients with lung adenocarcinoma brain metastases. The results showed that in the short-term efficacy model, the combined C-index of the training and validation sets (C\u0026thinsp;+\u0026thinsp;R) nomogram was 0.867 and 0.803, respectively, outperforming the individual clinical (C) and radiomics (R) nomograms. Wu et al.\u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e randomly divided 200 patients with locally limited small-cell lung cancer (SCLC) into a training group (n\u0026thinsp;=\u0026thinsp;140) and a testing group (n\u0026thinsp;=\u0026thinsp;60). They extracted and screened imaging features from simulated localization CT images and constructed radiomics models, clinical models, and a combined model that included both clinical factors and radscore. Their research showed that the combined model of radscore and clinical factors could effectively predict the prognosis of patients with limited-stage SCLC, exhibiting better performance than imaging and clinical models alone. In summary, it has been found that the efficacy and accuracy of predicting the prognosis of patients with NSCLC based solely on clinicopathological features, blood biochemical indicators, imaging features, and radiomics are not satisfactory. In contrast, a comprehensive predictive model that combines radiomics and clinical features shows higher AUC values and has better predictive performance. Based on the above research, we hypothesize that a nomogram model combining CT radiomics with clinical and pathological factors may have good performance in classification and prognosis prediction. This nomogram integrates important influencing factors related to tumor prognosis through multiple regression analysis and constructs intuitive graphs using statistical prediction models to provide numerical probabilities of clinical events\u003csup\u003e[\u003cspan additionalcitationids=\"CR31\" citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTherefore, the main objective of this study was to construct a predictive model for OS in the form of a nomogram based on CT radiomics features combined with clinicopathological factors. This model aims to predict prognosis and perform risk stratification, thereby improving the assessment of prognosis for patients with NSCLC and providing guidance for advancing personalized treatment and precision medicine.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatients and clinicopathological data\u003c/h2\u003e \u003cp\u003eThis retrospective study included a total of 216 patients who underwent chest CT scans at the First People's Hospital of Kunming from February 2019 to September 2023 and were diagnosed with NSCLC through pathology from the PACS system. This study was approved by the Ethics Committee of the First People's Hospital of Kunming(YLS2023-75), and patients' informed consent was waived.\u003c/p\u003e \u003cp\u003eThe inclusion criteria were as follows: (1) Diagnosis of NSCLC confirmed through puncture pathology; (2) Completion of a chest CT scan within one month before treatment, with complete imaging data available; (3) No prior needle therapy received for anti-tumor treatment before the CT examination.\u003c/p\u003e \u003cp\u003eThe exclusion criteria were: (1) Receipt of anti-tumor treatment before chest CT examination; (2) Incomplete clinical data or missing/poor-quality images; (3) Presence of concurrent malignant tumors in other organs; (4) Patients with missing follow-up data.\u003c/p\u003e \u003cp\u003eBased on the inclusion and exclusion criteria, a total of 39 patients who did not meet the requirements were excluded from this study. The final cohort comprised 177 patients, including 107 males and 70 females. Among these, there were 139 survivors and 38 patients who had died. There were 52 cases of squamous cell carcinoma and 125 cases of adenocarcinoma; 80 patients received surgical treatment, 99 patients received chemotherapy, and 49 patients received targeted therapy; 32 cases received radiotherapy and 19 cases received immunotherapy. The patients were randomly divided into a training set (n\u0026thinsp;=\u0026thinsp;124) and a validation set (n\u0026thinsp;=\u0026thinsp;53) in a 7:3 ratio. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the standard flow chart for patient inclusion and exclusion.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eImage acquisition and reconstruction parameters\u003c/h3\u003e\n\u003cp\u003eAll patients underwent multi-slice spiral CT scans (SOMATOM Definition AS 128, Siemens Healthineers; Perform chest CT plain scan using Brilliance iCT 256, Philips Healthcare. Before scanning, patients are required to remove any metal objects from their chest, such as metal zippers on clothing and any metallic objects on their undergarments. During the examination, the patient lies supine on the examination bed, with both arms raised, and is calm throughout the process. As the scan begins, the patient's head is advanced, and they are instructed to hold their breath after taking a deep breath in accordance with the machine\u0026rsquo;s instructions. The scanning range extends from the chest entrance to the level of the lung base. The scanning parameters were as follows: tube voltage of 120 kVp, tube current of 110/113mAs, FOV of 350 mm \u0026times; 400 mm, matrix of 512 \u0026times; 512, layer thickness of 1 mm, layer spacing of 1 mm, collimator width of 128 \u0026times; 0.6 mm, and tube rotation time of 0.5 s. On completion of the scan, the original images are reconstructed using standard techniques. The data are uploaded to the PACS system through a 1 mm thin layer and multi-plane reconstruction, and the patient's CT images are downloaded and saved in DICOM format.\u003c/p\u003e\n\u003ch3\u003eImage preprocessing and segmentation\u003c/h3\u003e\n\u003cp\u003eThe linear interpolation algorithm was used to resample all images to isotropic voxels with a size of 1 mm \u0026times; 1 mm \u0026times; 1 mm, using the B-spline interpolation algorithm for the resampling process. Using the Huiyi Huiying Data Artificial Intelligence Research Platform(\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://mics.huiyihuiying.com/#/\u003c/span\u003e\u003cspan address=\"https://mics.huiyihuiying.com/#/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), a radiologist with three years of experience, Zefei Peng, manually delineated the region of interest (ROI) of the tumor on a lung window with a thickness of 1 mm (window width: 1600HU, window position: -550HU). The entire tumor lesion area was outlined layer by layer. Once the delineation was completed, a radiology attending physician with 10 years of experience, Yang Fu, reviewed and corrected the delineated ROI.\u003c/p\u003e\n\u003ch3\u003eFeature extraction and selection\u003c/h3\u003e\n\u003cp\u003eAs shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, in the process of feature extraction, we extracted histogram features, morphological features, high-order texture features, clinical features, etc, respectively. A total of 1,688 radiomics features were extracted from the volume of interest of tumors in CT images, mainly divided into the following categories: (1) First-order statistical features: This category includes 324 local binary patterns (LBP) and wavelet filtering features that reflect the symmetry, uniformity, and local intensity distribution changes of the measured voxels. Key metrics include median, mean, minimum, maximum, standard deviation, skewness, and kurtosis. Kurtosis is used to quantify the steepness of pixel intensity distribution in tumor images, reflecting the heterogeneity and microstructure complexity of tumor tissues.(2) Shape features: A total of 14 morphological features quantitatively describe the geometric characteristics of the region of interest, such as the surface area, volume, surface area-to-volume ratio, sphericity, compactness, and three-dimensional diameter of the tumor. These features can describe the three-dimensional size and morphological information of the tumor. (3) Gray level co-occurrence matrix (GLCM) features: This category includes 432 features derived from LBP and wavelet, serving as a statistical tool to describe image texture. Key indicators of GLCM features, such as contrast, correlation, energy (or inverse entropy), homogeneity (or contrast), and entropy, provide different information about image texture and can be used for applications such as image classification, retrieval, and segmentation. The contrast reflects the sharp edge of the tumor, and high contrast indicates irregular boundary or invasive growth of the tumor; correlation is used to measure the consistency of tumor internal texture. Low correlation may correspond to high malignancy (such as GLCM characteristics of poorly differentiated squamous cell carcinoma); entropy represents the disorder of tumor tissue, and high entropy is common in necrotic or highly heterogeneous regions (the entropy of metastatic lymph nodes is significantly higher than that of benign lesions). (4) Neighborhood Gray Tone Difference Matrix (NGTDM) Features: A total of 90 NGTDM features are included, mainly used to quantify the grayscale differences between pixels or voxels and their adjacent counterparts within a predefined distance. The main characteristics of NGTDM include roughness, busyness, contrast, complexity, and texture intensity. (5) Gray Level Run Length Matrix (GLRLM) Features: This category has a total of 288 features, which serves as a statistical measure to describe the number of consecutive occurrences of gray levels in a specified direction in an image. These features assist in analyzing the texture and structural information of the image. (6) Gray Level Size Zone Matrix (GLSZM): This matrix consists of 288 units and is a statistical tool used to describe the size of texture regions in an image. (7) Gray Level Dependence Matrix (GLDM) Features: A total of 252 features are included, which describe the dependency relationship between grayscale values in images, and can aid in identifying and quantifying heterogeneity within tissues. Using the intra-class correlation coefficient (ICC) features with an ICC\u0026thinsp;\u0026gt;\u0026thinsp;0.9 were selected, while those that are highly correlated with other features were removed.\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eBefore performing feature selection, all radiomics features were standardized by removing the mean and dividing by standard deviation, resulting in a mean of zero and a variance of one for each set of feature values. Owing to the large number of radiomics features and the aim to mitigate issues related to model overfitting and multicollinearity, two methods were used to reduce dimensionality: Firstly, Spearman correlation analysis was used to eliminate features with a correlation of 0.9 or higher. Secondly, single factor Cox analysis was used to retain features with p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, leaving 137 features. Finally, COX-LASSO and ten fold cross validation were used to further determine key radiomics features, select the indicators most relevant to the research objectives, and obtain the weights of these indicators. The Radscore for each patient was calculated using the following formula: (radscore\u0026thinsp;=\u0026thinsp;0.230 * Kurtosis_firstorder_original\u0026thinsp;+\u0026thinsp;0.068 * Mean_firstorder_original\u0026thinsp;+\u0026thinsp;0.154 * JointAverage_glcm_original+-0.022 * DifferentiatedEntropy_glcm_original\u0026thinsp;+\u0026thinsp;0.200 * ClusterProminence_glcm_original\u0026thinsp;+\u0026thinsp;0.130 * DifferentiatedAverage_glcm_original\u0026thinsp;+\u0026thinsp;0.310 * LargeDependenciesHighGrayLevelEmphasis_gldm_original\u0026thinsp;+\u0026thinsp;0.199 * LongRunLowGrayLevelEmphasis_glrlm_original\u0026thinsp;+\u0026thinsp;0.178 * RootMeanSquare_firstorder_logarithm\u0026thinsp;+\u0026thinsp;0.021 * JoineEntropy_glcm_logarithm\u0026thinsp;+\u0026thinsp;0.300 * JointeEnergy_glcm_logarithm+-0.008 * Idm_glcm_logarithm\u0026thinsp;+\u0026thinsp;0.016 * DifferentiatedAverage_glcm_logarithm-0.025 * LargeDependenceEmphasis_gldm_logarithm-0.065 * LargeDependenceLowGrayLevelEmphasis_gldm_logarithm+-0.332 * SmallDependenceLowGrayLevelEmphasis_gldm_logarithm). The Radscore serves as a comprehensive reflection of the radiomics features and will be included in the subsequent model construction.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e\n\u003ch3\u003eEstablishment of predictive models\u003c/h3\u003e\n\u003cp\u003eFor the clinical pathological factors of patients in the training set, significant variables were first selected through univariate analysis. Subsequently, multivariate analysis was conducted to identify independent risk factors that are significantly correlated with prognosis, thereby establishing a clinical model. Simultaneously, radiomic features were screened, and key radiomics features were extracted to establish radiomics labels. Based on these Radscore, a radiomics model was developed. By combining the Radscore with the clinically independent risk factors, a joint prediction model was established using multiple regression analysis. A nomogram was constructed based on multiple logistic regression, and the receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of the prediction model. The AUC value was used as a quantitative indicator for evaluating the predictive performance of the model. According to the model\u0026rsquo;s formula, scores were calculated for each patient, allowing for their classification into high-risk and low-risk groups based on these scores. C-index was used to evaluate the predictive performance of the model, while the Kaplan-Meier (KM) method was used to estimate the median survival and survival rate of patients. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the workflow of the radiomics analysis.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eEstablishment and verification of radiomics nomogram\u003c/h2\u003e \u003cp\u003eA nomogram was established based on multivariate logistic regression analysis by combining clinically independent risk factors and Radscore (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The accuracy of the model\u0026rsquo;s predictions was evaluated using calibration curves, where the horizontal axis represents the predicted probability and the vertical axis represents the actual probability. The reference curve, which represents an ideal scenario where the predicted and actual probabilities are equal, serves as a benchmark; therefore, the closer the calibration curve is to the reference curve, the better the model\u0026rsquo;s consistency. Decision curve analysis (DCA) is used to evaluate the clinical practicability of the prediction model, also known as the evaluation of net benefit. DCA can integrate the preferences of decision makers and patients into the analysis. The vertical axis is the value of net benefit and the horizontal axis is the threshold probability. Under the same threshold probability, the higher the net benefit, the better the clinical efficacy of the model.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analysis was conducted using R language software (version 4.3.1; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.r-project.org\u003c/span\u003e\u003cspan address=\"https://www.r-project.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Radiomics features were subjected to statistical analysis, with continuous data that conform to a normal distribution typically represented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD). For skewed data distributions, the median and interquartile range (IQR) were used for description. Categorical variables were represented by the number of cases and the percentage [n(%)]. Depending on the distribution of data, either the t-test or the Mann-Whitney U test was used for inferential comparison. The statistical significance of categorical variables was assessed using the χ\u003csup\u003e2\u003c/sup\u003e test or Fisher's exact test, with p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 indicating statistically significant differences. The performance of each prediction model was quantified using the ROC curve and the AUC. Proportional Hazards (PH) is the core assumption of the Cox proportional hazards model in survival analysis, which refers to the risk ratios at different groups or covariate levels remaining constant throughout the entire follow-up period and not changing over time. We confirmed the validity of the PH hypothesis through Schoenfeld residual test (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05), and the model results were robust. A Cox regression model was established to analyze survival time, using hazard ratio (HR) values to establish the relationship between survival time and various variables. Meanwhile, C-index was used to evaluate the accuracy of the nomogram model. The KM method was used to estimate median survival and to calculate survival curves, with comparisons of survival curves (survival rates) conducted using a log-rank test.\u003c/p\u003e \u003c/div\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eClinicopathological characteristics\u003c/h2\u003e \u003cp\u003eThis study included a total of 177 patients with NSCLC, among whom 38 (21.47%) had died and 139 (78.53%) survived. The cohort comprised 107 males (60.45%) and 70 females (39.55%) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). All variables were included in the Cox regression analysis, and univariate analysis was performed to select the following factors for multivariate analysis: tumor size, lymph node metastasis, distant metastasis, tumor status, T stage, N stage, M stage, platelets, D-dimer, FDP, FIB, CA125, CRP, CA153, Radscore, and surgical treatment (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e). The results of the multivariate analysis showed that CRP (HR\u0026thinsp;=\u0026thinsp;1.010, 95% CI: 1.004\u0026ndash;1.019), Radscore (HR\u0026thinsp;=\u0026thinsp;2.960, 95% CI: 1.679\u0026ndash;5.225), and surgical treatment (HR\u0026thinsp;=\u0026thinsp;0.020, 95% CI: 0.054\u0026ndash;0.733) were independent risk factors significantly affecting the prognosis of NSCLC (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). Both CRP and the Radscore, along with surgical treatment, showed significant correlations with OS in patients with NSCLC (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\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\u003eClinical, pathological, and imaging characteristics of training and validation sets of non-small cell lung cancer patients\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\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 \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\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\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eValidation set\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSurvival\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDeath\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSurvival\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eDeath\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-Value\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBrain-metastases\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e75 (78.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (48.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33 (75.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7 (77.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 (21.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (51.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11 (25.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.804\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33 (34.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (44.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20 (45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e62 (65.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (55.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (54.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5 (55.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e60.0\u003c/p\u003e \u003cp\u003e[51.0;67.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e62.0\u003c/p\u003e \u003cp\u003e[58.0;69.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e58.0\u003c/p\u003e \u003cp\u003e[49.8;63.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e57.0 [54.0;68.0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.122\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.768\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e55 (57.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17 (58.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e26 (59.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e40 (42.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12 (41.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18 (40.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5 (55.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDrinking:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.879\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e69 (72.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22 (75.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (68.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6 (66.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26 (27.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (24.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (31.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFamily:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.974\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e91 (95.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28 (96.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42 (95.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9 (100%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 (4.21%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 (3.45%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (4.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLocation:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.94\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31 (32.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (24.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20 (45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1 (11.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9 (9.47%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (10.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (2.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19 (20.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (24.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9 (20.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21 (22.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (24.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (27.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15 (15.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5 (17.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (4.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEmpty:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.127\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e88 (92.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25 (86.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40 (90.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9 (100%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7 (7.37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (13.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (9.09%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLobation:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.586\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e65 (68.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (62.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31 (70.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6 (66.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (31.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11 (37.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (29.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBurr :\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.645\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e68 (71.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21 (72.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29 (65.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6 (66.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e27 (28.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (27.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15 (34.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePleural.traction:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.119\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e59 (62.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22 (75.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28 (63.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7 (77.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36 (37.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (24.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16 (36.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLNM:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e47 (49.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (20.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20 (45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e48 (50.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23 (79.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (54.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5 (55.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMetastasis:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e78 (82.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (51.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32 (72.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5 (55.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (17.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (48.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (27.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eType:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.212\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSquamous cell carcinoma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28 (29.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (34.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11 (25.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eadenocarcinoma\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e67 (70.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19 (65.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33 (75.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6 (66.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStage:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e38 (40.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 (3.45%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (31.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1 (11.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7 (7.37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (13.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (15.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (31.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (24.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (22.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 (21.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17 (58.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (29.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.034\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e37 (38.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (13.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20 (45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22 (23.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (34.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (22.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12 (12.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (20.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5 (11.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (25.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (31.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9 (20.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e45 (47.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5 (17.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (38.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (5.26%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (10.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (4.55%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e27 (28.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (44.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (27.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3 (33.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (17.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (27.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (29.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1 (1.05%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e76 (80.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (44.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32 (72.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5 (55.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19 (20.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (55.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (27.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOperation\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (44.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25 (86.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (54.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7 (77.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e53 (55.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (13.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20 (45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSurgery.for.brain.metastases:\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.435\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e93 (97.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26 (89.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e43 (97.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9 (100%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2 (2.11%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3 (10.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (2.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.963\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e43 (45.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12 (41.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19 (43.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e52 (54.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17 (58.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25 (56.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7 (77.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.526\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e72 (75.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19 (65.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (68.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7 (77.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23 (24.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (34.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14 (31.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.535\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e76 (80.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23 (79.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e37 (84.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9 (100%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19 (20.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6 (20.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (15.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0 (0.00%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\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\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.447\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e84 (88.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27 (93.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39 (88.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8 (88.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11 (11.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (6.90%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5 (11.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1 (11.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003e*0 represents none, 1 represents present.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable 2.1 Univariate analysis of survival related risk factors in non-small cell lung cancer patients\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 30.3797%;\"\u003e\n \u003cp\u003eCharacteristic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29.1139%;\"\u003e\n \u003cp\u003eHR-value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.2315%;\"\u003e\n \u003cp\u003e95%C-index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.2749%;\"\u003e\n \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003cem\u003e-value\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 30.3797%;\"\u003e\n \u003cp\u003eBrain-metastases*\u003c/p\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003cp\u003eSmoking\u003c/p\u003e\n \u003cp\u003eDrinking\u003c/p\u003e\n \u003cp\u003eFamily\u003c/p\u003e\n \u003cp\u003eNumber\u003c/p\u003e\n \u003cp\u003eSize\u003c/p\u003e\n \u003cp\u003eLocation\u003c/p\u003e\n \u003cp\u003eLung.cancer.size*\u003c/p\u003e\n \u003cp\u003eEmpty\u003c/p\u003e\n \u003cp\u003eLobation\u003c/p\u003e\n \u003cp\u003eBurr\u003c/p\u003e\n \u003cp\u003ePleural.traction\u003c/p\u003e\n \u003cp\u003eLNM*\u003c/p\u003e\n \u003cp\u003eMetastasis*\u003c/p\u003e\n \u003cp\u003eType*\u003c/p\u003e\n \u003cp\u003eStage\u003c/p\u003e\n \u003cp\u003eT*\u003c/p\u003e\n \u003cp\u003eN*\u003c/p\u003e\n \u003cp\u003eM*\u003c/p\u003e\n \u003cp\u003ePLT*\u003c/p\u003e\n \u003cp\u003eLeukocyte\u003c/p\u003e\n \u003cp\u003eNeutrophils\u003c/p\u003e\n \u003cp\u003eMonocyte\u003c/p\u003e\n \u003cp\u003eLymphocyte\u003c/p\u003e\n \u003cp\u003eC.reactive.protein*\u003c/p\u003e\n \u003cp\u003eD.Dimer*\u003c/p\u003e\n \u003cp\u003eFDP*\u003c/p\u003e\n \u003cp\u003eFIB*\u003c/p\u003e\n \u003cp\u003eNSE\u003c/p\u003e\n \u003cp\u003eCEA\u003c/p\u003e\n \u003cp\u003eCA199\u003c/p\u003e\n \u003cp\u003eCA125*\u003c/p\u003e\n \u003cp\u003eCA153*\u003c/p\u003e\n \u003cp\u003eSCC.Ag\u003c/p\u003e\n \u003cp\u003eCYFRA21.1\u003c/p\u003e\n \u003cp\u003eProGRP\u003c/p\u003e\n \u003cp\u003eOperation*\u003c/p\u003e\n \u003cp\u003eSurgery.for.brain.metastases\u003c/p\u003e\n \u003cp\u003eChemotherapy\u003c/p\u003e\n \u003cp\u003eTargeted.therapy\u003c/p\u003e\n \u003cp\u003eRadiotherapy\u003c/p\u003e\n \u003cp\u003eImmunotherapy\u003c/p\u003e\n \u003cp\u003eRadscore*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29.1139%;\"\u003e\n \u003cp\u003e2.245\u003c/p\u003e\n \u003cp\u003e0.906\u003c/p\u003e\n \u003cp\u003e1.027\u003c/p\u003e\n \u003cp\u003e1.113\u003c/p\u003e\n \u003cp\u003e0.934\u003c/p\u003e\n \u003cp\u003e1.045\u003c/p\u003e\n \u003cp\u003e1.048\u003c/p\u003e\n \u003cp\u003e1.211\u003c/p\u003e\n \u003cp\u003e1.066\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e1.297\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e2.250\u003c/p\u003e\n \u003cp\u003e1.228\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0.822\u003c/p\u003e\n \u003cp\u003e0.513\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e3.449\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e4.936\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0.613\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e2.474\u003c/p\u003e\n \u003cp\u003e1.351\u003c/p\u003e\n \u003cp\u003e1.648\u003c/p\u003e\n \u003cp\u003e4.744\u003c/p\u003e\n \u003cp\u003e1.004\u003c/p\u003e\n \u003cp\u003e1.104\u003c/p\u003e\n \u003cp\u003e0.997\u003c/p\u003e\n \u003cp\u003e0.907\u003c/p\u003e\n \u003cp\u003e0.970\u003c/p\u003e\n \u003cp\u003e1.014\u003c/p\u003e\n \u003cp\u003e1.138\u003c/p\u003e\n \u003cp\u003e1.030\u003c/p\u003e\n \u003cp\u003e1.366\u003c/p\u003e\n \u003cp\u003e1.0005\u003c/p\u003e\n \u003cp\u003e1.00006\u003c/p\u003e\n \u003cp\u003e0.991\u003c/p\u003e\n \u003cp\u003e1.001\u003c/p\u003e\n \u003cp\u003e1.016\u003c/p\u003e\n \u003cp\u003e0.688\u003c/p\u003e\n \u003cp\u003e0.995\u003c/p\u003e\n \u003cp\u003e1.008\u003c/p\u003e\n \u003cp\u003e0.124\u003c/p\u003e\n \u003cp\u003e1.613\u003c/p\u003e\n \u003cp\u003e0.977\u003c/p\u003e\n \u003cp\u003e1.284\u003c/p\u003e\n \u003cp\u003e0.745\u003c/p\u003e\n \u003cp\u003e0.572\u003c/p\u003e\n \u003cp\u003e5.045\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.2315%;\"\u003e\n \u003cp\u003e(1.075, 4.687)\u003c/p\u003e\n \u003cp\u003e(0.432, 1.900)\u003c/p\u003e\n \u003cp\u003e(0.992, 1.062)\u003c/p\u003e\n \u003cp\u003e(0.529, 2.338)\u003c/p\u003e\n \u003cp\u003e(0.398, 2.190)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;(0.141, 7.728)\u003c/p\u003e\n \u003cp\u003e(0.826, 1.330)\u003c/p\u003e\n \u003cp\u003e(0.926,1.585)\u003c/p\u003e\n \u003cp\u003e(0.831, 1.366)\u003c/p\u003e\n \u003cp\u003e(1.119, 1.504)\u003c/p\u003e\n \u003cp\u003e(0.773, 6.552)\u003c/p\u003e\n \u003cp\u003e(0.580, 2.602)\u003c/p\u003e\n \u003cp\u003e(0.363,1.858)\u003c/p\u003e\n \u003cp\u003e(0.219, 1.204)\u003c/p\u003e\n \u003cp\u003e(1.394, 8.531)\u003c/p\u003e\n \u003cp\u003e(2.260, 10.779)\u003c/p\u003e\n \u003cp\u003e(0.281,1.335)\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(1.590, 3.848)\u003c/p\u003e\n \u003cp\u003e(1.004, 1.819)\u003c/p\u003e\n \u003cp\u003e(1.187, 2.287)\u003c/p\u003e\n \u003cp\u003e(2.213, 10.172)\u003c/p\u003e\n \u003cp\u003e(1.0008,1.009)\u003c/p\u003e\n \u003cp\u003e(0.998,1.221)\u003c/p\u003e\n \u003cp\u003e(0.970,1.024)\u003c/p\u003e\n \u003cp\u003e(0.795,1.035)\u003c/p\u003e\n \u003cp\u003e(0.940,1.002)\u003c/p\u003e\n \u003cp\u003e(1.006,1.021)\u003c/p\u003e\n \u003cp\u003e(1.025,1.263)\u003c/p\u003e\n \u003cp\u003e(1.009,1.051)\u003c/p\u003e\n \u003cp\u003e(1.053,1.772)\u003c/p\u003e\n \u003cp\u003e(0.967,1.034)\u003c/p\u003e\n \u003cp\u003e(0.999,1.0006)\u003c/p\u003e\n \u003cp\u003e(0.973,1.010)\u003c/p\u003e\n \u003cp\u003e(1.0004,1.002)\u003c/p\u003e\n \u003cp\u003e(1.007,1.025)\u003c/p\u003e\n \u003cp\u003e(0.373,1.268)\u003c/p\u003e\n \u003cp\u003e(0.962,1.029)\u003c/p\u003e\n \u003cp\u003e(0.986,1.031)\u003c/p\u003e\n \u003cp\u003e(0.042,0.362)\u003c/p\u003e\n \u003cp\u003e(0.472,5.507)\u003c/p\u003e\n \u003cp\u003e(0.461,2.068)\u003c/p\u003e\n \u003cp\u003e(0.596,2.768)\u003c/p\u003e\n \u003cp\u003e(0.298,1.862)\u003c/p\u003e\n \u003cp\u003e(0.134,2.435)\u003c/p\u003e\n \u003cp\u003e(3.109,8.187)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.2749%;\"\u003e\n \u003cp\u003e0.032\u003c/p\u003e\n \u003cp\u003e0.795\u003c/p\u003e\n \u003cp\u003e0.113\u003c/p\u003e\n \u003cp\u003e0.777\u003c/p\u003e\n \u003cp\u003e0.874\u003c/p\u003e\n \u003cp\u003e0.965\u003c/p\u003e\n \u003cp\u003e0.697\u003c/p\u003e\n \u003cp\u003e0.194\u003c/p\u003e\n \u003cp\u003e0.612\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003cp\u003e0.175\u003c/p\u003e\n \u003cp\u003e0.593\u003c/p\u003e\n \u003cp\u003e0.633\u003c/p\u003e\n \u003cp\u003e0.106\u003c/p\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003cp\u003e0.121\u003c/p\u003e\n \u003cp\u003e0.046\u003c/p\u003e\n \u003cp\u003e0.0018\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003cp\u003e0.025\u003c/p\u003e\n \u003cp\u003e0.075\u003c/p\u003e\n \u003cp\u003e0.850\u003c/p\u003e\n \u003cp\u003e0.156\u003c/p\u003e\n \u003cp\u003e0.066\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003cp\u003e0.039\u003c/p\u003e\n \u003cp\u003e0.032\u003c/p\u003e\n \u003cp\u003e0.023\u003c/p\u003e\n \u003cp\u003e0.976\u003c/p\u003e\n \u003cp\u003e0.829\u003c/p\u003e\n \u003cp\u003e0.246\u003c/p\u003e\n \u003cp\u003e0.032\u003c/p\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003cp\u003e0.111\u003c/p\u003e\n \u003cp\u003e0.778\u003c/p\u003e\n \u003cp\u003e0.481\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003cp\u003e0.471\u003c/p\u003e\n \u003cp\u003e0.952\u003c/p\u003e\n \u003cp\u003e0.527\u003c/p\u003e\n \u003cp\u003e0.517\u003c/p\u003e\n \u003cp\u003e0.414\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e*LNM=lymph node metastasis;PLT=platelet;FDP=Fibrin Degradation Products；FIB=Fibrinogen；NSE=Neuron-Specific Enolase；CEA=Carcinoembryonic antigen\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 2.2 Multivariate analysis of risk factors related to overall survival in non-small cell lung cancer\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" class=\"fr-table-selection-hover\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 30.3797%;\"\u003e\n \u003cp\u003eCharacteristic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29.1139%;\"\u003e\n \u003cp\u003eHR-value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.2315%;\"\u003e\n \u003cp\u003e95%C-index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.2749%;\"\u003e\n \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003cem\u003e-value\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 30.3797%;\"\u003e\n \u003cp\u003ereactive.protein\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eOperation\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eRadscore\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 29.1139%;\"\u003e\n \u003cp\u003e1.010\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0.200\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e2.960\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.2315%;\"\u003e\n \u003cp\u003e(1.004, 1.019)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;(0.054, 0.733)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e(1.679, 5.225)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 16.2749%;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0.015\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\u003cp\u003eIn the high-risk group, the median survival time was 11 months (range: 8\u0026ndash;25 months), with postoperative survival rates at 1, 2, and 3 years of 74.87%, 57.55%, and 40.10%, respectively. In contrast, the low-risk group had a median survival period of 30 months (range: 10\u0026ndash;38 months), with postoperative survival rates at 1, 2, and 3 years of 97.70%, 95.23%, and 92.59%, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e3\u003c/span\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 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSurvival Probability and Median Survival Time\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\u003eGrouping\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOne year survival probability\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTwo year survival probability\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eThree year survival probability\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003emedian survival\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh-risk group\u003c/p\u003e \u003cp\u003eLow-risk group\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.748(0.645\u0026ndash;0.868)\u003c/p\u003e \u003cp\u003e0.977(0.933-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.575(0.443\u0026ndash;0.746)\u003c/p\u003e \u003cp\u003e0.952(0.889-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.401(0.232\u0026ndash;0.691)\u003c/p\u003e \u003cp\u003e0.925(0.848-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11(8-25.2)\u003c/p\u003e \u003cp\u003e30(10\u0026ndash;38)\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=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eScreening results of radiomics features\u003c/h2\u003e \u003cp\u003eA total of 1,688 radiomics features were extracted from the VOI of tumors in the CT images. Spearman correlation analysis was used to remove features with a correlation coefficient above 0.9. Following univariate Cox analysis features with p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were retained, resulting in 137 features. Thereafter, COX-LASSO regression and tenfold cross-validation were used for feature selection, leading to the identification of 16 key radiomics features (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eKey radiomics features and their correlation coefficients\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRadiomics features\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCoefficient\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKurtosis_firstorder_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMean_firstorder_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.068\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJointAverage_glcm_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.154\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDifferenceEntropy_glcm_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.022\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClusterProminence_glcm_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDifferenceAverage_glcm_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLargeDependenceHighGrayLevelEmphasis_gldm_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.31\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLongRunLowGrayLevelEmphasis_glrlm_original\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.199\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRootMeanSquared_firstorder_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.178\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJointEntropy_glcm_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJointEnergy_glcm_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIdm_glcm_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.008\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDifferenceAverage_glcm_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLargeDependenceEmphasis_gldm_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.025\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLargeDependenceLowGrayLevelEmphasis_gldm_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.065\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmallDependenceLowGrayLevelEmphasis_gldm_logarithm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.332\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=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePredictive model efficiency\u003c/h2\u003e \u003cp\u003eKM curves were generated based on three independent risk factors (CRP, Radscore and surgical treatment) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The KM curve graph showed that, regardless of whether in the training set or the validation set, the survival rates of patients in the low-risk group at 12, 24, and 36 months after surgery were significantly higher than those in the high-risk group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Notably, the survival rates of patients with low risk at 12, 24,and 36-months post-surgery were close to 1. Over time, the number of surviving patients in both the high-risk and low-risk groups decreased.\u003c/p\u003e \u003cp\u003eBased on multivariate Cox regression analysis, clinical models, radiomics models and combined models were constructed separately. The AUC of the clinical model in the training set and validation set were 0.858 (95% CI: 0.789\u0026ndash;0.927) and 0.775 (95% CI: 0.614\u0026ndash;0.935), respectively. The AUC values of the radiomics model in the training and validation sets were 0.859 (95% CI: 0.805\u0026ndash;0.914) and 0.778 (95% CI: 0.67\u0026ndash;0.887), respectively. The AUC values of the combined model in the training and validation sets were 0.892 (95% CI: 0.848\u0026ndash;0.936) and 0.838 (95% CI: 0.71\u0026ndash;0.965), respectively. The AUC values of both the radiomics model and the combined model were higher than those of the clinical model, with the combined model demonstrating the highest AUC values in both sets. Indicating that the predictive performance of the combined model is significantly superior to that of individual clinical and radiomics models. As shown in Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e5\u003c/span\u003eA and B, the calibration curve demonstrates a good fit, indicating good consistency between the predicted and actual values of the prediction model, as well as high predictive accuracy. The final research results showed that in both the training and validation sets, the comprehensive prediction models exhibited high AUC values for predicting patients' 1, 2, and 3 year survival rates, demonstrating good predictive performance, as shown in Figs.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and B. Since this is a small data set, we use bootstrapping to test the robustness and stability of features. Through 200 bootstrapping samples, the average C-index is 0.869/0.705 (the effectiveness of model prediction at 36 months, training/validation set).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eEvaluation of clinical utility\u003c/h2\u003e \u003cp\u003eThe clinical utility of radiomics model, clinical model and combined model was evaluated using a DCA, and the decision curve analysis results of these models in the training and validation sets are provided in Figs.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e7\u003c/span\u003eA and \u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e7\u003c/span\u003eB, respectively. DCA revealed that all three prediction models were clinically useful in both sets and the combined model had high overall net benefits within a reasonable threshold probability range.\u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThe aim of this study is to establish a nomogram model based on CT radiomics combined with clinicopathological factors to predict the survival rates of patients with NSCLC at 1, 2, and 3 years. Additionally, the study evaluates the survival rate of high-risk and low-risk groups using KM curves. By combining CT radiomics features with clinicopathological factors, univariate and multivariate Cox regression analyses were conducted on each variable to identify independent risk factors that affect patient survival. First, Cox regression was used to establish a clinical model based on surgical treatment and CRP. Radiomics features were then incorporated into this clinical model to establish a combined model. According to the median value of the Radscore, patients were divided into high-risk and low-risk groups. The column chart model established through three independent risk factors effectively visualized the prediction results and independent risk factors. The Cox regression analysis of the comprehensive model showed a C-index of 0.892 (range: 0.848\u0026ndash;0.936) for the training set and 0.838 (range 0.710\u0026ndash;0.965) for the validation set, demonstrating the good predictive performance of our model. The median survival time of high-risk patients with NSCLC was 11 months (range: 8\u0026ndash;25), with 1, 2, and 3 year survival rates of 74.87%, 57.55%, and 40.10%, respectively. In contrast, the median survival period of low-risk patients was 30 months (range: 10\u0026ndash;38), with 1, 2, and 3 year survival rates of 97.70%, 95.23%, and 92.59%, respectively. These findings indicate that the 1, 2, and 3year survival rates of low-risk patients are significantly higher. The KM curve analysis further supports this, showing that in both the training and validation sets, the survival rates of patients with NSCLC low-risk group at 1, 2, and 3 years are close to 100%. The predictive model established in this study provides valuable guidance for risk stratification and personalized diagnosis and treatment of patients with NSCLC, providing a basis for additional treatment options and a close follow-up basis for patients at high risk.\u003c/p\u003e \u003cp\u003eIn addition, this study conducted univariate and multivariate Cox regression analyses on clinicopathological factors, and the results showed that CRP (HR\u0026thinsp;=\u0026thinsp;1.010, 95% CI: 1.004\u0026ndash;1.019), Radscore (HR\u0026thinsp;=\u0026thinsp;2.960, 95% CI: 1.679\u0026ndash;5.225), and surgical treatment (HR\u0026thinsp;=\u0026thinsp;0.020, 95% CI: 0.054\u0026ndash;0.733) were independent risk factors affecting the prognosis of NSCLC. Previous studies have also shown that multiple factors can affect the survival of patients with NSCLC\u003csup\u003e[\u003cspan additionalcitationids=\"CR34 CR35 CR36 CR37 CR38\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e. For example, Guo et al.\u003csup\u003e[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e found that monitoring D-Dimer levels and addressing blood hypercoagulability in patients with NSCLC across different TNM stages can aid in early prevention and treatment, thereby enhancing patient survival rates. CRP is a common acute inflammatory response protein synthesized by liver cells and released into the bloodstream. Its relationship with malignant tumors is not fully understood\u003csup\u003e[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e. Buresova et al.\u003csup\u003e[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e found that CRP may affect the level of circulating tumor DNA (ctDNA) in NSCLC, thereby affecting the prognosis of patients through an investigation of the potential association between tumor markers, laboratory parameters, ctDNA, and survival rate. Wang et al.\u003csup\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e conducted a meta-analysis on the correlation between serum CRP levels and NSCLC prognosis and discovered that patients with elevated serum CRP levels had lower five-year survival rates. Frey et al.\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/sup\u003e studied 52 patients with primary NSCLC (UICC stage III) and found that a high baseline CRP to albumin ratio was associated with shorter progression-free survival (PFS) (p\u0026thinsp;=\u0026thinsp;0.038) and OS (p\u0026thinsp;=\u0026thinsp;0.022). This is consistent with the findings of Nassar\u003csup\u003e[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e, who found that the CRP concentration during the third treatment cycle and the difference in CRP concentration between the third and second treatment cycles were the strongest predictors of PFS and OS. Measuring longitudinal CRP can monitor inflammation levels and serve as a promising prognostic marker, particularly as inflammation levels decrease during treatment cycles. This finding is consistent with Zheng\u0026rsquo;s study\u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e, which noted that an increase in baseline CRP levels is significantly correlated with decreased PFS, especially in patients receiving chemotherapy combined with immune checkpoint inhibitors (ICI) treatment. While the predictive value of CRP is prominent, it may vary depending on the treatment regimen. Our research supports the potential of CRP as a broad prognostic biomarker, providing new insights into the complex interactions between inflammatory markers and cancer treatment responses and emphasizing the potential of CRP as a predictive and prognostic biomarker for NSCLC. In addition, these findings further highlight the necessity of exploring other variables that affect OS and validating these results in a broader patient population. Radiomics can be used to comprehensively and quantitatively evaluate the spatiotemporal heterogeneity of tumors, and when combined with clinicopathological factors, its predictive performance for prognosis may be improved\u003csup\u003e[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]\u003c/sup\u003e. Zheng et al.\u003csup\u003e[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]\u003c/sup\u003e showed that CT-based deep learning models exhibit good performance in predicting OS in patients with NSCLC. In terms of treatment, radiomics can non-invasively evaluate EGFR mutation status\u003csup\u003e[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]\u003c/sup\u003e, support targeted therapy, and predict microvascular invasion in patients with stage I NSCLC before surgery\u003csup\u003e[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]\u003c/sup\u003e, aiding in surgical method selection and individualized treatment planning. The study by Yang et al.\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e confirmed that a line chart based on \u003csup\u003e18\u003c/sup\u003eF-FDG PET/CT Radscore and clinicopathological factors has good predictive power for prognosis, effectively guiding individualized treatment for patients with NSCLC. This finding is highly consistent with our results. The study by Gonz\u0026aacute;lez et al.\u003csup\u003e[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]\u003c/sup\u003e showed that patients who underwent surgery had significantly longer survival times (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), also consistent with our findings. In the training set, 53 out of 95 patients who did not undergo surgery died, with a mortality rate of 55.8%; among the 29 patients who underwent surgery, four died, with a mortality rate of 13.8%. In the validation set, 20 out of 44 patients who did not undergo surgery died, with a mortality rate of 45.5%; among the nine patients who underwent surgery, two died, with a mortality rate of 22.2%. Therefore, surgical treatment is considered an independent protective factor for patients with NSCLC. Nakwan et al.\u003csup\u003e[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]\u003c/sup\u003e conducted a study on the survival factors among patients with cancer in a non-university hospital in Thailand from 2012 to 2021, confirming that surgical resection is an important predictor of patients with cancer survival. At present, ICI treatment has greatly changed the treatment prospects for NSCLC\u003csup\u003e[\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]\u003c/sup\u003e. The US Food and Drug Administration (FDA) approved nivolumab and pembrolizumab for the treatment of lung squamous cell carcinoma in 2015. In October 2016, the FDA authorized pembrolizumab as a first-line treatment for patients with PD-L1 overexpression (\u0026ge;\u0026thinsp;50%); atezolizumab has also been approved for patients with advanced NSCLC who have progressed after chemotherapy\u003csup\u003e[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]\u003c/sup\u003e. Although immunotherapy (HR\u0026thinsp;=\u0026thinsp;0.572, 95% CI: 0.134\u0026ndash;2.435, p\u0026thinsp;=\u0026thinsp;0.414) did not show statistical significance in our study, it does not mean that immunotherapy lacks value in NSCLC prognosis. At present, the application of immunotherapy in NSCLC primarily focuses on patients at the late stage and has shown good results. Lv et al.\u003csup\u003e[\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]\u003c/sup\u003e have shown that immunotherapy can reshape tumor microenvironment to inhibit tumor cells, block CTLA-4 and PD-1/PD-L1 immune checkpoints to alleviate T cell functional inhibition, and facilitate tumor clearance through the body's immune system. Its application in patients with late-stage NSCLC has broad clinical prospects.\u003c/p\u003e \u003cp\u003eFinally, we constructed a nomogram by combining the Radscore with clinicopathological factors. The calibration curve showed that the predicted survival probability closely aligns with the actual survival time of the patients, thereby validating the accuracy of the model. The validation of the OS nomogram demonstrated that as follow-up time increases, the AUC for predicting prognosis also gradually increases, further demonstrating the good predictive performance of the OS nomogram. Additionally, we used KM analysis to evaluate the reliability of the OS nomogram in predicting patient survival rates. The results of the KM analysis showed that the OS nomogram effectively distinguishes between patients at high and low risk, indicating its efficacy in predicting patients at high and low risk. Therefore, this nomogram is considered a robust and reliable model that can serve as strong evidence for additional treatment and close follow-up for patients with poor prognoses. A nomogram model based on CT radiomics characteristics and clinical pathological factors was constructed to achieve accurate risk stratification of NSCLC patients by integrating multi-dimensional data such as imaging score, surgical treatment and C-reactive protein. The model can guide the formulation of individualized treatment strategies: the high-risk group (such as stage Ⅲ with lymph node metastasis) is recommended to receive intensive treatment (neoadjuvant immunotherapy combined with chemotherapy), and the low-risk group (such as stage Ⅰ a well differentiated adenocarcinoma) is suitable for step-down treatment (surgery or targeted therapy). Through the accurate prediction of the 1, 2, and 3 year survival rates of patients with non-small cell lung cancer, dynamic monitoring of curative effect is achieved to adjust the optimization scheme, and its visual scoring system significantly improves the treatment accuracy and clinical net benefit\u003csup\u003e[\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCompared with previous studies, this study has the following advantages. First, this study extracts radiomics features based on CT and combines them with clinical pathological factors. Multimodal receipts help to improve the accuracy of the prediction model. Secondly, the nomogram model constructed by integrating multi-dimensional variables achieved a high AUC value in the prognosis prediction of NSCLC (AUC\u0026thinsp;=\u0026thinsp;0.892 in the training set and AUC\u0026thinsp;=\u0026thinsp;0.838 in the validation set), and achieved individualized risk stratification through the visual scoring system, which has the advantages of high accuracy, interpretability and clinical practicability\u003csup\u003e[\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]\u003c/sup\u003e. Meanwhile, our study has several limitations. First, it is a retrospective study, and there may be selection bias. Second, the follow-up period is relatively short, and some outcome events have not yet occurred, which may result in loss-to-follow-up bias. Third, as a single-center study, the generalizability of the results may be limited. In future research, we will further expand the sample size and conduct multicenter and prospective studies to explore the application of artificial intelligence in predicting the prognosis of NSCLC. This approach will help provide more robust support for improving the quality of life of patients with NSCLC.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eThe nomogram constructed by combining CT radiomics features with clinicopathological factors shows good predictive efficacy and application value in predicting the prognosis of patients with NSCLC. This model can effectively evaluate the prognosis of patients and provide important guidance for improving the quality of life of patients with NSCLC. In addition, the significantly higher survival rates at 1, 2, and 3 years post-surgery underscore the positive impact of surgical treatment on patient prognosis.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eCT：Computed tomography；NSCLC：non-small-cell lung cancer；VOI：volume of interest；OS：Overall survival；CRP： C-reactive protein；\u003cstrong\u003eICI\u003c/strong\u003e\u003cstrong\u003e：\u003c/strong\u003e immune-checkpoint-inhibitor\u003cstrong\u003e；\u003c/strong\u003ePFS：progression-free survival；TNM：Tumor-Node-Metastasis；MTV：Metabolic Tumor Volume；TLG：Total Lesion Glycolysis；LVI：lymphovascular invasion；AUC：area under curve；iPFS：intracranial progression free survival；KM：Kaplan-Meier；HR：Hazard ratio；ICC：Intra-class Correlation Coefficient；Radscore：radiomics score; TNM: tumor-node-metastasis; FDP: degradation products; NSE:neuron-specific enolase; CEA:carcinoembryonic antigen; CA125:cancer antigen 125; CA153:cancer antigen 153;MRI：magnetic resonance imaging；GLCM：Gray level co-occurrence matrix；NGTDM：Neighborhood Gray Tone Difference Matrix； GLRLM：Gray Level Run Length Matrix； GLSZM：Gray Level Size Zone Matrix ；GLDM：Gray Level Dependence Matrix；DCA：Decision curve analysis；ROC：receiver operating characteristic；ROI：region of interest；PH：Proportional Hazards；ctDNA：circulating tumor DNA\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Ethics Committee of the First People's Hospital of Kunming(YLS2023-75), and patients' informed consent was waived. All methods were carried out in accordance with relevant guidelines and regulations.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePatients signed informed consent regarding publishing their clinical data and images.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eFunding\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study was supported by National Natural Science Foundation of China Project (No. 82160348)；Yunnan Province Major Special Plan (No. 202302AA310018-D-8); Yunnan Province's \"Xingdian Talent Support Program\" Youth Talent Project (No. XDYC-QNRC-2022-0608)；Beijing Medical Award Foundation Ruiying Fund（No.\u0026nbsp;YXJL-2022-0665-0216）;Yunnan Provincial Department of Education Science Research Fund Project(No. 2025Y0383).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWang Yubo conceived the idea for this research. Wang Yubo, Peng Zefei, and Hu Hao were responsible for data collection. Ding Zhiyong, Li Xiandou，Li Jiageng and Fu Yang performed the image analysis. Wang Yubo wrote the manuscript, while Zhang Yisong、Xie bosen and Kong Mengxue conducted statistical analysis. Dr. Yang Bin edited and reviewed the manuscript. All authors discussed the results and provided comments on the manuscript. All authors contributed to this article and approved the final submitted version.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLi C, Lei S, Ding L, et al. Global burden and trends of lung cancer incidence and mortality. Chin Med J. 2023;136:1583\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/CM9.0000000000002529\u003c/span\u003e\u003cspan address=\"10.1097/CM9.0000000000002529\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEttinger DS, Wood DE, Aisner DL, et al. Non-Small Cell Lung Cancer, Version 3.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2022;20:497\u0026ndash;530. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.6004/jnccn.2022.0025\u003c/span\u003e\u003cspan address=\"10.6004/jnccn.2022.0025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchabath MB, Cote ML. Cancer Progress and Priorities: Lung Cancer. Cancer Epidemiol Biomarkers Prev. 2019;28:1563\u0026ndash;79. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1158/1055-9965.EPI-19-0221\u003c/span\u003e\u003cspan address=\"10.1158/1055-9965.EPI-19-0221\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJeon DS, Kim HC, Kim SH, et al. Five-year overall survival and prognostic factors in patients with lung cancer: results from the Korean Association of Lung Cancer Registry (KALC-R) 2015. CancerRes Treat. 2023;55:103\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4143/crt.2022.264\u003c/span\u003e\u003cspan address=\"10.4143/crt.2022.264\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArriola E, Trigo JM, S\u0026aacute;nchez-Gastaldo A, et al. Prognostic Value of Clinical Staging According to TNM in Patients With SCLC: A Real-World Surveillance Epidemiology and End-Results Database Analysis. JTO Clin Res Rep. 2021;3:100266. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jtocrr.2021.100266\u003c/span\u003e\u003cspan address=\"10.1016/j.jtocrr.2021.100266\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang P, Chen PL, Li ZH, et al. Association of smoking and polygenic risk with the incidence of lung cancer: a prospective cohort study. Br J Cancer. 2022;126:1637\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41416-022-01736-3\u003c/span\u003e\u003cspan address=\"10.1038/s41416-022-01736-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrant WB. Cancer Incidence Rates in the US in 2016\u0026ndash;2020 with Respect to Solar UVB Doses, Diabetes and Obesity Prevalence, Lung Cancer Incidence Rates, and Alcohol Consumption: An Ecological Study. Nutrients. 2024;16:1450doi. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/nu16101450\u003c/span\u003e\u003cspan address=\"10.3390/nu16101450\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAng L, Chan CPY, Yau WP, et al. Association between family history of lung cancer and lung cancer risk: a systematic review and meta-analysis. Lung Cancer. 2020;148:129\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.lungcan.2020.08.012\u003c/span\u003e\u003cspan address=\"10.1016/j.lungcan.2020.08.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRami-Porta R, Nishimura KK, Giroux DJ, et al. The International Association for the Study of Lung Cancer Lung Cancer Staging Project: Proposals for Revision of the TNM Stage Groups in the Forthcoming (Ninth) Edition of the TNM Classification for Lung Cancer. J Thorac Oncol. 2024;19:1007\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jtho.2024.02.011\u003c/span\u003e\u003cspan address=\"10.1016/j.jtho.2024.02.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang B, Zhong J, Zhong J, et al. Development and Validation of a Radiomics Nomogram Based on \u003csup\u003e18\u003c/sup\u003eF-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography and Clinicopathological Factors toPredict the Survival Outcomes of Patients With Non-Small Cell Lung Cancer. Front Oncol. 2020;10:1042. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fonc.2020.01042\u003c/span\u003e\u003cspan address=\"10.3389/fonc.2020.01042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCascone T, Fradette J, Pradhan M, et al. Tumor Immunology and Immunotherapy of Non-Small-Cell Lung Cancer. Cold Spring Harb Perspect Med. 2022;12:a037895. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1101/cshperspect.a037895\u003c/span\u003e\u003cspan address=\"10.1101/cshperspect.a037895\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRen S, Zeng G, Yi Y, et al. Combinations of plasma cfDNA concentration, integrity and tumor markers are promising biomarkers for early diagnosis of non-small cell lung cancer. Heliyon. 2023;9:e20851. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.heliyon.2023.e20851\u003c/span\u003e\u003cspan address=\"10.1016/j.heliyon.2023.e20851\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao X, Zheng YZ, Liao HY, et al. A clinical nomogram and heat map for assessing survival in patients with stage I non-smallcell lung cancer after complete resection. Ther Adv Med Oncol. 2020;12:1758835920970063. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1177/1758835920970063\u003c/span\u003e\u003cspan address=\"10.1177/1758835920970063\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReck M, Remon J, Hellmann MD. First-Line Immunotherapy for Non-Small-Cell Lung Cancer. J Clin Oncol. 2022;40:586\u0026ndash;97. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1200/JCO.21.01497\u003c/span\u003e\u003cspan address=\"10.1200/JCO.21.01497\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatel SA, Weiss J. Advances in the Treatment of Non-Small Cell Lung Cancer: Immunotherapy. Clin Chest Med. 2020;41:237\u0026ndash;47. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ccm.2020.02.010\u003c/span\u003e\u003cspan address=\"10.1016/j.ccm.2020.02.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNasser NJ, Gorenberg M, Agbarya A. First line Immunotherapy for Non-Small Cell Lung Cancer. Pharmaceuticals (Basel). 2020;13:373. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/ph13110373\u003c/span\u003e\u003cspan address=\"10.3390/ph13110373\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang T, Yang Y, Liu X, et al. Primary Invasive Mucinous Adenocarcinoma of the Lung: Prognostic Value of CT Imaging Features Combined with Clinical Factors. Korean J Radiol. 2021;22:652\u0026ndash;62. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3348/kjr.2020.0454\u003c/span\u003e\u003cspan address=\"10.3348/kjr.2020.0454\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJain P, Khorrami M, Gupta A, et al. Novel Non-Invasive Radiomic Signature on CT Scans Predicts Response to Platinum-Based Chemotherapy and Is Prognostic of Overall Survival in Small Cell Lung Cancer. Front Oncol. 2021;11:744724. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fonc.2021.744724\u003c/span\u003e\u003cspan address=\"10.3389/fonc.2021.744724\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMonaco L, Gemelli M, Gotuzzo I, et al. Metabolic Parameters as Biomarkers of Response to Immunotherapy and Prognosis in Non-Small Cell Lung Cancer (NSCLC): A Real World Experience. Cancers (Basel). 2021;13:1634. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cancers13071634\u003c/span\u003e\u003cspan address=\"10.3390/cancers13071634\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTankyevych O, Trousset F, Latappy C, et al. Development of Radiomic-Based Model to Predict Clinical Outcomes in Non-Small Cell Lung Cancer Patients Treated with Immunotherapy. Cancers (Basel). 2022;14:5931. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cancers14235931\u003c/span\u003e\u003cspan address=\"10.3390/cancers14235931\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaad MB, Hong L, Aminu M, et al. Predicting benefit from immune checkpoint inhibitors in patients with non-small-cell lung cancer by CT-based ensemble deep learning: a retrospective study. Lancet Digit Health. 2023;5:e404\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S2589-7500(23)00082-1\u003c/span\u003e\u003cspan address=\"10.1016/S2589-7500(23)00082-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoon JW, Yang E, Kim JH, et al. Predicting Non-Small-Cell Lung Cancer Survival after Curative Surgery via Deep Learning of Diffusion MRI. Diagnostics (Basel). 2023;13:2555. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/diagnostics13152555\u003c/span\u003e\u003cspan address=\"10.3390/diagnostics13152555\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi B, Yang L, Zhang H, et al. Outcome-Supervised Deep Learning on Pathologic Whole Slide Images for Survival Prediction of Immunotherapy in Patients with Non-Small Cell Lung Cancer. Mod Pathol. 2023;36:100208. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.modpat.2023.100208\u003c/span\u003e\u003cspan address=\"10.1016/j.modpat.2023.100208\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan H, Niimi T, Matsunaga T, et al. Preoperatively predicting survival outcome for clinical stage IA pure-solid non-small cell lung cancer by radiomics-based machine learning. J Thorac Cardiovasc Surg. 2024;S0022. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e5223(24)00441-0.doi: 10.1016/j.jtcvs.2024.05.010\u003c/span\u003e\u003cspan address=\"5223(24)00441-0.doi: 10.1016/j.jtcvs.2024.05.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDercle L, Fronheiser M, Rizvi NA, et al. Baseline Radiomic Signature to Estimate Overall Survival in Patients With NSCLC. J Thorac Oncol. 2023;18:587\u0026ndash;98. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jtho.2022.12.019\u003c/span\u003e\u003cspan address=\"10.1016/j.jtho.2022.12.019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYao W, Liao Y, Li X, et al. Noninvasive Method for Predicting the Expression of Ki67 and Prognosis in Non-Small-Cell Lung Cancer Patients: Radiomics. J Healthc Eng. 2022;2022:7761589. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1155/2022/7761589\u003c/span\u003e\u003cspan address=\"10.1155/2022/7761589\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Q, Shao J, Xue T, et al. Intratumoral and peritumoral radiomics nomograms for the preoperative prediction of lymphovascular invasion and overall survival in non-small cell lung cancer. Eur Radiol. 2023;33:947\u0026ndash;58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00330-022-09109-3\u003c/span\u003e\u003cspan address=\"10.1007/s00330-022-09109-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQi H, Hou Y, Zheng Z, et al. Clinical characteristics and MRI based radiomics nomograms can predict iPFS and short-term efficacy of third-generation EGFR-TKI in EGFR-mutated lung adenocarcinoma with brain metastases. BMC Cancer. 2024;24:362. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12885-024-12121-z\u003c/span\u003e\u003cspan address=\"10.1186/s12885-024-12121-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu J, Zhou Y, Xu C, et al. Effectiveness of CT radiomic features combined with clinical factors in predicting prognosis in patients with limited-stage small cell lung cancer. BMC Cancer. 2024;24:170. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12885-024-11862-1\u003c/span\u003e\u003cspan address=\"10.1186/s12885-024-11862-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu G, Liu X, Li Y, et al. The nomograms for predicting overall and cancer-specific survival in elderly patients with early-stage lung cancer: A population-based study using SEER database. Front Public Health. 2022;10:946299. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpubh.2022.946299\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2022.946299\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhu J, Shi H, Ran H, et al. Development and Validation of a Nomogram for Predicting Overall Survival in Patients with Second Primary Small Cell Lung Cancer After Non-Small Cell Lung Cancer: A SEER-Based Study. Int J Gen Med. 2022;15:3613\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2147/IJGM.S353045\u003c/span\u003e\u003cspan address=\"10.2147/IJGM.S353045\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang B, Teng M, You H, et al. A Nomogram for Predicting Survival in Advanced Non-Small-Cell Lung Carcinoma Patients: A Population-Based Study. Cancer Invest. 2023;41:672\u0026ndash;85. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/07357907.2023.2241547\u003c/span\u003e\u003cspan address=\"10.1080/07357907.2023.2241547\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo J, Gao Y, Gong Z, et al. Plasma D-Dimer Level Correlates with Age, Metastasis, Recurrence, Tumor-Node-Metastasis Classification (TNM), and Treatment of Non-Small-Cell Lung Cancer (NSCLC) Patients. Biomed Res Int. 2021;2021:9623571doi. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1155/2021/9623571\u003c/span\u003e\u003cspan address=\"10.1155/2021/9623571\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHart PC, Rajab IM, Alebraheem M, et al. C-Reactive Protein and Cancer-Diagnostic and Therapeutic Insights. Front Immunol. 2020;11:595835. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2020.595835\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2020.595835\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuresova M, Benesova L, Minarik M, et al. Circulating Tumor DNA correlates with Lactate Dehydrogenase, CYFRA 21\u0026thinsp;\u0026ndash;\u0026thinsp;1, and CRP levels in patients with advanced NSCLC. J Cancer. 2023;14:1\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7150/jca.78574\u003c/span\u003e\u003cspan address=\"10.7150/jca.78574\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang H, Yang R, Zhou K, et al. Association between pretreatment C-reactive protein level and survival in non-small cell lung cancer patients treated with immune checkpoint inhibitors: A meta-analysis. Int Immunopharmacol. 2023;124:110937. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.intimp.2023.110937\u003c/span\u003e\u003cspan address=\"10.1016/j.intimp.2023.110937\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrey A, Martin D, D'Cruz L, et al. C-Reactive Protein to Albumin Ratio as Prognostic Marker in Locally Advanced Non-Small Cell Lung Cancer Treated with Chemoradiotherapy. Biomedicines. 2022;10:598. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/biomedicines10030598\u003c/span\u003e\u003cspan address=\"10.3390/biomedicines10030598\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNassar YM, Ojara FW, P\u0026eacute;rez-Pitarch A, et al. C-Reactive Protein as an Early Predictor of Efficacy in Advanced Non-Small-Cell Lung Cancer Patients: A Tumor Dynamics-Biomarker Modeling Framework. Cancers (Basel). 2023;15:5429. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/cancers15225429\u003c/span\u003e\u003cspan address=\"10.3390/cancers15225429\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng X, Zhang L, Wu L, et al. Baseline C-reactive protein predicts efficacy of the first-line immune checkpoint inhibitors plus chemotherapy in advanced lung squamous cell carcinoma: a retrospective, multicenter study. BMC Cancer. 2023;23:1244. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12885-023-11737-x\u003c/span\u003e\u003cspan address=\"10.1186/s12885-023-11737-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVanguri RS, Luo J, Aukerman AT, et al. Multimodal integration of radiology, pathology and genomics for prediction of response to PD-(L)1 blockade in patients with non-small cell lung cancer. Nat Cancer. 2022;3:1151\u0026ndash;64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s43018-022-00416-8\u003c/span\u003e\u003cspan address=\"10.1038/s43018-022-00416-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Yang Z, Chen R, et al. Histopathology images-based deep learning prediction of prognosis and therapeutic response in small cell lung cancer. NPJ Digit Med. 2024;7:15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41746-024-01003-0\u003c/span\u003e\u003cspan address=\"10.1038/s41746-024-01003-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng X, Liu K, Shen N, et al. Predicting overall survival and prophylactic cranial irradiation benefit in small-cell lung cancer with CT-based deep learning: A retrospective multicenter study. Radiother Oncol. 2024;195:110221. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.radonc.2024.110221\u003c/span\u003e\u003cspan address=\"10.1016/j.radonc.2024.110221\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFelfli M, Liu Y, Zerka F, et al. Systematic Review, Meta-Analysis and Radiomics Quality Score Assessment of CT Radiomics-Based Models Predicting Tumor EGFR Mutation Status in Patients with Non-Small-Cell Lung Cancer. Int J Mol Sci. 2023;24:11433. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/ijms241411433\u003c/span\u003e\u003cspan address=\"10.3390/ijms241411433\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeng L, Tang HZ, Luo YW, et al. Preoperative CT Radiomics Nomogram for Predicting Microvascular Invasion in Stage I Non-Small Cell Lung Cancer. Acad Radiol. 2024;31:46\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.acra.2023.05.015\u003c/span\u003e\u003cspan address=\"10.1016/j.acra.2023.05.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonz\u0026aacute;lez M, Calvo V, Redondo I, et al. Overall survival for early and locally advanced non-small-cell lung cancer from one institution: 2000\u0026ndash;2017. Clin Transl Oncol. 2021;23:1325\u0026ndash;33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12094-020-02521-5\u003c/span\u003e\u003cspan address=\"10.1007/s12094-020-02521-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakwan N, Kumsuk K. Survival Analysis of Lung Cancer: A 10-Year Real-Life Experience in a Non-University-Based Hospital in Thailand (2012\u0026ndash;2021). Asian Pac J Cancer Prev. 2023;24:3021\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.31557/APJCP.2023.24.9.3021\u003c/span\u003e\u003cspan address=\"10.31557/APJCP.2023.24.9.3021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen P, Liu Y, Wen Y, et al. Non-small cell lung cancer in China. Cancer Commun (Lond). 2022;42:937\u0026ndash;70. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/cac2.12359\u003c/span\u003e\u003cspan address=\"10.1002/cac2.12359\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLahiri A, Maji A, Potdar PD, et al. Lung cancer immunotherapy: progress, pitfalls, and promises. Mol Cancer. 2023;22:40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12943-023-01740-y\u003c/span\u003e\u003cspan address=\"10.1186/s12943-023-01740-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLv B, Wang Y, Ma D, et al. Immunotherapy: Reshape the Tumor Immune Microenvironment. Front Immunol. 2022;13:844142. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2022.844142\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2022.844142\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNishioka R, Kawahara D, Imano N, et al. A nomogram-based survival prediction model for non-small cell lung cancer patients based on clinical risk factors and multiregion radiomics features. Clin Radiol. 2025;84:106826. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.crad.2025.106826\u003c/span\u003e\u003cspan address=\"10.1016/j.crad.2025.106826\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu L, Lou X, Kong N, et al. Can quantitative peritumoral CT radiomics features predict the prognosis of patients with non-small cell lung cancer? A systematic review. Eur Radiol. 2023;33(3):2105\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00330-022-09174-8\u003c/span\u003e\u003cspan address=\"10.1007/s00330-022-09174-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJia X, Wang Y, Zhang H, et al. Current status and quality of prognosis prediction models of non-small cell lung cancer constructed using computed tomography (CT)-based radiomics: a systematic review and radiomics quality score 2.0 assessment. Quant Imaging Med Surg. 2024;14(9):6978\u0026ndash;89. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21037/qims-24-22\u003c/span\u003e\u003cspan address=\"10.21037/qims-24-22\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Computed tomography, Non-small cell lung cancer, Radiomics Features, Nomogram model, Prognosis","lastPublishedDoi":"10.21203/rs.3.rs-7907062/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7907062/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThis study aims to investigate the use of computed tomography (CT) radiomics features combined with clinicopathological factors to establish and validate a radiomics nomogram for predicting overall survival (OS) in patients with non-small cell lung cancer (NSCLC).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis study included 177 patients with NSCLC, from whom CT images and clinicopathological data were collected (124 patients in the training set and 53 in the validation set). A total of 1,688 radiomics features were extracted from the volume of interest (VOI) of the tumors. Spearman correlation analysis and univariate Cox analysis were used for preliminary screening, followed by LASSO-COX regression combined with ten-fold cross-validation to further identify key radiomics features. Meanwhile, independent clinical risk factors were identified through Cox regression analysis. A nomogram was constructed based on the radiomics score (Radscore) combined with the independent clinical risk factors. The predictive performance of the model was evaluated using the C-index and calibration curves.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAmong the 177 patients with NSCLC, there were 107 males (60.45%)and 70 females (39.55%). In total, 16 key radiomics features were identified, and an OS nomogram was established based on the Radscore and clinical independent risk factors. The area under the curve(AUC) of the training and validation sets were 0.892 and 0.838, respectively. The calibration curve showed that the predicted OS values demonstrated good consistency with the actual values.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThe construction of a nomogram based on CT radiomics features combined with clinicopathological factors demonstrates good efficacy in predicting OS in patients with NSCLC and can provide valuable guidance for individualized treatment strategies.\u003c/p\u003e","manuscriptTitle":"Predict the prognosis of patients with non-small cell lung cancer based on CT radiomics and clinical pathological factors","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-22 09:13:43","doi":"10.21203/rs.3.rs-7907062/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"10c173b3-fbfb-4910-9eb9-a17f7b3ec867","owner":[],"postedDate":"December 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-11T11:41:23+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-22 09:13:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7907062","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7907062","identity":"rs-7907062","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}