Prediction Model For Lymph Node Metastasis Research Articles

A prospective study on the application of para-aortic lymph node metastasis prediction model in locally advanced cervical cancer and its effect on prognosis (CQGOG-0109)

A prospective study on the application of para-aortic lymph node metastasis prediction model in locally advanced cervical cancer and its effect on prognosis (CQGOG-0109)

Development of a multi-modal learning-based lymph node metastasis prediction model for lung cancer

PurposeThis study proposed a three-dimensional (3D) multi-modal learning-based model for the automated prediction and classification of lymph node metastasis in patients with non-small cell lung cancer (NSCLC) using computed tomography (CT) images and clinical information. MethodsWe utilized clinical information and CT image data from 4239 patients with NSCLC across multiple institutions. Four deep learning algorithm-based multi-modal models were constructed and evaluated for lymph node classification. To further enhance classification performance, a soft-voting ensemble technique was applied to integrate the outcomes of multiple multi-modal models. ResultsA comparison of the classification performance revealed that the multi-modal model, which integrated CT images and clinical information, outperformed the single-modal models. Among the four multi-modal models, the Xception model demonstrated the highest classification performance, with an area under the curve (AUC) of 0.756 for the internal test dataset and 0.736 for the external validation dataset. The ensemble model (SEResNet50_DenseNet121_Xception) exhibited even better performance, with an AUC of 0.762 for the internal test dataset and 0.751 for the external validation dataset, surpassing the multi-modal model's performance. ConclusionsIntegrating CT images and clinical information improved the performance of the lymph node metastasis prediction models in patients with NSCLC. The proposed 3D multi-modal lymph node prediction model can serve as an auxiliary tool for evaluating lymph node metastasis in patients with non-pretreated NSCLC, aiding in patient screening and treatment planning.

A CT-based radiomics model for predicting lymph node metastasis in hepatic alveolar echinococcosis patients to support lymph node dissection

BackgroundHepatic alveolar echinococcosis (AE) is a severe zoonotic parasitic disease, and accurate preoperative prediction of lymph node (LN) metastasis in AE patients is crucial for disease management, but it remains an unresolved challenge. The aim of this study was to establish a radiomics model for the preoperative prediction of LN metastasis in hepatic AE patients.MethodsA total of 100 hepatic AE patients who underwent hepatectomy and hepatoduodenal ligament LN dissection at Qinghai Provincial People's Hospital between January 2016 and August 2023 were included in the study. The patients were randomly divided into a training set and a validation set at an 8:2 ratio. Radiomic features were extracted from three-dimensional images of the hepatoduodenal ligament LNs delineated on arterial phase computed tomography (CT) scans of hepatic AE patients. Least absolute shrinkage and selection operator (LASSO) regression was applied for data dimensionality reduction and feature selection. Multivariate logistic regression analysis was performed to develop a prediction model, and the predictive performance of the model was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).ResultsA total of 7 radiomics features associated with LN status were selected using LASSO regression. The classification performances of the training set and validation set were consistent, with area under the operating characteristic curve (AUC) values of 0.928 and 0.890, respectively. The model also demonstrated good stability in subsequent validation.ConclusionIn this study, we established and evaluated a radiomics-based prediction model for LN metastasis in patients with hepatic AE using CT imaging. Our findings may provide a valuable reference for clinicians to determine the occurrence of LN metastasis in hepatic AE patients preoperatively, and help guide the implementation of individualized surgical plans to improve patient prognosis.

Radiomics analysis based on CT for predicting lymph node metastasis and prognosis in duodenal papillary carcinoma

ObjectivesRadiomics has been demonstrated to be strongly associated with TNM stage and patient prognosis. We aimed to develop a model for predicting lymph node metastasis (LNM) and survival.MethodsFor radiomics texture selection, 3D Slicer 5.0.3 software and the least absolute shrinkage and selection operator (LASSO) algorithm were used. Subsequently, the radiomics model, computed tomography (CT) image, and clinical risk model were compared. The performance of the three models was evaluated using receiver operating characteristic (ROC) curves, decision curve analysis (DCA), calibration plots, and clinical impact curves (CICs).ResultsFor the LNM prediction model, 224 patients with LNM information were used to construct a model that was applied to predict LNM. According to the CT data and clinical characteristics, we constructed a radiomics model, CT imaging model and clinical model. The radiomics model for evaluating LNM status showed excellent calibration and discrimination in the training cohort (AUC = 0.926, 95% CI = 0.869–0.982) and the validation cohort (AUC = 0.872, 95% CI = 0.802–0.941). DeLong’s test demonstrated that the difference among the three models was significant. Similarly, DCA and CIC showed that the radiomics model has better clinical utility than the CT imaging model and clinical model. Our model also exhibited good performance in predicting survival—in line with the findings of the model built with clinical risk factors.ConclusionsCT radiomics models exhibited better predictive performance for LNM than models built based on clinical risk characteristics and CT imaging and had comparative clinical utility for predicting patient prognosis.Critical relevance statementThe radiomics model showed excellent performance and discrimination for predicting LNM and survival of duodenal papillary carcinoma (DPC).Key PointsLNM status determines the most appropriate treatment for DPC.Our radiomics model for evaluating the LNM status of DPC performed excellently.The radiomics model had high sensitivity and specificity for predicting survival, exhibiting great clinical value.Graphical

A new clinical model for predicting lymph node metastasis in T1 colorectal cancer

PurposeLymph node metastasis (LNM) is a crucial factor that determines the prognosis of T1 colorectal cancer (CRC) patients. We aimed to develop a practical prediction model for LNM in T1 CRC.MethodsWe conducted a retrospective analysis of data from 825 patients with T1 CRC who underwent radical resection at a single center in China. All enrolled patients were randomly divided into a training set and a validation set at a ratio of 7:3 using R software. Risk factors for LNM were identified through multivariate logistic regression analyses. Subsequently, a prediction model was developed using the selected variables.ResultsThe lymph node metastasis (LNM) rate was 10.1% in the training cohort and 9.3% in the validation cohort. In the training set, risk factors for LNM in T1 CRC were identified, including depressed endoscopic gross appearance, sex, submucosal invasion combined with tumor grade (DSI-TG), lymphovascular invasion (LVI), and tumor budding. LVI emerged as the most potent predictor for LNM. The prediction model based on these factors exhibited good discrimination ability in the validation sets (AUC: 79.3%). Compared to current guidelines, the model could potentially reduce over-surgery by 48.9%. Interestingly, we observed that sex had a differential impact on LNM between early-onset and late-onset CRC patients.ConclusionsWe developed a clinical prediction model for LNM in T1 CRC using five factors that are easily accessible in clinical practice. The model has better predictive performance and practicality than the current guidelines and can assist clinicians in making treatment decisions for T1 CRC patients.

Oncologic outcomes of screen-detected and non-screen-detected T1 colorectal cancers.

The incidence of T1 colorectal cancer (CRC) has increased with the implementation of CRC screening programs. It is unknown whether the outcomes and risk models for T1 CRC based on non-screen-detected patients can be extrapolated to screen-detected T1 CRC. This study aimed to compare the stage distribution and oncologic outcomes of T1 CRC patients within and outside the screening program. Data from T1 CRC patients diagnosed between 2014 and 2017 were collected from 12 hospitals in the Netherlands. The presence of lymph node metastasis (LNM) at diagnosis was compared between screen-detected and non-screen-detected patients using multivariable logistic regression. Cox proportional hazard regression was used to analyze differences in the time to recurrence (TTR), metastasis-free survival (MFS), cancer-specific survival (CSS), and overall survival. Additionally, the performance of conventional risk factors for LNM was evaluated across the groups. 1803 patients were included (1114 [62%] screen-detected), with median follow-up of 51 months (interquartile range 30). The proportion of LNM did not significantly differ between screen- and non-screen-detected patients (12.6% vs. 8.9%; odds ratio 1.41; 95%CI 0.89-2.23); a prediction model for LNM performed equally in both groups. The 3- and 5-year TTR, MFS, and CSS were similar for patients within and outside the screening program. However, overall survival was significantly longer in screen-detected T1 CRC patients (adjusted hazard ratio 0.51; 95%CI 0.38-0.68). Screen-detected and non-screen-detected T1 CRCs have similar stage distributions and oncologic outcomes and can therefore be treated equally. However, screen-detected T1 CRC patients exhibit a lower rate of non-CRC-related mortality, resulting in longer overall survival.

An Intelligent System of Predicting Lymph Node Metastasis in Colorectal Cancer Using 3D CT Scans

In colorectal cancer (CRC), accurately predicting lymph node metastasis (LNM) contributes to developing appropriate treatment plans and serves as the key to long‐term survival of patients. In the clinical settings, preoperative LNM diagnosis in CRC predominantly depends on computed tomography (CT). Nevertheless, lymph nodes are small in size and difficult to identify on 3D CT scans, and CT‐based diagnosis of metastatic lymph nodes is prone to a significant misdiagnosis rate and lacks consistency across clinicians. Currently, there is no automatic system available for LNM prediction in CRC via 3D CT scans. In addition, existing deep learning‐ (DL‐) based lymph node detection models present low detection accuracy and high false‐positive rates, and most existing DL‐based lymph node metastasis prediction models mainly use tumor area characteristics but fail to adequately utilize lymph node information, thus not yielding satisfactory results. To tackle these issues, we propose an intelligent diagnosis system for this challenging task, mainly including a lymph node detection (LND) model and a lymph node metastasis prediction (LNMP) model. In detail, the LND model utilizes an encoder‐decoder network to detect lymph nodes, and the LNMP model employs an innovative attention‐based multiple instance learning (MIL) network. An instance‐level self‐attention feature enhancement module is designed to extract and augment lymph node features as a bag of instances. Furthermore, a bag‐level MIL prediction module is employed to extract instance features and create a bag representation for the ultimate LNM prediction. As far as we know, the proposed intelligent system represents the pioneering method for addressing this complex clinical challenge. In experiments, our proposed intelligent system achieves the AUC of 75.4% and the accuracy of 73.9%, showcasing a significant enhancement compared to physicians specialising in CRC and highlighting its strong clinical applicability. The accessible code can be found at https://github.com/SCU-MI/IS-LNM.

Evaluation of a three-gene methylation model for correlating lymph node metastasis in postoperative early gastric cancer adjacent samples.

Lymph node metastasis (LNM) has a profound impact on the treatment and prognosis of early gastric cancer (EGC), yet the existing evaluation methods lack accuracy. Recent research has underscored the role of precancerous lesions in tumor progression and metastasis. The objective of this study was to utilize the previously developed EGC LNM prediction model to further validate and extend the analysis in paired adjacent tissue samples. We evaluated the model in a monocentric study using Methylight, a methylation-specific PCR technique, on postoperative fresh-frozen EGC samples (n = 129) and paired adjacent tissue samples (n = 129). The three-gene methylation model demonstrated remarkable efficacy in both EGC and adjacent tissues. The model demonstrated excellent performance, with areas under the curve (AUC) of 0.85 and 0.82, specificities of 85.1% and 80.5%, sensitivities of 83.3% and 73.8%, and accuracies of 84.5% and 78.3%, respectively. It is noteworthy that the model demonstrated superior performance compared to computed tomography (CT) imaging in the adjacent tissue group, with an area under the curve (AUC) of 0.86 compared to 0.64 (p < 0.001). Furthermore, the model demonstrated superior diagnostic capability in these adjacent tissues (AUC = 0.82) compared to traditional clinicopathological features, including ulceration (AUC = 0.65), invasional depth (AUC = 0.66), and lymphovascular invasion (AUC = 0.69). Additionally, it surpassed traditional models based on these features (AUC = 0.77). The three-gene methylation prediction model for EGC LNM is highly effective in both cancerous and adjacent tissue samples in a postoperative setting, providing reliable diagnostic information. This extends its clinical utility, particularly when tumor samples are scarce, making it a valuable tool for evaluating LNM status and assisting in treatment planning.

Radiomics and Clinicopathological Characteristics for Predicting Lymph Node Metastasis in Testicular Cancer

Accurate prediction of lymph node metastasis (LNM) in patients with testicular cancer is highly relevant for treatment decision-making and prognostic evaluation. Our study aimed to develop and validate clinical radiomics models for individual preoperative prediction of LNM in patients with testicular cancer. We enrolled 91 patients with clinicopathologically confirmed early-stage testicular cancer, with disease confined to the testes. We included five significant clinical risk factors (age, preoperative serum tumour markers AFP and B-HCG, histotype and BMI) to build the clinical model. After segmenting 273 retroperitoneal lymph nodes, we then combined the clinical risk factors and lymph node radiomics features to establish combined predictive models using Random Forest (RF), Light Gradient Boosting Machine (LGBM), Support Vector Machine Classifier (SVC), and K-Nearest Neighbours (KNN). Model performance was assessed by the area under the receiver operating characteristic (ROC) curve (AUC). Finally, the decision curve analysis (DCA) was used to evaluate the clinical usefulness. The Random Forest combined clinical lymph node radiomics model with the highest AUC of 0.95 (±0.03 SD; 95% CI) was considered the candidate model with decision curve analysis, demonstrating its usefulness for preoperative prediction in the clinical setting. Our study has identified reliable and predictive machine learning techniques for predicting lymph node metastasis in early-stage testicular cancer. Identifying the most effective machine learning approaches for predictive analysis based on radiomics integrating clinical risk factors can expand the applicability of radiomics in precision oncology and cancer treatment.

A clinically practical model for the preoperative prediction of lymph node metastasis in bladder cancer: a multicohort study.

The aim of this study was to construct a clinically practical model to precisely predict lymph node (LN) metastasis in bladder cancer patients. Four independent cohorts were included. The least absolute shrinkage and selection operator regression with multivariate logistic regression were applied. The diagnostic efficacy of LN score and CT/MRI was compared by accuracy, sensitivity, specificity, and area under curve (AUC). A total of 606 patients were included to develop a basic prediction model. After multistep gene selection, the LN metastasis prediction model was constructed with 5 genes. The model can accurately predict LN metastasis with an AUC of 0.781. For clinically practical use, we transformed the model into a Fast LN Scoring System using the SYSMH cohort (n = 105). High LN score patients exhibited a 72.2% LN metastasis rate, while low LN score patients showed a 3.4% LN metastasis rate. The LN score achieved a superior accuracy than CT/MRI (0.882 vs. 0.727). Application of LN score can correct the diagnosis of 88% (22/25) patients who were misdiagnosed by CT/MRI. The clinically practical LN score can precisely, rapidly, and conveniently predict LN status, which will assist preoperative diagnosis for LN metastasis and guide precise therapy.

Multi-center study on predicting breast cancer lymph node status from core needle biopsy specimens using multi-modal and multi-instance deep learning

The objective of our study is to develop a deep learning model based on clinicopathological data and digital pathological image of core needle biopsy specimens for predicting breast cancer lymph node metastasis. We collected 3701 patients from the Fourth Hospital of Hebei Medical University and 190 patients from four medical centers in Hebei Province. Integrating clinicopathological data and image features build multi-modal and multi-instance (MMMI) deep learning model to obtain the final prediction. For predicting with or without lymph node metastasis, the AUC was 0.770, 0.709, 0.809 based on the clinicopathological features, WSI and MMMI, respectively. For predicting four classification of lymph node status (no metastasis, isolated tumor cells (ITCs), micrometastasis, and macrometastasis), the prediction based on clinicopathological features, WSI and MMMI were compared. The AUC for no metastasis was 0.770, 0.709, 0.809, respectively; ITCs were 0.619, 0.531, 0.634, respectively; micrometastasis were 0.636, 0.617, 0.691, respectively; and macrometastasis were 0.748, 0.691, 0.758, respectively. The MMMI model achieved the highest prediction accuracy. For prediction of different molecular types of breast cancer, MMMI demonstrated a better prediction accuracy for any type of lymph node status, especially in the molecular type of triple negative breast cancer (TNBC). In the external validation sets, MMMI also showed better prediction accuracy in the four classification, with AUC of 0.725, 0.757, 0.525, and 0.708, respectively. Finally, we developed a breast cancer lymph node metastasis prediction model based on a MMMI model. Through all cases tests, the results showed that the overall prediction ability was high.

A novel approach to quantify calcifications of thyroid nodules in US images based on deep learning: predicting the risk of cervical lymph node metastasis in papillary thyroid cancer patients.

Based on ultrasound (US) images, this study aimed to detect and quantify calcifications of thyroid nodules, which are regarded as one of the most important features in US diagnosis of thyroid cancer, and to further investigate the value of US calcifications in predicting the risk of lymph node metastasis (LNM) in papillary thyroid cancer (PTC). Based on the DeepLabv3+ networks, 2992 thyroid nodules in US images were used to train a model to detect thyroid nodules, of which 998 were used to train a model to detect and quantify calcifications. A total of 225 and 146 thyroid nodules obtained from two centers, respectively, were used to test the performance of these models. A logistic regression method was used to construct the predictive models for LNM in PTCs. Calcifications detected by the network model and experienced radiologists had an agreement degree of above 90%. The novel quantitative parameters of US calcification defined in this study showed a significant difference between PTC patients with and without cervical LNM (p < 0.05). The calcification parameters were beneficial to predicting the LNM risk in PTC patients. The LNM prediction model using these calcification parameters combined with patient age and other US nodular features showed a higher specificity and accuracy than the calcification parameters alone. Our models not only detect the calcifications automatically, but also have value in predicting cervical LNM risk of PTC patients, thereby making it possible to investigate the relationship between calcifications and highly invasive PTC in detail. Due to the high association of US microcalcifications with thyroid cancers, our model will contribute to the differential diagnosis of thyroid nodules in daily practice. • We developed an ML-based network model for automatically detecting and quantifying calcifications within thyroid nodules in US images. • Three novel parameters for quantifying US calcifications were defined and verified. • These US calcification parameters showed value in predicting the risk of cervical LNM in PTC patients.

Landscape of gene mutation in Chinese thyroid cancer patients: Construction and validation of lymph node metastasis prediction model based on clinical features and gene mutation marker.

Reporting the clinicopathological information of thyroid cancer (TC) patients from a central medical center of east China, and constructing the nomogram predicting lymph node metastasis (LNM). We collected the patients who underwent thyroid cancer surgery in our institute from July 1, 2019 to July 31, 2021, a total of 253 subjects were enrolled. We used HiPure FFPE DNA Kit to extract DNA and RNApure FFPE Kit to extract RNA from the paraffin sections of tumor tissue, the extracted DNA samples and RNA samples were used for NGS sequencing. The clinical and pathological information of TCGA-THCA cohort was obtained as the validation cohort. Multivariate logistic regression analysis was performed to identify the independent prognostic factor, and the nomogram was subsequently constructed by "rms" R package. Secondary cases contained more mutation of BRAF (90.48% vs. 62.07%) and TERT (33.0% vs. 3.0%), as compared with primary cases. Primary patients with positive lymph node were younger (40.9 ± 10.8 vs. 45.3 ± 11.8, p = 0.0031) and contained advanced TI-RADS levels (4c: 22.8% vs. 8.3%, 5: 6.5% vs. 0/0%, p = 1.878e-03), as well as more RET genetic alteration (16.3% vs. 2.7%, p = 2.566e-03). We chose age, tumor diameter, RET fusion, and gender to construct the LNM predicting nomogram. Calibration plot, DCA curve, and the clinical impact plot verified the preferable prognostic value of the nomogram, with an AUC value of 0.724 (0.656-0.792). We successfully validated the prognostic value of the nomogram in TCGA-THCA cohort. RET fusion might impact the process of protein digestion and absorption, cytokine-cytokine receptor interaction, ECM-receptor interaction, focal adhesion. We provide a novel nomogram to predict the LNM for TC patients, including the features of patient's age, gender, tumor diameter, and RET alteration. Further studies from multiple medical centers are essential to validate the nomogram.

Nomogram established using risk factors of early gastric cancer for predicting the lymph node metastasis

For the prognosis of patients with early gastric cancer (EGC), lymph node metastasis (LNM) plays a crucial role. A thorough and precise evaluation of the patient for LNM is now required. To determine the factors influencing LNM and to construct a prediction model of LNM for EGC patients. Clinical information and pathology data of 2217 EGC patients downloaded from the Surveillance, Epidemiology, and End Results database were collected and analyzed. Based on a 7:3 ratio, 1550 people were categorized into training sets and 667 people were assigned to testing sets, randomly. Based on the factors influencing LNM determined by the training sets, the nomogram was drawn and verified. Based on multivariate analysis, age at diagnosis, histology type, grade, T-stage, and size were risk factors of LNM for EGC. Besides, nomogram was drawn to predict the risk of LNM for EGC patients. Among the categorical variables, the effect of grade (well, moderate, and poor) was the most significant prognosis factor. For training sets and testing sets, respectively, area under the receiver-operating characteristic curve of nomograms were 0.751 [95% confidence interval (CI): 0.721-0.782] and 0.786 (95%CI: 0.742-0.830). In addition, the calibration curves showed that the prediction model of LNM had good consistency. Age at diagnosis, histology type, grade, T-stage, and tumor size were independent variables for LNM in EGC. Based on the above risk factors, prediction model may offer some guiding implications for the choice of subsequent therapeutic approaches for EGC.

A Multi-modal Heterogeneous Graph Forest to Predict Lymph Node Metastasis of Non-small Cell Lung Cancer.

Lymph node metastasis (LNM) is critical for treatment decision-making for cancer patients, but it is difficult to diagnose accurately before surgery. Machine learning can learn nontrivial knowledge from multi-modal data to support accurate diagnosis. In this paper, we proposed a Multi-modal Heterogeneous Graph Forest (MHGF) approach to extract the deep representations of LNM from multi-modal data. Specifically, we first extracted the deep image features from CT images to represent the pathological anatomic extent of the primary tumor (pathological T stage) using a ResNet-Trans network. And then, a heterogeneous graph with six vertices and seven bi-directional relations was defined by medical experts to describe the possible relations between the clinical and image features. After that, we proposed a graph forest approach to construct the sub-graphs by removing each vertex in the complete graph iteratively. Finally, we used graph neural networks to learn the representations of each sub-graph in the forest to predict LNM and averaged all the prediction results as final results. We conducted experiments on 681 patients' multi-modal data. The proposed MHGF achieves the best performances with a 0.806 AUC value and 0.513 AP value compared with state-of-art machine learning and deep learning methods. The results indicate that the graph method can explore the relations between different types of features to learn effective deep representations for LNM prediction. Moreover, we found that the deep image features about the pathological anatomic extent of the primary tumor are useful for LNM prediction. And the graph forest approach can further improve the generalization ability and stability of the LNM prediction model.

Comparison of MRI and CT-based radiomics for preoperative prediction of lymph node metastasis in pancreatic ductal adenocarcinoma.

The preoperative prediction of lymph node metastasis (LNM) in pancreatic ductal adenocarcinoma (PDAC) is essential in prognosis and treatment strategy formulation. To compare the performance of computed tomography (CT) and magnetic resonance imaging (MRI) radiomics models for the preoperative prediction of LNM in PDAC. In total, 160 consecutive patients with PDAC were retrospectively included, who were divided into the training and validation sets (ratio of 8:2). Two radiologists evaluated LNM basing on morphological abnormalities. Radiomics features were extracted from T2-weighted imaging, T1-weighted imaging, and multiphase contrast enhanced MRI and multiphase CT, respectively. Overall, 1184 radiomics features were extracted from each volume of interest drawn. Only features with an intraclass correlation coefficient ≥0.75 were included. Three sequential feature selection steps-variance threshold, variance thresholding and least absolute shrinkage selection operator-were repeated 20 times with fivefold cross-validation in the training set. Two radiomics models based on multiphase CT and multiparametric MRI were built with the five most frequent features. Model performance was evaluated using the area under the curve (AUC) values. Multiparametric MRI radiomics model achieved improved AUCs (0.791 and 0.786 in the training and validation sets, respectively) than that of the CT radiomics model (0.672 and 0.655 in the training and validation sets, respectively) and of the radiologists' assessment (0.600-0.613 and 0.560-0.587 in the training and validation sets, respectively). Multiparametric MRI radiomics model may serve as a potential tool for preoperatively evaluating LNM in PDAC and had superior predictive performance to multiphase CT-based model and radiologists' assessment.

Construction and Evaluation of a Preoperative Prediction Model for Lymph Node Metastasis of cIA Lung Adenocarcinoma Using Random Forest.

Background Lymph node metastasis (LNM) is the main route of metastasis in lung adenocarcinoma (LA), and preoperative prediction of LNM in early LA is key for accurate medical treatment. We aimed to establish a preoperative prediction model of LNM of early LA through clinical data mining to reduce unnecessary lymph node dissection, reduce surgical injury, and shorten the operation time. Methods We retrospectively collected imaging data and clinical features of 1121 patients with early LA who underwent video-assisted thoracic surgery at the First Hospital of China Medical University from 2004 to 2021. Logistic regression analysis was used to select variables and establish the preoperative diagnosis model using random forest classifier (RFC). The prediction results from the test set were used to evaluate the prediction performance of the model. Results Combining the results of logistic analysis and practical clinical application experience, nine clinical features were included. In the random forest classifier model, when the number of nodes was three and the n-tree value is 500, we obtained the best prediction model (accuracy = 0.9769), with a positive prediction rate of 90% and a negative prediction rate of 98.69%. Conclusion We established a preoperative prediction model for LNM of early LA using a machine learning random forest method combined with clinical and imaging features. More excellent predictors may be obtained by refining imaging features.

380. UTILITY OF PET/CT FOR LYMPH NODE METASTASIS DETECTION IN PT1 EARLY ESOPHAGEAL SQUAMOUS CELL CARCINOMA

Abstract Nodal staging for early esophageal squamous cell carcinoma (ESCC) was mandatory to decide therapy strategy. Utility of preoperative Positron Emission Tomography/Computed Tomography (PET/CT) for lymph node metastasis (LNM) detection in early ESCC was evaluated in the study. 186 consecutive patients with pathologically confirmed T1 tumors undergoing preoperative PET/CT at Zhongshan Hospital between 2015 August to 2020 December were retrospectively enrolled. Clinical characteristics and PET/CT results were collected and compared between patients with or without LNM. A PET integrated risk score model for prediction of LNM was established and validated. Survivals were compared between high risk group and low risk group by Log Rank test. 40 out of 186 patients were diagnosed with LNM on PET-CT. PET/CT alone for LNM detection was with a 53.3% sensitivity, 84.65% specificity, and 90.4% negative predictive value. Tumor size &amp;gt;2.5 cm (P = 0.015), tumor depth (P = 0.012), LVI (P &amp;lt; 0.001), differentiation (P = 0.066) and PET LN (+) (P &amp;lt; 0.001) were included in the prediction model for LNM. The AUC of ROC in derivation cohort and validation cohort was 0.860 and 0.846 respectively. High risk group scoring &amp;gt;3 points showed poorer 5-year recurrence free survival (62.6% VS. 74.7%, P = 0.046) and similar overall survival (87.6% VS. 86.9%, P = 0.942) as compared to low risk group. PET/CT for LNM detection in early ESCC was of 53.3% sensitivity and 84.65% specificity. PET integrated risk score model for LNM was of fair property with 0.860 AUC and was predictive of recurrence.

Prognostic Role of Preoperative Vascular Cell Adhesion Molecule-1 Plasma Levels in Urothelial Carcinoma of the Bladder Treated With Radical Cystectomy

BackgroundAngiogenesis-related marker vascular cell adhesion molecule-1 (VCAM-1) has been shown to be elevated in urothelial carcinoma of the bladder (UCB), but its predictive/prognostic role has not been determined. Thus, this study aimed to investigate the predictive/prognostic role of VCAM-1 for patients who have UCB treated with radical cystectomy (RC).MethodsThe study enrolled 1036 patients with clinically non-metastatic advanced UCB who underwent RC, and plasma VCAM-1 was evaluated preoperatively. The correlation of plasma VCAM-1 with pathologic and survival outcomes was assessed using binominal logistic regression and multivariable Cox regression analyses. Discrimination was assessed using the area under the curve and concordance indices. The clinical net benefit was evaluated using decision curve analysis (DCA).ResultsPreoperative VCAM-1 was significantly elevated in patients with adverse pathologic features. Higher VCAM-1 levels were independently associated with increased risk of lymph-node-metastasis (LNM), ≥pT3 disease, and non-organ-confined disease (NOCD (p < 0.001 for each). Preoperative plasma VCAM-1 was independently associated with recurrence-free survival (RFS), cancer-specific survival (CSS), and overall survival (OS) in pre- and postoperative multivariable models. Adding VCAM-1 to these predictive models improved their discriminatory ability to predict all outcomes by a significant margin. In the DCA, VCAM-1 addition to the reference models for prediction of LNM, NOCD, RFS, and CSS resulted in relevant improvement.ConclusionsElevated plasma VCAM-1 was associated with biologically and clinically aggressive UCB disease features. After validation, preoperative VCAM-1 may serve as a biomarker to help identify patients likely to benefit from intensified/multimodal therapy. In addition, VCAM-1 improved the discriminatory power of predictive/prognostic models and can be used to refine personalized clinical decision-making.

A nomogram for predicting lymph node metastasis in early gastric cancer

Objective: To explore the independent risk factors of lymph node metastasis (LNM) in early gastric cancer, and to use nomogram to construct a prediction model for above LNM. Methods: A retrospective cohort study was conducted. Inclusion criteria: (1) primary early gastric cancer as stage pT1 confirmed by postoperative pathology; (2) complete clinicopathological data. Exclusion criteria: (1) patients with advanced gastric cancer, stump gastric cancer or history of gastrectomy; (2) early gastric cancer patients confirmed by pathology after neoadjuvant chemotherapy; (3) other types of gastric tumors, such as lymphoma, neuroendocrine tumor, stromal tumor, etc.; (4) primary tumors of other organs with gastric metastasis. According to the above criteria, 1633 patients with early gastric cancer who underwent radical gastrectomy at the Department of General Surgery of the Chinese PLA General Hospital First Medical Center from December 2005 to December 2020 were enrolled as training set, meanwhile 239 patients with early gastric cancer who underwent gastrectomy at the Department of General Surgery of the Chinese PLA General Hospital Fourth Medical Center from December 2015 to December 2020 were enrolled as external validation set. Risk factors of LNM in early gastric cancer were identified by using univariate and multivariate logistic regression analyses. A nomogram prediction model was established with significant factors screened by multivariate analysis. Area under the receiver operating characteristic curve (AUC) was used for assessing the predictive value of the model. Calibration curve was drawn for external validation. Results: Among 1633 patients in training set, the mean number of retrieved lymph nodes was 20 (13-28), and 209 patients (12.8%) had lymph node metastasis. Univariate analysis showed that gender, resection range, tumor location, tumor morphology, lymph node clearance, vascular invasion, lymphatic cancer thrombus, tumor length, tumor differentiation, microscopic presence of signet ring cells and depth of tumor invasion were associated with LNM (all P<0.05). Multivariate analysis revealed that females, tumor morphology as ulcer type, vascular invasion, lymphatic cancer thrombus, tumor length≥3 cm, deeper invasion of mucosa, and poor differentiation were independent risk factors for LNM in early gastric cancers (all P<0.05). Receiver operating characteristic curve indicated that AUC of training set was 0.818 (95%CI: 0.790-0.847) and AUC of external validation set was 0.765 (95%CI: 0.688-0.843). The calibration curve showed that the LNM probability predicted by nomogram was consistent with the actual situation (C-index: 0.818 in training set and 0.765 in external validation set). Conclusions: Females, tumor morphology as ulcer type, vascular invasion, lymphatic cancer thrombus, tumor length≥3 cm, deeper invasion of mucosa and poor differentiation are independent risk factors for LNM of early gastric cancer. The establishment of a nomogram prediction model for LNM in early gastric cancer has great diagnostic value and can provide reference for treatment selection.