Construction,validation and,visualization of a web-based nomogram for predicting the overall survival and cancer-specific survival of leiomyosarcoma patients with lung metastasis

BackgroundThis study sought to assess the prognostic factors for leiomyosarcoma (LMS) patients with lung metastasis and construct web-based nomograms to predict overall survival (OS) and cancer-specific survival (CSS).MethodPatients diagnosed with LMS combined with lung metastasis between 2010 and 2016 were identified in the Surveillance, Epidemiology, and End Results (SEER) database. The patients were randomly divided into a training set and a testing set. The X-tile analysis provides the best age and tumor size cut-off point, and changes continuous variables into categorical variables. The independent prognostic factors were determined by Cox regression analysis, and 2 nomograms were established. Receiver operating characteristic curves and calibration curves were used to evaluate the nomograms. Based on the nomograms, 2 web-based nomograms were established.ResultsTwo hundred and twenty-eight cases were included in the OS nomogram construction, and were randomly divided into a training set (n=160) and a validation set (n=68). Age, T stage, bone metastasis, surgery, chemotherapy, marital status, tumor size, and tumor site were found to be correlated with OS. One hundred and eighty-three cases were enrolled in the CSS nomogram construction, and randomly divided into a training set (n=129) and a validation set (n=54). Age, bone metastasis, surgery, chemotherapy, tumor size, and tumor site were found to be correlated with CSS. Two nomograms were established to predict OS and CSS. In the training set, the areas under the curve of the nomogram for predicting 1-, 2-, and 3-year OS were 0.783, 0.830, and 0.832, respectively, and those for predicting 1-, 2-, and 3-year CSS were 0.889, 0.777, and 0.884, respectively. Two web-based nomograms were established to predict OS (https://wenn23.shinyapps.io/lmslmosapp/), and CSS (https://wenn23.shinyapps.io/lmslmcssapp/).ConclusionThe developed web-based nomogram is a useful tool for accurately analyzing the prognosis of LMS patients with lung metastasis, and could help clinical doctors to make personalized clinical decisions.     

Application of machine learning approaches to predict the 5-year survival status of patients with esophageal cancer

BackgroundAccurate prognostic estimation for esophageal cancer (EC) patients plays an important role in the process of clinical decision-making. The objective of this study was to develop an effective model to predict the 5-year survival status of EC patients using machine learning (ML) algorithms.MethodsWe retrieved the information of patients diagnosed with EC between 2010 and 2015 from the Surveillance, Epidemiology, and End Results (SEER) Program, including 24 features. A total of 8 ML models were applied to the selected dataset to classify the EC patients in terms of 5-year survival status, including 3 newly developed gradient boosting models (GBM), XGBoost, CatBoost, and LightGBM, 2 commonly used tree-based models, gradient boosting decision trees (GBDT) and random forest (RF), and 3 other ML models, artificial neural networks (ANN), naive Bayes (NB), and support vector machines (SVM). A 5-fold cross-validation was used in model performance measurement.ResultsAfter excluding records with missing data, the final study population comprised 10,588 patients. Feature selection was conducted based on the χ² test, however, the experiment results showed that the complete dataset provided better prediction of outcomes than the dataset with removal of non-significant features. Among the 8 models, XGBoost had the best performance [area under the receiver operating characteristic (ROC) curve (AUC): 0.852 for XGBoost, 0.849 for CatBoost, 0.850 for LightGBM, 0.846 for GBDT, 0.838 for RF, 0.844 for ANN, 0.833 for NB, and 0.789 for SVM]. The accuracy and logistic loss of XGBoost were 0.875 and 0.301, respectively, which were also the best performances. In the XGBoost model, the SHapley Additive exPlanations (SHAP) value was calculated and the result indicated that the four features: reason no cancer-directed surgery, Surg Prim Site, age, and stage group had the greatest impact on predicting the outcomes.ConclusionsThe XGBoost model and the complete dataset can be used to construct an accurate prognostic model for patients diagnosed with EC which may be applicable in clinical practice in the future.     

Refining the definition of perioperative mortality following hepatectomy using death within 90 days as the standard criterion

ObjectivesDefining perioperative mortality as death that occurs within 30 days of surgery may underestimate ‘true’ mortality among patients undergoing hepatic resection. To better define perioperative mortality, trends in the risk for death during the first 90 days after hepatectomy were assessed.MethodsSurveillance, Epidemiology and End Results (SEER) Medicare data were used to identify 2597 patients who underwent hepatic resection during 1991–2006. Data on their clinicopathological characteristics, surgical management and perioperative mortality were collected and survival was assessed at 30, 60 and 90 days post-surgery.ResultsOverall, 5.7% of patients died within the first 30 days. Postoperative mortality at 60 and 90 days were 8.3% and 10.1%. In-hospital mortality after hepatic resection was greater among patients with hepatocellular carcinoma (HCC) than among those with colorectal liver metastases (CRLM) (8.9% and 3.8%, respectively; P < 0.001). In CRLM patients, mortality increased from 4.3% at 30 days to 8.4% at 90 days, whereas mortality in HCC patients increased from 9.7% at 30 days to 15.0% at 90 days (both P < 0.05). Patients with HCC were twice as likely as CRLM patients to die within 30 days [odds ratio (OR) 2.03], 60 days (OR = 1.74) and 90 days (OR = 1.71) (all P < 0.001). Differences in 30- and 90-day mortality were greatest among HCC patients undergoing major hepatic resection (P < 0.05).ConclusionsReporting deaths that occur within a maximum of 30 days of surgery underestimates the mortality associated with hepatic resection. Traditional 30-day definitions of mortality are misleading and surgeons should report all perioperative outcomes that occur within 90 days of hepatic resection.     

Association of marital status with cardiovascular outcome in patients with breast cancer

BackgroundThe influences of marital status on cardiovascular death risk in patients with breast cancer remained unclear. This study aimed to evaluate the associations of different marital status with cardiovascular death risk in patients with breast cancer.MethodsA total of 182,666 female breast cancer patients were enrolled in this study from the Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2014, and was divided into two groups: married (N=107,043) and unmarried (N=75,623). A 1:1 propensity score matching (PSM) was applied to reduce inter-group bias between the two groups. Competing-risks model was used to assess the associations between different marital status and cardiovascular death risk in patients with breast cancer.ResultsAfter PSM, marital status was an independent predictor for cardiovascular death in patients with breast cancer. Unmarried condition was associated with increased cardiovascular death risk than married condition among breast cancer patients [unadjusted model: hazard ratio (HR) =2.012, 95% confidence interval (CI): 1.835–2.208, P<0.001; Model 1: HR =1.958, 95% CI: 1.785–2.148, P<0.001; Model 2: HR =1.954, 95% CI: 1.781–2.144, P<0.001; Model 3: HR =1.920, 95% CI: 1.748–2.107, P<0.001]. With the exception of separated condition (adjusted HR =0.886, 95% CI: 0.474–1.658, P=0.705), further unmarried subgroups analysis showed that the other three unmarried status were associated with increased cardiovascular death risk as follows: single (adjusted HR =1.623, 95% CI: 1.421–1.853, P<0.001), divorced (adjusted HR =1.394, 95% CI: 1.209–1.608, P<0.001), and widowed (adjusted HR =2.460, 95% CI: 2.227–2.717, P<0.001). In particularly, widowed condition showed the highest cardiovascular death risk in all 4 unmarried subgroups.ConclusionsUnmarried condition (e.g., single, divorced and widowed) was associated with elevated cardiovascular death risk compared with their married counterparts in patients with breast cancer, suggesting that more attention and humanistic care should be paid to unmarried breast cancer patients (especially the widowed patients) in the management of female breast cancer patients.     

Diagnostic and prognostic factors,and two nomograms for endometrial cancer patients with bone metastasis: A large cohort retrospective study

Patients with endometrial cancer (EC) who develop bone metastasis (BM) always imply a poorer prognosis. However, reliable predictive models associated with BM from EC are currently limited.We retrospectively analyzed data on 54,077 patients diagnosed with primary EC in the Surveillance, Epidemiology, and End Results database. Multivariate logistic regression analysis was used to determine independent predictors of BM from EC. Univariate and multivariate Cox regression analyses were used to determine independent prognostic factors for EC with BM. Based on independent predictors and prognostic factors, we constructed a diagnostic nomogram and prognostic nomogram separately. Besides, calibration curves, receiver operating characteristic curves, and decision curve analysis were used to evaluate the models.A total of 54,077 patients with EC from the Surveillance, Epidemiology, and End Results database were included in this study, 364 of whom had BM. Multivariate analysis in the logistic model showed that lung metastasis, liver metastasis, brain metastasis, N stage, T stage, histologic grade, and race were risk factors for BM from EC. Multivariate analysis in the Cox model showed that liver metastasis, brain metastasis, chemotherapy, surgery, and histologic type had a significant effect on overall survival. Moreover, the receiver operating characteristic curve, calibration curve, and decision curve analysis indicated the good performance of both diagnostic and prognostic nomograms.Two clinical prediction model was constructed and validated to predict individual risk and overall survival for EC with BM, respectively. Diagnostic nomogram and prognostic nomogram are complementary, improving the clinician''s ability to assess the patient''s prognosis and enhancing prognosis-based decision making.     

Predictors for survival in patients with bone metastasis of small cell lung cancer: A population-based study

The objective of the current study is to analyze the clinical and demographic characteristics of patients with bone metastasis of small cell lung cancer (SCLC) and explore their survival predictors.We retrospectively extracted patients with bone metastasis of SCLC from the Surveillance, Epidemiology, and End Results database. We applied Cox regression analyses to identify independent survival predictor of overall survival (OS) and cancer-specific survival (CSS). Only significant predictors from univariable analysis were included for multivariable Cox analysis. Kaplan–Meier method was used to evaluate survival differences between groups by the log–rank test.A total of 5120 patients with bone metastasis of SCLC were identified and included for survival analysis. The 1-year OS and CSS rates of bone metastasis of SCLC were 19.8% and 21.4%, respectively. On multivariable analysis, gender, age, radiotherapy, chemotherapy, liver metastasis, brain metastasis, insurance status, and marital status independently predicted OS and CSS. There was no significant difference of OS and CSS in terms of race and tumor size.Independent predictors of survival were identified among patients with bone metastasis of SCLC, which could be valuable to clinicians in treatment decision. Patients with bone metastasis of SCLC may benefit from radiotherapy and chemotherapy.     

Prognostic model for predicting overall and cancer-specific survival among patients with superficial spreading melanoma: A SEER based study

Skin malignant melanoma is one of the most aggressive skin tumors. Superficial spreading melanoma (SSM) is the most common histological type, which can originate from different body skin sites, and some patients can still accumulate regional lymph nodes and even have distant metastasis in some cases. This study used the relevant data from the monitoring, epidemiology and results database of the National Cancer Institute database to study the overall survival (OS) and cancer-specific survival (CSS) of SSM patients and established an SSM nomogram to evaluate the prognosis of patients. A total of 13,922 patients were collected from the monitoring, epidemiology and results database of the National Cancer Institute and randomly divided into a training cohort (8353 cases) and a validation cohort (5569 cases). Univariate and multivariate Cox regression analysis were used to determine prognostic factors, and these factors were used to construct OS and CSS nomograms for patients with SSM. Finally, the discrimination and consistency of the nomogram model were evaluated by the consistency index (C-index), area under the curve (AUC) and calibration curve. Multivariate Cox regression analysis suggested that age, sex, tumor site, the American joint committee on cancer T stage and the first primary melanoma were independent predictors of OS and CSS in patients with SSM and that the American joint committee on cancer N stage was also an independent predictor of CSS in patients with SSM. Based on the above prognostic factors, this study constructed a predictive model. The C-index of the model OS and CSS for this training cohort was 0.805 [95% CI: 0.793–0.817] and 0.896 [95% CI: 0.878–0.913], respectively. The AUC values for 1-, 3-, and 5-year OS were 0.822, 0.820, and 0.821, respectively, and the AUC values for CSS were 0.914, 0.922, and 0.893, respectively. The data indicated that both nomograms showed better predictive accuracy. The calibration curves of the training cohort and the validation cohort were in good agreement. The nomogram has superior predictive performance in predicting 1-, 3-, and 5-year OS and CSS prognosis in patients with SSM and can provide a reference for individualized treatment and clinical counseling of SSM.     

Clinicopathological features and prognostic analysis of metastatic pulmonary sarcomatoid carcinoma: a SEER analysis

BackgroundPulmonary sarcomatoid carcinoma (PSC) is a rare type of non-small cell lung cancer (NSCLC). Metastases are often detected at the first diagnosis. Despite high rates of distant metastasis, there is insufficient data describing the characteristics of PSC metastasis.MethodsWe performed a Surveillance, Epidemiology, and End Results (SEER) database-based analysis of clinicopathological features and prognosis of distant metastasis in PSC patients. Data queried for this analysis included PSC patients in the database between 2010 and 2016.ResultsA total of 934 patients met the criteria for inclusion in the analysis and included, at the time of diagnosis, 512 (54.8%) patients with metastasis, including bone (n=152; 16.3%), brain (n=108; 11.6%), liver (n=70; 7.5%), lung (n=142; 15.2%) metastases. Binary logistic regression showed that patients with giant cell carcinoma [odds ratio (OR) 4.023, 95% confidence interval (CI): 2.113–7.661, P<0.001] and spindle cell carcinoma (OR 3.151, 95% CI: 1.699–5.843, P<0.001) were associated with metastasis. Log-rank test and Kaplan-Meier plots indicated poor prognosis in metastatic patients [the 1-, 3-, and 5-year overall survival (OS) rates were 14.1%, 5.5%, and 4.8%, respectively]. Multivariable analysis showed younger and chemotherapy as improved prognostic factors of PSC patients with single metastasis site.ConclusionsThe SEER database-based analysis revealed the clinical features of distant metastasis of PSC and showed that different histological types posed distinct metastasis potential. Besides, age and chemotherapy were the independent prognostic factors of PSC patients with single metastasis site.     

Preoperative prevalence of Barrett's esophagus in esophageal adenocarcinoma: a systematic review.

BACKGROUND & AIMS: The public health impact of past screening and surveillance practices on the outcomes of Barrett's related cancers has not previously been quantified. Our purpose was to determine the prior prevalence of Barrett's esophagus in reported cases of incident adenocarcinoma undergoing resection, as an indirect measure of impact. METHODS: We performed a systematic review of the literature from 1966 to 2000. Studies were included if they reported: (1) the number of consecutive adenocarcinomas resected, and (2) the number of those resected who had a previously known diagnosis of Barrett's. We generated summary estimates using a random effects model. RESULTS: We identified and reviewed 752 studies. Twelve studies representing a total of 1503 unique cases of resected adenocarcinomas met inclusion criteria. Using a random effects model, the overall percentage of patients undergoing resection who had a prior diagnosis of Barrett's was 4.7% +/- 2.9%. CONCLUSIONS: The low prior prevalence (approximately 5%) of Barrett's esophagus in this study population provides indirect evidence to suggest that recent efforts to identify patients with Barrett's-whether through endoscopic screening or evaluation of symptomatic patients-have had minimal public health impact on esophageal adenocarcinoma outcomes. The potential benefits of endoscopic surveillance seem to have been limited to only a fraction of those individuals at risk. These data thus provide a clear and compelling rationale for the development of effective screening strategies to identify patients with Barrett's esophagus.     

Pulmonary lymphoepithelioma-like carcinoma: a Surveillance,Epidemiology, and End Results database analysis

Background

Pulmonary lymphoepithelioma-like carcinoma (LELC) is one of the rare histological non-small cell lung cancers. Only a few case reports have been published. The knowledge of its characteristics and prognosis in western population is limited. Based on the data of the Surveillance, Epidemiology, and End Results database (SEER), an analysis was performed to fill the gap of our knowledge.

Methods

Characteristics, treatment and outcomes of all pulmonary LELC patients was extracted both from the SEER database from 1973 to 2011 using SEER*Stat 8.2.1 statistical analysis was performed using SPSS 16.0 and GraphPad Prism 5.

Results

A total of 62 patients with pulmonary LELC are identified and analyzed. The median age at diagnosis is 65. Among them, the majority was male (64.4%). Early stage patients account for the largest proportion (67.8%). The median survival of all LELC patients is 107 months [95% confidence interval (CI), 67–147]. The 1, 3 and 5 years survival rates of LELC are 85.6%, 74.5% and 55.2%. In the comparisons incorporating with other types of large cell lung cancer (LCC), adenocarcinoma (AD) and squamous cell lung cancer (SQ), the overall survival (OS) of LELC is superior to others. Most of the early stage (localized and regional) LELC patients (37/45, 82.2%) received surgical resection as the primary treatment. Patients older than 65 years predicted a worse prognosis.

Conclusions

Pulmonary LELC is a rare pathological type of lung cancer. In this cohort, most LELC cases were male and in early stage. Majority of early stage LELC patients have received surgical resection. Patients older than 65 years had worse survival. Unfortunately, no other prognostic factor has been identified in our study. In addition, we observed that LELC had an ideal prognosis comparing to other types of LCC, AD and SQ. In order to understand pulmonary LELC more thoroughly, more cases are required.     

Colorectal cancer in patients under close colonoscopic surveillance

BACKGROUND & AIMS: Colonoscopic polypectomy is considered effective for preventing colorectal cancer (CRC), but the incidence of cancer in patients under colonoscopic surveillance has rarely been investigated. We determined the incidence of CRC in patients under colonoscopic surveillance and examined the circumstances and risk factors for CRC and adenoma with high-grade dysplasia. METHODS: Patients were drawn from 3 adenoma chemoprevention trials. All underwent baseline colonoscopy with removal of at least one adenoma and were deemed free of remaining lesions. We identified patients subsequently diagnosed with invasive cancer or adenoma with high-grade dysplasia. The timing, location, and outcome of all cases of cancer and high-grade dysplasia identified are described and risks associated with their development explored. RESULTS: CRC was diagnosed in 19 of the 2915 patients over a mean follow-up of 3.7 years (incidence, 1.74 cancers/1000 person-years). The cancers were located in all regions of the colon; 10 were at or proximal to the hepatic flexure. Although most of the cancers (84%) were of early stage, 2 participants died of CRC. Seven patients were diagnosed with adenoma with high-grade dysplasia during follow-up. Older patients and those with a history of more adenomas were at higher risk of being diagnosed with invasive cancer or adenoma with high-grade dysplasia. CONCLUSIONS: CRC is diagnosed in a clinically important proportion of patients following complete colonoscopy and polypectomy. More precise and representative estimates of CRC incidence and death among patients undergoing surveillance examinations are needed.     

Prognostic factors for survival in terminal lung cancer patients

Objective: To determine the factors that are related to short-term survival and to develop a model that can be used to estimate prognosis in terminal lung cancer patients. Design: Longitudinal cohort study of hospice lung cancer patients followed from date of admission to hospice until death. Setting: Community-based nonprofit home hospice service. Patients: Three hundred ten consecutive lung cancer patients admitted to hospice, with a separate validation sample of 78 consecutive hospice lung cancer patients. Measurements: The relationships between survival and admission demographic characteristics, information from the history and physical examination, assessments of performance and nutrition, particular symptoms, and the presence of a living will were evaluated. Results: Mean survival was 51 days, with a median survival of 27 days. Shorter survival was independently associated with those who had no living will on admission to hospice (p=0.008), those who had tissue types other than squamous cell or adenocarcinoma (p=0.008), those who had liver metastases (p=0.04), those who were tachycardic (p<0.001), those who required assistance or were dependent in their toileting (p<0.001) and feeding (p=0.001), those who had dry mouths (p=0.01), and those who had severe or incapacitating pain (p<0.05). A model estimating survival time based on the number of these significant variables present is reported (r=0.53 in the original sample; r=0.38 in the validation sample). Conclusions: Multiple factors, including tissue type, the presence of metastases, assessments of functional status, specific symptoms, and the presence of a living will, were related to short-term survival in terminal lung cancer patients admitted to hospice. A model utilizing these specific factors allows useful estimates of short-term survival for these patients. Presented at the 14th Annual Meeting and Symposium of the National Hospice Organization, Nashville, Tennessee, October 31, 1992.     

Small intestinal neuroendocrine tumors: Prognostic factors and survival

Objective. Small intestinal neuroendocrine tumors (SI-NETs) make up 38% of gastroenteropancreatic neuroendocrine tumors. We report our experience with SI-NETs at the National Center for Neuroendocrine Tumors in Norway, focusing on prognostic factors and survival. Material and methods. The medical records of 258 patients with SI-NETs diagnosed between 1983 and 2007 were retrospectively reviewed. Demographic, clinical and tumor characteristics were registered in a database. Results. Median age at diagnosis was 62 years (range 28–84); 53% of patients were men. Median survival was 9.3 years [95% confidence interval (CI) 7.6;10.8]. Survival did not improve for patients diagnosed between 1998 and 2007 compared with those diagnosed between 1990 and 1997 (p=0.44), median survival 8.1 [7.1;9.1] versus 6.8 [4.0;9.5] years. Overall 5-year survival was 72%, while expected 5-year survival in the general population was 92%. The corresponding relative 5-year survival for the patient group was 78%. Distant metastases, urinary 5-hydroxyindoleacetic acid ratio ≥3.7 times the upper limit of normal, chromogranin A ratio ≥6.2 times the upper limit of normal, age ≥64, male gender, carcinoid heart disease, and Ki-67 ≥5% were associated with decreased survival. Using multivariate analysis, only distant metastases (hazard ratio (HR) 1.98 [1.04;3.76], p=0.04), chromogranin A ratio ≥6.2 (HR 1.90 [1.12;3.20], p=0.02), and age ≥64 (3.12 [1.93;5.04], p<0.001) remained independent predictors. Conclusions. Survival did not improve over the study period. Overall and relative 5-year survival compared favorably with that in population-based studies. Distant metastases, elevated chromogranin A levels, and advanced age were the only independent predictors of poor survival.     

The incidence and prognosis of thymic squamous cell carcinoma: A Surveillance,Epidemiology, and End Results Program population-based study

Background:Thymic carcinoma represents a rare type of malignant mediastinal tumor and has been the subject of controversy. Although independent prognostic factors related to thymic carcinoma have been investigated previously, few studies have focused specifically on the survival outcomes associated with thymic squamous cell carcinoma (TSCC). This study aims at presenting a survival analysis in this rare malignant disease at population level.Methods:We extracted the data of 216 patients with TSCC recorded from 1973 to 2015 from the Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute. The patients’ demographic features, clinical traits, and treatment factors were analyzed in order to identify prognostic factors, which correlate overall survival using the Kaplan–Meier method as well as a multivariate Cox regression model, for TSCC.Results:The majority of patients were male, Caucasian, married, and insured. Furthermore, 58.3%, 54.6%, and 59.7% of patients TSCC underwent surgery, radiotherapy, and chemotherapy respectively. In a multivariate analysis, age of diagnosis (hazard ratio [HR]: 1.022, 95% confidence interval [CI]: 1.003–1.040, P = .020), surgical treatment (HR: 0.282, 95% CI: 0.164–0.484, P = .000), and stage (regional vs distant HR: 0.532, 95% CI: 0.324–0.872, P = .013; localized vs distant HR: 0.297, 95% CI: 0.133–0.664, P = .003) correlated with increased overall survival, whereas adjuvant therapy, including chemotherapy and radiotherapy, did not correlate with survival. Among surgically treated patients, age of diagnosis and stage were associated with better overall survival, while chemotherapy and radiotherapy did not contribute significantly to overall survival.Conclusion:Surgery, age of diagnosis, and stage were associated with better overall survival among TSCC.     

A visualized dynamic prediction model for overall survival in patients diagnosed with brain metastases from lung squamous cell carcinoma

Introduction

Patients presenting with brain metastases (BMs) from lung squamous cell carcinoma (LUSC) often encounter an extremely poor prognosis. A well-developed prognostic model would assist physicians in patient counseling and therapeutic decision-making.

Methods

Patients with LUSC who were diagnosed with BMs between 2000 and 2018 were reviewed in the Surveillance, Epidemiology, and End Results (SEER) database. Using the multivariate Cox regression approach, significant prognostic factors were identified and integrated. Bootstrap resampling was used to internally validate the model. An evaluation of the performance of the model was conducted by analyzing the area under the curve (AUC) and calibration curve.

Results

A total of 1812 eligible patients' clinical data was retrieved from the database. Patients' overall survival (OS) was significantly prognosticated by five clinical parameters. The nomogram achieved satisfactory discrimination capacity, with 3-, 6-, and 9-month AUC values of 0.803, 0.779, and 0.760 in the training cohort and 0.796, 0.769, and 0.743 in the validation cohort. As measured by survival rate probabilities, the calibration curve agreed well with actual observations. There was also a substantial difference in survival curves between the different prognostic groups stratified by prognostic scores. For ease of access, the model was deployed on a web-based server.

Conclusions

In this study, a nomogram and a web-based predictor were developed to assist physicians with personalized clinical decisions and treatment of patients who presented with BMs from LUSC.