Surveillance, Epidemiology, And End Results Dataset Research Articles

A comparative performance assessment of artificial intelligence based classifiers and optimized feature reduction technique for breast cancer diagnosis

Breast cancer (BC) is a catastrophic global health concern that causes numerous fatalities worldwide. Early detection of breast cancer may mitigate death rates; however, the prevailing diagnostic procedure for the malignancy necessitates numerous multifaceted laboratory tests that must be performed by medical professionals. In this article machine learning, a branch of Artificial Intelligence (AI), has been employed to improve cancer diagnosis, prognoses and survival rates while reducing the vulnerability of humans. Support Vector Machine (SVM), K-nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF) and Grey Wolf Optimizer (GWO) are implemented to prognosticate breast cancer. Comprehensive insights into the efficacy of these approaches for breast cancer prognosis are provided by the performance assessment that is accomplished using the confusion matrix, Receiver Operating Characteristic (ROC) curves and parallel coordinate plots. Both UCI (University of California Irvine) and SEER (Surveillance, Epidemiology and End Results) datasets have been utilized to confirm the investigation's findings and ensure their generalizability across diverse data sources. The results conclusively demonstrate that SVM is the cohort's most accurate classifier. With a stupendous accuracy rate of 99.1 %, the GWO-SVM compares favorably to all other algorithms. Furthermore, feature reduction approaches such as Minimum Redundancy Maximum Relevance (mRMR), ReliefF and Principal Component Analysis (PCA) are utilized. ReliefF has demonstrated exceptional effectiveness with a maximum accuracy of 98.2 %.

Development and validation of a deep learning-based survival prediction model for pediatric glioma patients: A retrospective study using the SEER database and Chinese data

ObjectiveDevelop a time-dependent deep learning model to accurately predict the prognosis of pediatric glioma patients, which can assist clinicians in making precise treatment decisions and reducing patient risk. Study designThe study involved pediatric glioma patients from the Surveillance, Epidemiology, and End Results (SEER) Registry (2000–2018) and Tangdu Hospital in China (2010–2018) within specific time frames. For training, we selected two neural network-based algorithms (DeepSurv, neural multi-task logistic regression [N-MTLR]) and one ensemble learning-based algorithm (random survival forest [RSF]). Additionally, a multivariable Cox proportional hazard (CoxPH) model was developed for comparison purposes. The SEER dataset was randomly divided into 80 % for training and 20 % for testing, while the Tangdu Hospital dataset served as an external validation cohort. Super-parameters were fine-tuned through 1000 repeated random searches and 5-fold cross-validation on the training cohort. Model performance was assessed using the concordance index (C-index), Brier score, and Integrated Brier Score (IBS). Furthermore, the accuracy of predicting survival at 1, 3, and 5 years was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and the area under the ROC curves (AUC). The generalization ability of the model was assessed using the C-index of the Tangdu Hospital data, ROC curves for 1, 3, and 5 years, and AUC values. Lastly, decision curve analysis (DCA) curves for 1, 3, and 5-year time frames are provided to assess the net benefits across different models. ResultsA total of 9532 patients with pediatric glioma were included in this study, comprising 9274 patients from the SEER database and 258 patients from Tangdu Hospital in China. The average age at diagnosis was 9.4 ± 6.2 years, and the average survival time was 96 ± 66 months. Through comprehensive performance comparison, the DeepSurv model demonstrated the highest effectiveness, with a C-index of 0.881 on the training cohort. Furthermore, it exhibited excellent accuracy in predicting the 1-year, 3-year, and 5-year survival rates (AUC: 0.903–0.939). Notably, the DeepSurv model also achieved remarkable performance and accuracy on the Chinese dataset (C-index: 0.782, AUC: 0.761–0.852). Comprehensive analysis of DeepSurv, N-MTLR, and RSF revealed that tumor stage, radiotherapy, histological type, tumor size, chemotherapy, age, and surgical method are all significant factors influencing the prognosis of pediatric glioma. Finally, an online version of the pediatric glioma survival predictor based on the DeepSurv model has been established and can be accessed through https://pediatricglioma-tangdu.streamlit.app. ConclusionsThe DeepSurv model exhibits exceptional efficacy in predicting the survival of pediatric glioma patients, demonstrating strong performance in discrimination, calibration, stability, and generalization. By utilizing the online version of the pediatric glioma survival predictor, which is based on the DeepSurv model, clinicians can accurately predict patient survival and offer personalized treatment options.

Survival benefits of radiotherapy in locally advanced unresectable and metastatic pancreatic cancer: a single-institution cohort and SEER database analysis.

Chemotherapy (CT) remains the primary treatment for locally advanced unresectable pancreatic cancer (LAUPC) and metastatic pancreatic cancer (MPC). The role of radiotherapy (RT) in these conditions remains unclear. This study compares the outcomes of CT alone versus CT combined with RT (combined-modality therapy [CMT]) in LAUPC and MPC patients. We conducted a retrospective analysis of LAUPC and MPC patients treated with either CT or CMT from a single institution and Surveillance, Epidemiology, and End Results (SEER) database. Kaplan-Meier curves and Cox hazards models evaluated the association between treatment modalities and overall survival (OS). Propensity score matching (PSM) ensured balanced comparisons. Landmark analysis addressed immortal time bias. Subgroup analyses were based on clinical characteristics. eXtreme Gradient Boosting (XGBoost) and Shapley Additive Explanations (SHAP) assessed outcome prediction and influence of significant predictors. The study included 102 patients receiving CMT and 155 receiving CT at single institution, along with 1733 CMT and 9310 CT patients from the SEER dataset. In the single-institution cohort, CMT showed superior survival compared to CT both before (median OS: 20.5 vs. 11.5 months, hazard ratio [HR]: 0.47, 95% CI: 0.34-0.65, P=0.001) and after PSM (median OS: 22.2 vs. 11.8 months, HR: 0.49, 95% CI: 0.30-0.79, P=0.003). Multivariate analyses confirmed that CMT was independently associated with improved OS both before (HR: 0.54, 95% CI: 0.38-0.77, P=0.001) and after PSM (HR: 0.45, 95% CI: 0.27-0.73, P=0.001). Landmark analysis indicated better OS for patients receiving CMT compared to CT alone. Subgroup analysis revealed an OS benefit for CMT across most subgroups. SHAP value analysis indicated that CMT was the most significant contributor to survival outcomes. SEER database validation confirmed these findings. This study demonstrates that CMT significantly improves OS in LAUPC and MPC patients compared to CT alone. Integrating RT with CT could be beneficial for treating LAUPC and MPC.

Comparative prognosis analysis of ovarian squamous cell carcinoma versus serous carcinoma: Insights from the SEER database.

The aim of this study was to identify survival rates and potential prognostic factors of ovarian squamous cell carcinoma (OSCC), offering valuable insights for clinical decision making. Leveraging the Surveillance, Epidemiology, and End Results (SEER) database, we selected 11 078 serous carcinoma (SC) patients and 198 OSCC patients based on predetermined criteria diagnosed from 2000 to 2020. We compared the overall survival (OS) and cancer-specific survival (CSS) before and after propensity score matching (PSM) in two groups. Prognostic differences were also compared between OSCC and SC groups at different stages. Univariate and multivariate Cox regression analyses were performed to investigate the impact of clinical and pathologic variables on the survival of patients with OSCC. Finally, we developed and validated a nomogram predictive model. OSCC tumors exhibited distinct characteristics, being relatively larger, more frequently unilateral, and better differentiated than SC tumors. After PSM, Kaplan-Meier analysis revealed significantly lower survival rates for OSCC patients in Stages IIB-IV, while Stages IA-IC displayed comparable survival. Independent risk factors for OSCC patients included advanced age, single marital status, higher tumor stage, and increased tumor size. Conversely, higher median household income and chemotherapy emerged as independent protective factors. Our predictive model and nomogram accurately forecasted patient survival rates in both SEER and internal validation datasets. OSCC patients face significantly poorer prognosis than their SC counterparts, except in the very early stages. Higher median household income was associated with better OSCC survival.

Evolving patterns in systemic treatment utilization and survival among older patients with advanced cutaneous melanoma.

In the last decade, melanoma treatment has improved significantly. However, data on population-level treatment utilization and survival trends among older patients is limited. This study aimed to analyze trends in systemic anticancer therapy (Rx), including the uptake of immune checkpoint inhibitors (ICIs), in conjunction with trends in cause-specific survival among older patients (66+) diagnosed with advanced melanoma (2008-2019). We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare Condensed Resource to assess any Rx utilization among patients first diagnosed with advanced melanoma in 2008-2010, 2011-2014, and 2015-2019, stratified by stage, and type of first-line Rx among patients receiving Rx. The SEER dataset was used to evaluate trends in cause-specific survival by year of diagnosis. Rx utilization (any type) almost doubled, from 28.6% (2008-2010) to 55.4% (2015-2019) for stage 3 melanoma, and from 35.5% to 68.0% for stage 4 melanoma. In 2008-2010, the standard first-line treatment was cytokines/cytotoxic chemotherapy/other. By 2015-2019, only 5.1% (stage 3) and <3.6% (stage 4) of patients receiving Rx received these agents, as ICIs emerged as the dominant treatment. Both 1-year and 5-year cause-specific survival significantly improved since 2010 for stage 4 and since 2013 for stage 3. This study shows a significant rise in Rx utilization and a rapid transition from cytokines/cytotoxic chemotherapy to ICIs, reflecting a rapid uptake of highly effective treatment in a previously challenging disease with limited options before 2011. The documented survival improvement aligns with the adoption of these novel treatments, underscoring their significant impact on real-world patient outcomes.

Critical Risk Assessment, Diagnosis, and Survival Analysis of Breast Cancer.

Breast cancer is the most prevalent type of cancer in women. Risk factor assessment can aid in directing counseling regarding risk reduction and breast cancer surveillance. This research aims to (1) investigate the relationship between various risk factors and breast cancer incidence using the BCSC (Breast Cancer Surveillance Consortium) Risk Factor Dataset and create a prediction model for assessing the risk of developing breast cancer; (2) diagnose breast cancer using the Breast Cancer Wisconsin diagnostic dataset; and (3) analyze breast cancer survivability using the SEER (Surveillance, Epidemiology, and End Results) Breast Cancer Dataset. Applying resampling techniques on the training dataset before using various machine learning techniques can affect the performance of the classifiers. The three breast cancer datasets were examined using a variety of pre-processing approaches and classification models to assess their performance in terms of accuracy, precision, F-1 scores, etc. The PCA (principal component analysis) and resampling strategies produced remarkable results. For the BCSC Dataset, the Random Forest algorithm exhibited the best performance out of the applied classifiers, with an accuracy of 87.53%. Out of the different resampling techniques applied to the training dataset for training the Random Forest classifier, the Tomek Link exhibited the best test accuracy, at 87.47%. We compared all the models used with previously used techniques. After applying the resampling techniques, the accuracy scores of the test data decreased even if the training data accuracy increased. For the Breast Cancer Wisconsin diagnostic dataset, the K-Nearest Neighbor algorithm had the best accuracy with the original dataset test set, at 94.71%, and the PCA dataset test set exhibited 95.29% accuracy for detecting breast cancer. Using the SEER Dataset, this study also explores survival analysis, employing supervised and unsupervised learning approaches to offer insights into the variables affecting breast cancer survivability. This study emphasizes the significance of individualized approaches in the management and treatment of breast cancer by incorporating phenotypic variations and recognizing the heterogeneity of the disease. Through data-driven insights and advanced machine learning, this study contributes significantly to the ongoing efforts in breast cancer research, diagnostics, and personalized medicine.

Optimizing feature selection and parameter tuning for breast cancer detection using hybrid GAHBA-DNN framework

Breast cancer has been life-threatening for many years as it is the common cause of fatality among women. The challenges of screening such tumors through manual approaches can be overcome by computer-aided diagnosis, which aids radiologists in making precise decisions. The selection of significant features is crucial for the estimation of prediction accuracy. This work proposes a hybrid Genetic Algorithm (GA) and Honey Badger Algorithm (HBA) based Deep Neural Network (DNN), HGAHBA-DNN for the concurrent optimal features selection and parameter optimization; further, the optimal features and parameters extracted are fed into the DNN for the prediction of the breast cancer. It fuses the benefits of HBA with parallel processing and efficient feedback with GA’s excellent global convergent rate during the processing stages. The aforementioned method is evaluated on the Wisconsin Original Breast Cancer (WOBC), Wisconsin Diagnostic Breast Cancer (WDBC), and the Surveillance, Epidemiology, and End Results (SEER) datasets. Subsequently, the performance is validated using several metrics like accuracy, precision, Recall, and F1-score. The experimental result shows that HGAHBA-DNN obtains accuracy of 99.42%, 99.84%, and 92.44% for the WOBC, WDBC, and SEER datasets respectively, which is much superior to the other state-of-the-art methods.

Development and validation of a deep learning model for predicting postoperative survival of patients with gastric cancer

BackgroundDeep learning (DL), a specialized form of machine learning (ML), is valuable for forecasting survival in various diseases. Its clinical applicability in real-world patients with gastric cancer (GC) has yet to be extensively validated.MethodsA combined cohort of 11,414 GC patients from the Surveillance, Epidemiology and End Results (SEER) database and 2,846 patients from a Chinese dataset were utilized. The internal validation of different algorithms, including DL model, traditional ML models, and American Joint Committee on Cancer (AJCC) stage model, was conducted by training and testing sets on the SEER database, followed by external validation on the Chinese dataset. The performance of the algorithms was assessed using the area under the receiver operating characteristic curve, decision curve, and calibration curve.ResultsDL model demonstrated superior performance in terms of the area under the curve (AUC) at 1, 3, and, 5 years post-surgery across both datasets, surpassing other ML models and AJCC stage model, with AUCs of 0.77, 0.80, and 0.82 in the SEER dataset and 0.77, 0.76, and 0.75 in the Chinese dataset, respectively. Furthermore, decision curve analysis revealed that the DL model yielded greater net gains at 3 years than other ML models and AJCC stage model, and calibration plots at 3 years indicated a favorable level of consistency between the ML and actual observations during external validation.ConclusionsDL-based model was established to accurately predict the survival rate of postoperative patients with GC.

Influenza activity and regional mortality for non-small cell lung cancer

Lung cancer is the leading cause of cancer deaths in the United States and worldwide. While influenza illness is known to be particularly dangerous for frail and elderly patients, the relationship between influenza illness and outcomes in patients with cancer remains largely unknown. The Surveillance, Epidemiology, and End Results (SEER) database was queried to identify patients with non-small cell lung cancer (NSCLC) diagnosed between 2009 and 2015. Influenza-like illness (ILI) activity, provided by the Outpatient Influenza-like Illness Surveillance Network of the Center of Disease for Control and Prevention, was merged with the SEER dataset on the state-month level. Regional monthly mortality rates were compared during low versus high flu months in this ecological cohort study. 202,485 patients with NSCLC from 13 SEER-reporting states were included in the analysis. 53 of 1049 state-months (5.1%) had high flu activity. Monthly mortality rates during low and high flu months were 0.041 (95% CI 0.041–0.042) and 0.051 (95% CI 0.050–0.053), respectively (RR 1.24 [95% CI 1.21–1.27]). The association between ILI activity and mortality was observed at the individual state level and in all clinical and regional subgroups. Increased regional influenza activity is associated with higher mortality rates for patients with NSCLC. Vaccine-directed initiatives and increased awareness amongst providers will be necessary to address the growing but potentially preventable burden of influenza-related lung cancer deaths in the U.S.

Risk of second cancer in esophageal squamous cell carcinoma and adenocarcinoma survivors: a population-based analysis in SEER dataset.

Previous studies have reported increased risk of second cancer in both esophageal squamous cell cancer (ESCC) and esophageal adenocarcinoma (EAC) survivors. This study aimed to examine the risk and influential factors of second cancer in ESCC and EAC patients. This population-based cohort study included 7,297 ESCC patients and 11,812 EAC patients who were in 1992-2019 from the Surveillance, Epidemiology, and End Results (SEER) program in the United States. These patients were followed up until diagnosis of second cancer, death, or end of the study (December 31, 2019). We calculated standard incidence ratio (SIR) and 95% confidence interval (CI) of second cancer and performed competing-risk regression to estimate the subdistribution hazard ratios (sHR) comparing categories of patients' characteristics. After a total of 49,509.38 person-years of follow-up, 431 (5.9%) ESCC patients and 636 (5.9%) EAC patients developed a second cancer. An overall increased risk of second cancer was observed in both ESCC patients (SIR: 1.66, 95% CI: 1.51-1.83) and EAC patients (SIR: 1.11, 95% CI: 1.02-1.20). ESCC patients were at increased risk of second malignancy in oral cavity and pharynx (SIR: 12.57, 95% CI: 9.87-15.79), stomach (SIR: 3.03, 95% CI: 1.77-4.85), nose and larynx (SIR: 4.79, 95% CI: 2.47-8.37), and lung and bronchus (SIR: 2.44, 95% CI: 1.96-2.99), but decreased risk of prostate cancer (SIR: 0.73, 95% CI: 0.52-0.99). EAC patients had increased risk of second malignancies in stomach (SIR: 4.41, 95% CI: 3.23-5.89), lung and bronchus (SIR: 1.26, 95% CI: 1.02-1.54), and kidney (SIR: 1.57, 95% CI: 1.05-2.25). The risk of second cancer was higher in female ESCC patients than in males (sHR: 1.34, 95% CI: 1.11-1.63) and decreased with more advanced tumor stage in both ESCC patients (sHR: 0.62, 95% CI: 0.50-0.76 for regional stage; sHR: 0.27, 95% CI: 0.20-0.36 for distant stage) and EAC patients (sHR: 0.47, 95% CI: 0.40-0.56 for regional stage; sHR: 0.10, 95% CI: 0.07-0.13 for distant stage). Both ESCC and EAC patients are at considerable risk of certain types of second cancer.

Clinicopathological features and prognostic significance of pulmonary adenocarcinoma with signet ring cell components: meta-analysis and SEER analysis.

Pulmonary adenocarcinoma is a common type of lung cancer that has been on the rise in recent years. Signet ring cell components (SRCC) can be present in various patterns of pulmonary adenocarcinoma, including papillary, acinar, and solid patterns. "Signet ring cell carcinoma" is a distinct subtype in the 2014 WHO classification of lung neoplasms, subsequent WHO classifications in 2015 and 2021 have deemed signet ring cells as accompanying morphological features with no clinical significance. The prognostic and clinical implications of SRCC in pulmonary adenocarcinoma remain controversial. Therefore, we conducted a meta-analysis to investigate the clinicopathological features and prognostic factors of SRCC in pulmonary adenocarcinoma. We conducted a comprehensive search in PubMed, EMBASE, and Web of Science to identify studies that examined the clinicopathological features and prognostic implications of pulmonary adenocarcinoma with SRCC. We used both fixed- and random-effects models to analyze the data and calculate the pooled hazard ratio (HR) and odds ratio (OR) with 95% confidence intervals (CIs). Additionally, we explored the prognostic significance of SRCC in pulmonary adenocarcinoma using the Surveillance, Epidemiology, and End Results (SEER) database. Our meta-analysis included 29 studies with pulmonary adenocarcinoma and SRCC components. The results showed that pulmonary adenocarcinoma with SRCC was associated with larger tumor size (OR = 1.99; 95% CI, 1.62-2.44, p < 0.001), advanced overall stage (OR = 5.18, 95% CI, 3.28-8.17, p < 0.00001) and lymph node stage (OR = 5.79, 95% CI, 1.96-17.09, p = 0.001), and worse overall survival (OS) compared to those without SRCC (HR = 1.80, 95% CI, 1.50-2.16, p < 0.00001). Analysis using the SEER dataset confirmed these findings. Our meta-analysis provides evidence that pulmonary adenocarcinoma with SRCC is associated with distinct clinicopathological features and a poorer prognosis. These findings have important implications for the management and treatment of patients. However, further studies are needed to validate these findings and explore the significance of SRCC in various subtypes of pulmonary adenocarcinoma.

Secondary Breast Angiosarcoma After a Primary Diagnosis of Breast Cancer: A Retrospective Analysis of the Surveillance, Epidemiology, and End Results (SEER) Database.

Angiosarcoma is a rare complication of breast-conserving therapy. This study evaluated the change in incidence between 1992 and 2016 of secondary breast angiosarcoma (SBA) in patients with a history of breast cancer and the impact of management strategies for the original breast carcinoma on angiosarcoma treatment. Breast cancer and angiosarcoma cases were abstracted from the Surveillance, Epidemiology, and End Result (SEER) database. SBAs were defined as angiosarcomas located in the breast occurring after a prior breast cancer diagnosis. Primary breast angiosarcomas (PBAs) were defined as an angiosarcoma diagnosis listed as "one primary only." Incidence rates were estimated using a proportion of the US total population. Survival was analyzed by the Kaplan-Meier method, and Cox proportional hazard models were used to assess the association of clinicopathologic characteristics on overall survival. Between 1992 and 2016, 193 cases of SBA were reported in the SEER dataset in patients with a prior history of breast cancer. The incidence of breast angiosarcoma in patients with a prior diagnosis of breast cancer increased 3-fold from about 10 cases per 100,000 person-years to about 30 cases per 100,000 person-years over this same period ( P =0.0037). For treatment of SBA (n=193), almost all (95%) had surgery. Nine percent received radiation (compared with 35% of patients with PBA, P <0.001) and 23% received chemotherapy (vs. 45% for PBA, P =0.11). We demonstrate an increasing incidence of SBA over the study period. These data can help inform shared decision-making for optimal management of locoregional breast cancer and raise awareness of secondary angiosarcoma.

Oncologic Outcomes in Nipple-sparing Mastectomy with Immediate Reconstruction and Total Mastectomy with Immediate Reconstruction in Women with Breast Cancer: A Machine-Learning Analysis.

This study used a single-institution cohort, the Severance dataset, validated the results by using the surveillance, epidemiology, and end results (SEER) database, adjusted with propensity-score matching (PSM), and analyzed by using a machine learning method. To determine whether the 5-year, disease-free survival (DFS) and overall survival (OS) of patients undergoing nipple-sparing mastectomy (NSM) with immediate breast reconstruction (IBR) are not inferior to those of women treated with total mastectomy/skin-sparing mastectomy (TM/SSM). The Severance dataset enrolled 611 patients with early, invasive breast cancer from 2010 to 2017. The SEER dataset contained data for 485,245 patients undergoing TM and 14,770 patients undergoing NSM between 2000 and 2018. All patients underwent mastectomy and IBR. Intraoperative, frozen-section biopsy for the retro-areolar tissue was performed in the NSM group. The SEER dataset was extracted by using operation types, including TM/SSM and NSM. The primary outcome was DFS for the Severance dataset and OS for the SEER dataset. PSM analysis was applied. Survival outcomes were analyzed by using the Kaplan-Meier method and Cox proportional hazard (Cox PH) regression model. We implemented XGBSE to predict mortality with high accuracy and evaluated model prediction performance using a concordance index. The final model inspected the impact of relevant predictors on the model output using shapley additive explanation (SHAP) values. In the Severance dataset, 151 patients underwent NSM with IBR and 460 patients underwent TM/SSM with IBR. No significant differences were found between the groups. In multivariate analysis, NSM was not associated with reduced oncologic outcomes. The same results were observed in PSM analysis. In the SEER dataset, according to the SHAP values, the individual feature contribution suggested that AJCC stage ranks first. Analyses from the two datasets confirmed no impact on survival outcomes from the two surgical methods. NSM with IBR is a safe and feasible procedure in terms of oncologic outcomes. Analysis using machine learning methods can be successfully applied to identify significant risk factors for oncologic outcomes.

Investigating the Efficacy of HPV Vaccines in Preventing Cervical Cancer from 2006 to 2018 in the US: A SEER Data Set Analysis.

Human papillomavirus (HPV) is the most common sexually transmitted infection in the US.The first HPV vaccine was introduced in 2006. There are three different HPV vaccines that commonly target high-risk HPV types. This study compares HPV vaccine efficacy based on alternative endpoints with the most recently available cervical cancer incidence data from the Surveillance, Epidemiology and End Results (SEER) program and SEER*Stat statistical software. The incidence of cervical cancer, mined from the most recent April 2021 SEER data set, was stratified according to age and racial groups. Trend analysis reporting cervical cancer incidence percentage change (PC) and annual percentage change (APC) was calculated by SEER*Stat statistical software. A total of 46,583 cases of cervical cancer were reported, with an average of about 3,580 incidents of cervical cancer per year, with an overall decrement of about 60 cases over the period of 12 years. The percentage change according to age and race groups varied between -15.9 among 40- 44 years old (yo) and +13.8 among 30-34 yo, and from -12 among non-Hispanic White women to +13 among Hispanic women. Statistically significant APC was observed for five of the nine age groups and four of the five racial groups. There seems to be little if any, correlation between cervical cancer incidence and the HPV vaccine program in the US. HPV vaccine efficacy based on alternative endpoints, such as nucleic acid testing and cytological, surgical, and seropositivity endpoints, is fair. Therefore, it is important to emphasize such alternative testing and surrogate endpoints.

Web-based nomogram to predict postresection risk of distant metastasis in patients with leiomyosarcoma: retrospective analysis of the SEER database and a Chinese cohort.

This study investigated risk factors and constructed an online tool to predict distant metastasis (DM) risk in patients with leiomyosarcoma (LMS) after surgical resection. Data regarding patients with LMS who underwent surgical resection between 2010 and 2018 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Data were collected regarding patients with LMS who underwent surgical resection at Tianjin Medical University Cancer Hospital and Institute (TJMUCH) between October 2010 and July 2018. Patients were randomly divided into training and validation sets. Logistic regression analyses were performed; a nomogram was established. The area under the curve (AUC) and calibration curve were used to evaluate the nomogram, which served as the basis for a web-based nomogram. This study included 4461 and 76 patients from the SEER database and TJMUCH, respectively. Age, ethnicity, grade, T stage, N stage, radiotherapy, and chemotherapy were associated with DM incidence. C-index values were 0.815 and 0.782 in the SEER and Chinese datasets, respectively; corresponding AUC values were 0.814 and 0.773, respectively. A web-based nomogram (https://weijunqiang-leimyosarcoma-seer.shinyapps.io/dynnomapp/) was established. Our web-based nomogram is an accurate and user-friendly tool to predict DM risk in patients with LMS; it can aid clinical decision-making.

Clinicopathological characteristics and cancer-specific prognosis of primary pulmonary lymphoepithelioma-like carcinoma: a population study of the US SEER database and a Chinese hospital.

Primary pulmonary lymphoepithelioma-like carcinoma (PPLELC) is a rare histological type of non-small cell lung cancer (NSCLC), which accounts for less than 1% of NSCLC. Currently, there is no well-recognized treatment guideline for PPLELC. We identified PPLELC patients from the Surveillance, Epidemiology, and End Results (SEER) dataset between 2000 and 2015 (n = 72) as well as from our medical center between 2014 and 2020 (n = 16). All diagnoses were confirmed by pathological testing, and the clinicopathological characteristics of patients were retrieved and summarized. Survival analyses were conducted using the Kaplan-Meier analysis and log-rank tests. Multivariate survival analysis was performed with the Cox regression hazards model. The median age at diagnosis of the PPLELC cohort was 64 years, ranging from 15 to 86 years. The percentages of patients with TNM stages I, II, III, and IV were 52.3%, 10.2%, 20.5%, and 17.0%, respectively. Among the 88 cases, lesion resection was performed in 69 cases (78.4%), 16 cases (18.1%) received beam radiation, and 40 cases (45.5%) underwent chemotherapy. In the SEER dataset of lung cancer, the percentage of PPLELC in the Asian race (0.528‰) was almost 10 times higher than that in the white (0.065‰) and black (0.056‰) races. Patients with TNM stage III-IV exhibited a worse prognosis than those with TNM stage I-II (p = 0.008), with a 5-year cancer-specific survival (CSS) rate of 81.8% for TNM stage I-II and 56.2% for TNM stage III-IV. Specifically, the N stage and M stage were the leading prognostic factors, not the T stage and tumor size. Moreover, patients who underwent surgery had significantly better outcomes than those who did not (p = 0.014). Additional multivariate analysis indicated that the TNM stage was an independent prognosis factor for CSS (HR, 3.31; 95% CI, 1.08-10.14). PPLELC is a rare tumor with Asian susceptibility. Although the prognosis of PPLELC is better than that of other subtypes of NSCLC, it remains unsatisfactory for advanced-stage disease. The current treatment options for PPLELC include surgical resection, chemotherapy, radiotherapy, and immune therapy. Among these options, patients with surgical resection have better survival rates in this study. However, large-scale clinical research trials will be necessary to develop effective treatment guidelines for PPLELC.

Application of machine learning algorithm in predicting distant metastasis of T1 gastric cancer

Distant metastasis (DM) is relatively uncommon in T1 stage gastric cancer (GC). The aim of this study was to develop and validate a predictive model for DM in stage T1 GC using machine learning (ML) algorithms. Patients with stage T1 GC from 2010 to 2017 were screened from the public Surveillance, Epidemiology and End Results (SEER) database. Meanwhile, we collected patients with stage T1 GC admitted to the Department of Gastrointestinal Surgery of the Second Affiliated Hospital of Nanchang University from 2015 to 2017. We applied seven ML algorithms: logistic regression, random forest (RF), LASSO, support vector machine, k-Nearest Neighbor, Naive Bayesian Model, Artificial Neural Network. Finally, a RF model for DM of T1 GC was developed. The AUC, sensitivity, specificity, F1-score and accuracy were used to evaluate and compare the predictive performance of the RF model with other models. Finally, we performed a prognostic analysis of patients who developed distant metastases. Independent risk factors for prognosis were analysed by univariate and multifactorial regression. K-M curves were used to express differences in survival prognosis for each variable and subvariable. A total of 2698 cases were included in the SEER dataset, 314 with DM, and 107 hospital patients were included, 14 with DM. Age, T-stage, N-stage, tumour size, grade and tumour location were independent risk factors for the development of DM in stage T1 GC. A combined analysis of seven ML algorithms in the training and test sets found that the RF prediction model had the best prediction performance (AUC: 0.941, Accuracy: 0.917, Recall: 0.841, Specificity: 0.927, F1-score: 0.877). The external validation set ROCAUC was 0.750. Meanwhile, survival prognostic analysis showed that surgery (HR = 3.620, 95% CI 2.164–6.065) and adjuvant chemotherapy (HR = 2.637, 95% CI 2.067–3.365) were independent risk factors for survival prognosis in patients with DM from stage T1 GC. Age, T-stage, N-stage, tumour size, grade and tumour location were independent risk factors for the development of DM in stage T1 GC. ML algorithms had shown that RF prediction models had the best predictive efficacy to accurately screen at-risk populations for further clinical screening for metastases. At the same time, aggressive surgery and adjuvant chemotherapy can improve the survival rate of patients with DM.

Abstract P6-03-06: Determining prognostic factors and optimal surgical intervention for young patients with triple-negative breast cancer

Abstract Background Triple-negative breast cancer (TNBC) is more frequently diagnosed in young patients, with an incidence of 26% of this population compared to 12% overall, and is characterized by high malignancy and poor prognosis. Limited data are available that contribute to a comprehensive summarization of the prognostic factors and the determination of surgical strategy that are associated with young patients with TNBC. We aimed to determine the optimal surgical approach (breast-conserving versus mastectomy) for patients aged &amp;lt; 40 years with TNBC and establish a prognostic model. Methods We performed a cohort study with a median follow-up of 31 months using the Surveillance, Epidemiology, and End Results (SEER) data of young patients &amp;lt; 40 years diagnosed with stage I–III TNBC between 2010 and 2016. A Cox proportional hazards model was used to investigate the effects of baseline characteristics on breast cancer-specific survival (BCSS) and overall survival (OS). To ensure that differences in outcomes were not based on baseline differences in demographic and clinical characteristics, we performed Kaplan–Meier analysis before and after propensity score matching (1:1). Subgroup analyses stratified by TNM stage as well as further propensity score matching analyses were performed. A nomogram was constructed from the multivariate logistic regression to incorporate all the prognostic factors to predict the BCSS rates of patients at 3 years and 5 years. Young patients &amp;lt; 40 years diagnosed with stage I–III TNBC between 2006 and 2016 in Shenzhen Second People’s Hospital (SSPH) were enrolled as external validation. Results A total of 2,854 patients from SEER dataset and 250 from SSPH were included in this study. On multivariable analysis, unmarried status, lack of health insurance, advanced T stage, advanced N stage, invasive lobular carcinoma or mixed histologic type, were all significantly associated with poor BCSS and OS. Young patients with TNBC were more likely to undergo mastectomy than breast-conserving surgery. Notably, patients with T1N0M0 or T2-4N+M0 tumors who underwent breast-conserving surgery achieved longer BCSS and OS than those who underwent mastectomy; however, the type of surgery did not influence survival rates among patients with T1N+M0 or T2-4N0M0 tumors. The nomogram was constructed by the five variables and passed the calibration and validation steps (C-index: 0.774 for training cohort and 0.768 for validation cohort). The area under the receiver operating characteristic curves (AUCs) predicting the 3-year and 5-year BCSS rates were calculated (0.783 and 0.774 in training cohort; 0.786 and 0.772 in validating cohort). Conclusions A localized surgical approach may be a superior option for young patients with TNBC, especially those with T1N0M0 and T2-4N+M0 tumors. Marital status, health insurance status, T stage, N stage, and histological type were independent prognostic factors, and a nomogram established based on these variables successfully predicted the 3- and 5-year survival probabilities among these patients. Citation Format: Yi-Zi Zheng, Yan Liu, Zhen-Han Deng, Guo-Wen Liu, Ni Xie. Determining prognostic factors and optimal surgical intervention for young patients with triple-negative breast cancer [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-03-06.

Risk Stratification for Management of Solitary Fibrous Tumor/Hemangiopericytoma of the Central Nervous System.

Solitary fibrous tumor/hemangiopericytoma (SFT/HPC) of the central nervous system (CNS) is a rare meningeal tumor. Given the absence of prospective or randomized data, there are no standard indications for radiotherapy. Recently, the NRG Oncology and EORTC cooperative groups successfully accrued and completed the first prospective trials evaluating risk-adapted adjuvant radiotherapy strategies for meningiomas. Using a similar framework, we sought to develop prognostic risk categories that may predict the survival benefit associated with radiotherapy, using two large national datasets. We queried the National Cancer Database (NCDB) and the Surveillance, Epidemiology, and End Results (SEER) databases for all newly diagnosed cases of SFT/HPC within the CNS. Risk categories were created, as follows: low risk-grade 1, with any extent of resection (EOR) and grade 2, with gross-total resection; intermediate risk-grade 2, with biopsy/subtotal resection; high risk-grade 3 with any EOR. The Kaplan-Meier method and Cox proportional hazards regressions were used to determine the association of risk categories with overall and cause-specific survival. We then determined the association of radiotherapy with overall survival in the NCDB, stratified by risk group. We identified 866 and 683 patients from the NCDB and SEER databases who were evaluated, respectively. In the NCDB, the 75% survival times for low- (n = 312), intermediate- (n = 239), and high-risk (n = 315) patients were not reached, 86 months (HR 1.60 (95% CI 1.01-2.55)), and 55 months (HR 2.56 (95% CI 1.68-3.89)), respectively. Our risk categories were validated for overall and cause-specific survival in the SEER dataset. Radiotherapy was associated with improved survival in the high- (HR 0.46 (0.29-0.74)) and intermediate-risk groups (HR 0.52 (0.27-0.99)) but not in the low-risk group (HR 1.26 (0.60-2.65)). The association of radiotherapy with overall survival remained significant in the multivariable analysis for the high-risk group (HR 0.55 (0.34-0.89)) but not for the intermediate-risk group (HR 0.74 (0.38-1.47)). Similar results were observed in a time-dependent landmark sensitivity analysis. Risk stratification based on grade and EOR is prognostic of overall and cause-specific survival for SFT/HPCs of the CNS and performs better than any individual clinical factor. These risk categories appear to predict the survival benefit from radiotherapy, which is limited to the high-risk group and, potentially, the intermediate-risk group. These data may serve as the basis for a prospective study evaluating the management of meningeal SFT/HPCs.

Racial disparities in recommendations for surgical resection of primary brain tumours: a registry-based cohort analysis

Disparities in treatment and outcomes disproportionately affect minority ethnic and racial populations in many surgical fields. Although substantial research in racial disparities has focused on outcomes, little is known about how surgeon recommendations can be influenced by patient race. The aim of this study was to investigate racial and socioeconomic disparities in the surgical management of primary brain tumors. In this registry-based cohort study, we used data from the Surveillance, Epidemiology, and End Results (SEER) database (1975-2016) and the American College of Surgeons National Cancer Database (NCDB) in the USA for independent analysis. Adults (aged ≥20 years) with a new diagnosis of meningioma, glioblastoma, pituitary adenoma, vestibular schwannoma, astrocytoma, and oligodendroglioma, with information on tumour size and surgical recommendation were included in the analysis. The primary outcome of this study was the odds of a surgeon recommending against surgical resection at diagnosis of primary brain neoplasms. This outcome was determined using multivariable logistic regression with clinical, demographic, and socioeconomic factors. This study included US national data from the SEER (1975-2016) and NCDB (2004-17) databases of adults with a new diagnosis of meningioma (SEER n=63 674; NCDB n=222 673), glioblastoma (n=35 258; n=104 047), pituitary adenoma (n=27 506; n=87 772), vestibular schwannoma (n=11 525; n=30 745), astrocytoma (n=5402; n=10 631), and oligodendroglioma (n=3977; n=9187). Independent of clinical and demographic factors, including insurance status and rural-urban continuum code, Black patients had significantly higher odds of recommendation against surgical resection of meningioma (adjusted odds ratio 1·13, 95% CI 1·06-1·21, p<0·0001), glioblastoma (1·14, 1·01-1·28, p=0·038), pituitary adenoma (1·13, 1·05-1·22, p<0·0001), and vestibular schwannoma (1·48, 1·19-1·84, p<0·0001) when compared with White patients in the SEER dataset. Additionally, patients of unknown race had significantly higher odds of recommendation against surgical resection for pituitary adenoma (1·80, 1·41-2·30, p<0·0001) and vestibular schwannoma (1·49, 1·10-2·04, p=0·011). Performing a validation analysis using the NCDB dataset confirmed these significant results for Black patients with meningioma (1·18, 1·14-1·22, p<0·0001), glioblastoma (1·19, 1·12-1·28, p<0·0001), pituitary adenoma (1·21, 1·16-1·25, p<0·0001), and vestibular schwannoma (1·19, 1·04-1·35, p=0·0085), and indicated and indicated that the findings are independent of patient comorbidities. When further restricted to the most recent decade in SEER, these inequities held true for Black patients, except those with glioblastoma (meningioma [1·18, 1·08-1·28, p<0·0001], pituitary adenoma [1·20, 1·09-1·31, p<0·0001], and vestibular schwannoma [1·54, 1·16-2·04, p=0·0031]). Racial disparities in surgery recommendations in the USA exist for patients with primary brain tumours, independent of potential confounders including clinical, demographic, and select socioeconomic factors. Further studies are needed to understand drivers of this bias and enhance equality in surgical care. None.