Importance. Treatment of women with stage IV breast cancer (BC) extends population-averaged survival by only a few months. These investigations will identify individual circumstances where appropriate therapy will extend survival while minimizing adverse events.Objective. Our goal is to develop high confidence deep learning (DL) models for predicting survival in individual stage IV breast cancer patients based on their unique circumstances generated by patient, cancer, treatment and adverse events variables. Our plan is to improve the predictive accuracy of deep learning (DL)-based predictive models by considering time-fixed covariates, i.e., patient and cancer data at the time of diagnosis, together with time-varying events that occur after initial diagnosis. Design, Setting, and Participants. We used the SEER-Medicare linked dataset from 1991-2016 to investigate women diagnosed with stage IV BC who enrolled at 65 years or older for age eligibility. We outlined time-fixed variables, including date of diagnosis, age, race, marital status, breast cancer stage, tumor grade, laterality, estrogen receptor (ER), progesterone receptor (PR), and human epidermal receptor 2 (HER2) status and comorbidity index, prior therapy, adverse events, and changes in comorbidity. We delineated the time-varying covariates at each visit, including administered treatments, adverse events, comorbidity index, and age. We extended four DL-based predictive survival models (DeepSurv and DeepHit. Nnet-survival and Cox-Time) that deal with right-censored time-to-event data to consider both a patient's time-fixed covariates and a patient's time-varying covariates. We predicted the survival of five hypothetical patients to demonstrate the model's utility. We found high concordance between the performance metric, time-dependent concordance, and each of the model's hyperparameters to demonstrated prediction validity.Results. By incorporating time-varying variables with the time-fixed variables, the models reduced the error rates of the concordance index, the most commonly applied evaluation metric in survival analysis, from 28-38% to 2-12%, and significantly improved the integrated Brier score, a metric of the model's discrimination and calibration. Conclusions and Relevance. By combining the consideration of time-fixed with those time-varying variables in our predictive models, we decreased the predictive error rate to under 10% in the predicted survival of stage IV BC patients. Our established models can predict survival in individual stage IV patients with high confidence based on their circumstance-specific situations generated by considering their unique patient, cancer, treatment and adverse events variabels. These models will serve as an important adjunct to treatment decisions in patients with stage IV BC to optimize therapy for extending patient lives and minimizing adverse events.

(1) Background: We aim to construct a machine learning (ML) algorithm to predict the risk of distant metastasis (DM) of T1 and T2gallbladder cancer (GBC); (2) Demographic and clinical pathological data of T1 and T2 GBC patients were extracted from the National Institutes of Health (NIH)’s Surveillance, Epidemiology, and End Results (SEER) database between 2004 and 2015 to develop seven ML algorithm models. Models were evaluated based on accuracy, precision, recall rate, F1- score, and area under the receiver operating characteristic curve (AUC); (3) Results:A total of 4371 patients were included in the study, of whom 764 (17.4%) developed DM. Multivariate logistic regression showed that age, histology, tumor size, T and N stages were independent factors in GBC with DM. A novel nomogram was established to predict distant metastasis in early T stage GBC patients. Evaluation indicators of the best model Random Forest (RF) were as follows: accuracy (0.828), recall rate (0.862), precision (0.811), F1- score (0.836), and AUC value (0.913); (4) Con-clusions: The RF model constructed in this study could accurately predict distant metastasis in GBC patients, which may provide clinicians with more personalized clinical decision-making recom-mendations.

The real-world outcomes of patients with metastatic prostate cancer (mPCa) are largely unexplored. We investigated the improvements in overall survival (OS) and cancer-specific survival (CSS) in patients with de novo mPCa in latest years. The USA SEER Research Data (2000-2017) were analyzed using the SEER*Stat software. The Kaplan-Meier method and Cox regression were used. Patients with de novo mPCa were allocated to 3 cohorts based on year of diagnosis: A (2000-2003), B (2004-2010), C (2011-2014). Maximum follow-up was fixed to 5 years. Overall, 26434 patients were included. Age, race and metastatic stage significantly affected OS and CSS. After adjustment for age and race, patients in cohort C showed 9% reduced risk of death (HR:0.91 [95% CI, 0.87-0.95], p<0.001) and 8% reduced risk of cancer-specific death (HR:0.91 [95% CI, 0.87-0.95], p<0.001) compared to those in cohort A. After adjustment for age, race and metastatic stage, patients in cohort C showed an improvement in OS and CSS compared to cohort B (HR:0.94 [95% CI, 0.91-0.97], p=0.001 and HR:0.89 [95% CI, 0.85-0.92], p<0.001). Patients with M1c disease had a more pronounced improvement in OS and CSS compared with the other stages. No differences were found between cohort B and C. In conclusion, the prognosis of de novo mPCa remains poor with a median OS of 30 months and a median CSS of 35 months. Limited OS and CSS improvements were observed in latest years.

Background: Gastrointestinal stromal tumors (GIST) are rare mesenchymal neoplasms of the gastrointestinal tract (GIT) that represent approximately 1 to 2 percent of primary gastrointestinal (GI) cancers. Owing to their rarity, very little is known about the overall epidemiology and prognostic factors of the pathology. The purpose of this study is to investigate the clinical characteristics, survival outcomes, and independent prognostic factors of patients with GIST in the past decade. Methods: A total of 2,374 patients diagnosed with GIST, between 2010 and 2017, were ultimately enrolled in our study by retrieving the Surveillance, Epidemiology, and End Results (SEER) database. We analyzed demographics, clinical characteristics, and overall mortality (OM) as well as cancer-specific mortality (CSM) of GIST. Variables with a p value < 0.01 in the univariate Cox regression were incorporated into the multivariate Cox model to determine the independent prognostic factors, with a hazard ratio (HR) of greater than 1 representing adverse prognostic factors. Results: Multivariate cox proportional hazard regression analyses of factors affecting all-cause mortality and GIST related mortality among US patients between 2010 and 2017 revealed higher overall mortality in Non-Hispanic Blacks (HR= 1.516, 95% CI 1.172-1.961, p= 0.002), age 80+ (HR= 9.783, 95% CI4.185-22.868, p= 0), followed by age 60-79 (HR= 3.408, 95% CI 1.488-7.807, p=0.004); male patients (HR= 1.795 , 95% CI 1.461-2.206, p=0); advanced disease with distant metastasis (HR= 3.865 , 95% CI 2.977-5.019, p=0), followed by regional involvement by both direct extension and lymph node involvement (HR= 3.853, 95% CI 1.551-9.57, p=0.004); and widowed patients (HR= 1.975, 95% CI 1.494-2.61, p= 0), followed by single patients (HR= 1.53, 95% CI 1.154-2.028, p=0.003). The highest CSM was observed in the same groups, except widowed patients and patients aged 60-79. The highest CSM was also observed among patients that underwent chemotherapy (HR= 1.687, 95% CI 1.19-2.392, p= 0.003). Conclusion: In this United States population-based retrospective cohort study using the SEER database, we found that non-Hispanic blacks, male patients, and patients older than 60 years have a higher mortality with GIST. Furthermore, patients who received chemotherapy have a higher GIST specific mortality and married patients had a lower mortality. However, we do not know to what extent these independent prognostic factors interact with each other to influence mortality. This study paves the way for future studies addressing those interactions. The results of this study may help treating clinicians to identify patient populations associated with dismal prognosis as those may require closer follow-up and more intensive therapy; furthermore, with married patients having a better survival, we hope to encourage clinicians to involve family members of the affected patients early in the disease course as the social support might impact the prognosis.

Background: Primary cardiac sarcomas (PCS) are extremely rare malignant tumors involving the heart. Only isolated case reports have been described. There is a paucity of data on the epidemiological characteristics of PCS. This study has the objective of investigating the epidemiologic characteristics, survival outcomes, and independent prognostic factors of PCS. Methods: We enrolled a total of 362 patients with PCS, between 2000 and 2017, by retrieving the Surveillance, Epidemiology, and End Results (SEER) database. We analyzed demographics, clinical characteristics, and overall mortality (OM) as well as cancer-specific mortality (CSM) of PCS. Variables with a p-value < 0.1 in the univariate Cox regression were incorporated into the multivariate Cox model to determine the independent prognostic factors, with a hazard ratio (HR) of greater than 1 representing adverse prognostic factors. Results: Crude analysis revealed a high OM in age 80+ (HR=5.958, 95% CI 3.357-10.575, p=0), followed by age 60-79 (HR=1.429, 95% CI 1.028-1.986, p=0.033); and PCS with distant metastases (HR=1.888, 95% CI 1.389-2.566, p=0). Patients that underwent surgical resection of the primary tumor and patients with malignant fibrous histiocytomas (HR=0.657, 95% CI 0.455-0.95, p=0.025) had a better OM (HR=0.606, 95% CI 0.465-0.791, p=0). The highest cancer-specific mortality was observed in age 80+ (HR=5.037, 95% CI 2.606-9.736, p=0) and patients with distant metastases (HR=1.953, 95% CI 1.396-2.733, p=0). Patients with malignant fibrous histiocytomas (HR=0.572, 95% CI 0.378-0.865, p=0.008) and those who underwent surgery (HR=0.581, 95% CI 0.436-0.774, p=0) had a lower CSM. Multivariate Cox proportional hazard regression analyses revealed higher OM in the age group 80+ (HR=13.261, 95% CI 5.839-30.119, p=0) and advanced disease with distant metastases (HR=2.013, 95% CI 1.355-2.99, p=0.001). Lower OM was found in patients with rhabdomyosarcoma (HR=0.364, 95% CI 0.154-0.86, p=0.021) and widowed patients (HR=0.506, 95% CI 0.263-0.977, p=0.042). Multivariate Cox proportional hazard regression analyses of CSM also revealed higher mortality in the same groups, and lower mortality in patients with Rhabdomyosarcoma. Conclusion: In this United States population-based retrospective cohort study using the SEER database, we found that cardiac rhabdomyosarcoma was associated with the lowest CSM and OM. Furthermore, as expected, age and advanced disease at diagnosis were independent factors predicting poor prognosis. Surgical resection of the primary tumor showed lower CSM and OM in the crude analysis but when adjusted for covariates in the multivariate analysis, it did not significantly impact the overall mortality or the cancer-specific mortality. These findings allow for treating clinicians to recognize patients that should be referred to palliative/hospice care at the time of diagnosis and avoid any surgical interventions as they did not show any differences in mortality. Surgical resection in patients with poor prognoses should be reserved as a palliative measure rather than an attempt to cure the disease.

Background. This study aimed to establish a prognostic nomogram for predicting overall survival (OS) and cancer-specific survival (CSS) of patients with early-onset endometrial cancer (EOEC). Methods. Patients diagnosed with EOEC during 2004-2015 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate Cox regression analyses were performed to ascertain the independent factors impacting the prognosis of EOEC patients, while a prognostic nomogram was established. Its performance was verified by the receiver operating characteristic (ROC) curves, and calibration plots. The prediction accuracy and clinical net benefit between the model, TNM stage and SEER stage were further compared by the concordance index (C-index) and decision curve analysis (DCA). Results. A total of 4415 eligible patients with EOEC were identified and enrolled in this study. Independent prognostic factors for OS and CSS were confirmed by multivariate Cox regression, which were further incorporated to constructed a nomogram. ROC curves and calibration curves indicated an optimal accuracy of the nomogram. Meanwhile, the C-index and DCA showed the predictive capability and the clinical applicability of the nomogram was superior over the TNM stage and SEER stage. Conclusions. Nomogram based on risk factors was developed as an individualized tool to predict OS and CSS for EOEC patients. The novel tool will assist in patient counseling and selection of appropriate therapeutic regimens.

Background: Atypical teratoid/rhabdoid tumor (AT/RT) is an uncommon and aggressive pediatric central nervous system neoplasm. However, a universal clinical consensus or reliable prognostic evaluation system for this malignancy is lacking. Our study aimed to develop a risk model based on comprehensive clinical data to assist in clinical decision-making. Methods: We conducted a retrospective study by examining data from the Surveillance, Epidemiology, and End Results (SEER) repository, spanning 2000 to 2019. The external validation cohort was sourced from the Children's Hospital Affiliated to Chongqing Medical University, China. To discern independent factors affecting overall survival (OS) and cancer-specific survival (CSS), we applied Least Absolute Shrinkage and Selection Operator (LASSO) and Random Forest (RF) regression analyses. Based on these factors, we structured nomogram survival predictions and initiated a dynamic online risk-evaluation system. To contrast survival outcomes among diverse treatments, we used propensity score matching (PSM) methodology. Molecular data with the most common mutations in AT/RT were extracted from the Catalogue of Somatic Mutations in Cancer (COSMIC) database. Results: The annual incidence of AT/RT showed an increasing trend (APC, 2.86%; 95% CI:0.75-5.01). Our prognostic study included 316 SEER database participants and 27 external validation patients. The entire group had a median OS of 18 months (range 11.5 to 24 months) and median CSS of 21 months (range 11.7 to 29.2). Evaluations involving C-statistics, DCA, and ROC analysis underscored the distinctive capabilities of our prediction model. An analysis via PSM highlighted that individuals undergoing triple therapy (integrating surgery, radiotherapy, and chemotherapy) had discernibly enhanced OS and CSS. The most common mutations of AT/RT identified in the COSMIC database were SMARCB1, BRAF, SMARCA4, NF2, and NRAS. Conclusions: In this study, we devised a predictive model that effectively gauges the prognosis of AT/RT and briefly analyzed its genomic features, which might offer a valuable tool to address existing clinical challenges.