A Web-Based Calculator to Predict Early Death Among Patients With Bone Metastasis Using Machine Learning Techniques: Development and Validation Study.

BACKGROUND: Patients with bone metastasis often experience a significantly limited survival time, and a life expectancy of <3 months is generally regarded as a contraindication for extensive invasive surgeries. In this context, the accurate prediction of survival becomes very important since it serves as a crucial guide in making clinical decisions. OBJECTIVE: This study aimed to develop a machine learning-based web calculator that can provide an accurate assessment of the likelihood of early death among patients with bone metastasis. METHODS: This study analyzed a large cohort of 118,227 patients diagnosed with bone metastasis between 2010 and 2019 using the data obtained from a national cancer database. The entire cohort of patients was randomly split 9:1 into a training group (n=106,492) and a validation group (n=11,735). Six approaches-logistic regression, extreme gradient boosting machine, decision tree, random forest, neural network, and gradient boosting machine-were implemented in this study. The performance of these approaches was evaluated using 11 measures, and each approach was ranked based on its performance in each measure. Patients (n=332) from a teaching hospital were used as the external validation group, and external validation was performed using the optimal model. RESULTS: In the entire cohort, a substantial proportion of patients (43,305/118,227, 36.63%) experienced early death. Among the different approaches evaluated, the gradient boosting machine exhibited the highest score of prediction performance (54 points), followed by the neural network (52 points) and extreme gradient boosting machine (50 points). The gradient boosting machine demonstrated a favorable discrimination ability, with an area under the curve of 0.858 (95% CI 0.851-0.865). In addition, the calibration slope was 1.02, and the intercept-in-large value was -0.02, indicating good calibration of the model. Patients were divided into 2 risk groups using a threshold of 37% based on the gradient boosting machine. Patients in the high-risk group (3105/4315, 71.96%) were found to be 4.5 times more likely to experience early death compared with those in the low-risk group (1159/7420, 15.62%). External validation of the model demonstrated a high area under the curve of 0.847 (95% CI 0.798-0.895), indicating its robust performance. The model developed by the gradient boosting machine has been deployed on the internet as a calculator. CONCLUSIONS: This study develops a machine learning-based calculator to assess the probability of early death among patients with bone metastasis. The calculator has the potential to guide clinical decision-making and improve the care of patients with bone metastasis by identifying those at a higher risk of early death.

A New Treatment Strategy for Spinal Metastasis: The "Systemic Conditions, Effectiveness of Systemic Treatment, Neurology, and Oncology" Decision Framework System.

BACKGROUND AND OBJECTIVES: Treating metastatic spinal tumors poses a significant challenge because there are currently no universally applied guidelines for managing spinal metastases. This study aims to propose a new decision framework for the 12-point epidural spinal cord compression grading system to treat patients with metastatic spinal tumors and investigate its clinical effectiveness in a multicenter analysis. METHODS: This study analyzed 940 patients with metastatic spinal tumors between December 2017 and March 2023. The study provided the clinical evidence for the systemic conditions, effectiveness of systemic treatment, neurology, and oncology (SENO) decision framework among spine metastases. The SENO decision framework was launched in January 2021 in our hospitals, classifying patients into 2 groups: The non-SENO group (n = 489) consisted of patients treated between December 2017 and January 2021, while the SENO group (n = 451) comprised patients treated from January 2021 to March 2023. RESULTS: Patients in the SENO group were more likely to receive minimally invasive surgery (67.85% vs 58.69%) and less chance of receiving spinal cord circular decompression surgery (14.41% vs 24.74%) than patients in the non-SENO group ( P < .001). Furthermore, patients in the SENO group experienced fewer perioperative complications (9.09% vs 15.34%, P = .004), incurred lower hospitalization costs ( P < .001), had shorter length of hospitalization ( P < .001), and received systematic treatments for tumors earlier ( P < .001). As a result, patients in the SENO group (329.00 [95% CI: 292.06-365.94] days) demonstrated significantly improved survival outcomes compared with those in the non-SENO group (279.00 [95% CI: 256.91-301.09], days) ( P < .001). At 3 months postdischarge, patients in the SENO group reported greater improvements in their quality of life, encompassing physical, social, emotional, and functional well-being, when compared with patients in the non-SENO group. CONCLUSION: The SENO decision framework is a promising approach for treating patients with metastatic spinal tumors.

Promoting postoperative recovery in patients with metastatic epidural spinal cord compression based on the concept of ERAS: a multicenter analysis of 304 patients.

BACKGROUND CONTEXT: Enhanced recovery after surgery (ERAS) has proven beneficial for patients undergoing orthopedic surgery. However, the application of ERAS in the context of metastatic epidural spinal cord compression (MESCC) remains undefined. PURPOSE: This study aims to establish a medical pathway rooted in the ERAS concept, with the ultimate goal of scrutinizing its efficacy in enhancing postoperative outcomes among patients suffering from MESCC. STUDY DESIGN/SETTING: An observational cohort study. PATIENT SAMPLE: A total of 304 patients with MESCC who underwent surgery were collected between January 2016 and January 2023 at two large tertiary hospitals. OUTCOME MEASURES: Surgery-related variables, patient quality of life, and pain outcomes. Surgery-related variables in the study included surgery time, surgery site, intraoperative blood loss, and complication. METHODS: From January 2020 onwards, ERAS therapies were implemented for MESCC patients in both institutions. Thus, the ERAS cohort included 138 patients with MESCC who underwent surgery from January 2020 to January 2023, whereas the traditional cohort consisted of 166 patients with MESCC who underwent surgery from January 2016 to December 2019. Clinical baseline characteristics, surgery-related features, and surgical outcomes were collected. Patient quality of life was evaluated using the Functional Assessment of Cancer Therapy-General Scale (FACT-G), and pain outcomes were assessed using the Visual Analogue Scale (VAS). RESULTS: Comparison of baseline characteristics revealed that the two cohorts were similar (all p>.050), indicating comparable distribution of clinical characteristics. In terms of surgical outcomes, patients in the ERAS cohort exhibited lower intraoperative blood loss (p<.001), shorter postoperative hospital stays (p<.001), lower perioperative complication rates (p=.020), as well as significantly shorter time to ambulation (P<0.001), resumption of regular diet (p<.001), removal of urinary catheter (p<.001), initiation of radiation therapy (p<.001), and initiation of systemic internal therapy (p<.001) compared with patients in the traditional cohort. Regarding pain outcomes and quality of life, patients undergoing the ERAS program demonstrated significantly lower VAS scores (p<.010) and higher scores for physical (p<.001), social (p<.001), emotional (p<.001), and functional (p<.001) well-being compared with patients in the traditional cohort. CONCLUSIONS: The ERAS program, renowned for its ability to expedite postoperative recuperation, emerges as a promising approach to ameliorate the recovery process in MESCC patients. Not only does it exhibit potential in enhancing pain management outcomes, but it also holds the promise of elevating the overall quality of life for these individuals. Future investigations should delve deeper into the intricate components of the ERAS program, aiming to unravel the precise mechanisms that underlie its remarkable impact on patient outcomes.

Establishment and validation of an interactive artificial intelligence platform to predict postoperative ambulatory status for patients with metastatic spinal disease: a multicenter analysis.

BACKGROUND: Identification of patients with high-risk of experiencing inability to walk after surgery is important for surgeons to make therapeutic strategies for patients with metastatic spinal disease. However, there is a lack of clinical tool to assess postoperative ambulatory status for those patients. The emergence of artificial intelligence (AI) brings a promising opportunity to develop accurate prediction models. METHODS: This study collected 455 patients with metastatic spinal disease who underwent posterior decompressive surgery at three tertiary medical institutions. Of these, 220 patients were collected from one medical institution to form the model derivation cohort, while 89 and 146 patients were collected from two other medical institutions to form the external validation cohorts 1 and 2, respectively. Patients in the model derivation cohort were used to develop and internally validate models. To establish the interactive AI platform, machine learning techniques were used to develop prediction models, including logistic regression (LR), decision tree (DT), random forest (RF), extreme gradient boosting machine (eXGBM), support vector machine (SVM), and neural network (NN). Furthermore, to enhance the resilience of the study's model, an ensemble machine learning approach was employed using a soft-voting method by combining the results of the above six algorithms. A scoring system incorporating 10 evaluation metrics was used to comprehensively assess the prediction performance of the developed models. The scoring system had a total score of 0 to 60, with higher scores denoting better prediction performance. An interactive AI platform was further deployed via Streamlit. The prediction performance was compared between medical experts and the AI platform in assessing the risk of experiencing postoperative inability to walk among patients with metastatic spinal disease. RESULTS: Among all developed models, the ensemble model outperformed the six other models with the highest score of 57, followed by the eXGBM model (54), SVM model (50), and NN model (50). The ensemble model had the best performance in accuracy and calibration slope, and the second-best performance in precise, recall, specificity, area under the curve (AUC), Brier score, and log loss. The scores of the LR model, RF model, and DT model were 39, 46, and 26, respectively. External validation demonstrated that the ensemble model had an AUC value of 0.873 (95% CI: 0.809-0.936) in the external validation cohort 1 and 0.924 (95% CI: 0.890-0.959) in the external validation cohort 2. In the new ensemble machine learning model excluding the feature of the number of comorbidities, the AUC value was still as high as 0.916 (95% CI: 0.863-0.969). In addition, the AUC values of the new model were 0.880 (95% CI: 0.819-0.940) in the external validation cohort 1 and 0.922 (95% CI: 0.887-0.958) in the external validation cohort 2, indicating favorable generalization of the model. The interactive AI platform was further deployed online based on the final machine learning model, and it was available at https://postoperativeambulatory-izpdr6gsxxwhitr8fubutd.streamlit.app/ . By using the AI platform, researchers were able to obtain the individual predicted risk of postoperative inability to walk, gain insights into the key factors influencing the outcome, and find the stratified therapeutic recommendations. The AUC value obtained from the AI platform was significantly higher than the average AUC value achieved by the medical experts ( P <0.001), denoting that the AI platform obviously outperformed the individual medical experts. CONCLUSIONS: The study successfully develops and validates an interactive AI platform for evaluating the risk of postoperative loss of ambulatory ability in patients with metastatic spinal disease. This AI platform has the potential to serve as a valuable model for guiding healthcare professionals in implementing surgical plans and ultimately enhancing patient outcomes.

[Research progress in separation surgery combined with stereotactic radiotherapy of spinal metastases].

With the continuous improvement of cancer treatment, the survival of patients with spinal metastases has been significantly prolonged. Currently, the treatment of spinal metastases presents a trend of multi-mode. Clinical surgical methods include vertebral tumor resecting spinal canal decompression and internal fixation surgery, separation surgery, minimally invasive surgery and percutaneous ablation technology, etc. Radiotherapy techniques include traditional external radiation therapy, stereotactic radiotherapy and brachytherapy, etc. The risk of vertebral tumor resecting spinal canal decompression and internal fixation surgery, and the incidence of intraoperative and postoperative complications is high. The extension of postoperative recovery period may lead to delay of follow-up radiotherapy and other medical treatment, which has a serious impact on patients' survival and treatment confidence. However, the precision of traditional external radiation therapy is not high, and the limitation of tolerance of spinal cord makes it difficult to achieve the goal of controlling insensitive tumor. With the development of radiotherapy and surgical technology, stereotactic radiotherapy with higher accuracy and separation surgery with smaller surgical strike have become the focus of many clinical experts at present. This article reviews the progress of Hybrid treatment of separation surgery combined with stereotactic radiotherapy.

[Progress of decompression surgery for spinal metastases].

The spine is the most common site of bone metastases from malignant tumors, with metastatic epidural spinal cord compression occurring in about 10% of patients with spinal metastases. Palliative radiotherapy and simple laminectomy and decompression have been the main treatments for metastatic spinal cord compression. The former is ineffective and delayed for radiation-insensitive tumors, and the latter often impairs spinal stability. With the continuous improvement of surgical techniques and instrumentation in recent years, the treatment model of spinal metastases has changed a lot. Decompression surgery underwent open decompression, separation surgery, minimally invasive surgery and laser interintermal thermal ablation decompression. However, no matter what kind of surgical plan is adopted, it should be assessed precisely according to the specific situation of the patient to minimize the risk of surgery as far as possible to ensure the smooth follow-up radiotherapy. This paper reviews the research progress of decompression for spinal metastases.

Development of a novel 12-point grading system for evaluating epidural spinal cord compression and its clinical implications.

BACKGROUND CONTEXT: The assessment of epidural spinal cord compression (ESCC) plays a crucial role in clinical decision-making, yet the current grading system lacks reliability and requires improvements. PURPOSE: The study aims to develop a reliable grading system for evaluating ESCC and to investigate its association with the neurological status of patients. STUDY DESIGN/SETTING: A prospective cohort study. PATIENT SAMPLE: A total of 330 patients with metastatic spinal disease were included in the study. OUTCOME MEASURES: The main outcome was the neurological status evaluated using the American Spinal Injury Association (ASIA) scale. METHODS: We proposed a novel grading system, called the 12-point ESCC grading system, to evaluate ESCC based on findings from spinal magnetic resonance imaging (MRI). This new grading system consists of 12 grades, ranging from Grade 0 to 3, with higher grades indicating more severe ESCC. The detailed information about the sagittal image of the spine and the severity of spinal cord swelling was considered in this new grading system. The Spearman correlation analysis and logistic regression analysis were employed to investigate the correlation between the previous 6-point grading system and ASIA, as well as between the new 12-point ESCC grading system and ASIA. The prediction effectiveness was evaluated using the area under curve (AUC) analysis. RESULTS: Patients with higher grades in the 12-point ESCC grading system exhibited a higher likelihood of experiencing a worse neurological condition. Specifically, patients with grades 2a to 2d and 3a to 3d according to the new 12-point ESCC grading system were significantly associated with more complete paralysis (p<.001) compared with patients with grade 0. The Spearman correlation coefficient was 0.729 between the previous 6-point ESCC grading system and ASIS and 0.750 between the new 12-point ESCC grading system and ASIS. When categorizing ASIS into complete paralysis and other neurological statuses, the 6-point ESCC score yielded an AUC of 0.820, which increased to 0.860 with the new 12-point ESCC grading system. Furthermore, when ASIS was divided into normal and abnormal neurological statuses, the AUC increased from 0.889 to 0.906. Additionally, spinal cord swelling was significantly associated with more complete paralysis (p<.001) and abnormal neurological status (p<.001) based on the new 12-point ESCC grading system. CONCLUSIONS: The new 12-point ESCC grading system provides more detailed information and further improves the prediction effectiveness for evaluating neurological status compared with the previous 6-point ESCC grading system. In the new 12-point ESCC grading system, higher grades or the presence of spinal cord swelling are indicative of a worse neurological condition.

Treatment of patients with metastatic epidural spinal cord compression using an enhanced recovery after surgery program.

Purpose: The aims of this study were to introduce a new medical, pathway based on the concept of "enhanced recovery after surgery" (ERAS) for patients with metastatic epidural spinal cord compression (MESCC), and to test whether the ERAS program could improve clinical metrics among such patients. Methods: Data from patients with MESCC (n = 98), collected between December 2016 and December 2019 (Non-ERAS cohort), and from 86 patients with metastatic epidural spinal cord compression collected between January 2020 and December 2022 (ERAS cohort), were retrospectively analyzed. Patients were treated by decompressive surgery combined with transpedicular screw implantation and internal fixation. Patient baseline clinical characteristics were collected and compared between the two cohorts. Surgical outcomes analyzed included operation time; intraoperative blood loss; postoperative length of hospital stay; time to ambulation, regular diet, urinary catheter removal, and radiation therapy; perioperative complications; anxiety; depression; and satisfaction with treatment. Results: No significant differences in clinical characteristics were found between the non-ERAS and enhanced recovery after surgery cohorts (all p > 0.050), indicating that the two cohorts were comparable. Regarding surgical outcomes, the enhanced recovery after surgery cohort had significantly less intraoperative blood loss (p < 0.001); shorter length of postoperative hospital stay (p < 0.001); shorter time to ambulation (p < 0.001), regular diet (p < 0.001), urinary catheter removal (p < 0.001), radiation administration (p < 0.001), and systemic internal therapy (p < 0.001); lower perioperative complication rate (p = 0.024); less postoperative anxiety (p = 0.041); and higher score for satisfaction with treatment (p < 0.001); whereas operation time (p = 0.524) and postoperative depression (p = 0.415) were similar between the two cohorts. Compliance analysis demonstrated that ERAS interventions were successfully conducted in the vast majority of patients. Conclusion: The enhanced recovery after surgery intervention is beneficial to patients with metastatic epidural spinal cord compression, according to data on intraoperative blood loss; length of hospital stay; time to ambulation, regular diet, urinary catheter removal, radiation exposure, and systemic internal therapy; perioperative complication; alleviation of anxiety; and improvement of satisfaction. However, clinical trials to investigate the effect of enhanced recovery after surgery are needed in the future.

Machine learning-based algorithms to predict severe psychological distress among cancer patients with spinal metastatic disease.

BACKGROUND CONTEXT: Metastatic spinal disease is an advanced stage of cancer patients and often suffer from terrible psychological health status; however, the ability to estimate the risk probability of this adverse outcome using current available data is very limited. PURPOSE: The goal of this study was to propose a precise model based on machine learning techniques to predict psychological status among cancer patients with spinal metastatic disease. STUDY DESIGN/SETTING: A prospective cohort study. PATIENT SAMPLE: A total of 1043 cancer patients with spinal metastatic disease were included. OUTCOME MEASURES: The main outcome was severe psychological distress. METHODS: The total of patients was randomly divided into a training dataset and a testing dataset on a ratio of 9:1. Patients' demographics, lifestyle choices, cancer-related features, clinical manifestations, and treatments were collected as potential model predictors in the study. Five machine learning algorithms, including XGBoosting machine, random forest, gradient boosting machine, support vector machine, and ensemble prediction model, as well as a logistic regression model were employed to train and optimize models in the training set, and their predictive performance was assessed in the testing set. RESULTS: Up to 21.48% of all patients who were recruited had severe psychological distress. Elderly patients (p<0.001), female (p =0.045), current smoking (p=0.002) or drinking (p=0.003), a lower level of education (p<0.001), a stronger spiritual desire (p<0.001), visceral metastasis (p=0.005), and a higher Eastern Cooperative Oncology Group (ECOG) score (p<0.001) were significantly associated with worse psychological health. With an area under the curve (AUC) of 0.865 (95% CI: 0.788-0.941) and an accuracy of up to 0.843, the gradient boosting machine algorithm performed best in the prediction of the outcome, followed by the XGBooting machine algorithm (AUC: 0.851, 95% CI: 0.768-0.934; Accuracy: 0.826) and ensemble prediction (AUC: 0.851, 95% CI: 0.770-0.932; Accuracy: 0.809) in the testing set. In contrast, the AUC of the logistic regression model was only 0.836 (95% CI: 0.756-0.916; Accuracy: 0.783). CONCLUSIONS: Machine learning models have greater predictive power and can offer useful tools to identify individuals with spinal metastatic disease who are experiencing severe psychological distress.

Prediction of postoperative health-related quality of life among patients with metastatic spinal cord compression secondary to lung cancer.

Background: Health-related quality of life (HRQoL) is a critical aspect of overall well-being for patients with lung cancer, particularly those with metastatic spinal cord compression (MSCC). However, there is currently a lack of universal evaluation of HRQoL in this specific patient population. The aim of this study was to develop a nomogram that can accurately predict HRQoL outcomes in patients with lung cancer-related MSCC. Methods: A total of 119 patients diagnosed with MSCC secondary to lung cancer were prospectively collected for analysis in the study. The least absolute shrinkage and selection operator (LASSO) regression analysis, along with 10-fold cross-validation, was employed to select the most significant variables for inclusion in the nomogram. Discriminative and calibration abilities were assessed using the concordance index (C-index), discrimination slope, calibration plots, and goodness-of-fit tests. Net reclassification index (NRI) and integrated discrimination improvement (IDI) analyses were conducted to compare the nomogram's performance with and without the consideration of comorbidities. Results: Four variables were selected to construct the final nomogram, including the Eastern Cooperative Oncology Group (ECOG) score, targeted therapy, anxiety scale, and number of comorbidities. The C-index was 0.87, with a discrimination slope of 0.47, indicating a favorable discriminative ability. Calibration plots and goodness-of-fit tests revealed a high level of consistency between the predicted and observed probabilities of poor HRQoL. The NRI (0.404, 95% CI: 0.074-0.734, p = 0.016) and the IDI (0.035, 95% CI: 0.004-0.066, p = 0.027) confirmed the superior performance of the nomogram with the consideration of comorbidities. Conclusions: This study develops a prediction nomogram that can assist clinicians in evaluating postoperative HRQoL in patients with lung cancer-related MSCC. This nomogram provides a valuable tool for risk stratification and personalized treatment planning in this specific patient population.

Quality of Life and Mental Health Status Among Cancer Patients With Metastatic Spinal Disease.

This study aimed to investigate the quality of life and mental health status and further to identify relevant risk factors among advanced cancer patients with spine metastases. This study prospectively included and analyzed 103 advanced cancer patients with spine metastases. Patient's basic information, lifestyles, comorbidities, tumor characteristics, therapeutic strategies, economic conditions, quality of life, anxiety, and depression were collected. Patient's quality of life was assessed using the Functional Assessment of Cancer Therapy-General Scale (FACT-G), and anxiety and depression were evaluated using the Hospital Anxiety and Depression Scale (HADS). Subgroup analysis was performed based on different age groups, and a multivariate analysis was performed to test the ability of 20 potential risk factors to predict quality of life, anxiety, and depression. The mean total FACT-G score was only 61.38 ± 21.26. Of all included patients, 52.43% had skeptical or identified anxiety and 53.40% suffered from skeptical or identified depression. Patients had an age of 60 or more and <70 years had the lowest FACT-G score (54.91 ± 19.22), highest HADS anxiety score (10.25 ± 4.22), and highest HADS depression score (10.13 ± 4.94). After adjusting all other potential risk factors, age was still significantly associated with quality of life (OR = 0.57, 95%CI: 0.38-0.86, p < 0.01) and depression (OR = 1.55, 95%CI: 1.00-2.42, p = 0.05) and almost significantly associated with anxiety (OR = 1.52, 95%CI: 0.94-2.43, p = 0.08). Besides, preference to eating vegetables, time since knowing cancer diagnosis, surgical treatment at primary cancer, hormone endocrine therapy, and economic burden due to cancer treatments were found to be significantly associated with the quality of life. A number of comorbidities and economic burden due to cancer treatments were significantly associated with anxiety. Advanced cancer patients with spine metastases suffer from poor quality of life and severe anxiety and depression, especially among patients with an age of 60 or more and <70 years. Early mental health care and effective measures should be conducted to advanced cancer patients with spine metastases, and more attention should be paid to take care of patients with an age of 60 or more and <70 years in terms of their quality of life and mental health status.

Preoperative embolization in the treatment of patients with metastatic epidural spinal cord compression: A retrospective analysis.

Purpose: The purpose of the study was to assess the effectiveness and safety of preoperative embolization in the treatment of patients with metastatic epidural spinal cord compression (MESCC). Methods: A retrospective analysis of 138 MESCC patients who underwent decompressive surgery and spine stabilization was performed in a large teaching hospital. Among all enrolled patients, 46 patients were treated with preoperative embolization (the embolization group), whereas 92 patients did not (the control group). Patient's baseline clinical characteristics, surgery-related characteristics, and postoperative neurological status, complications, and survival prognoses were collected and analyzed. Subgroup analysis was performed according to the degree of tumor vascularity between patients with and without preoperative embolization. Results: Patients with severe hypervascularity experienced more mean blood loss in the control group than in the embolization group, and this difference was statistically significant (P=0.02). The number of transfused packed red cells (PRC) showed a similar trend (P=0.01). However, for patients with mild and moderate hypervascularity, both blood loss and the number of PRC transfusion were comparable across the two groups. Regarding decompressive techniques, the embolization group (64.29%, 9/14) had a higher proportion of circumferential decompression in comparison to the control group (30.00%, 9/30) among patients with severe hypervascularity (P=0.03), whereas the rates were similar among patients with mild (P=0.45) and moderate (P=0.54) hypervascularity. In addition, no subgroup analysis revealed any statistically significant differences in operation time, postoperative functional recovery, postoperative complications, or survival outcome. Multivariate analysis showed that higher tumor vascularity (OR[odds ratio]=3.69, 95% CI [confident interval]: 1.30-10.43, P=0.01) and smaller extent of embolization (OR=4.16, 95% CI: 1.10-15.74, P=0.04) were significantly associated with more blood loss. Conclusions: Preoperative embolization is an effective and safe method in treating MESCC patients with severe hypervascular tumors in terms of intra-operative blood loss and surgical removal of metastatic tumors. Preoperative tumor vascularity and extent of embolization are independent risk factors for blood loss during surgery. This study implies that MESCC patients with severe hypervascular tumors should be advised to undergo preoperative embolization.

A machine learning-Based model to predict early death among bone metastatic breast cancer patients: A large cohort of 16,189 patients.

Purpose: This study aims to develop a prediction model to categorize the risk of early death among breast cancer patients with bone metastases using machine learning models. Methods: This study examined 16,189 bone metastatic breast cancer patients between 2010 and 2019 from a large oncological database in the United States. The patients were divided into two groups at random in a 90:10 ratio. The majority of patients (n = 14,582, 90%) were served as the training group to train and optimize prediction models, whereas patients in the validation group (n = 1,607, 10%) were utilized to validate the prediction models. Four models were introduced in the study: the logistic regression model, gradient boosting tree model, decision tree model, and random forest model. Results: Early death accounted for 17.4% of all included patients. Multivariate analysis demonstrated that older age; a separated, divorced, or widowed marital status; nonmetropolitan counties; brain metastasis; liver metastasis; lung metastasis; and histologic type of unspecified neoplasms were significantly associated with more early death, whereas a lower grade, a positive estrogen receptor (ER) status, cancer-directed surgery, radiation, and chemotherapy were significantly the protective factors. For the purpose of developing prediction models, the 12 variables were used. Among all the four models, the gradient boosting tree had the greatest AUC [0.829, 95% confident interval (CI): 0.802-0.856], and the random forest (0.828, 95% CI: 0.801-0.855) and logistic regression (0.819, 95% CI: 0.791-0.847) models came in second and third, respectively. The discrimination slopes for the three models were 0.258, 0.223, and 0.240, respectively, and the corresponding accuracy rates were 0.801, 0.770, and 0.762, respectively. The Brier score of gradient boosting tree was the lowest (0.109), followed by the random forest (0.111) and logistic regression (0.112) models. Risk stratification showed that patients in the high-risk group (46.31%) had a greater six-fold chance of early death than those in the low-risk group (7.50%). Conclusion: The gradient boosting tree model demonstrates promising performance with favorable discrimination and calibration in the study, and this model can stratify the risk probability of early death among bone metastatic breast cancer patients.