Prognostic Factors for Survival in Patients with Gastric Cancer using a Random Survival Forest

Comparative evaluation of outcomes amongst different radiosurgery management paradigms for patients with large brain metastasis

INTRODUCTION

This study compares four management paradigms for large brain metastasis (LMB): fractionated SRS (FSRS), staged SRS (SSRS), resection and postoperative-FSRS (postop-FSRS) or preoperative-SRS (preop-SRS).

METHODS

Patients with LBM (≥ 2 cm) between July 2017 and January 2022 at a single tertiary institution were evaluated. Primary endpoints were local failure (LF), radiation necrosis (RN), leptomeningeal disease (LMD), a composite of these variables, and distant intracranial failure (DIF). Gray's test compared cumulative incidence, treating death as a competing risk with a random survival forests (RSF) machine-learning model also used to evaluate the data.

RESULTS

183 patients were treated to 234 LBMs: 31.6% for postop-FSRS, 28.2% for SSRS, 20.1% for FSRS, and 20.1% for preop-SRS. The overall 1-year composite endpoint rates were comparable (21 vs 20%) between nonoperative and operative strategies, but 1-year RN rate was 8 vs 4% (p = 0.012), 1-year overall survival (OS) was 48 vs. 69% (p = 0.001), and 1-year LMD rate was 5 vs 10% (p = 0.052). There were differences in the 1-year RN rates (7% FSRS, 3% postop-FSRS, 5% preop-SRS, 10% SSRS, p = 0.037). With RSF analysis, the out-of-bag error rate for the composite endpoint was 47%, with identified top-risk factors including widespread extracranial disease, > 5 total lesions, and breast cancer histology.

CONCLUSION

This is the first study to conduct a head-to-head retrospective comparison of four SRS methods, addressing the lack of randomized data in LBM literature amongst treatment paradigms. Despite patient characteristic trends, no significant differences were found in LF, composite endpoint, and DIF rates between non-operative and operative approaches.

Mortality prediction and influencing factors for intensive care unit patients with acute tubular necrosis: random survival forest and cox regression analysis

Background: Patients with acute tubular necrosis (ATN) not only have severe renal failure, but also have many comorbidities, which can be life-threatening and require timely treatment. Identifying the influencing factors of ATN and taking appropriate interventions can effectively shorten the duration of the disease to reduce mortality and improve patient prognosis. Methods: Mortality prediction models were constructed by using the random survival forest (RSF) algorithm and the Cox regression. Next, the performance of both models was assessed by the out-of-bag (OOB) error rate, the integrated brier score, the prediction error curve, and area under the curve (AUC) at 30, 60 and 90 days. Finally, the optimal prediction model was selected and the decision curve analysis and nomogram were established. Results: RSF model was constructed under the optimal combination of parameters (mtry = 10, nodesize = 88). Vasopressors, international normalized ratio (INR)_min, chloride_max, base excess_min, bicarbonate_max, anion gap_min, and metastatic solid tumor were identified as risk factors that had strong influence on mortality in ATN patients. Uni-variate and multivariate regression analyses were used to establish the Cox regression model. Nor-epinephrine, vasopressors, INR_min, severe liver disease, and metastatic solid tumor were identified as important risk factors. The discrimination and calibration ability of both predictive models were demonstrated by the OOB error rate and the integrated brier score. However, the prediction error curve of Cox regression model was consistently lower than that of RSF model, indicating that Cox regression model was more stable and reliable. Then, Cox regression model was also more accurate in predicting mortality of ATN patients based on the AUC at different time points (30, 60 and 90 days). The analysis of decision curve analysis shows that the net benefit range of Cox regression model at different time points is large, indicating that the model has good clinical effectiveness. Finally, a nomogram predicting the risk of death was created based on Cox model. Conclusion: The Cox regression model is superior to the RSF algorithm model in predicting mortality of patients with ATN. Moreover, the model has certain clinical utility, which can provide clinicians with some reference basis in the treatment of ATN and contribute to improve patient prognosis.

Comparison of risk prediction models for the progression of pelvic inflammatory disease patients to sepsis: Cox regression model and machine learning model

Introduction

The present study presents the development and validation of a clinical prediction model using random survival forest (RSF) and stepwise Cox regression, aiming to predict the probability of pelvic inflammatory disease (PID) progressing to sepsis.

Methods

A retrospective cohort study was conducted, gathering clinical data of patients diagnosed with PID between 2008 and 2019 from the Medical Information Mart for Intensive Care (MIMIC)-IV database. Patients who met the Sepsis 3.0 diagnostic criteria were selected, with sepsis as the outcome. Univariate Cox regression and stepwise Cox regression were used to screen variables for constructing a nomogram. Moreover, an RSF model was created using machine learning algorithms. To verify the model's performance, a calibration curve, decision curve analysis (DCA), and receiver operating characteristic (ROC) curve were utilized. Furthermore, the capabilities of the two models for estimating the incidence of sepsis in PID patients within 3 and 7 days were compared.

Results

A total of 1064 PID patients were included, of whom 54 had progressed to sepsis. The established nomogram highlighted dialysis, reduced platelet (PLT) counts, history of pneumonia, medication of glucocorticoids, and increased leukocyte counts as significant predictive factors. The areas under the curve (AUCs) of the nomogram for prediction of PID progression to sepsis at 3-day and 7-day (3-/7-day) in the training set and the validation set were 0.886/0.863 and 0.824/0.726, respectively, and the C-index of the model was 0.8905. The RSF displayed excellent performance, with AUCs of 0.939/0.919 and 0.712/0.571 for 3-/7-day risk prediction in the training set and validation set, respectively.

Conclusion

The nomogram accurately predicted the incidence of sepsis in PID patients, and relevant risk factors were identified. While the RSF model outperformed the Cox regression models in predicting sepsis incidence, its performance exhibited some instability. On the other hand, the Cox regression-based nomogram displayed stable performance and improved interpretability, thereby supporting clinical decision-making in PID treatment.

Deep learning model for predicting the survival of patients with primary gastrointestinal lymphoma based on the SEER database and a multicentre external validation cohort

PURPOSE

Due to the rarity of primary gastrointestinal lymphoma (PGIL), the prognostic factors and optimal management of PGIL have not been clearly defined. We aimed to establish prognostic models using a deep learning algorithm for survival prediction.

METHODS

We collected 11,168 PGIL patients from the Surveillance, Epidemiology, and End Results (SEER) database to form the training and test cohorts. At the same time, we collected 82 PGIL patients from three medical centres to form the external validation cohort. We constructed a Cox proportional hazards (CoxPH) model, random survival forest (RSF) model, and neural multitask logistic regression (DeepSurv) model to predict PGIL patients' overall survival (OS).

RESULTS

The 1-, 3-, 5-, and 10-year OS rates of PGIL patients in the SEER database were 77.1%, 69.4%, 63.7%, and 50.3%, respectively. The RSF model based on all variables showed that the top three most important variables for predicting OS were age, histological type, and chemotherapy. The independent risk factors for PGIL patient prognosis included sex, age, race, primary site, Ann Arbor stage, histological type, symptom, radiotherapy, and chemotherapy, according to the Lasso regression analysis. Using these factors, we built the CoxPH and DeepSurv models. The DeepSurv model's C-index values were 0.760 in the training cohort, 0.742 in the test cohort, and 0.707 in the external validation cohort, which demonstrated that the DeepSurv model performed better compared to the RSF model (0.728) and the CoxPH model (0.724). The DeepSurv model accurately predicted 1-, 3-, 5- and 10-year OS. Both calibration curves and decision curve analysis curves demonstrated the superior performance of the DeepSurv model. We developed the DeepSurv model as an online web calculator for survival prediction, which can be accessed at http://124.222.228.112:8501/ .

CONCLUSIONS

This DeepSurv model with external validation is superior to previous studies in predicting short-term and long-term survival and can help us make better-individualized decisions for PGIL patients.

Bioinformatics analysis reveals immune prognostic markers for overall survival of colorectal cancer patients: a novel machine learning survival predictive system

Objectives

Immune microenvironment was closely related to the occurrence and progression of colorectal cancer (CRC). The objective of the current research was to develop and verify a Machine learning survival predictive system for CRC based on immune gene expression data and machine learning algorithms.

Methods

The current study performed differentially expressed analyses between normal tissues and tumor tissues. Univariate Cox regression was used to screen prognostic markers for CRC. Prognostic immune genes and transcription factors were used to construct an immune-related regulatory network. Three machine learning algorithms were used to create an Machine learning survival predictive system for CRC. Concordance indexes, calibration curves, and Brier scores were used to evaluate the performance of prognostic model.

Results

Twenty immune genes (BCL2L12, FKBP10, XKRX, WFS1, TESC, CCR7, SPACA3, LY6G6C, L1CAM, OSM, EXTL1, LY6D, FCRL5, MYEOV, FOXD1, REG3G, HAPLN1, MAOB, TNFSF11, and AMIGO3) were recognized as independent risk factors for CRC. A prognostic nomogram was developed based on the previous immune genes. Concordance indexes were 0.852, 0.778, and 0.818 for 1-, 3- and 5-year survival. This prognostic model could discriminate high risk patients with poor prognosis from low risk patients with favorable prognosis.

Conclusions

The current study identified twenty prognostic immune genes for CRC patients and constructed an immune-related regulatory network. Based on three machine learning algorithms, the current research provided three individual mortality predictive curves. The Machine learning survival predictive system was available at: https://zhangzhiqiao8.shinyapps.io/Artificial_Intelligence_Survival_Prediction_for_CRC_B1005_1/, which was valuable for individualized treatment decision before surgery.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04657-3.

Accuracy of Endoscopic Ultrasonography for Gastric Cancer Staging

Gastric cancer remains a health problem, with treatment indications varying with the TNM stage. We aimed in this study to highlight the role of EUS in GC patients and also to calculate the accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of EUS for T and N staging in our group of patients with this disease. In this study, we included 41 GC patients, and individual values for every T stage accuracy, sensitivity, specificity, PPV, NPV, correct staging, understaging, and overstaging were calculated. EUS overall accuracy for T staging was 58.53%, with the highest sensitivity reached for the T4 stage, 95.83%. For N+vs. N-staging, EUS accuracy was 68.29%, with a sensitivity of 75% and a specificity of 44.44%. The positive and negative predicted values for the presence or absence of nodal disease were 82.75%, respectively 33.33%. In conclusion, this study confirmed the importance of EUS for the assessment of GC T and N stage and highlighted the role of this tool in the detection of liver micrometastasis unrevealed by other imaging techniques like abdominal ultrasound or MSCT.

Identifying Factors Affecting Cancer-related Death in Patients with Adenocarcinoma Gastric Cancer; An Analysis in the Presence of Competing Risks

Background: Adenocarcinoma is the most common type of gastric cancer that has shorter survival than other types of gastric cancer. The death of patients with this type of cancer may be due to the progression of cancer or other related causes. Objectives: The aim of this study is to determine the factors affecting death due to the cancer progression in gastric cancer patients with the diagnosis of adenocarcinoma, using competing risk models. Methods: This retrospective cohort study was performed on 306 gastric cancer patients diagnosed with adenocarcinoma referring to Imam Khomeini clinic in Hamadan from 2002 to 2017. Death due to the cancer progression was considered an interest event and death due to without progression as a competing event. To determine the effect of covariates on hazard, the cause-specific and subdistribution hazard regression models were used. Data analysis was performed, using R3.6.1 software and cmprsk and survival packages. Results: The mean (SD) age of patients was 62.3 (12.5) years and 74.3% were male. The effect of the stage, the number of involved lymphomas, and the type of treatment were significant on the hazard of death due to the disease progression in both cause-specific and subdistribution hazard models. Conclusions: The results showed that most deaths occur in the first 3 years of follow-up. The higher stage and higher number of lymph nodes have increased the hazard of death but supplementary treatment significantly decreased the hazard of death due to cancer progression in adenocarcinoma gastric cancer patients in both competing risk models.

Potential Prognostic Immune Biomarkers of Overall Survival in Ovarian Cancer Through Comprehensive Bioinformatics Analysis: A Novel Artificial Intelligence Survival Prediction System

Background: The tumour immune microenvironment plays an important role in the biological mechanisms of tumorigenesis and progression. Artificial intelligence medicine studies based on big data and advanced algorithms are helpful for improving the accuracy of prediction models of tumour prognosis. The current research aims to explore potential prognostic immune biomarkers and develop a predictive model for the overall survival of ovarian cancer (OC) based on artificial intelligence algorithms.

Methods: Differential expression analyses were performed between normal tissues and tumour tissues. Potential prognostic biomarkers were identified using univariate Cox regression. An immune regulatory network was constructed of prognostic immune genes and their highly related transcription factors. Multivariate Cox regression was used to identify potential independent prognostic immune factors and develop a prognostic model for ovarian cancer patients. Three artificial intelligence algorithms, random survival forest, multitask logistic regression, and Cox survival regression, were used to develop a novel artificial intelligence survival prediction system.

Results: The current study identified 1,307 differentially expressed genes and 337 differentially expressed immune genes between tumour samples and normal samples. Further univariate Cox regression identified 84 prognostic immune gene biomarkers for ovarian cancer patients in the model dataset (GSE32062 dataset and GSE53963 dataset). An immune regulatory network was constructed involving 63 immune genes and 5 transcription factors. Fourteen immune genes (PSMB9, FOXJ1, IFT57, MAL, ANXA4, CTSH, SCRN1, MIF, LTBR, CTSD, KIFAP3, PSMB8, HSPA5, and LTN1) were recognised as independent risk factors by multivariate Cox analyses. Kaplan-Meier survival curves showed that these 14 prognostic immune genes were closely related to the prognosis of ovarian cancer patients. A prognostic nomogram was developed by using these 14 prognostic immune genes. The concordance indexes were 0.760, 0.733, and 0.765 for 1-, 3-, and 5-year overall survival, respectively. This prognostic model could differentiate high-risk patients with poor overall survival from low-risk patients. According to three artificial intelligence algorithms, the current study developed an artificial intelligence survival predictive system that could provide three individual mortality risk curves for ovarian cancer.

Conclusion: In conclusion, the current study identified 1,307 differentially expressed genes and 337 differentially expressed immune genes in ovarian cancer patients. Multivariate Cox analyses identified fourteen prognostic immune biomarkers for ovarian cancer. The current study constructed an immune regulatory network involving 63 immune genes and 5 transcription factors, revealing potential regulatory associations among immune genes and transcription factors. The current study developed a prognostic model to predict the prognosis of ovarian cancer patients. The current study further developed two artificial intelligence predictive tools for ovarian cancer, which are available at https://zhangzhiqiao8.shinyapps.io/Smart_Cancer_Survival_Predictive_System_17_OC_F1001/ and https://zhangzhiqiao8.shinyapps.io/Gene_Survival_Subgroup_Analysis_17_OC_F1001/. An artificial intelligence survival predictive system could help improve individualised treatment decision-making.

Identification of Lifestyle Behaviors Associated with Recurrence and Survival in Colorectal Cancer Patients Using Random Survival Forests

Current lifestyle recommendations for cancer survivors are the same as those for the general public to decrease their risk of cancer. However, it is unclear which lifestyle behaviors are most important for prognosis. We aimed to identify which lifestyle behaviors were most important regarding colorectal cancer (CRC) recurrence and all-cause mortality with a data-driven method. The study consisted of 1180 newly diagnosed stage I-III CRC patients from a prospective cohort study. Lifestyle behaviors included in the current recommendations, as well as additional lifestyle behaviors related to diet, physical activity, adiposity, alcohol use, and smoking were assessed six months after diagnosis. These behaviors were simultaneously analyzed as potential predictors of recurrence or all-cause mortality with Random Survival Forests (RSFs). We observed 148 recurrences during 2.6-year median follow-up and 152 deaths during 4.8-year median follow-up. Higher intakes of sugary drinks were associated with increased recurrence risk. For all-cause mortality, fruit and vegetable, liquid fat and oil, and animal protein intake were identified as the most important lifestyle behaviors. These behaviors showed non-linear associations with all-cause mortality. Our exploratory RSF findings give new ideas on potential associations between certain lifestyle behaviors and CRC prognosis that still need to be confirmed in other cohorts of CRC survivors.

Inhibition of Topoisomerase IIα and Induction of Apoptosis in Gastric Cancer Cells by 19-Triisopropyl Andrographolide

Gastric cancer is the most common cancer in Eastern Asia. Increasing chemoresistance and general systemic toxicities have complicated the current chemotherapy leading to an urgent need of more effective agents. The present study reported a potent DNA topoisomerase IIα inhibitory activity of an andrographolide analogue (19-triisopropyl andrographolide, analogue-6) in gastric cancer cells; MKN-45, and AGS cells. The analogue was potently cytotoxic to both gastric cancer cell lines with the half maximal inhibitory concentration (IC50 values) of 6.3±0.7 μM, and 1.7±0.05 μM at 48 h for MKN-45, and AGS cells, respectively. It was more potent than the parent andrographolide and the clinically used, etoposide with the IC50 values of >50 μM in MKN-45 and 11.3±2.9 μM in AGS cells for andrographolide and 28.5±4.4 μM in MKN-45 and 4.08±0.5 μM in AGS cells for etoposide. Analogue-6 at 2 μM significantly inhibited DNA topoisomerase IIα enzyme in AGS cells, induced DNA damage, activated cleaved PARP-1, and Caspase3 leading to late cellular apoptosis. Interestingly, the expression of tumor suppressor p53 was not activated. These results show the importance of 19-triisopropyl-andrographolide in its emerging selectivity to primary target on topoisomerase IIα enzyme, inducing DNA damage and apoptosis by p53- independent mechanism. Thereby, the results provide insights of the potential of 19-triisopropyl andrographolide as an anticancer agent for gastric cancer. The chemical transformation of andrographolide is a promising strategy in drug discovery of a novel class of anticancer drugs from bioactive natural products.