Random Survival Forests

A prediction model based on random survival forest analysis of the overall survival of elderly female papillary thyroid carcinoma patients: a SEER-based study

OBJECTIVE

Papillary thyroid carcinoma (PTC) is a common malignancy whose incidence is three times greater in females than in males. The prognosis of ageing patients is poor. This research was designed to construct models to predict the overall survival of elderly female patients with PTC.

METHODS

We developed prediction models based on the random survival forest (RSF) algorithm and traditional Cox regression. The data of 4539 patients were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Twelve variables were analysed to establish the models. The C-index and the Brier score were selected to evaluate the discriminatory ability of the models. Time-dependent receiver operating characteristic (ROC) curves were also drawn to evaluate the accuracy of the models. The clinical benefits of the two models were compared on the basis of the DCA curve. In addition, the Shapley Additive Explanations (SHAP) plot was used to visualize the contribution of the variables in the RSF model.

RESULTS

The C-index of the RSF model was 0.811, which was greater than that of the Cox model (0.781). According to the Brier score and the area under the ROC curve (AUC), the RSF model performed better than the Cox model. On the basis of the DCA curve, the RSF model demonstrated fair clinical benefit. The SHAP plot showed that age was the most important variable contributing to the outcome of PTC in elderly female patients.

CONCLUSIONS

The RSF model we developed performed better than the Cox model and might be valuable for clinical practice.

Development and validation of novel interpretable survival prediction models based on drug exposures for severe heart failure during vulnerable period

BACKGROUND

Severe heart failure (HF) has a higher mortality during vulnerable period while targeted predictive tools, especially based on drug exposures, to accurately assess its prognoses remain largely unexplored. Therefore, this study aimed to utilize drug information as the main predictor to develop and validate survival models for severe HF patients during this period.

METHODS

We extracted severe HF patients from the MIMIC-IV database (as training and internal validation cohorts) as well as from the MIMIC-III database and local hospital (as external validation cohorts). Three algorithms, including Cox proportional hazards model (CoxPH), random survival forest (RSF), and deep learning survival prediction (DeepSurv), were applied to incorporate the parameters (partial hospitalization information and exposure durations of drugs) for constructing survival prediction models. The model performance was assessed mainly using area under the receiver operator characteristic curve (AUC), brier score (BS), and decision curve analysis (DCA). The model interpretability was determined by the permutation importance and Shapley additive explanations values.

RESULTS

A total of 11,590 patients were included in this study. Among the 3 models, the CoxPH model ultimately included 10 variables, while RSF and DeepSurv models incorporated 24 variables, respectively. All of the 3 models achieved respectable performance metrics while the DeepSurv model exhibited the highest AUC values and relatively lower BS among these models. The DCA also verified that the DeepSurv model had the best clinical practicality.

CONCLUSIONS

The survival prediction tools established in this study can be applied to severe HF patients during vulnerable period by mainly inputting drug treatment duration, thus contributing to optimal clinical decisions prospectively.

Nomogram Predicts Prognostic Factors for Head and Neck Cutaneous Melanoma: A Population-Based Analysis

BACKGROUND

The head and neck cutaneous melanoma (HNCM) accounts for 20% of newly diagnosed melanoma. Research on prognostic models for their survival yet remains largely unexplored. This study employed a nomogram approach to develop and validate a predictive model for both overall survival (OS) and disease-specific survival (DSS) in patients with HNCM.

METHODS

This study analyzed the HNCM patients diagnosed between 2004 and 2014 from Surveillance, Epidemiology, and End Results database. To identify independent prognostic factors for HNCM, we integrated results from univariate Cox regression analysis, random survival forests, and LASSO regression with cross-validation. A nomogram was designed and validated based on the identified characteristics to predict the 3-, 5-, and 8-year OS and DSS of patients with HNCM.

RESULTS

Age, Stage, Ulceration, Thickness, Chemotherapy, lymph node metastasis, and Radiation were identified as independent prognostic factors. The nomogram achieved a satisfactory performance with C-indices of 0.824(DSS) and 0.757(OS) in the training cohort and 0.827(DSS) and 0.749(OS) in the validation cohort, respectively. The area under the curves for the OS at 3, 5, and 8 years were 0.789, 0.788, and 0.794 for the training cohort, and 0.778, 0.776, and 0.795 for the validation cohort, respectively. For DSS, the area under the curves at 3, 5, and 8 years were 0.859, 0.842, and 0.828 in the training cohort, and 0.864, 0.844, and 0.834 in the validation cohort, respectively. The calibration curve showed that there was a strong correlation between the observed outcomes and the predicted survival probability.

CONCLUSIONS

This study established and validated predictive nomograms for HNCM patients with robust predictive performance.

Deep learning models for predicting the survival of patients with medulloblastoma based on a surveillance, epidemiology, and end results analysis

Medulloblastoma is a malignant neuroepithelial tumor of the central nervous system. Accurate prediction of prognosis is essential for therapeutic decisions in medulloblastoma patients. We analyzed data from 2,322 medulloblastoma patients using the SEER database and randomly divided the dataset into training and testing datasets in a 7:3 ratio. We chose three models to build, one based on neural networks (DeepSurv), one based on ensemble learning that Random Survival Forest (RSF), and a typical Cox Proportional-hazards (CoxPH) model. The DeepSurv model outperformed the RSF and classic CoxPH models with C-indexes of 0.751 and 0.763 for the training and test datasets. Additionally, the DeepSurv model showed better accuracy in predicting 1-, 3-, and 5-year survival rates (AUC: 0.767-0.793). Therefore, our prediction model based on deep learning algorithms can more accurately predict the survival rate and survival period of medulloblastoma compared to other models.

Explainable machine learning predicts survival of retroperitoneal liposarcoma: A study based on the SEER database and external validation in China

OBJECTIVE

We have developed explainable machine learning models to predict the overall survival (OS) of retroperitoneal liposarcoma (RLPS) patients. This approach aims to enhance the explainability and transparency of our modeling results.

METHODS

We collected clinicopathological information of RLPS patients from The Surveillance, Epidemiology, and End Results (SEER) database and allocated them into training and validation sets with a 7:3 ratio. Simultaneously, we obtained an external validation cohort from The First Affiliated Hospital of Naval Medical University (Shanghai, China). We performed LASSO regression and multivariate Cox proportional hazards analysis to identify relevant risk factors, which were then combined to develop six machine learning (ML) models: Cox proportional hazards model (Coxph), random survival forest (RSF), ranger, gradient boosting with component-wise linear models (GBM), decision trees, and boosting trees. The predictive performance of these ML models was evaluated using the concordance index (C-index), the integrated cumulative/dynamic area under the curve (AUC), and the integrated Brier score, as well as the Cox-Snell residual plot. We also used time-dependent variable importance, analysis of partial dependence survival plots, and the generation of aggregated survival SHapley Additive exPlanations (SurvSHAP) plots to provide a global explanation of the optimal model. Additionally, SurvSHAP (t) and survival local interpretable model-agnostic explanations (SurvLIME) plots were used to provide a local explanation of the optimal model.

RESULTS

The final ML models are consisted of six factors: patient's age, gender, marital status, surgical history, as well as tumor's histopathological classification, histological grade, and SEER stage. Our prognostic model exhibits significant discriminative ability, particularly with the ranger model performing optimally. In the training set, validation set, and external validation set, the AUC for 1, 3, and 5 year OS are all above 0.83, and the integrated Brier scores are consistently below 0.15. The explainability analysis of the ranger model also indicates that histological grade, histopathological classification, and age are the most influential factors in predicting OS.

CONCLUSIONS

The ranger ML prognostic model exhibits optimal performance and can be utilized to predict the OS of RLPS patients, offering valuable and crucial references for clinical physicians to make informed decisions in advance.

Mitochondrial RNA modification-based signature to predict prognosis of lower grade glioma: a multi-omics exploration and verification study

Mitochondrial RNA modification (MRM) plays a crucial role in regulating the expression of key mitochondrial genes and promoting tumor metastasis. Despite its significance, comprehensive studies on MRM in lower grade gliomas (LGGs) remain unknown. Single-cell RNA-seq data (GSE89567) was used to evaluate the distribution functional status, and correlation of MRM-related genes in different cell types of LGG microenvironment. We developed an MRM scoring system by selecting potential MRM-related genes using LASSO regression analysis and the Random Survival Forest algorithm, based on multiple bulk RNA-seq datasets from TCGA, CGGA, GSE16011, and E-MTAB-3892. Analysis was performed on prognostic and immunological features, signaling pathways, metabolism, somatic mutations and copy number variations (CNVs), treatment responses, and forecasting of potential small-molecule agents. A total of 35 MRM-related genes were selected from the literature. Differential expression analysis of 1120 normal brain tissues and 529 LGGs revealed that 22 and 10 genes were upregulated and downregulated, respectively. Most genes were associated with prognosis of LGG. METLL8, METLL2A, TRMT112, and METTL2B were extensively expressed in all cell types and different cell cycle of each cell type. Almost all cell types had clusters related to mitochondrial RNA processing, ribosome biogenesis, or oxidative phosphorylation. Cell-cell communication and Pearson correlation analyses indicated that MRM may promoting the development of microenvironment beneficial to malignant progression via modulating NCMA signaling pathway and ICP expression. A total of 11 and 9 MRM-related genes were observed by LASSO and the RSF algorithm, respectively, and finally 6 MRM-related genes were used to establish MRM scoring system (TRMT2B, TRMT11, METTL6, METTL8, TRMT6, and TRUB2). The six MRM-related genes were then validated by qPCR in glioma and normal tissues. MRM score can predict the malignant clinical characteristics, abundance of immune infiltration, gene variation, clinical outcome, the enrichment of signaling pathways and metabolism. In vitro experiments demonstrated that silencing METTL8 significantly curbs glioma cell proliferation and enhances apoptosis. Patients with a high MRM score showed a better response to immunotherapies and small-molecule agents such as arachidonyl trifluoromethyl ketone, MS.275, AH.6809, tacrolimus, and TTNPB. These novel insights into the biological impacts of MRM within the glioma microenvironment underscore its potential as a target for developing precise therapies, including immunotherapeutic 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.

Artificial intelligence-based prognostic model accurately predicts the survival of patients with diffuse large B-cell lymphomas: analysis of a large cohort in China

BACKGROUND

Diffuse large B-cell lymphomas (DLBCLs) display high molecular heterogeneity, but the International Prognostic Index (IPI) considers only clinical indicators and has not been updated to include molecular data. Therefore, we developed a widely applicable novel scoring system with molecular indicators screened by artificial intelligence (AI) that achieves accurate prognostic stratification and promotes individualized treatments.

METHODS

We retrospectively enrolled a cohort of 401 patients with DLBCL from our hospital, covering the period from January 2015 to January 2019. We included 22 variables in our analysis and assigned them weights using the random survival forest method to establish a new predictive model combining bidirectional long-short term memory (Bi-LSTM) and logistic hazard techniques. We compared the predictive performance of our "molecular-contained prognostic model" (McPM) and the IPI. In addition, we developed a simplified version of the McPM (sMcPM) to enhance its practical applicability in clinical settings. We also demonstrated the improved risk stratification capabilities of the sMcPM.

RESULTS

Our McPM showed superior predictive accuracy, as indicated by its high C-index and low integrated Brier score (IBS), for both overall survival (OS) and progression-free survival (PFS). The overall performance of the McPM was also better than that of the IPI based on receiver operating characteristic (ROC) curve fitting. We selected five key indicators, including extranodal involvement sites, lactate dehydrogenase (LDH), MYC gene status, absolute monocyte count (AMC), and platelet count (PLT) to establish the sMcPM, which is more suitable for clinical applications. The sMcPM showed similar OS results (P < 0.0001 for both) to the IPI and significantly better PFS stratification results (P < 0.0001 for sMcPM vs. P = 0.44 for IPI).

CONCLUSIONS

Our new McPM, including both clinical and molecular variables, showed superior overall stratification performance to the IPI, rendering it more suitable for the molecular era. Moreover, our sMcPM may become a widely used and effective stratification tool to guide individual precision treatments and drive new drug development.

Advancing precision medicine in immunoglobulin light-chain amyloidosis: a novel prognostic model incorporating multi-organ indicators

To develop a more accurate prognostic model that incorporates indicators of multi-organ involvement for immunoglobulin light-chain (AL) Amyloidosis patients. Biopsy-proven AL amyloidosis patients between January 1, 2012, and February 28, 2023, were enrolled and randomly divided into a training set and a test set at a ratio of 7:3. Prognostic indicators that comprehensively cover cardiac, renal, and hepatic involvement were identified in the training set by random survival forest (RSF). Then, RSF and Cox models were established. The Concordance index (C-index) and integrated brier scores (IBS) were applied to evaluate the models' performance in the test set. Besides, the net reclassification index (NRI) and integrated discrimination improvement (IDI) were calculated. A total of 173 eligible patients were included. After a median follow-up of 25.9 (9.2, 50.3) months, 48 (27.7%) patients died. Creatine kinase-MB, estimated glomerular filtration rate ≤ 50 mL/min/1.73 m2, interventricular septum ≥ 15 mm, ejection fraction, alanine aminotransferase and Live involved were selected to develop prediction models. The RSF model based on the above indicators achieved C-index and IBS values of 0.834 (95% CI 0.725-0.915) and 0.151 (95% CI 0.1402-0.181), respectively. At last, the NRI and IDI of the RSF model were 0.301 (95% CI 0.048-0.546, P = 0.012) and 0.157 (95% CI 0.041-0.269, P < 0.001) at 5-year by comparing the RSF model with the Cox model which is based on the Mayo 2012 staging system. The RSF model that incorporates indicators of multi-organ involvement had a great performance, which may be helpful for physicians' decision-making and more accurate overall survival prediction.

Deep learning model for predicting postoperative survival of patients with gastric cancer

Background

Prognostic prediction for surgical treatment of gastric cancer remains valuable in clinical practice. This study aimed to develop survival models for postoperative gastric cancer patients.

Methods

Eleven thousand seventy-five patients from the Surveillance, Epidemiology, and End Results (SEER) database were included, and 122 patients from the Chinese database were used for external validation. The training cohort was created to create three separate models, including Cox regression, RSF, and DeepSurv, using data from the SEER database split into training and test cohorts with a 7:3 ratio. Test cohort was used to evaluate model performance using c-index, Brier scores, calibration, and the area under the curve (AUC). The new risk stratification based on the best model will be compared with the AJCC stage on the test and Chinese cohorts using decision curve analysis (DCA), the net reclassification index (NRI), and integrated discrimination improvement (IDI).

Results

It was discovered that the DeepSurv model predicted postoperative gastric cancer patients' overall survival (OS) with a c-index of 0.787; the area under the curve reached 0.781, 0.798, 0.868 at 1-, 3- and 5- years, respectively; the Brier score was below 0.25 at different time points; showing an advantage over the Cox and RSF models. The results are also validated in the China cohort. The calibration plots demonstrated good agreement between the DeepSurv model's forecast and actual results. The NRI values (test cohort: 0.399, 0.288, 0.267 for 1-, 3- and 5-year OS prediction; China cohort:0.399, 0.288 for 1- and 3-year OS prediction) and IDI (test cohort: 0.188, 0.169, 0.157 for 1-, 3- and 5-year OS prediction; China cohort: 0.189, 0.169 for 1- and 3-year OS prediction) indicated that the risk score stratification performed significantly better than the AJCC staging alone (P < 0.05). DCA showed that the risk score stratification was clinically useful and had better discriminative ability than the AJCC staging. Finally, an interactive native web-based prediction tool was constructed for the survival prediction of patients with postoperative gastric cancer.

Conclusion

In this study, a high-performance prediction model for the postoperative prognosis of gastric cancer was developed using DeepSurv, which offers essential benefits for risk stratification and prognosis prediction for each patient.

The development of a prediction model based on deep learning for prognosis prediction of gastrointestinal stromal tumor: a SEER-based study

Accurately predicting the prognosis of Gastrointestinal stromal tumor (GIST) patients is an important task. The goal of this study was to create and assess models for GIST patients' survival patients using the Surveillance, Epidemiology, and End Results Program (SEER) database based on the three different deep learning models. Four thousand five hundred thirty-eight patients were enrolled in this study and divided into training and test cohorts with a 7:3 ratio; the training cohort was used to develop three different models, including Cox regression, RSF, and DeepSurv model. Test cohort was used to evaluate model performance using c-index, Brier scores, calibration, and the area under the curve (AUC). The net benefits at risk score stratification of GIST patients based on the optimal model was compared with the traditional AJCC staging system using decision curve analysis (DCA). The clinical usefulness of risk score stratification compared to AJCC tumor staging was further assessed using the Net Reclassification Index (NRI) and Integrated Discrimination Improvement (IDI). The DeepSurv model predicted cancer-specific survival (CSS) in GIST patients showed a higher c-index (0.825), lower Brier scores (0.142), and greater AUC of receiver operating characteristic (ROC) analysis (1-year ROC:0.898; 3-year:0.853, and 5-year ROC: 0.856). The calibration plots demonstrated good agreement between the DeepSurv model's forecast and actual results. The NRI values ( training cohort: 0.425 for 1-year, 0.329 for 3-year and 0.264 for 5-year CSS prediction; test cohort:0.552 for 1-year,0.309 for 3-year and 0.255 for 5-year CSS prediction) and IDI (training cohort: 0.130 for 1-year,0.141 for 5-year and 0.155 for 10-year CSS prediction; test cohort: 0.154 for 1-year,0.159 for 3-year and 0.159 for 5-year CSS prediction) indicated that the risk score stratification performed significantly better than the AJCC staging alone (P < 0.001). DCA demonstrated the risk score stratification as more clinically beneficial and discriminatory than AJCC staging. Finally, an interactive native web-based prediction tool was constructed for the survival prediction of GIST patients. This study established a high-performance prediction model for projecting GIST patients based on deep learning, which has advantages in predicting each person's prognosis and risk stratification.

Development and validation of a prospective study to predict the risk of readmission within 365 days of respiratory failure: based on a random survival forest algorithm combined with COX regression modeling

BACKGROUND

There is a need to develop and validate a widely applicable nomogram for predicting readmission of respiratory failure patients within 365 days.

METHODS

We recruited patients with respiratory failure at the First People's Hospital of Yancheng and the People's Hospital of Jiangsu. We used the least absolute shrinkage and selection operator regression to select significant features for multivariate Cox proportional hazard analysis. The Random Survival Forest algorithm was employed to construct a model for the variables that obtained a coefficient of 0 following LASSO regression, and subsequently determine the prediction score. Independent risk factors and the score were used to develop a multivariate COX regression for creating the line graph. We used the Harrell concordance index to quantify the predictive accuracy and the receiver operating characteristic curve to evaluate model performance. Additionally, we used decision curve analysiso assess clinical usefulness.

RESULTS

The LASSO regression and multivariate Cox regression were used to screen hemoglobin, diabetes and pneumonia as risk variables combined with Score to develop a column chart model. The C index is 0.927 in the development queue, 0.924 in the internal validation queue, and 0.922 in the external validation queue. At the same time, the predictive model also showed excellent calibration and higher clinical value.

CONCLUSIONS

A nomogram predicting readmission of patients with respiratory failure within 365 days based on three independent risk factors and a jointly developed random survival forest algorithm has been developed and validated. This improves the accuracy of predicting patient readmission and provides practical information for individualized treatment decisions.

Identification of multiomics map and key biomarkers in uveal melanoma with chromosome 3 loss

Purpose

Chromosome 3 loss is an independent risk factor for uveal melanoma (UM), but its exact molecular mechanisms remain unclear. This study was designed to investigate the relationship between chromosome 3 loss and molecular alterations at multiple levels to construct a prognostic model.

Methods

Forty-four UM cases with chromosome 3 loss (chr3 del group) and 36 UM cases without copy number variation on chromosome 3 (chr3 wt group) were collected from the Cancer Genome Atlas (TCGA). The TCGA dataset was subjected to a univariate Cox regression analysis to identify different expressed genes, and a subsequent random forest algorithm analysis revealed significant changes in different expressed genes, which were used to develop key biomarkers for UM. Following that, the immune cell infiltration analysis and drug sensitivity analyses were carried out. The UM cell line was then utilized to investigate the potential functions of the key biomarker via cell apoptosis, proliferation, cycle assays, WB, and RT-qPCR.

Results

By analyzing the 80 cases data in TCGA, the authors unveiled molecular changes relevant to loss of chromosome 3 in UM as well as their poor survival. In addition, machine learning analysis identified three hub genes (GRIN2A, ACAN, and MMP9) as potential therapeutic targets. The differentially enriched pathways between the two groups were mainly about immune-system activity, and hub genes expression was also highly correlated with immune infiltration levels.

Conclusion

Chromosome 3 loss has considerable clinical significance for UM, and GRIN2A may be useful in diagnosing, treating, and prognosticating the condition.

Bioinformatics and Experimental Validation for Identifying Biomarkers Associated with AMG510 (Sotorasib) Resistance in KRASG12C-Mutated Lung Adenocarcinoma

The Kirsten rat sarcoma viral oncogene homolog (KRAS)G12C mutation is prevalent in lung adenocarcinoma (LUAD), driving tumor progression and indicating a poor prognosis. While the FDA-approved AMG510 (Sotorasib) initially demonstrated efficacy in treating KRASG12C-mutated LUAD, resistance emerged within months. Data from AMG510 treatment-resistant LUAD (GSE204753) and single-cell datasets (GSE149655) were analyzed. Gene set variation analysis (GSVA) and gene set enrichment analysis (GSEA) were used to explore enriched signaling pathways, nomogram models were constructed, and transcription factors predicting resistance biomarkers were predicted. CIBERSORT identified immune cell subpopulations, and their association with resistance biomarkers was assessed through single-cell analysis. AMG510-resistant LUAD cells (H358-AR) were constructed, and proliferative changes were evaluated using a CCK-8 assay. Key molecules for AMG510 resistance, including SLC2A1, TLE1, FAM83A, HMGA2, FBXO44, and MTRNR2L12, were recognized. These molecules impacted multiple signaling pathways and the tumor microenvironment and were co-regulated by various transcription factors. Single-cell analysis revealed a dampening effect on immune cell function, with associations with programmed cell death ligand 1 (PDL1) expression, cytokine factors, and failure factors. The findings indicate that these newly identified biomarkers are linked to the abnormal expression of PDL1 and have the potential to induce resistance through immunosuppression. These results highlight the need for further research and therapeutic intervention to address this issue effectively.

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.

Homologous recombination deficiency reflects the heterogeneity and monitoring treatment response for patients with breast cancer

BACKGROUND

In breast cancer (BC), homologous recombination defect (HRD) is a common carcinogenic mechanism. It is meaningful to classify BC according to HRD biomarkers and to develop a platform for identifying BC molecular features, pathological features and therapeutic responses.

METHODS

In total, 109 HRD genes were collected and screened by univariate Cox regression analysis to determine the prognostic genes, which were used to construct a consensus matrix to identify BC subtype. Differentially expressed genes (DEGs) were filtered by the Limma package and screened by random forest analysis to build a model to analyze the immunotherapy response and sensitivity and prognosis of patients suffering from BC to different drugs.

RESULTS

Thirteen out of 109 HRD genes were prognostic genes of BC, and BC was classified into two subgroups based on their expression. Cluster 1 had a significantly backward survival outcome and a significantly higher adaptive immunity score relative to cluster 2. Six genes were identified by random forest analysis as factors for developing the model. The model provided a prediction called risk score, which showed a significant stratification effect on BC prognosis, immunotherapy response and IC50 values of 62 drugs.

CONCLUSIONS

In the present study, two HRD subtypes of BC were successfully identified, for which mutation and immunological features were determined. A model based on differential genes of HRD subtypes was established, which was a potential predictor of prognosis, immunotherapy response and drug sensitivity of BC.

Clinical and therapeutical significances of the cluster and signature based on oxidative stress for osteosarcoma

BACKGROUND

It is of great clinical significance to find out the ideal tumor biomarkers and therapeutic targets to improve the prognosis of patients with osteosarcoma (OS). Oxidative stress (OXS) can directly target intracellular macromolecules and exhibit dual effects of tumor promotion and suppression.

METHODS

OXS-related genes (OXRGs) were extracted from public databases, including TARGET and GEO. Univariate Cox regression analysis, Random Survival Forest algorithm, and LASSO regression were performed to identify prognostic genes and establish the OXS-signature. The efficacy of the OXS-signature was further evaluated by Kaplan-Meier curves and timeROC package. Evaluation of immunological characteristics was achieved based on ESTIMATE algorithm and ssGSEA. Submap algorithm was used to explore the response to anti-PD1 and anti-CTLA4 therapy for OS. Drug response prediction was conducted by using pRRophetic package. The expression values of related genes in the OXS-signature were detected with PCR assays.

RESULTS

Two OXS-clusters were identified for OS, with remarkable differences of clusters presented in prognosis. Kyoto Encyclopedia of Genes Genomes (KEGG) analysis showed that differentially expressed genes (DEGs) between the OXS-clusters were significantly enriched in several immune-related pathways. Patients with lower OS-scores attained better clinical outcomes, and presented more sensitivity to ICB therapy. By contrast, OS patients with higher OS-scores revealed more sensitivity to certain drugs. Furthermore, critical genes, RHBDL2 and CGREF1 from the model, were significantly higher expressed in OS cell lines.

CONCLUSIONS

Our study identified the clusters and signature based on OXS, which would lay the foundation for molecular experimental research, disease prevention and treatment of OS.

Potential value of mitochondrial regulatory pathways in the clinical application of clear cell renal cell carcinoma: a machine learning-based study

BACKGROUND

Renal clear cell carcinoma (RCC) is a common cancer in urinary system with increasing incidence. At present, targeted therapy and immunotherapy are the main therapeutic programs in clinical therapy. To develop novel drugs and provide new ideas for clinical therapy, the identification of potential ccRCC subtypes and potential target genes or pathways has become a current research focus.

AIM

The aim of this study was to explore the underlying mechanisms of mitochondrial function in ccRCC. This regulatory pathway is closely related to tumor development and metastasis in ccRCC patients, and their abnormal changes may affect the prognosis of cancer patients. Therefore, we decided to construct a prognostic model of ccRCC patients based on mitochondrial regulatory genes, aiming to provide new methods and ideas for clinical therapy.

RESULT

The 5-year survival prediction model based on iterative LASSO reached 0.746, and the cox model based on coxph reached C-index = 0.77, integrated c/D AUC = 0.61, and integrated brier score = 0.14. The rsf model based on randomForestSRC was built with C-index = 0.82, integrated c/D AUC = 0.69, and integrated brier score = 0.11. The results show that mitochondrial regulatory pathway is a potential target pathway for clinical therapy of ccRCC, which can provide guidelines for clinical targeted therapy, immunotherapy and other first-line therapy.

Machine-Learning Algorithms Using Systemic Inflammatory Markers to Predict the Oncologic Outcomes of Colorectal Cancer After Surgery

BACKGROUND

This study aimed to investigate the clinical significance of machine-learning (ML) algorithms based on serum inflammatory markers to predict survival outcomes for patients with colorectal cancer (CRC).

METHODS

The study included 941 patients with stages I to III CRC. Based on random forest algorithms using 15 compositions of inflammatory markers, four different prediction scores (DFS score-1, DFS score-2, DFS score-3, and DFS score-4) were developed for the Yonsei cohort (training set, n = 803) and tested in the Ulsan cohort (test set, n = 138). The Cox proportional hazards model was used to determine correlation between prediction scores and disease-free survival (DFS). Harrell's concordance index (C-index) was used to compare the predictive ability of prediction scores for each composition.

RESULTS

The multivariable analysis showed the DFS score-4 to be an independent prognostic factor after adjustment for clinicopathologic factors in both the training and test sets (hazard ratio [HR], 8.98; 95% confidence interval [CI] 6.7-12.04; P < 0.001 for the training set and HR, 2.55; 95% CI 1.1-5.89; P = 0.028 for the test set]. With regard to DFS, the highest C-index among single compositions was observed in the lymphocyte-to-C-reactive protein ratio (LCR) (0.659; 95% CI 0.656-0.662), and the C-index of DFS score-4 (0.727; 95% CI 0.724-0.729) was significantly higher than that of LCR in the test set. The C-index of DFS score-3 (0.725; 95% CI 0.723-0.728) was similar to that of DFS score-4, but higher than that of DFS score-2 (0.680; 95% CI 0.676-0.683).

CONCLUSIONS

The ML-based approaches showed prognostic utility in predicting DFS. They could enhance clinical use of inflammatory markers in patients with CRC.

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.

Molecular subtypes of lung adenocarcinoma patients for prognosis and therapeutic response prediction with machine learning on 13 programmed cell death patterns

BACKGROUND

Lung adenocarcinoma (LUAD) seriously threatens people's health worldwide. Programmed cell death (PCD) plays a critical role in regulating LUAD growth and metastasis as well as in therapeutic response. However, currently, there is a lack of integrative analysis of PCD-related signatures of LUAD for accurate prediction of prognosis and therapeutic response.

METHODS

The bulk transcriptome and clinical information of LUAD were obtained from TCGA and GEO databases. A total of 1382 genes involved in regulating 13 various PCD patterns (apoptosis, necroptosis, pyroptosis, ferroptosis, cuproptosis, netotic cell death, entotic cell death, lysosome-dependent cell death, parthanatos, autophagy-dependent cell death, oxeiptosis, alkaliptosis and disulfidptosis) were included in the study. Weighted gene co-expression network analysis (WGCNA) and differential expression analysis were performed to identify PCD-associated differential expression genes (DEGs). An unsupervised consensus clustering algorithm was used to explore the potential subtypes of LUAD based on the expression profiles of PCD-associated DEGs. Univariate Cox regression analysis, Least Absolute Shrinkage and Selection Operator (LASSO) regression, Random Forest (RF) analysis and stepwise multivariate Cox analysis were performed to construct a prognostic gene signature. The "oncoPredict" algorithm was utilized for drug-sensitive analysis. GSVA and GSEA were utilized to perform function enrichment analysis. MCPcounter, quanTIseq, Xcell and ssGSEA algorithms were used for tumor immune microenvironment analysis. A nomogram incorporating PCDI and clinicopathological characteristics was established to predict the prognosis of LUAD patients.

RESULTS

Forty PCD-associated DEGs related to LUAD were obtained by WGCNA analysis and differential expression analysis, followed by unsupervised clustering to identify two LUAD molecular subtypes. A programmed cell death index (PCDI) with a five-gene signature was established by machine learning algorithms. LUAD patients were then divided into a high PCDI group and a low PCDI group using the median PCDI as a cutoff. Survival and therapeutic analysis revealed that the high PCDI group had a poor prognosis and was more sensitive to targeted drugs but less sensitive to immunotherapy compared to the low PCDI group. Further enrichment analysis showed that B cell-related pathways were significantly downregulated in the high PCDI group. Accordingly, the decreased tumor immune cell infiltration and the lower tumor tertiary lymphoid structure (TLS) scores were also found in the high PCDI group. Finally, a nomogram with reliable predictive performance PCDI was constructed by incorporating PCDI and clinicopathological characteristics, and a user-friendly online website was established for clinical reference ( https://nomogramiv.shinyapps.io/NomogramPCDI/ ).

CONCLUSION

We performed the first comprehensive analysis of the clinical relevance of genes regulating 13 PCD patterns in LUAD and identified two LUAD molecular subtypes with distinct PCD-related gene signature which indicated differential prognosis and treatment sensitivity. Our study provided a new index to predict the efficacy of therapeutic interventions and the prognosis of LUAD patients for guiding personalized treatments.

Screening strategies and dynamic risk prediction models for Alzheimer's disease

BACKGROUND

Characterizing the progression from Mild cognitive impairment (MCI) to Alzheimer's disease (AD) is essential for early AD prevention and targeted intervention. Our goal was to construct precise screening schemes for individuals with different risk of AD and to establish prognosis models for them.

METHODS

We constructed a retrospective cohort by reviewing individuals with baseline diagnosis of MCI and at least one follow-up visits between November 2005 and May 2021. They were stratified into high-risk and low-risk groups with longitudinal cognitive trajectory. Then, we established a screening framework and obtained optimal screening strategies for two risk groups. Cox and random survival forest (RSF) models were developed for dynamic prognosis prediction.

RESULTS

In terms of screening strategies, the combination of Clinical Dementia Rating Sum of Boxes (CDRSB) and hippocampus volume was recommended for the high-risk MCI group, while the combination of Alzheimer's Disease Assessment Scale Cognitive 13 items (ADAS13) and FAQ was recommended for low-risk MCI group. The concordance index (C-index) of the Cox model for the high-risk group was 0.844 (95% CI: 0.815-0.873) and adjustments for demographic information and APOE ε4. The RSF model incorporating longitudinal ADAS13, FAQ, and demographic information and APOE ε4 performed for the low-risk group.

CONCLUSION

This precise screening scheme will optimize allocation of medical resources and reduce the economic burden on individuals with low risk of MCI. Moreover, dynamic prognosis models may be helpful for early identification of individuals at risk and clinical decisions, which will promote the secondary prevention of AD.

The identification of genes associated T-cell exhaustion and construction of prognostic signature to predict immunotherapy response in lung adenocarcinoma

T-cell exhaustion (Tex) is considered to be a reason for immunotherapy resistance and poor prognosis in lung adenocarcinoma. Therefore, we used weighted correlation network analysis to identify Tex-related genes in the cancer genome atlas (TCGA). Unsupervised clustering approach based on Tex-related genes divided patients into cluster 1 and cluster 2. Then, we utilized random forest and the least absolute shrinkage and selection operator to identify nine key genes to construct a riskscore. Patients were classified as low or high-risk groups. The multivariate cox analysis showed the riskscore was an independent prognostic factor in TCGA and GSE72094 cohorts. Moreover, patients in cluster 2 with high riskscore had the worst prognosis. The immune response prediction analysis showed the low-risk group had higher immune, stromal, estimate scores, higher immunophenscore (IPS), and lower tumor immune dysfunction and exclusion score which suggested a better response to immune checkpoint inhibitors (ICIs) therapy in the low-risk group. In the meantime, we included two independent immunotherapy cohorts that also confirmed a better response to ICIs treatment in the low-risk group. Besides, we discovered differences in chemotherapy and targeted drug sensitivity between two groups. Finally, a nomogram was built to facilitate clinical decision making.

Preoperative Radiomics Nomogram Based on CT Image Predicts Recurrence-Free Survival After Surgical Resection of Hepatocellular Carcinoma

RATIONALE AND OBJECTIVES

To construct preoperative models based on CT radiomics, radiologic and clinical features to predict recurrence-free survival (RFS) after liver resection (LR) of BCLC 0 to B stage hepatocellular carcinoma (HCC) and to classify the prognosis.

MATERIALS AND METHODS

This study retrospectively analyzed 161 HCC patients who underwent radical LR. Two methods, the least absolute shrinkage and selection operator and random survival forest analysis, were performed for radiomics signature (RS) construction. Univariate and multivariate stepwise Cox regression analyses were performed to establish a combined nomogram (RCN) of RS and clinical parameters and a clinical nomogram (CN). The performance of the models was assessed comprehensively using Harrell's concordance index (C-index), the calibration curve, and decision curve analysis. The discrimination accuracy of the models was compared using integrated discrimination improvement index (IDI). The risk stratification effect was assessed with Kaplan-Meier survival analysis and subgroup analysis.

RESULTS

The RCN achieved a C-index of 0.792/0.758 in the training/validation set, which was higher than the CN, RS, and BCLC stage system. The discriminatory accuracy of the RCN was improved when compared to the CN, RS, and BCLC staging systems (IDI > 0). Decision curve analysis reflected the clinical net benefit of the RCN. The RCN allows risk stratification of patients in different clinical subgroups.

CONCLUSION

The integrated model combining RS and clinical factors can more effectively predict RFS after LR of BCLC 0 to B stage HCC patients and can effectively stratify the prognostic risk.

Development of preoperative and postoperative machine learning models to predict the recurrence of huge hepatocellular carcinoma following surgical resection

Resection has been commonly utilized for treating huge hepatocellular carcinoma (HCC) with a diameter of ≥10 cm; however, a high rate of mortality is reported due to recurrence. The present study was designed to predict the recurrence following resection based on preoperative and postoperative machine learning models. In total, 1,082 patients with HCC who underwent liver resection in the Eastern Hepatobiliary Surgery Hospital cohort between January 2008 and December 2016 were divided into a training cohort and an internal validation cohort. In addition, 164 patients from Mengchao Hepatobiliary Hospital cohort between January 2014 and December 2016 served as an external validation cohort. The demographic information, and serological, MRI, and pathological data were obtained from each patient prior to and following surgery, followed by evaluating the model performance using the concordance index, time-dependent receiver operating characteristic curves, prediction error cures, and a calibration curve. A preoperative random survival forest (RSF) model and a postoperative RSF model were constructed based on the training set, which outperformed the conventional models, such as the Barcelona Clinic Liver Cancer (BCLC), the 8th edition of the American Joint Committee on Cancer (AJCC 8th) staging systems, and the Chinese stage systems. In addition, the preoperative and postoperative RSF models could also re-stratify patients with BCLC stage A/B/C or AJCC 8th stage IB/II/IIIA/IIIB or Chinese stage IB/IIA/IIB/IIIA into low-risk, intermediate-risk, and high-risk groups in the training and the two validation cohorts. The preoperative and postoperative RSF models were effective for predicting recurrence in patients with huge HCC following hepatectomy.

Identification of macrophage-related genes in sepsis-induced ARDS using bioinformatics and machine learning

Sepsis-induced acute respiratory distress syndrome (ARDS) is one of the leading causes of death in critically ill patients, and macrophages play very important roles in the pathogenesis and treatment of sepsis-induced ARDS. The aim of this study was to screen macrophage-related biomarkers for the diagnosis and treatment of sepsis-induced ARDS by bioinformatics and machine learning algorithms. A dataset including gene expression profiles of sepsis-induced ARDS patients and healthy controls was downloaded from the gene expression omnibus database. The limma package was used to screen 325 differentially expressed genes, and enrichment analysis suggested enrichment mainly in immune-related pathways and reactive oxygen metabolism pathways. The level of immune cell infiltration was analysed using the ssGSEA method, and then 506 macrophage-related genes were screened using WGCNA; 48 showed differential expression. PPI analysis was also performed. SVM-RFE and random forest map analysis were used to screen 10 genes. Three key genes, SGK1, DYSF and MSRB1, were obtained after validation with external datasets. ROC curves suggested that all three genes had good diagnostic efficacy. The nomogram model consisting of the three genes also had good diagnostic efficacy. This study provides new targets for the early diagnosis of sepsis-induced ARDS.

Predicting the survival of patients with pancreatic neuroendocrine neoplasms using deep learning: A study based on Surveillance, Epidemiology, and End Results database

BACKGROUND

The study aims to evaluate the performance of three advanced machine learning algorithms and a traditional Cox proportional hazard (CoxPH) model in predicting the overall survival (OS) of patients with pancreatic neuroendocrine neoplasms (PNENs).

METHOD

The clinicopathological dataset obtained from the Surveillance, Epidemiology, and End Results database was randomly assigned to the training set and testing set at a ratio of 7:3. The concordance index (C-index) and integrated Brier score (IBS) were used to compare the predictive performance of the models. The accuracy of the model in predicting the 5-year and 10-year survival rates was compared using the receiver operating characteristic curve, decision curve analysis (DCA) and calibration curve.

RESULTS

This study included 3239 patients with PNENs in total. The DeepSurv model had the highest C-index of 0.7882 in the testing set and training set and the lowest IBS of 0.1278 in the testing set compared with the CoxPH, neural multitask logistic and random survival forest models (C-index = 0.7501, 0.7616, and 0.7612, respectively; IBS = 0.1397, 0.1418, and 0.1432, respectively). Moreover, the DeepSurv model had the highest accuracy in predicting 5- and 10-year OS rates (area under the curve: 0.87 and 0.90). DCA showed that the DeepSurv model had high potential for clinical decisions in 5- and 10-year OS models. Finally, we developed an online application based on the DeepSurv model for clinical use (https://whuh-ml-neuroendocrinetumor-app-predict-oyw5km.streamlit.app/).

CONCLUSIONS

All four models analyzed above can predict the prognosis of PNENs well, among which the DeepSurv model has the best prediction performance.

Systematic analysis of transcriptome signature for improving outcomes in lung adenocarcinoma

PURPOSE

The updated guidelines highlight gene expression-based multigene panel as a critical tool to assess overall survival (OS) and improve treatment for lung adenocarcinoma (LUAD) patients. Nevertheless, genome-wide expression signatures are still limited in real clinical utility because of insufficient data utilization, a lack of critical validation, and inapposite machine learning algorithms.

METHODS

2330 primary LUAD samples were enrolled from 11 independent cohorts. Seventy-six algorithm combinations based on ten machine learning algorithms were applied. A total of 108 published gene expression signatures were collected. Multiple pharmacogenomics databases and resources were utilized to identify precision therapeutic drugs.

RESULTS

We comprehensively developed a robust machine learning-derived genome-wide expression signature (RGS) according to stably OS-associated RNAs (OSRs). RGS was an independent risk element and remained robust and reproducible power by comparing it with general clinical parameters, molecular characteristics, and 108 published signatures. RGS-based stratification possessed different biological behaviors, molecular mechanisms, and immune microenvironment patterns. Integrating multiple databases and previous studies, we identified that alisertib was sensitive to the high-risk group, and RITA was sensitive to the low-risk group.

CONCLUSION

Our study offers an appealing platform to screen dismal prognosis LUAD patients to improve clinical outcomes by optimizing precision therapy.

Machine Learning-Based Prognostic Model for Patients After Lung Transplantation

Importance

Although numerous prognostic factors have been found for patients after lung transplantation (LTx) over the years, an accurate prognostic tool for LTx recipients remains unavailable.

Objective

To develop and validate a prognostic model for predicting overall survival in patients after LTx using random survival forests (RSF), a machine learning algorithm.

Design, Setting, and Participants

This retrospective prognostic study included patients who underwent LTx between January 2017 and December 2020. The LTx recipients were randomly assigned to training and test sets in accordance with a ratio of 7:3. Feature selection was performed using variable importance with bootstrapping resampling. The prognostic model was fitted using the RSF algorithm, and a Cox regression model was set as a benchmark. The integrated area under the curve (iAUC) and integrated Brier score (iBS) were applied to assess model performance in the test set. Data were analyzed from January 2017 to December 2019.

Main Outcomes And Measures

Overall survival in patients after LTx.

Results

A total of 504 patients were eligible for this study, consisting of 353 patients in the training set (mean [SD] age, 55.03 [12.78] years; 235 [66.6%] male patients) and 151 patients in the test set (mean [SD] age, 56.79 [10.95] years; 99 [65.6%] male patients). According to the variable importance of each factor, 16 were selected for the final RSF model, and postoperative extracorporeal membrane oxygenation time was identified as the most valuable factor. The RSF model had excellent performance with an iAUC of 0.879 (95% CI, 0.832-0.921) and an iBS of 0.130 (95% CI, 0.106-0.154). The Cox regression model fitted by the same modeling factors to the RSF model was significantly inferior to the RSF model with an iAUC of 0.658 (95% CI, 0.572-0.747; P < .001) and an iBS of 0.205 (95% CI, 0.176-0.233; P < .001). According to the RSF model predictions, the patients after LTx were stratified into 2 prognostic groups displaying significant difference, with mean overall survival of 52.91 months (95% CI, 48.51-57.32) and 14.83 months (95% CI, 9.44-20.22; log-rank P < .001), respectively.

Conclusions and relevance

In this prognostic study, the findings first demonstrated that RSF could provide more accurate overall survival prediction and remarkable prognostic stratification than the Cox regression model for patients after LTx.

Risk prediction score for clinical outcome in atrial fibrillation and stable coronary artery disease

OBJECTIVE

Antithrombotic therapy is essential for patients with atrial fibrillation (AF) and stable coronary artery disease (CAD) because of the high risk of thrombosis, whereas a combination of antiplatelets and anticoagulants is associated with a high risk of bleeding. We sought to develop and validate a machine-learning-based model to predict future adverse events.

METHODS

Data from 2215 patients with AF and stable CAD enrolled in the Atrial Fibrillation and Ischaemic Events With Rivaroxaban in Patients With Stable Coronary Artery Disease trial were randomly assigned to the development and validation cohorts. Using the random survival forest (RSF) and Cox regression models, risk scores were developed for net adverse clinical events (NACE) defined as all-cause death, myocardial infarction, stroke or major bleeding.

RESULTS

Using variables selected by the Boruta algorithm, RSF and Cox models demonstrated acceptable discrimination and calibration in the validation cohort. Using the variables weighted by HR (age, sex, body mass index, systolic blood pressure, alcohol consumption, creatinine clearance, heart failure, diabetes, antiplatelet use and AF type), an integer-based risk score for NACE was developed and classified patients into three risk groups: low (0-4 points), intermediate (5-8) and high (≥9). In both cohorts, the integer-based risk score performed well, with acceptable discrimination (area under the curve 0.70 and 0.66, respectively) and calibration (p>0.40 for both). Decision curve analysis showed the superior net benefits of the risk score.

CONCLUSIONS

This risk score can predict the risk of NACE in patients with AF and stable CAD.

TRIAL REGISTRATION NUMBERS

UMIN000016612, NCT02642419.

Integrated analysis of single-cell and bulk RNA sequencing identifies a signature based on macrophage marker genes involved in prostate cancer prognosis and treatment responsiveness

In the tumor microenvironment, tumor-associated macrophages (TAMs) interact with cancer cells and contribute to the progression of solid tumors. Nonetheless, the clinical significance of TAM-related biomarkers in prostate cancer (PCa) is largely unexplored. The present study aimed to construct a macrophage-related signature (MRS) for predicting PCa patient prognosis based on macrophage marker genes. Six cohorts comprising 1056 PCa patients with RNA-Seq and follow-up data were enrolled. Based on macrophage marker genes identified by single-cell RNA-sequencing (scRNA-seq) analysis, univariate analysis, least absolute shrinkage and selection operator (Lasso)-Cox regression, and machine learning procedures were performed to derive a consensus MRS. Receiver operating characteristic curve (ROC), concordance index, and decision curve analyses were used to confirm the predictive capacity of the MRS. The predictive performance of the MRS for recurrence-free survival (RFS) was stable and robust, and the MRS outperformed traditional clinical variables. Furthermore, high-MRS-score patients presented abundant macrophage infiltration and high-expression levels of immune checkpoints (CTLA4, HAVCR2, and CD86). The frequency of mutations was relatively high in the high-MRS-score subgroup. However, the low-MRS-score patients had a better response to immune checkpoint blockade (ICB) and leuprolide-based adjuvant chemotherapy. Notably, abnormal ATF3 expression may be associated with docetaxel and cabazitaxel resistance in PCa cells, T stage, and the Gleason score. In this study, a novel MRS was first developed and validated to accurately predict patient survival outcomes, evaluate immune characteristics, infer therapeutic benefits, and provide an auxiliary tool for personalized therapy.

Illustrating the biological functions and diagnostic value of transmembrane protein family members in glioma

Background

It is well-established that patients with glioma have a poor prognosis. Although the past few decades have witnessed unprecedented medical advances, the 5-year survival remains dismally low.

Objective

This study aims to investigate the role of transmembrane protein-related genes in the development and prognosis of glioma and provide new insights into the pathogenesis of the disease

Methods

The datasets of glioma patients, including RNA sequencing data and relative clinical information, were obtained from The Cancer Genome Atlas (TCGA), Chinese Glioma Genome Atlas (CGGA) and Gene Expression Omnibus (GEO) databases. Prognostic transmembrane protein-related genes were identified by univariate Cox analysis. New disease subtypes were recognized based on the consensus clustering method, and their biological uniqueness was verified via various algorithms. The prognosis signature was constructed using the LASSO-Cox regression model, and its predictive power was validated in external datasets by receiver operating characteristic (ROC) curve analysis. An independent prognostic analysis was conducted to evaluate whether the signature could be considered a prognostic factor independent of other variables. A nomogram was constructed in conjunction with traditional clinical variables. The concordance index (C-index) and Decision Curve Analysis (DCA) were used to assess the net clinical benefit of the signature over traditional clinical variables. Seven different softwares were used to compare the differences in immune infiltration between the high- and low-risk groups to explore potential mechanisms of glioma development and prognosis. Hub genes were found using the random forest method, and their expression was based on multiple single-cell datasets.

Results

Four molecular subtypes were identified, among which the C1 group had the worst prognosis. Principal Component Analysis (PCA) results and heatmaps indicated that prognosis-related transmembrane protein genes exhibited differential expression in all four groups. Besides, the microenvironment of the four groups exhibited significant heterogeneity. The 6 gene-based signatures could predict the 1-, 2-, and 3-year overall survival (OS) of glioma patients. The signature could be used as an independent prognosis factor of glioma OS and was superior to traditional clinical variables. More immune cells were infiltrated in the high-risk group, suggesting immune escape. According to our signature, many genes were associated with the content of immune cells, which revealed that transmembrane protein-related genes might influence the development and prognosis of glioma by regulating the immune microenvironment. TMEM158 was identified as the most important gene using the random forest method. The single-cell datasets consistently showed that TMEM158 was expressed in multiple malignant cells.

Conclusion

The expression of transmembrane protein-related genes is closely related to the immune status and prognosis of glioma patients by regulating tumor progression in various ways. The interaction between transmembrane protein-related genes and immunity during glioma development lays the groundwork for future studies on the molecular mechanism and targeted therapy of glioma.

Which model is better in predicting the survival of laryngeal squamous cell carcinoma?: Comparison of the random survival forest based on machine learning algorithms to Cox regression: analyses based on SEER database

Prediction of postoperative survival for laryngeal carcinoma patients is very important. This study attempts to demonstrate the utilization of the random survival forest (RSF) and Cox regression model to predict overall survival of laryngeal squamous cell carcinoma (LSCC) and compare their performance. A total of 8677 patients diagnosed with LSCC from 2004 to 2015 were obtained from surveillance, epidemiology, and end results database. Multivariate imputation by chained equations was applied to filling the missing data. Lasso regression algorithm was conducted to find potential predictors. RSF and Cox regression were used to develop the survival prediction models. Harrell's concordance index (C-index), area under the curve (AUC), Brier score, and calibration plot were used to evaluate the predictive performance of the 2 models. For 3-year survival prediction, the C-index in training set were 0.74 (0.011) and 0.84 (0.013) for Cox and RSF respectively. For 5-year survival prediction, the C-index in training set were 0.75 (0.022) and 0.80 (0.011) for Cox and RSF respectively. Similar results were found in validation set. The AUC were 0.795 for RSF and 0.715 for Cox in the training set while the AUC were 0.765 for RSF and 0.705 for Cox in the validation set. The prediction error curves for each model based on Brier score showed the RSF model had lower prediction errors both in training group and validation group. What's more, the calibration curve displayed similar results of 2 models both in training set and validation set. The performance of RSF model were better than Cox regression model. The RSF algorithms provide a relatively better alternatives to be of clinical use for estimating the survival probability of LSCC patients.

New Staging System and Prognostic Model for Malignant Phyllodes Tumor Patients without Distant Metastasis: A Development and Validation Study

PURPOSE

To build a new staging system and new prognostic models for MPTB.

METHODS

We performed a comprehensive analysis of the data from the SEER database.

RESULTS

We discussed the characteristics of MPTB by comparing 1085 MPTB cases with 382,718 invasive ductal carcinoma cases. We established a new stage- and age-stratification system for MPTB patients. Furthermore, we built two prognostic models for MPTB patients. The validity of these models was confirmed through multifaceted and multidata verification.

CONCLUSIONS

Our study provided a staging system and prognostic models for MPTB patients, which can not only help to predict patient outcomes, but also enhance the understanding of the prognostic factors associated with MPTB.

A systematic and comprehensive analysis of T cell exhaustion related to therapy in lung adenocarcinoma tumor microenvironment

Background: T cell exhaustion (TEX) is an important immune escape mechanism, and an in-depth understanding of it can help improve cancer immunotherapy. However, the prognostic role of TEX in malignant lung adenocarcinoma (LUAD) remains unclear. Methods: Through TCGA and GEO datasets, we enrolled a total of 498 LUAD patients. The patients in TCGA-LUAD were unsupervised clustered into four clusters according to TEX signaling pathway. WGCNA analysis, survival random forest analysis and lasso regression analysis were used to select five differentially expressed genes among different clusters to construct a TEX risk model. The risk model was subsequently validated with GEO31210. By analyzing signaling pathways, immune cells and immune checkpoints using GSEA, GSVA and Cibersortx, the relationship between TEX risk score and these variables was evaluated. In addition, we further analyzed the expression of CCL20 at the level of single-cell RNA-seq and verified it in cell experiments. Results: According to TEX signaling pathway, people with better prognosis can be distinguished. The risk model constructed by CD109, CCL20, DKK1, TNS4, and TRIM29 genes could further accurately identify the population with poor prognosis. Subsequently, it was found that dendritic cells, CD44 and risk score were closely related. The final single-cell sequencing suggested that CCL2O is a potential therapeutic target of TEX, and the interaction between TEX and CD8 + T is closely related. Conclusion: The classification of T cell depletion plays a crucial role in the clinical decision-making of lung adenocarcinoma and needs to be further deepened.

A novel prognostic model related to oxidative stress for treatment prediction in lung adenocarcinoma

Background

The prognostic model based on oxidative stress for lung adenocarcinoma (LUAD) remains unclear.

Methods

The information of LUAD patients were acquired from TCGA dataset. We also collected two external datasets from GEO for verification. Oxidative stress-related genes (ORGs) were extracted from Genecards. We performed machine learning algorithms, including Univariate Cox regression, Random Survival Forest, and Least Absolute Shrinkage and Selection Operator (Lasso) analyses on the ORGs to build the OS-score and OS-signature. We drew the Kaplan-Meier and time-dependent receiver operating characteristic curve (ROC) to evaluate the efficacy of the OS-signature in predicting the prognosis of LUAD. We used GISTIC 2.0 and maftool algorithms to explore Genomic mutation of OS-signature. To analyze characteristic of tumor infiltrating immune cells, ESTIMATE, TIMER2.0, MCPcounter and ssGSEA algorithms were applied, thus evaluating the immunotherapeutic strategies. Chemotherapeutics sensitivity analysis was based on pRRophetic package. Finally, PCR assays was also used to detect the expression values of related genes in the OS-signature in cell lines.

Results

Ten ORGs with prognostic value and the OS-signature containing three prognostic ORGs were identified. The significantly better prognosis of LUAD patients was observed in LUAD patients. The efficiency and accuracy of OS-signature in predicting prognosis for LUAD patients was confirmed by survival ROC curves and two external validation data sets. It was clearly observed that patients with high OS-scores had lower immunomodulators levels (with a few exceptions), stromal score, immune score, ESTIMATE score and infiltrating immune cell populations. On the contrary, patients with higher OS-scores were more likely to have higher tumor purity. PCR assays showed that, MRPL44 and CYCS were significantly higher expressed in LUAD cell lines, while CAT was significantly lower expressed.

Conclusion

The novel oxidative stress-related model we identified could be used for prognosis and treatment prediction in lung adenocarcinoma.

A prediction model for the 5-year, 10-year and 20-year mortality of medullary thyroid carcinoma patients based on lymph node ratio and other predictors

Aim

To explore the predictive value of lymph node ratio (LNR) for the prognosis of medullary thyroid carcinoma (MTC) patients, and constructed prediction models for the 5-year, 10-year and 20-year mortality of MTC patients based on LNR and other predictors.

Methods

This cohort study extracted the data of 2,093 MTC patients aged ≥18 years undergoing total thyroidectomy and neck lymph nodes dissection. Kaplan-Meier curves and log-rank tests were performed to compare survival curves between LNR < 15% group and LNR ≥ 15% group. All data was divided into the training set (n = 1,465) and the testing set (n = 628). The random survival forest model was constructed in the training set and validated in the testing set. The area under the curve (AUC) was employed for evaluating the predictive ability of the model.

Results

The 5-year, 10-year and 20-year overall survival (OS) and cause-specific survival (CSS) of MTC patients with LNR <15% were higher than those with LNR ≥15%. The OS was 46% and the CSS was 75% after 20 years' follow-up. The AUC of the model for the 5-year, 10-year, and 20-year OS in MTC patients was 0.878 (95%CI: 0.856-0.900), 0.859 (95%CI: 0.838-0.879) and 0.843 (95%CI: 0.823-0.862) in the training set and 0.845 (95%CI: 0.807-0.883), 0.841 (95%CI: 0.807-0.875) and 0.841 (95%CI: 0.811-0.872) in the testing set. In the training set, the AUCs were 0.869 (95%CI: 0.845-0.892), 0.843 (95%CI: 0.821-0.865), 0.819 (95%CI: 0.798-0.840) for the 5-year, 10-year and 20-year CCS in MTC patients, respectively. In the testing set, the AUCs were 0.857 (95%CI: 0.822-0.892), 0.839 (95%CI: 0.805-0.873) and 0.826 (95%CI: 0.794-0.857) for the 5-year CCS, 10-year CCS and 20-year CCS in MTC patients, respectively.

Conclusion

The models displayed good predictive performance, which might help identify MTC patients might have poor outcomes and appropriate interventions should be applied in these patients.

Predicting prognosis and clinical efficacy of immune checkpoint blockade therapy via interferon-alpha response in muscle-invasive bladder cancer

Background: Immune checkpoint blockade (ICB) can prompt durable and robust responses in multiple cancers, involving muscle-invasive bladder cancer (MIBC). However, only a limited fraction of patients received clinical benefit. Clarifying the determinants of response and exploring corresponding predictive biomarkers is key to improving outcomes. Methods: Four independent formerly published cohorts consisting of 641 MIBC patients were enrolled in this study. We first analyzed the associations between various cancer hallmarks and ICB therapy response in two immunotherapeutic cohorts to identify the leading prognostic hallmark in MIBC. Furthermore, advanced machine learning methods were performed to select robust and promising predictors from genes functioning in the above leading pathway. The predictive ability of selected genes was also validated in multiple MIBC cohorts. Results: We identified and verified IFNα response as the leading cancer hallmark indicating better treatment responses, favorable overall survival, and an inflamed tumor microenvironment with higher infiltration of immune effector cells in MIBC patients treated with ICB therapy. Subsequently, two commonly selected genes, CXCL10 and LAMP3, implied better therapy response and the CXCL10^highLAMP3^high patients would benefit more from ICB therapy, which was comprehensively validated from the perspective of gene expression, clinical response, patient survival and immune features. Conclusion: Higher IFNα response primarily predicted better ICB therapeutic responses and reflected an inflamed microenvironment in MIBC. A composite of CXCL10 and LAMP3 expression could serve as promising predictive biomarkers for ICB therapeutic responses and be beneficial for clinical decision-making in MIBC.

Development and validation of survival prediction model for gastric adenocarcinoma patients using deep learning: A SEER-based study

Background

The currently available prediction models, such as the Cox model, were too simplistic to correctly predict the outcome of gastric adenocarcinoma patients. This study aimed to develop and validate survival prediction models for gastric adenocarcinoma patients using the deep learning survival neural network.

Methods

A total of 14,177 patients with gastric adenocarcinoma from the Surveillance, Epidemiology, and End Results (SEER) database were included in the study and randomly divided into the training and testing group with a 7:3 ratio. Two algorithms were chosen to build the prediction models, and both algorithms include random survival forest (RSF) and a deep learning based-survival prediction algorithm (DeepSurv). Also, a traditional Cox proportional hazard (CoxPH) model was constructed for comparison. The consistency index (C-index), Brier score, and integrated Brier score (IBS) were used to evaluate the model's predictive performance. The accuracy of predicting survival at 1, 3, 5, and 10 years was also assessed using receiver operating characteristic curves (ROC), calibration curves, and area under the ROC curve (AUC).

Results

Gastric adenocarcinoma patients were randomized into a training group (n = 9923) and a testing group (n = 4254). DeepSurv showed the best performance among the three models (c-index: 0.772, IBS: 0.1421), which was superior to that of the traditional CoxPH model (c-index: 0.755, IBS: 0.1506) and the RSF with 3-year survival prediction model (c-index: 0.766, IBS: 0.1502). The DeepSurv model produced superior accuracy and calibrated survival estimates predicting 1-, 3- 5- and 10-year survival (AUC: 0.825-0.871).

Conclusions

A deep learning algorithm was developed to predict more accurate prognostic information for gastric cancer patients. The DeepSurv model has advantages over the CoxPH and RSF models and performs well in discriminative performance and calibration.

A transient receptor potential channel-related model based on machine learning for evaluating tumor microenvironment and immunotherapeutic strategies in acute myeloid leukemia

Background

Acute myeloid leukemia (AML) is an aggressive hematopoietic malignancy. Transient receptor potential (TRP) channels in AML still need to be further explored. A TRP channel-related model based on machine learning was established in this study.

Methods

The data were downloaded from TCGA-LAML and Genome-Tissue Expression (GTEx). TRP-related genes (TRGs) were extracted from previous literature. With the use of Single-Sample Gene Set Enrichment Analysis (ssGSEA), TRP enrichment scores (TESs) were calculated. The limma package was used to identify differentially expressed genes (DEGs), and univariate Cox regression analysis was performed to identify prognostic DEGs. The above prognostic DEGs were analyzed by Random Survival Forest and least absolute shrinkage and selection operator (Lasso) analysis to create the TRP signature. The Kaplan-Meier and receiver operating characteristic (ROC) curves were plotted to investigate the efficiency and accuracy of prognostic prediction. Moreover, genomic mutation analysis was based on GISTIC analysis. Based on ESTIMATE, TIMER, MCPcounter, and ssGSEA, the tumor microenvironment and immunological characteristics were expressly evaluated to explore immunotherapeutic strategies. Enrichment analysis for TRP signature was based on the Kyoto Encyclopedia of Genes Genomes (KEGG), Gene Ontology (GO), over-representation analysis (ORA), and Gene Set Enrichment Analysis (GSEA). Genomics of Drug Sensitivity in Cancer (GDSC) and pRRophetic were used to carry out drug sensitivity analysis. Conclusively, SCHIP1 was randomly selected to perform in vitro cyto-functional experiments.

Results

The worse clinical outcomes of patients with higher TESs were observed. There were 107 differentially expressed TRGs identified. Our data revealed 57 prognostic TRGs. Eight TRGs were obtained to establish the prognostic TRP signature, and the worse clinical outcomes of patients with higher TRP scores were found. The efficiency and accuracy of TRP signature in predicting prognosis were confirmed by ROC curves and five external validation datasets. Our data revealed that the mutation rates of DNMT3A, IDH2, MUC16, and TTN were relatively high. The level of infiltrating immune cell populations, stromal, immune, and ESTIMATE scores increased as the TRP scores increased. Nevertheless, AML patients with lower TRP scores exhibited more tumor purity. The TRP scores were found to be correlated with immunomodulators and immune checkpoints, thus revealing immune characteristics and immunotherapeutic strategies. The IC50 values of six chemotherapeutics were lower in the high TRP score (HTS) group. Finally, it was found that SCHIP1 may be the oncogenic gene.

Conclusion

The results of this study will help in understanding the role of TRP and SCHIP1 in the prognosis and development of AML.

Cisplatin-based concurrent chemoradiotherapy improved the survival of locoregionally advanced nasopharyngeal carcinoma after induction chemotherapy by reducing early treatment failure

PURPOSE

The aims of this study focusing on Locoregionally advanced nasopharyngeal carcinoma (LANPC) were mainly two-fold: on the one hand, to establish a cut-off value to differentiate early and late failure based on prognosis after recurrence or metastasis; and on the other hand, to investigate the duration of concurrent cisplatin benefit over follow-up time. The results of our study have the potential to guide clinical practice and follow-up.

METHODS

In total, 3123 patients with stage III-IVa NPC receiving Induction chemotherapy followed by concurrent cisplatin or not were analysed. The cut-off value of treatment failure was calculated using the minimum P-value approach. Random survival forest (RSF) model was to simulate the cumulative probabilities of treatment failure (locoregional recurrence and /or distant metastasis) over-time, as well as the monthly time-specific, event-occurring probabilities, for patients at different treatment groups.

RESULTS

Based on subsequent prognosis, early locoregional failure (ELRF) should be defined as recurrence within 14 months (P = 1.47 × 10 - 3), and early distant failure (EDF) should be defined as recurrence within 20 months (P = 1.95 × 10 - 4). A cumulative cisplatin dose (CCD) > 200 mg/m² independently reduced the risk of EDF (hazard ratio, 0.351; 95% confidence interval (CI), 0.169-0.732; P = 0.005). Better failure-free survival (FFS) and overall survival (OS) were observed in concurrent chemotherapy settings ([0 mg/m² vs. 1-200 mg/m² vs. >200 mg/m²]: FFS: 70.4% vs. 74.4% vs. 82.6%, all P < 0.03; OS: 79.5% vs. 83.8% vs. 90.8%, all P < 0.01). In the monthly analysis, treatment failure mainly occurred during the first 4 years, and the risk of distant failure in patients treated with concurrent chemotherapy never exceeded that of patients without concurrent chemotherapy.

CONCLUSION

Locoregional failure that developed within 14 months and/or distant failure within 20 months had poorer subsequent survival. Concurrent chemotherapy provides a significant FFS benefit, primarily by reducing EDF, translating into a long-term OS benefit.

Comprehensive evaluation of breast cancer immunotherapy and tumor microenvironment characterization based on interleukin genes-related risk model

Breast cancer (BRCA) is the most prevalent malignancy and the leading cause of death in women. Interleukin (IL) genes are critical in tumor initiation and control. Nevertheless, the prognosis value of the IL in BRCA remains unclear. We collected data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO), and 94 IL genes were identified from GeneCard. Based on the random forest (RF), least absolute shrinkage and selection operator (LASSO) analysis, and multivariate Cox regression analysis, we constructed an IL signature. GSE22219, GSE25065, and GSE21653 were derived as validation sets. The expression differences in the tumor microenvironment (TME), immunotherapy, and chemosensitivity of BRCA between the high- and low-risk groups were evaluated. Overall, 21 IL genes were selected to construct an IL risk model, of which IL18BP, IL17D, and IL23A were the first time identified as prognostic genes in BRCA. IL score could distinguish BRCA patients with inferior outcomes, and AUC of it was 0.70, 0.76, and 0.72 for 1-,3- and 5- years, respectively, which was also verified in GSE22219, GSE25065, and GSE21653 cohorts. Meanwhile, compared to luminal A and luminal B, HER2-positive and TNBC had significantly higher IL score. Besides, the high-risk group had a significantly higher prevalence of TP53 and TTN but a lower prevalence of PIK3CA, as well as higher tumor mutation burden (TMB) and neoantigen level. High- and low-risk groups exhibited notable differences in immunomodulators and tumor infiltrates immune cells (TIICs), and the high-risk group had significantly lower Tumor Immune Dysfunction and Exclusion (TIDE) score. Additionally, the high-risk group has more responders to immune or anti-HER2 combination therapy, whereas the low-risk group has higher sensitivity to docetaxel and paclitaxel. Consequently, we constructed a reliable risk model based on the IL genes, which can provide more information on both the risk stratification and personalizing management strategies for BRCA.

Prognostic analysis of pT1-T2aN0M0 cervical adenocarcinoma based on random survival forest analysis and the generation of a predictive nomogram

Background

The efficacy of adjuvant radiotherapy for postoperative patients with early-stage cervical adenocarcinoma who are lymph node-negative is still inconclusive. Establishing a nomogram to predict the prognosis of such patients could facilitate clinical decision-making.

Methods

We recruited 4636 eligible patients with pT1-T2aN0M0 cervical adenocarcinoma between 2004 and 2016 from the Surveillance, Epidemiology and End Results (SEER) database. Random survival forest (RSF) and conditional survival forest (CSF) model was used to assess the prognostic importance of each clinical characteristic variable. We identified independent prognostic factors associated with overall survival (OS) by univariate and multivariate Cox regression risk methods and then constructed a nomogram. We stratified patients based on nomogram risk scores and evaluated the survival benefit of different adjuvant therapies. To reduce confounding bias, we also used propensity score matching (PSM) to match the cohorts before performing survival analyses.

Results

The RSF and CSF model identified several important variables that are associated with prognosis, including grade, age, radiotherapy and tumor size. Patients were randomly divided into training and validation groups at a ratio of 7:3. Multivariate cox analysis revealed that age, grade, tumor size, race, radiotherapy and histology were independent prognostic factors for overall survival. Using these variables, we then constructed a predictive nomogram. The C-index value for evaluating the prognostic nomogram fluctuated between 0.75 and 0.91. Patients were divided into three subgroups based on risk scores, and Kaplan-Meier (K-M) survival analysis revealed that in the low-risk group, postoperative chemotherapy alone was associated with a significantly worse OS than surgery alone. Following PSM, survival analysis showed that compared with surgery alone, radiotherapy was associated with a worse OS in the training group although there was no significant difference in the validation group.

Conclusions

For patients with pT1-T2aN0M0 cervical adenocarcinoma, adjuvant treatments such as postoperative radiotherapy or chemotherapy, compared with surgery alone, are of no benefit with regards to patient survival. Our prognostic nomogram exhibits high accuracy for predicting the survival of patients with early-stage postoperative cervical adenocarcinoma.

Characterization of Epithelial-Mesenchymal Transition Identifies a Gene Signature for Predicting Clinical Outcomes and Therapeutic Responses in Bladder Cancer

Purpose. The complex etiological variables and high heterogeneity of bladder cancer (BC) make prognostic prediction challenging. We aimed to develop a robust and promising gene signature using advanced machine learning methods for predicting the prognosis and therapy responses of BC patients. Methods. The single-sample gene set enrichment analysis (ssGSEA) algorithm and univariable Cox regression were used to identify the primary risk hallmark among the various cancer hallmarks. Machine learning methods were then combined with survival and differential gene expression analyses to construct a novel prognostic signature, which would be validated in two additional independent cohorts. Moreover, relationships between this signature and therapy responses were also identified. Functional enrichment analysis and immune cell estimation were also conducted to provide insights into the potential mechanisms of BC. Results. Epithelial-mesenchymal transition (EMT) was identified as the primary risk factor for the survival of BC patients (HR=1.43, 95% CI: 1.26-1.63). A novel EMT-related gene signature was constructed and validated in three independent cohorts, showing stable and accurate performance in predicting clinical outcomes. Furthermore, high-risk patients had poor prognoses and multivariable Cox regression analysis revealed this to be an independent risk factor for patient survival. CD8+ T cells, Tregs, and M2 macrophages were found abundantly in the tumor microenvironment of high-risk patients. Moreover, it was anticipated that high-risk patients would be more sensitive to chemotherapeutic drugs, while low-risk patients would benefit more from immunotherapy. Conclusions. We successfully identified and validated a novel EMT-related gene signature for predicting clinical outcomes and therapy responses in BC patients, which may be useful in clinical practice for risk stratification and individualized treatment.

Multidimensional characteristics, prognostic role, and preoperative prediction of peritoneal sarcomatosis in retroperitoneal sarcoma

Background

Peritoneal sarcomatosis (PS) could occur in patients with retroperitoneal sarcomas (RPS). This study aimed to expand the understanding of PS on its characteristics and prognostic role, and develop a nomogram to predict its occurrence preoperatively.

Methods

Data of 211 consecutive patients with RPS who underwent surgical treatment between 2011 and 2019 was retrospectively reviewed. First, the clinicopathological characteristics of PS were summarized and analyzed. Second, the disease-specific survival (DSS) and recurrence-free survival (RFS) of patients were analyzed to evaluate the prognostic role of PS. Third, preoperative imaging, nearly the only way to detect PS preoperatively, was combined with other screened risk factors to develop a nomogram. The performance of the nomogram was assessed.

Results

Among the 211 patients, 49 (23.2%) patients had PS with an incidence of 13.0% in the primary patients and 35.4% in the recurrent patients. The highest incidence of PS occurred in dedifferentiated liposarcoma (25.3%) and undifferentiated pleomorphic sarcoma (25.0%). The diagnostic sensitivity of the preoperative imaging was 71.4% and its specificity was 92.6%. The maximum standardized uptake value (SUVmax) was elevated in patients with PS (P&lt;0.001). IHC staining for liposarcoma revealed that the expression of VEGFR-2 was significantly higher in the PS group than that in the non-PS group (P = 0.008). Survival analysis (n =196) showed significantly worse DSS in the PS group than in non-PS group (median: 16.0 months vs. not reached, P &lt; 0.001). In addition, PS was proven as one of the most significant prognostic predictors of both DSS and RFS by random survival forest algorithm. A nomogram to predict PS status was developed based on preoperative imaging combined with four risk factors including the presentation status (primary vs. recurrent), ascites, SUVmax, and tumor size. The nomogram significantly improved the diagnostic sensitivity compared to preoperative imaging alone (44/49, 89.8% vs. 35/49, 71.4%). The C-statistics of the nomogram was 0.932, and similar C-statistics (0.886) was achieved at internal cross-validation.

Conclusion

PS is a significant prognostic indicator for RPS, and it occurs more often in recurrent RPS and in RPS with higher malignant tendency. The proposed nomogram is effective to predict PS preoperatively.

A novel hypoxia- and lactate metabolism-related signature to predict prognosis and immunotherapy responses for breast cancer by integrating machine learning and bioinformatic analyses

Background

Breast cancer is the most common cancer worldwide. Hypoxia and lactate metabolism are hallmarks of cancer. This study aimed to construct a novel hypoxia- and lactate metabolism-related gene signature to predict the survival, immune microenvironment, and treatment response of breast cancer patients.

Methods

RNA-seq and clinical data of breast cancer from The Cancer Genome Atlas database and Gene Expression Omnibus were downloaded. Hypoxia- and lactate metabolism-related genes were collected from publicly available data sources. The differentially expressed genes were identified using the “edgeR” R package. Univariate Cox regression, random survival forest (RSF), and stepwise multivariate Cox regression analyses were performed to construct the hypoxia-lactate metabolism-related prognostic model (HLMRPM). Further analyses, including functional enrichment, ESTIMATE, CIBERSORTx, Immune Cell Abundance Identifier (ImmuCellAI), TIDE, immunophenoscore (IPS), pRRophetic, and CellMiner, were performed to analyze immune status and treatment responses.

Results

We identified 181 differentially expressed hypoxia-lactate metabolism-related genes (HLMRGs), 24 of which were valuable prognostic genes. Using RSF and stepwise multivariate Cox regression analysis, five HLMRGs were included to establish the HLMRPM. According to the medium-risk score, patients were divided into high- and low-risk groups. Patients in the high-risk group had a worse prognosis than those in the low-risk group (P &lt; 0.05). A nomogram was further built to predict overall survival (OS). Functional enrichment analyses showed that the low-risk group was enriched with immune-related pathways, such as antigen processing and presentation and cytokine-cytokine receptor interaction, whereas the high-risk group was enriched in mTOR and Wnt signaling pathways. CIBERSORTx and ImmuCellAI showed that the low-risk group had abundant anti-tumor immune cells, whereas in the high-risk group, immunosuppressive cells were dominant. Independent immunotherapy datasets (IMvigor210 and GSE78220), TIDE, IPS and pRRophetic analyses revealed that the low-risk group responded better to common immunotherapy and chemotherapy drugs.

Conclusions

We constructed a novel prognostic signature combining lactate metabolism and hypoxia to predict OS, immune status, and treatment response of patients with breast cancer, providing a viewpoint for individualized treatment.

The Development of a Prediction Model Based on Random Survival Forest for the Postoperative Prognosis of Pancreatic Cancer: A SEER-Based Study

Simple Summary

Surgery is the main treatment to cure pancreatic cancer (PC). However, the 5-year survival rate of surgical resection is only 10–20%. The aim of our study was to develop a prediction model with the novel machine learning algorithm random survival forest (RSF) and to offer easy-to-use prediction tools, including risk stratification and individual prognosis. The study would benefit patients and physicians in postoperative management and facilitate personalized medicine.

Abstract

Accurate prediction for the prognosis of patients with pancreatic cancer (PC) is a emerge task nowadays. We aimed to develop survival models for postoperative PC patients, based on a novel algorithm, random survival forest (RSF), traditional Cox regression and neural networks (Deepsurv), using the Surveillance, Epidemiology, and End Results Program (SEER) database. A total of 3988 patients were included in this study. Eight clinicopathological features were selected using least absolute shrinkage and selection operator (LASSO) regression analysis and were utilized to develop the RSF model. The model was evaluated based on three dimensions: discrimination, calibration, and clinical benefit. It found that the RSF model predicted the cancer-specific survival (CSS) of the postoperative PC patients with a c-index of 0.723, which was higher than the models built by Cox regression (0.670) and Deepsurv (0.700). The Brier scores at 1, 3, and 5 years (0.188, 0.177, and 0.131) of the RSF model demonstrated the model’s favorable calibration and the decision curve analysis illustrated the model’s value of clinical implement. Moreover, the roles of the key variables were visualized in the Shapley Additive Explanations plotting. Lastly, the prediction model demonstrates value in risk stratification and individual prognosis. In this study, a high-performance prediction model for PC postoperative prognosis was developed, based on RSF The model presented significant strengths in the risk stratification and individual prognosis prediction.

FOXM1 Variant Contributes to Gefitinib Resistance via Activating Wnt/β-Catenin Signal Pathway in Patients with Non-Small Cell Lung Cancer

PURPOSE

Although gefitinib prolonged the progression-free survival (PFS) of patients with non-small cell lung cancer (NSCLC), unpredictable resistance limited its clinical efficacy. Novel predictive biomarkers with explicit mechanisms are urgently needed.

EXPERIMENTAL DESIGN

A total of 282 patients with NSCLC with gefitinib treatment were randomly assigned in a 7:3 ratio to exploratory (n = 192) and validation (n = 90) cohorts. The candidate polymorphisms were selected with Haploview4.2 in Hapmap and genotyped by a MassARRAY system, and the feature variables were identified through Randomforest Survival analysis. Tanswell and clonogenic assays, base editing and cell-derived tumor xenograft model were performed to uncover the underlying mechanism.

RESULTS

We found that the germline missense polymorphism rs3742076 (A>G, S628P), located in transactivation domain of FOXM1, was associated with PFS in exploratory (median PFS: GG vs. GA&AA, 9.20 vs. 13.37 months, P = 0.00039, HR = 2.399) and validation (median PFS: GG vs. GA&AA, 8.13 vs. 13.80 months, P = 0.048, HR = 2.628) cohorts. We elucidated that rs3742076_G conferred resistance to gefitinib by increasing protein stability of FOXM1 and facilitating an aggressive phenotype in vitro and in vivo through activating wnt/β-catenin signaling pathway. Meanwhile, FOXM1 level was highly associated with prognosis in patients with EGFR-mutant NSCLC. Mechanistically, FOXM1 rs3742076_G upregulated wnt/β-catenin activity by directly binding to β-catenin in cytoplasm and promoting transcription of β-catenin in nucleus. Remarkably, inhibition of β-catenin markedly reversed rs3742076_G-induced gefitinib resistance and aggressive phenotypes.

CONCLUSIONS

These findings characterized rs3742076_G as a gain-of-function polymorphism in mediating gefitinib resistance and tumor aggressiveness, and highlighted the variant as a predictive biomarker in guiding gefitinib treatment.

The cuproptosis-related signature associated with the tumor environment and prognosis of patients with glioma

Background

Copper ions are essential for cellular physiology. Cuproptosis is a novel method of copper-dependent cell death, and the cuproptosis-based signature for glioma remains less studied.

Methods

Several glioma datasets with clinicopathological information were collected from TCGA, GEO and CGGA. Robust Multichip Average (RMA) algorithm was used for background correction and normalization, cuproptosis-related genes (CRGs) were then collected. The TCGA-glioma cohort was clustered using ConsensusClusterPlus. Univariate Cox regression analysis and the Random Survival Forest model were performed on the differentially expressed genes to identify prognostic genes. The cuproptosis-signature was constructed by calculating CuproptosisScore using Multivariate Cox regression analysis. Differences in terms of genomic mutation, tumor microenvironment, and enrichment pathways were evaluated between high- or low-CuproptosisScore. Furthermore, drug response prediction was carried out utilizing pRRophetic.

Results

Two subclusters based on CRGs were identified. Patients in cluster2 had better clinical outcomes. The cuproptosis-signature was constructed based on CuproptosisScore. Patients with higher CuproptosisScore had higher WHO grades and worse prognosis, while patients with lower grades were more likely to develop IDH mutations or MGMT methylation. Univariate and Multivariate Cox regression analysis demonstrated CuproptosisScore was an independent prognostic factor. The accuracy of the signature in prognostic prediction was further confirmed in 11 external validation datasets. In groups with high-CuproptosisScore, PIK3CA, MUC16, NF1, TTN, TP53, PTEN, and EGFR showed high mutation frequency. IDH1, TP53, ATRX, CIC, and FUBP1 demonstrated high mutation frequency in low-CuproptosisScore group. The level of immune infiltration increased as CuproptosisScore increased. SubMap analysis revealed patients with high-CuproptosisScore may respond to anti-PD-1 therapy. The IC50 values of Bexarotene, Bicalutamide, Bortezomib, and Cytarabine were lower in the high-CuproptosisScore group than those in the low-CuproptosisScore group. Finally, the importance of IGFBP2 in TCGA-glioma cohort was confirmed.

Conclusion

The current study revealed the novel cuproptosis-based signature might help predict the prognosis, biological features, and appropriate treatment for patients with glioma.

A novel deep learning prognostic system improves survival predictions for stage III non-small cell lung cancer

Background

Accurate prognostic prediction plays a crucial role in the clinical setting. However, the TNM staging system fails to provide satisfactory individual survival prediction for stage III non‐small cell lung cancer (NSCLC). The performance of the deep learning network for survival prediction in stage III NSCLC has not been explored.

Objectives

This study aimed to develop a deep learning‐based prognostic system that could achieve better predictive performance than the existing staging system for stage III NSCLC.

Methods

In this study, a deep survival learning model (DSLM) for stage III NSCLC was developed based on the Surveillance, Epidemiology, and End Results (SEER) database and was independently tested with another external cohort from our institute. DSLM was compared with the Cox proportional hazard (CPH) and random survival forest (RSF) models. A new prognostic system for stage III NSCLC was also proposed based on the established deep learning model.

Results

The study included 16,613 patients with stage III NSCLC from the SEER database. DSLM showed the best performance in survival prediction, with a C‐index of 0.725 in the validation set, followed by RSF (0.688) and CPH (0.683). DSLM also showed C‐indices of 0.719 and 0.665 in the internal and real‐world external testing datasets, respectively. In addition, the new prognostic system based on DSLM (AUROC = 0.744) showed better performance than the TNM staging system (AUROC = 0.561).

Conclusion

In this study, a new, integrated deep learning‐based prognostic model was developed and evaluated for stage III NSCLC. This novel approach may be valuable in improving patient stratification and potentially provide meaningful prognostic information that contributes to personalized therapy.