Prediction Of Clinical Prognosis Research Articles

Comprehensive analysis of diverse programmed cell death patterns in the prognosis, tumor microenvironment and drug sensitivity in hepatocellular carcinoma.

Treatment failure in patients with liver hepatocellular carcinoma (LIHC) is primarily caused by tumor progression and therapy resistance. Tumor immunity plays a crucial role in regulating the homeostasis of cells through the process of programmed cell death (PCD). However, the expression profile and clinical significance of PCD-related genes in LIHC require further investigation. In this study, we analyzed twelve commonly observed PCD patterns to construct a prognostic model. We collected RNA-seq data, genomics, and clinical information from TCGA-LIHC and GSE14520 cohorts to validate the prognostic gene signature. We discovered 75 PCD-related differentially expressed genes (DEGs) with prognostic significance in LIHC. Using these genes, we constructed a PCD-related score (PCDscore) with an 11-gene signature through LASSO COX regression analysis. Validation in the GSE14520 cohort demonstrated that LIHC patients with high PCDscore had poorer prognoses. Unsupervised clustering based on the 11 model genes revealed 3 molecular subtypes of LIHC with distinct prognoses. By incorporating PCDscore with clinical features, we constructed a highly predictive nomogram. Additionally, PCDscore was correlated with immune checkpoint genes and immune cell infiltration. LIHC patients with high PCDscore exhibited sensitivity to common chemotherapy drugs (such as cisplatin and docetaxel). To summarize, our study developed a novel PCDscore model that comprehensively analyzed different cell death modes, providing an accurate prediction of clinical prognosis and drug sensitivity for LIHC patients.

Identification of circRNA-lncRNA-miRNA-mRNA competitive endogenous RNA networks as prognostic bio-markers in head and neck squamous cell carcinoma.

Head and neck squamous cell carcinoma (HNSCC) is one of the most prevalent neoplasms of malignancy. The pathogenesis of HNSCC is involved in the change of gene expression and the abnormal interaction between genes. The purpose of our research is to find genes that affect the survival time of patients and construct circRNA-lncRNA-miRNA-mRNA network. This paper aims to reveal the development mechanism of HNSCC and provide new ideas for clinical prognosis prediction. Transcriptome data were obtained from TCGA and GEO databases. A total of 556 samples were included. We screened the differentially expressed lncRNAs, circRNAs and mRNAs to predict interactions between lncRNA-miRNA, circRNA-miRNA and miRNA-mRNA. Then, we selected the hub mRNAs to predict the prognosis of patients. We identified 8 hub prognostic mRNAs (FAM111A, IFIT2, CAV1, KLF9, OCIAD2, SLCO3A1, S100A16 and APOL3) by a series of bioinformatics analyses and established the prognostic ceRNA network of 8 mRNAs, 3 miRNAs, 2 circRNAs and 1 lncRNAs according to the targeting relationship by using databases. We established the circRNA-lncRNA-miRNA-mRNA gene interaction network in HNSCC. We illuminated the molecular mechanism underlying the gene regulation associated with the pathogenesis of HNSCC and predicted the biomarkers related to prognosis.

A novel prognostic N7-methylguanosine-related long non-coding RNA signature in clear cell renal cell carcinoma

Clear cell renal cell carcinoma (ccRCC) is regulated by methylation modifications and long noncoding RNAs (lncRNAs). However, knowledge of N7-methylguanosine (m7G)-related lncRNAs that predict ccRCC prognosis remains insufficient. A prognostic multi-lncRNA signature was created using LASSO regression to examine the differential expression of m7G-related lncRNAs in ccRCC. Furthermore, we performed Kaplan–Meier analysis and area under the curve (AUC) analysis for diagnosis. In all, a model based on five lncRNAs was developed. Principal component analysis (PCA) indicated that the risk model precisely separated the patients into different groups. The IC50 value for drug sensitivity divided patients into two risk groups. High-risk group of patients was more susceptible to A.443654, A.770041, ABT.888, AMG.706, and AZ628. Moreover, a lower tumor mutation burden combined with low-risk scores was associated with a better prognosis of ccRCC. Quantitative real-time polymerase chain reaction (qRT-PCR) exhibited that the expression levels of LINC01507, AC093278.2 were very high in all five ccRCC cell lines, AC084876.1 was upregulated in all ccRCC cell lines except 786-O, and the levels of AL118508.1 and DUXAP8 were upregulated in the Caki-1 cell line. This risk model may be promising for the clinical prediction of prognosis and immunotherapeutic responses in patients with ccRCC.

Biomarker of Pulmonary Inflammatory Response in LUAD: miR-584-5p Targets RAB23 to Suppress Inflammation Induced by LPS in A549 Cells

Background: Pulmonary inflammatory response (PIR) is one of the prognostic risk factors of lung adenocarcinoma (LUAD), with a high mortality rate. Objective: This study aims to investigate prognostic microRNA (miRNA) to improve clinical prognosis prediction and postoperative inflammation treatment in LUAD patients. Methods: About 201 differentially expressed microRNAs (DE-miRNAs) in LUAD were mined by differential analysis. Univariate/multivariate Cox analyses established and validated prognostic risk miRNAs in TCGA-LUAD. KEGG and GO were used to link risk signatures and biological functions. After 48 hours of exposure to 50 ng/mL LPS, the miR-584-5p/RAB23 regulatory network was verified in qRT-PCR, Western Blotting, and the Luciferase Reporter Assay in A549 cells. Results: MiR-584-5p and miR-101-3p were validated as riskscore correlated with LUAD patients’ 1-year survival (p < 0.001) and participate in multiple inflammation-related pathways. RAB23, a RAS oncogene, is involved in inflammatory MAPK signaling. Evidence suggests that miR-584-5p regulates inflammation in LUAD by targeting RAB23. A549 cells were transfected with the mimic and inhibitor of miR-584-5p, confirming the negative regulatory relationship between miR-584-5p and RAB23. In the A549 induced by LPS, either over-expression of miR-584-5p or knock-down of RAB23 expression decreased the expression of inflammatory factors and increased cell viability. Conclusion: Prognostic-related risk miR-584-5p can regulate the expression of RAB23 at both the mRNA and protein levels, thereby influencing the development of a PIR in LUAD. This will have significant implications for the clinical prognosis prediction and therapy decision-making of LUAD patients with PIR.

Comprehensive analysis of the RSK gene family in acute myeloid leukemia determines a prognostic signature for the prediction of clinical prognosis and treatment responses

ABSTRACTObjectiveThe prognosis of acute myeloid leukemia (AML) remains poor although the basic and translational research has been highly productive in understanding the genetics and pathopoiesis of AML and a plethora of targeted therapies have been developed. Consequently, it is crucial to deepen the knowledge of molecular pathogenesis underlying AML for the advancement of new treatment options.MethodA RSK gene family-related signature was constructed to investigate whether RSK gene family members were useful in predicting the prognosis of AML patients. The relationship between the RSK gene family-related signature and the infiltration of immune cells was further assessed using the CIBERSORT algorithm. The ‘oncoPredict’ package was used to analyze relationships between the RSK gene family-related signature and the sensitivity to drugs or small molecules.ResultsPatients were classified into two groups using the RSK gene family-related signature following the median risk score. Overall survival (OS) was significantly longer in patients with low-risk scores than that in patients with high-risk scores as showed by both training and validation datasets. Moreover, the signature was helpful in predicting 1-year, 3-year, and 5-year OS in training and validation datasets. In addition, it was identified that low-risk patients exhibited greater sensitivity to 20 drugs or small molecules and that high-risk patients had higher sensitivity to 38 drugs or small molecules.ConclusionRSK gene family members, particularly RPS6KA1 and RPS6KA4, may help to predict prognosis for AML patients. Furthermore, RPS6KA1 may serve as a novel drug target for AML.

Construction of a risk model and prediction of prognosis and immunotherapy based on cuproptosis-related LncRNAs in the urinary system pan-cancer

BackgroundUrinary pan-cancer system is a general term for tumors of the urinary system including renal cell carcinoma (RCC), prostate cancer (PRAD), and bladder cancer (BLCA). Their location, physiological functions, and metabolism are closely related, making the occurrence and outcome of these tumors highly similar. Cuproptosis is a new type of cell death that is different from apoptosis and plays an essential role in tumors. Therefore, it is necessary to study the molecular mechanism of cuproptosis-related lncRNAs to urinary system pan-cancer for the prognosis, clinical diagnosis, and treatment of urinary tumors.MethodIn our study, we identified 35 co-expression cuproptosis-related lncRNAs (CRLs) from the urinary pan-cancer system. 28 CRLs were identified as prognostic-related CRLs by univariate Cox regression analysis. Then 12 CRLs were obtained using lasso regression and multivariate cox analysis to construct a prognostic model. We divided patients into high- and low-risk groups based on the median risk scores. Next, Kaplan–Meier analysis, principal component analysis (PCA), functional rich annotations, and nomogram were used to compare the differences between the high- and low-risk groups. Finally, the prediction of tumor immune dysfunction and rejection, gene mutation, and drug sensitivity were discussed.ConclusionFinally, the candidate molecules of the urinary system pan-cancer were identified. This CRLs risk model may be promising for clinical prediction of prognosis and immunotherapy response in urinary system pan-cancer patients.

Use of Real-World Data in Population Science to Improve the Prevention and Care of Diabetes-Related Outcomes.

The past decade of population research for diabetes has seen a dramatic proliferation of the use of real-world data (RWD) and real-world evidence (RWE) generation from non-research settings, including both health and non-health sources, to influence decisions related to optimal diabetes care. A common attribute of these new data is that they were not collected for research purposes yet have the potential to enrich the information around the characteristics of individuals, risk factors, interventions, and health effects. This has expanded the role of subdisciplines like comparative effectiveness research and precision medicine, new quasi-experimental study designs, new research platforms like distributed data networks, and new analytic approaches for clinical prediction of prognosis or treatment response. The result of these developments is a greater potential to progress diabetes treatment and prevention through the increasing range of populations, interventions, outcomes, and settings that can be efficiently examined. However, this proliferation also carries an increased threat of bias and misleading findings. The level of evidence that may be derived from RWD is ultimately a function of the data quality and the rigorous application of study design and analysis. This report reviews the current landscape and applications of RWD in clinical effectiveness and population health research for diabetes and summarizes opportunities and best practices in the conduct, reporting, and dissemination of RWD to optimize its value and limit its drawbacks.

LncRNA PCBP1-AS1 mediated downregulation of ITGAL as a prognostic biomarker in lung adenocarcinoma.

Integrin alpha L (ITGAL) seemed to play a critical role in carcinogenesis and immune regulation. Nevertheless, the effects of ITGAL on non-small cell lung cancer (NSCLC) remain elusive. The present paper intended to determine the effects of ITGAL in NSCLC via the integration of bioinformatic analyses. In this study, we found that the mRNA and protein levels of ITGAL were downregulated in NSCLC tissues. Significantly, low ITGAL expression was related to poorer prognosis and increased malignancy of NSCLC. In addition, GO analysis and KEGG pathway analysis revealed that the coexpressed genes of ITGAL were predominantly associated with various immune-associated signaling pathways, like the T cell receptor signaling pathway, Th17 cell differentiation, chemokine signaling pathway, and NF-κB signaling pathway. Our result indicated that lncRNA-mediated downregulation of integrin alpha L expression was tightly related to immunocyte infiltration, immune modulators, and chemotactic factors in NSCLC, which potentially serves as a biomarker for clinical prognosis prediction and immunotherapy of NSCLC. This is the first comprehensive analysis of ITGAL in the prognosis, immune microenvironment, and immunotherapy of lung adenocarcinoma. ITGAL are promising biomarkers for predicting clinical outcomes and immunotherapy responses in patients with NSCLC.

PI3K pathway mutation predicts an activated immune microenvironment and better immunotherapeutic efficacy in head and neck squamous cell carcinoma

BackgroundPI3K pathway is the most frequently mutated pathway in head and neck squamous cell carcinoma (HNSC), which plays a crucial role in tumorigenesis and progression. In the present study, we aimed to investigate the role of PI3K pathway mutation in clinical prognosis prediction and the relationship with immune microenvironment and response rate to immunotherapy.MethodsWe collected 129 samples with immunotherapy information from MSKCC-2019 cohort as well as 501 and 40 samples from TCGA-HNSC and MD-Anderson non-immunotherapy cohorts, respectively. Somatic mutation data was utilized to characterize the mutational status of the PI3K pathway. Subsequently, we further analyzed the differences in prognosis, immunotherapy response, genomic alterations, functional characteristics, and immune microenvironment between the mutation and wild groups.ResultsThe Kaplan-Meier survival curves displayed that PI3K pathway mutation predicted observably prolonged overall survival (OS) in the immunotherapy cohort MSKCC-2019 (p = 0.012) but did not reach statistical significance in the non-immunotherapy cohorts TCGA-HNSC (p = 0.68) and MD-Anderson (p = 0.68). After incorporating several clinicopathologic features such as age, gender, and tumor mutation burden (TMB), the results of multivariate Cox regression analysis also demonstrated that the PI3K pathway mutation could indicate better immunotherapy outcomes in HNSC patients with a hazard ratio (HR) of 0.533 (95% CI: 0.313–0.910; p = 0.021) in the immunotherapy cohort MSKCC-2019, compared with 0.888 (95% CI: 0.636–1.241; p = 0.487) and 1.939 (95% CI: 0.483–7.781; p = 0.351) in the non-immunotherapy cohorts TCGA-HNSC and MD-Anderson. In addition, the results of the subclass mapping (SubMap) and the tumor immune dysfunction and exclusion (TIDE) also consistently suggested that patients in the mutation group are more likely to benefit from immunotherapy. And further studies showed that the mutation group owned significantly higher TMB, activated immune-related pathways, richer abundance of immune cells, and higher expression levels of immunomodulators. To improve the prognosis of the wild group, we identified five relatively sensitive potential drugs for the wild group, including “BMS-536924,” “linsitinib,” “NVP-TAE684,” “PLX-4720,” and “clonazepam.”ConclusionsThe PI3K pathway mutation status could be considered as a potential biomarker to predict better immunotherapeutic efficacy and clinical outcomes after immunotherapy in HNSC patients.

Pan-cancer analysis: predictive role of TAP1 in cancer prognosis and response to immunotherapy

BackgroundTransporter associated with antigen processing 1 (TAP1) is a molecule involved in processing and presentation of major histocompatibility complex class I restricted antigens, including tumor-associated antigens. TAP1 participates in tumor immunity, and is aberrantly expressed in multiple cancer types;MethodsTranscriptome profiles were obtained from The Cancer Genome Atlas and Genotype-Tissue Expression databases. Genetic alterations, protein distribution, and interaction information for TAP1 were downloaded from cBioPortal, Human Protein Atlas and Compartmentalized Protein–Protein Interaction, respectively. Single-cell analyses of TAP1 across cancers were conducted via the Tumor Immune Single-cell Hub website. Gene set enrichment analysis was employed to investigate TAP1-associated functional mechanisms and processes. Immune cell infiltration was explored using Tumor Immune Estimation Resource 2.0. Pan-cancer correlations between TAP1 expression and immunotherapy biomarkers were explored using the Spearman’s correlation test. Associations with immunotherapy responses were also investigated using clinicopathological and prognostic information from cohorts of patients with cancer receiving immune checkpoint inhibitors.ResultsTAP1 expression was elevated in most cancer types and exhibited distinct prognostic value. Immune cells expressed more TAP1 than malignant cells within most tumors. TAP1 expression was significantly correlated with immune-related pathways, T-lymphocyte infiltration, and immunotherapeutic biomarkers. Clinical cohort validation revealed a significant correlation with immune therapeutic effects and verified the prognostic role of TAP1 in immunotherapy. Western blot assay indicated that TAP1 is upregulated in glioblastoma compared with adjacent normal brain tissues.ConclusionTAP1 is a robust tumor prognostic biomarker and a novel predictor of clinical prognosis and immunotherapeutic responses in various cancer types.

Development and validation of a novel ferroptosis‑related lncRNA prognostic signature for pancreatic adenocarcinoma.

Long non‑coding RNAs (lncRNAs) serve a pivotal role in the regulation of cancer cell ferroptosis. However, the prognostic value of ferroptosis‑related lncRNAs in pancreatic adenocarcinoma (PAAD) largely remains unclear. We aimed at constructing a lncRNA‑based signature to improve the prognosis prediction of PAAD. In the present study, the transcriptome profiling data and clinical information of patients with PAAD were obtained from The Cancer Genome Atlas (TCGA) and International Cancer Gene Consortium (ICGC) databases. Univariate Cox regression analysis of the TCGA cohort demonstrated that 26 ferroptosis‑related lncRNAs had significant prognostic value for PAAD (all P<0.01). Least absolute shrinkage and selection operator regression and multivariate Cox proportional hazards regression analyses were performed to construct a prognostic ferroptosis‑related lncRNA signature (FRLS) comprising nine ferroptosis‑related lncRNAs. The efficacy of this FRLS was verified in the training (TCGA) and validation (ICGC) cohorts. Based on the risk model, high risk scores were significantly correlated with poor overall survival (OS) (hazard ratio, 1.314; 95% confidence interval, 1.218‑1.418; P<0.001). The receiver operating characteristic curves and principal component analysis further demonstrated the robust prognostic ability of the FRLS. Furthermore, a nomogram with favorable predictive efficacy for the prediction of OS was constructed based on the FRLS and clinical features. Gene set enrichment analysis demonstrated that the genes in the FRLS participated in a number of cancer‑associated immunoregulatory pathways. Importantly, it was demonstrated that immune infiltration and response to cancer immunotherapy differed significantly between the high and low‑risk groups according to the FRLS. In conclusion, the risk signature based on the FRLS has potential for the clinical prediction of prognosis and immunotherapy response in patients with PAAD.

Identification of a prognostic six-immune-gene signature and a nomogram model for uveal melanoma

BackgroundTo identify an immune-related prognostic signature and find potential therapeutic targets for uveal melanoma.MethodsThe RNA-sequencing data obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets. The prognostic six-immune-gene signature was constructed through least absolute shrinkage and selection operator and multi-variate Cox regression analyses. Functional enrichment analysis and single sample GSEA were carried out. In addition, a nomogram model established by integrating clinical variables and this signature risk score was also constructed and evaluated.ResultsWe obtained 130 prognostic immune genes, and six of them were selected to construct a prognostic signature in the TCGA uveal melanoma dataset. Patients were classified into high-risk and low-risk groups according to a median risk score of this signature. High-risk group patients had poorer overall survival in comparison to the patients in the low-risk group (p < 0.001). These findings were further validated in two external GEO datasets. A nomogram model proved to be a good classifier for uveal melanoma by combining this signature. Both functional enrichment analysis and single sample GSEA analysis verified that this signature was truly correlated with immune system. In addition, in vitro cell experiments results demonstrated the consistent trend of our computational findings.ConclusionOur newly identified six-immune-gene signature and a nomogram model could be used as meaningful prognostic biomarkers, which might provide uveal melanoma patients with individualized clinical prognosis prediction and potential novel treatment targets.

Retracted: Identification of an m6A RNA Methylation Regulator Risk Score Model for Prediction of Clinical Prognosis in Astrocytoma

Retracted: Identification of an m6A RNA Methylation Regulator Risk Score Model for Prediction of Clinical Prognosis in Astrocytoma

A Prognostic Model for Predicting Liver Cancer Patients Based On Immune Checkpoint Gene-Related Basement Membrane Genes, And Analyzing Immunity and Potential Drug Candidates

Abstract:&#x0D; Objective: Hepatocellular carcinoma is one of the most common malignant tumors in the world. The expression of immune checkpoint genes in tumor cells prevents the immune system from eliminating tumors. Basement membrane-related genes are genes that are closely related to human diseases obtained in the latest research.&#x0D; Methods: First, basement membrane-related genes were extracted from immune checkpoint genes, and a prognosis model of immune checkpoint-related basement membrane genes was constructed. C-index curves and ROC were drawn by survival analysis, progression-free survival analysis, and independent prognostic analysis. Curves, principal component analysis, and validation of the clinical grouping model were performed to verify its accuracy, enrichment analysis, immune analysis, and tumor mutation burden survival analysis were performed to further explore the potential functions of the model, and finally, potential drugs targeting the model were discussed.&#x0D; Results: A prognostic model for predicting the survival time of liver cancer patients was constructed, and the predictive ability of the model was verified. GO and KEGG enrichment analysis revealed differences in the functions and pathways of differential genes. Four differentially expressed immune functions were found. The top 4 genes mutated in the high and low risk groups were compared. Twenty-five drugs with significant differences in drug sensitivity between high- and low-risk groups were explored.&#x0D; Conclusion: The risk-prognostic model based on the association of basement membrane genes and immune checkpoint genes in this study may be promising for clinical prediction of prognosis and immunotherapy response in patients with liver cancer.

Exploration of the role of Cuproptosis genes and their related long non-coding RNA in clear cell renal cell carcinoma: a comprehensive bioinformatics study

Clear cell renal cell carcinoma is a common malignant tumor of the urinary system. The mechanism of its occurrence and development is unknown, and there is currently few effective comprehensive predictive markers for prognosis and treatment response. With the discovery of a new cell death process – cuproptosis drew the attention of researchers. We constructed a model for the prediction of clinical prognosis and immunotherapy response through integrative analysis of gene expression datasets from KIRC samples in The Cancer Genome Atlas (TCGA) database. During the course of the study, we found that cuproptosis genes are significantly differentially expressed between clear cell renal cell carcinoma samples and normal samples. Based on this, we put forward the prognostic model for cuproptosis gene related-long non-coding RNA. And through various statistic and external independent cohorts, we proved that the model is accurate and stable, worthy of clinical application and further exploration and validation.

Identification of tryptophan metabolic gene-related subtypes, development of prognostic models, and characterization of tumor microenvironment infiltration in gliomas.

Epigenetic regulation and immunotherapy of tumor microenvironment (TME) is a hot topic in recent years. However, the potential value of tryptophan metabolism genes in regulating TME and immunotherapy is still unclear. A comprehensive study of glioma patients was carried out based on 40 tryptophan metabolic genes. Subsequently, these prognostic tryptophan metabolic genes are systematically associated with immunological characteristics and immunotherapy. A risk score model was constructed and verified in the Cancer Genome Atlas (TCGA) and the Chinese Glioma Genome Atlas (CGGA) cohorts to provide guidance for prognosis prediction and immunotherapy of glioma patients. We described the changes of tryptophan metabolism genes in 966 glioma samples from genetic and transcriptional fields and evaluated their expression patterns from two independent data sets. We identified two different molecular subtypes and found that two subtypes were associated with clinicopathological features, prognosis, TME cell infiltration, and immune checkpoint blockers (ICBs). Then, four genes (IL4I1, CYP1A1, OGDHL, and ASMT) were screened out by univariate and multivariate cox regression analysis of tryptophan metabolism genes, and a risk score model for predicting the overall survival (OS) of glioma patients was constructed. And its predictive ability is verified using the CGGA database. At the same time, we verified the expression of IL4I1, CYP1A1, OGDHL, and ASMT four genes in glioma specimens and cell lines in GES4260 and GES15824. Therefore, we constructed a nomogram to improve the clinical applicability of the risk assessment model. The high risk score group, characterized by increased TMB and immune cell infiltration, was also sensitive to temozolomide immunotherapy. Our comprehensive analysis of tryptophan metabolic genes in gliomas shows that they play a potential role in tumor immune stromal microenvironment, clinicopathological features, and prognosis. Tryptophan metabolism genes play an indispensable role in the complexity, diversity, and prognosis of TME. This risk score model based on tryptophan metabolism gene is a new predictor of clinical prognosis and immunotherapy response of glioma, and guides a more appropriate immunotherapy strategy for glioma patients.

Exploring risk factors for cervical lymph node metastasis in papillary thyroid microcarcinoma: construction of a novel population-based predictive model

BackgroundMachine learning was a highly effective tool in model construction. We aim to establish a machine learning-based predictive model for predicting the cervical lymph node metastasis (LNM) in papillary thyroid microcarcinoma (PTMC).MethodsWe obtained data on PTMC from the SEER database, including 10 demographic and clinicopathological characteristics. Univariate and multivariate logistic regression (LR) analyses were applied to screen the risk factors for cervical LNM in PTMC. Risk factors with P < 0.05 in multivariate LR analysis were used as modeling variables. Five different machine learning (ML) algorithms including extreme gradient boosting (XGBoost), random forest (RF), adaptive boosting (AdaBoost), gaussian naive bayes (GNB) and multi-layer perceptron (MLP) and traditional regression analysis were used to construct the prediction model. Finally, the area under the receiver operating characteristic (AUROC) curve was used to compare the model performance.ResultsThrough univariate and multivariate LR analysis, we screened out 9 independent risk factors most closely associated with cervical LNM in PTMC, including age, sex, race, marital status, region, histology, tumor size, and extrathyroidal extension (ETE) and multifocality. We used these risk factors to build an ML prediction model, in which the AUROC value of the XGBoost algorithm was higher than the other 4 ML algorithms and was the best ML model. We optimized the XGBoost algorithm through 10-fold cross-validation, and its best performance on the training set (AUROC: 0.809, 95%CI 0.800–0.818) was better than traditional LR analysis (AUROC: 0.780, 95%CI 0.772–0.787).ConclusionsML algorithms have good predictive performance, especially the XGBoost algorithm. With the continuous development of artificial intelligence, ML algorithms have broad prospects in clinical prognosis prediction.

A novel nomogram based on the prognostic nutritional index for predicting postoperative outcomes in patients with stage I-III gastric cancer undergoing robotic radical gastrectomy.

The inflammation and nutrition status are crucial factors influencing the outcome of patients with gastric cancer. This study aims to investigate the prognostic value of the preoperative prognostic nutritional index (PNI) in patients with stage I-III gastric cancer undergoing robotic radical gastrectomy combined with Enhanced Recovery after Surgery (ERAS), and further to create a clinical prognosis prediction model. 525 patients with stage I-III gastric cancer who underwent ERAS combined with RRG from July 2010 to June 2018 were included in this work, and were divided randomly into training and validating groups in a 7-to-3 ratio. The association between PNI and overall survival (OS) was assessed by Kaplan-Meier analysis and the log-rank test. Independent risk factors impacting postoperative survival were analyzed with the Cox proportional hazards regression model. A nomogram for predicting OS was constructed based on multivariate analysis, and its predictive performance was evaluated using Harrell's concordance index (C-index), calibration plots, ROC curve, decision curve analysis (DCA), and time-dependent ROC curve analysis. Survival analyses revealed the presence of a significant correlation between low preoperative PNI and shortened postoperative survival (P = 0.001). According to multivariate analysis, postoperative complications (P < 0.001), pTNM stage (II: P = 0.007; III: P < 0.001), PNI (P = 0.048) and lymph node ratio (LNR) (P = 0.003) were independent prognostic factors in patients undergoing ERAS combined with RRG. The nomogram constructed based on PNI, pTNM stage, complications, and LNR was superior to the pTNM stage model in terms of predictive performance. The C-indexes of the nomogram model were respectively 0.765 and 0.754 in the training and testing set, while AUC values for 1-year, 3-year, and 5-year OS were 0.68, 0.71, and 0.74 in the training set and 0.60, 0.67, and 0.72 in the validation set. Preoperative PNI is an independent prognostic factor for patients with stage I-III gastric cancer undergoing ERAS combined with robotic radical gastrectomy. Based on PNI, we constructed a nomogram for predicting postoperative outcomes of gastric cancer patients, which might be utilized clinically.

A novel prognostic model based on urea cycle-related gene signature for colorectal cancer.

Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in the world. This study aimed to develop a urea cycle (UC)-related gene signature that provides a theoretical foundation for the prognosis and treatment of patients with CRC. Differentially expressed UC-related genes in CRC were confirmed using differential analysis and Venn diagrams. Univariate Cox and least absolute shrinkage and selection operator regression analyses were performed to identify UC-related prognostic genes. A UC-related signature was created and confirmed using distinct datasets. Independent prognostic predictors were authenticated using Cox analysis. The Cell-type Identification by Estimating Relative Subsets of RNA Transcripts algorithm and Spearman method were applied to probe the linkage between UC-related prognostic genes and tumor immune-infiltrating cells. The Human Protein Atlas database was used to determine the protein expression levels of prognostic genes in CRC and normal tissues. Verification of the expression levels of UC-related prognostic genes in clinical tissue samples was performed using real-time quantitative polymerase chain reaction (qPCR). A total of 49 DEUCRGs in CRC were mined. Eight prognostic genes (TIMP1, FABP4, MMP3, MMP1, CD177, CA2, S100P, and SPP1) were identified to construct a UC-related gene signature. The signature was then affirmed using an external validation set. The risk score was demonstrated to be a credible independent prognostic predictor using Cox regression analysis. Functional enrichment analysis revealed that focal adhesion, ECM-receptor interaction, IL-17 signaling pathway, and nitrogen metabolism were associated with the UC-related gene signature. Immune infiltration and correlation analyses revealed a significant correlation between UC-related prognostic genes and differential immune cells between the two risk subgroups. Finally, the qPCR results of clinical samples further confirmed the results of the public database. Taken together, this study authenticated UC-related prognostic genes and developed a gene signature for the prognosis of CRC, which will be of great significance in the identification of prognostic molecular biomarkers, clinical prognosis prediction, and development of treatment strategies for patients with CRC.

The development of a novel signature based on the m6A RNA methylation regulator-related ceRNA network to predict prognosis and therapy response in sarcomas.

Background: N6 methyladenosine (m6A)-related noncoding RNAs (including lncRNAs and miRNAs) are closely related to the development of cancer. However, the gene signature and prognostic value of m6A regulators and m6A-associated RNAs in regulating sarcoma (SARC) development and progression remain largely unexplored. Therefore, further research is required. Methods: We obtained expression data for RNA sequencing (RNA-seq) and miRNAs of SARC from The Cancer Genome Atlas (TCGA) datasets. Correlation analysis and two target gene prediction databases (miRTarBase and LncBase v.2) were used to deduce m6A-related miRNAs and lncRNAs, and Cytoscape software was used to construct ceRNA-regulating networks. Based on univariate Cox regression and least absolute shrinkage and selection operator (LASSO) Cox regression analyses, an m6A-associated RNA risk signature (m6Ascore) model was established. Prognostic differences between subgroups were explored using Kaplan–Meier (KM) analysis. Risk score-related biological phenotypes were analyzed in terms of functional enrichment, tumor immune signature, and tumor mutation signature. Finally, potential immunotherapy features and drug sensitivity predictions for this model were also discussed. Results: A total of 16 miRNAs, 104 lncRNAs, and 11 mRNAs were incorporated into the ceRNA network. The risk score was obtained based on RP11-283I3.6, hsa-miR-455-3p, and CBLL1. Patients were divided into two risk groups using the risk score, with patients in the low-risk group having longer overall survival (OS) than those in the high-risk group. The receiver operating characteristic (ROC) curves indicated that risk characteristic performed well in predicting the prognosis of patients with SARC. In addition, lower m6Ascore was also positively correlated with the abundance of immune cells such as monocytes and mast cells activated, and several immune checkpoint genes were highly expressed in the low-m6Ascore group. According to our analysis, lower m6Ascore may lead to better immunotherapy response and OS outcomes. The risk signature was significantly associated with the chemosensitivity of SARC. Finally, a nomogram was constructed to predict the OS in patients with SARC. The concordance index (C-index) for the nomogram was 0.744 (95% CI: 0.707–0.784). The decision curve analysis (DCA), calibration plot, and ROC curve all showed that this nomogram had good predictive performance. Conclusion: This m6Ascore risk model based on m6A RNA methylation regulator-related RNAs may be promising for clinical prediction of prognosis and might contain potential biomarkers for treatment response prediction for SARC patients.