Prognosis Prediction In Lung Adenocarcinoma Research Articles

Combination of genomic instability score and TP53 status for prognosis prediction in lung adenocarcinoma

The genomic instability (GI) /homologous recombination deficiency (HRD) score, calculated as the sum of the events of loss of heterozygosity (LOH), large-scale state transition (LST) and telomere allele imbalance (TAI), is used to guide the choice of treatment in several cancers, but its relationship with genomic features, clinicopathological characteristics and prognosis in lung cancer is poorly understood, which could lead to population bias in prospective studies. We retrospectively analyzed 1011 lung cancer patients whose tumor samples were successfully profiled by high-throughput sequencing panel including GI/HRD score. Alterations of many cancer suppressor genes were associated with higher GI/HRD scores, biallelic inactivation of TP53 was correlated with a high GI/HRD score. A combination of two gene alterations exhibited a higher GI/HRD scores than single gene alterations. The GI/HRD score was associated with advanced stages in lung adenocarcinoma but not in lung squamous cell carcinoma. Furthermore, patients with higher GI/HRD scores had significantly shorter overall survival and progression-free survival than patients with lower GI/HRD scores. Finally, patients with a combination of a higher GI/HRD scores and TP53 alteration exhibited an extremely poor prognosis compared with patients with a lower GI/HRD scores and wild-type TP53 (overall survival, training cohort, hazard ratio (HR) = 8.56, P < 0.001; validation cohort, HR = 6.47, P < 0.001; progression-free survival, HR = 4.76, P < 0.001). Our study revealed the prognostic value of the GI/HRD score in lung adenocarcinoma, but not for all lung cancer. Moreover, the combination of the GI/HRD score and TP53 status could be a promising strategy to predict the prognosis of patients with lung adenocarcinoma.

Optimal 18F-FDG PET/CT radiomics model development for predicting EGFR mutation status and prognosis in lung adenocarcinoma: a multicentric study.

To develop and interpret optimal predictive models to identify epidermal growth factor receptor (EGFR) mutation status and subtypes in patients with lung adenocarcinoma based on multicentric 18F-FDG PET/CT data, and further construct a prognostic model to predict their clinical outcome. The 18F-FDG PET/CT imaging and clinical characters of 767 patients with lung adenocarcinoma from 4 cohorts were collected. Seventy-six radiomics candidates using cross-combination method to identity EGFR mutation status and subtypes were built. Further, Shapley additive explanations and local interpretable model-agnostic explanations were used for optimal models' interpretation. Moreover, in order to predict the overall survival, a multivariate Cox proportional hazard model based on handcrafted radiomics features and clinical characteristics was constructed. The predictive performance and clinical net benefit of the models were evaluated via area under receiver operating characteristic (AUC), C-index and decision curve analysis. Among the 76 radiomics candidates, light gradient boosting machine classifier (LGBM) combined with recursive feature elimination wrapped LGBM feature selection method achieved best performance in predicting EGFR mutation status (AUC reached 0.80, 0.61, 0.71 in the internal test cohort and two external test cohorts, respectively). And extreme gradient boosting classifier combined with support vector machine feature selection method achieved best performance in predicting EGFR subtypes (AUC reached 0.76, 0.63, 0.61 in the internal test cohort and two external test cohorts, respectively). The C-index of the Cox proportional hazard model achieved 0.863. The integration of cross-combination method and the external validation from multi-center data achieved a good prediction and generalization performance in predicting EGFR mutation status and its subtypes. The combination of handcrafted radiomics features and clinical factors achieved good performance in predicting prognosis. With the urgent needs of multicentric 18F-FDG PET/CT trails, robust and explainable radiomics models have great potential in decision making and prognosis prediction of lung adenocarcinoma.

Development of a cuproptosis-related signature for prognosis prediction in lung adenocarcinoma based on WGCNA.

Cuproptosis is a novel mitochondrial respiration-dependent cell death mechanism induced by copper that can kill cancer cells via copper carriers in cancer therapy. However, the clinical significance and prognostic value of cuproptosis in lung adenocarcinoma (LUAD) remains unclear. We performed a comprehensive bioinformatics analysis of the cuproptosis gene set, including copy number aberration, single-nucleotide variation, clinical characteristics, survival analysis, etc. Cuproptosis-related gene set enrichment scores (cuproptosis Z-scores) were calculated in The Cancer Genome Atlas (TCGA)-LUAD cohort using single-sample gene set enrichment analysis (ssGSEA). Modules significantly associated with cuproptosis Z-scores were screened by weighted gene co-expression network analysis (WGCNA). The hub genes of the module were then further screened by survival analysis and least absolute shrinkage and selection operator (LASSO) analysis, in which TCGA-LUAD (497 samples) and GSE72094 (442 samples) were used as the training and validation cohorts, respectively. Finally, we analyzed the tumor characteristics, immune cell infiltration levels, and potential therapeutic agents. Missense mutation and copy number variant (CNV) events were general in the cuproptosis gene set. We identified 32 modules, of which the MEpurple (107 genes) and MEpink (131 genes) modules significantly positively and negatively correlated with cuproptosis Z-scores, respectively. We identified 35 hub genes significantly related to overall survival and constructed a prognostic model consisting of 7 cuproptosis-related genes in patients with LUAD. Compared with the low-risk group, patients in the high-risk group had a worse overall survival and gene mutation frequency, as well as significantly higher tumor purity. In addition, infiltration of immune cells was also significantly different between the 2 groups. Furthermore, the correlation between the risk scores and half-maximum inhibitory concentration (IC50) of antitumor drugs in the Genomics of Drug Sensitivity in Cancer (GDSC) v. 2 database was explored, revealing differences in drug sensitivity across the 2 risk groups. Our study provided a valid prognostic risk model for LUAD and improved understanding of its heterogeneity, which may aid in the development of personalized treatment strategies.

Identifying M1-like macrophage related genes for prognosis prediction in lung adenocarcinoma based on a gene co-expression network

Macrophages are one of the most important players in the tumor microenvironment. But the contribution of macrophages to lung adenocarcinoma (LUAD) is still controversial. The current study aimed to display an immune landscape to clarify the function of macrophages and detect prognostic hub genes in LUAD. The transcriptome data were adopted to screen differently expressed genes (DEGs) in The Cancer Genome Atlas database (TCGA). The cell type identification by estimating relative subsets of RNA transcripts (CIBERSORT) algorithm was used to reveal the immune landscape. Weighted gene co-expression network analysis (WGCNA) analysis was performed to identify the hub module associated with macrophages. Function Enrichment analysis was conducted on hub module genes. Moreover, univariate and multivariate Cox regression analyses were performed to identify prognostic hub genes. Kaplan-Meier (KM) and Time-dependent receiver operating characteristic (ROC) curves were plotted to assess the prognostic capacity of the four prognostic hub genes. The GES1196959 dataset from the Gene Expression Omnibus (GEO) database was downloaded to verify the differential expression of the 4 prognostic hub genes.

T Cell-Mediated Tumor Killing-Related Classification of the Immune Microenvironment and Prognosis Prediction of Lung Adenocarcinoma

Although immune checkpoint inhibitors (ICI) are a promising therapeutic strategy for lung adenocarcinoma (LUAD), individual subgroups that might benefit from them are yet to be identified. As T cell-mediated tumor killing (TTK) is an underlying mechanism of ICI, we identified subtypes based on genes associated with TTK sensitivity and assessed their predictive significance for LUAD immunotherapies. Using high-throughput screening techniques, genes regulating the sensitivity of T cell-mediated tumor killing (GSTTK) with differential expression and associations with prognosis were discovered in LUAD. Furthermore, patients with LUAD were divided into subgroups using unsupervised clustering based on GSTTK. Significant differences were observed in the tumor immune microenvironment (TIME), genetic mutation and immunotherapy response across subgroups. Finally, the prognostic significance of a scoring algorithm based on GSTTK was assessed. A total of 6 out of 641 GSTTK exhibited differential expression in LUAD and were associated with prognosis. Patients were grouped into two categories based on the expression of the six GSTTK, which represented different TTK immune microenvironments in LUAD. Immune cell infiltration, survival difference, somatic mutation, functional enrichment and immunotherapy responses also varied between the two categories. Additionally, a scoring algorithm accurately distinguished overall survival rates across populations. TTK had a crucial influence on the development of the varying TIME. Evaluation of the varied TTK modes of different tumors enhanced our understanding of TIME characteristics, wherein the changes in T cell activity in LUAD are reflected. Thus, this study guides the development of more effective therapeutic methods.

An Inflammatory Response-Related Gene Signature Can Predict the Prognosis and Impact the Immune Status of Lung Adenocarcinoma

Lung adenocarcinoma (LUAD) accounts for a cancer with high heterogeneity and poor prognostic outcome. Nonetheless, it is still unknown about the relation between inflammatory response-related genes (IRGs) and LUAD. This study used LASSO-Cox regression for establishing the multigene prognostic signature based on TCGA and the GSE31210 cohorts. In addition, gene set enrichment analysis (GSEA) was performed for GO and KEGG analyses. By contrast, single-sample GSEA (ssGSEA) investigated immune cell infiltration scores as well as the immune pathway activity. We also conducted qRT-PCR and IHC to evaluate prognostic gene expression at protein and mRNA levels within LUAD and adjacent healthy samples. As a result, a novel prognostic signature involving 10 IRGs was identified. Furthermore, the signature has been validated as being important in functional analysis, TME, drug sensitivity, and prognosis prediction in LUAD. Moreover, prognostic genes showed significant expression at protein and mRNA levels in LUAD compared with normal samples. The signature involving 10 IRGs could potentially predict LUAD prognosis.

Systematic Analysis and Identification of Molecular Subtypes of TRP-Related Genes and Prognosis Prediction in Lung Adenocarcinoma.

Background Transient receptor potential channel (TRP) is a superfamily of nonselective cation channels, which is a member of calcium ion channels with a vital role in different calcium ion signal transduction pathways. TRP channel expression is often changed in the tumor, although the role of TRP proteins in lung cancer is unknown. Methods Molecular Signatures Database (MsigDB) provided the TRP gene set. Univariate Cox regression analysis was performed on The Cancer Genome Atlas Lung Adenocarcinoma (TCGA-LUAD) data collection set employing the coxph function of R package survival to find prognosis-related genes. The R package ConsumusClusterPlus was employed for doing the consistency cluster analysis of TCGA-LUAD samples according to the prognosis-related TRP gene. The R-package limma was utilized for investigating the differential expression of TRP subtypes. According to the differentially expressed genes between subtypes, the least absolute shrinkage and selection operator (LASSO) regression was employed to find the major genes and develop the risk model. CIBERPORT algorithm, R package maftools, gene set variation analysis (GSVA), and pRRophetic of R-package were employed for measuring the proportion of immune cells among subtypes, genomic mutation difference, pathway enrichment score, and drug sensitivity analysis. Results A total of 15 TRP-related genes associated with the prognosis of lung adenocarcinoma were found. According to the expression value of 15 genes, lung adenocarcinoma can be sorted into two subcategories. The prognosis of cluster1 is considerably better in comparison with that of cluster2. There were 123 differentially expressed genes between C1 and C2 subtypes, including 6 up- and 117 downregulated genes. There were major variations in the tumor microenvironment between C1 and C2 subtypes. The proportion of CD8 T cells in the C1 subtype was considerably enhanced in comparison with that in the C2 subtype. We further discovered 123 differentially expressed genes among subtypes, and 8 key genes were obtained at the end. The risk score (RS) model developed by the 8-gene signature had good strength in the TCGA validation set, overall set, and Gene Expression Omnibus (GEO) external dataset. There were major variations in immune checkpoint gene expression, patient sensitivity to immunotherapeutic drugs, immune infiltration, and genomic mutations between high and low groups on the basis of RS. Conclusions The risk model developed on the basis of TRP-related genes can help in predicting the prognosis of patients suffering from lung adenocarcinoma and guide immunotherapy.

Identification and Validation of a Three Pyroptosis-Related lncRNA Signature for Prognosis Prediction in Lung Adenocarcinoma.

Pyroptosis, defined as programmed cell death, results in the release of inflammatory mediators. Recent studies have revealed that pyroptosis plays essential roles in antitumor immunity and immunotherapy efficacy. Long noncoding RNAs (lncRNAs) are involved in a variety of biological behaviors in tumor cells, although the roles and mechanisms of lncRNAs in pyroptosis are rarely studied. Our study aimed to establish a novel pyroptosis-related lncRNA signature as a forecasting tool for predicting prognosis and ascertaining immune value. Based on lung adenocarcinoma (LUAD) patients from The Cancer Genome Atlas (TCGA), we performed Pearson’s correlation analysis to identify pyroptosis-related lncRNAs. After differentially expressed gene analysis and univariate Cox regression analysis, we selected prognosis-related and differentially expressed lncRNAs. Finally, we performed multivariate Cox regression analysis to establish the three pyroptosis-related lncRNA signature. Kaplan–Meier (KM) survival analyses and receiver operating characteristic (ROC) curves indicated the excellent performance for predicting the prognosis of LUAD patients. At the same time, we applied multidimensional approaches to further explore the functional enrichment, tumor microenvironment (TME) landscape, and immunotherapy efficacy among the different risk groups. A nomogram was constructed by integrating risk scores and clinical characteristics, which was validated using calibrations and ROC curves. Three lncRNAs, namely, AC090559.1, AC034102.8, and AC026355.2, were involved in this signature and used to classify LUAD patients into low- and high-risk groups. Overall survival time (OS) was higher in the low-risk group than in the high-risk group, which was also validated in our LUAD cohort from Shandong Provincial Hospital. TME landscape analyses revealed that a higher abundance of infiltrating immune cells and a greater prevalence of immune-related events existed in the low-risk group. Meanwhile, higher expression of immune checkpoint (ICP) genes, higher immunophenoscore (IPSs), and greater T cell dysfunction in the low-risk group demonstrated a better response to immunotherapy than the high-risk group. Combined with predictions from the Tumor Immune Dysfunction and Exclusion (TIDE) website, we found that LUAD patients in the low-risk group significantly benefited from programmed cell death 1 (PD-1) and cytotoxic T-lymphocyte–associated protein 4 (CTLA4) immune checkpoint blockade (ICB) therapy compared with those in the high-risk group. Furthermore, drug susceptibility analysis identified potential sensitive chemotherapeutic drugs for each risk group. In this study, a novel three pyroptosis-related lncRNA signature was constructed, which could accurately predict the immunotherapy efficacy and prognosis in LUAD patients.

Identification of metastasis-related long non-coding RNAs in lung cancer through a novel tumor mesenchymal score

Long non-coding RNAs (lncRNAs) have been proven to play critical roles in epithelial-mesenchymal transition (EMT) and metastasis of lung cancer. However, the biological functions and related mechanisms of lncRNAs are unclear. In addition, the EMT-based prognosis prediction in lung cancer still lacks investigation. Here, we established the methodology of identifying critical metastasis-related lncRNAs using comprehensive datasets of cancer transcriptome, genome and epigenome, and also provided tools for prognosis prediction in lung cancer. Initially, important mesenchymal marker genes were identified to compose the tumor mesenchymal score, which predicted patient prognosis in lung cancer, especially lung adenocarcinoma (LUAD). The score was also correlated with several crucial biological and physiological processes, such as tumor immune and hypoxia. Based on the score, lung cancer patients was classified into epithelial and mesenchymal subtypes, and lncRNAs which exhibited expressional dysregulation, promotor methylation alteration and copy number variation between the two subtypes in LUAD were identified and underwent further prognostic analyses. Finally, we identified 14 lncRNAs as EMT-related and significant biomarkers in prognosis prediction of LUAD. As validation, lncRNA RBPMS-AS1 was proven to be co-expressed with epithelial biomarkers, suppressive for A549 cell migration, invasion and EMT, and also significantly associated with better outcomes of LUAD patients, suggesting the potential of RBPMS-AS1 to serve as a lncRNA epithelial biomarker in metastasis of LUAD. Based on the identified lncRNAs, an EMT-linked lncRNA prognostic signature was further established. Taken together, our study provides robust predictive tools, potential lncRNA targets and feasible screening strategies for future study of lung cancer metastasis.

GPR87 promotes tumor cell invasion and mediates the immunogenomic landscape of lung adenocarcinoma

The purpose of this study is to examine the association between G protein-coupled receptor 87 (GPR87) and lung adenocarcinoma (LUAD) metastasis and immune infiltration. The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets extract clinical data. According to the TCGA database, increased GPR87 expression predicts poor overall survival, progression-free interval, and disease-specific survival in LUAD patients. The meta-analysis also reveals a significant association between high GPR87 expression and poor overall survival. Moreover, functional experiments demonstrate that GPR87 silencing reduces LUAD cell invasion and migration. Immunoblotting shows that GPR87 knockdown decreased Vimentin and N-cadherin expression and increased E-cadherin expression in LUAD cells. GPR87 expression in LUAD is positively correlated with immune infiltration. In addition, GPR87 expression is associated with immune and chemotherapy resistance in LUAD patients. Our findings indicate that GPR87 promotes tumor progression and is correlated with immune infiltration, suggesting GPR87 as a possible biomarker for prognosis prediction in LUAD.

Construction of a hypoxic immune microenvironment associated gene-based model for prognosis prediction of lung adenocarcinoma.

Lung adenocarcinoma (LUAD) accounts for the majority of cancer deaths worldwide, with a high incidence rate and mortality. It is highly important to develop biomarker model to accurately predict the prognosis. RNA-Seq data and clinical follow-up data of LUAD were downloaded from The Cancer Genome Atlas (TCGA) database. Hypoxia-related gene sets were collected from the Gene Set Enrichment Analysis (GSEA) website. A gene signature model was established using the Limma package in the R software, univariate and multivariate survival analyses, and least absolute shrinkage and selection operator (LASSO) algorithms. Two hypoxia subtypes (C1 and C2) were classified according to the expressions of 55 prognostic hypoxic-related genes. Differentially expressed genes (DEGs) between two hypoxia subtypes and immune group were analyzed. Then, 390 DEGs related to hypoxic immune microenvironment were filtered. According to hypoxia type and immune type, the samples were classified into hypoxia-high & immune-low group, hypoxia-low & immune-high group. Based on these differentially expressed genes (DEGs), a 5-genes signature model, which showed a stable prediction performance on datasets of different platforms and immunotherapy datasets, was finally developed. Meanwhile, it demonstrated a better performance compared with other existing models. The AUC of the 5-gene signature was high in both the training dataset and 4 independent validation datasets and was confirmed as a clinical feature-independent prognostic model. This study developed a hypoxic immune microenvironment associated gene-based model for prognostic prediction of LUAD, providing clinicians with a reliable prognostic assessment tool and facilitating clinical treatment decision-making.

A Novel Mitochondrial-Related Gene Signature for the Tumor Immune Microenvironment Evaluation and Prognosis Prediction in Lung Adenocarcinoma.

Lung adenocarcinoma (LUAD) remains the most common deadly disease and has a poor prognosis. More and more studies have reported that mitochondrial-related genes (MTRGs) were associated with the clinical outcomes of multiple tumors solely. In this study, we aimed to develop a novel prognostic model based on MTRGs. Differentially expressed MTRGs were identified from TCGA-LUAD and GSE31210 cohorts. Univariate Cox regression analysis was utilized to screen differentially expressed MTRGs that were related to prognosis of LUAD. Then, LASSO Cox regression analysis was used to develop a prognostic signature. ESTIMATE was used for estimating the fractions of immune cell types. In this study, we identified 44 overlapping differentially expressed MTRGs in TCGA-LUAD and GSE31210 cohorts. Among 44 overlapping differentially expressed MTRGs, nine genes were associated with prognosis of LUAD. When the penalty parameter lambda was the minimum, there were six genes meeting the conditions of constructing the signature, including SERPINB5, CCNB1, FGR MAOB, SH3BP5, and CYP24A1. The survival analysis suggested that prognosis of patients in the high-risk group was significantly worse than that in the low-risk group. Cox regression analyses showed that the risk score was an independent predictor of LUAD prognosis. As with the results of ESTIMATE score, the degree of immune cell infiltration in the low-risk group was higher than that in the high-risk group, such as TIL, Treg, and B cells. In addition, TMB and cancer stem cell infiltration were higher in the low-risk group than the high-risk group. In conclusion, we developed a novel MTRG signature acting as a negative independent prognostic factor. In the future, individualized treatments and medical decision-making may benefit from using the predicted model.

Review of EMT network-based feature selection improves prognosis prediction in lung adenocarcinoma

Review of EMT network-based feature selection improves prognosis prediction in lung adenocarcinoma

DNA methylation molecular subtypes for prognosis prediction in lung adenocarcinoma

AimsLung cancer is one of the main results in tumor-related mortality. Methylation differences reflect critical biological features of the etiology of LUAD and affect prognosis.MethodsIn the present study, we constructed a prediction prognostic model integrating various DNA methylation used high-throughput omics data for improved prognostic evaluation.ResultsOverall 21,120 methylation sites were identified in the training dataset. Overall, 237 promoter genes were identified by genomic annotation of 205 CpG loci. We used Akakike Information Criteria (AIC) to obtain the validity of data fitting, but to prevent overfitting. After AIC clustering, specific methylation sites of cg19224164 and cg22085335 were left. Prognostic analysis showed a significant difference among the two groups (P = 0.017). In particular, the hypermethylated group had a poor prognosis, suggesting that these methylation sites may be a marker of prognosis.ConclusionThe model might help in the identification of unknown biomarkers in predicting patient prognosis in LUAD.

Multi-Omics Signatures Identification for LUAD Prognosis Prediction Model Based on the Integrative Analysis of Immune and Hypoxia Signals.

Lung adenocarcinoma (LUAD) is the most common histological subtype of lung cancer with heterogeneous outcomes and diverse therapeutic responses. However, the understanding of the potential mechanism behind LUAD initiation and progression remains limited. Increasing evidence shows the clinical significance of the interaction between immune and hypoxia in tumor microenvironment. To mine reliable prognostic signatures related to both immune and hypoxia and provide a more comprehensive landscape of the hypoxia-immune genome map, we investigated the hypoxia-immune-related alteration at the multi-omics level (gene expression, somatic mutation, and DNA methylation). Multiple strategies including lasso regression and multivariate Cox proportional hazards regression were used to screen the signatures with clinical significance and establish an incorporated prognosis prediction model with robust discriminative power on survival status on both the training and test datasets. Finally, combing all the samples, we constructed a robust model comprising 19 signatures for the prognosis prediction of LUAD patients. The results of our study provide a comprehensive landscape of hypoxia-immune related genetic alterations and provide a robust prognosis predictor for LUAD patients.

Identification of a Four-Gene Signature Associated with the Prognosis Prediction of Lung Adenocarcinoma Based on Integrated Bioinformatics Analysis.

Lung adenocarcinoma (LUAD) is often diagnosed at an advanced stage, so it is necessary to identify potential biomarkers for the early diagnosis and prognosis of LUAD. In our study, a gene co-expression network was constructed using weighted gene co-expression network analysis (WGCNA) in order to obtain the key modules and genes correlated with LUAD prognosis. Four hub genes (HLF, CHRDL1, SELENBP1, and TMEM163) were screened out using least absolute shrinkage and selection operator (LASSO)–Cox regression analysis; then, a prognostic model was established for predicting overall survival (OS) based on these four hub genes..Furthermore, the prognostic values of this four-gene signature were verified in four validation sets (GSE26939, GSE31210, GSE72094, and TCGA-LUAD) as well as in the GEPIA database. To assess the prognostic values of hub genes, receiver operating characteristic (ROC) curves were constructed and a nomogram was created. We found that a higher expression of four hub genes was associated with a lower risk of patient death. In a training set, it was demonstrated that this four-gene signature was a better prognostic factor than clinical factors such as age and stage of disease. Moreover, our results revealed that these four genes were suppressor factors of LUAD and that their high expression was associated with a lower risk of death. In summary, we demonstrated that this four-gene signature could be a potential prognostic factor for LUAD patients. These findings provide a theoretical basis for exploring potential biomarkers for LUAD prognosis prediction in the future.

A Novel Inflammatory-Related Gene Signature Based Model for Risk Stratification and Prognosis Prediction in Lung Adenocarcinoma.

Inflammation is an important hallmark of cancer and plays a role in both neogenesis and tumor development. Despite this, inflammatory-related genes (IRGs) remain to be poorly studied in lung adenocarcinoma (LUAD). We aim to explore the prognostic value of IRGs for LUAD and construct an IRG-based prognosis signature. The transcriptomic profiles and clinicopathological information of patients with LUAD were obtained from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO). Least absolute shrinkage and selection operator (LASSO) analysis and multivariate Cox regression were applied in the TCGA set to generate an IRG risk signature. LUAD cases with from the GSE31210 and GSE30219 datasets were used to validate the predictive ability of the signature. Analysis of the TCGA cohort revealed a five-IRG risk signature consisting of EREG, GPC3, IL7R, LAMP3, and NMUR1. This signature was used to divide patients into two risk groups with different survival rates. Multivariate Cox regression analysis verified that the risk score from the five-IRG signature negatively correlated with patient outcome. A nomogram was developed using the IRG risk signature and stage, with C-index values of 0.687 (95% CI: 0.644–0.730) in the TCGA training cohort, 0.678 (95% CI: 0.586–0.771) in GSE30219 cohort, and 0.656 (95% CI: 0.571–0.740) in GSE30219 cohort. Calibration curves were consistent between the actual and the predicted overall survival. The immune infiltration analysis in the TCGA training cohort and two GEO validation cohorts showed a distinctly differentiated immune cell infiltration landscape between the two risk groups. The IRG risk signature for LUAD can be used to predict patient prognosis and guide individual treatment. This risk signature is also a potential biomarker of immunotherapy.

Histopathological Images and Multi-Omics Integration Predict Molecular Characteristics and Survival in Lung Adenocarcinoma.

Histopathological images and omics profiles play important roles in prognosis of cancer patients. Here, we extracted quantitative features from histopathological images to predict molecular characteristics and prognosis, and integrated image features with mutations, transcriptomics, and proteomics data for prognosis prediction in lung adenocarcinoma (LUAD). Patients obtained from The Cancer Genome Atlas (TCGA) were divided into training set (n = 235) and test set (n = 235). We developed machine learning models in training set and estimated their predictive performance in test set. In test set, the machine learning models could predict genetic aberrations: ALK (AUC = 0.879), BRAF (AUC = 0.847), EGFR (AUC = 0.855), ROS1 (AUC = 0.848), and transcriptional subtypes: proximal-inflammatory (AUC = 0.897), proximal-proliferative (AUC = 0.861), and terminal respiratory unit (AUC = 0.894) from histopathological images. Moreover, we obtained tissue microarrays from 316 LUAD patients, including four external validation sets. The prognostic model using image features was predictive of overall survival in test and four validation sets, with 5-year AUCs from 0.717 to 0.825. High-risk and low-risk groups stratified by the model showed different survival in test set (HR = 4.94, p < 0.0001) and three validation sets (HR = 1.64–2.20, p < 0.05). The combination of image features and single omics had greater prognostic power in test set, such as histopathology + transcriptomics model (5-year AUC = 0.840; HR = 7.34, p < 0.0001). Finally, the model integrating image features with multi-omics achieved the best performance (5-year AUC = 0.908; HR = 19.98, p < 0.0001). Our results indicated that the machine learning models based on histopathological image features could predict genetic aberrations, transcriptional subtypes, and survival outcomes of LUAD patients. The integration of histopathological images and multi-omics may provide better survival prediction for LUAD.

Immune-related eight-lncRNA signature for improving prognosis prediction of lung adenocarcinoma.

BackgroundLung adenocarcinoma (LUAD) is the leading cause of cancer‐related deaths worldwide. Therefore, the identification of a novel prediction signature for predicting the prognosis risk and survival outcomes is urgently demanded.MethodsWe integrated a machine‐learning frame by combing the Cox regression and Least Absolute Shrinkage and Selection Operator (LASSO) regression model to identify the LUAD‐related long non‐coding RNA (lncRNA) survival biomarkers. Subsequently, the Spearman correlation test was employed to interrogate the relationships between lncRNA signature and tumor immunity and constructed the competing endogenous RNA (ceRNA) network.ResultsHerein, we identified an eight‐lncRNA signature (PR‐lncRNA signature, NPSR1‐AS1, SATB2‐AS1, LINC01090, FGF12‐AS2, AC005256.1, MAFA‐AS1, BFSP2‐AS1, and CPC5‐AS1), which contributes to predicting LUAD patient's prognosis risk and survival outcomes. The PR‐lncRNA signature has also been confirmed as the robust signature in independent datasets. Further parsing of the LUAD tumor immune infiltration showed the PR‐lncRNAs were closely associated with the abundance of multiple immune cells infiltration and the expression of MHC molecules. Furthermore, by constructing the PR‐lncRNA–related ceRNA network, we interrogated more potential anti‐cancer therapy targets.ConclusionlncRNAs, as emerging cancer biomarkers, play an important role in a variety of cancer processes. Identification of PR‐lncRNA signatures allows us to better predict patient's survival outcomes and disease risk. Finally, the PR‐lncRNA signatures could help us to develop novel LUAD anti‐cancer therapeutic strategies.

An Inflammation-Related Nine-Gene Signature to Improve Prognosis Prediction of Lung Adenocarcinoma

Background A novel predictive model was rarely reported based on inflammation-related genes to explore clinical outcomes of lung adenocarcinoma (LUAD) patients. Methods Using TCGA database, we screened nine inflammation-related genes with a prognostic value, and LASSO regression was applied for model construction. The predictive value of the prognostic signature developed from inflammation-related genes was assessed by survival assays and multivariate assays. PCA and t-SNE analysis were performed to demonstrate clustering abilities of risk scores. Results Thirteen inflammation-related genes (BTG2, CCL20, CD69, DCBLD2, GPC3, IL7R, LAMP3, MMP14, NMUR1, PCDH7, PIK3R5, RNF144B, and TPBG) with prognostic values were finally identified. LASSO regression further screened nine candidates (BTG2, CCL20, CD69, IL7R, MMP14, NMUR1, PCDH7, RNF144B, and TPBG). Then, a prognostic prediction model using the above nine genes was constructed. A reliable clustering ability of risk score was demonstrated by PCA and t-SNE assays in 500 LUAD patients. The survival assays revealed that the overall survivals of the high-risk group were distinctly poorer than those of the low-risk group with 1-, 3-, and 5-year AUC values of 0.695, 0.666, and 0.694, respectively. Finally, multivariate assays demonstrated the scoring system as an independent prognostic factor for overall survival. Conclusions Our study shows that the signature of nine inflammation-related genes can be used as a prognostic marker for LUAD.