Gene Instability-Related lncRNA Prognostic Model of Melanoma Patients via Machine Learning Strategy

Machine learning combined with single-cell analysis reveals predictive capacity and immunotherapy response of T cell exhaustion-associated lncRNAs in uterine corpus endometrial carcinoma

BACKGROUND

The exhaustion of T-cells is a primary factor contributing to immune dysfunction in cancer. Long non-coding RNAs (lncRNAs) play a significant role in the advancement, survival, and treatment of Uterine Corpus Endometrial Carcinoma (UCEC). Nevertheless, there has been no investigation into the involvement of lncRNAs associated with T-cell exhaustion (TEXLs) in UCEC. The goal of this work is to establish predictive models for TEXLs in UCEC and study their related immune features.

METHODS

Using transcriptome and single-cell sequencing data from The Cancer Genome Atlas and Gene Expression Omnibus databases, we employed co-expression analysis and univariate Cox regression to identify prognostic-associated TEXLs (pTEXLs). The prognostic model was developed using the Least Absolute Contraction and Selection Operator. The immunotherapy characteristics of the prognostic model risk score were studied. Then molecular subgroups were identified through non-negative Matrix Factorization based on pTEXLs. The identification of co-expressed genes was done using a weighted correlation network analysis. Subsequently, a diagnostic model for UCEC was created. In-depth investigations, both in vitro and in vivo, were carried out to elucidate the molecular mechanism of the key gene within the diagnostic model.

RESULTS

Receiver operating characteristic curve, calibration curve, and decision curve analysis proved the validity of the predictive models established according to pTEXLs. The subgroup with lower risk scores in the prognostic model has better responses to blocking immune checkpoint therapy. Single-cell analysis suggests that the expression level of MIEN1 is relatively high in immune cells among diagnostic genes. Furthermore, the targeted suppression of MIEN1 via sh-MIEN1 diminishes the proliferative, migratory, and invasive capacities of UCEC cells, potentially associated with CD8+ T cell exhaustion.

CONCLUSIONS

The association between TEXLs and UCEC was methodically elucidated by our investigation. A stable pTEXLs risk prediction model and a diagnosis model for UCEC were also established.

The obesity-linked human lncRNA AATBC stimulates mitochondrial function in adipocytes

Adipocytes are critical regulators of metabolism and energy balance. While white adipocyte dysfunction is a hallmark of obesity-associated disorders, thermogenic adipocytes are linked to cardiometabolic health. As adipocytes dynamically adapt to environmental cues by functionally switching between white and thermogenic phenotypes, a molecular understanding of this plasticity could help improving metabolism. Here, we show that the lncRNA Apoptosis associated transcript in bladder cancer (AATBC) is a human-specific regulator of adipocyte plasticity. Comparing transcriptional profiles of human adipose tissues and cultured adipocytes we discovered that AATBC was enriched in thermogenic conditions. Using primary and immortalized human adipocytes we found that AATBC enhanced the thermogenic phenotype, which was linked to increased respiration and a more fragmented mitochondrial network. Expression of AATBC in adipose tissue of mice led to lower plasma leptin levels. Interestingly, this association was also present in human subjects, as AATBC in adipose tissue was inversely correlated with plasma leptin levels, BMI, and other measures of metabolic health. In conclusion, AATBC is a novel obesity-linked regulator of adipocyte plasticity and mitochondrial function in humans.

Senescence-Related lncRNA Signature Predicts Prognosis, Response to Immunotherapy and Chemotherapy in Skin Cutaneous Melanoma

Skin cutaneous melanoma (SKCM) is a highly malignant and aggressive cancer. Previous studies have shown that cellular senescence is a promising therapeutic strategy to limit melanoma cell progression. However, models to predict the prognosis of melanoma based on senescence-related lncRNAs and the efficacy of immune checkpoint therapy remain undefined. In this study, we developed a predictive signature consisting of four senescence-related lncRNAs (AC009495.2, U62317.1, AATBC, MIR205HG), and we then classified patients into high- and low-risk groups. GSEA (Gene set enrichment analysis) showed different activation of immune-related pathways in two groups. In addition, there were significant differences between the scores of tumor immune microenvironment, tumor burden mutation, immune checkpoint expression, and chemotherapeutic drug sensitivity between the two groups of patients. It provides new insights to guide more personalized treatment for patients with SKCM.

Integration of lncRNAs, Protein-Coding Genes and Pathology Images for Detecting Metastatic Melanoma

Melanoma is a lethal skin disease that develops from moles. This study aimed to integrate multimodal data to predict metastatic melanoma, which is highly aggressive and difficult to treat. The proposed EnsembleSKCM method evaluated the prediction performances of long noncoding RNAs (lncRNAs), protein-coding messenger genes (mRNAs) and pathology images (images) for metastatic melanoma. Feature selection was used to screen for metastatic biomarkers in the lncRNA and mRNA datasets. The integrated EnsembleSKCM model was built based on the weighted results of the lncRNA-, mRNA- and image-based models. EnsembleSKCM achieved 0.9444 in the prediction accuracy of metastatic melanoma and outperformed the single-modal prediction models based on the lncRNA, mRNA and image data. The experimental data suggest the importance of integrating the complementary information from the three data modalities. WGCNA was used to analyze the relationship of molecular-level features and image features, and the results show connections between them. Another cohort was used to validate our prediction.

Novel GIRlncRNA Signature for Predicting the Clinical Outcome and Therapeutic Response in NSCLC

Background: Non–small cell lung cancer (NSCLC) is highly malignant with driver somatic mutations and genomic instability. Long non-coding RNAs (lncRNAs) play a vital role in regulating these two aspects. However, the identification of somatic mutation-derived, genomic instability-related lncRNAs (GIRlncRNAs) and their clinical significance in NSCLC remains largely unexplored.

Methods: Clinical information, gene mutation, and lncRNA expression data were extracted from TCGA database. GIRlncRNAs were screened by a mutator hypothesis-derived computational frame. Co-expression, GO, and KEGG enrichment analyses were performed to investigate the biological functions. Cox and LASSO regression analyses were performed to create a prognostic risk model based on the GIRlncRNA signature (GIRlncSig). The prediction efficiency of the model was evaluated by using correlation analyses with mutation, driver gene, immune microenvironment contexture, and therapeutic response. The prognostic performance of the model was evaluated by external datasets. A nomogram was established and validated in the testing set and TCGA dataset.

Results: A total of 1446 GIRlncRNAs were selected from the screen, and the established GIRlncSig was used to classify patients into high- and low-risk groups. Enrichment analyses showed that GIRlncRNAs were mainly associated with nucleic acid metabolism and DNA damage repair pathways. Cox analyses further identified 19 GIRlncRNAs to construct a GIRlncSig-based risk score model. According to Cox regression and stratification analyses, 14 risk lncRNAs (AC023824.3, AC013287.1, AP000829.1, LINC01611, AC097451.1, AC025419.1, AC079949.2, LINC01600, AC004862.1, AC021594.1, MYRF-AS1, LINC02434, LINC02412, and LINC00337) and five protective lncRNAs (LINC01067, AC012645.1, AL512604.3, AC008278.2, and AC089998.1) were considered powerful predictors. Analyses of the model showed that these GIRlncRNAs were correlated with somatic mutation pattern, immune microenvironment infiltration, immunotherapeutic response, drug sensitivity, and survival of NSCLC patients. The GIRlncSig risk score model demonstrated good predictive performance (AUCs of ROC for 10-year survival was 0.69) and prognostic value in different NSCLC datasets. The nomogram comprising GIRlncSig and tumor stage exhibited improved robustness and feasibility for predicting NSCLC prognosis.

Conclusion: The newly identified GIRlncRNAs are powerful biomarkers for clinical outcome and prognosis of NSCLC. Our study highlights that the GIRlncSig-based score model may be a useful tool for risk stratification and management of NSCLC patients, which deserves further evaluation in future prospective studies.

Ferroptosis-related lncRNA signature predicts prognosis and immunotherapy efficacy in cutaneous melanoma

Purpose

Ferroptosis-related lncRNAs are promising biomarkers for predicting the prognosis of many cancers. However, a ferroptosis-related signature to predict the prognosis of cutaneous melanoma (CM) has not been identified. The purpose of this study was to construct a ferroptosis-related lncRNA signature to predict prognosis and immunotherapy efficacy in CM.

Methods

Ferroptosis-related differentially expressed genes (FDEGs) and lncRNAs (FDELs) were identified using TCGA, GTEx, and FerrDb datasets. We performed Cox and LASSO regressions to identify key FDELs, and constructed a risk score to stratify patients into high- and low-risk groups. The lncRNA signature was evaluated using the areas under the receiver operating characteristic curves (AUCs) and Kaplan-Meier analyses in the training, testing, and entire cohorts. Multivariate Cox regression analyses including the lncRNA signature and common clinicopathological characteristics were performed to identify independent predictors of overall survival (OS). A nomogram was developed for clinical use. We performed gene set enrichment analyses (GSEA) to identify significantly enriched pathways. Differences in the tumor microenvironment (TME) between the 2 groups were assessed using 7 algorithms. To predict the efficacy of immune checkpoint inhibitors (ICI), we analyzed the association between PD1 and CTLA4 expression and the risk score. Finally, differences in Tumor Mutational Burden (TMB) and molecular drugs Sensitivity between the 2 groups were performed.

Results

We identified 5 lncRNAs (AATBC, AC145423.2, LINC01871, AC125807.2, and AC245041.1) to construct the risk score. The AUC of the lncRNA signature was 0.743 in the training cohort and was validated in the testing and entire cohorts. Kaplan-Meier analyses revealed that the high-risk group had poorer prognosis. Multivariate Cox regression showed that the lncRNA signature was an independent predictor of OS with higher accuracy than traditional clinicopathological features. The 1-, 3-, and 5-year survival probabilities for CM patients were 92.7%, 57.2%, and 40.2% with an AUC of 0.804, indicating a good accuracy and reliability of the nomogram. GSEA showed that the high-risk group had lower ferroptosis and immune response. TME analyses confirmed that the high-risk group had lower immune cell infiltration (e.g., CD8+ T cells, CD4+ memory-activated T cells, and M1 macrophages) and lower immune functions (e.g., immune checkpoint activation). Low-risk patients whose disease expressed PD1 or CTLA4 were likely to respond better to ICIs. The analysis demonstrated that the TMB had significantly difference between low- and high- risk groups. Chemotherapy drugs, such as sorafenib, Imatinib, ABT.888 (Veliparib), Docetaxel, and Paclitaxel showed Significant differences in the estimated IC50 between the two risk groups.

Conclusion

Our novel ferroptosis-related lncRNA signature was able to accurately predict the prognosis and ICI outcomes of CM patients. These ferroptosis-related lncRNAs might be potential biomarkers and therapeutic targets for CM.

Identification of LncRNA Prognostic Signature Associated With Genomic Instability in Pancreatic Adenocarcinoma

Background

Genomic instability (GI) is a critical feature of cancer which plays a key role in the occurrence and development of pancreatic adenocarcinoma (PAAD). Long non-coding RNA (LncRNA) is an emerging prognostic biomarker because it is involved in regulating GI. Recently, researchers used such GI-related LncRNAs (GILncRNAs) to establish a prognostic signature for patients with cancer and helped in predicting the overall prognosis of the patients. However, it is evident that patients with PAAD still lack such prognostic signature constructed with GILncRNA.

Methods

The present study screened GILncRNAs from 83 patients with PAAD. Prognosis-related GILncRNAs were identified by univariate Cox regression analysis. The correlation coefficients of these GILncRNAs were obtained by multivariate Cox regression analysis and used to construct a signature. The signature in the present study was then assessed through survival analysis, mutation correlation analysis, independent prognostic analysis, and clinical stratification analysis in the training set and validated in the testing as well as all TCGA set. The current study performed external clinical relevance validation of the signature and validated the effect of AC108134.2 in GILncSig on PAAD using in vitro experiments. Finally, the function of GILncRNA signature (GILncSig) dependent on Gene Ontology enrichment analysis was explored and chemotherapeutic drug sensitivity analysis was also performed.

Results

Results of the present study found that a total of 409 GILncRNAs were identified, 5 of which constituted the prognostic risk signature in this study, namely, AC095057.3, AC108134.2, AC124798.1, AL606834.1, and AC104695.4. It was found that the signature of the present study was better than others in predicting the overall survival and applied to patients with PAAD of all ages, genders, and tumor grades. Further, it was noted that the signature of the current study in the GSE102238, was correlated with tumor length, and tumor stage of patients with PAAD. In vitro, functional experiments were used in the present study to validate that AC108134.2 is associated with PAAD genomic instability and progression. Notably, results of the pRRophetic analysis in the current study showed that the high-risk group possessed reverse characteristics and was sensitive to chemotherapy.

Conclusions

In conclusion, it was evident that the GILncSig used in the present study has good prognostic performance. Therefore, the signature may become a potential sensitive biological indicator of PAAD chemotherapy, which may help in clinical decision-making and management of patients with cancer.