Background: As a caspase-independent type of cell death, necroptosis plays a significant role in the initiation, and progression of gastric cancer (GC). Numerous studies have confirmed that long non-coding RNAs (lncRNAs) are closely related to the prognosis of patients with GC. However, the relationship between necroptosis and lncRNAs in GC remains unclear.Methods: The molecular profiling data (RNA-sequencing and somatic mutation data) and clinical information of patients with stomach adenocarcinoma (STAD) were retrieved from The Cancer Genome Atlas (TCGA) database. Pearson correlation analysis was conducted to identify the necroptosis-related lncRNAs (NRLs). Subsequently, univariate Cox regression and LASSO-Cox regression were conducted to establish a 12-NRLs signature in the training set and validate it in the testing set. Finally, the prognostic power of the 12-NRLs signature was appraised via survival analysis, nomogram, Cox regression, clinicopathological characteristics correlation analysis, and the receiver operating characteristic (ROC) curve. Furthermore, correlations between the signature risk score (RS) and immune cell infiltration, immune checkpoint molecules, somatic gene mutations, and anticancer drug sensitivity were analyzed.Results: In the present study, a 12-NRLs signature comprising REPIN1-AS1, UBL7-AS1, LINC00460, LINC02773, CHROMR, LINC01094, FLNB-AS1, ITFG1-AS1, LASTR, PINK1-AS, LINC01638, and PVT1 was developed to improve the prognosis prediction of STAD patients. Unsupervised methods, including principal component analysis and t-distributed stochastic neighbor embedding, confirmed the capability of the present signature to separate samples with RS. Kaplan-Meier and ROC curves revealed that the signature had an acceptable predictive potency in the TCGA training and testing sets. Cox regression and stratified survival analysis indicated that the 12-NRLs signature were risk factors independent of various clinical parameters. Additionally, immune cell infiltration, immune checkpoint molecules, somatic gene mutations, and half-inhibitory concentration differed significantly among different risk subtypes, which implied that the signature could assess the clinical efficacy of chemotherapy and immunotherapy.Conclusion: This 12-NRLs risk signature may help assess the prognosis and molecular features of patients with STAD and improve treatment modalities, thus can be further applied clinically.

Peritoneal metastasis is the most frequent cause of death in patients with advanced gastric carcinoma (GC). The phosphatase of regenerating liver-3 (PRL-3) is recognized as an oncogene and plays an important role in GC peritoneal metastasis. However, the mechanism of how PRL-3 regulates GC invasion and metastasis is unknown. In the present study, we found that PRL-3 presented with high expression in GC with peritoneal metastasis, but phosphatase and tensin homologue (PTEN) was weakly expressed. The p-PTEN/PTEN ratio was also higher in GC with peritoneal metastasis than that in the normal gastric tissues. We also found the same phenomenon when comparing the gastric mucosa cell line with the GC cell lines. After constructing a wild-type and a mutant-type plasmid without enzyme activity and transfecting them into GC SGC7901 cells, we showed that only PRL-3 had enzyme activity to downregulate PTEN and cause PTEN phosphorylation. The results also showed that PRL-3 increased the expression levels of MMP-2/MMP-9 and promoted the migration and invasion of the SGC7901 cells. Knockdown of PRL-3 decreased the expression levels of MMP-2/MMP-9 significantly, which further inhibited the migration and invasion of the GC cells. PRL-3 also increased the expression ratio of p-Akt/Akt, which indicated that PRL-3 may mediate the PI3K/Akt pathway to promote GC metastasis. When we transfected the PTEN siRNA plasmid into the PRL-3 stable low expression GC cells, the expression of p-Akt, MMP-2 and MMP-9 was reversed. In conclusion, our results provide a bridge between PRL-3 and PTEN; PRL-3 decreased the expression of PTEN as well as increased the level of PTEN phosphorylation and inactivated it, consequently activating the PI3K/Akt signaling pathway, and upregulating MMP-2/MMP-9 expression to promote GC cell peritoneal metastasis.

Background. The tumor microenvironment (TME) is associated with disease outcomes and treatment response in colon cancer. Here, we constructed a TME‐related gene signature that is prognosis of disease survival and may predict response to immunotherapy in colon cancer. Methods. We calculated immune and stromal scores for 385 colon cancer samples from The Cancer Genome Atlas (TCGA) database using the ESTIMATE algorithm. We identified nine TME‐related prognostic genes using Cox regression analysis. We evaluated associations between protein expression, extent of immune cell infiltrate, and patient survival. We calculated risk scores and built a clinical predictive model for the TME‐related gene signature. Receiver operating characteristic (ROC) curves were generated to assess the predictive power of the signature. We estimated the half‐maximal inhibitory concentration (IC50) of chemotherapeutic drugs in patients using the pRRophetic algorithm. The expression of immune checkpoint genes was evaluated. Results. High immune and stromal scores are significantly associated with poor overall survival (p < 0.05). We identified 773 differential TME‐related prognostic genes associated with survival; these genes were enriched in immune‐related pathways. Nine key prognostic genes were identified and were used to construct a TME‐related prognostic signature: CADM3, LEP, CD1B, PDE1B, CCL22, ABI3BP, IGLON5, SELE, and TGFB1. This signature identified a high‐risk group with worse survival outcomes, based on Kaplan‐Meier analysis. A nomogram composed of clinicopathological factors and risk score exhibited good accuracy. Drug sensitivity analysis identified no difference in sensitivity between the high‐risk and low‐risk groups. High‐risk patients had higher expression of PD‐1, PDL‐1, and CTLA‐4 and lower expression of LAG‐3 and VSIR. Infiltration of dendritic cells was higher in the high‐risk group. Conclusions. We identified a novel prognostic TME‐related gene expression signature in colon cancer. Stratification of patients based on this gene signature could be used to improve outcomes and guide better therapy for colon cancer patients.

Background: This study aims to develop and validate a predictive model combining deep transfer learning, radiomics, and clinical features for lymph node metastasis (LNM) in early gastric cancer (EGC).Materials and methods: This study retrospectively collected 555 patients with EGC, and randomly divided them into two cohorts with a ratio of 7:3 (training cohort, n = 388; internal validation cohort, n = 167). A total of 79 patients with EGC collected from the Second Affiliated Hospital of Soochow University were used as external validation cohort. Pre-trained deep learning networks were used to extract deep transfer learning (DTL) features, and radiomics features were extracted based on hand-crafted features. We employed the Spearman rank correlation test and least absolute shrinkage and selection operator regression for feature selection from the combined features of clinical, radiomics, and DTL features, and then, machine learning classification models including support vector machine, K-nearest neighbor, random decision forests (RF), and XGBoost were trained, and their performance by determining the area under the curve (AUC) were compared.Results: We constructed eight pre-trained transfer learning networks and extracted DTL features, respectively. The results showed that 1,048 DTL features extracted based on the pre-trained Resnet152 network combined in the predictive model had the best performance in discriminating the LNM status of EGC, with an AUC of 0.901 (95% CI: 0.847–0.956) and 0.915 (95% CI: 0.850–0.981) in the internal validation and external validation cohorts, respectively.Conclusion: We first utilized comprehensive multidimensional data based on deep transfer learning, radiomics, and clinical features with a good predictive ability for discriminating the LNM status in EGC, which could provide favorable information when choosing therapy options for individuals with EGC.

Ferroptosis is a newly defined mode of programmed oxidative cell death. Knowledge of ferroptosis-related long noncoding (lnc) RNA in the tumor immune microenvironment of colon cancer is lacking. We systematically analyzed the correlations between ferroptosis-related lncRNAs and the tumor microenvironment, immune cell infiltration, and patient prognosis for 379 colon cancer samples in the Cancer Genome Atlas (TCGA). Using consensus clustering, we divided the 379 colon cancer patients into two subgroups (clusters 1 and 2) based on the differentially expressed ferroptosis-related lncRNAs. Cluster 1 was preferentially associated with longer overall survival, upregulated immune checkpoint inhibitor expressions, higher immunoscores, higher stromal scores, higher estimated scores, and distinct immune cell infiltration. Cancer- and metabolism-related pathways were enriched by gene set enrichment analyses. We constructed a prognostic signature of 15 ferroptosis-related lncRNAs (ZEB1-AS1, LINC01011, AC005261.3, LINC01063, LINC02381, ELFN1-AS1, AC009283.1, LINC02361, AC105219.1, AC002310.1, AL590483.1, MIR4435-2HG, NKILA, AC021054.1, and AL450326.1) and divided the patients into the high- and low-risk-score groups. The signature was validated using TCGA training and testing cohorts. The risk signature was an independent prognostic factor for predicting survival and excellently predicted the prognoses of patients with colon cancer. Moreover, the risk signature was related to immune characteristics. Chemosensitivity analyses showed that low-risk-score patients were more sensitive to sorafenib. In summary, our work revealed the important role of ferroptosis-related lncRNAs in the tumor microenvironment and immune cell infiltration and may help determine personalized prognoses and treatment for patients with colon cancer.