Background Long non-coding RNAs (lncRNAs) have been implicated in diagnosis and prognosis in various cancers. However, few lncRNA signatures have been established for prediction of gastrointestinal stromal tumors (GIST). We aimed to explore a lncRNA signature profile that associated with clinical relevance by mining data from Gene Expression Ominus (GEO) and Surveillance, Epidemiology, and End Results (SEER) Program. Methods Using a lncRNA-mining approach, we performed non-negative matrix factorization (NMF) consensus algorithm in Gastrointestinal stromal tumors (GISTs) cohorts (61 patients from GSE8167 and GSE17743) to cluster LncRNA expression profiles. Comparative markers selection, and Gene Set Enrichment Analysis (GSEA) algorithm were performed between distinct molecular subtypes of GIST. The survival rate of GIST patients from SEER stratified by gender were compared by Kaplan–Meier method and log-rank analysis. lncRNA-mRNA co-expression analysis was performed by Pearson correlation coefficients (PCC) using R package LINC. Somatic copy number alterations of GIST patients (GSE40966) were analyzed via web server GenePattern GISTIC2 algorithm. Results A total of four lncRNA molecular subtypes of GIST were identified with distinct biological pathways and clinical characteristics. LncRNA expression profiles well clustered the GIST samples into small size (<5 mm) and large size tumors (>5 mm), which is a fundamental index for GIST malignancy diagnosis. Several lncRNAs with abundant expression (LRRC75A-AS1, HYMAI, NEAT1, XIST and FTX) were closely associated with tumor size, which may suggest to be biomarkers for the GIST malignancy. Particularly, LRRC75A-AS1 was positively associated with tumor diameters and suggested an oncogene in GIST. Co-expression analysis suggested that chromosome region 17p11.2–p12 may contribute to the oncogenic process in malignant GIST. Interestingly, the gender had a strong influence on clustering by lncRNA expression profile. Data from the Surveillance, Epidemiology, and End Results (SEER) Program were further explored and 7983 patients who were diagnosed with GISTs from 1973 to 2014 were enrolled for analysis. The results also showed the favorable prognosis for female patients. The survival rate between male and female with GIST was statistically significant (P < 0.0001). Gene set enrichment analysis (GSEA) indicated distinct pathways between female and male, and malignant GIST was associated with several cancer metabolism and cell cycle associated pathways. Conclusions This lncRNAs-based classification for GISTs may provide a molecular classification applicable to individual GIST that has implications to influence lncRNA markers selection and prediction of tumor progression.
Background Early non-invasive identification of patients at risk of developing postoperative sleep disorder (PSD), which is common after surgery, is an essential step in reducing surgery stress and an important part of enhanced recovery after surgery. Objective We used smart HRV patches to (1) explore different HRV parameters as potential PSD biomarkers and (2) develop and validate a prognostic model for the early prediction of PSD including change of autonomic function in early postoperative period. Methods This is a prospective cohort study where we assessed autonomic function in a separate sample of 51 patients who underwent DaVinci robotic/laparoscopic radical surgery for gastrointestinal cancer with and without insomnia. Results In this study, 22(43.137%) of 51 patients experienced PSD. Multivariate logistic regression analysis showed that ICU, POD3 nocturnal LF/HF and SD daytime pNN50 were risk predictors of postoperative sleep quality. The risk factor prediction model was established using ICU (P = 0.013, OR = 0.030), 120h SDNN (P = 0.072, OR = 0.954), POD3 daytime LF/HF (P = 0.096, OR = 3.894), POD3 nocturnal LF/HF (P = 0.025, OR = 1.235), POD2 24h LF/HF (P = 0.256, OR = 4.370), and SD daytime pNN50 (P = 0.039, OR = 0.828). The AUC was 0.969. Conclusion Circadian rhythm and activity of ANS was involved in PSD. HRV based on remote measurement technology and long-range monitor have potential as digital biomarkers for helping predict PSD.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.