Search

Background: Little is known about patients' intention for participation in end-of-life decisions (EOLD) in three coastal provinces in southern China. This study aimed to explore the willingness of patients with cancer pain to participate in EOLD and potential influencing factors. Methods: A multi-center cross-sectional study was performed in three coastal provinces in southern China. Two hundred and thirty patients with cancer pain were recruited and consented to fill out the questionnaires. The patients' willingness to participate in EOLD, demographic and disease-related data was surveyed. Results: In total, 223 patients completed and returned the survey (response rate = 96.95%). One hundred four cases (46.64%) were willing to participate in EOLD. 119 (54.36%) cases not willing to participate in EOLD, respectively. Multivariate logistic regression analysis shows that educational level (OR: 0.683, 95% CI: 0.482–0.966), history of alcoholism (OR: 8.353, 95%CI: 2.535–27.525), Eastern Cooperative Oncology Group (ECOG) score (OR: 0.645, 95% CI: 0.450–0.925) and experience of explosive pain (OR: 6.367, 95% CI: 3.103–13.062) and clinical rescue (OR: 3.844, 95% CI: 1.722–8.577) had significant effects on EOLD intention (P < 0.05). Finally, a predictive model combined above five factors was established, which showed a good discrimination (area under receiver operating characteristic curve: 0.849, 95% CI: 0.796–0.899, P < 0.001) and calibration (Hosmer-Lemeshow Test: Chi-square = 10.103, P = 0.258) for which patients more willing to participate in EOLD. Conclusions: The willingness of patients with cancer pain to participate in EOLD is at a modest level in three coastal provinces in southern China. Patients with lower educational level, history of alcoholism, better health status and experience of explosive pain and clinical rescue may be more prone to participate in EOLD. [ABSTRACT FROM AUTHOR]

Immune system dysregulation is associated with tumor incidence and growth. Here, we established an RNA-based individualized immune signature associated with prognosis for nonsmall cell lung cancer (NSCLC) to guide adjuvant therapy. We downloaded publicly accessible data on RNA expression and clinical characteristics of NSCLC from the Cancer Genome Atlas (TCGA). From immune-related genes (IRGs) retrieved from the immunology database and analysis portal (ImmPort) database, we then screened differentially expressed immune-related genes (DEIRGs). Using overall survival (OS) as a clinical endpoint, we identified 26 prognostic DEIRGs via univariate and multivariate Cox regression analysis, and then developed a risk model based on these 26 IRGs with an area under the curve (AUC) of 0.701, and its predictive ability independent from other clinical factors. We also downloaded tumor immune infiltrate data and analyzed the correlations between lymphocytic infiltration with our risk scores, but found no significant association. Furthermore, we retrieved 86 differentially expressed transcription factors (TFs) to assess their regulatory relationships with the 26 prognostic DEIRGs. In summary, we developed a robust risk model to predict survival in patients with NSCLC, based on the expression of 26 IRGs. It provides novel predictive and therapeutic molecular targets. [ABSTRACT FROM AUTHOR]