Background: To explore the cellular immunity and cytokines status of NCP patients and to predict the correlation between the cellular immunity levels, cytokines and the severity of patients. Methods: 123 NCP patients were divided into mild and severe groups. Peripheral blood was collected, lymphocyte subsets and cytokines were detected. Correlation analysis was performed on the lymphocyte subsets and cytokines, and the differences between the indexes of the two groups were analyzed. Results: 102 mild and 21 severe patients were included. Lymphocyte subsets were reduced in two groups. The proportion of CD8 + T reduction in the mild and severe group was 28.43% and 61.9%, respectively; The proportion of B cell reduction was 25.49% and 28.57%; The proportion of NK cell reduction was 34.31% and 47.62%; The detection value of IL-6 was 0 in 55.88% of the mild group, mild group has a significantly lower proportion of patients with IL-6 higher than normal than severe group; There was no significant linear correlation between the lymphocyte subsets and cytokines, while significant differences were noticed between the two groups in CD4 + T, CD8 + T, IL-6 and IL-10. Conclusions: Low levels of CD4+T and CD8+T are common in severe NCP. IL-6 and IL-10 levels were higher in severe patients. T cell subsets and cytokines can be used as one of the basis for predicting the transition from mild to severe. Large number of samples are still needed to confirm the “warning value” of CD4 + T, CD8 + T IL-6 and IL-10.

Accumulating evidence suggests that hypoxia microenvironment and long non-coding lncRNAs (lncRNAs) exert critical roles in tumor development. Herein, we aim to develop a hypoxia-related lncRNA (HRL) model to predict the survival outcomes of patient with lower-grade glioma (LGG). The RNA-sequencing data of 505 LGG samples were acquired from The Cancer Genome Atlas (TCGA). Using consensus clustering based on the expression of hypoxia-related mRNAs, these samples were divided into three subsets that exhibit distinct hypoxia content, clinicopathologic features, and survival status. The differentially expressed lncRNAs across the subgroups were documented as candidate HRLs. With LASSO regression analysis, eight informative lncRNAs were selected for constructing the prognostic HRL model. This signature had a good performance in predicting LGG patients’ overall survival in the TCGA cohort, and similar results could be achieved in two validation cohorts from the Chinese Glioma Genome Atlas. The HRL model also showed correlations with important clinicopathologic characteristics such as patients’ age, tumor grade, IDH mutation, 1p/19q codeletion, MGMT methylation, and tumor progression risk. Functional enrichment analysis indicated that the HLR signature was mainly involved in regulation of inflammatory response, complement, hypoxia, Kras signaling, and apical junction. More importantly, the signature was related to immune cell infiltration, estimated immune score, tumor mutation burden, neoantigen load, and expressions of immune checkpoints and immunosuppressive cytokines. Finally, a nomogram was developed by integrating the HRL signature and clinicopathologic features, with a concordance index of 0.852 to estimate the survival probability of LGG patients. In conclusion, our study established an effective HRL model for prognosis assessment of LGG patients, which may provide insights for future research and facilitate the designing of individualized treatment.