Comprehensive Analysis and Reinforcement Learning of Hypoxic Genes Based on Four Machine Learning Algorithms for Estimating the Immune Landscape, Clinical Outcomes, and Therapeutic Implications in Patients With Lung Adenocarcinoma

Identification of hypoxia-related gene signatures based on multi-omics analysis in lung adenocarcinoma

Lung adenocarcinoma (LUAD) is the most common type of lung cancer and one of the malignancies with the highest incidence rate and mortality worldwide. Hypoxia is a typical feature of tumour microenvironment (TME), which affects the progression of LUAD from multiple molecular levels. However, the underlying molecular mechanisms behind LUAD hypoxia are not fully understood. In this study, we estimated the level of hypoxia by calculating a score based on 15 hypoxia genes. The hypoxia scores were relatively high in LUAD patients with poor prognosis and were bound up with tumour node metastasis (TNM) stage, tumour size, lymph node, age and gender. By comparison of high hypoxia score group and low hypoxia score group, 1820 differentially expressed genes were identified, among which up-regulated genes were mainly about cell division and proliferation while down-regulated genes were primarily involved in cilium-related biological processes. Besides, LUAD patients with high hypoxia scores had higher frequencies of gene mutations, among which TP53, TTN and MUC16 had the highest mutation rates. As for DNA methylation, 1015 differentially methylated probes-related genes were found and may play potential roles in tumour-related neurobiological processes and cell signal transduction. Finally, a prognostic model with 25 multi-omics features was constructed and showed good predictive performance. The area under curve (AUC) values of 1-, 3- and 5-year survival reached 0.863, 0.826 and 0.846, respectively. Above all, our findings are helpful in understanding the impact and molecular mechanisms of hypoxia in LUAD.

Prognostic markers of ferroptosis-related long non-coding RNA in lung adenocarcinomas

Ferroptosis is a recently established type of iron-dependent programmed cell death. Growing studies have focused on the function of ferroptosis in cancers, including lung adenocarcinoma (LUAD). However, the factors involved in the regulation of ferroptosis-related genes are not fully understood. In this study, we collected data from lung adenocarcinoma datasets of the Cancer Genome Atlas (TCGA-LUAD). The expression profiles of 60 ferroptosis-related genes were screened, and two differentially expressed ferroptosis subtypes were identified. We found the two ferroptosis subtypes can predict clinical outcomes and therapeutic responses in LUAD patients. Furthermore, key long non-coding RNAs (lncRNAs) were screened by single factor Cox and least absolute shrinkage and selection operator (LASSO) based on which co-expressed with the 60 ferroptosis-related genes. We then established a risk score model which included 13 LUAD ferroptosis-related lncRNAs with a multi-factor Cox regression. The risk score model showed a good performance in evaluating the outcome of LUAD. What's more, we divided TCGA-LUAD tumor samples into two groups with high- and low-risk scores and further explored the differences in clinical characteristics, tumor mutation burden, and tumor immune cell infiltration among different LUAD tumor risk score groups and evaluate the predictive ability of risk score for immunotherapy benefit. Our findings provide good support for immunotherapy in LUAD in the future.

Development and Validation of a Prognostic Nomogram for Lung Adenocarcinoma: A Population-Based Study

PURPOSE

To establish an effective and accurate prognostic nomogram for lung adenocarcinoma (LUAD). Patients and Methods. 62,355 LUAD patients from 1975 to 2016 enrolled in the Surveillance, Epidemiology, and End Results (SEER) database were randomly and equally divided into the training cohort (n = 31,179) and the validation cohort (n = 31,176). Univariate and multivariate Cox regression analyses screened the predictive effects of each variable on survival. The concordance index (C-index), calibration curves, receiver operating characteristic (ROC) curve, and area under the ROC curve (AUC) were used to examine and validate the predictive accuracy of the nomogram. Kaplan-Meier curves were used to estimate overall survival (OS).

RESULTS

10 prognostic factors associated with OS were identified, including age, sex, race, marital status, American Joint Committee on Cancer (AJCC) TNM stage, tumor size, grade, and primary site. A nomogram was established based on these results. C-indexes of the nomogram model reached 0.777 (95% confidence interval (CI), 0.773 to 0.781) and 0.779 (95% CI, 0.775 to 0.783) in the training and validation cohorts, respectively. The calibration curves were well-fitted for both cohorts. The AUC for the 3- and 5-year OS presented great prognostic accuracy in the training cohort (AUC = 0.832 and 0.827, respectively) and validation cohort (AUC = 0.835 and 0.828, respectively). The Kaplan-Meier curves presented significant differences in OS among the groups.

CONCLUSION

The nomogram allows accurate and comprehensive prognostic prediction for patients with LUAD.