Research on prognostic models and biomarkers of lung adenocarcinoma using integrated proteomics and transcriptomics

Objective To screen prognostic biomarkers for lung adenocarcinoma by integrating proteomics and transcriptomics.

Methods Proteomics, transcriptomics and clinical characteristics data of lung adenocarcinoma patients were downloaded from the TCGA public database. The dataset was split into training set and validation set at a ratio of 7∶3. Univariate prognostic analysis of protein expression was conducted in the training set based on patients' clinical survival time, survival status, and protein expression data. A prognostic model for lung adenocarcinoma patients was constructed using the lasso-step cox method, and risk scores were calculated. Patients were divided into high-risk and low-risk groups based on the median risk score, and the prognosis of the two groups was analyzed. A prognostic nomogram model and calibration curves were constructed to clinically validate and correlate the model. The protein expression of the model was analyzed based on the HPA database, and enrichment analysis was performed on the risk proteins. Immunohistochemical and clinical characteristic analyses were conducted in 20 newly diagnosed lung adenocarcinoma patients from our hospital.

Results Five proteins associated with prognosis were screened out, and a risk protein model was constructed. The risk score had a predictive effect on the prognosis of lung adenocarcinoma patients. The risk model demonstrated strong and independent prognostic predictive ability. The nomogram model showed high accuracy in predicting individual prognosis. Furthermore, there were intrinsic relationship of the risk model and its calculated risk scores with clinical staging characteristics. HPA database analysis revealed significant overexpression of CD38, CD49B, ADAR1, and cdc25C4 in lung adenocarcinoma tissues. The 20 clinical specimens from our hospital validated the high expression of CD49B in newly diagnosed lung adenocarcinoma patients with distant metastasis and its sensitivity to treatment.

Conclusion The combined analysis of proteomics and transcriptomics for prognostic biomarkers of lung adenocarcinoma yields reliable results. CD49B plays a crucial role in lung adenocarcinoma, and the prognostic prediction model based on this gene is expected to provide important references for clinical treatment of lung adenocarcinoma.