Expression of long non-coding RNA SFTA1P in lung adenocarcinoma and its predictive value for prognosis

Abstract

Objective To investigate the expression of long non-coding (lnc) RNA SFTA1P in lung adenocarcinoma, and its association with clinical prognosis in order to construct and evaluate a prognostic prediction model based on SFTA1P. Methods Real-time fluorescence quantitative PCR was used to detect the expression of SFTA1P in 62 pairs of non-small-cell lung cancer (NSCLC) tissues and corresponding normal lung tissues adjacent to the cancer for the association between SFTA1P expression level and clinical prognosis. The data sets of patients diagnosed with lung adenocarcinoma were downloaded from TCGA database, the required variable data were extracted, and the clinical parameter information and SFTA1P expression of the 471 patients with lung adenocarcinoma were sorted out. Subsequently, variables were screened using univariate and multivariate Cox proportion risk regression models, and a nomogram model was constructed based on the final prognostic prediction model. For the internal validation of the prediction model, Bootstrap resampling method was adopted, and concordance index (C-index) and calibration curve were used respectively to evaluate the discrimination and calibration of the prediction model. Results The SFTA1P expression was significantly decreased in the lung cancer tissues than the para-cancerous tissues (0.692±0.103 vs 1.765±0.149, P < 0.05). There were totally 3 selected independent clinical variables correlating to survival rates in the patients, that is, pathological stage, radiotherapy and SFTA1P expression. In the nomogram model, the calibration curves of 1-year and 3-year survival rates showed that the predicted values had good consistency with the actual values, with a C-index of 0.69 and 95% confidence interval of 0.63~0.75. Conclusion The expression of lnc RNA SFTA1P can be used as an independent prognostic factor for lung adenocarcinoma. The nomogram model constructed from the above independent variables (including age) can predict the survival rate of the patients.

Keywords

• non-small cell lung cancer
• lung adenocarcinoma
• sfta1p
• prognosis
• proportional hazard model
• nomogram