LRG1 predicts the prognosis and is associated with immune infiltration in thyroid cancer: a bioinformatics study.

Abstract

The diagnostic and prognostic value of the leucine-rich alpha-2-glycoprotein 1 (LRG1) gene in thyroid cancer remains unclear. Using the Cancer Genome Atlas (TCGA) database, we conducted a bioinformatics analysis to determine the role of LRG1 in thyroid cancer. Data from 512 patients with thyroid cancer and 59 normal individuals were collected from TCGA database. The Kruskal-Wallis test and logistic analysis were used to examine the relationship between LRG1 expression and clinicopathologic characteristics. Cox regression and Kaplan-Meier analysis were used to determine the predictive value of LRG1 on clinical outcomes. Single-sample gene set enrichment analysis (ssGSEA) was used to reveal associations between LRG1 expression and immune infiltration levels in thyroid cancer. LRG1 was highly expressed in thyroid cancer (P < 0.001) and could effectively distinguish tumor tissue (area under the curve = 0.875) from normal tissue. Moreover, LRG1 was significantly correlated with pathological N stage (odds ratio (OR) = 2.411 (1.659-3.505), P < 0.001). Kaplan-Meier survival analysis revealed that patients with high LRG1 expression had better overall survival (hazard ratio (HR) = 0.30, P = 0.038). Cox regression analysis indicated that pathological M stage was a risk factor for progression-free interval (HR = 5.964 (2.010-17.694), P < 0.001). Using ssGSEA, we found that LRG1 expression was positively correlated with the number of T helper 1 cells (R = 0.435, P < 0.001), dendritic cells (R = 0.442, P < 0.001), and macrophages (R = 0.459, P < 0.001). LRG1 may be an important biomarker for predicting the prognosis of thyroid cancer and represent a suitable target for immunotherapy associated with immune infiltration.