Molecular determinants for lymph node metastasis in clinically early-stage endometrial cancer

Abstract

Patients with occult lymph node metastasis in endometrioid-type endometrial cancer (EC) are prone to the development of recurrences and have worse outcomes compared with patients without lymph node metastasis. In the current study, the aim was to identify molecular parameters associated with lymph node metastasis in EC clinically early-stage disease. A univariate analysis of differentially expressed genes, proteins and clinicopathological parameters (including myometrial invasion and tumor grade) was performed, comparing EC patients with and without lymph node metastasis (n=262 patients from The Cancer Genome Atlas). Significant parameters were introduced in a multivariate model and a gene expression pathway analysis. Lymph node metastasis was associated with expression of 268 unique genes (P<0.001), 19 unique proteins (P<0.05), tumor grade and myometrial invasion in univariate analysis. Multivariate analysis demonstrated 10 genes independently associated with lymph node metastasis and 4 independently associated proteins. Myometrial invasion was the only independent clinicopathological parameter associated with lymph node status. The enrichment pathway analysis demonstrated that expression of epidermal growth factor receptor, Bcl2 antagonist of cell death and phosphatase and tensin homolog pathways were significantly involved in lymph node metastasis (P≤0.001). A gene expression signature to predict lymph node status in EC was created for future validation. Few studies have focused on the association between EC's molecular characteristics and nodal metastasis. Defining molecular risk factors for EC lymphatic nodal metastasis may help to individualize treatment and improve patient outcomes.

Keywords: clinicopathological and molecular parameters; endometrial cancer; lymph node metastasis.

Figure 1.

Gene expression signature as predictor of lymph node (LN) status. (A) Heatmap of the signature predictive of lymph node status. (B) Performance of the model in the training dataset measured by the receiver operating characteristic (ROC) curve. (C) Contingency table of level of agreement between TCGA data and the predictive model: kappa coefficient of 0.2. Patients classified correctly (Neg-Neg and Pos-Pos) are on blue background. Neg, patients with negative lymph node status; Pos, patients with positive lymph node status.

Figure 2.

Volcano plot of significant genes from univariate analysis. LN, lymph node. Black points represent significant genes; white points represent nonsignificant genes.