The autophagy-related gene PEA15 is a potential prognostic biomarker for early-stage endometrial carcinoma
Background
The Cancer Genome Atlas (TCGA) molecular classification has advanced risk stratification for endometrial carcinoma but has demonstrated comparable survival outcomes between the microsatellite instability (MSI) and copy-number low (CN-L) subtypes. In this study, we aimed to identify potential autophagy-related molecular signatures to increase the precision of TCGA-based prognostic stratification in early-stage endometrial carcinoma.
Methods
Univariate Cox regression analysis of the TCGA-Uterine Corpus Endometrial Carcinoma cohort was used to identify autophagy-related genes associated with survival outcomes in patients with endometrial carcinoma. The candidates were analyzed by the Kaplan–Meier method. Multivariate Cox regression was used to assess whether PEA15 served as an independent prognostic factor, especially for the MSI and CN-L subtypes. We examined the correlation between PEA15 protein expression and patient survival through immunohistochemical analysis of tissue microarrays from our institutional cohort of stage I endometrial cancer patients.
Results
Univariate analysis revealed that NRG3, PEA15, DNAJB1, BAK1, DRAM1, KLHL24, ATF6, CDKN2A, MBTPS2, and UVRAG were significantly associated with survival outcomes in early-stage endometrial carcinoma patients. Multivariate analysis established PEA15 as an independent prognostic factor. Immunohistochemical analysis of tissue microarrays revealed that elevated PEA15 expression was significantly correlated with poorer overall survival and disease-free survival. Both univariate and multivariate Cox regression confirmed high PEA15 expression as an independent prognostic factor for recurrence in patients with stage I endometrioid adenocarcinoma.
Conclusions
The autophagy-related gene PEA15 is an independent prognostic biomarker in early-stage endometrial carcinoma, improving risk stratification between the MSI and CN-L subtypes. Immunohistochemical detection has clinical potential for molecular classification, offering opportunities for personalized postoperative management strategies.