Construction of a lipid metabolism-based prognostic gene signature in cervical squamous cell carcinoma and validation of LIPG's oncogenic role

Background: Cervical Cancer, in which cervical squamous cell carcinoma (CSCC) accounts for 60-70% of cases, has a poor prognosis and poses a significant health threat to global patients. Lipid metabolism reprogramming is a key driver of tumor progression and tumor microenvironment (TME) regulation, making it a promising target for improving the efficacy of immunotherapy. This study aimed to construct a lipid metabolism prognostic signature (LMPS) in CSCC and identify key genes involved in tumor progression.

Methods: Through RNA-sequencing and clinical data from TCGA and GTEx databases, we identified differentially expressed lipid metabolism-related genes (DLMGs) and constructed the LMPS using machine learning algorithms. Next, the value of the LMPS was validated using the HTMCP database and the GEO database. Furthermore, the relationship between the LMPS and the TME was analyzed, including immune cell infiltration, immune checkpoint gene expression, and drug sensitivity. The key gene Lipase G (LIPG) was identified through machine learning methods and validated through cellular and Molecular Biology experiments.

Results: A total of 60 DLMGs were identified, with 9 DLMGs showing prognostic value. The LMPS was constructed using 6 genes (ACOT4, PLA2G2D, GAL3ST1, ALOX12B, PLA2G3, and LIPG), which effectively predicted patients' survival (AUC: 0.76, 0.75, 0.68 at 1, 3, 5 years, respectively). High LMPS was correlated with an immune-suppressive TME, reduced immune cell infiltration, lower human leukocyte antigen (HLA) and immune checkpoint gene expression, and higher IC50 values for common chemotherapy drugs. LIPG was identified as a key gene, showing higher expression in advanced Cancer stages. As revealed by functional experiments, LIPG promoted lipid accumulation and phosphatidylcholine (PC) hydrolysis in CSCC cells. Additionally, LIPG facilitated tumor progression through activation of the MAPK-p38 signaling pathway.

Conclusions: The LMPS was a valuable prognostic tool and was correlated with the TME and drug sensitivity. LIPG was a key regulator of lipid metabolism and facilitated CSCC development by hydrolyzing PC into lysophosphatidylcholine (LPC) and activating the MAPK-p38 signaling pathway. These findings may highlight the potential of targeting lipid metabolism for therapeutic intervention in CSCC.

Cervical squamous cell carcinoma; LIPG; Lipid metabolism; Machine learning; Prognostic signature; Tumor microenvironment.