Sarcomas are heterogeneous connective tissue malignancies that have been historically categorized into soft tissue and bone cancers. Although multimodal therapies are implemented, many sarcoma subtypes are still difficult to treat. Lipids play vital roles in cellular activities; however, ectopic levels of lipid metabolites have an impact on tumor recurrence, metastasis, and drug resistance. Thus, precision therapies targeting lipid metabolism in sarcoma need to be explored. In this study, we performed a comprehensive analysis of molecular stratification based on lipid metabolism-associated genes (LMAGs) using both public datasets and the data of patients in our cohort and constructed a novel prognostic model consisting of squalene epoxidase (SQLE) and tumor necrosis factor (TNF). We first integrated information on gene expression profile and survival outcomes to divide TCGA sarcoma patients into high- and low-risk subgroups and further revealed the prognosis value of the metabolic signature and immune infiltration of patients in both groups, thus proposing various therapeutic recommendations for sarcoma. We observed that the low-risk sarcoma patients in the TCGA-SARC cohort were characterized by high proportions of immune cells and increased expression of immune checkpoint genes. Subsequently, this lipid metabolic signature was validated in four external independent sarcoma datasets including the CHCAMS cohort. Notably, SQLE, a rate-limiting enzyme in cholesterol biosynthesis, was identified as a potential therapeutic target for sarcoma. Knockdown of SQLE substantially inhibited cell proliferation and colony formation while promoting the apoptosis of sarcoma cells. Terbinafine, an inhibitor of SQLE, displayed similar tumor suppression capacity in vitro. The prognostic predictive model and the potential drug target SQLE might serve as valuable hints for further in-depth biological, diagnostic, and therapeutic exploration of sarcoma.