STARD5 as a potential clinical target of hepatocellular carcinoma.

To reveal whether STARD5 is a potential biomarker for diagnosis and prognosis of HCC. Using gene expression omnibus and the cancer genome atlas (TCGA) to screen differentially expressed genes in HCC and STARD5 was selected by LASSO algorithm. Then, we analyzed the association between STARD5 and clinical characteristics of HCC patients in TCGA and International Cancer Genome Consortium. Meanwhile, the mRNA and protein level of STARD5 was also verified by collecting 87 cases of HCC patients' liver tissues using qRT-PCR and WB. Next, we applied gene set enrichment analysis (GSEA) for pathways analysis of STARD5. Finally, TIMER1.0 and TISIDB were used to explore the correlation of STARD5 with immune cell infiltration. The expression of STARD5 was lower in HCC and negatively correlated with tumor grade (p < 0.05), while high expression of STARD5 suggested a better prognosis for HCC patients (p < 0.01) and it could be an independent prognostic predictor (p < 0.001). Meanwhile, STARD5 also had strong diagnostic accuracy for HCC patients. GSEA revealed that STARD5-related genes were mainly enriched in E2F targets, G2M checkpoint and KRAS signaling. The TIMER1.0 and TISIDB databases found a negative correlation between STARD5 and tumor immune infiltrating cells. STARD5 could be used as a potential target for HCC diagnosis and prognosis.