Abstract
BACKGROUND: Hepatocellular carcinoma (HCC) outcomes remain suboptimal. Myeloid-derived suppressor cells (MDSCs) exhibit notable immunosuppressive and pro-tumorigenic properties. However, the relationship with HCC remains insufficiently explored. METHODS: Using TCGA data, we developed an MDSC-associated lncRNA prognostic model and a clinical nomogram. To explore the model’s mechanistic basis and clinical significance, enrichment analysis, tumor mutation burden (TMB) analysis, tumor microenvironment (TME) evaluation, immunotherapy response prediction, and drug sensitivity assessment were performed. RT-qPCR was utilized to confirm lncRNA expression. RESULTS: A 7-lncRNA prognostic signature was established. High-risk patients exhibited significantly worse survival (p < 0.001). The nomogram demonstrated superior accuracy over models excluding the risk evaluation. Enrichment analysis exposed that metabolic and immune-associated pathways were predominant within low-risk cohort, while cell proliferation and gene expression regulation dominant across the high-risk population. Meanwhile, an increased TMB and a degraded TME appeared in the high-risk population. The drug responsiveness evaluation revealed that sorafenib, axitinib, and others exhibited enhanced effectiveness among the low-risk population. High-risk individuals displayed enhanced reactions to medications like staurosporine, savolitinib, and others. CONCLUSIONS: The prognostic model constructed based on the seven MDSCs-associated lncRNAs showed good application value in assessing prognosis and guiding clinical therapy. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s41065-026-00664-z.