Hepatocellular carcinoma (HCC) is a common digestive system disease with no curative treatment. Zinc finger protein 385D antisense RNA 2 (ZNF385D‑AS2) is a long non‑coding RNA (lncRNA) that has been predicted to function in human diseases, including several types of cancer. Yet, it has not been investigated in relation to liver cancer. Thus, the present study was designed with an aim to elucidate the prognostic significance of lncRNA ZNF385D‑AS2 in HCC. The Cancer Genome Atlas‑Liver Hepatocellular Carcinoma (TCGA‑LIHC) collection of data was utilized to analyze the expression of lncRNA ZNF385D‑AS2 in liver cancer. Then Chi‑square tests were used to evaluate the correlation between clinical characteristics and lncRNA ZNF385D‑AS2 expression. The significance of lncRNA ZNF385D‑AS2 in patient prognosis was evaluated using Kaplan‑Meier curves and Cox analysis. Concomitantly, Gene Set Enrichment Analysis (GSEA) was performed to analyze the most closely related cytological behavior. Finally, we used the Database for Annotation, Visualization and Integrated Discovery (DAVID) and KOBAS software and data from the Gene Expression Omnibus (GEO) database to analyze the possible competing endogenous RNA (ceRNA) network pattern as well as the co‑expression network in liver cancer. Based on the results, analysis of RNA‑Seq gene expression data for 303 patients with primary tumors revealed low expression of ZNF385D‑AS2 in liver cancer. Low expression of ZNF385D‑AS2 was found to be significantly associated with sex (P=0.050), T stage (P=0.049), M stage (P=0.040), N stage (P<0.001) and clinical stage (P=0.037). Patients with ZNF385D‑AS2 low‑expression liver cancers had a shorter median overall survival compared with the patients with ZNF385D‑AS2 high‑expression liver cancers (P=0.0079). Cox analysis identified ZNF385D‑AS2 low‑expression as an independent prognostic variable (AUC=0.594) for overall survival in liver cancer patients. Co‑expression and ceRNA predictive analysis data suggested that there may be a regulatory signaling axis between ZNF385D‑AS2 and miR‑96 and miR‑182. In conclusion, our results suggests that low expression of ZNF385D‑AS2 is predictive of a poor prognosis of liver cancer patients.