In recent years, immune checkpoint inhibitor has achieved remarkable success in multiple cancer treatment. However, how to pre‐judge which patients are suitable for immune checkpoint inhibitor is a difficult problem. We use the existing public bioinformatics database to comprehensively analyze the relationship between clinical data of various cancers with immune checkpoint blocking molecules and long non‐coding RNAs (lncRNAs), and try to find the potential predictive value of lncRNA for immunotherapy with checkpoint inhibitors. In this study, we found that: (a) high expression of lncRNA MIR155 host gene (MIR155HG) was closely related to better overall survival (OS) in cholangiocarcinoma (CHOL), lung adenocarcinoma (LUAD), and skin cutaneous melanoma (SKCM), and have better disease‐free survival (DFS) in CHOL. Meanwhile, the high level of MIR155HG was associated with poorer OS in glioblastoma multiforme (GBM), kidney renal clear cell carcinoma (KIRC), brain lower grade glioma (LGG), and uveal melanoma (UVM). (b) The expression of MIR155HG was significantly correlated with infiltrating levels of immune cells and immune molecules, especially with immune checkpoint molecules such as programmed cell death protein 1 (PD‐1), PD‐1 ligand 1 (PD‐L1), and cytotoxic T lymphocyte‐associated antigen 4 (CTLA4) in most kinds of cancers. (c) Detection of clinical CHOL and liver hepatocellular carcinoma tissues confirmed that there was a strong positive correlation between MIR155HG expression and the levels of CTLA4 and PD‐L1. MIR155 host gene can be used as a prognostic marker in multiple cancers, and of great value in predicting the curative effect of immune checkpoint inhibitor therapy owing to it is closely related with immune cells infiltration and immune checkpoint molecules expression.