Hepatocellular carcinoma (HCC) is a highly heterogeneous disease, which makes the prognostic prediction challenging. Abnormal peroxisomes can promote the development of cancers. This study aimed to construct a prognostic model based on peroxisome-related genes and identify its prognostic prediction and immune distinction abilities in HCC. The prognostic model was constructed based on The Cancer Genome Atlas (TCGA) and The International Cancer Genome Consortium (ICGC). Kaplan-Meier curve, time-dependent receiver operating characteristic curve and Cox analysis were used to evaluate the model. The immune status, tumor microenvironment, drug sensitivity and expression levels of the mRNA and protein between HCC and adjacent non-tumorous tissues were analyzed and compared. A prognostic model of 9 peroxisome-related genes was established and validated. Overall survival was markedly higher in the low-risk group relative to the high-risk group. The risk score was an independent prognostic factor. Tumor-related pathways were enriched in the high-risk group and the HCC patients in high-risk group showed depleted immune status. Furthermore, immune checkpoint-related genes, cell cycle-related genes, and multidrug resistance-related genes were overexpressed in the high-risk group. The expression levels of prognostic genes were negatively related to the anti-tumor drugs sensitivity. In addition, the expression level of each prognostic gene in HCC tissues was higher than that in adjacent non-tumorous tissues in an independent sample cohort and the similar results were found in most cancer types. A signature based on the nine peroxisome-related genes is a promising biomarker of HCC and is beneficial to the realization of individualized treatment.