Dysregulation of apoptosis occurs in different types of malignant tumors and is likely to influence the tumor evolution, as well as clinical prognosis. However, the limited number of studies investigating the predictive power of apoptosis-related genes (ARGs) in gastric cancer indicates a gap in the current research. 174 ARGs who differentially expressed were screened using public databases, including the Gene Expression Omnibus and the Molecular Signatures Database. Univariate and LASSO regression analyses were rigorous approaches to recognize the 12 optimal genes (CTHRC1, PDGFRL, VCAN, GJA1, LOX, UPP1, ANGPT2, CRIM1, HIF1A, APOD, RNase1, and ID1) that make up the prognostic risk model. Molecular mutations, related signaling pathways, and immune system characteristics in different subgroups defined by the risk model were analyzed using different R packages. Moreover, based on the database of Genomics of Drug Sensitivity in Cancer, chemotherapy sensitivity was predicted among the risk subgroups. As a result, there were differences in mutation profiles, signaling pathways, and infiltrated immune cells between patients in various risk groups. Moreover, the low-risk group displayed greater sensitivity to chemotherapy than the high-risk group. Risk model provided a better prognostic value than the T, N, and M stages, according to the receiver operating characteristic curve. Finally, in a nomogram, the risk model and clinical factors were combined to visualize the survival rates of patients with GC. In response to the differential expression of apoptosis-related genes, a novel model for predicting the prognosis of GC patients was developed. This model may be highly valuable for guiding doctors to deliver treatment plans tailored to the need of patients with GC.