Objective To explore the efficacy of SOX regimen neoadjuvant chemotherapy (NAT) for advanced gastric cancer and its influencing factors. Methods The clinical data of 128 patients with advanced gastric cancer who received SOX regimen NAT (NAT group) and 128 patients with advanced gastric cancer who underwent surgical treatment without NAT (control group) in our hospital from Feb. 2010 to Apr. 2016 were analyzed retrospectively. A multivariate Cox proportional hazard regression model was used to analyze the factors affecting the overall survival, and a nomogram prediction model for predicting the efficacy of NAT was established based on related variables. Results The median survival time of patients in the NAT group was 31 months and that in the control group was 21 months. There was significant difference between the 2 groups (P<0.05). Multivariate Cox proportional hazard regression model analysis showed that age, tumor location, tumor size, carcinoembryonic antigen, carbohydrate antigen 19-9 (CA 19-9), clinical T stage, clinical N stage, lymph node metastasis and not receiving SOX regimen NAT were the independent risk factors for the prognosis of gastric cancer patients (all P<0.05). The nomogram prediction model was successfully constructed. It had been verified internally that the model had a good coincidence rate, good discrimination and accuracy. By comparing the prediction model with the TNM staging, it was found that the nomogram prediction model had high sensitivity and specificity. Conclusion SOX regimen NAT can improve the condition of patients with advanced gastric cancer. According to age, tumor size, tumor site, carcinoembryonic antigen, CA 19-9, clinical T stage, clinical N stage, lymph node metastasis and whether or not to receive SOX regimen NAT, the prediction model established by the nomogram can better predict the prognosis of advanced gastric cancer patients.