Objective:To investigate the correlation between serum microRNA (miR)-15a-5p and prognosis, neoadjuvant chemotherapy (NAC) response in patients with locally advanced gastric cancer (LAGC).Methods:The clinical data of 122 patients with LAGC who underwent surgery after NAC in Eastern Theater Air Force Hospital of the Chinese People′s Liberation Army from May 2016 to April 2020 were analyzed retrospectively. The general clinical data and laboratory examination results of the patients were recorded. The expression level of serum miR-15a-5p was detected by real-time fluorescence quantitative polymerase chain reaction, and the relationship between the expression of miR-15a-5p and different clinical characteristics in patients with LAGC was analyzed. The pathological response was evaluated by Becker tumor regression grading, in which patients with grade 1a, 1b and 2 were sensitive group and patients with grade 3 were resistant group.Results:The patients with LAGC were divided into high expression (>1.038) and low expression (≤1.038) according to the median miR-15a-5p of 1.038 with 61 cases each. The expression level of serum miR-15a-5p was related to the preference for spicy food, endoscopic ultrasonography (EUS)-T stage and EUS-N stage ( P<0.01 or <0.05). According to the evaluation result of pathological reaction, there were 47 cases in resistance group and 74 cases in sensitive group. The serum miR-15a-5p in resistance group was significantly higher than that in sensitive group: 1.69 (1.39, 1.97) vs. 0.99 (0.96, 1.02), and there was statistical difference ( Z =-8.55, P<0.01). The receiver operating characteristic curve analysis result showed that the area under the curve of serum miR-15a-5p predicting NAC response was 0.959 (95% CI 0.929 to 0.990), the optimal cut-off value was 1.049, the sensitivity was 100.0%, and the specificity was 85.1%. Multivariate Logistic regression analysis result showed that miR-15a-5p was an independent risk factor for NAC response in patients with LAGC ( HR = 1 880.840, 95% CI 123.510 to 28 641.846, P<0.01). Kaplan-Meier survival curve analysis result showed that the median overall survival time and median progression free survival time in patients with high expression of miR-15a-5p were significantly shorter than those in patients with low expression of miR-15a-5p (19 months vs. 62 months and 12 months vs. 51 months), and there were statistical differences (log-rank χ2 = 41.99 and 61.97, P<0.01); the 10-year overall survival rate and 10-year progression free survival rate in patients with high expression of miR-15a-5p were significantly lower than those in patients with low expression of miR-15a-5p (4.9% vs. 52.5% and 24.6% vs. 85.2%), and there were statistical differences (log-rank χ2 = 33.70 and 45.32, P<0.01). Multivariate Cox regression analysis result showed that R 0 resection and miR-15a-5p were the independent risk factors affecting the overall survival time and progression free survival time in patients with LAGC (overall survival time: HR = 1.945 and 3.487, 95% CI 1.033 to 3.660 and 2.112 to 5.759, P<0.05 or <0.01; progression free survival time: HR = 2.427 and 6.335, 95% CI 1.069 to 5.510 and 3.341 to 12.013, P<0.05 or <0.01). Conclusions:The increase of serum miR-15a-5p level is related to NAC response and poor prognosis in patients with LAGC. It can be used as a reliable biomarker to predict the prognosis and NAC response of LAGC.