To evaluate the short-term efficacy and safety of neoadjuvant synchronous chemoradiotherapy (paclitaxel plus carboplatin regimen) in stage III adenocarcinoma of esophagogastric junction (AEG). Forty cases clinically diagnosed as stage III AEG were prospectively enrolled at the Department of Gastrointestinal Oncology Surgery, the First Affiliated Hospital of Hebei North University from December 2014 to November 2017 and then were randomly divided into paclitaxel plus carboplatin combined with synchronous radiotherapy group(neoadjuvant group) and direct operation group. Inclusion criteria was as follows:(1) AEG was diagnosed by gastroscopic biopsy and III stage was confirmed by ultrasound endoscopy and spiral CT;(2) physical strength score ≥70, and age ≤75 years old; (3) no contraindications of chemoradiotherapy and operation. Exclusion criteria was as follows:(1) patients voluntarily withdrew or refused the treatment;(2) occurrence of severe anaphylaxis; (3) uncontrollable events happened during treatment and treatment was unable to continue;(4) tumor developed obviously during treatment. Preoperative neoadjuvant synchronous chemoradiotherapy used TP regimen: paclitaxel 80 mg/m², drug concentration-time area under curve of carboplatin= 1.5 mg×ml⁻¹×min⁻¹, once per week for 9 weeks; radiotherapy began at the second week, 40 Gy/20 F, completed within 4 weeks. Operative procedure of both groups was radical resection of cardiac cancer(D2). Postoperative chemotherapy regimen was oral Tegafur(Gimeracil and Oteracil potassium). The side effects, diet situation, change of gastroscopic image after treatment in patients of neoadjuvant group were observed and efficacy evaluation of chemotherapy was performed according to solid tumor efficacy evaluation criteria of US National Cancer Institute. Operation-associated parameters, including R0 resection rate, lymph node metastasis, operative mortality and postoperative complications, were compared between two groups. There were no significant differences in baseline information between the two group (all P>0.05). One case in neoadjuvant group was excluded because of perforation at lesion site 7 weeks after chemotherapy. The side effects of 19 cases in neoadjuvant group were mainly alopecia (100%) and marrow inhibition (68.4%), while 3-4 degree side effects were alopecia(8/19,42.1%), leukopenia (3/19, 15.8%) and neutropenia(3/19, 15.8%). Complete remission was observed in 4 cases; partial remission was observed in 13 cases and stable disease in 2 cases, with an objective response rate of 89.5% and a disease control rate of 100%. Before neoadjuvant chemotherapy, 16 cases were difficult to take liquid diet and 3 cases received liquid diet only, while after 12 weeks of neoadjuvant chemotherapy, all the 19 cases received normal diet. Besides, after neoadjuvant chemotherapy, gastroscopic examination showed close healing of cardiac ulcer, disappearance of swelling, and renewal of normal mucosa. Compared to direct operation group, neoadjuvant group had less number of positive lymph node (4.9±3.6 vs. 8.8±2.8, P<0.05) and higher R0 resection rate (94.7% vs. 50.0%, P<0.05). Total number of harvested lymph node was not significantly different between two groups (19.1±2.5 vs. 18.6±7.0, t=0.326, P=0.746). There was no surgical death in either group. One case in direct operation group developed postoperative inflammatory obstruction. No associated complication was found in neoadjuvant group. Paclitaxel plus carboplatin combined with synchronous radiotherapy can elevate the R0 resection rate of patients with stage III esophagogastric junction adenocarcinoma, without increasing operative mortality and postoperative complications.