Tumor regression and survival after perioperative MAGIC-style chemotherapy in carcinoma of the stomach and gastroesophageal junction.

Background: We assessed the effectiveness of perioperative MAGIC-style chemotherapy in our series focused on the tumor regression grade and survival rate.

Methods: We conducted a retrospective study of 53 patients following a perioperative regimen of epirubicin, cisplatin, and fluorouracil or capecitabine (ECF/X). Forty-four (83 %) neoplasias were located in the stomach and 9 (17 %) were located at the esophagogastric junction. Perioperative chemotherapy completion, resection, TNM staging, the tumor regression grade (Becker's classification) and survival were analyzed.

Results: Forty-five patients (85 %) completed the 3 preoperative cycles. R0 resection was achieved in 42 (79 %) patients. Thirty-five (66 %) patients completed the 3 postoperative cycles. Nine carcinomas (17 %) were considered major responders after preoperative chemotherapy. With multivariate analysis, only completion of perioperative chemotherapy (HR: 0.25; 95%CI: 0.08 - 0.79; p = 0.019) was identified as an independent prognostic factor for disease-specific survival. However, the protective effect of perioperative therapy was lost in patients with ypT3-4 and more than 4 positive lymph nodes (HR: 1.16; 95%CI: 1.02 - 1.32; p = 0.029). The tumor regression grade (major vs minor responders) was at the limit of significance only with univariate analysis. The 5-year overall and disease-specific survival rates were 18 % and 22 % respectively.

Conclusions: The percentage of major responder tumors after preoperative chemotherapy was low. Completion of perioperative ECF/X chemotherapy may benefit patients with gastric carcinomas that do not invade the subserosa with few positive lymph nodes.