Indications for adjuvant chemotherapy in patients with AJCC stage IIa T3N0M0 and T1N2M0 gastric cancer—an east and west multicenter study

Abstract Purpose To determine the indications for adjuvant chemotherapy (AC) in patients with stage IIa gastric cancer (T3N0M0 and T1N2M0) according to the 7th American Joint Committee on Cancer (AJCC). Methods A total of 1593 patients with T3N0M0 or T1N2M0 stage gastric cancer were identified from the Surveillance, Epidemiology, and End Results (SEER) database for the period 1988.1–2012.12. Cox multiple regression, nomogram and decision curve analyses were performed. External validation was performed using databases of the Fujian Medical University Union Hospital (FJUUH) (n = 241) and Italy IMIGASTRIC center (n = 45). Results Cox multiple regression analysis showed that the risk factors that affected OS in patients receiving AC were age > 65 years old, T1N2M0, LN dissection number ≤ 15, tumor size > 20 mm, and nonadenocarcinoma. A nomogram was constructed to predict 5-year OS, and the patients were divided into those predicted to receive a high benefit (points ≤ 188) or a low benefit from AC (points > 188) according to a recursive partitioning analysis. OS was significantly higher for the high-benefit patients in the SEER database and the FJUUH dataset than in the non-AC patients (Log-rank < 0.05), and there was no significant difference in OS between the low-benefit patients and non-AC patients in any of the three centers (Log-rank = 0.154, 0.470, and 0.434, respectively). The decision curve indicated that the best clinical effect can be obtained when the threshold probability is 0–92%. Conclusion Regarding the controversy over whether T3N0M0 and T1N2M0 gastric cancer patients should be treated with AC, this study presents a predictive model that provides concise and accurate indications. These data show that high-benefit patients should receive AC.