Proximal gastrectomy may be a reasonable choice for patients with selected proximal advanced gastric cancer: A propensity score-matched analysis.

BACKGROUND:Whether proximal gastrectomy (PG) can be applied to patients with proximal advanced gastric cancer (AGC) remains controversial. We aimed to explore the oncological safety of PG for proximal AGC in this study. METHODS:452 patients undergoing surgery for proximal AGC in the Affiliated Cancer Hospital of Nanjing Medical University were enrolled in this study. 329 patients with AGC were finally analyzed, of which 254 patients underwent total gastrectomy (TG) and 75 patients underwent PG. We used propensity score-matched (PSM) analysis to reduce biases. RESULTS:After PSM, 67 patients with proximal AGC were included in the PG group and TG group, respectively. The estimated 5-year OS rates for TG and PG group after PSM were 64.3% and 74.9%, respectively, and no significant difference in OS existed between the two groups (p = 0.275). Multivariate analysis showed that PG was not an independent prognostic factor. Incidence of metastasis in No.5 or 6 lymph node (LN) station was significantly higher in the patients with pathological T4 and Borrmann III tumors (9.9% and 10.6%) than those with pathological T2/3 and Borrmann I/II tumors (2.2% and 3.3%). No metastasis was observed in No.5 or 6 LN station in patients with pathological T2/3 tumors or Borrmann I/II tumors when tumor size was ≤4 cm. CONCLUSIONS:PG is a reasonable choice for patients with selected proximal AGC, especially for those with tumors of size ≤4 cm, Borrmann type I/II, and pathological T2/3. Future prospective randomized trials should be conducted first in patients with these specific proximal tumors.