[Prognostic analysis of neoadjuvant chemotherapy for locally advanced gastric cancer with propensity score matching method]

Abstract

Objective: To compare the effects of neoadjuvant chemotherapy and adjuvant chemotherapy on the prognosis of patients with locally advanced gastric cancer using propensity score matching method.

Methods: Clinical data of patients with locally advanced gastric cancer undergoing open D2 radical gastrectomy between January 2012 and December 2014 at the Affiliated Tumor Hospital of Zhengzhou University were analyzed retrospectively. Sixty-five patients receiving neoadjuvant chemotherapy (NAC) were allocated into the NAC group and 1243 patients receiving postoperative adjuvant chemotherapy (AC) were allocated into the AC group.

Inclusion criteria: (1) age ranged from 18 to 75 years old, and biopsy specimen was confirmed as adenocarcinoma; (2) all the operative procedures were open D2 radical gastrectomy;(3)image examinations showed no distant metastasis or other unresectable factors.

Exclusion criteria: no open D2 radical gastrectomy, undergoing laparoscopic surgery, neoadjuvant chemotherapy course <2 cycles, without adjuvant chemotherapy, history of other malignancies, severe complications, incomplete data. SOX (tegafur-gimeracil-oteracil plus oxaliplatin) or XELOX (capecitabine plus oxaliplatin) was used as neoadjuvant and postoperative adjuvant chemotherapy regimen. One-to-two propensity score matching was performed to balance the covariance between two groups. Survival was analyzed using the Kaplan-Meier method. Differences between the curves were tested using log-rank test.

Results: After balancing the covariates including gender, age, tumor location, degree of differentiation, clinical stage, chemotherapy regimen, chemotherapy course and surgical approach, 195 patients were enrolled, including 65 patients of the NAC group and 130 patients of the AC group. The number of harvested lymph nodes in NAC and AC group was 22.3±4.6 and 22.6±5.1 respectively, without statistically significant difference(t=1.125, P=0.263). Pathological response assessment for NAC group showed TRG0 in 6 cases, TRG1 in 8 cases, TRG2 in 17 cases, TRG3 in 34 cases; sensitive (TRG 0 to 2) in 31 cases (47.7%), non-sensitive in 34 cases (52.3%). The 3-year progression-free survival rate of NAC and AC group was 73.6%(95%CI: 62.8-84.3) and 69.9%(95%CI:62.1-77.7) respectively, which was not significantly different(P=0.361). The 3-year overall survival rate of NAC and AC group was 80.0%(95%CI:70.2-89.8) and 74.6%(95%CI:67.2-82.0) respectively, which was not significantly different as well(P=0.387). Subgroup analysis revealed that the 3-year progression-free survival rate and 3-year overall survival rate of sensitive patients in NAC group were 83.3%(95%CI:70.0-96.6) and 87.1%(95%CI:75.3-98.9) respectively, which were significantly higher than 62.4%(95%CI:46.1-78.7, P=0.037) and 70.2%(95%CI:54.7-85.7, P=0.033) of non-sensitive patients in NAC group, and those in AC group(P=0.044 and P=0.040).

Conclusions: Effects of neoadjuvant chemotherapy and postoperative adjuvant chemotherapy on the prognosis of patients with locally advanced gastric cancer are similar. Patients who are sensitive to neoadjuvant chemotherapy have better prognosis. It may be beneficial to improve prognosis that some appropriate patients with locally advanced gastric cancer are screened for neoadjuvant chemotherapy.