[Comparison of clinicopathological features and prognosis between adenocarcinoma of esophagogastric junction and adenocarcinoma of gastric antrum].

Objective: To compare the clinicopathological features and the prognosis between patients with adenocarcinoma of esophagogastric junction (AEG) and with adenocarcinoma of gastric antrum (AGA), and to investigate the prognostic factors of AEG and AGA.

Methods: A retrospective cohort study was performed on clinicopathological data of 239 AEG patients (AEG group) and 313 AGA patients selected simultaneously (AGA group) undergoing operation at Harbin Medical University Cancer Hospital from January 2001 to December 2012.

Inclusion Criteria: (1) receiving radical surgery (R0 resection); (2) AEG or AGA confirmed by pathological examination of postoperative tissue specimens; (3) without preoperative neoadjuvant radiotherapy or chemotherapy; (4) complete clinicopathological and follow-up data; (5) patients who died of non-tumor-related causes were excluded. Chi-square test and independent samples t-test were used to determine differences in clinicopathological factors between two groups. The overall survival (OS) of patients was compared by Kaplan-Meier method and Log-rank test. Multivariate prognosis analysis was performed using Cox proportional hazards regression model.

Results: As compared to AGA group, AEG group had higher proportion of male [82.0%(196/239) vs. 65.2%(204/313),χ²=19.243,P<0.001], older age [(60±10) years vs. (55±12) years, t=4.895, P<0.001], larger tumor diameter [(5.6±2.4) cm vs. (5.0±3.3) cm, t=2.480,P=0.013], more T4 stage[64.8%(155/239) vs. 55.6%(174/313),Z=-3.998, P<0.001], and more advanced tumor stage [stage III:60.7%(145/239) vs. 55.6%(174/313),Z=-2.564,P=0.010]. There were no statistically significant differences in serum albumin or hemoglobin between two groups (all P>0.05). The 5-year OS rate was 33.5% and 56.9% in AEG group and AGA group respectively and the median OS was 60.0(3.0-60.0) months and 33.6(3.0-60.0) months respectively; the difference was statistically significant (P<0.001). In AEG group, univariate analysis showed that differences of hemoglobin level (5-year OS rate: 24.0% for <130 g/L, 39.9% for ≥130 g/L, P=0.006), tumor diameter (5-year OS rate: 41.9% for <5 cm,28.8% for ≥5 cm, P=0.014), N stage (5-year OS rate: 42.2% for N0, 40.9% for N1, 31.7% for N2, 15.8% for N3a, 9.0% for N3b, P<0.001) and TNM stage (5-year OS rate: 56.2% for stage I, 38.5% for stage II, 28.3% for stage III,P=0.017) were statistically significant (all P<0.05); multivariate analysis revealed that the worse N stage was an independent risk factor of prognosis survival for AEG patients(HR=1.404,95%CI:1.164-1.693, P<0.001), and serum hemoglobin level ≥130 g/L was an independent protective factor of prognosis survival for AEG patients (HR=0.689,95%CI:0.501-0.946,P=0.021). In AGA group, univariate analysis showed that differences of serum albumin (5-year OS rate: 49.1% for <40 g/L, 61.1% for ≥ 40 g/L, P=0.021), tumor diameter (5-year OS rate: 74.2% for <5 cm, 39.9% for ≥ 5 cm, P<0.001), T stage (5-year OS rate: 98.3% for T1,83.3% for T2,50.0% for T3,36.8% for T4, P<0.001), N stage (5-year OS rate: 89.0% for N0, 62.3% for N1, 50.0% for N2, 33.9% for N3a, 10.3% for N3b, P<0.001) and TNM stage (5-year OS rate: 97.3% for stage I, 75.8% for stage II, 32.8% for stage III, P<0.001) were statistically significant (all P<0.05); multivariate analysis revealed that the worse T stage (HR=1.516,95%CI:1.060-2.167,P=0.023) and the worse N stage (HR=1.453,95%CI:1.209-1.747,P<0.001) were independent risk factors for prognosis of AGA patients.

Conclusions: As compared to AGA, AEG presents have poorer prognosis,and is easier to present with later pathological stage and larger tumor diameter. N stage and hemoglobin level are independent factors associated with the OS of AEG patients. T stage and N stage are independent factors associated with the OS of AGA patients.