Prognostic Factors of Esophagogastric Adenocarcinoma in a Single-Center Veteran Population

Introduction: Esophagogastric adenocarcinoma (EGA) includes esophageal (EA), gastroesophageal junctional (GEJA), and distal gastric (DGA) adenocarcinomas. Prognostic significance of clinicopathologic factors in those tumors remain obscure, especially for GEJA that has been inconsistently classified and staged. The aim of this study was therefore to determine prognostic features of EGA in 3 geographic tumor groups in a large single center cohort. Methods: This is a retrospective study of patients with proven EGA treated within the VA Boston Healthcare System over a 20-year period. Patient charts were reviewed to classify EGA tumors into EA, GEJA, and DGA subgroups, based on the tumor epicenter location information. All relevant clinical-pathological data were extracted from charts and tabulated using student t-test, Chi square and Fisher exact test. Histopathologic features of each EGA case with available histologic slides were determined by a single gastrointestinal pathologist. Overall survival (OS) was estimated with the Kaplan-Meier methods. Results: 347 eligible patients with a median age of 70 years (range: 48-94) were identified. All patients were male and 97.1% were White. Based on tumor epicenter location, EGAs were divided into EA (over 2 cm above the GEJ; N=63, 18.1%), GEJA (within 2 cm above and 3 cm below the GEJ; N=231, 66.6%), and DGA (over 3 cm below the GEJ; N=53, 15.3%). The median overall survival (OS) was the longest in EA (62.9 months), compared to GEJA (33.4), and DGA (38.1) (P < 0.001; Figure 1). Significant risk factors for OS (Table 1) included tumor location (P = 0.018), size (P < 0.001), differentiation (P < 0.001), adenocarcinoma subtype (P < 0.001), TNM stages (P < 0.001). Independent risk factors for OS included low-grade papillary adenocarcinoma (P < 0.05), mixed adenocarcinoma (P < 0.05), adenosquamous carcinoma (P < 0.05), and TNM stages (P < 0.05). Based on tumor differentiation, EGA was further divided into low-risk (common well-moderately differentiated tubular and low-grade papillary adenocarcinomas) and high-risk (uncommon adenocarcinoma subtypes) subgroups, whereby median OS was significantly longer in the low-risk (83 months) than in the high-risk (10 months) subgroups (P < 0.001; data not shown). Conclusion: EGA is heterogeneous with significantly better OS in EA than in GEJA or DGA and could be stratified into low- and high-risk subgroups with significantly different outcomes.Figure 1.: Overall survival is shown for 3 groups. The Kaplan-Meier method was used with a log rank test to estimate OS. Survivors were censored. OS was significantly longer in EA than in GEJA and DGA groups (P = 0.005), but not different between GEJA and DGA groups (P = 0.604). Table 1. - Prognostic value of clinicopathologic variables in overall survival of patients with esophagogastric adenocarcinoma Characteristic Total Number (%) Mean Overall Survival (months, mean + Standard Error) P value Number 347 (100) 70.0 Age (yrs) 0.094 < 59 43 (12.4) 58.7 ± 11.8 60-69 118 (34.0) 41.3 ± 5.7 70-79 121 (34.9) 35.3 ± 4.5 >80 65 (18.7) 29.7 ± 19.2 Tumor location 0.018 Distal Esophagus ( > 2 cm above GEJ) 63 (18.1) 62.9 ± 9.9 GEJ (2 cm above/3 cm below GEJ 231 (66.6) 33.4 ± 3.3 Distal stomach ( >3 cm below GEJ) 53 (15.3) 38.1 ± 7.5 Tumor size (cm) <0.001 < 0.9 34 (9.8) 84.7 ± 13.9 1.0-2.9 77 (22.2) 57.2 ± 6.8 3.0-3.9 158 (45.5) 50.5 ± 10.5 >4.0 158 (45.5) 20.6 ± 2.9 Unknown 37 (10.7).001 22.4 ± 5.4 Tumor differentiation < 0.001 Well-moderate differentiation 141 (40.6) 61.5 ± 5.8 Poor differentiation 206 (59.4) 24.1 ± 18.1 Tumor invasion depth < 0.001 T1 87 (25.1) 85.5 ± 8.0 T2 50 (14.4) 41.4 ± 6.4 T3-4 138 (39.8) 25.5 ± 3.8 Unknown 72 (20.7) 9.6 ± 1.7 Nodal metastasis < 0.001 N0 126 (36.3) 72.6 ±8.5 N1-3 142 (40.9) 23.2 ±3.8 Unknown 79 (22.8) 13.2 ±5.7 Distant metastasis < 0.001 Absent 110 (31.7) 73.5 ± 7.0 Present 152 (43.8) 14.7 ± 1.5 Unknown 74 (21.3) 34.7 ± 5.9 GEJ denotes Gastroesophageal Junction.