Background: A recent Intergroup trial demonstrated a significant survival advantage of postgastrectomy chemoradiation in gastric cancer patients, primarily because of a reduction of a relative locoregional recurrence (LRR) rate exceeding 70% in control patients. Radical gastrectomy with extended lymphadenectomy may reduce LRR, possibly affecting adjuvant treatment strategies.
Methods: Information on patients undergoing gastrectomy for potentially curable gastric cancer between 1990 and 2000 was reviewed. Patterns of first disease recurrence, survival, and disease-free survival were calculated, and predictors were identified.
Results: Gastrectomies were performed in 73 patients, with R0 resections in 82%. The median lymph node count was 24; positive nodes were found in 64% of patients. The median actuarial survival was 27 months, with a 5-year survival of 37%. Disease recurred in 35 patients (48%) after a median interval of 7 months (range,.5-67). Recurrent disease patterns included distant only (37%), peritoneal only (23%), peritoneal/locoregional (17%), all sites combined (14%), locoregional only (6%), and distant/locoregional (3%). Recurrence predictors were N3 category for distant recurrence (hazard ratio [HR], 10.2; P =.005), T3/4 category for peritoneal recurrence (HR, 4.8; P =.008), peritoneal relapse (HR, 40; P =.002), and a prior abdominal operation for LRR (HR, 3.2; P =.01). N2 disease had a distant failure risk similar to N1 status and an intraperitoneal failure risk similar to an N3 category.
Conclusions: Isolated LRR of gastric cancer after gastrectomy and extended lymphadenectomy is rare in this series. Most recurrences appeared diffusely at distant or peritoneal sites, and most LRRs occurred in conjunction with relapse at extraregional sites. Pathologic predictors of intraperitoneal (T3/4) or systemic failure (>N1) could be used to guide individualized, risk-oriented, adjuvant treatment.