Return To Abstract Listing

Gastrointestinal Imaging

E2749. Intestinal and Diffuse Gastric Cancer:Retrospective StudyComparing Primary Sites and Spread Patterns

Korivi B,  Sun J,  Asran A,  Patnana M,  Bhosale P. MD Anderson Cancer Center, Houston, TX

Address correspondence to B. Korivi (

Objective: To assess for differences in the primary sites and spread patterns of intestinal and diffuse gastric carcinoma subtypes.

Materials and Methods: A retrospective IRB approved study was performed. Imaging data base was mined for patients with gastric cancer. 105 treatment naïve patients were included, 69 patients had intestinal, and 36 patients had diffuse gastric carcinoma. Patient demographics, pathologic data, tumor stage, primary tumor site and metastases sites were recorded. Summary statistics were performed. Two-sample t-test was used to compare two different histologic subtypes. Pearson’s chi squared test was used to correlate tumor pathology with metastatic sites. Kaplan–Meier survival curves were generated. Log-rank test was used to compare survival curves. To evaluate the relationship between the primary sites and metastatic sites, a heat map with hierarchical clustering was created based on the observed frequencies of the primary tumors and the metastases. The cluster ordering was adjusted by frequency with the dendrogram structure imposed. P-values of 0.05 or less were considered statistically significant.

Results: Intestinal gastric carcinoma subtype was significantly associated with hepatic metastases p<0.001. The odds ratio is 10.8 with a 95% confidence interval ranging from 4.19 to 27.84. The risk of an intestinal cancer developing hepatic metastasis is about ten times higher than the risk of a diffuse cancer developing a hepatic metastasis. Diffuse gastric carcinoma was associated with omental metastases, gastrosplenic and transverse mesocolonic implants (p<0.026). Heat map measuring the observed frequencies of primary gastric cancer sites to their metastatic site demonstrated that primary site of tumor in the antrum was associated with peritoneal metastases. Patients with baseline peritoneal metastases had a significantly lower survival rate than patients without peritoneal metastases p=0.001. Patients without involvement of the omentum, gastrohepatic and gastrosplenic ligament has a better survival. p=0.018, p=0.004 p=0.001 respectively. Presence of hepatic metastases had no effect on survival p=0.136. Primary gastric tumors occurring at the greater curvature had better survival than those with primary sites at the GE junction, antrum, and lesser curvature (p<0.004).

Conclusion: There is a significant difference in spread patterns of intestinal gastric carcinoma when compared with diffuse gastric carcinoma, which will aid in radiographic detection of disease. The intestinal gastric cancer subtype is associated with hepatic metastases. The presence of hepatic metastases does not affect overall survival however peritoneal disease is associated with a shorter overall survival.