: Since endoscopic findings of primary gastric lymphoma are ambiguous and diverse, it is not easy to distinguish them from gastric adenocarcinoma or secondary gastric lymphoma. The aim of this study was to investigate the difference in clinical and endoscopic features between primary gastric lymphoma and gastric involvement of lymphoma.

: Forty-eight patients were enrolled in this retrospective study between June 2008 and February 2017. The patients were divided into primary gastric lymphoma group (primary group, n = 18) and gastric involvement group (secondary group, n = 30) based on whether or not they carried gastric lesions alone. Patients’ clinical characteristics, endoscopic findings and pathologic data were retrospectively reviewed based on electronic medical records.

: The mean age of patients was 63.3 ± 13.1 years and 29 patients were female (60.4%). Diffuse large B-cell lymphoma pathology (81.3%), gastric body involvement (47.9%) and ulceroinfiltrative morphology on endoscopy (43.8%) were common features. Regardless of the two groups, the initial endoscopic diagnosis was considered as lymphoma only in 41.7%. Compared with the primary group, fundus (P = 0.035) and regional lymph node (P < 0.001) were significantly associated with the secondary group. However, there was no significant difference in endoscopic findings including location, size, number, and morphology of lesion.

: Endoscopic diagnosis of gastric lymphoma is a challenge. There is no difference in endoscopic findings between the primary and secondary groups even when confirmed separately. However, when the lesion is present in the fundus, we keep in mind the possibility of secondary gastric lymphoma.

There is still no consensus on the optimal treatment for primary gastrointestinal lymphoma (PGIL). The aim of this study was to compare surgery combined with chemotherapy and chemotherapy alone in PGIL.

We retrospectively reviewed and analyzed the treatment outcomes of 107 patients with primary gastrointestinal lymphoma diagnosed between March 1999 and December 2009 at Kosin University Gospel Hospital. Patients were divided into two groups: 35 patients who underwent surgery combined with chemotherapy (group A) and 72 patients who were treated with chemotherapy alone (group B). And we analyzed prognostic factors associated with short survival.

The 5-year progression free survival rates (PFS) of group A and B were 86.7% and 66.1%, respectively (P = 0.037), while the 5-year overall survival rates (OS) were 86.8% and 68.4%, respectively (P = 0.129). In multivariate analysis, Both PFS and OS were not changed by treatment strategies (surgery combined with chemotherapy or chemotherapy only). The international prognostic index (IPI) was the only independent predictive factor for PFS.

In our study, surgery combined with chemotherapy and chemotherapy only make no difference of survival rate. And further randomized prospective studies are needed to confirm a treatment strategies at improving survival outcomes in PGIL patients.