A single-center study of treatment outcomes and survival in patients with primary gastric lymphomas between 1990 and 2003.

Primary gastric lymphomas are the most common extranodal non-Hodgkin's lymphomas and are divided into indolent (low grade) and aggressive (high grade) types. They are mainly the disease of middle age, with a male predominance reported by most of the studies. For several years, surgery played a central role in diagnosis, staging, and treatment of this entity, yet recently there has been a move away from a surgical approach to conservative treatment. To determine the role of surgery as the initial treatment modality, we performed this retrospective single-center research on 245 patients with primary gastric lymphoma who were treated according to our protocol between 1990 and 2003. The patients' characteristics, distribution of histological types, treatment results, and disease-specific survival were followed. According to the histology, 59.2% had diffuse large B-cell lymphoma (DLCL), 26.1% MALT lymphoma, 9.8% mixed lymphoma (indolent and aggressive at the same time), while other types were infrequent. In total, 161 patients (65.7%) were treated with surgical resection as the initial treatment, which was then followed or not by additional therapy (chemotherapy, chemotherapy and radiotherapy, radiotherapy) depending on the histological type of lymphoma and the extent of residual disease after surgery. In 84 patients (34.3%), the treatment approach was conservative. The selection of treatment (chemotherapy, chemotherapy and radiotherapy, radiotherapy or Helicobacter pylori eradication only) was based on the histological type of lymphoma, considering also the patients' physical condition. The disease-specific survival in the group of patients who underwent surgery was statistically significantly better than in patients who were treated conservatively (p=0.049). At 5 years, it was 96.9% for the group treated with surgery and 89.8% in patients treated conservatively. However, the results were biased, as the patients who were treated conservatively were either in a worse performance status or presented with a more extensive disease. Similarly, in the DLCL type the disease-specific survival was better in the surgically treated group (97.2%) than in the conservatively treated patients (89.2%). The difference was barely significant (p=0.046) and again the results have to be considered with caution due to the selection of patients in a worse performance status or with a more extensive disease for conservative treatment. In the MALT lymphoma and mixed lymphoma types, there were no differences in the disease-specific survival between both treatment groups. Regarding the statement that for conservative treatment patients were selected who were unsuitable for the resection on account of concomitant diseases or due to the fact that the process was inoperable, we believe that the conservative approach gives comparable outcomes to the approach including initial surgery. The existing evidence thus no longer justifies surgery as the standard initial treatment and preference should be given to conservative treatment approaches.