R-CHOP has been recognized as standard therapy of DLBCL. In Wohrer’s report, R-CHOP for the early stage of gastric DLBCL resulted in 87% CR, with 12 of 15 patients alive at 15 months after chemotherapy . Although our study was small sample size, all the stage I patients obtained CR, with both OS and PFS at 100%, and also stage II-IV patients obtained long survival. We should consider cases with bleeding and/or potential bleeding from deep ulcerative lesions, always paying attention to the possibility of repeated bleeding or perforation of stomach after chemotherapy. Although such cases were included in this study, we were able to complete R-CHOP safely.
The dose of radiotherapy should be discussed. In the SWOG0014 study, the dose of involved field radiation was 40-46 Gy . A Japanese multi-center phase II trial reported using 40.5 Gy of radiation . In another phase II study, four cycles of CHOP with IFRT of 40 Gy was adjusted to early stage of primary gastric DLBCL . The IELSG4 study compared chemotherapy only without rituximab treatment versus chemotherapy with involved field radiotherapy. In this study, cases were limited to stage I to stageIIE, and they evaluated after four cycles of CHOP-like regimens. Then if the cases obtained CR, they were randomized to an additional two cycles of chemotherapy and 30 Gy involved field RT. There were no significant differences in OS between the two arms, though DFS was significantly better than radiotherapy groups, and four cases (18%) relapsed in the chemotherapy-alone group . In our study, we planned the dose of radiation as 30.6 Gy in CR cases and 39.6-40 Gy in PR cases based on NCCN guidelines for DLBCL  and there were no relapses. This consensus was based on expert opinion that the volume and dose of radiation therapy should be as minimimal as possible because the gastric DLBCL is surrounded with at-risk organs such as liver, kidneys and heart. Radiation-induced acute toxicities such as nausea and appetite loss and delayed toxicities such as gastric ulcer and renal or liver dysfunction could be reduced by lower dose of radiation. The risk of secondary malignancy should be evaluated after 10 years following this treatment. It is a very important point of treatment for gastric DLBCL, although it is difficult to conduct a randomized study specifically focusing on the doses of radiation therapy, mainly because of patient accrual.
Furthermore, we administered eight doses of rituximab in the treatment, compared to four doses in the SWOG0014 study . To complete infusion of eight cycles of rituximab during chemotherapy, we decided on weekly administration of rituximab. The eight doses of rituximab might have contributed to our good results.
In this our present report, most patients dropping chemotherapy were stage IV and high score of aa-IPI, and their case were complicated by severe general condition at diagnosis. All patients who completed R-CHOP with or without radiotherapy were able to obtain long PFS and OS. We therefore recommend R-CHOP therapy as a promising treatment for gastric lymphoma, carefully selecting the dose of radiotherapy dependent on chemo-sensitivity.