Abstract 1628 Introduction: Standard treatment for localized diffuse large B cell lymphoma (DLBCL) has been rather a short cycle of immunochemotherapy followed by involved field radiotherapy or prolonged cycles of immunochemotherapy. There is no convincing evidence in favor of either strategy. This retrospective analysis is an attempt to compare these treatment options. Methods: Patients were eligible if they had histologicaly newly diagnosed localized DLBCL by the Osaka Lymphoma Study Group (OLSG) central review panel and registered between 2003 and 2011, and received rituximab plus cyclophosphamide, doxorubicin, vincristine and prednisolone (R-CHOP) like immunochemotherapy with more than three consecutive courses as initial therapy. In this study, we defined a localized DLBCL as DLBCL with Ann Arbor stage I or non-bulky ( Results: With a median follow-up time of 34 months, neither OS nor PFS differ between these treatment groups. The 3-year OS were 85.5% in Chemo group and 96.2% in Chemo+RT group, respectively (P=0.225). The 3-year PFS were 74.3% in Chemo group and 89.7% in Chemo+RT group, respectively (P=0.185). A multivariate Cox regression model showed that Chemo+RT group have a tendency to improve PFS [hazard ratio =0.33; 95% confidence interval 0.10–1.07; P =0.066] of localized DLBCL compared with Chemo group. Grade 3 to 4 neutropenia and neutopenic fever were more frequent in patients with Chemo group (P Conclusion: For the treatment of localized DLBCL, although the difference between two treatment options was not significant in efficacy, short cycle of immunochemotherapy followed by radiation therapy seems to be superior to prolonged cycles of immunochemotherapy in terms of safety. Further studies are needed to define the optimal treatment option for localized DLBCL in the rituximab era. Disclosures: No relevant conflicts of interest to declare.