Background: Approximately 80% of patients with diffuse large B-cell lymphoma (DLBCL) achieve a complete response (CR) after rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone therapy (R-CHOP) as first-line treatment but approximately 25% of patients with CR experience relapse. Disease recurrence generally occurs in the first 2 years following the first CR, but some patients relapse in the 5 years following the first CR. The optimal follow-up strategy for patients with non-Hodgkin's lymphoma has been under consideration. There are limitations to surveillance imaging, including risk of increase in healthcare costs, and increased anxiety due to the possibility of recurrence. To assess the role of surveillance imaging in the rituximab era, we retrospectively analyzed DLBCL patients, treated with a uniform treatment regimen, who relapsed after achieving the first CR.Patients and Methods: Between 2004 and 2009, DLBCL patients were newly diagnosed and were treated with 6 to 8 cycles of full-dose R-CHOP therapy (50 mg/m2 doxorubicin, 750 mg/m2 cyclophosphamide and 1.4 mg/m2 [maximum 2.0 mg/body] vincristine on day 1, 100 mg/body per day of prednisolone on days 1 to 5, and 375 mg/m2 rituximab per treatment cycle; treatment cycles were repeated every 3 weeks) at the 8 institutions of the Yokohama City University Hematology Group. Physical examinations were performed every 1 to 3 months, and periodical CT was generally performed every 6 months for 5 years after achieving CR. The relapses were categorized into 2 groups based on the modality of detection. One group consisted of the relapses that were detected by surveillance imaging (SI) in the absence of symptoms. The other group was composed of relapses that were detected by clinical symptoms, abnormal laboratory data, or SI in the presence of symptoms. Overall survival (OS) was measured from the start of the initial treatment until the time of the last follow-up or death. Using Kaplan-Meier survival analysis and the log-rank test, we assessed OS based on the modality of relapse detection.Results: In total, 315 patients with DLBCL were diagnosed and treated with R-CHOP therapy. Further, 289 patients achieved the first CR. We identified 52 patients who relapsed after the first CR. The patients who relapsed were at a more progressed clinical stage and had a higher International Prognostic Index (IPI) score than the patients in CR, although there were no significant differences in the age, gender, or median duration of follow-up between the patients with or without CR. Of the 52 relapsed patients, 19 relapses (37%) were detected by routine SI in the absence of clinical symptoms. The remaining 33 relapses (63%) were detected by SI in the presence of clinical symptoms (CS) or were detected following unscheduled imaging due to the presence of clinical symptoms or abnormal laboratory findings. The most frequently observed clinical symptom in the CS group was lymphadenopathy (n = 13, 39%). Median observation period after the first relapse in the 24 survivors (10 in SI group and 14 in CS group) was 62 months. There were no significant differences in age, gender, clinical stage, IPI, or median relapse-free duration between the SI and CS groups. The OS curve is shown in Figure below. There was no significant difference (P = 0.23) in the 5-year OS rates in the SI and the CS groups.Conclusion: Periodic surveillance imaging does not prolong the OS in relapsed patients. There are more to be discussed about the follow up strategy for DLBCL. DisclosuresNo relevant conflicts of interest to declare.