Background: Diffuse large B-cell lymphoma (DLBCL) is the most common type of lymphoma, mainly affecting elderly population. Treatment of elderly patients is challenging, and many clinical trials have been conducted to identify the optimal strategy. While rituximab plus CHOP (R-CHOP) therapy is considered as the standard, elderly patients are often frail and sometimes unable to tolerate the standard dose of CHOP therapy. In our institution, 5/6 (83%) and 7/12 (58%) doses of standard CHOP have been administered for patients aged 65-79 years and aged 80 or older, respectively. To evaluate the efficacy and tolerability of our reduced-dose R-CHOP therapy, we conducted a retrospective analysis of 100 patients with median age of 74 (65-86) years and presented the results in ASH 2016 meeting (Ono et al., abstract #4211). The outcome was quite satisfactory. In summary, the overall response rate (ORR) was 93%, and the complete response (CR) rate was 81%. Three-year overall survival (3-yr OS) was 78%. These results seem to be better than the outcome of LNH-98.5 trial (median age 69 years, ORR 83%, CR 75%, 2-yr OS 70%, Coiffier et al., NEJM 2002) and LNH03-6B trial (median age 70 years, ORR 86%, CR 74%, 3-yr OS 72%, Delarue et al., Lancet Oncol 2013), in which 8 cycles of standard-dose R-CHOP were scheduled for patients older than 60 years. There was no significant difference in patient characteristics between our data and those big trials. To find out factors that could play an important role in achieving this favorable outcome, we performed an additional analysis of our data. Methods: We applied the same database that we previously analyzed for ASH 2016. This database included clinical information of 100 patients aged 65 years or older with newly diagnosed DLBCL, who underwent R-CHOP therapy from August 2010 to December 2013 in our institution. We calculated relative dose intensity (RDI) from the doses of cyclophosphamide and doxorubicin and the intervals between each course compared with the standard dose (750 mg/m2 cyclophosphamide and 50 mg/m2 doxorubicin every 21 days). Comorbid conditions were classified by Charlson comorbidity index (CCI). We also analyzed time from diagnosis to initiation of chemotherapy (diagnosis to treatment interval; DTI) in 81 patients. Results: As previously reported, 14 patients were very elderly (≥80 years old) and 57 and 5 patients had an International Prognostic Index (IPI) score of at least 3 and CCI score of at least 3, respectively. Scheduled cycles of R-CHOP were completed in 87 patients and the median RDI was 0.81 in patients aged 65-79 years and 0.58 in patients aged 80 or older. Reduced-dose R-CHOP described above was completed as planned in most of patients. Survival predictors were analyzed using univariate analysis followed by Cox regression model. Multivariate analysis demonstrated that age and RDI had no significant impact on OS. Instead, the independent factors negatively impacting on OS were IPI score ≥3 (p < 0.001), CCI score ≥3 (p = 0.03), incidence of febrile neutropenia (p = 0.04) and not completing the scheduled cycles of R-CHOP (p = 0.001). The median DTI was 17 days. Patients with higher IPI score tended to receive chemotherapy immediately after diagnosis (p = 0.002). Patients with delay in therapy (DTI ≥17 days) had a better outcome (2-year OS was 97% vs. 72%, p = 0.02). Conclusion: In general, older age and low RDI are accepted as the negative prognostic factors. However, they had no impact on OS in this study. The current findings suggest that the completion of 6 or more cycles of R-CHOP plays an important role in achieving a better outcome even when the dose of CHOP is reduced. Although full-dose R-CHOP is still golden standard in the treatment of DLBCL, the appropriate dose reduction can be considered for older and frail patients. Disclosures No relevant conflicts of interest to declare.