Background: The Chimeric antigen receptor (CAR) T-cell therapy has shown very promising future in relapsed/refractory diffused large B cell lymphoma (R/R DLBCL) patients resisted to second-line chemotherapy and failed to auto hematopoietic cell transplantation. Trials with murine CD19-CAR T-cell therapy have shown only 40%-54% complete remission (CR) rate in patients with R/R DLBCL. However, 30%-60% of patients with previous CR relapse within 1 year. It is critical to develop new strategies to improve survival of R/R DLBCL patients failed to murine CD19-CAR T cells therapy. Forty-two R/R DLBCL patients failed to murine CD19-CAR T cells therapy who accepted secondary humanized CAR T cells salvage therapy in our center were recruited to this studyAim: 1. Investigate the effect and safety of Secondary humanized CAR T cells salvage therapy in R/R DLBCL patients failed to murine CD19-CAR T cells therapy. 2. Analyze the influence factor of curative effect of secondary humanized CAR T cells, to predict the outcome of treatment.Method: We retrospectively analyzed 42 cases treated with secondary humanized CAR T cells salvage therapy after failing to murine CD19-CAR T cells therapy in our center, enrolled from 05/01/2018 to 05/01/2020. According to the new result of tumor immunohistochemical analysis, five patients still chose to target CD19, twenty-one targeted CD20, and sixteen targeted CD22. Pretreatment regimen before infusion of CAR T cells were chemotherapy comprising fludarabine (three doses of 30 mg/m2 daily) and cyclophosphamide (three doses of 300mg/m2 daily). The T-cells with specific CAR expression in peripheral blood were dynamically detected by flow cytometry. After CAR T-cells infusion, PET-CT were performed every 3 months to evaluate the state of disease. Patients were followed up till 06/15/2021Results: The incidence of cytokine release syndrome (CRS) was 84%, of which 22% was severe CRS (≥grade 3), as the incidence of immune effector cells associated neurologic toxicity (ICANS) was 7.14%, of which 2.83% was severe ICANS. The incidence of target organ damage was 12%. The rate of treatment-related mortality was 7%. In 3 months, complete remission rate was 26.2%, partial remission rate (PR) was 33.3% and the overall response rate (ORR) was 59.5%. Ten patients (26.2%) remained in complete remission at the cutoff date. The median of progression-free survival (PFS) length was 4.42 months (95%CI: 1.87-6.02). The median of overall survival (OS) length was 9.24 months (95%CI: 4.44-~). The 1-year overall survival rate was 38.1%. We also found high level of LDH, heavy tumor burden, no less than 4 IPI score, and double expression of MYC and BCL2 were risk factors of OS, PFS and CR. The patients who had achieved CR or PR after murine CD19-CAR T cells therapy had a favorable OS and PFS. The longer the patients had PFS during murine CD19-CAR T cells therapy, the longer PFS and OS during humanized CAR T cells therapy.Conclusion: 1. Secondary humanized CAR T cells salvage therapy improves survival of R/R DLBCL patients failed to murine CD19-CAR T cells therapy. 2. The incidence of severe CRS and treatment-related causality were relatively low. 3. The patients who got better curative effect during murine CAR T cells therapy prone to a longer survival time.Keywords: Secondary humanized CAR T cells therapy, murine CAR T cells therapy, R/R DLBCL, CRS, ICANS. DisclosuresNo relevant conflicts of interest to declare.