Abstract Background: Following infusion of tisagenlecleucel, quantitative polymerase chain reaction (qPCR) can be used to measure in vivo kinetics of CAR transgene. The utility of qPCR to inform individual pt treatment decisions and measure functional persistence is an active area of research. Objective: To determine if CAR transgene detection by qPCR can be used to inform treatment decisions. Methods: Transgene levels in blood measured by qPCR from pivotal phase II studies in pts with relapsed/refractory pediatric and young adult acute lymphoblastic leukemia (pALL; ELIANA NCT02435849, N=75 and ENSIGN NCT02228096, N=29) and adult diffuse large B-cell lymphoma (DLBCL; JULIET NCT02445248, N=93) were used to investigate the relationship between transgene persistence and clinical response. Results: In both pALL and DLBCL, there were detectable CAR transgene levels by qPCR in both responders and nonresponders. The geometric mean maximal expansion (Cmax) was similar between responding and nonresponding DLBCL pts, while a 1.7-fold difference was observed in pALL pts. For both DLBCL and pALL, high inter-individual variability in transgene levels was noted. Similarly, higher CAR-T cell expansion from flow cytometry data pooled from responding pALL and pediatric chronic lymphocytic leukemia pts was observed vs nonresponding pts (Mueller KT, et al. Blood 2017), while the levels in DLBCL pts were comparatively lower in blood, likely due to partitioning of functional CAR-T cells to target sites including lymph nodes. The median time to maximal transgene level (Tmax) ranged from 9-10 days in DLBCL responders and nonresponders and pALL responders, while nonresponding pALL pts showed delayed expansion with median Tmax of 20 days. The median time corresponding to last quantifiable transgene level (Tlast) and the half-life estimated from the terminal slope of the cellular kinetic profile, two indicators of persistence, were higher in responding vs nonresponding pts with DLBCL and pALL. Despite this general trend, in some cases, transgene levels were not detectable at later time points in pts with continued response. Although the majority of responding pts show persistent transgene levels, some pts maintained a favorable clinical response despite a decline in transgene levels to below the level of quantification. Conclusion: In both pALL and DLBCL, CAR transgene is initially detected at high levels with high variability in both responders and nonresponders. While the majority of responding pts tend to have persistent transgene levels, some pts maintain favorable clinical responses despite a lack of quantifiable transgene. These results indicate that transgene levels in peripheral blood are not definitive to guide clinical decisions in pALL and DLBCL. Further analysis is needed to understand how CAR transgene levels relate to disease burden and duration of response. Citation Format: Rakesh Awasthi, Karen Thudium Mueller, Gregory A. Yanik, Constantine S. Tam, Susana Rives, Joseph P. McGuirk, Michael A. Pulsipher, Michael W. Boyer, Ulrich Jaeger, Andre Baruchel, Gary D. Myers, Peter Borchmann, Stephen J. Schuster, Heather Stefanski, Michael Bishop, Edward R. Waldron, Ozlem Anak, Abhijit Chakraborty, Eric Bleickardt, Stephane Wong, Lida Bubuteishvili Pacaud, Edmund K. Waller, Shannon L. Maude. Evaluation of in vivo chimeric antigen receptor (CAR) transgene levels in patients (pts) treated with tisagenlecleucel abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr CT237.