Tisagenlecleucel cellular kinetics, dose, and immunogenicity in relation to clinical factors in relapsed/refractory DLBCL

Abstract

The anti-CD19 chimeric antigen receptor (CAR)–T cell therapy tisagenlecleucel was evaluated in the global, phase 2 JULIET study in adult patients with relapsed/refractory diffuse large B-cell lymphoma (DLBCL). We correlated tisagenlecleucel cellular kinetics with clinical/product parameters in 111 patients treated in JULIET. Tisagenlecleucel persistence in responders and nonresponders, respectively, was demonstrated for 554 and 400 days maximum by flow cytometry and for 693 and 374 days maximum by quantitative polymerase chain reaction (qPCR). No relationships were identified between cellular kinetics (qPCR) and product characteristics, intrinsic/extrinsic factors, dose, or immunogenicity. Most patients with 3-month response had detectable transgene at time of response and continued persistence for ≥6 months. Expansion (maximal expansion of transgene/CAR-positive T-cell levels in vivo postinfusion [Cmax]) was potentially associated with response duration but this did not reach statistical significance (hazard ratio for a twofold increase in Cmax, 0.79; 95% confidence interval, 0.61-1.01). Tisagenlecleucel expansion was associated with cytokine-release syndrome (CRS) severity and tocilizumab use; no relationships were observed with neurologic events. Transgene levels were associated with B-cell levels. Dose was associated with CRS severity, but this was not statistically significant after adjusting for baseline tumor burden. In contrast to the results from B-cell precursor acute lymphoblastic leukemia (B-ALL) and chronic lymphocytic leukemia, similar exposure was observed in DLBCL in this study regardless of response and expansion was lower in DLBCL than B-ALL, likely from differences in cancer location and/or T-cell intrinsic factors. Relationships between expansion and CRS severity, and lack of relationships between dose and exposure, were similar between DLBCL and B-ALL. Tisagenlecleucel cellular kinetics in adult relapsed/refractory DLBCL improve current understanding of in vivo expansion and its relationships with safety/efficacy endpoints. This trial was registered at www.clinicaltrials.gov as #NCT02445248.

Chimeric Antigen Receptor T Cell Therapies: A Review of Cellular Kinetic‐Pharmacodynamic Modeling Approaches

“…16,17 Because we are still in the early stages of CAR-T cell model development and because rich cellular kinetic data are collected for each patient, standard noncompartmental analysis approaches for characterizing the dose-exposure-response relationship are the main analysis used to support drug development. 4,5 These relatively simple approaches are suitable for answering many questions about dose selection and intrinsic and extrinsic factors that impact exposure and response. However, some questions, such as predicting the optimal timing of a drug combination or predicting a safe and effective first-in-human dose based on preclinical data, can benefit from the more mechanistic models shown in Figure 1, which account for target engagement, immunophenotype differentiation, and biodistribution.…”

“…A model that includes CAR-T cell distribution to target tissue ( Figure 1C) could address why different profiles for cellular kinetics, safety, and efficacy are observed for the same product (tisagenlecleucel) in different indications (eg, pediatric acute lymphoblastic leukemia, diffuse large B cell lymphoma, chronic lymphoblastic leukemia). 5,37,38 These differences may be because of the location(s) of the target, sensitivity of the targeted tumor, or variability in T cell function across the patient population. A physiological model for activated, non-targeted T cell distribution in mice has been developed, 12 which may also be modified to apply to humans.…”

“…Parameters like C max and AUC have been used to characterize the dose-exposureresponse relationship to support dose selection and for exploring the impact of intrinsic and extrinsic factors on exposure, safety, and efficacy. 4,5 However, such analysis lacks the mechanistic interaction between tumor and CAR-T cell, which may be important in addressing questions such as the optimal time of tocilizumab dosing to treat cytokine-release syndrome or the optimal time of anti-PD-1 dosing to help reduce exhaustion and provide a better response. 6 One challenge in the application of model-informed drug development principles to CAR-T cell therapy, as demonstrated by almost every publication that we reviewed, is that there is no consensus yet in the pharmacometrics and systems pharmacology fields on the properties that a CAR-T cell mathematical model needs to have.…”

TCR Redirected T Cells for Cancer Treatment: Achievements, Hurdles, and Goals

“…An immune response against the murine-derived CD19 CAR was described before and after cell therapy. In the JULIET study (283), the majority of treated patients showed detectable levels of pre-existing anti-murine CD19-specific antibodies, that further increased upon CAR-T cells infusion. Nonetheless, the kinetics of engraftment was unmodified, and rejection barely detected, questioning the relevance of these markers in predicting CAR-T cells persistence.…”

Model-based Cellular Kinetic Analysis of Chimeric Antigen Receptor-T Cells in Humans

“…When comparing multiple trial results in the dose range of 10 7 ~ 10 9 , neither CAR-T cell expansion nor contraction was associated with CAR-T doses (Fig S9). Although a steep dose-response relationship was speculated 2 , a solid dose-dependency has not yet been found in many previous analyses 30,36 , potentially a result of variability in the T cell subset composition. Nevertheless, in the trial of MM 15 , cohorts 2 and 3 were treated with cyclophosphamide before the injection of a low (1-5 x 10 7 ) or high dose (1-5 x 10 8 ) of CAR-T cells.…”