David Ternant
INSERM UMR1327 ISCHEMIA, Tours, France
Objectives: CD19-targeted Chimeric Antigen Receptor (CAR) T-cells are recombinant receptors for CD19 antigen that redirect patients’ T lymphocytes toward CD19+ cell recognition and elimination. The kinetics of CAR T-cells was described in previous works[1-3], but neither using CART-cell counts nor with a “pharmacometrician” eye. This study aimed to describe the pharmacokinetic interindividual variability of CD19-targeted CART-cells in relapsed/refractory diffuse large B-cell lymphoma (DLBCL) patients.
Methods: In this prospective observational cohort (NCT04290000), 61 patients with DLBCL refractory to rituximab and chemotherapy treated with CD19-targeted CART-cells (Axicabtagene Ciloleucel or tisagenlecleucel) between May 2020 and April 2023 at Montpellier University Hospital (Montpellier, France), were assessed. Patients received a single injection of 2 million cells/kg. Blood samples were collected to measure CART-cell counts using flow cytometry. Joint kinetics of CART-cells (L) and their (latent) targets (R) was described using Friberg transit models.[4] These models consisted in a first proliferating compartment, three maturation compartments and a fifth compartment of circulating cells. For each transit model, parameters to be estimated were CART-cell first-order absorption (kL0), proliferation (resp. kLR and kprol), death (resp. kL and kD) and transit rate constants, the lasts being estimated as mean transit time (resp. MTTL and MTTR). CART-cell proliferation and CART-cell-induced target elimination were assumed to depend on both circulating target (kLR.L1.R5) and CART-cell counts (kD.L5.R5). Parameters were estimated using population analysis. Relationship between CAR T-cell kinetics and time-to-progression was described using multivariate parametric analysis using CAR T-cell kinetics or CD4 and CD8 peak counts. The following factors were tested on kinetic and survival parameters as covariates using likelihood ratio test (LRT): body weight (BW), age, sex, number of treatment line, delay since diagnosis (DIAG), TEP-scan metabolic tumor volume (TMTV), maximum standardized uptake value (SUVmax), age-adjusted international prognostic index (aaIPI) and denutrition.
Results: A total 531 CART-cell counts were available and were satisfactorily described by the double transit model. A peak of CART-cell counts was observed 10 days after injection, followed by a long-term residence. CART-cell kinetics was highly variable among patients. Estimated parameters were central distribution volume (V=8.4 L), kL0=0.030 day-1, kLR=2.2 mm-3.day-1, kD=143 mm-3.day-1, MTTL=4.7 days and MTTR=13.1 days. Older age was associated with an increased CAR T-cell proliferation (p<0.001) together with a decreased mean transit time (p<0.001); high total metabolic tumor volume was associated with increased mean transit time (p=0.0018) and targets death (p=0.023) whereas a longer time from diagnostic was associated with increased CAR T-cell death (p=0.0017). Higher CD4 CAR T-cell peaks (HR=0.0085, [0.0001–0.69], p=0.034) and a lower risk age-adjusted international prognostic index (HR=0.26 [0.10–0.67], p=0.0061) were significantly associated with decreased hazard of progression.
Conclusions: This is the first model that describes recursive interactions between CAR T-cell and target kinetics. The model explained CART-cell long-term residence by death-and-rebirth phenomenon, which may be due to repeated target-induced reactivation of memory CART-cells and suggested that higher CD4 CAR T-cell peak is associated with decreased hazard to progression.
References:
[1] Stein AM, et al. Tisagenlecleucel Model-Based Cellular Kinetic Analysis of Chimeric Antigen Receptor-T Cells. CPT Pharmacometrics Syst Pharmacol. 2019;8:285-295.
[2] Liu C, et al. Model-Based Cellular Kinetic Analysis of Chimeric Antigen Receptor-T Cells in Humans. Clin Pharmacol Ther 2021,109:716-727.
[3] Paixão EA, et al. Modeling Patient-Specific CAR-T Cell Dynamics: Multiphasic Kinetics via Phenotypic Differentiation. Cancers (Basel). 2022;14:5576.
[4] Friberg LE, et al. Model of chemotherapy-induced myelosuppression with parameter consistency across drugs. J Clin Oncol. 2002;20:4713-21.
Reference: PAGE 32 (2024) Abstr 10801 [www.page-meeting.org/?abstract=10801]
Poster: Drug/Disease Modelling - Oncology