050: Modeling Effects of Lymphodepletion Therapy on CAR T-cell & Cytokine Dynamics

Statement of Purpose, Innovation or Hypothesis: Lymphodepletion of host T-cells is thought to improve the ability of CAR T-cells to expand and persist. Recent studies have shown a higher probability of leukemia-free survival and lower risk of relapse in patients achieving adequate fludarabine (FLU) exposure [1, 2]. A mathematical model was developed to understand the dynamics between host T-cells, CAR T-cells, and cytokines and was used to test hypotheses on how changes in lymphodepletion affected CAR T-cell persistence and cytokine dynamics. The model was parameterized using data from an early phase clinical study of CD 19 CAR T-cells.

Description of Methods & Materials: This study used data from 24 patients ≤21 years old with relapsed/refractory B-ALL treated with CD19.41 BBz-CAR T-cells (NCT03573700) [3]. Results from the first 12 patients were presented in Talleur, AC et al. [3]. Data included lymphodepletion dosage of FLU, serial CAR T-cell concentrations and 38 different cytokine levels for at least 28 days post infusion. The mathematical model consisted of a system of nonlinear ordinary differential equations used to describe the kinetics between the host lymphocytes Panetta et al. [4], cytokines (indirect response model), and CAR T-cells. Parameters were estimated using nonlinear mixed-effects modeling methods and were used to calculated the extent of CAR T-cell exposure, area under the concentration curve (AUC) over 28 days, and the increase in cytokine levels relative to pre-therapy levels. Simulations were used to quantify how changes in FLU dosage affects the dynamics of cytokines and CAR T-cells.

Data & Results: The parameterized model was used to quantify changes in CAR T-cells and cytokines. We observed a median (range) AUC over 28 days of CAR T-cells of 1.4x106 3 copies*days/µg genomic DNA of (1.1x105, 6.0x10) a median (range) increase in and IL15 levels of 4.6X (1.3X, 9.5X) and 4.1X (0.5X, 22X).We did not observe a difference inCAR T-cell AUC between individuals infused with 1x106 cells/kg or 3x106 cells/kg CAR+ T cells. However, CAR T-cell AUC trended higher in the group of individuals with complete response at week 4 (1.5X higher; p=0.12 or probability of week 4 response > 80% if CAR T-cell AUC > 2x106 vs between 40 and 80% otherwise). Next, simulations evaluated the effects of FLU cumulative exposure on CAR T-cell exposure using the median expected cumulative FLU exposure simulated in three age groups: 10 mg×hr/L for age < 6; 15 mg×hr/L for age between 10 and 15; and, 25 mg×hr/L for age greater than 15, given a FLU dose of 25 mg/m2X3 QD. The simulations showed an increased CAR T-cell AUC of 70% when the FLU exposure increased from 10 to 15 mg×hr/L and an additional 54% when the FLU exposure increased from 15 to 25 mg×hr/L.

Interpretation, Conclusion or Significance: The simulations based on these data and provide insights into how changes in lymphodepletion affect CAR T-cell exposure. The significance of this model is that it provides us with an efficient method to test hypotheses on how changes in the model parameters (e.g. lymphodepletion or cytokine related parameters) can affect CAR T-cell exposure and persistence.

Disclosures:

Citations/References: 1. Fabrizio, V.A., et al., Optimal fludarabine lymphodepletion is associated with improved outcomes after CAR T-cell therapy. Blood Adv, 2022. 6(7): p. 1961-1968. 2. Dekker, L., et al., Fludarabine exposure predicts outcome after CD19 CAR T-cell therapy in children and young adults with acute leukemia. Blood Adv, 2022. 6(7): p. 1969-1976. 3. Talleur, A.C., et al., Preferential expansion of CD8+ CD19-CAR T cells postinfusion and the role of disease burden on outcome in pediatric B-ALL. Blood Advances, 2022. 6(21): p. 5737-5749. 4. Panetta, J.C., et al., Using Pharmacokinetic and Pharmacodynamic Modeling and Simulation to Evaluate Importance of Schedule in Topotecan Therapy for Pediatric Neuroblastoma. Clinical Cancer Research, 2008. 14(1): p. 318-325.

Additional Information/Authors: none