PWII-004 - DEVELOPMENT OF A QUANTITATIVE SYSTEMS PHARMACOLOGY (QSP) MODEL TO EVALUATE THE THERAPEUTIC POTENTIAL OF ANTI-CD19 CAR-T IN SYSTEMIC LUPUS ERYTHEMATOSUS (SLE)

H. Park¹, G. Mugundu², A. Singh²; ¹Department of Pharmaceutical Sciences, College of Pharmacy, University of Tennessee Health Sciences Center, ²Takeda Pharmaceuticals.

Background: CAR-T cell therapy has proven enormous success in different oncology indications with approval of 6 drugs in hematological malignancies. More recently, CAR-T cells have been explored as a therapeutic option for autoimmune disorders. Repurposing some of the approved CD19 targeted CAR-T cells seems feasible for B-Cell mediated autoimmune diseases such as systemic lupus erythematosus (SLE). Consequently, numerous clinical trials are currently investigating CD19 CAR-Ts for the treatment of SLE. Overall pharmacological action of CD19 CAR-Ts in autoimmune disorders is still unknown. With very limited clinical dataset, the dose-exposure-response of CAR-T cells in autoimmune disorders is still unclear.

Methods: Under low data environments, we have developed a QSP model (Figure), which integrates dataset(s) such as 1) the cellular kinetics of CAR-T subsets and immunophenotypes, 2) lymphodepletion chemotherapy (LDC) mediated depletion of the host T-cells, 3) CAR-mediated depletion of CD19+ B cells, 4) dynamics of disease biomarkers including interferon-gamma, compliment C and proteinuria as well as 5) disease activity scores in SLE patients (SLEDAI). The study consisted of four female and one male patient aged between 18 and 24 years who were dosed with 1 million CD19 CAR+ T-cells/kg post standard LDC (1,000 mg/m2/day cyclophosphamide from day -5 to day -3 and 25 mg/m2/day fludarabine on day -3 before CAR-T cell infusion).

Results: The model was able to adequately describe the overall cellular kinetics of CAR+ cells, host-T cell kinetics post LDC. Differential kinetics of CAR-T subsets & phenotypes along with CD19+ B cell depletion were also well characterized. Based on correlation between each surrogate biomarker and disease score examined from multiple literature, model-predicted biomarker output was translated to SLEDAI score and compared with observed SLEDAI score before and after treatment to evaluate predictability of the model-based outcome.

Conclusion: Proposed QSP model is expected to be leveraged towards adequately describing dose-exposure-response relationship of CD19 CAR-T cells and facilitate further clinical development in autoimmune disorders.