(T1030-06-38) Optimizing Flow Cytometric Immune Phenotyping and CAR Persistence Determination to Enable Immune Monitoring of Patients under CAR T Cell Therapy

Purpose: Immune monitoring of patients that received autologous CAR T cell treatment is a central bioanalytical tool during clinical trials (e.g., for pharmacokinetic analysis), as well as for the therapeutic monitoring of patients treated in clinical routine. Using a combination of flow cytometric immune phenotyping and CAR persistence determination allows for a holistic immune monitoring approach. Precise and robust methods are the basis for measurements on patient samples in order to obtain reliable results.

Methods: Blood samples for method optimization were derived from 11 healthy donors and CD19 CAR T cells were spiked in with varying target frequencies among CD3⁺ viable T cells. To prepare samples for flow cytometric immune phenotyping, a lyse-no-wash protocol was used, which enables a precise cell count determination. For flow cytometric CAR persistence determination a bulk lysis approach including washing steps was chosen to establish the required assay sensitivity for a rare cell analysis ( < 1 % CAR⁺ among CD3⁺ viable T cells). To ensure a robust CAR persistence determination, the individual experimental steps of the assay (bulk lysis of whole blood, fluorescence staining procedure, and flow cytometric measurement) were evaluated in robustness experiments using a design of experiments (DOE) based quality-by-design (QbD) approach. For this, critical key parameters (factors, X) and parameter interactions were identified using a low level risk assessment, and different relevant responses (Y) were defined as readouts. Additionally, the repeatability of the persistence determination (N=3, n=3), as well as the immune phenotyping (N=3, n=6) was investigated to ensure the robustness of the method for immune monitoring in CAR clinical trials.

Results: For CAR persistence determination, 8 different parameters were tested to obtain a design for the most robust protocol possible. Here, we focused on 5 parameters affecting the first 2 critical steps of the method: bulk lysis of whole blood [i) ratio of whole blood to lysis solution, ii) lysis time, iii) hold time of cells after lysis] and fluorescence staining procedure [iv) incubation time of the CD19 CAR detection reagent, and v) number of washing steps after detection reagent staining]. All in all, the results obtained during robustness testing indicate a feasible, robust method with suitable parameter optima and edge of failure ranges (indicating the design space). Subsequent repeatability analysis showed a high level of repeatability for all cell subsets of interest for immune phenotyping (viable CD45⁺ cells, CD3⁺ cells, CD3⁺CD4⁺ cells, CD3⁺CD8⁺ cells, CD19⁺ B cells) as well as for CAR persistence determination with varying levels (high, medium, low) of spiked in CAR cells (CAR⁺ cells, CAR⁺CD4⁺ cells, CAR⁺ CD8⁺ cells).

Conclusion: In summary, a comprehensive, robust characterization of immune cell composition and CAR T persistence is possible using the reported method, rendering this approach useful for both clinical trial and clinical routine patient immune monitoring.

parameter optima and edge of failure analysis for bulk lysis of whole blood

parameter optima and edge of failure analysis for fluorescence staining procedure

Reapeatability analysis of immune phenotyping and CAR persistence determination