Detection of CAR-T19 cells in peripheral blood and cerebrospinal fluid: An assay applicable to routine diagnostic laboratories

Abstract

Background: Chimeric antigen receptor-modified T-cells targeting CD19 (CAR-T19) are licensed for treating relapsed/refractory diffuse large B-cell lymphoma and B-acute lymphoblastic leukemia. Predicting treatment responses and toxicity (e.g., cytokine release syndrome and neurotoxicity) remains a big challenge. CAR-T19 monitoring could increase our understanding of treatment responses and be of relevance to patient management. A robust method for accurate CAR-T19 detection is therefore extremely desirable. Methods: An assay that uses fluorochrome-conjugated human recombinant soluble CD19 was tested against two commercially available CAR-T19 therapies and a CAR-T19 cell line developed in-house. Precision, concordance, and analyte stability were tested using peripheral blood obtained from CAR-T19-treated patients and controls. Results: The assay showed good accuracy, and had a limit of blank for whole blood samples of 0.13%. Reproducibility and inter-operator concordance were satisfactory (CVs <15%). The assay distinguished CAR-T19 from reactive T-cells in cerebrospinal fluid (CSF) from patients with suspected immune effector cell-associated neurotoxicity syndrome (ICANS), and was adapted to study memory T-cell compartments in treated patients. Conclusion: The assay enabled routine monitoring of CAR-T19 in blood and CSF samples. Despite profound cytopenia in many lymphoma patients, results were obtained regularly from only 4 ml of blood. The assay can be adapted easily to characterize the memory and exhaustion status of CAR-T19 and native T-cells. Importantly, it does not rely on CAR construct specificity; thus, it can be used to detect any CD19-targeted CAR cell. Finally, our validation process can serve as a blueprint for other fluorochrome proteins used to detect CAR cells.