CYTO Virtual Interactive 2021 Oral Presentation - Antigen Density Qantification of Neuroblastoma Cell surface Targets to Inform the Development of CAR T cells potent towards clinically relevant levels
Background: Antigen density has emerged as a major factor determining the potency of CAR T cells. Traditional designs, such as KYMRIAH-based constructs have shown impressive clinical benefit when targeting B lineage antigens but manifest a high antigen density threshold for full activation, increasing the risk of relapse with antigen low variants, as observed following CARs targeting CD22 (Fry et al., 2018). In solid tumors, striking clinical results are lacking. Challenges include that truly tumor-specific antigens are rare and typically expressed at lower and more heterogeneous antigen densities. An accurate understanding of molecules/cell on tumor cells is crucial, as signal strength can be tuned by modifying CAR design towards an optimal therapeutic window and disease models used for CAR T cell development must be validated to express clinically relevant levels of target antigen, if they are to inform predicted potency against disease in humans.
Methods: Using flow cytometry to directly measure molecules/cell of immunotherapy-relevant targets including Glypican-2 (GPC2) on neuroblastoma cells in patients, we developed and applied a multicolor antigen density quantification assay on nine bone marrow samples infiltrated with metastatic tumor obtained and tested at 2 different institutions. Hematopoietic cells were excluded from the analysis based upon CD45, and bone marrow stromal cells based upon CD13 expression (Theodorakos et al., 2019). Gated CD45–CD13– cells co-expressed NCAM and GD2, known to be expressed on neuroblastoma (Warzynski et al., 2002) and were quantified in addition to GPC2, L1CAM and ALK. The fluorescence signal from the saturating antibody-fluorophore conjugates was converted into molecules/cell by using the reagent F/P ratio along with calibrated BD QuantiBRITETM beads and Custom Quantitation Beads containing known numbers of fluorophore molecules per bead.
Results: Metastatic neuroblastoma in the bone marrow demonstrated 4262±703 GPC2 molecules/cell, which is significantly lower than levels measured on the majority of neuroblastoma cell lines (p=0.0236). In comparison to several other immunotherapy targets we observed that GPC2 antigen density is lower than GD2, NCAM and L1CAM, but higher than anaplastic lymphoma kinase (ALK). The highest antigen density was observed for the disialoganglioside GD2, which demonstrated almost 30x more molecules than GPC2 and was expressed at significantly higher levels than the other targets tested. Levels of GPC2 on clinical specimens were approximating the level present on the neuroblastoma cell line SMS-SAN and found to be below the threshold required to control tumor growth for traditionally designed GPC2 CAR T cells. Sequential modifications to the CAR architecture, such as CD28-derived hinge-transmembrane and signaling domains and the overexpression of cJun resulted in a potent lead GPC2-CAR capable of impressive tumor regression mediating long-term complete responses in the absence of toxicity.
Conclusion: Our data emphasize that overexpressed tumor antigens can vary widely in the level and variation of expression and illustrate the profound impact of antigen density on the potency of CAR T cells. We highlight an essential role for the accurate quantitation of antigen density on clinical samples and show that selecting disease models for CAR T cell development informed by these measurements resulted in a promising lead candidate available for testing in early phase clinical trials.
University of Hong Kong
Shobana Stassen received her Bachelor of Science in Engineering (Hons) from Princeton University in 2010 and her Master of Science in Electrical and Electronic Engineering (Distinction) from the University of Hong Kong in 2019. She is currently a researcher at the Applied Life Photonics group at the University of Hong Kong. Her research interests are primarily in the development graph based statistical methods suitable for multi-omic and scalable single-cell analysis.
CMLE Credit: 1.0