Aim: Chimeric antigen receptor (CAR) T cell therapy (CTT) is an autologous T cell-based treatment for B cell-based malignancies. During CTT, the patient’s T cells must first be activated in vitro with either anti-CD3 or anti-CD3/CD28 monoclonal antibodies (mAb). Dysfunctional T cells, such as terminally differentiated CD28- T cells, could hinder the efficacy of CTT and are found at higher frequencies in older patients. This study aims to define the optimal activation protocol for CAR T cell generation with older donors and to test whether CD28 expression on patient PBMCs can predict CAR T cell product quality.
Methods: Samples from young (20-30yo) and aged (60+) healthy donors were used to generate CAR T cells using both anti-CD3 and anti-CD3/CD28 mAb activation protocols. CD28 expression was assessed pre-activation and CAR T cell quality was tracked, in terms of yield, phenotype and function. Correlative analyses were performed to determine whether CD28 expression could predict CAR T cell product quality with either anti-CD3 or anti-CD3/CD28 mAb activation.
Results: Aged PBMC samples generated more differentiated CAR T cell products than young samples, and anti-CD3/CD28 mAb activation further exacerbated differentiation as compared to anti-CD3 mAb activation in both young and aged donors. Anti-CD3/CD28 mAb activation also generated a higher frequency of IFNg- and TNF-producing CAR T cells, increased killing capacity and augmented proliferation potential for both donor groups. Finally, we defined a novel metric of CD28 abundance with high predictive value for CAR T cell product quality.
Conclusions: In summary, anti-CD3/CD28 mAb activation protocols exacerbate T cell differentiation and may not be optimal for CAR T cell generation with aged samples. Accordingly, CD28 abundance on patient PBMCs can predict CAR T cell product quality and may be used to identify patients who would benefit from anti-CD3/CD28 mAb activation protocols.