Introduction: Adoptive transfer of T cells expressing chimeric antigen receptors (CARs) is an effective immunotherapy for hematological cancers but requires a rethinking for clinical efficacy against solid tumors, where CARs have largely failed. Lymphodepletive preconditioning regimens can enhance CAR activity in vivo by promoting T-cell expansion and depleting immunoinhibitory cells that counteract cellular immunity. These nonspecific regimens, however, can be exceedingly toxic and contribute to poor quality of life. Novel strategies are needed to bypass the intratumoral inhibition of CARs. CD4+FoxP3+ regulatory T cells (Tregs) play a critical role in treatment failure, and importantly, CARs have been shown to inadvertently potentiate Tregs by providing a local source of IL-2 for Treg consumption. We explored whether specific disruption of this axis would confer efficacy against solid tumors.
Methods: We developed second- (CD28z) and third- (CD28-4-1BBz) generation CARs targeting the tumor-specific mutation, EGFRvIII. To eliminate secretion of IL-2, two amino acid substitutions were introduced in the PYAP Lck binding-motif of the CD28 domain (xCD28) of CAR transgenes. We evaluated these modified second and third generation CARs against established B16 melanomas expressing EGFRvIII.
Results: Second generation CD28z CARs fail to expand in vivo. Addition of 4-1BB in third generation CARs improves expansion, but this modification alone was insufficient for CD28-4-1BBz CARs to treat tumors without prior host lymphodepletion. CARs deficient in Lck signaling, however, significantly retarded tumor growth in immune-intact mice without prior lymphodepletion, and this was dependent on inclusion of 4-1BB in CAR design. To determine if deficient Lck signaling altered CAR vulnerability to Tregs, we lymphodepleted mice and transferred CARs +/- Tregs. Co-transfer was sufficient to abrogate the efficacy of CD28-4-1BBz CARs, whereas the efficacy of xCD28-4-1BBz CARs remained unperturbed.
Conclusions: xCD28-4-1BBz CARs may be an effective immunotherapy for solid tumors infiltrated with Treg and may mitigate the need for toxic lymphodepletive preconditioning.
Patient Care: This work identifies and circumvents a major obstacle to effective T-cell immunotherapy for the treatment of solid tumors. In this study, we present novel modifications to CAR design that enable engineered T cells to bypass in vivo immune inhibition, resulting in antitumor efficacy against highly established tumors in fully immunocompetent hosts. This holds tremendous promise in the clinical translation of this therapy for patients bearing solid tumors, particularly those known to be heavily infiltrated with regulatory T cells (e.g. glioblastoma), as it may obviate the need for preparative lymphotoxic regimens using chemotherapy and/or radiation.
Learning Objectives: By the conclusion of this session, participants should be able to: 1) Understand and identify the challenges and obstacles to achieving antitumor efficacy with adoptively transferred T cells, 2) Describe intratumoral immune interactions that can promote or impede immunotherapy, and methods to manipulate and/or exploit these interactions to favorably shape CAR activity.