Background Bioengineered T cell treatments for acute myeloid leukemia (AML) are challenged by near universal expression of leukemia antigens on normal hematopoietic stem/progenitor cells:^{1 2} ‘on target/off tumor‘ activity may cause myelosuppression while sustained antigen exposure can lead to T cell exhaustion.³ In addition, splicing variants may allow antigen escape. We hypothesize that by using a novel CD33-C2-specific single domain VHH antibody as the antigen targeting domain in dimerizing agent-regulated immunoreceptor complex T cells (DARIC T cells), we will enable pharmacologically-controllable targeting of CD33, allowing eradication of leukemia expressing either of the major splice variants of CD33: i.e., full-length CD33 or CD33ΔE2.

Methods We engineered DARIC-expressing lentiviral vectors containing encoding separated CD33-C2-specific antigen binding and 41BB-CD3zeta signaling chains that heterodimerize following addition of rapamycin via embedded FKBP12 and FRB* domains.⁴ Peripheral blood mononuclear cells were stimulated with IL-2, anti-CD3, and anti-CD28 antibodies 24h prior to transduction with DARIC33 lentiviral vector. Surface expression of antigen binding or signaling chains was assessed using biotinylated CD33, or antibodies to VHH-domains or FRB* respectively. Rapamycin-dependent in vitro activity was measured by IFNg release. To evaluate in vivo activity, NSG mice injected with 1 × 10⁵ MOLM-14/luc cells were treated 5-7 days later with 1 × 10⁷ DARIC33 T cells in the presence or absence of rapamycin and tumor progression followed by luciferase activity.

Results DARIC33+ T cells bound biotinylated-CD33, anti-VHH and anti-FRB* antibodies. Rapamycin addition increased expression of both signaling and antigen-recognition chains, suggesting augmented receptor stability in the presence of dimerizing drug. In the presence of rapamycin, coculture of DARIC33 T cells with cell lines expressing either full length or CD33ΔE2⁵ showed equivalent rapamycin-dependent activation, demonstrating DARIC33 responds to both splice variants. Titration experiments showed rapamycin-dependent activation with EC₅₀ = 25pM. Negligible IFNg release was observed in the absence of drug. DARIC33 T cells significantly extended survival of AML-bearing mice, but only when treated with rapamycin. The DARIC33 T cells were activated in vivo by sub-immunosuppressive rapamycin dosing, as weekly or 0.1 mg/kg QOD dosing led to similar levels of tumor suppression.

Conclusions DARIC33 T cells appear to be potent antileukemic agents: they are activated by AML cell lines in vitro as demonstrated by cytokine release and cytotoxicity, and significantly extend survival in an aggressive xenograft model. Temporal control provided by the DARIC architecture promises to enhance safety and potentially efficacy of CAR T therapy for AML, for example by enabling hematopoietic recovery or providing T cell rest.

References

1. Perna F, Berman SH, Soni RK, Mansilla-Soto J, Eyquem J, Hamieh M, et al. Integrating proteomics and transcriptomics for systematic combinatorial chimeric antigen receptor therapy of AML. Cancer Cell 2017 Oct 9;32(4):506–519.e5.

2. Haubner S, Perna F, Köhnke T, Schmidt C, Berman S, Augsberger C, et al. Coexpression profile of leukemic stem cell markers for combinatorial targeted therapy in AML. Leukemia. 2019 Jan;33(1):64.

3. Lamarche C, Novakovsky GE, Qi CN, Weber EW, Mackall CL, Levings MK. Repeated stimulation or tonic-signaling chimeric antigen receptors drive regulatory T cell exhaustion. bioRxiv. 2020 Jun 28;2020.06.27.175158.

4. Leung W-H, Gay J, Martin U, Garrett TE, Horton HM, Certo MT, et al. Sensitive and adaptable pharmacological control of CAR T cells through extracellular receptor dimerization. JCI Insight [Internet]. 2019 Jun 6 [cited 2019 Jun 11];4(11). Available from: https://insight.jci.org/articles/view/124430

5. Pérez-Oliva AB, Martínez-Esparza M, Vicente-Fernández JJ, Corral-San Miguel R, García-Peñarrubia P, Hernández-Caselles T. Epitope mapping, expression and post-translational modifications of two isoforms of CD33 (CD33M and CD33m) on lymphoid and myeloid human cells. Glycobiology 2011;21(6):757–770.