Abstract

Introduction: T-cell based therapies have emerged as one of the major breakthroughs in anticancer treatment: Immune checkpoint inhibitors, chimeric antigen receptor gene-modified T-cells (CAR-T-cells), and T-cell engaging bispecific antibodies (BsAb) are leading the advances. In the era of personalized medicine, T-cells offer alternative strategies to overcome resistance to chemotherapy or small molecules. Yet, hurdles for such therapy can be crippling, such as inability of T cells to infiltrate "cold tumors", cytokine release syndrome following T cell-based therapies, neurologic toxicity, and on-target off-tumor effects. To address these hurdles, polyclonal T-cells armed with GD2xCD3 or HER2xCD3 BsAb for cytotherapy hold promise. Ganglioside GD2 and HER2 are tumor associated surface antigens expressed in a broad spectrum of aggressive malignancies, while being restricted in normal tissues. Phase I trials of T-cells armed with a chemical conjugate of hu3F8 x mouse OKT3 (NCT02173093) or trastuzumab x mouse OKT3 (NCT00027807) demonstrated the safety of 160 x 106/kg/injection x 8 doses (or 1.28 x 109/cycle) with suggestion of clinical benefit. Here, we report the safety and efficacy of adoptive T-cell therapy armed with the recombinant forms of these BsAb for the treatment of GD2(+) and/or HER2(+) tumors in preclinical models.

Methods: Recombinant anti-GD2 BsAb and anti-HER2 BsAb were made using the IgG(L)-scFv platform (Can Immunol Res, 3:266, 2015, Oncoimmunology, PMID:28405494). T-cells from normal volunteer donors were isolated, activated and expanded by CD3/CD28 beads in the presence of 100 IU/mL of interleukin 2 (IL-2). Between day 7 and day 14, activated T cells (ATCs) were harvested and armed for 20 minutes at room temperature with -GD2-BsAb or HER2-BsAb. After washing, armed ATCs were tested for cell surface density of BsAb and antibody dependent T cell mediated cytotoxicity (ADTC) in vitro. In vivo anti-tumor potencies of armed T cells were tested against GD2(+) or HER2(+) cell lines or patient derived xenografts (PDXs) in BALB-Rag2-/-IL-2R-γϲ-KO (DKO) mice.

Results: GD2-BsAb of the IgG(L)-scFv form showed superior potency over other bispecific platforms in vitro and in vivo. GD2-BsAb or HER2-BsAb armed ATCs showed potent antigen-specific cytotoxicity against GD2 or HER2 positive tumors such as neuroblastoma, melanoma and osteosarcoma in vitro over a range of antibody dose (5 to 500 ng/106 cells). Optimal arming per T cell required 25,000 to 45,000 idiotype(+) molecules. There was no evidence of activation induced cell death when confronted by antigen or tumor target. In vivo GD2-BsAb or HER2-BsAb armed ATCs could ablate neuroblastoma, malignant melanoma, and osteosarcoma tumors over a range of cell doses (10x106, 20x106 and 40x106 per injection, one to three times a week for 2 to 4 weeks) with a range of BsAb doses (5 ng to 500 ng/million of T-cells) without significant toxicities in DKO mice. By immunohistochemistry, the frequency of tumor infiltrating CD3(+) T-cells strongly correlated with tumor response.

Conclusions:

Using the IgG(L)-scFv format, GD2-BsAb or HER2-BsAb armed ATCs could provide a potent and economical cytotherapy platform against GD2(+) or HER2(+) tumors without the complexity of gene modification (as in chimeric antigen receptor modified T cells). At such low arming doses, where BsAb is T cell bound, where ADTC is not induced, and T cell expansion is not required for anti-tumor effect, clinical toxicity is expected to be low.

Disclosures

Cheung:Ymabs: Patents & Royalties.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract