Background: In 1 in 3 metastatic breast cancers, the ErbB2 gene is amplified. Metastatic ErbB2-expressing breast cancer remains incurable despite aggressive treatment.

We have developed a chimeric antigen receptor (CAR) named I28z that re-targets T-cell specificity against ErbB2-expressing breast cancer. This CAR is targeted using the high affinity ICR12 antibody and enables targeted T-cells to destroy both ErbB2 amplified and non-amplified tumour cell types.

A major concern with clinical development of this approach is the potential likelihood of on-target off-tumour toxicity, due to the recognition of low levels of ErbB2 in healthy tissues. With expression both in cardiac tissue and pulmonary endothelium, a method to minimise this potential risk is of paramount importance.

Preliminary Work: This projects aim is to explore the potential of maximising the therapeutic index by co-expressing the highly potent I28z CAR with a PD-1 based inhibitory chimeric antigen receptor (ICAR) that recognises other members of the ErbB family which are co expressed in healthy tissues with low ErbB2.

Contrastingly, in ErbB2 amplified breast cancer, levels of ErbB2 far exceed those of other ErbB types. The hypothesis is that inhibitory (ICAR) and activating (CAR) signals will neutralize each other in healthy tissue, whereas stimulatory I28z signals will dominate in the tumour microenvironment. Consequently, it is predicted that effector function of the transduced T cells will be restricted to tumour sites. We have designed and engineered various ErbB engaging ICAR constructs with targeting moieties derived from natural ligands, an antibody ScFv and an ErbB antagonist. These have been co-expressed using SFG retroviral vectors with the ErbB2-specific CAR, I28z and its truncated control ITR in activated T cells.

Results: In-vitro efficacy testing of CAR/ICAR-engineered T-cells against a panel of tumour cell lines and antigen presenting cells with various expression ratios of ErbB2 and ErbB1/4. Both cytotoxicity and IL-2/IFNy cytokine production have been investigated. The results to date have shown CAR T-cell mediated cytotoxic activity is significantly reduced when co-expressed with a PD-1 ICAR in an antigen dependent manner. Furthermore obstacles specific to ICAR design have been highlighted and investigated.

Conclusion: In conclusion, we are developing a clinically relevant model for the use of an ICAR to alleviate on-target off-tumour toxicity when using ErbB2 targeted CAR immunotherapy, with highly promising results in in vitro testing to date.

Citation Format: Roseanna Maria Petrovic, Scott Wilkie, John Maher. Developing a PD-1 based inhibitory chimeric antigen receptor (ICAR) for co-expression, to overcome off-tumor toxicity when targeting ErbB2 using engineered T cells. [abstract]. In: Proceedings of the CRI-CIMT-EATI-AACR Inaugural International Cancer Immunotherapy Conference: Translating Science into Survival; September 16-19, 2015; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(1 Suppl):Abstract nr A082.