Unexpected In Vivo Toxicity Supports Use of mRNA in CAR T-Cell Therapy
July 2015
Researchers at Oslo University have discovered an important finding that may change the way scientists design certain types of T-cell therapies. T cells play a key role in cell-mediated immunity and in recent years scientists have learned how to isolate a patient’s own T cells, engineer the receptors to attack specific cells, such as cancer cells, and reintroduce them back into the patient.
The team, led by Jon Kyte, has been working on a T-cell therapy platform that utilizes transient expression of chimeric antigen receptors (CARs) to target leukemia and other blood-related cancers. CARs are recombinant receptors comprised of an extracellular antigen binding domain derived from a monoclonal antibody and intracellular signaling domains from the T-cell receptor complex. To connect the two domains, CARs frequently contain a spacer region which is typically based on a constant region of either IgG1 or IgG4. Newer generations also include one or two co-stimulatory molecules, such as CD28 or OX40.
Previously, most T-cell therapies utilized viral vectors, which permanently integrate into the host genome. This allows the CAR to replicate and expand with the T cell population. However, during clinical trials the persistent expression and expansion of the CAR resulted in severe toxicity, implicating a need for safer methods. To combat this, Kyte and colleagues utilized mRNA to transiently express the chimeric proteins. mRNA is a safer alternative because it does not integrate into the genome and is naturally metabolized by the cell.
In their previous study, the researchers developed an mRNA encoding the CD19-specific CAR, fmc63-IgG1Fc-CD28-OXO40-CD3ξ (19-IgFc-28OXξ). The mRNA was unmodified and capped with anti-reverse cap analogue (ARCA), a guanosine triphosphate analogue that initiates transcription. They showed that T cells expressing this CAR efficiently killed primary leukemia and lymphoma cells in vitro with minimal toxicity.
In their most recent paper, the researchers sought to validate their in vitro findings in an in vivo mouse model and unexpectedly found that their preferred CAR construct had only a marginal effect on tumor growth and caused substantial toxicity. Notably, 4 out of 5 mice treated with CAR-mRNA transduced T cells fully recovered. Mice that received stably transduced T cells died within one week due to persistent toxicity.
Subsequent experiments showed that the toxicity did not stem from the co-administered IL-2 adjuvant, so they evaluated the CAR at low and high expression levels. Neither condition appeared to reduce leukemia burden though toxicity did subside at lower expression levels.
Interestingly, when they tested a CAR variant with a short spacer they found great improvement in anti-leukemia activity. Strikingly, the variant without the IgG1-Fc spacer eradicated the leukemia without significant toxicity. This result has great implications for changing current therapeutic designs.
Next, the authors constructed and tested multiple CARs that contained variants of each domain. The authors were able to confirm that the CARs containing the shortest spacer exerted the greatest anti-leukemia activity and showed minimal toxicity. Similar to previous work, the variants containing the full spacer did not exert anti-leukemia activity and displayed significant cytotoxicity. Other domains did not appear to influence leukemia burden or toxicity.
Finally the authors sought to find the mechanism of in vivo toxicity for the full IgG1-Fc spacer. The full spacer contains a Fcγ receptor (FcγR) binding domain and may be interacting with mouse macrophages. The hypothesis was corroborated by immunohistochemistry experiments that showed strong binding of the full IgG1-Fc spacer with CD64, a high affinity FcγR. They did not see strong binding with shorter IgG1-Fc variants without the FcγR binding domain. Notably, the full spacer did not interact substantially to human FcγRs.
Taken together, these results suggest an importance to the IgG1-Fc spacer design for adoptive T-cell therapies. The fact that the full spacer did not interact significantly with human FcγRs demonstrates the complexity of the matter. Optimal design may not be fully understood until tried in humans. This highlights the safety advantages of mRNA and makes it a particularly attractive platform for future T-cell therapies.
Featured TriLink Products: Anti-Reverse Cap Analogue (ARCA). If you are interested in therapeutic grade ARCA, please contact us for a consultation. Please also see our wide selection of modified NTPs.Â
Have a question? Visit Ask An Expert.
Â
References