Can mRNA-encoded Antibodies Revolutionize Immunotherapy?

What are the benefits and challenges of encoding antibodies using mRNA?

The concept of using exogenous mRNA to express a protein in vivo is not new. Its feasibility was demonstrated in 1990¹. However, due to several challenges, it only recently matured to a therapeutic approach. At the same time, the COVID 19 pandemic propelled the field to new heights by exposing the possibility of manufacturing and delivering an mRNA to an incredible number of patients at very low cost. This has triggered various companies and academic groups to explore whether other therapeutics can be “packaged” into an mRNA to reduce costs.

One important avenue has been therapeutic antibodies, where the average cost of a traditional antibody per patient is US$100,000 vs $100 for an mRNA, an astounding difference. Other advantages of mRNA include proper assembly and natural glycosylation patterns of the antibody when produced directly in humans and rapid preclinical ‘design-build-test’ cycles. Despite these clear advantages, there are still some technical challenges, such as optimal mRNA design to minimize innate immune reactions and to maximize the expression of the encoded antibody, developing better understanding of the relationship between the dose of mRNA and the expression and pharmacokinetics of the resultant antibody, and looking more towards the future, selective delivery of LNPs beyond the liver. These collective challenges will be further discussed in the sections below.

How is mRNA designed to encoded antibodies?

The journey of an mRNA-encoded antibody to the clinic inevitably starts with the design. A mature mRNA is typically composed of a 5’ cap, 5’ untranslated region, coding sequence, 3’ untranslated region and a polyadenylation tail (polyA). The biological activity and stability of the mRNA are dictated by the 5’ cap (typically a 7- methyl guanosine or synthetic analog) and the 3’ poly(A) tail. These elements can be added during in vitro transcription or enzymatically after. The UTRs and the coding sequence can be engineered to enhance protein production, increase stability, and reduce immunogenicity of the mRNA (Figure 1).

Unmodified mRNA and the byproducts from in vivo transcription may stimulate RNA receptors including RIG-I like receptors and toll-like receptors, thereby eliciting immunogenic responses and unavoidable deleterious side effects. The efforts to decrease innate immunogenicity have focused on two aspects: base modifications and transcription/purification methods. Substitution of uridine and cytosine for the naturally occurring pseudouridine and 5-methylcytosine are very effective for this purpose and additionally, increase protein production. However, there are concerns that the fidelity of mRNA translation may be affected by such modifications. High purity mRNA product has been shown to minimize the potential immune response. This can be achieved either by employing engineered T7 polymerases to reduce generation of undesirable byproducts or by very stringent purification via high-performance chromatography.

Figure 1. mRNA-encoded antibody design. Numerous antibody formats can be encoded using mRNA that include full-length antibodies, nanobodies or various bispecific architectures. Adapted from Deal et al, 2021, Vaccines 9(2):108

How are mRNA-encoded antibodies delivered and formulated?

One of the key challenges for all mRNA medicines - including mRNA-encoded antibodies - is the efficient cytosolic delivery of mRNA to desired target cells and efficient protein translation. Typically, large mRNA molecules do not easily move back and forth across lipid bilayers due to their negative charge, and they are incredibly susceptible to degradation from RNases. These hurdles have largely been overcome by encapsulating mRNA into a complex mix of lipids called lipid nanoparticles (LNPs). These serve as the most common delivery method for mRNA drugs and vaccines to date, including several of the COVID-19 vaccines.

Importantly, mRNAs formulated into LNPs are shielded from degradation, improve mRNA plasma half-life, and provide superior cellular uptake which results in enhanced translation. LNP formulations typically consist of a cationic ionizable lipid capable of modulating their charge depending on the environmental pH, a PEGylated lipid, cholesterol, and a helper lipid; each component impacts the overall stability, tolerability and delivery of the nanoparticle.

Currently, many mRNA-encoded antibodies are delivered systemically and are primarily translated in the liver, which acts as a ‘bioreactor’ for antibody production (Figure 2). The propensity of LNPs for hepatic distribution is due to both a tendency to drift into and get trapped by large endothelial fenestrations of the liver, as well as a ‘coat’ of serum proteins acquired in vivo that promotes binding to liver cell receptors and uptake by liver-resident macrophages (Kupffer cells).

However, where indications require protective antibodies at distinct locations, particularly for passive immunotherapy in viral lung infections—local mRNA delivery and expression may be favorable. Considerable progress has been made in developing both inhalable mRNA-LNP drug formulations to directly target mucosal surfaces in addition to rational design of synthetic lipids that divert LNPs to the lungs. As the application of mRNA-encoded antibodies diversify, so too will the need to rationally engineer both LNPs and mRNA to efficiently target relevant organs.  

Figure 2. mRNA-encoded immunotherapy. mRNA-LNPs are typically administered intravenously and target the liver, where hepatocytic uptake leads to translation of encoded antibodies, secretion into the circulation nd ultimately binding of cognate antigens. Adapted from Schlake et al, 2018. Cell Mol Life Sci 76:301-328.

What are the challenges testing the efficacy of mRNA-encoded antibodies in translational models?

Understanding the safety profile and efficacy of mRNA-encoded antibodies will largely follow the same course as describe in a recent Eureka article for recombinant antibodies. However, several unique differences arise. Ultimately approaches will be dictated by disease indication. For instance, many of the initial binding and functional studies will require purifying mRNA-encoded antibodies from relevant producer cell lines or animals to ascertain mechanisms of action. The transient nature of mRNA expression, and the need to optimize ratios of more than one mRNA if encoding heavy and light chains separately, could potentially impact yield and bioavailability of antibodies for downstream analysis.

Investigating strategies to express mRNA-encoded antibodies in vivo—such as the choice to encode full-length antibodies or smaller, more complex formats—will be key to defining success in terms of antibody expression kinetics, pharmacodynamics (PD), and efficacy. Careful consideration of preclinical models will be dependent on the mechanism of action of the encoded antibody, clinical indication and expected duration of treatment. For example, peak serum levels of mRNA-encoded antibody can significantly differ depending on the species when using equivalent dosing and mRNA ratios2 . Moreover, it is desirable to benchmark against recombinant antibody equivalents to demonstrate comparable pharmacokinetics/pharmacodynamics (PK/PD) and efficacy of the mRNA approach.

From a biological standpoint, a key distinction compared to recombinant antibodies is the inherent need of a delivery system, namely LNPs. As detailed below, LNPs themselves pose an additional entity or ’excipient’ to understand as well as the antibody itself and will require interrogating repeat dosing effects, biodistribution and immunogenicity of the final mRNA drug product.

Preclinical Safety Considerations for developing an mRNA-Encoded Antibody

An Investigational New Drug (IND)-enabling toxicology package for an mRNA encoding a standard format or bispecific antibody involves multiple considerations for the final product (antibody), including the encoding mRNA, and the means of delivery.  For example, one must consider affinity to the primary target, selectivity and cross species binding as well as intended on- and potential off-target pharmacology of the antibody candidate.

Also, important to consider are potential acute toxicity of the intact nanoparticle, the potential immune stimulating effects of the mRNA, the intended biodistribution of the mRNA and the potential formation anti-drug antibodies in response to the resultant antibody.

Before initiating nonclinical safety studies, the binding equilibrium dissociation constant (KD) to the human target protein and selectivity of the antibody candidates can be assessed by Surface Plasmon Resonance, Retrogenix^® and in vitro functional assays to select the optimal antibody for development.

Thereafter, cross-species binding of the candidate antibody can be assessed by SPR or Retrogenix^® to determine relevant affinity of the antibody across the nonclinical safety species.  If candidate antibodies manufactured by conventional means are characterized using in vitro functional assays, candidate antibodies produced by in vitro mRNA translation should be characterized in the same cell lines using the same functional assays to establish equivalent performance characteristics.

Acute infusion reactions associated with LNPs are mediated by complement activation, in part due to the development of anti-LNP (polyethylene glycol; PEG) antibodies following repeat dosing regimens, and the observation of cytokine release is not uncommon.  In addition, some nanoparticles can increase the risk of coagulation and thrombus formation.  Candidate LNPs/mRNAs can also be compared in single dose toxicity studies.  Thereafter the best candidates can be selected for further toxicity studies using the most appropriate species for human antibodies.

To prevent acute immune stimulating effects of LNP delivered mRNAs, modifications can be made to the LNP as well the mRNA backbone to reduce immunogenicity.  Several predictive in silico tools are commonly used to de-risk attributes associated with mRNA backbone and optimization of production and purification processes, as well as co-administration of innate immune inhibitors should also be considered.  However, these latter aspects should be considered in the design of the LNP/mRNA before in vivo studies are conducted.

Pharmacokinetic/toxicokinetic characterization of the LNP constituents as well as the mRNA platform should be assessed in the nonclinical toxicity studies by Liquid chromatography–mass spectrometry (LCMS) and polymerase chain reaction (PCR), respectively, whereas the PK of the resultant antibody and anti-drug antibodies can be assessed by electrochemiluminescence or ELISA assays.

Since clearance of the nanoparticle is rapid compared to the antibody product, pharmacokinetic or toxicokinetic studies should be performed to determine the exposure profile of each, and to correlate the dose of the intact LNP/mRNA nanoparticle with resultant antibody expression to assess the PK/antibody exposure relationship.  This information can be used inform Physiologically Based Pharmacokinetic Modeling to select the most effective starting dose in humans.

Based on FDA guidelines and ICHS12 guidance for cell and gene therapies, biodistribution of the mRNA should be determined by PCR at the time of peak expression and at the end of the study in the following tissues: blood, gonads, brain, liver, kidneys, lung, heart, and spleen, at minimum, and additional tissues as appropriate for the program.  The formation of anti-drug antibodies should also be assessed and correlated to the clearance and exposure to the antibody product over time as this may also inform the toxicity assessment and whether longer term toxicity studies will be required to support later stages of the clinical program.

Since the acute toxicity of the LNP/mRNA nanoparticle are inextricably linked, the final clinical candidate should be tested in dose-ranging and repeat dose toxicity studies.  Based on the above noted considerations, cytokine release and possibly complement activation should be at minimum assessed after the first dose. The design of in vivo toxicity studies should be customized based on the expected pharmacologic mode of action of the resultant antibody with inclusion of typical in-life toxicity endpoints, safety pharmacology and histopathology according to the ICHS6 Guidance. 

Conclusion

Though mRNA encoded antibodies are in nascent stages of research and development the advantages of scale up, lower manufacturing costs compared to conventional antibody production, and identical glycosylation patterns to endogenous human antibodies may increase patient accessibility and improve safety by decreasing anti-drug antibody responses.  These attributes combined with extended antibody production from the liver may also increase the duration of exposure, lower the Cmax to trough ratios and thereby minimize Cmax mediated toxicities such as cytokine release associated with some antibody platforms such as bispecific antibodies.

The format possibilities and potential clinical indications will inevitably increase with advances in mRNA design and delivery technologies.  Overall, this approach could represent the next disruptive technology for antibody and protein therapeutics in general.

References:
1.    Wolff JA et al, Science, 1990
2.    Kasiewicz, L. N. et al. Lipid nanoparticles incorporating a GalNAc ligand enable in vivo liver ANGPTL3 editing in wild-type and somatic LDLR knockout non-human primates. bioRxiv 2021.11.08.467731 (2021) doi:10.1101/2021.11.08.467731.
3.    Kasiewicz, L. N. et al. GalNAc-Lipid nanoparticles enable non-LDLR dependent hepatic delivery of a CRISPR base editing therapy. Nat. Commun. 14, 2776 (2023).
4.    Kose et al, 2019, Sci Immunol 4(35):eaaw6647

Stan Spence, PhD, DABT, is a Senior Principal Scientific Advisor, Roxana Redis, PhD, is an Associate Director of the Biotherapeutics Division and Dan Rocca, PhD is a Research Leader in the Biotherapeutics Division at Charles River Laboratories.

Oncology & Immuno-Oncology Studies

To reach the clinic in record time, it’s vital to test your oncology therapies in systems that reflect the disease seen in humans. We are here to help with a range of translational oncology studies, including in vitro assays and in vivo models, that mirror human cancers.

How Can We Support Your Program?