Unlike conventional IgG antibodies from human and mouse, which are tetramers of paired heavy and light chains, VHHs are composed solely of the single variable domain of a heavy chain.

The compact nature and small size of VHHs makes them the ideal building blocks for antibody engineering.

In the same way that most proteins are made up of many domains, multiple VHH monomers can be chained together to create multi-specific antibodies, offering exciting therapeutic possibilities without the risk of chain mispairing seen in conventional bi-specific IgGs.

Along with their simplicity, the small coding sequences of VHHs (<375 bp compared with 800-900 bp for a typical IgG-derived scFv) fit more easily in the vectors used for cell and gene therapy engineering such as CAR-Ts.

Find out more about the advantages of using VHHs for CAR engineering.

Immunogenicity is a universal concern in the development of antibody-based therapeutics.

Foreign sequences within therapeutic antibodies may trigger a new immune response generating anti-drug antibodies (ADAs), limiting drug effectiveness in the short or long term and potentially causing side effects.

As a result, most mouse monoclonal antibodies have been through extensive humanisation to make them suitable for use in patients.

Given that they are originally derived from camelids, there are concerns that VHHs may also have similar problems.

It is true that the C-terminus of a VHH antibody can appear similar to a broken IgG ‘hinge’ fragment and can therefore be target for immune recognition by pre-existing ADAs in some patients. This has been identified as a risk factor for immunogenicity, but can be screened for straightforwardly in vitro (Rossotti et al., 2021).

Nonetheless, the small size of VHHs and similarity to human IGHV3 gene products – comparable to humanised mouse IgG heavy chain domains – favours low immunogenicity.

This is also borne out in the clinic, with most trials of VHHs showing minimal immunogenicity (with a few notable exceptions). So while it’s certainly not something to ignore, immunogenicity is unlikely to be a major problem with VHH-based therapeutics compared with conventional monoclonals.

Interestingly, excessive humanisation of VHH antibodies may even be counterproductive, and some fully humanised VHHs have been shown to trigger the production of treatment-emergent ADAs. Furthermore, the introduction of too much human sequence to VHHs can remove special VHH “hallmark” residues, increasing the risk of antibody aggregation and reducing binding affinity.

As VHHs become more prevalent in clinical use, more data will be available to help us reduce VHH immunogenicity even further. For example, our sophisticated fully synthetic VHH libraries allow us to screen out some of these risks from the very beginning of the discovery process.

For more detailed information about VHHs and immunogenicity, take a look at Ackaert et al.,2021, and Rossotti et al., 2021.

One of the most useful properties of small format single domain VHH antibodies is their ability to be chained together into one long amino acid string that then folds into its component recognition domains.

Importantly, it is possible to link together only the VHH domains of camelid antibodies to create functional bi- and multi-specific antibodies without the need for their associated Fc domains, which can influence a number of biological processes.

Manufacturing these antibodies in one long, self-folding string also means that the resulting product is homogeneous, unlike paired Fc-based antibodies where there is a risk that the monomers are not produced in equal proportions or pair improperly, leading to a heterogenous product.

This offers the opportunity to create bi- or multi-specific antibodies with a lower cost of goods compared with bi-specific IgG antibodies, since multi-specific VHHs can be purified more straightforwardly, and are even simple enough to express from microbial systems such as yeast.

We have years of expertise in engineering VHHs to create potent novel biotherapeutics. We know that each multi-specific antibody is different, and it’s important to test VHH orientations, different linker lengths and flexibilities. See Vasilenko et al. 2020 for more information about VHH engineering and our application note on bi-specific VHH biotherapeutics for the treatment of solid tumours.

We have three different VHH libraries at Isogenica.

Our standard LlamdA^® library contains more than 10¹³ diverse VHH sequences, equivalent to the circulating antibody repertoire of a million llamas. We also have our Humanised huLlamdA™ and V_HHantage™ libraries, which have been engineered to have more favourable properties for drug development.

Although the name LlamdA^® might sound like these antibodies come directly from llamas, all our libraries are purely synthetic. We use standard VHH frameworks, only varying the CDR regions to generate a high degree of diversity where it’s needed.

Not only does this mean we can begin screening with any antigen immediately, instead of waiting for an animal immune response to develop, it also simplifies the subsequent antibody engineering. Our fully synthetic VHHs are tailor-made for cloning into different expression systems or building bi- and multi-specifics using established tools.

Synthetic libraries are especially useful when targeting difficult-to-immunise targets. This includes antigens that are expensive or time-consuming to produce in bulk or are toxic to the species being immunised. It also gets around the problems caused by antigens with very high cross-species conservation, such as phospholamban, an important regulator of cardiac function (De Genst et al., 2022).

Only a few years ago, high affinity was viewed by many in the research community as the single most important property of a therapeutic antibody, with desired binding affinities in the picomolar range. Today it’s a rare request.

Now it is widely recognised that the epitope, avidity, and biological mode of action are significantly more important for a drug to be successful than its affinity alone (Rudnick and Adams, 2009; Oostindie et al. 2022). We are also able to create bi-valent or bi-paratopic VHH multimers, which target two epitopes within the same antigen, to exploit the power of avidity interactions.

While an animal immunisation approach is probably more likely to generate an extremely high affinity antibody than any synthetic library – we usually see binding affinity in the low nanomolar range – the high diversity, speed of screening, and ease of engineering more than make up for it.

The small size and high affinity of single-domain VHH antibodies offers the opportunity to target a wide range of challenging or intractable targets with high specificity and low immunogenicity.

However, current VHH-based therapeutics have a very short half-life in the body due to rapid clearance by the kidney, and require multiple, frequent infusions. This is ideal for some applications such as radiolabelled diagnostics, but not for the treatment of long-term diseases like cancer.

To solve this problem we have created our ISOXTEND® cross-species serum albumin binders, validated in vivo, suitable for half-life extension of VHH therapeutics and other biologics.

ISOXTEND® protects biotherapeutics from kidney clearance by ‘piggybacking’ onto the serum albumin endocytic recycling pathway, significantly extending therapeutic half-life.

ISOXTEND® has a demonstrated half-life of up to 26 hrs in mouse in vivo studies, equating to approximately 16-19 days in humans. It can be easily engineered to create mono-, bi or multi-specific antibodies, and is cross-reactive across multiple species including humans, mice and monkeys.