Home › Services › Cross-Species Reactivity Engineering

Your antibody does not bind the cyno ortholog. Do you need a surrogate, or can the lead itself be taught to cross-react?

In most cases the lead can be engineered. Abwiz Bio adds cynomolgus, mouse or other ortholog cross-reactivity to your existing antibody during humanization or STEM™ affinity maturation, by selecting the variant library against both the human and the animal antigen at once. The result is one molecule that works in both the tox species and the clinic — which means the safety package describes the drug itself rather than a proxy for it, and you avoid the cost, calendar time and interpretive ambiguity of raising, characterizing and defending a separate surrogate.

The problem with surrogate antibodies

When a lead is human-specific, the conventional path is a surrogate: a second antibody raised against the animal ortholog, used to generate the toxicology and pharmacology package. It is an accepted route, and it is expensive in ways that compound.

A surrogate is a different molecule with a different paratope, different affinity, different epitope, different effector engagement and different pharmacokinetics. Every conclusion drawn from it has to be bridged back to the clinical candidate with an argument rather than a measurement, and reviewers are entitled to probe every step of that argument. Meanwhile the surrogate itself has to be discovered, characterized and manufactured — a second program running beside the first.

Engineering cross-reactivity into the lead collapses that duplication. One molecule, one set of analytics, one manufacturing process, and animal data that describes the actual clinical candidate.

How we engineer ortholog binding

• Dual-antigen selection. The variant library is panned against your human antigen and the ortholog together, so only clones that engage both are enriched. Cross-reactivity is a survival condition, not an outcome hoped for and screened afterwards.
• Built into humanization or maturation. Cross-species selection is normally folded into an engineering pass you were running anyway — developability optimization, affinity maturation, or humanization — rather than added as a separate campaign, so it costs a filter rather than a project.
• Designed CDR diversity. STEM™ libraries vary CDR positions using human antibody amino-acid usage, guided by a local AlphaFold instance plus germline and structural bioinformatics, giving the selection enough sequence space to find a paratope that tolerates the divergent residues in the ortholog epitope.
• Affinity balanced, not sacrificed. Stringency is set so that human affinity is maintained while ortholog binding is acquired, and both are measured on the delivered clones.
• Biophysical filters throughout. Thermostability and polyreactivity filters run in every stage, so a cross-reactive clone does not arrive with new developability problems.

Species and starting points

Where a conserved epitope is required across many orthologs at once, single-domain antibodies are often the better starting format: their small paratope reaches clefts, catalytic sites and conserved surfaces that a conventional Fab cannot, which is exactly where sequence conservation between species tends to be highest.

You receive

• Cross-species binding data for the delivered clones against both human and ortholog antigen
• Comparison against the parent antibody, so the affinity trade is explicit
• Full VH/VL sequences and recombinant IgG
• ELISA data as standard; BLI kinetics optional after IgG purification
• No downstream royalties, stage-gated approval before each next step

Frequently asked questions

Can you make an existing antibody cross-react with the cyno ortholog?

Usually yes. We build a variant library around your lead and pan it against the human antigen and the cynomolgus ortholog simultaneously, so only clones binding both survive. This is normally folded into a humanization or STEM affinity maturation pass rather than run as a separate campaign.

Why engineer cross-reactivity instead of raising a surrogate antibody?

A surrogate is a different molecule with a different epitope, affinity, effector profile and pharmacokinetics, so every animal result has to be bridged back to the clinical candidate by argument rather than measurement, and the surrogate itself requires its own discovery, characterization and manufacturing. A cross-reactive lead gives you animal data on the actual candidate, with one molecule and one set of analytics.

Will adding ortholog binding weaken binding to the human target?

Selection stringency is set to maintain human affinity while ortholog binding is acquired, and both affinities are measured on every delivered clone against the parent, so any trade-off is visible in the data rather than assumed. In practice a panel is delivered so you can choose the balance your program needs.

Which species do you support?

Cynomolgus monkey and mouse are the most frequently requested. Other orthologs of the same antigen are supported where the antigen can be supplied, including rat and dog.

What do you need to start?

The VH/VL amino-acid sequence of your lead and both antigens, human and ortholog. No animals and no new discovery work are involved. Send your sequence to info@abwizbio.com for a scoped quote.