Nona Biosciences and Lonza have entered a strategic collaboration to develop a fully human, single-domain antibody-based blood-brain barrier crossing platform for central nervous system therapeutics. The programme will combine Nona Biosciences’ HCAb Harbour Mice discovery technology with Lonza’s protein development, GS Gene Expression System and GlycoConnect bioconjugation capabilities, while its commercial structure includes undisclosed upfront and option payments and shared revenue from future licences.
The significance lies less in any individual molecule and more in an attempt to create a reusable delivery component that other drug developers could attach to antibodies, proteins or conjugated therapeutic payloads. Most large biologic medicines enter the brain only at very low levels because the blood-brain barrier is designed to restrict the movement of circulating molecules into neural tissue. A platform that improves transport without creating unacceptable peripheral toxicity could therefore influence several neurological drug pipelines rather than only one programme.
However, the collaboration remains at an early technology-development stage. Nona Biosciences and Lonza have not disclosed the endothelial receptor that the single-domain antibody will target, the amount of improvement in brain exposure being sought, a lead therapeutic payload, a central nervous system indication or a timetable for preclinical and clinical development. The claim that the resulting technology could become best in class will consequently depend on comparative pharmacology, safety and manufacturability data that have not yet been presented.
Why this collaboration targets the delivery problem that still limits CNS biologics
The blood-brain barrier is a central obstacle in neurological drug development because successful target identification does not automatically produce adequate exposure at the site of disease. Conventional antibodies can remain largely confined to the bloodstream, forcing developers to administer substantial systemic doses to obtain relatively modest concentrations in the brain. That imbalance can raise manufacturing costs, increase exposure in non-neural tissues and narrow the therapeutic window of otherwise promising biological medicines.
Blood-brain barrier shuttles are intended to change that equation by engaging receptors on brain endothelial cells and using cellular transport processes to move therapeutic molecules across the barrier. The most established experimental strategies involve receptor-mediated transcytosis, in which a shuttle binds a receptor, enters the endothelial cell and is transported towards the brain-facing side. Transferrin receptor 1 has attracted considerable attention, although other receptors, including CD98 and insulin-like growth factor receptors, are also being evaluated across the industry.
The challenge is that receptor binding alone does not guarantee productive delivery. A molecule may bind too strongly and become trapped inside endothelial cells, undergo lysosomal degradation or alter the normal function of the transport receptor. Conversely, insufficient affinity may produce weak uptake and inconsistent brain exposure. Researchers must optimise affinity, valency, binding site, molecular architecture and release behaviour as an integrated system rather than treating the shuttle as a simple targeting tag.
The undisclosed receptor strategy is therefore one of the most important unanswered questions surrounding the Nona Biosciences and Lonza collaboration. Until the partners identify the transport mechanism and show quantitative biodistribution data, it will be difficult to determine whether the programme offers a genuine mechanistic advantage or represents another entrant into an increasingly competitive receptor-mediated delivery field.
What fully human single-domain antibodies could change in blood-brain barrier shuttle design
Nona Biosciences’ HCAb Harbour Mice platform generates fully human heavy chain-only antibodies and fully human variable heavy single-domain binders. These compact binding domains can be incorporated into bispecific or multispecific molecules without requiring the complete heavy-chain and light-chain structure used by conventional antibodies. Their relatively small structural footprint may make them useful as modular shuttle components attached to a separate therapeutic antibody or biological payload.
This architecture could offer practical advantages. A compact domain may reduce the increase in molecular size created when a transport arm is added to an existing therapeutic antibody. Single-domain binders can also provide greater flexibility in selecting fusion positions, adjusting valency and engineering multispecific constructs that need to engage both a blood-brain barrier receptor and a disease-related target. Fully human binders may additionally reduce the amount of humanisation work required compared with camelid-derived single-domain antibodies.
Those theoretical benefits do not remove the normal development risks. Fully human origin does not guarantee the absence of immunogenicity because new junctions, fusion sequences, aggregates and post-translational modifications can still trigger immune responses. Compact domains can also create stability or aggregation problems when transferred from an isolated discovery construct into a larger multispecific molecule. Their pharmacokinetic behaviour may change depending on whether the domain is used alone, fused to an antibody or connected to another therapeutic modality.
The decisive question will be whether the Nona Biosciences platform can repeatedly generate shuttles that combine strong receptor selectivity with suitable affinity, efficient transcytosis and acceptable developability. Discovery breadth is valuable, but the commercial value of a blood-brain barrier platform depends on reproducible performance across multiple payloads and disease models rather than exceptional results from a single construct.
Why Lonza’s manufacturing and conjugation role matters before a candidate reaches the clinic
Lonza’s involvement expands the collaboration beyond antibody discovery. The GS Gene Expression System is designed to support recombinant protein production in Chinese hamster ovary cells, while Lonza’s development infrastructure can address cell-line construction, expression, purification, analytical characterisation and eventual scale-up. These capabilities matter because blood-brain barrier shuttles are likely to involve more complex molecular formats than standard monoclonal antibodies.
A molecule can demonstrate promising activity in an academic-scale experiment and still fail during process development. Multispecific proteins may produce unwanted chain combinations, low expression yields, aggregates or difficult-to-remove variants. Changes introduced to improve transport can affect solubility, thermal stability, Fc function or formulation behaviour. Early integration of discovery and manufacturing expertise could allow the partners to remove candidates with unacceptable production characteristics before they consume substantial preclinical resources.
GlycoConnect adds another potential layer by enabling site-specific attachment through an antibody’s native glycan. In principle, precise conjugation could support the delivery of therapeutic payloads that extend beyond conventional antibodies, including oligonucleotides, proteins or other biologically active components. A consistent attachment site may also reduce product heterogeneity and make analytical control more manageable than less selective conjugation methods.
The exact role of GlycoConnect in the programme has not been defined, and it should not be assumed that every future blood-brain barrier construct will be a conjugate. The partners must still show that conjugation does not interfere with receptor binding, transport, payload activity or pharmacokinetics. Site-specific chemistry can improve molecular consistency, but it cannot compensate for a shuttle that fails to release its payload efficiently into brain tissue.
How the platform compares with transferrin receptor and other receptor-mediated shuttles
The collaboration enters a field in which several developers have already established preclinical or clinical evidence for receptor-mediated brain delivery. Transferrin receptor-based platforms are among the most advanced because the receptor is abundant on brain endothelial cells and can support transport of engineered antibodies. Roche’s trontinemab, which combines an amyloid-targeting antibody with a transferrin receptor-binding brain shuttle, has generated clinical evidence of substantial amyloid plaque reduction and has moved into late-stage development.
That progress strengthens the biological case for blood-brain barrier shuttles, but it also raises the competitive standard. A new platform may need to demonstrate not only increased brain exposure but also a meaningful improvement in dose efficiency, speed of target engagement, safety, dosing convenience or compatibility with different therapeutic payloads. Incremental transport gains may not be sufficient if established platforms already have clinical data and manufacturing strategies.
Alternative receptors could offer differentiation, particularly if they provide broader distribution across brain regions, lower peripheral expression or better performance in specific patient populations. The receptor selected by Nona Biosciences and Lonza will affect tissue distribution, safety monitoring and the range of therapeutic formats that can be supported. A shuttle intended for chronic neurodegenerative diseases may require a different balance of potency and tolerability than one designed for aggressive brain tumours or rare paediatric disorders.
Direct comparisons will also require carefully standardised experiments. Brain concentration measurements can be misleading when residual drug remains inside blood vessels rather than entering neural tissue. Developers must separate vascular and parenchymal exposure, measure intact and functional payload, assess regional distribution and demonstrate target engagement. Without those controls, apparent blood-brain barrier crossing may reflect endothelial accumulation rather than therapeutically useful delivery.
Which preclinical evidence will determine whether the technology is truly differentiated
The first meaningful proof point will be evidence that selected single-domain binders cross the blood-brain barrier through a defined and reproducible mechanism. The partners will need to disclose receptor identity, binding affinity, epitope, valency and transport behaviour. Studies should establish whether the construct reaches brain parenchyma rather than remaining associated with capillaries, and whether delivery is sustained across clinically relevant dose levels.
Cross-species activity will be another major issue. Receptor expression, binding epitopes and transport efficiency can differ between rodents, non-human primates and humans. A shuttle that performs strongly in a genetically modified mouse may produce weaker exposure in primates or require a species-specific surrogate molecule. These differences can complicate toxicology, dose prediction and interpretation of efficacy models.
Payload testing will determine whether the technology is genuinely modular. A platform may work well with one antibody yet perform differently when attached to an enzyme, oligonucleotide, cytokine or multispecific protein. Changes in molecular weight, charge, valency and pharmacokinetics can alter receptor engagement and transport. Demonstrating comparable delivery improvements across several payload classes would provide stronger support for a licensing platform than a single proof-of-concept experiment.
Safety data must examine receptor biology outside the brain as well as inside it. Blood-brain barrier transport receptors may also be expressed in blood cells, liver, muscle or other tissues. Repeated dosing could affect receptor function, alter iron or nutrient handling, produce peripheral sequestration or cause tissue-specific toxicity. These risks may be especially important in chronic neurological conditions where patients could receive treatment for years.
What regulators will expect from a modular blood-brain barrier delivery platform
A blood-brain barrier shuttle is not evaluated independently from the therapeutic molecule to which it is attached. Each resulting product will require evidence supporting its complete molecular architecture, pharmacology, biodistribution, toxicology, manufacturing consistency and clinical dose rationale. A shuttle that has been used successfully in one programme may reduce uncertainty, but it will not eliminate the need to assess a new payload and construct.
Regulators are likely to focus closely on whether increased brain penetration changes the distribution of the therapeutic target or creates effects in healthy neural tissue. Developers will need assays capable of distinguishing the shuttle, intact therapeutic construct, released payload and degradation products. Conventional plasma pharmacokinetics may provide an incomplete picture when the main objective is controlled exposure within specific brain compartments.
Manufacturing changes could also become more consequential than they are for simpler antibodies. Alterations in cell line, glycosylation, conjugation efficiency or aggregate levels might influence receptor binding and transport. The collaboration’s integration of discovery, expression and conjugation capabilities could help establish stronger control strategies, although that advantage must be demonstrated through reproducible batches and validated analytical methods.
Clinical development will require biomarkers that show the platform is performing as intended. Cerebrospinal fluid measurements may support exposure assessments but do not necessarily reflect concentrations throughout the brain. Imaging, soluble biomarkers, pharmacodynamic readouts and disease-specific endpoints may therefore need to be combined. Programmes that cannot demonstrate early target engagement could struggle to separate delivery failure from failure of the therapeutic mechanism.
Why the licensing model could create value before a proprietary CNS pipeline emerges
The financial structure suggests that Nona Biosciences and Lonza are seeking to build a licensable technology rather than limiting the collaboration to one internally owned drug. Nona Biosciences is eligible for upfront and option payments, while both parties are positioned to share revenue from future licensing agreements. This creates the possibility of commercial returns from external programmes before either partner independently launches a central nervous system medicine.
For Nona Biosciences, the agreement extends the role of the HCAb Harbour Mice platform from antibody discovery into therapeutic delivery. A validated blood-brain barrier shuttle could increase demand for the biotechnology firm’s discovery services and create licensing opportunities across neurodegeneration, rare neurological diseases, neuroinflammation and neuro-oncology. Lonza could gain a differentiated technology that attracts clients into its broader development and manufacturing network.
The model also creates dependency on external adoption. Pharmaceutical companies may already be developing proprietary brain-shuttle technologies or may prefer platforms with human clinical validation. Licensees will examine freedom to operate, receptor exclusivity, payload compatibility, manufacturing economics and the ability to transfer the technology into existing pipelines. Broad claims of modularity will carry limited weight without comparative datasets and clear intellectual property protection.
Commercial success may therefore emerge in stages. The initial milestones are likely to involve binder selection, preclinical validation and option exercises rather than clinical product revenue. Larger value inflection points would require external licences, investigational new drug applications and human evidence that the technology improves brain exposure without introducing unacceptable safety risks.
What clinicians and drug developers should watch as the collaboration advances
The most important upcoming disclosure will be the identity of the blood-brain barrier receptor and the rationale for choosing it. That information will allow industry observers to evaluate whether the partners are competing directly with transferrin receptor shuttles or pursuing a less crowded transport mechanism. Receptor choice will also indicate possible peripheral safety liabilities and the types of diseases or patient populations that may be prioritised.
A second indicator will be whether Nona Biosciences and Lonza select a demonstration payload that can produce an unambiguous pharmacodynamic signal. A well-characterised antibody or enzyme may provide a clearer test of delivery than an experimental payload whose biological activity is uncertain. The strongest validation would show increased parenchymal exposure, improved target engagement and comparable or reduced systemic toxicity relative to a non-shuttled version.
The collaboration is strategically credible because it links antibody discovery, molecular engineering, conjugation and manufacturing. It also reflects a wider industry shift towards treating delivery technology as a platform capable of supporting multiple neurological assets. However, the programme remains preclinical, its mechanism has not been disclosed and its best-in-class ambition is not yet supported by comparative evidence.
Nona Biosciences and Lonza will ultimately be judged by whether their single-domain antibody shuttle delivers active therapeutics into relevant brain regions at clinically practical doses. Until that evidence emerges, the partnership should be viewed as a potentially important enabling-technology initiative rather than proof that the blood-brain barrier bottleneck has been solved.
Why the Nona Biosciences and Lonza alliance matters for brain-penetrant biologics added by Pallavi Madhiraju on
View all posts by Pallavi Madhiraju →