The Molecular Trojan Horse: How BBB-Penetrating Antibodies Cross the Blood-Brain Barrier
Key takeaways
- The blood-brain barrier (BBB) blocks most biologics from reaching the central nervous system, making BBB-penetrating antibodies an important strategy for overcoming one of the defining bottlenecks in CNS drug development.
- BBB-penetrating antibodies use receptor-mediated transcytosis (RMT) to engage transport receptors on brain endothelial cells and move therapeutic cargo into the brain.
- Clinical validation is strongest today in lysosomal storage disorders, while Alzheimer’s disease programs such as Trontinemab remain investigational and require Phase 3 confirmation.
- Brain-shuttle design is a narrow engineering problem: receptor affinity, release kinetics, peripheral receptor expression, payload biology, manufacturability and safety must all be balanced.
- AI can support this workflow by prioritizing shuttle binders, modeling binder-target interfaces, optimizing affinity and developability, and flagging off-target or immunogenicity risks before wet-lab validation.
Why the blood-brain barrier blocks biologics
The blood-brain barrier is not simply a wall. It is a highly selective transport interface formed by specialized brain endothelial cells, tight junctions, pericytes, astrocytic end-feet and the broader neurovascular unit. Its job is to protect the brain’s microenvironment while allowing tightly controlled exchange of nutrients, ions and signaling molecules.
That selectivity is essential for neural function, but it also limits therapeutic access. Small molecules must satisfy demanding physicochemical constraints to cross efficiently. Large biologics, including monoclonal antibodies, enzymes and many recombinant proteins, are even more restricted. In the BBB literature, systemic antibody exposure in brain tissue is often described as only a small fraction of plasma exposure, commonly not exceeding 0.1% of injected dose depending on the molecule, species, assay and tissue compartment measured [1-3].
For CNS disease, that creates a familiar development dilemma. Higher systemic doses may increase brain exposure, but they can also increase peripheral toxicity, cost and treatment burden. The central question for antibody engineers is therefore no longer only “what target should the antibody bind?” It is also “how can the molecule reach the right brain compartment at a therapeutically useful exposure?”
The Trojan Horse mechanism: receptor-mediated transcytosis
A new generation of BBB-penetrating antibodies aims to use the brain’s own transport machinery. Instead of nonspecifically opening the BBB, these molecules exploit receptor-mediated transcytosis, an endogenous process in which selected macromolecules bind receptors on the blood-facing side of brain endothelial cells, are internalized, trafficked across the cell and released toward the brain side [2,3].
The most widely explored receptors include transferrin receptor 1 (TfR1), insulin receptor, insulin-like growth factor 1 receptor (IGF1R) and other transport-associated receptors under active investigation. In practice, many brain-shuttle therapeutics are built as bispecific antibodies or fusion proteins with two functional modules:
- The shuttle module binds a BBB transport receptor, often TfR1, in a way that supports internalization without blocking the receptor’s normal biological function.
- The therapeutic module carries disease-relevant activity, such as amyloid-beta binding, enzyme replacement, progranulin delivery or another CNS-directed payload.
This Trojan Horse design turns the barrier into a selective gateway. The challenge is that transport is not guaranteed by receptor binding alone. If receptor affinity is too high, the complex can be trapped in endothelial cells or routed to lysosomes. If affinity is too low, the molecule may fail to engage the transport pathway. Productive delivery often depends on a balanced or pH-sensitive interaction: enough binding on the blood side, efficient intracellular trafficking and release on the brain side.
Figure 1. Receptor-mediated transport of a bispecific brain-shuttle antibody across the blood-brain barrier.
Clinical validation: from Hunter syndrome to Alzheimer’s trials
The clearest clinical proof points for BBB-crossing biologics are currently emerging in lysosomal storage disorders, where the therapeutic cargo and biomarker strategy are well defined. Pabinafusp alfa (JR-141 / IZCARGO) is approved in Japan for mucopolysaccharidosis type II (MPS II, Hunter syndrome) and uses a transferrin receptor-based delivery concept to bring iduronate-2-sulfatase across the BBB [4]. In the United States, tividenofusp alfa-eknm (DNL310 / AVLAYAH) received FDA accelerated approval in March 2026 for neurologic manifestations of Hunter syndrome in certain pediatric patients, with continued approval contingent on confirmatory evidence [5,6].
In Alzheimer’s disease, Trontinemab (RG6102) is one of the most visible investigational brain-shuttle antibodies. It combines an amyloid-beta antibody backbone with Roche’s Brainshuttle technology to increase delivery across the BBB. Non-human primate modeling reported higher brain exposure than the parent antibody format, and Roche/Genentech reported Phase Ib/IIa data in 2025 showing 91% amyloid PET negativity at 28 weeks in the 3.6 mg/kg cohort, with ARIA-E remaining below 5% in blinded safety data [7,8]. Those data are promising, but they are not yet proof of clinical efficacy. Trontinemab is now being evaluated in Phase 3 TRONTIER studies in early symptomatic Alzheimer’s disease [9,10].
In frontotemporal dementia, DNL593 (PTV:PGRN; formerly TAK-594) is designed to deliver progranulin across the BBB for GRN-related FTD. Denali announced in April 2026 that Takeda terminated the collaboration for strategic reasons, not due to safety or efficacy data, and that Denali regained full rights while continuing clinical development. Results from the ongoing Phase 1/2 study are expected by the end of 2026 [11].
The same delivery logic is relevant in neuro-oncology, but the clinical evidence is less mature. A 2026 review of biologic delivery to tumors in the brain notes that the BBB can remain functional in micro-tumors and infiltrative tumor margins, limiting biologic distribution even when the tumor core appears more permeable [12]. For this reason, brain-shuttle platforms may eventually support antibody, enzyme, oligonucleotide or immune-modulating payloads in brain tumors, but claims about clinical use in glioblastoma should be framed as exploratory unless tied to named, sourced programs.
Candidate | Developer / mechanism | Indication | Publication-safe status |
|---|---|---|---|
Pabinafusp alfa / JR-141 / IZCARGO | JCR Pharmaceuticals; anti-human TfR antibody fused to iduronate-2-sulfatase | Hunter syndrome / MPS II | Approved in Japan; named patient access in certain countries where not commercially available [4]. |
Tividenofusp alfa-eknm / DNL310 / AVLAYAH | Denali Therapeutics; TfR-enabled Enzyme TransportVehicle fused to iduronate-2-sulfatase | Neurologic manifestations of Hunter syndrome / MPS II | FDA accelerated approval in the U.S. for specified pediatric patients; confirmatory COMPASS study ongoing [5,6]. |
Trontinemab / RG6102 | Roche / Genentech; Brainshuttle anti-amyloid-beta antibody using TfR1-mediated delivery | Early symptomatic Alzheimer’s disease | Investigational; Phase 3 TRONTIER 1 and TRONTIER 2 studies recruiting [8-10]. |
DNL593 / PTV:PGRN / formerly TAK-594 | Denali Therapeutics; Protein TransportVehicle delivering progranulin | GRN-related frontotemporal dementia | Investigational Phase 1/2; Denali regained full rights after Takeda terminated collaboration; results expected by end of 2026 [11]. |
Table 1. Selected receptor-mediated transcytosis biologics and brain-shuttle programs
Engineering the shuttle: affinity, release and safety
- Affinity must be optimized, not maximized. For a standard antagonist antibody, tighter binding can be beneficial. For a shuttle arm, excessive affinity may reduce productive transcytosis by retaining the molecule at the receptor or routing it into degradation pathways. Monovalent, moderate-affinity or pH-dependent designs can help balance receptor engagement with release.
- The shuttle must avoid blocking normal receptor function. TfR1 and related receptors exist because the brain needs iron, amino acids and other essential inputs. A therapeutic shuttle should bind an epitope and geometry that supports transport without disrupting endogenous ligand biology.
- Peripheral receptor expression matters. TfR1 is not exclusive to the BBB. It is expressed in peripheral tissues, including erythroid lineage cells. This creates potential safety and pharmacokinetic liabilities, such as peripheral sink effects or hematologic toxicity, especially when receptor engagement is too strong or too broad.
- Payload biology still determines therapeutic value. Improved brain exposure does not guarantee efficacy. The payload must engage the right target, in the right disease stage, at a concentration and duration that changes disease biology.
- Developability remains central. Brain-shuttle constructs can be multispecific, fusion-based and structurally complex. They must still satisfy expression yield, stability, solubility, aggregation risk, immunogenicity, manufacturability and scalable analytics.
Where AI fits in brain-shuttle antibody design
AI should not be framed as replacing wet-lab validation or compressing development from years to months. A more credible view is that AI can reduce the number of poor candidates entering synthesis and testing, while helping teams understand trade-offs earlier.
For BBB-penetrating antibodies, AI-enabled workflows can support five design missions:
- Prioritizing shuttle binders: rank receptor-binding candidates by predicted affinity, specificity, epitope accessibility and developability before experimental screening.
- Designing the “Goldilocks” affinity window: guide mutations that tune binding strength, avidity and pH dependence.
- Modeling multispecific architecture: identify steric clashes, unfavorable linker geometries or conformations that may reduce binding or manufacturability.
- Flagging safety and developability risk: screen for aggregation-prone regions, sequence liabilities, immunogenicity risk, off-target binding and tissue-expression concerns.
- Mining new shuttle biology: use single-cell, proteomic and transcriptomic datasets to identify BBB-enriched receptors or disease-context transport pathways worth validating experimentally.
This is where the topic fits Ardigen’s AI for Biologics positioning. Ardigen’s biologics offering already maps to lead generation, hit screening, binder-target interface analysis, protein structure modeling, lead optimization, immunogenicity assessment and developability evaluation [14]. For brain-shuttle programs, those capabilities can be framed as a decision-support layer for candidate prioritization rather than a promise of instant clinical translation.
What this means for CNS biologics development
The field is moving from a simple access problem to a systems-engineering problem. A successful BBB-penetrating biologic must combine transport, payload potency, brain distribution, peripheral safety, disease-stage selection and practical manufacturability.
For lysosomal storage disorders, approved and late-stage programs show that BBB-crossing biologics can move beyond platform claims into regulatory reality. For Alzheimer’s disease, Trontinemab is testing whether higher brain exposure can improve the benefit-risk profile of anti-amyloid therapy. For neuro-oncology and other CNS indications, the rationale is strong, but programs should be described as emerging or exploratory unless clinical validation is available.
The strongest conclusion is not that the BBB has been “solved.” It is that receptor-mediated delivery has become one of the most important engineering frontiers in CNS biologics — and that AI can help teams navigate the design space with greater precision before committing to expensive experimental cycles.
Conclusion: from physiological wall to engineered gateway
BBB-penetrating antibodies are changing how CNS biologics are designed. By using receptor-mediated transcytosis, they aim to transport large therapeutic molecules across a barrier that historically excluded most antibody-based drugs from the brain.
The opportunity is substantial, but the bar is high. Brain-shuttle antibodies must demonstrate not only improved exposure but also safety, manufacturability and clinical benefit. The next era will belong to teams that treat BBB delivery as an integrated design challenge: transport receptor biology, antibody engineering, payload pharmacology, translational biomarkers and AI-guided optimization working together.
Next step
Designing a brain-shuttle biologic? Ardigen’s AI for Biologics team can help prioritize shuttle receptors, model binder-target interactions, assess developability and optimize affinity or stability before wet-lab validation. Book a 30-minute biologics design review to map your CNS delivery concept against an AI-enabled antibody engineering workflow.
Frequently Asked Questions
What are BBB-penetrating antibodies?
BBB-penetrating antibodies are engineered biologics designed to cross the blood-brain barrier more efficiently than standard monoclonal antibodies. Many use receptor-mediated transcytosis, where one part of the molecule binds a transport receptor on brain endothelial cells while another part carries the therapeutic activity.
What is receptor-mediated transcytosis?
Receptor-mediated transcytosis is a cellular transport process in which a molecule binds a receptor, is internalized into vesicles, moves across the cell and is released on the other side. Brain-shuttle antibodies use this process to move biologics across brain endothelial cells without broadly disrupting the BBB.
Why is transferrin receptor used for brain-shuttle antibodies?
Transferrin receptor 1 is expressed on brain endothelial cells and participates in iron-related transport biology. Antibody engineers can design shuttle arms that bind TfR1 while aiming to preserve normal receptor function, but affinity, epitope and peripheral receptor expression must be carefully optimized.
Are any BBB-crossing biologics approved?
Yes, selected BBB-crossing enzyme replacement therapies are approved for Hunter syndrome. Pabinafusp alfa is approved in Japan, and tividenofusp alfa-eknm / AVLAYAH received FDA accelerated approval in the U.S. for neurologic manifestations of Hunter syndrome in specified pediatric patients [4-6]. Alzheimer’s disease brain-shuttle programs such as Trontinemab remain investigational.
What is Trontinemab?
Trontinemab is Roche/Genentech’s investigational Brainshuttle anti-amyloid-beta antibody for Alzheimer’s disease. It is designed to improve delivery across the BBB via TfR1-mediated transport and is being evaluated in Phase 3 TRONTIER studies [8-10].
What are the main risks in brain-shuttle antibody design?
The main risks include insufficient brain exposure, receptor trapping, lysosomal degradation, peripheral receptor binding, hematologic or other systemic toxicity, immunogenicity, aggregation, poor manufacturability and lack of clinical benefit despite improved target engagement.
How can AI help design BBB-penetrating antibodies?
AI can help prioritize receptor binders, model binder-target interfaces, tune affinity, evaluate multispecific architecture, predict developability and mine omics data for new transport receptors. These methods do not replace experiments, but they can reduce the number of low-probability candidates that enter wet-lab testing.
What should be validated experimentally?
Teams should validate receptor binding, competition with endogenous ligands, internalization, transcytosis, brain exposure, tissue distribution, payload activity, safety, immunogenicity and manufacturability. AI predictions are most useful when they are connected to assays that can confirm or reject the design hypothesis.
Technical editing: Ardigen expert: Joanna Marczyńska-Grzelak, PhD
References
- Kadry H, Noorani B, Cucullo L. A blood-brain barrier overview on structure, function, impairment, and biomarkers of integrity. Fluids and Barriers of the CNS. 2020;17:69. doi:10.1186/s12987-020-00230-3.
- Terstappen GC, Meyer AH, Bell RD, Zhang W. Strategies for delivering therapeutics across the blood-brain barrier. Nature Reviews Drug Discovery. 2021;20:362-383. doi:10.1038/s41573-021-00139-y.
- Zhao P, Zhang N, An Z. Engineering antibody and protein therapeutics to cross the blood-brain barrier. Antibody Therapeutics. 2022;5(4):311-331. doi:10.1093/abt/tbac028.
- JCR Pharmaceuticals. NPS Program for IZCARGO; Product information for IZCARGO. Accessed 3 June 2026. https://www.jcrpharm.co.jp/en/company/nps-program/ and https://www.jcrpharm.co.jp/en/ir/product/
- U.S. Food and Drug Administration. FDA Approves Drug to Treat Neurologic Manifestations of Hunter Syndrome. 25 March 2026. https://www.fda.gov/news-events/press-announcements/fda-approves-drug-treat-neurologic-manifestations-hunter-syndrome
- Denali Therapeutics. Denali Therapeutics Announces U.S. FDA Approval of AVLAYAH (tividenofusp alfa-eknm) for Treatment of Hunter Syndrome (MPS II). 25 March 2026. https://investors.denalitherapeutics.com/news-releases/news-release-details/denali-therapeutics-announces-us-fda-approval-avlayahtm
- Grimm HP, Schumacher V, Schäfer M, et al. Delivery of the Brainshuttle amyloid-beta antibody fusion trontinemab to non-human primate brain and projected efficacious dose regimens in humans. mAbs. 2023;15(1):2261509. doi:10.1080/19420862.2023.2261509.
- Genentech. Genentech and Roche Present New Insights in Alzheimer’s Disease Research Across Its Diagnostics and Pharmaceutical Portfolios at AAIC. 27 July 2025. https://www.gene.com/media/press-releases/15072/2025-07-27/genentech-and-roche-present-new-insights
- ClinicalTrials.gov. NCT07169578, TRONTIER 1: A Study of Trontinemab in Participants With Early Symptomatic Alzheimer’s Disease. Accessed 3 June 2026. https://clinicaltrials.gov/study/NCT07169578
- ClinicalTrials.gov. NCT07170150, TRONTIER 2: A Clinical Trial of Trontinemab in Participants With Early Symptomatic Alzheimer’s Disease. Accessed 3 June 2026. https://clinicaltrials.gov/study/NCT07170150
- Denali Therapeutics. Denali Therapeutics Regains Full Rights to Investigational Therapy DNL593 (PTV:PGRN) for GRN-related Frontotemporal Dementia (FTD-GRN). 3 April 2026. https://investors.denalitherapeutics.com/news-releases/news-release-details/denali-therapeutics-regains-full-rights-investigational-therapy
- Gampa G, Vadlakonda R, Stefanich E, Kamath AV, Sadekar S, Shivva V. Bridging the blood-brain barrier: strategies to improve delivery of biologics to tumors in the brain. Fluids and Barriers of the CNS. 2026. doi:10.1186/s12987-026-00760-2.
- Denali Therapeutics. Second Quarter 2023 Financial Results and Business Highlights / SEC Exhibit 99.1. 8 August 2023. https://www.sec.gov/Archives/edgar/data/1714899/000171489923000119/ex991pressreleaseq22023.htm
- Ardigen. AI for Biologics. Accessed 3 June 2026. https://ardigen.com/ai-for-biologics/
