BX LAB TECHNICAL ARTICLE
Antibody-Oligonucleotide Conjugates: Three Coupled Development Problems
An AOC is usually drawn as antibody, linker and oligonucleotide. In development, each choice changes the behavior and test burden of the other two.
Key takeaways
Receptor binding and internalization do not guarantee productive cytosolic or nuclear delivery.
Conjugation site, linker and oligonucleotide-to-antibody ratio define a product population, not just a structure on paper.
Clinical progress is real, but pipeline stages must be dated and checked against current company and registry records.
An AOC is not an ADC with the payload swapped
The familiar three-part drawing invites an easy analogy: antibody, linker, payload. The analogy stops being useful when the payload is a large, highly charged oligonucleotide rather than a hydrophobic cytotoxin. An AOC must reach a selected cell, enter it, avoid unproductive trafficking and place an intact siRNA, ASO, PMO or other nucleic acid where its molecular machinery operates. The antibody can improve tissue targeting. It cannot, by itself, complete that sequence of events.
Target binding is only the first gate
A useful receptor needs the right tissue distribution, sufficient surface exposure and an internalization route compatible with repeat dosing. Antibody format then changes the rest of the molecule. A full IgG brings long circulation and bivalent binding but also size and tissue-penetration constraints. A Fab is smaller and clears differently. TfR1-directed muscle programs illustrate both approaches, but a receptor that carries an AOC into an endosome is not automatically a receptor that delivers functional oligonucleotide to the cytosol or nucleus.
Uptake is not productive delivery
After receptor-mediated uptake, much of an internalized conjugate can remain in endosomal compartments or continue toward lysosomal processing. Fluorescence inside a cell therefore does not prove that the payload reached RISC, nuclear pre-mRNA or another intended site of action. Development assays need to distinguish binding, total uptake, intracellular trafficking, payload release and functional gene modulation. Combining those measurements is slower than reporting one uptake number, but it identifies whether the limitation sits in the antibody, linker or oligonucleotide.
Conjugation chemistry sets the population you dose
Random lysine coupling and partial reduction of native disulfides can produce broad distributions of attachment sites and oligonucleotide-to-antibody ratios. That may be acceptable for an early screen, but it complicates comparability, purification and release testing. Glycan remodeling, engineered cysteines, enzyme-mediated ligation and other site-specific methods can narrow the population. They also introduce their own raw materials, reaction controls and analytical questions. Site-specific does not mean process-free; it moves variability into a more deliberate part of the process.
The payload changes the release problem
siRNA, gapmer ASOs, splice-switching oligonucleotides and PMOs do not share one intracellular destination or one stability profile. Their backbone chemistry, sugar modifications, length, terminal handles and counterions affect conjugation, charge, nuclease resistance and analysis. A linker that is stable for one payload may give poor release or aggregation with another. This is why antibody, linker and oligonucleotide cannot be optimized in separate workstreams and joined at the end.
Pipeline labels need a date stamp
The source article's pipeline table already needs updates. As checked on 20 July 2026, Avidity lists del-desiran in the Phase 3 HARBOR trial and del-brax in Phase 1/2, not Phase 3. Dyne initiated the Phase 3 HARMONIA trial of z-basivarsen in March 2026, while the ClinicalTrials.gov record for Tallac's TAC-001 describes a Phase 1/2 study that is active but not recruiting. Novartis completed its $12 billion acquisition of Avidity in February 2026. That is enough to show movement. It is not a basis for saying every leading AOC is late-stage or close to approval.
CMC has to join the project early
A development specification needs more than sequence and antibody identity. It should define attachment site, oligonucleotide-to-antibody ratio and distribution, free oligonucleotide, unconjugated antibody, aggregates, fragments, charge variants, payload integrity and a potency method tied to the intended mechanism. Stability studies must follow both halves of the conjugate and the linker between them. If those methods arrive after the lead has been selected, the project may discover that its best biological construct is the hardest one to manufacture reproducibly.
A useful supplier brief starts at the conjugation interface
For the oligonucleotide component, specify the complete sequence, backbone and sugar pattern, terminal reactive handle, spacer, required counterion, purity basis and the analytical method used to quantify free or damaged payload. State whether the supplier is delivering an unconjugated oligonucleotide, an activated intermediate or a finished conjugate. Include the intended conjugation chemistry and scale, because a handle that performs cleanly in a small model reaction may behave differently with a charged therapeutic sequence and a full antibody. That brief gives chemistry, protein and analytical teams the same molecule to discuss.
References
Technical context is supported by the peer-reviewed literature below.
- 生物学长 (16 July 2026) 抗体-核酸偶联物（AOC）全球研究进展与发展综述
- Fan et al. (2026) Research progress and development strategies of antibody-oligonucleotide conjugates. Gene Therapy
- Novartis (27 February 2026) Completion of the Avidity Biosciences acquisition
- Avidity Biosciences: clinical development programs and current phase information
- Dyne Therapeutics (8 March 2026) Initiation of the Phase 3 HARMONIA trial of z-basivarsen in DM1
- ClinicalTrials.gov NCT05399654: Phase 1/2 INCLINE-101 study of TAC-001