The advent of antibody–drug conjugates (ADCs) has already transformed the landscape of targeted cancer therapy, offering a compelling strategy for directing cytotoxic payloads specifically to tumour cells. Yet ADCs carry well-documented limitations: the large molecular weight of immunoglobulins constrains their penetration into solid tumour tissue, and their reliance on a single surface biomarker for recognition leaves room for off-target effects and the emergence of resistance. A study published in Nature Biotechnology on 27 March 2026 by Chen, López-Tena, Russo, and colleagues at the University of Geneva proposes a conceptually novel solution to these constraints, one in which DNA itself is repurposed as a molecular computing device to regulate drug release with unprecedented selectivity.
The core innovation of the system lies in the integration of two established but previously separate technologies: DNA-based logic circuits and hybridization chain reactions (HCRs). The researchers designed affibody–DNA and aptamer–DNA conjugates, each engineered to recognise a distinct cell-surface biomarker — specifically the epidermal growth factor receptor (EGFR) and programmed death-ligand 1 (PD-L1), both of which are well-characterised targets in oncology. Critically, rather than acting independently, the two biomarker-binding modules are designed to function as a Boolean AND gate: the therapeutic response is initiated only when both biomarkers are simultaneously present in close proximity on the target cell surface. This dual-recognition requirement substantially narrows the specificity window, potentially reducing the risk of inadvertent activation in healthy tissues that may express only one of the two markers.
When co-engagement occurs, the proximity of the two conjugates generates an initiator sequence that triggers an HCR — a cascade of stepwise DNA hairpin assembly events that proceeds isothermally and without enzymatic input. Each round of the reaction incorporates additional DNA strands carrying cytotoxic drug molecules attached via cathepsin-cleavable linkers. The assembled DNA–drug complex subsequently undergoes endocytosis; once internalized, the cathepsin-mediated cleavage releases the payload within the lysosomal compartment. The authors report a signal amplification exceeding 100-fold relative to the initial biomarker input, a figure that could translate into meaningful therapeutic gain at low receptor densities.
Beyond its elegance as a molecular mechanism, the platform demonstrates considerable modularity. Different cytotoxic agents may be conjugated to the DNA hairpins, and the study illustrates that drug combinations can be incorporated within the same HCR assembly. The system was also shown capable of recruiting generic antibodies to the target cell surface, further broadening its potential utility.
Limitations, however, merit careful consideration. The present study is restricted to in vitro cell-line models, and the translation of a multicomponent DNA-based architecture to in vivo conditions raises important questions regarding systemic stability, immunogenicity, and pharmacokinetic behavior. The reliance on EGFR and PD-L1 co-expression as a proof-of-concept also reflects an idealized tumor antigen profile not universally encountered in clinical practice. Furthermore, the efficient internalization of large DNA assemblies — and their subsequent endosomal processing — will require validation across diverse tumor types.
Nevertheless, the conceptual framework introduced here could prove influential. By merging programmable DNA chemistry with amplified drug delivery, this approach suggests that therapeutic molecules might one day ‘compute’ their biological environment before acting — a prospect with profound implications not only for oncology but for any clinical context in which the simultaneous interrogation of multiple molecular signals would confer diagnostic and therapeutic advantage.
Paolo Rega
Leave a Reply