The central insight of the paper is a distinction that the authors believe the field has too often blurred. Payload persistence describes how long the delivered RNA, and the protein or editing activity it directs, remains present in the body. Biological-outcome persistence describes how long the therapeutic effect itself endures, whether that is circulating protection against a virus, the survival of genetically corrected cells or a stabilised tissue state. These two quantities can diverge dramatically. In vivo base editing, for instance, is deliberately transient at the payload level, because limiting the exposure time of editing enzymes is a safety feature, yet its intended outcome is a permanent genomic correction that persists for the life of the patient. Durability, the authors insist, should be judged on outcomes rather than on the fleeting presence of the therapeutic molecule itself.

Existing nanomedicine evaluation frameworks, the paper argues, are structurally incomplete. Performance is typically measured through particle size, polydispersity, encapsulation efficiency, surface charge, biodistribution, cellular uptake, endosomal escape and early protein expression. These metrics are necessary, but they cannot explain why some biological effects persist while others rapidly decline. Cells do not receive RNA in a neutral environment. They interpret the delivery event through innate immune sensing, antigen presentation, metabolic tone, chromatin state, tissue inflammation and repair history. It is within this interpretive layer, the authors contend, that the next conceptual advance in RNA nanomedicine may lie, transforming durability from a secondary observation into an active design problem.

To avoid a category error, the framework carefully bounds what it means by biological memory. Adaptive immune memory resides in antigen-specific B and T lymphocytes and long-lived plasma cells, persists for months to years, and is measured through serology, germinal-centre activity and recall assays. Trained innate immunity resides in myeloid cells and their bone-marrow progenitors, is encoded epigenetically and metabolically over weeks to months, and is tracked by chromatin accessibility, histone marks and cytokine recall responses. Epigenetic states in non-immune cells operate on chromatin at disease-relevant loci, while regenerative repair functions at the level of tissue niches and resident progenitors. The hypothesis does not claim these systems share a single molecular mechanism. Rather, it groups them by a shared functional property: each can store a change induced around the time of delivery and influence the biological response long after the RNA itself has been degraded.

The evidence that mRNA lipid nanoparticle platforms can indeed generate durable biological memory is already substantial. In a landmark human study, SARS-CoV-2 mRNA vaccines induced persistent germinal-centre responses in draining lymph nodes, with spike-binding germinal-centre B cells still detectable at least twelve weeks after vaccination. A separate study documented durable memory B cell and T cell responses lasting at least six months, even as antibody profiles continued to evolve. These findings demonstrate that durability is inherently multidimensional: antibody titres, memory cell persistence and protection against variants do not always move together. Gene therapy offers a parallel narrative. CRISPR-based medicines have entered clinical practice, exemplified by exagamglogene autotemcel for severe haemoglobinopathies, and an individualised in vivo base-editing therapy for a rare genetic disease recently showed how quickly bespoke RNA-guided editing can now be developed and delivered via lipid nanoparticles.

The conceptual heart of the paper is the claim that therapeutic durability reflects a balance between biological forgetting and biological retention. Forgetting occurs when expression fades, antibody titres decline, edited cells disappear or a repaired tissue relapses into a pathological baseline. Retention occurs when memory cells, corrected cell populations or stable expression programmes keep the desired effect biologically available. Because nanocarriers reshape biodistribution, tissue exposure, pharmacokinetic behaviour and the immune context in which RNA is interpreted, the authors argue they can tip this balance. Lipid nanoparticles can stimulate innate immune pathways, interact with antigen-presenting cells and influence T follicular helper responses, properties that are advantageous for vaccines but potentially problematic for repeated protein replacement dosing.

Crucially, the hypothesis is constructed to be falsifiable rather than merely rhetorical. The authors define nanocarrier memory capacity as a carrier-attributable contrast: the difference in a durability endpoint between two formulations matched on identical RNA cargo and titrated to equivalent early exposure, defined as matching peak protein expression and the expression curve over the first twenty-four to seventy-two hours. Only then are the formulations allowed to differ in carrier variables such as ionisable lipid identity, helper-lipid ratio, surface chemistry or particle size. Any divergence in longitudinal durability endpoints under these conditions constitutes the carrier-attributable signal. Adjuvanticity is not assumed away but measured directly through early innate activation readouts. If adequately powered matched comparisons show no reproducible differences, then nanocarrier memory capacity, as defined, does not exist as a separable determinant, and durability is better explained by cargo design, dose, target-cell longevity and host context. The authors explicitly present this negative outcome as a realistic and reportable possibility.

The framework also introduces a quantifiable measure of clinical relevance: the therapeutic persistence window. This is defined as the period during which a biological response remains strong enough to provide clinical benefit, and it requires three specified elements: a measured durability quantity, a fitted decay model and a prespecified clinical-benefit threshold. Candidate models include monophasic exponential decay, describing a single dominant loss process, and biphasic decay, describing rapid early loss followed by a slower-decaying durable compartment, the shape expected when short-lived antibody-secreting cells decline while long-lived plasma cells and memory cells persist. The authors propose a minimum durability reporting set comprising peak response, decay slope, persistence window, recall capacity and the biological mechanism of loss or maintenance, arguing that adopting such a standard would let the field compare formulations on the shape of their durability rather than on a single early timepoint.

Safety receives symmetrical treatment in the framework, which is unusual and notable. Because trained innate immunity reprogrammes myeloid cells and their progenitors epigenetically and metabolically, the same imprinting that might support durable host defence could, if poorly regulated, contribute to chronic or maladaptive inflammation, a genuine concern for repeat-dose lipid nanoparticle platforms. The authors propose that repeat-dose development include innate-memory safety readouts alongside durability metrics: longitudinal myeloid training signatures, cytokine-recall profiles after standardised secondary stimulation, markers of progenitor-compartment skewing and tolerability across repeated doses. A formulation that raises a durability metric while simultaneously raising a maladaptive-imprinting metric should be flagged rather than rewarded. The desirable memory profile, in this framing, is one that maximises intended retention while keeping innate imprinting within a defined safety envelope.

The authors are candid about the limits of portability. A formulation optimised for the liver may not translate to spleen, bone marrow, muscle, lung, tumour or central nervous system compartments, and memory biology itself differs markedly across these tissues. They therefore hold that nanocarrier memory capacity is not a portable property but must be defined per tissue and per indication, with extrahepatic platforms evaluated for their memory consequences in the specific compartments they reach. Durability, they argue, is best modelled as a property of the carrier-by-host interaction, estimated within host strata defined by prior immune history, age, sex and pharmacogenomic background. If formulation rankings reorder across host contexts, that interaction is itself the finding. The hypothesis reframes RNA nanomedicine around a deceptively simple proposition: the future of the field may depend not only on delivering better instructions, but on helping cells and tissues remember the right instructions for long enough to matter. Whether matched-comparison experiments ultimately vindicate nanocarrier memory capacity or relegate durability to cargo and host biology, the framework promises to make that answer measurable, and the authors contend that either outcome would genuinely clarify the science.

Subject of Research: A hypothesis framework proposing that lipid nanocarriers can influence biological memory systems to improve the durability of mRNA vaccines and gene therapies

Article Title: The biological memory buffer hypothesis: can programmable nanocarriers shape cellular memory to improve the durability of mRNA vaccines and gene therapies?

Article References: Ugwu, O. P.-C., Ugwu, C. N., Anyanwu, G. E., Basajja, M., & Ebuoh, M. C. (2026). The biological memory buffer hypothesis: can programmable nanocarriers shape cellular memory to improve the durability of mRNA vaccines and gene therapies?. Frontiers in Drug Delivery, 6, Article 1902184. https://doi.org/10.3389/fddev.2026.1902184

Image Credits: AI Generated

DOI: 10.3389/fddev.2026.1902184

Keywords: mRNA vaccines, lipid nanoparticles, gene therapy, therapeutic durability, biological memory, trained immunity, nanomedicine, persistence window, memory B cells, CRISPR, epigenetic regulation, RNA therapeutics