Intranasal vs Encapsulated Delivery: An Evidence Comparison

Verified against No identity values apply to the page subject, so all are set to null rather than borrowed from an excipient. Representative excipient identities were verified individually against PubChem PUG-REST and are given in the excipient section and references: sodium caprate CID 4457968, salcaprozate sodium CID 23669833, 2-hydroxypropyl-beta-cyclodextrin CID 14049689, benzalkonium chloride CID 3014024. Fields marked “not verified” are ones we could not confirm against a primary chemical database — we leave them blank rather than guess.

Why this is a route comparison, not a compound comparison

The two formats do not share a molecular target, a mechanism or a class. What separates them is which biological barrier the payload has to cross, and the barriers are entirely different problems.

The intranasal format deposits an aqueous solution onto nasal respiratory and olfactory mucosa. Absorption from that surface bypasses gastrointestinal proteolysis and hepatic first-pass metabolism. The constraints are residence time, enzymatic activity in the mucosa, molecular size, and the small volume a metered spray can deliver.

The encapsulated format carries a much larger absolute mass in a solid dosage form, but the payload must survive gastric acid and pepsin, then gastric, intestinal and pancreatic proteases, then cross an epithelium that large hydrophilic molecules traverse poorly, then survive hepatic first pass. Research formats address this with enteric coating and intestinal permeation enhancers.

Because the subject is a route, chemical identity fields such as CAS number, molecular formula and molecular weight have no referent at the page level. Assigning an excipient's numbers to the page subject would be a category error, so they are reported separately and attributed to the excipient they belong to. The four excipients below were each verified individually against PubChem.

The excipients that carry the mechanism

On the encapsulated side, two intestinal permeation enhancers dominate the literature and have been compared directly for more than thirty years.

Salcaprozate sodium, usually called SNAC, is PubChem CID 23669833, molecular formula C15H20NNaO4, molecular weight 301.31, CAS 203787-91-1. Sodium caprate, the C10 enhancer, is CID 4457968, C10H19NaO2, 194.25, CAS 1002-62-6.

Twarog and colleagues reviewed the two side by side in Pharmaceutics in 2019 and reached a more equivocal position than the secondary literature usually reports (PMID 30781867). Sodium caprate is described as acting through tight-junction opening and membrane perturbation. For SNAC, the older explanation was increased transcellular passage via lipophilicity from non-covalent complexation; the more recent account, developed for oral semaglutide, is a pH-elevating, monomer-inducing and pepsin-inhibiting effect in the stomach. The review found equivocal evidence for discrete mechanisms at the epithelial level, particularly at the high quantities used in vivo, and concluded there is no convincing evidence that either enhancer is more efficacious than the other.

On the intranasal side the representative excipients are different in kind: solubilisers and preservatives rather than permeation enhancers. 2-hydroxypropyl-beta-cyclodextrin is CID 14049689, C63H112O42, 1541.5, CAS 107745-73-3. Benzalkonium chloride is CID 3014024; note that the PubChem record gives C22H40ClN at 354.0, a single homolog, while commercial benzalkonium chloride is a homolog mixture.

What limits absorption through the nose

Three constraints recur across the nasal delivery literature: residence time, molecular size and volume.

Merkus and colleagues reviewed nasal mucociliary clearance as a factor in nasal drug delivery in Advanced Drug Delivery Reviews (PMID 10837578). The system transports the mucus layer covering the nasal epithelium toward the nasopharynx by ciliary beating, and material deposited there is cleared rapidly. The review is notably cautious about the corollary that is usually asserted: it states that more experimental evidence is needed to support the conclusion that improved absorption from residence-time-extending formulations is actually caused by longer residence time.

On size, Huang and Donovan compiled absorption data for large molecules in the same journal issue (PMID 10837585). Their finding is specific and quantitative: for compounds larger than 1000 Da, nasal bioavailability can be predicted directly from molecular weight, and for these large molecules bioavailability generally falls between 0.5 and 5 percent. Particles up to approximately 1 micrometre were shown to enter the bloodstream after intranasal administration.

Al Bakri and colleagues later built a database of intranasally delivered peptides from a retrospective analysis and reported that peptides with molecular weights up to 6000 Da have been delivered by this route, with permeation enhancers and mucoadhesives showing promise in improving bioavailability (PMID 30173579). The 1000 Da figure is therefore a slope inflection, not a wall.

Nose-to-brain: the distinguishing claim, and its evidence grade

The feature that separates intranasal delivery from every other non-invasive route is the proposed direct pathway to the central nervous system, bypassing the blood-brain barrier. It is also where the evidence is thinnest relative to how confidently it is stated.

Qiu and colleagues reviewed the mechanisms in MedComm in 2025 (PMID 40487748), describing transport along olfactory and trigeminal nerve routes together with a respiratory pathway, and surveying delivery systems including nanocarriers and exosomes across neurodegenerative disease, acute neurological disease, brain tumours and psychiatric disorders. This is a review, not a primary measurement.

The worked example most often cited in this context also requires a correction. Aly and Waszczak's paper on intranasal gene delivery for Parkinson's disease, published in Expert Opinion on Drug Delivery (PMID 26289676), is indexed as a review, not as a primary study. It summarises evidence for intranasal delivery of biomolecules and describes the authors' own laboratory finding that intranasal glial cell line-derived neurotrophic factor, or plasmid DNA nanoparticles encoding it, was neuroprotective in a rat model of Parkinson's disease, with pericytes the likely transfected cell type. The authors themselves note that few genes encoding neurotrophic factors have been tested by this route and shown to be neuroprotective in a disease model.

Whether any given molecule reaches brain tissue at a meaningful concentration by this route is molecule-specific and is not established generally.

The encapsulated route's human proof point

The asymmetry between the two formats is clearest at the level of human validation, and it favours the encapsulated permeation-enhancer approach decisively.

Oral semaglutide is delivered as a tablet co-formulated with SNAC, and that formulation has been through large Phase 3 evaluation. NCT04707469, the PIONEER PLUS trial sponsored by Novo Nordisk, randomised 1,606 participants with type 2 diabetes to once-daily oral semaglutide at 14, 25 or 50 mg for 68 weeks. It started in January 2021 and completed in March 2023.

That programme functions as a proof of concept for the format itself: a permeation-enhancer capsule or tablet can make a peptide orally absorbable at a scale that supports regulatory approval. It does not generalise automatically to other payloads.

The magnitude should be stated plainly alongside it. Twarog and colleagues report that tablets containing these enhancers induce single-digit and highly variable increases in the oral bioavailability of macromolecule payloads in human trials, while noting this may be adequate for potent macromolecules. They also record that SNAC holds generally-regarded-as-safe status in the United States and sodium caprate has food-additive status, that no evidence of co-absorption of microorganisms had emerged from clinical trials to that point, and that effects of repeat administration beyond six months had not been assessed.

By comparison, nasal delivery to the central nervous system remains largely preclinical to early clinical.

What each format's evidence base can address

Neither format is superior in the abstract, and the literatures answer different questions.

The intranasal evidence base is strongest on mechanism and weakest on human outcome. It offers a well-characterised set of barriers, a quantitative relationship between molecular weight and bioavailability above 1000 Da, and a proposed direct central nervous system pathway supported mainly by tracer and animal work. It is the relevant literature for questions about small potent payloads and about central nervous system exposure by a non-invasive route.

The encapsulated permeation-enhancer evidence base is strongest on human outcome and, unusually, unsettled on mechanism. After thirty years of comparison the reviewers found equivocal evidence for discrete epithelial mechanisms between the two leading enhancers, yet the format carries a completed 1,606-participant Phase 3 programme. It is the relevant literature for questions about larger payload mass and about whether a solid oral dosage form can deliver a macromolecule at all.

Two things are not determined by either literature. Absolute human bioavailability by either route is molecule-specific and cannot be inferred from the format. And where a given payload sits relative to the nasal molecular-weight relationship depends on the specific sequence and its physicochemical properties, not on class membership.

A separate category of claims attaches to the handling of these formats, and most of those numbers have no primary source. They are listed below rather than repeated.

What is not known

Route comparisons are unusually prone to overreach because a finding for one payload is easily restated as a property of the format. Very little here generalises. Absolute human bioavailability by either route is molecule-specific, and no source located defines it for any payload other than the ones actually studied. The molecular-weight relationship for nasal absorption above 1000 Da is a statistical relationship derived from a compilation, not a threshold that predicts any individual sequence. The nose-to-brain pathway is supported mainly by tracer and animal work; whether a given molecule crosses at a concentration that matters is unestablished for arbitrary peptides, and the leading review on the mechanism is a review rather than a primary measurement. For the encapsulated format the human validation is real but narrow: it rests substantially on one molecule, semaglutide, in one formulation, and does not demonstrate that the approach transfers to other macromolecules. Mechanistic understanding of the two leading permeation enhancers remains contested after thirty years of direct comparison. Effects of repeat administration beyond six months had not been assessed for either enhancer at the time of the 2019 review, and no source located here updates that. Nothing retrieved addresses how these formats behave in any population defined by age, sex or disease state, because the delivery-science literature is largely organised around molecules and barriers rather than people.

Questions