Semaglutide nanoparticles: 2026 HIP study

Getting semaglutide into the body orally is harder than it looks. The molecule is a 4,114-dalton GLP-1 receptor agonist, and like most peptide drugs it is broken down by proteolytic enzymes in the stomach and small intestine before it can cross into the bloodstream. Novo Nordisk solved this commercially with Rybelsus, an oral tablet that pairs semaglutide with a permeation enhancer called sodium N-(8-[2-hydroxybenzoyl] amino) caprylate (SNAC). But SNAC-based absorption is notoriously variable, and the bioavailability of oral Rybelsus in clinical conditions runs around 1% compared with the subcutaneous injection forms^[1]. That gap between a working injectable and a consistently absorbable tablet is what a group of Italian researchers set out to address in a paper published in the International Journal of Pharmaceutics on 8 August 2026.

What hydrophobic ion pairing does

The paper describes a two-stage strategy. First, the researchers used hydrophobic ion pairing (HIP) to render semaglutide more lipid-compatible. HIP works by mixing a charged peptide with an oppositely charged lipid; the resulting complex is less water-soluble and more oil-soluble than the peptide alone. The cationic lipid chosen was DOTAP, a synthetic phospholipid that carries a permanent positive charge. Semaglutide, which carries negative charges at physiological pH, forms tight electrostatic complexes with DOTAP when the two are combined in defined molar ratios^[1]. The team tested ratios from semaglutide:DOTAP of 1:0 (no DOTAP, a control) up to 1:18. As the DOTAP proportion rose, the complex became progressively more lipophilic.

The second stage was to load the semaglutide-DOTAP complex into solid lipid nanoparticles (SLNs). SLNs are submicron particles made from lipids that are solid at body temperature. They are established as pharmaceutical carriers because they protect encapsulated drugs from enzymatic degradation, can be surface-modified to interact with mucus or epithelial cells, and are produced from generally recognised as safe lipid excipients. The lipid matrix chosen here was cetyl palmitate, a waxy solid with a good safety record^[1].

The role of the microfluidic device

Producing uniform, small nanoparticles at scale is one of the persistent manufacturing challenges in nanomedicine. Conventional batch processes produce particles with wide size distributions. The team used a herringbone microfluidic mixing device to generate the SLNs. In microfluidic production, small volumes of lipid and aqueous solutions are pumped through channels that force rapid, controlled mixing; the result is more uniform particle populations than batch emulsification. The herringbone channel geometry creates chaotic advection that folds the two fluid streams together efficiently^[1]. This production method is commercially significant: it can in principle be parallelised, which makes it a step toward manufacturing scale rather than a purely laboratory curiosity.

What the optimal formulation achieved

The best-performing formulation, labelled F10 in the paper, used a semaglutide:DOTAP molar ratio of 1:18 and a peptide loading of 10% by weight. The resulting particles were under 300 nm in diameter, carried a positive zeta potential (important for mucoadhesion), and showed almost complete encapsulation efficiency, meaning essentially all the semaglutide that was added ended up inside the particles rather than remaining in solution^[1]. In cell-based permeability assays using intestinal epithelial models, F10 produced approximately a 6-fold increase in apparent permeability compared to free semaglutide not processed through the HIP-SLN system^[1]. The formulation was also stable in simulated gastrointestinal conditions and showed acceptable biocompatibility at the concentrations tested in cell cultures.

What was not tested

The entire dataset is in vitro. No animal pharmacokinetic studies are reported, and there are no human data. In vitro permeability models using cell monolayers are useful screening tools, but they do not capture the full complexity of oral absorption: mucus turnover rates, intestinal transit, first-pass liver metabolism, and inter-individual variability in gut physiology all affect how much of any formulation actually reaches the portal circulation. A 6-fold improvement in a Caco-2 or similar cell assay does not translate directly to a 6-fold improvement in human bioavailability.

The paper also does not report what happens to the formulation over time. Shelf life, temperature sensitivity, and whether the cetyl palmitate matrix remains solid and intact during cold-chain storage or at room temperature are practical questions that matter for a drug that is often prescribed to patients who travel or live in warm climates.

Where this fits in the oral GLP-1 landscape

Oral delivery of GLP-1 receptor agonists is an active area of pharmaceutical research. The EMA authorised oral semaglutide as Rybelsus in 2020 for type-2 diabetes^[2], and that authorisation validated demand for a non-injectable GLP-1 option, but the bioavailability ceiling of the SNAC system has driven parallel work on lipid-based and nanoparticle-based delivery systems. Several research groups are pursuing lipid nanoparticles, polymeric nanoparticles, exosome-mimicking carriers, and mucoadhesive systems. HIP applied to GLP-1 receptor agonists is not new, but combining HIP with microfluidic SLN production for semaglutide specifically, with the herringbone device geometry, is the distinct contribution of this paper.

From a clinical translation standpoint, the gap between a 6-fold permeability improvement in cells and a viable oral product that competes with subcutaneous semaglutide on efficacy is wide. For context, oral semaglutide at 14 mg daily (the highest approved Rybelsus dose) produced weight loss roughly half that of subcutaneous semaglutide 2.4 mg weekly in comparative analyses, reflecting the bioavailability difference. Any lipid nanoparticle approach that wants to challenge the subcutaneous route would need to demonstrate not just improved cell permeability but sustained plasma concentrations in an animal model, then in humans. The Italian team's work is a methodologically interesting step in that direction, not a clinical result.

What this means for readers following GLP-1 research

Formulation science papers like this one rarely make news headlines, but they represent the slow groundwork beneath future oral GLP-1 drugs. If a microfluidic HIP-SLN system or something similar eventually reaches Phase 1 in humans and shows meaningful bioavailability, it will be because groups like this one worked through the in vitro parameters first. For now, the takeaway is that researchers have a formulation that looks promising in cells and has a scalable production method; the next meaningful data point is an in vivo pharmacokinetic study. This article is educational. It does not constitute medical advice, and semaglutide in any form is a prescription medicine. Discuss any treatment decisions with a qualified healthcare provider.