David Brayden of University College Dublin opened by recapping the field’s milestones: the approvals of oral semaglutide (Rybelsus, and the higher-dose Wegovy pill) and oral octreotide (Mycapssa), plus the clinical progress of macrocyclic peptides such as Merck’s MK-0616 (enlicitide decanoate), whose New Drug Application was submitted in 2026, and Johnson & Johnson’s JNJ-2113 (icotrokinra), approved in 2026 for psoriasis. He also flagged cautionary cases, including Chugai’s LUNA18, a Phase I oral RAS inhibitor that was terminated. Brayden raised a subtle question: when weekly subcutaneous injections of the same molecule exist, as with GLP-1 analogues, what exactly does the oral route offer? Patient preference surveys favour oral dosing, but strict dosing requirements—fasting, specific fluid volumes, waiting times before food—can erode that convenience. His conservative conclusion was that oral formulations offer an alternative route for some patients while others stay on injectables, yet oral administration remains the favoured option for most patients with chronic conditions.

A recurring theme was the poor predictive power of preclinical models. Aqueous enhancer mixtures perform differently in rat gut instillations than solid tablets do in dogs or macaques, and cell-based systems cannot tolerate the high excipient concentrations that reach the intestine in vivo. Brayden defended rat instillation models as informative despite their limits, offering clues for large-animal work. He then reviewed mechanistic thinking on the two gold-standard medium-chain fatty acid enhancers: salcaprozate sodium (SNAC), which holds generally-regarded-as-safe status, and sodium caprate (C10), which has phase II and III trial history. Molecular dynamics simulations now model how these agents interact with plasma membranes amid mixed micelles and bile salts, and cryo-electron microscopy has revealed that C10 forms micelles and vesicles whose structures shift with the pH of different gut regions. Despite these advances, Brayden argued that our picture of the enhancer–peptide relationship at the epithelium—and the path the peptide takes across it—remains rudimentary. He also highlighted data on solubilising excipients such as Labrasol, Labrafac, and Capryol-90, whose medium-chain fatty acid glycerides achieved insulin blood levels in rats equivalent to C10, attractive because of their established safety profiles.

Tahnee Dening of Genentech presented mechanistic work on how macrocyclic peptides interact with bile salts and biorelevant colloids, using nuclear magnetic resonance spectroscopy and artificial membranes. Focusing on octreotide, a low-permeability, high-solubility peptide, she showed that bile salts strongly interact with aromatic and lysine residues, forming colloids, micelles, and mixed micelles depending on critical micellar concentrations in fed-state and fasted-state simulated intestinal fluids. Dihydroxy bile salts impede peptide flux across membranes more than trihydroxy versions, yet offer better protection against peptidases. NMR revealed that complexation occurs more readily for octreotide than for desmopressin, which lacks the relevant residues. The counterintuitive implication: bile salts are surfactants with low CMC values, but whether they enhance absorption is complicated by their binding potential with some peptides—contrary to what surfactant-type enhancer logic would predict.

Stephen Buckley of Novo Nordisk reviewed barriers to oral macromolecule delivery and made predictions about future candidates. He emphasised that oral administration offers advantages for chronic indications and potential pharmacological benefits by routing drugs to the liver via the portal vein rather than flooding peripheral circulation. The Rybelsus story, he argued, was the decisive proof of concept: improving the potency and reducing the clearance of semaglutide, then matching it with SNAC, produced an oral product achieving similar pharmacologic endpoints in type 2 diabetes as the injectable, despite low bioavailability. Buckley described efforts to develop oral amycretin, a dual GLP-1 and amylin receptor agonist that showed weight loss and acceptable safety in trials with weekly subcutaneous doses up to 50 mg. Since the workshop, the FDA approved a 25 mg oral semaglutide for obesity (Wegovy Oral) and a second-generation Rybelsus with lower doses for diabetes. Buckley also reviewed bioinspired device attempts to achieve double-digit bioavailability, noting uncertain regulatory paths and questions about patient acceptability.

Ricardo Diaz de Leon Ortega of Quotient Sciences showed how physiologically based pharmacokinetic (PBPK) modelling integrates compound properties, in vitro metabolism data, and formulation dissolution with physiological parameters such as tissue volumes and blood flow. In one retrospective case with a poorly permeable Bristol Myers Squibb peptide, a Weibull model fitted dissolution data from enteric-coated tablets containing 300 or 600 mg SNAC, with dog and human PK data as inputs. Key conclusions: functional excipients like enhancers must be included in dissolution profiles and plasma measurements, and even biorelevant dissolution approximates in vivo conditions sub-optimally, weakening the model. Regarding SNAC’s mechanism, the data indicated a complex process with channels or pores created through plasma membrane fluidisation, mimicking aspects of paracellular flux. In a second case, modelling explored interactions of oral semaglutide with ten co-administered molecules including metformin and furosemide. Because semaglutide slows gastric transit and its formulation contains SNAC, both features could affect co-administered drug PK. Predicted bioavailability increases were 33 to 43 percent for metformin and 49 to 59 percent for furosemide, with clinical importance depending on each victim molecule’s therapeutic index.

Thomas Von Erlach of Vivtex presented GI-ORIS technology, a robotic system mounting porcine small intestinal mucosae in mini-Ussing chambers to screen multiple enhancers and combinations with selected peptides. AI can identify enhancer properties for particular peptides, and screening can evaluate liquid formulations in rat models before solid-dose testing in large animals. A case study for oral vancomycin produced a solid-dose formulation after five iterations that delivered more peptide than a C10-based formulation in dogs, and a proprietary immediate-release tablet outperformed Rybelsus in a dog PK study. But success in large animals does not guarantee success in humans: Novo Nordisk researchers showed a PCSK9 inhibitor peptide delivered successfully from a C10-SNAC tablet in dogs but failed in humans, highlighting gaps in knowledge about optimal in vivo release profiles and species-specific concentrations of peptide and enhancer across gut regions. Von Erlach confirmed that LUNA18, the orally bioavailable RAS inhibitor macrocycle, showed high basal permeability in the ex vivo assay, as predicted from its structure—suggesting custom formulations can handle macrocycles where permeability is less of an issue.

Randall Mrsny of the University of Bath described exploiting endogenous epithelial pathways for paracellular uptake and vesicular transcytosis. His lab’s permeant inhibitor of phosphatase (PIP) peptides open tight junctions by shifting the myosin light-chain kinase and phosphatase activation balance; rational design with amino acid replacement and d-versions produced lead candidates with good stability, enhancing capacity, and low toxicity. His second approach mimics Cholix, a truncated toxin from Vibrio cholerae that uses transcytosis to cross GI epithelia. Macromolecules conjugated to the truncated protein can transcytose the epithelium as nature intended. A fused Cholix–human interleukin-10 conjugate (AMT-101) delivered in an enteric tablet showed acceptable safety on repeat administration and indications of benefit in pouchitis patients in a recent phase II trial. By including the IgG1 CH2 element, a new Cholix–human growth hormone complex achieved 5 percent bioavailability in rats by promoting exocytosis across basolateral membranes.

Douglas Johns of Merck detailed the development of MK-0616, the oral macrocycle PCSK9 inhibitor for hypercholesterolaemia. With the LDL-C–PCSK9 receptor complex structure in hand, the team surmised small molecules might lack sufficient selectivity, whereas a cyclic peptide could achieve high affinity. mRNA display screened libraries of 10^14 peptides with unnatural amino acids, and structure-based design produced MK-0616 with picomolar potency, stability, and solubility—but poor permeability. Orally administered to cynomolgus monkeys in tablets with a permeation enhancer, sufficient bioavailability reduced plasma LDL-C, enabling clinical development. Merck chose C10 over Labrasol-ALF as enhancer. Phase II data showed LDL-C reductions at week 8 with daily tablets of 20–30 mg enlicitide plus 180 mg C10, and phase III CORALreef data confirmed lower LDL-cholesterol than placebo at 24 weeks. If approved, it will be the second oral peptide product containing an enhancer other than SNAC, since C8 was a component of Mycapssa’s oily suspension in 2020.

Takamitsu Ueto of Chugai described the LUNA18 programme, which reached phase I before termination. Chugai used mRNA display with in silico design to produce high-affinity RAS binders with 9–11 residues, amide cyclization, cLog P around 12, and no polar amino acids. LUNA18 (paluratide, 1438 Da) is lipophilic with good passive permeability across Caco-2 monolayers, achieving oral bioavailability over 21 percent in four animal species without enhancers—including 26 percent in monkeys and 47 percent in dogs. But solving permeability created a solubility problem: LUNA18 crystals dissolve poorly in water. Following the cyclosporine (Neoral) playbook, Chugai turned to self-emulsifying drug delivery systems. SEDDS number 85, comprising oleic acid, propylene glycol monocaprylate, Kolliphor EL, and dl-alpha-tocopherol, achieved 60 mg/mL solubility, uniform dispersion in five minutes in fasted-state simulated intestinal fluid, and maintained solubility after in vitro digestion. A PK study in rats with LUNA18 in a size-9 capsule produced high AUC values comparable to an oleic acid solution, and the formulation advanced to phase I in monkeys.

Karsten Lindhardt of Biograil presented the BIONDD device, a size-00 capsule that attaches to the stomach wall and activates a flip-over spike temporarily inserting into the stomach lining. Proof of concept in dogs: 2.5 mg of a model peptide entrapped in a fully dissolvable spike achieved plasma exposure and variability similar to subcutaneous administration. Endoscopy showed attachment at about five minutes and detachment at 24 hours in all canine subjects, with all devices accounted for at three days of GI transit. The system is also being evaluated with adalimumab, a greater challenge due to its much higher molecular weight. With a target dose of 150 mg per month, spikes entrapping 5.1 mg for daily dosing achieved reliable insertions in isolated porcine stomach mucosae, and in dogs the 2.5 and 5 mg oral doses produced PK profiles similar to subcutaneous adalimumab with relative bioavailability above 70 percent, with all devices actuating. Pegfilgrastim and mRNA in lipid nanoparticles are also being assessed, and injection moulding manufacturing has begun for scale-up toward GLP safety and first-in-human studies.

The roundtable, facilitated by Brayden with panellists from Halozyme, Novo Nordisk, Protagonist, Bristol Myers Squibb, and Bath, distilled practical guidance. For starting a new oral peptide programme: advantageous medicinal chemistry, molecular weight below 5000 Da, high potency, good stability, long half-life, and a clear PK–PD relationship. On bioassay screens, the consensus was that no single assay offers predictive capacity, but combining several provides direction—while noting that excipient–peptide interactions in tablets and release rates from dosage forms are not evaluated by cell and tissue bioassays or rodent studies. On animal models, panellists favoured dogs over pigs (exceptionally long gastroretentive times for solid dosage forms) and macaques (expensive, limited availability, ethical considerations), though macaques were useful in Merck’s MK-0616 programme before phase I. On whether enhancer performance has hit a ceiling, the view was that no one size fits all, and new agents would need substantial absorption benefits over SNAC and C10 to justify regulatory risks and additional toxicology—though new enhancers are being discovered, many repurposed from existing excipient solubilisers and the food industry, and combinations using different mechanisms are being examined. On nanotechnology, with a history of overpromise, panellists felt it must outperform standard enhancers, offer simple scale-up, and rely on GRAS materials, though particle synthesis and targeting science has improved considerably since earlier efforts.

The workshop’s overall message was that oral peptide delivery has moved from aspiration to commercial reality, but the science remains hard. Permeation enhancers work, yet their mechanisms are incompletely understood. Macrocyclic peptides can be engineered for permeability, but solubility then becomes the bottleneck. Devices can bypass enhancers entirely, but regulatory and acceptability questions remain. Preclinical models provide direction but not prediction. As more oral peptide products reach the market and more programmes advance through clinical trials, the field’s central challenge is shifting from whether oral peptides are possible to how to make them reliable, affordable, and acceptable to patients at scale.

Subject of Research: The fourth controlled release society workshop on oral peptide administration

Article Title: The fourth controlled release society workshop on oral peptide administration

Article References: Brayden, D. J., Buckley, S. T., & Lewis, A. L. (2026). The fourth controlled release society workshop on oral peptide administration. Frontiers in Drug Delivery, 6, Article 1860678. https://doi.org/10.3389/fddev.2026.1860678

Image Credits: AI Generated

DOI: 10.3389/fddev.2026.1860678

Keywords: fourth, controlled, release, society, workshop, oral, peptide, administration, scientific research