Why Oral Peptide Delivery Is So Hard
Oral Peptide Delivery
<1%
Oral bioavailability of semaglutide even with SNAC absorption enhancement technology, compared to near-complete absorption via injection.
Buckley et al., Clin Diabetes, 2024
Buckley et al., Clin Diabetes, 2024
View as imageNearly every peptide drug on the market requires an injection. Insulin, semaglutide, teriparatide, octreotide: patients inject them because swallowing them would destroy them. The global peptide therapeutics market reached $42.8 billion in 2023, yet fewer than a handful of peptide drugs can be taken by mouth. The reason is not a lack of effort. Pharmaceutical companies have spent decades and billions of dollars trying to solve oral peptide delivery. The problem is that evolution built your gastrointestinal tract specifically to break proteins and peptides into amino acids. Making a peptide survive that process intact and then cross the intestinal wall into the bloodstream remains one of the hardest problems in drug development.
For details on the one oral peptide success story, see oral semaglutide (Rybelsus): how a peptide survives your stomach. This article covers why the problem is so hard in the first place. For a broader view of how the race for oral tirzepatide is pushing these boundaries, see the pillar article.
Key Takeaways
- Oral peptide bioavailability is typically below 1-2%, compared to near-complete absorption from subcutaneous injection, because the GI tract evolved to digest peptides
- Three sequential barriers destroy oral peptides: enzymatic degradation (pepsin, trypsin, chymotrypsin), the mucus layer (physical trapping), and the intestinal epithelium (tight junctions blocking passage)
- SNAC, the absorption enhancer in oral semaglutide, works by locally raising stomach pH to inhibit pepsin and promoting transcellular absorption, but still achieves less than 1% bioavailability
- Peptides larger than 3 kDa face disproportionately worse oral bioavailability because they cannot pass through tight junctions and are more susceptible to enzymatic cleavage
- Orforglipron, a non-peptide small-molecule GLP-1 agonist, sidesteps the entire oral peptide problem by mimicking GLP-1 receptor activation without using a peptide structure
- Emerging technologies including cyclic peptide platforms, inflammation-triggered conjugates, and nanoparticle carriers each address different barriers but no single approach solves all three simultaneously
Barrier 1: Enzymatic Destruction in the Stomach and Intestine
The moment a peptide enters the stomach, it encounters an environment designed to destroy it. Gastric acid (pH 1-3) denatures the three-dimensional structure that gives most peptides their biological activity. Pepsin, the stomach's primary protease, cleaves peptide bonds adjacent to hydrophobic amino acids like phenylalanine, tyrosine, and leucine. Most therapeutic peptides contain these residues.
The destruction continues in the small intestine. Trypsin cleaves at lysine and arginine residues. Chymotrypsin targets aromatic amino acids. Elastase breaks bonds near small, nonpolar residues. Carboxypeptidases trim peptides from their C-terminus. These enzymes work in sequence and in parallel, and they are present at high concentrations because the intestine processes dietary protein constantly.[1]
A 2025 review of oral peptide delivery barriers estimated that enzymatic degradation alone eliminates 95-99% of an orally administered peptide dose before it reaches the intestinal wall.[1] The remaining 1-5% must still cross the mucus layer and epithelium.
Molecular Weight Matters
Peptides with a molecular weight above 3 kDa (roughly 25-30 amino acids) are degraded significantly faster during gastric transit than smaller peptides. This is partly because larger peptides present more cleavage sites to proteases, and partly because their three-dimensional structures are more vulnerable to acid-induced denaturation. Semaglutide, at approximately 4.1 kDa, sits right at this threshold, which is one reason its oral bioavailability is so low even with absorption enhancement technology.
Barrier 2: The Mucus Layer
Even if a peptide survives the enzymatic gauntlet, it must cross a mucus layer that lines the entire GI tract. This gel-like barrier is 100-800 micrometers thick in the stomach and 15-450 micrometers in the small intestine. It is continuously secreted by goblet cells and renewed every 1-4 hours.
The mucus layer traps peptides through two mechanisms. First, its mesh-like glycoprotein network physically blocks molecules above a certain size from diffusing through. Second, electrostatic and hydrophobic interactions between the peptide and mucin glycoproteins cause adhesion, holding peptides in the mucus where they are eventually shed with the layer or degraded by brush-border enzymes embedded in the mucus itself.[2]
This is a barrier that small-molecule drugs largely avoid. Conventional drugs like metformin or ibuprofen are small enough (under 500 Da) to diffuse through mucus readily. Therapeutic peptides, ranging from 500 Da to over 5,000 Da, are too large for easy diffusion and too chemically complex to avoid interaction with mucin.
Barrier 3: The Intestinal Epithelium
The final barrier is the intestinal epithelium itself: a single layer of cells that separates the gut lumen from the bloodstream. These cells are joined by tight junctions, protein complexes that seal the space between adjacent cells. The tight junctions allow water and small ions to pass but block molecules larger than approximately 600 Da.
This means peptides cannot take the paracellular route (between cells) unless the tight junctions are opened pharmacologically. The transcellular route (through cells) requires the peptide to cross two lipid bilayer membranes: the apical membrane facing the gut lumen and the basolateral membrane facing the blood side. Peptides, being hydrophilic and charged at physiological pH, do not cross lipid membranes efficiently.[1]
A 2025 study published in Nature Communications demonstrated how permeation enhancers work at this barrier: they create transient, recoverable defects in cell membranes that allow peptides to pass through the transcellular route.[3] The membrane defects heal within minutes after the enhancer concentration drops, limiting damage to the epithelial layer. This work provided the first direct visualization of the mechanism that SNAC and similar enhancers use to enable oral peptide absorption.
Barrier 4: First-Pass Metabolism
Even after a peptide crosses the intestinal wall, it enters the portal vein and passes through the liver before reaching systemic circulation. The liver contains peptidases that degrade circulating peptides, further reducing the amount that reaches target tissues. This hepatic first-pass effect is a barrier that injectable peptides bypass entirely by entering the bloodstream directly through subcutaneous tissue.
The combined effect of all four barriers: out of every 100 molecules of a peptide that a patient swallows, typically fewer than 1-2 reach the bloodstream intact.
How Oral Semaglutide Works (and Why It Is Still Limited)
Oral semaglutide (Rybelsus) is the most commercially successful oral peptide to date. It uses SNAC (sodium N-[8-(2-hydroxybenzoyl)amino] caprylate) as a co-formulated absorption enhancer. SNAC addresses two of the four barriers simultaneously:
- Local pH buffering: As the tablet erodes in the stomach, SNAC creates a local increase in pH around the dissolving semaglutide. This reduces pepsinogen-to-pepsin conversion, protecting semaglutide from enzymatic cleavage.
- Transcellular absorption enhancement: SNAC promotes the monomerization of semaglutide (preventing it from clumping into oligomers) and facilitates its passage across the gastric epithelium through the transcellular route.
Niu et al. (2025) tested combining SNAC with C10 (sodium caprate), another permeation enhancer, in a single tablet formulation. The combination showed improved peptide absorption compared to SNAC alone in preclinical models, suggesting that multi-enhancer approaches may push bioavailability higher.[4]
Even with SNAC, oral semaglutide achieves less than 1% bioavailability. This means the oral dose (14 mg for diabetes, 50 mg for obesity) must be vastly larger than the injected dose (1 mg for diabetes, 2.4 mg for obesity) to achieve equivalent blood levels. The waste is enormous, but the convenience of a pill over an injection drives patient demand. For a deep dive on this technology, see the SNAC technology behind oral peptide absorption.
The Orforglipron Approach: Skipping the Problem Entirely
One way to solve the oral peptide delivery problem is to stop using peptides. Orforglipron, developed by Eli Lilly, is a non-peptide small-molecule GLP-1 receptor agonist. At approximately 500 Da, it is small enough to absorb orally like a conventional drug, passing through the mucus layer and crossing the intestinal epithelium without any absorption enhancer.
Phase 3 trials (ATTAIN program) showed that orforglipron produced up to 14.7% weight loss at 72 weeks, approaching (though not matching) injectable tirzepatide's results.[5][6] Orforglipron does not require fasting before taking it (oral semaglutide requires a 30-minute fast with only a sip of water), removing another practical limitation.
The tradeoff: small molecules that mimic peptide receptor activation are much harder to design. GLP-1 is a 30-amino-acid peptide that contacts its receptor across a large binding surface. Finding a small molecule that can activate the same receptor with similar potency required years of medicinal chemistry and computational design. This approach is not generalizable to all peptide targets, especially those with large, flat binding interfaces. For more on this competitive landscape, see Danuglipron: Pfizer's oral GLP-1 and its rocky development path.
Emerging Technologies: What the Pipeline Looks Like
Cyclic Peptide Platforms
Chikamatsu et al. (2026) developed a cyclic peptide-based technology that enables efficient oral delivery of peptide cargo through the small intestine.[7] Cyclic peptides resist enzymatic degradation because their circular backbone lacks the free termini that exo-peptidases require as substrates. The cyclization also constrains the peptide's shape, reducing the entropy penalty for membrane crossing. This platform is designed as a modular carrier: the cyclic peptide shuttle can be conjugated to different therapeutic peptide payloads.
Inflammation-Triggered Conjugates
Cheng et al. (2026) engineered self-immolative conjugates that protect peptides during GI transit and release them in response to inflammatory conditions in the gut.[8] This approach is particularly relevant for peptide delivery to inflamed intestinal tissue (as in inflammatory bowel disease), where the drug needs to reach the site of pathology rather than systemic circulation.
Nanoparticle Carriers
Qian et al. (2026) used ferritin protein nanocages to encapsulate GLP-1 peptides for oral delivery, protecting the cargo from gastric degradation while facilitating intestinal uptake through receptor-mediated endocytosis.[9] Similarly, Liang et al. (2025) developed bacterial spore-based nanoparticle generators for oral insulin delivery, using the spore's natural resistance to gastric acid as a protective shell.[10]
The challenge with all nanoparticle approaches: manufacturing complexity and cost. Building reproducible nanoparticles at pharmaceutical scale is orders of magnitude harder than making a simple tablet.
For a broader view of what oral peptide drugs are currently in development, see the future of oral peptide drugs: what's in the pipeline.
Why This Problem Persists
The oral peptide delivery problem has been recognized since the discovery of insulin in 1921. A century later, it remains largely unsolved. The fundamental issue is that biology optimized the GI tract to do exactly what drug developers do not want: disassemble peptides into their component amino acids for nutrition.
Every solution to one barrier creates a new constraint. Enzyme inhibitors protect the peptide but may impair normal protein digestion. Permeation enhancers open tight junctions but risk allowing pathogens and toxins to enter. Nanoparticle encapsulation protects the cargo but limits the dose that can be loaded per carrier. SNAC achieves less than 1% bioavailability and requires strict fasting protocols. No single technology addresses all four barriers simultaneously.
The result: most peptide drugs will continue to be injected for the foreseeable future. The rare exceptions (oral semaglutide, orforglipron) required either massive dose compensation or an entirely different molecular approach. For the hundreds of peptide drugs in clinical trials, injection remains the default because it works, even if patients wish it did not.
The Bottom Line
Oral peptide delivery remains one of pharma's hardest problems because the GI tract presents four sequential barriers: enzymatic degradation (eliminating 95-99% of the dose), the mucus layer (trapping peptides by size and charge), the intestinal epithelium (tight junctions blocking paracellular transport), and hepatic first-pass metabolism. SNAC-enabled oral semaglutide achieves less than 1% bioavailability despite being the most successful oral peptide product. Emerging approaches including cyclic peptide platforms, nanoparticle carriers, and non-peptide small-molecule agonists each address different pieces of the problem, but no technology solves all barriers at once.