Why are most peptides injected rather than swallowed?

Last updated: October 2026

Most peptides are injected because the gut is built to digest them. Stomach acid and pepsin, then pancreatic and brush-border enzymes, break peptides into amino acids, and the survivors are too large and water-loving to cross the intestinal lining¹. Oral peptides that rely on permeation enhancers, such as oral semaglutide, reach the blood at about 1% or less, so injection remains the default^3,4.

What happens to a peptide when it is swallowed?

Peptides are short chains of amino acids. A common convention calls chains of 2 to 50 amino acids peptides and longer chains proteins, though the line is blurred: human insulin, with 51 amino acids, is described both ways¹. Because of poor stability and limited permeability in the gut, therapeutic peptides and proteins are usually given by injection¹.

The first obstacle is acid. Gastric fluid in a healthy adult sits at about pH 1.5 to 3.5, rising to about pH 5 to 6 in the duodenum and pH 7 to 8 in the lower small intestine¹. Pepsin, the stomach's protein-cutting enzyme, works best at pH 2 to 3 and splits peptide bonds into smaller fragments¹. In simulated gastric fluid, no insulin could be detected after 30 minutes¹.

The small intestine adds pancreatic enzymes, including trypsin, chymotrypsin, carboxypeptidase and elastase, and peptidases on the brush border then cut what remains into tripeptides, dipeptides and single amino acids¹. In one comparison of 17 peptide drugs, only 3 remained after 30 minutes in human intestinal fluid: cyclosporine (99% intact), desmopressin (25%) and octreotide (22%), and all three are now sold in oral forms¹.

Why can't peptides cross the gut wall?

A peptide that survives digestion still has to get through mucus, a sticky gel secreted by goblet cells that traps foreign material; in humans the thickest layers are in the stomach, at about 180 micrometers, and the colon¹. Beneath the mucus lies a single layer of epithelial cells, about 90% of them absorptive enterocytes, sealed together by tight junctions that make the lining a gatekeeper against large molecules¹.

Peptides are poorly suited to either route across that layer. Going through cells (transcellular) is predicted to be extremely slow, because peptides are water-loving, carry many hydrogen-bonding groups and usually weigh far more than 500 daltons¹. Going between cells (paracellular) means passing water-filled pores of about 3 to 10 angstroms, which molecules above about 500 daltons generally cannot use; even when opened by permeation enhancers, these pores stay under 20 nanometers and cover only 0.01% to 0.1% of the intestinal surface¹.

People also differ. Mucus condition, enzyme secretion and gut motility vary between individuals, and that variability is itself a barrier for oral peptides¹.

Which peptides already work by mouth?

A small group of peptide medicines is approved for oral use, and each gets around the gut in a different way. Cyclosporine, a cyclic peptide with a molecular weight of 1202, was the first oral peptide formulation approved in the United States, in 1990, and an improved formulation followed five years later¹. Linaclotide, a 14-amino-acid peptide, avoids the absorption problem entirely: it acts locally on the inner surface of the intestine and is minimally absorbed, with negligible systemic availability⁸.

Two newer products use permeation enhancers. Oral semaglutide was approved in 2019, and an oral octreotide capsule using a transient permeability enhancer technology followed in 2020¹. Octreotide capsules are enteric-coated and contain sodium caprylate among their inactive ingredients⁷. In the label's single-administration study in healthy subjects, the capsule needed about 200 times more octreotide than the injection to reach similar overall exposure, and food cut absorption by about 90%⁷.

How does oral semaglutide compare with the injection?

Semaglutide is a GLP-1 analog with 94% sequence homology to the human hormone, modified to bind albumin and resist breakdown by the enzyme DPP-4, which gives it a half-life of about one week⁴. The tablet is co-formulated with salcaprozate sodium (SNAC), which helps semaglutide absorb, mostly in the stomach⁴. Its estimated absolute bioavailability is about 0.4% to 1% for the original tablets and about 1% to 2% for a newer tablet formulation on the same label⁴. The weekly injection under the skin has an absolute bioavailability of 89%⁵.

The tablet works only under strict conditions. Its label specifies the tablet on an empty stomach in the morning with no more than 4 ounces of water, followed by at least 30 minutes before food, drinks or other oral medicines⁴. A pharmacokinetic analysis of clinical pharmacology trials estimated oral bioavailability at 0.8% under those recommended conditions, rising with a longer fasting time after the tablet and falling with more water⁶. Within-person variability in bioavailability was 137%, which daily use and the long half-life reduce to about 33% variability in steady-state exposure⁶.

What technologies are used to make peptides work orally?

Researchers attack the barriers one at a time. Strategies include permeation enhancers, inhibitors of gut enzymes, and mucus-penetrating and cell-penetrating peptides². Other approaches include enteric coatings, nanoparticles and intestinal microdevices¹. Enzyme inhibitors work by binding reversibly or irreversibly to the enzymes that would otherwise cut the peptide¹. Oral versions of insulin, calcitonin, parathyroid hormone and vasopressin have all reached clinical trials².

Results so far are modest. Oral semaglutide, octreotide and salmon calcitonin have completed Phase III trials, but oral peptide formulations built on traditional permeation enhancers have low and variable bioavailability of about 1%³. Nanoparticle formats have largely disappointed, with few prototypes reaching clinical trials because of low loading, poor release control, low uptake by intestinal cells and difficult scale-up³. Medicinal chemists are also designing smaller, stable macrocyclic peptides that may be absorbed better than linear ones³.

Why is injection still the default for most peptides?

Because an oral product must lose most of its peptide to work at all. With bioavailability around 1%, only potent peptides with an affordable cost of synthesis are realistic oral candidates³. The best candidates have a wide therapeutic index, some stability in the gut, a long elimination half-life and low clearance³. Semaglutide fits that profile, with its week-long half-life^4,3.

Injection has real costs, too. Long-term injections raise adherence problems, including pain, aversion to needles and local irritation, which is why so much research has gone into alternatives^1,2. For most peptides, though, low membrane permeability, metabolism and rapid clearance still stand in the way of a practical tablet³.