Secretin
Human secretin, Synthetic human secretin, Porcine secretin, SCT
Secretin was the first hormone ever discovered, identified by Bayliss and Starling in 1902, and it is the molecule that gave the word 'hormone' its meaning. It is a 27-amino-acid gut peptide that makes the pancreas secrete bicarbonate-rich fluid. Today it is used only as a diagnostic agent, and it is best known outside medicine for a spectacular sequence of failed autism trials.
Mechanism
Secretin is released from enteroendocrine S cells in the duodenal and upper jejunal mucosa when acidic gastric chyme lowers luminal pH. It binds the secretin receptor, a class B G-protein-coupled receptor closely related to the receptors for glucagon, vasoactive intestinal peptide and glucagon-like peptide-1, which couples through Gs to activate adenylyl cyclase and raise intracellular cyclic AMP. In pancreatic duct epithelium, the resulting protein kinase A activity phosphorylates and opens the cystic fibrosis transmembrane conductance regulator chloride channel, driving chloride efflux which is coupled through anion exchangers to bicarbonate secretion, with water following osmotically. The output is a large volume of alkaline pancreatic juice that neutralises duodenal acid and provides the pH at which pancreatic enzymes work. This CFTR dependence is why secretin-stimulated testing is informative in cystic fibrosis and CFTR-related disease.
Secretin has a second set of actions that underpin its diagnostic uses. It stimulates biliary bicarbonate secretion from cholangiocytes, potentiates cholecystokinin-driven enzyme secretion from acinar cells, inhibits gastrin release and gastric acid output, and relaxes the sphincter of Oddi while transiently increasing pancreatic ductal flow, which is what makes the ampulla of Vater easier to identify at endoscopy and what distends the pancreatic duct during secretin-enhanced magnetic resonance cholangiopancreatography. In gastrinoma it produces a paradoxical response: normal gastric G cells are inhibited by secretin, but gastrinoma cells respond with a sharp rise in gastrin, and this paradox is the basis of the secretin stimulation test for Zollinger-Ellison syndrome. Secretin receptors are also present in the central nervous system, which is the slender biological thread from which the autism hypothesis was hung.
What the research shows
Secretin's diagnostic value is real, though unevenly quantified. Secretin-enhanced magnetic resonance cholangiopancreatography is the best-supported use: pooled sensitivity 86% and specificity 97% for pancreas divisum across 10 studies and 1,474 patients, against 52% sensitivity for unenhanced imaging. Endoscopic secretin-stimulated pancreatic function testing, measuring peak duodenal bicarbonate, is conceptually sound and can separate advanced chronic pancreatitis from abdominal pain without it, but the foundational endoscopic study enrolled only 18 patients and the evidence base here is thinner than the technique's routine description suggests. The secretin stimulation test for gastrinoma exploits a genuine paradoxical physiology and remains part of the diagnostic algorithm for Zollinger-Ellison syndrome, though proton pump inhibitor use complicates interpretation and must be accounted for.
The autism episode deserves separate treatment because it is one of the most instructive negative results in modern paediatrics. A 1998 report of three children whose autistic symptoms apparently improved after diagnostic secretin during endoscopy, amplified by television coverage, led thousands of families in the United States to pursue secretin infusions, sometimes at considerable cost. The response of the research community was exemplary: the Sandler trial in 1999 found nothing, and over the following decade a series of randomised placebo-controlled trials replicated that null result, culminating in a Cochrane review of 16 studies recommending against its use entirely. One detail from the Sandler trial is worth stating precisely, because it is often garbled: the placebo group actually improved more on the primary behavioural scale than the secretin group did, and after being told the results 69% of parents in the study said they remained interested in secretin anyway. That is a measure of how tenaciously expectation survives contrary evidence, which is a slightly different and more interesting point than the one usually made. Any current marketing of secretin for autism is contradicted by an unusually complete body of evidence.
Evidence assessment
High-quality evidence
Synthetic human secretin holds FDA-approved labelling for three diagnostic indications, and the best-quantified of its uses, secretin-enhanced magnetic resonance cholangiopancreatography, is supported by a meta-analysis of 10 studies covering 1,474 patients. The tier applies to those diagnostic uses. One correction from the draft: the evidence for endoscopic secretin-stimulated pancreatic function testing is thinner than implied, resting substantially on an 18-patient prospective study. The tier emphatically does not apply to autism, where the evidence is strong in the opposite direction: a Cochrane review of randomised trials found no benefit, and that null finding is itself high-quality evidence.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Lack of benefit of a single dose of synthetic human secretin in the treatment of autism and pervasive developmental disorder Preclinical only
No benefit on any outcome measure. Mean decreases in the Autism Behaviour Checklist total score over four weeks were 8.9 in the secretin group and 17.8 in the placebo group (mean difference -8.9, 95% CI -19.4 to 1.6, P=0.11), so the placebo group improved more, not less. No child had treatment-limiting adverse effects. Corrected during audit: the draft claimed that 'roughly 70% of parents in both arms, including the placebo arm, wanted continued treatment'. What the paper actually reports is that after being told the results, 69% of the parents of children in the study said they remained interested in secretin as a treatment. That is a single figure for the whole cohort, not a per-arm comparison, and it describes continuing interest despite a null result rather than a perceived response.
Intravenous secretin for autism spectrum disorders (ASD) Preclinical only
Definitively negative. No evidence that single or multiple doses of intravenous secretin improve social interaction, communication or restricted and repetitive behaviours. The reviewers concluded that secretin should not be recommended or administered as a treatment for autism spectrum disorder, and that further trials would be justified only if new high-quality evidence established a plausible mechanism or identified a defined subgroup. Rarely does a systematic review reach a conclusion this unambiguous.
An endoscopic pancreatic function test with synthetic porcine secretin for the evaluation of chronic abdominal pain and suspected chronic pancreatitis Preclinical only
Median peak duodenal bicarbonate concentrations were 87, 72 and 35 meq/l across the three groups respectively (P=0.010), and the test was safe and well tolerated. Corrected during audit: the draft described this as a validation study that 'established' the endoscopic test and 'defined the bicarbonate thresholds used in subsequent practice'. With 18 patients and 5 to 7 per group, that is a substantial overstatement. It is a feasibility study showing the endoscopic approach can separate advanced chronic pancreatitis from pain without it.
Safety
Intravenous secretin is well tolerated in the diagnostic setting. The most common effects are transient: nausea, abdominal discomfort, flushing and a brief sensation of warmth. In the 60-child autism trial, no participant had a treatment-limiting adverse effect. Hypersensitivity reactions can occur, as with any injected peptide, and the earlier porcine-derived product carried a greater theoretical immunogenic risk than the current synthetic human sequence, which was one reason for the switch. Because secretin transiently increases pancreatic ductal secretion and pressure, there is a small theoretical risk of provoking pancreatitis, and it should not be given during an episode of acute pancreatitis.
Secretin is inactive by mouth. It is a 27-residue peptide degraded by gastric and pancreatic proteases before absorption, so oral secretin products have no plausible pharmacological effect regardless of what is claimed for them. The peptide is not licensed in the United Kingdom and is used only under specialist hospital supervision through unlicensed import routes. Given the definitive negative evidence in autism, any product offered for that purpose should be regarded as unsupported by the literature; parents encountering such offers may wish to discuss them with a paediatrician.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | Not licensed by the MHRA. Secretin is obtained as an unlicensed imported medicinal product for use in specialist gastroenterology and radiology centres, principally for secretin-enhanced magnetic resonance cholangiopancreatography and, less often, for pancreatic function testing. Not available outside hospital practice. |
| United States | Synthetic human secretin was approved by the FDA in 2004 for stimulation of pancreatic secretions to aid in the diagnosis of exocrine pancreatic dysfunction, for stimulation of gastrin secretion to aid in the diagnosis of gastrinoma, and for stimulation of pancreatic secretions to facilitate identification of the ampulla of Vater and accessory papilla during endoscopic retrograde cholangiopancreatography. A synthetic porcine product was approved in 2002 and subsequently discontinued. Prescription-only, for intravenous hospital use. |
| WADA (sport) | Not prohibited. Secretin does not appear on any section of the current Prohibited List and has no plausible ergogenic mechanism; it is an intravenous diagnostic agent affecting pancreatic and biliary bicarbonate secretion. |
Questions
Yes. William Bayliss and Ernest Starling demonstrated in 1902 that acid placed in a denervated loop of dog jejunum still triggered pancreatic secretion, proving that a blood-borne chemical messenger, not a nerve reflex, carried the signal. Starling coined the word 'hormone' for this class of messenger a few years later. Secretin is therefore the founding molecule of endocrinology.
No. This is one of the most thoroughly settled questions in paediatric therapeutics. Following a widely publicised 1998 report of three children, a series of randomised placebo-controlled trials tested intravenous secretin, and a Cochrane review of 16 studies concluded it should not be recommended or administered for autism spectrum disorder. In the first major trial the placebo group actually improved more than the secretin group on the main behavioural measure, and 69% of parents said afterwards that they remained interested in secretin anyway.
No. It is a 27-amino-acid peptide with no oral bioavailability, degraded by gastric and pancreatic proteases before it could reach the circulation. Every licensed and validated use of secretin is by intravenous injection in a hospital setting. Oral secretin products cannot deliver the hormone to its receptors.
Diagnosis, not treatment. Its approved uses are stimulating pancreatic secretion to assess exocrine pancreatic function, stimulating gastrin to help diagnose gastrinoma in Zollinger-Ellison syndrome, and helping identify the ampulla of Vater during endoscopic retrograde cholangiopancreatography. It is also widely used off-label to enhance magnetic resonance cholangiopancreatography, where the evidence for improved ductal visualisation is actually the strongest in its whole file.