
Larazotide Side Effects: What the Trials Actually Measured
Written by Marcus Reinhardt · Lead Analyst, Peptide Insider · LinkedIn
Evidence-first peptide coverage — every claim on this page is linked to its primary source and reviewed against our editorial policy
Citations verified against PubMed · Published: 2026-09-15 · Last updated: 2026-09-15
Larazotide side effects are, unusually for a research-market peptide, something you can look up: five published randomized trials — four in celiac disease, one in children with a post-COVID inflammatory syndrome — recorded adverse events against placebo, and none of them reported a higher adverse-event rate on larazotide than on placebo — two state comparability outright, one reports no serious events, and the children’s trial reports no events attributed to the drug. The catch is that all five tested a controlled pharmaceutical product in people with a diagnosed condition, and none tested the compounded capsules or research vials most people asking this question are holding. This page lays out what the trials measured, where the dose-response ran backwards, why the Phase 3 stopped, and where the data ends.
Quick Answer: The Most Common Larazotide Side Effects
Across the published trials, no abstract reports a higher adverse-event rate on larazotide than on placebo. The 86-person dose-ranging study named headache and urinary tract infection as the most common events and recorded no serious adverse events (PMID 22825365). The 184-person gluten-challenge trial found adverse-event rates “similar between larazotide acetate and placebo groups” (PMID 23163616). The 342-person trial reported safety “comparable with placebo” — and its 0.5 mg arm actually reduced headache and tiredness compared with placebo (PMID 25683116). A 2022 meta-analysis of four trials and 626 patients found larazotide reduced gluten-related diarrhea in patients undergoing gluten challenge, with other adverse events comparable between groups (PMID 34339872). In 12 children with MIS-C, “no larazotide-related adverse events were reported” over 24 weeks of follow-up (PMID 40737433). What no trial measured: anything about compounded capsules, research vials, “leaky gut” without a diagnosis, or use beyond 12 weeks.
The people who have taken larazotide under monitoring were adults with diagnosed celiac disease on a gluten-free diet — in the largest trial, for 12 months or longer and still symptomatic — plus smaller groups deliberately fed 2.4 to 2.7 grams of gluten a day for two to six weeks to provoke a reaction, and 12 hospitalized children. Their arc, as the abstracts describe it: symptoms tracked the gluten, not the drug. In the 342-person trial the lowest dose cut symptomatic days by 26% and increased improved-symptom days by 31%, while the two higher doses did nothing measurable — and the 0.5 mg arm reported less headache and tiredness than placebo, not more.
What those participants had that a capsule buyer does not: a verified diagnosis, a product made to pharmaceutical standards, a fixed schedule, and a study nurse recording every complaint against a placebo group. The trials tell you the molecule is tolerable in that setting. They cannot tell you what an unverified capsule does in a gut that was never examined.
Source: PMIDs 25683116, 23163616, 22825365. When we publish an individual account here, it will be a real person, named with permission. We don’t write fiction into this slot.

Looking for Lab-Tested Peptides?
Neither vendor we review stocks larazotide (checked September 15, 2026). Our top-rated vendor publishes batch COAs and third-party testing on every compound it does carry — get up to 30% off your first order.
Affiliate link — it doesn’t change what we report. All compounds sold strictly for non-human research use. Read our full Amino Club review.
In This Guide
- What Is Larazotide?
- How It Works — and Why “Non-Systemic” Shapes the Side Effects
- The Trial Numbers That Matter
- The Real Rates: What Was Reported Against Placebo
- Headache and Urinary Tract Infection: The Two Named Events
- The Gut Itself: Diarrhea, Symptom Scores and a Backwards Dose Curve
- Serious Adverse Events — and Why the Phase 3 Really Stopped
- When Side Effects Happen — and What the Trials Can Say About Fading
- What 12 Weeks Can and Can’t Tell You
- Who the Trials Enrolled — and Who They Didn’t
- Dose and Side Effects: The Correlation Runs Backwards
- What People Taking Larazotide Report
- Stacked With BPC-157 and KPV: What the Data Covers
- How Larazotide Compares to BPC-157 and KPV on Safety Data
- Where the Trial Data Stops: Compounded Capsules and Research Vials
- Sourcing and Verification
- US Regulatory Status
- Frequently Asked Questions
- Glossary of Terms
What Is Larazotide?
Larazotide acetate (development codes AT-1001 and INN-202) is a single-chain peptide of eight amino acids that regulates the tight junctions between intestinal cells (PMID 33881350; the code list is catalogued in PMID 33397225). Its structure is derived from a protein secreted by Vibrio cholerae, the cholera bacterium (PMID 17697209); a peptide derived from that bacterium’s zonula occludens toxin opens tight junctions in cell models, and larazotide blocks that opening (PMID 22401908). It is taken by mouth, and the Massachusetts General Hospital trial registry describes it as “a locally acting, non-systemic, octapeptide inhibitor of the zonulin receptor” (NCT05747534).
The drug spent nearly two decades in development for celiac disease. Innovate Biopharmaceuticals licensed it from Alba Therapeutics in 2016 and in August 2019 announced the first patient dosed in what it described as the first Phase 3 trial for celiac disease; 9 Meters Biopharma ran that trial to its end. The company stated in 2019 that larazotide held FDA Fast Track designation for celiac disease and had been “exposed to nearly 600 subjects in clinical trials” (company statement, August 2019). The Phase 3 was stopped in June 2022 — for futility, not safety, which matters and is covered below. Since then the molecule has led a second life in two places: academic trials in post-COVID conditions at Mass General, and the compounding-pharmacy and research-vial market, where “leaky gut” buyers are the main audience. Our peptides for leaky gut page covers that framing; this page covers what happens to the body.
How It Works — and Why “Non-Systemic” Shapes the Side Effects
Zonulin is a protein that loosens the tight junctions sealing the spaces between intestinal cells, letting larger molecules through; tight-junction opening is believed to be part of how gluten fragments reach the immune tissue beneath (PMID 22825365). Larazotide is described as a zonulin antagonist: in cell work it blocked the rearrangement of the junction protein ZO-1 and the actin skeleton provoked by gluten fragments and by a junction-opening toxin fragment, prevented that fragment from lowering transepithelial electrical resistance, blocked a gluten 13-mer from crossing a cell layer, and blunted the permeability increase caused by cytokines; in gluten-sensitized mice it preserved normal junction structure (PMID 22401908). A 2021 review adds a second proposed mechanism: inhibition of myosin light chain kinase, which would ease tension on the actin filaments that pull junctions open (PMID 33881350).
The design point that governs its side-effect profile is that larazotide is meant to act in the gut lumen and stay there. The developer described it as “not systemically absorbed” (company statement); the trial registry calls it “locally acting, non-systemic.” A drug that does not reach the bloodstream in meaningful amounts has no mechanism for the systemic effects that dominate the side-effect pages of injected metabolic peptides such as retatrutide. What it can do is change what happens in the gut — which is why the trial data below is almost entirely about gut symptoms, and why the interesting finding is that most of them moved in the right direction.
KEY TAKEAWAY
Larazotide’s trial side-effect profile is placebo-shaped because the molecule was built to stay in the gut and act on a single junction-regulating pathway. That profile belongs to the pharmaceutical product tested in celiac patients — not to a capsule of unknown origin, and not to a condition the trials never studied.
The Trial Numbers That Matter
Seven human studies define what is known. Five have published results; two have not posted any — the Phase 3, terminated at an interim analysis, and the Long COVID trial, completed in June 2026 — as of September 15, 2026. The regimens below are what the trials administered, reproduced for reference; none of them is a recommendation, and none was studied outside a monitored trial.
| Trial | Who | What was administered | What the record says about safety |
|---|---|---|---|
| Proof of concept, 2007 (PMID 17697209) | Coeliac patients, in-patient (7 placebo and 14 AT-1001 in the cytokine analysis) | Single 12 mg doses with acute gluten exposure | “No increase in adverse events” vs placebo; GI symptoms more frequent on placebo (P = 0.018) |
| Dose-ranging, 2012 (PMID 22825365) | 86 celiac patients on a gluten-free diet | 0.25, 1, 4 or 8 mg three times daily, 14 days, with or without 2.4 g/day gluten | “Generally well tolerated. No serious adverse events were observed.” Most common: headache, urinary tract infection |
| Gluten challenge, 2013 (PMID 23163616) | 184 coeliac patients | 1, 4 or 8 mg three times daily, 6 weeks, 2.7 g/day gluten | “Adverse event rates were similar between larazotide acetate and placebo groups” |
| Persistent symptoms, 2015 (PMID 25683116) | 342 celiac adults, gluten-free ≥12 months | 0.5, 1 or 2 mg three times daily, 12 weeks (4-week placebo run-in and run-out) | “Safety was comparable with placebo.” 0.5 mg reduced headache and tiredness (P = .010) |
| Phase 3 CedLara, 2019–2022 (NCT03569007) | 307 celiac adults enrolled (525 planned) | 0.25 or 0.5 mg vs placebo, 12-week double-blind phase | Terminated by sponsor at interim analysis for futility; no results posted; no safety finding cited |
| MIS-C phase 2a, 2025 (PMID 40737433) | 12 children, median age 5.7 | Four times daily for 3 weeks, alongside standard treatment | “No larazotide-related adverse events were reported” through 24 weeks |
| Long COVID phase 2a, 2023–2026 (NCT05747534) | 107 participants aged 7–50 | 250 or 500 µg four times daily for 21 days | Completed June 18, 2026; primary outcome includes adverse-event profiling; results not yet posted |
Pooled, a 2022 systematic review found four randomized trials meeting its criteria, totaling 626 patients — 465 on larazotide and 161 on placebo (PMID 34339872). That is a small safety database by approved-drug standards and a large one by research-peptide standards: BPC-157 and KPV, the two compounds most often sold alongside it, have no published placebo-controlled human safety data at all.
The Real Rates: What Was Reported Against Placebo
Here is the honest shape of the evidence. The published abstracts report larazotide’s safety comparatively — “no increase,” “similar,” “comparable with placebo” — rather than as a per-event percentage table. The per-event tables exist inside the full papers and their supplements; we cite what the abstracts state and do not reconstruct numbers we have not verified. Three findings are quantified in the abstracts themselves:
- Gut symptoms after acute gluten, 2007: more frequent in the placebo group than the AT-1001 group (P = 0.018); interferon-gamma rose in 4 of 7 placebo patients (57%) versus 4 of 14 on the drug (29%); intestinal permeability rose 70% on placebo and did not rise on AT-1001 (PMID 17697209).
- Symptom days, 2015: at 0.5 mg, a 26% decrease in symptomatic days (P = .017), a 31% increase in improved-symptom days (P = .034), and a 50%-or-more reduction in weekly abdominal-pain score for at least 6 of 12 weeks (P = .022) (PMID 25683116).
- Gluten-related diarrhea, pooled: reduced on larazotide versus placebo in patients undergoing gluten challenge, with “other AEs were comparable between both AT-1001 and placebo groups” (PMID 34339872).
Notice what those numbers are: the drug reducing the symptoms a side-effect page normally lists. For a compound built to blunt a gluten reaction in people being fed gluten, that is the expected direction — and it is why most “larazotide side effects” pages either invent a list or borrow one from an unrelated peptide.
PERSONAL OBSERVATION — PI EDITORIAL
Reading the five abstracts side by side, the thing that stood out was how little the safety language changed across fifteen years and 626 pooled patients: “no increase,” “similar,” “comparable.” No investigator found a signal worth a sentence. The second thing that stood out is why that is hard to improve on: the population has a symptomatic baseline. Headache, tiredness, bloating and diarrhea are the complaints celiac patients bring into a trial, which is why the 2015 investigators measured them as outcomes rather than adverse events — and why a reader with celiac disease should treat any confident side-effect list for this compound, ours included, as a statement about attribution first and the molecule second.
Headache and Urinary Tract Infection: The Two Named Events
Only one abstract names specific adverse events: the 2012 dose-ranging study, in which “the most common adverse events were headache and urinary tract infection” (PMID 22825365). Two things about that sentence deserve a plain reading. First, “most common” is not “caused by” — the abstract does not report whether either event was more frequent on larazotide than on placebo, and the same abstract calls the drug “generally well tolerated” with no serious adverse events. Second, in our reading, a 14-day trial that names headache and urinary tract infection as its leading events is telling you it found nothing more specific.
The 2015 trial then complicates the headache story in the drug’s favor. Among its exploratory endpoints, the 0.5 mg dose produced “a decrease in the nongastrointestinal symptoms of headache and tiredness (P = .010)” relative to placebo (PMID 25683116). The 2015 investigators measured headache and tiredness as nongastrointestinal symptoms of celiac disease, not as adverse events; in a population still symptomatic on a gluten-free diet, a drug that reduces gluten-driven symptoms could plausibly reduce them. Whether the 2012 headaches were a drug effect, chance, or celiac disease itself, the larger and longer trial pointed the other way. That is the complete published record on the two named events. Anyone offering you percentages for them is not citing an abstract we can find.
The Gut Itself: Diarrhea, Symptom Scores and a Backwards Dose Curve
Because larazotide acts in the gut, gut symptoms are where its side effects would appear if it had any of note — and the trials went looking. The 2007 in-patient study found gastrointestinal symptoms more frequent on placebo than on drug after acute gluten (PMID 17697209). The 2013 gluten-challenge trial found that 1 mg three times daily limited gluten-induced symptoms on the Gastrointestinal Symptom Rating Scale (P = 0.002), while the ratio of anti-tissue-transglutaminase IgA over baseline was 19.0 on placebo against 5.78, 3.88 and 7.72 at 1, 4 and 8 mg (PMID 23163616). The meta-analysis singled out one adverse event that larazotide reduced: gluten-related diarrhea (PMID 34339872).
The oddity — and it is reported as an oddity, not smoothed over — is that the dose-response ran backwards. In 2012, larazotide appeared to limit symptom worsening “at some lower doses but not at the higher dose.” In 2013 the 1 mg arm, not the 8 mg arm, carried the symptom result. In 2015 the 0.5 mg arm met the primary endpoint while 1 mg and 2 mg “were no different than placebo for any end point.” The investigators did not explain this in the abstracts, and neither will we; a locally acting peptide whose effect fades at higher concentrations is a real pharmacological pattern, but the published record simply shows the curve and stops. What it means for side effects is straightforward: the doses with any evidence of benefit were the smallest ones; no abstract reports harm at the larger doses, and none reports a benefit either.
Serious Adverse Events — and Why the Phase 3 Really Stopped
The 2012 abstract states it directly: “No serious adverse events were observed” (PMID 22825365). The 2015 abstract states safety was “comparable with placebo” over 12 weeks in 342 people (PMID 25683116). The MIS-C trial followed 12 children for 24 weeks after three weeks of dosing and reported no larazotide-related adverse events (PMID 40737433). Nothing in the published record describes a serious event attributed to the drug.
The Phase 3 is where readers assume something went wrong, so it is worth being exact. CedLara randomized celiac patients to 0.25 mg, 0.5 mg or placebo, with 525 planned. At an interim analysis conducted at roughly half the target enrollment, after the 12-week double-blind phase, the sponsor concluded that “the additional number of patients needed to determine a significant clinical outcome between placebo and larazotide is too large to support trial continuation” (9 Meters Biopharma, June 21, 2022). The registry lists the reason as “Trial terminated by Sponsor,” with 307 participants enrolled and no results posted (NCT03569007). That is a futility stop: the drug was not separating from placebo on the symptom endpoint. The press release reports no safety data at all — which, in a discontinuation announcement, is itself information. A safety-driven stop reads very differently.
Two caveats keep this honest. A futility stop means the largest trial ever launched for larazotide — 525 planned, 307 actually randomized — never published its adverse-event table, so a block of 307 participants is missing from the safety literature. And “no serious adverse events in 86 people over 14 days” is a statement about detection power, not a guarantee: rare events need thousands of exposures to surface, and larazotide has had hundreds.
When Side Effects Happen — and What the Trials Can Say About Fading
We do not publish week-by-week side-effect calendars unless a trial reports one, and none of the larazotide trials did. What the designs do tell you is the window in which anything was observed at all: single doses given in an in-patient setting (2007); 14 days (2012); 6 weeks (2013); 12 weeks of treatment bracketed by 4-week placebo periods (2015); 3 weeks of dosing with 24 weeks of follow-up (MIS-C); and 21 days of dosing with outcomes measured over 8 weeks (Long COVID, per its registry entry). Within those windows the abstracts report comparability with placebo and nothing about onset, peak or resolution — because there was nothing to time.
One structural detail from the 2015 trial is relevant to anyone judging their own experience: it included a 4-week placebo run-in before treatment and a 4-week placebo run-out after (PMID 25683116). The abstract records the design, not the rationale, but run-in periods of that kind exist to separate a drug effect from the improvement symptomatic patients show on placebo — and from symptoms in the first weeks of any new regimen that belong to the condition, not the compound. A gut that has been symptomatic for years generates plenty of events to misattribute.
What 12 Weeks Can and Can’t Tell You
The longest controlled exposure in the published literature is 12 weeks (PMID 25683116); the Phase 3 double-blind phase was also 12 weeks. The developer’s description of “a favorable safety profile comparable to placebo for long-term chronic administration” is a company statement from 2019 (Innovate Biopharmaceuticals), and “long-term” there means the trial durations above, not years. Nobody has published data on a person taking larazotide continuously for a year, and the compounding-pharmacy users described below are, in effect, running that experiment without a control group.
The mechanistic argument for low long-term risk is real but should be labeled as an argument: a peptide that stays in the gut lumen and is degraded there has limited opportunity to accumulate or act elsewhere. The counter-argument is equally real: tight junctions are a whole-body system, and the celiac trials could not have detected a slow effect on nutrient absorption or the microbiome. Neither argument is data.
Who the Trials Enrolled — and Who They Didn’t
Every celiac trial enrolled adults with diagnosed celiac disease maintained on a gluten-free diet; the 2015 trial required at least 12 months of that diet and ongoing symptoms (PMID 25683116). The MIS-C trial enrolled hospitalized children with a median age of 5.7 (PMID 40737433). The Long COVID trial enrolled ages 7 to 50 with a Long COVID diagnosis (NCT05747534).
Compare that to who is actually asking about larazotide side effects, judging by where the conversation happens online: long-COVID and ME/CFS communities, SIBO and IBS forums, mast-cell-activation groups, and people self-describing “leaky gut” — none of which is a population any published trial studied, and the last of which is not a recognized diagnosis. The safety data transfers to those readers only by assumption. That is not a reason to expect harm; it is a reason not to quote a celiac trial as if it were about you.
Dose and Side Effects: The Correlation Runs Backwards
Across the program the trials administered, three times daily, 0.25 to 8 mg in 2012, 1 to 8 mg in 2013 and 0.5 to 2 mg in 2015, then 0.25 or 0.5 mg up to three times daily in the Phase 3; the 2007 study used single 12 mg doses; the Mass General Long COVID trial used 250 or 500 µg four times daily (NCT05747534), and the MIS-C abstract reports four-times-daily dosing without stating an amount. The direction of travel is unmistakable: the doses studied fell by more than an order of magnitude over fifteen years, because the low doses carried the efficacy signal and the high doses carried nothing. On the safety side, no abstract reports a dose-related increase in adverse events at any level tested — and none breaks adverse events down by dose arm, so nothing more can be said in either direction.
We do not translate any of that into a suggestion. There is no approved dose of larazotide, the trials disagreed with each other about the effective one, and the products people can obtain are not the trial product. If you are working with a research vial, our peptide calculator converts whatever amount you enter — it does not propose one.
DID YOU KNOW
“AT-1001” is not a unique name. It is larazotide’s original code, but the same code was used by Amicus Therapeutics for migalastat, a Fabry-disease drug (NCT01218659), and by a separate research program for an α3β4 nicotinic receptor ligand studied for nicotine dependence (PMID 22278092, PMID 26162864). Three unrelated molecules, one label. A vial marked “AT-1001” with no sequence or mass-spectrometry identity on its certificate is not telling you which one it is.
What People Taking Larazotide Report
We searched Reddit on September 15, 2026 and read 91 larazotide-specific posts, 53 of them with active comment threads. Post volume has risen sharply since 2024 — 10 posts that year, 31 in 2025, 28 so far in 2026 — and lives mostly in long-COVID, general peptide and celiac communities, with smaller clusters in IBS, SIBO and ME/CFS forums. Everything below is paraphrased and linked to its thread; reports are hypothesis fuel, not evidence — no dose control, no blinding, self-selected posters, and products of unknown identity.
- The clearest intolerance report comes from a mast-cell-activation thread: one commenter started with a fraction of a capsule and, about a month later, reported “a horrific day” and never took a second dose. In the same thread another commenter describes splitting capsules, taking a portion before meals, and crediting it with tolerating a separate medication the next day (thread). Same product class, opposite experiences, no way to adjudicate.
- Headaches, then gas: a gut-health poster ten days in described introducing new foods without reactions and a few headaches they considered expected; about seven weeks later they reported constant gas and burping, which they attributed to a SIBO recurrence rather than the peptide (thread). Headache is the one event the 2012 trial also named.
- Trial participants speak: in the same MCAS thread, a commenter in Boston on their last day of study medication in a larazotide trial reported noticing nothing good or bad and assumed placebo; another said the drug had helped them with food (thread). A long-COVID poster invited to the trial asked for others’ experiences and was pointed to the MIS-C result (thread); another had signed up but aged out before it began (thread).
- How people are getting it: a compounded oral capsule recommended by an osteopath, taken before restaurant meals (thread); a compounding pharmacy at a quoted three dollars per pill on a doctor’s recommendation, and a parent describing a combined capsule of BPC-157, larazotide and KPV given to a child (thread); celiac patients locating it at compounding pharmacies after the Phase 3 stopped (thread); and SIBO patients frustrated at finding it at all (thread).
The pattern, such as it is: no cluster around any specific side effect, one clear intolerance, a few headaches, and many people asking where to buy it. That fits a compound with a placebo-like trial profile — and equally a compound too few people have taken for a pattern to show.
Stacked With BPC-157 and KPV: What the Data Covers
Larazotide is sold and compounded alongside BPC-157 and KPV as a “gut stack,” and the Reddit threads above include a child given all three in one capsule. No published study has tested larazotide with either compound, in humans or animals. The three have different proposed mechanisms — junction regulation, a gastric-protein fragment with rodent healing data, and an anti-inflammatory tripeptide — which is the marketing logic, and also the reason nothing can be said about how they interact. Each component’s own safety record is covered on its page: larazotide’s is on this one; BPC-157’s human record is empty; KPV’s is empty. A stack’s safety is not the sum of its parts when two of the parts are zero.
How Larazotide Compares to BPC-157 and KPV on Safety Data
On the single axis of published human safety data, larazotide is in a different category from the compounds it is sold next to. It has five randomized, placebo-controlled trials with adverse-event reporting, a meta-analysis, a Phase 3 terminated for futility and a completed Long COVID trial. BPC-157 has a large rodent literature and no placebo-controlled human trial with published adverse-event data; KPV has mouse colitis data — both covered on their own pages. The comparison cuts both ways: larazotide’s cleaner safety story is a product of having been tested, and its efficacy story ended in a futility stop that the other two have never had to face because they were never put to the test. Our peptides for gut health page ranks the three on evidence; if the question is which one has had its side effects measured in people, the answer is only this one.
Where the Trial Data Stops: Compounded Capsules and Research Vials
Everything above describes a pharmaceutical product manufactured under a company’s control and given in monitored trials. Two other products carry the name. Compounding pharmacies, according to multiple Reddit accounts, dispense larazotide capsules on a practitioner’s order; whether a given pharmacy may lawfully compound it depends on the FDA’s bulk-drug-substance categories under Section 503A, which we have not been able to verify for larazotide — it was not among the seven peptides in the July 2026 advisory committee vote, and a pharmacy should be able to tell you the basis on which it compounds. Research-chemical vendors sell vials labeled larazotide or AT-1001 for non-human research use, with no regulatory review of identity or purity.
The trial safety data attaches to neither product. It attaches to a specific peptide, at a verified purity, in a formulation designed to survive the stomach and act in the intestine. A capsule or vial inherits that record only to the extent that it contains the same molecule — which is exactly what a certificate of analysis exists to establish. Our guide to reading a peptide certificate of analysis covers what identity confirmation looks like; the AT-1001 name problem above is a reason to insist on it for this compound in particular.
PERSONAL OBSERVATION — PI EDITORIAL
When we checked the two vendors we review for this page on September 15, 2026, neither listed larazotide. That is unusual — most compounds with a Reddit following are stocked somewhere we track — and it fits what the threads say: this is a compounding-pharmacy compound first and a research-vial compound second. It also means the usual PI move, comparing batch COAs across vendors, is not available here. A compounding pharmacy is regulated differently from a research vendor, and the question to ask it is not “is this tested” but “where did the raw peptide come from, and can I see its certificate.”
PRACTICAL NOTES — WHAT THE TRIALS DID, NOT INSTRUCTIONS
- Every trial that states a route gave larazotide by mouth; the developer described the trial product as taken before meals, and no study has examined injection of this compound in anyone
- The 2015 trial bracketed treatment with 4-week placebo periods — the standard way to separate a drug effect from placebo improvement, and a reminder that early changes on any new regimen are hard to attribute
- Compounded capsules and research vials are different products from the trial drug; a batch-matched certificate confirming identity is the minimum, per our lab-testing guide
- Diagnosed celiac disease is managed with a gastroenterologist — none of the trials replaced the gluten-free diet, and neither should a page
Sourcing and Verification
As of September 15, 2026, neither Amino Club nor Mile High Compounds lists larazotide, so we have no vendor to compare and no batch certificates to review for it. Where it does appear on the research market, the standard applies with extra force: a certificate of analysis for the specific lot, showing identity by mass spectrometry and purity by HPLC, from a named laboratory. Because “AT-1001” denotes three different molecules, identity is the line item that matters most. Our framework for spotting a fake peptide website and our vendor reviews cover the rest. Everything sold in this market is labeled for non-human research use, and for larazotide that label describes the total regulatory oversight of the product.
US Regulatory Status
Larazotide is not an FDA-approved drug. Its developer stated in 2019 that it held FDA Fast Track designation for celiac disease (company statement); Fast Track speeds review but approves nothing, and the Phase 3 that would have supported an application was terminated in June 2022 (NCT03569007). The compound’s active clinical life is now academic: the Mass General Long COVID trial completed on June 18, 2026 with 107 participants and had not posted results as of its August 18, 2026 registry update (NCT05747534). Its primary outcome includes adverse-event profiling, so when results appear they will add the largest post-2015 safety dataset to this page. We will update it.
Other compounds we cover in the same honest register, from one end of the evidence scale to the other: retatrutide and eloralintide with large trial datasets; emideltide and 5-Amino-1MQ with thin ones; ENDO-205, a pipeline peptide with none yet; and CAQK, AHK-Cu and dihexa, where no human has been studied at all.
Frequently Asked Questions
What are the side effects of larazotide?
In five published randomized trials, none reported a higher adverse-event rate on larazotide than on placebo; two state the rates were similar or comparable outright. The one study that named events listed headache and urinary tract infection as most common, with no serious adverse events; the 342-person trial found the 0.5 mg dose reduced headache and tiredness versus placebo. No trial reported a dose-related increase in side effects.
Is larazotide safe?
In celiac adults and in a small group of children, over exposures of up to 12 weeks, the trials found safety comparable with placebo and no serious drug-related events. That record covers the pharmaceutical product in monitored patients. It does not cover compounded capsules, research vials, use beyond 12 weeks, or conditions the trials never studied.
Why did the larazotide Phase 3 trial stop?
For futility, not safety. At a planned interim analysis in June 2022 the sponsor concluded the number of additional patients needed to show a difference from placebo was too large to justify continuing. The announcement reported no safety concern, and the registry lists the reason as “Trial terminated by Sponsor.”
Does larazotide cause headaches?
Headache was one of the two most common adverse events in the 86-person 2012 trial, but the abstract does not say it was more frequent than on placebo. In the larger 2015 trial the 0.5 mg dose reduced headache and tiredness compared with placebo. The published record does not support headache as an established side effect.
Is larazotide FDA approved?
No. The developer stated it held Fast Track designation for celiac disease, but the Phase 3 that would have supported an application was terminated in 2022. It remains an investigational compound; a Long COVID trial at Massachusetts General Hospital completed in June 2026 with results pending.
Can you buy larazotide?
Not from either vendor we review, as of September 15, 2026. Online accounts describe compounding pharmacies dispensing larazotide capsules on a practitioner’s order, and some research-chemical sites list vials labeled larazotide or AT-1001 for non-human research use. Because “AT-1001” also names two unrelated molecules, a certificate confirming identity matters more than usual.
Glossary of Terms
- Larazotide acetate (AT-1001, INN-202): an eight-amino-acid oral peptide developed as a tight-junction regulator for celiac disease.
- Zonulin: a protein released by the gut lining that loosens tight junctions; larazotide is described as its antagonist.
- Tight junctions: protein seals between adjacent intestinal cells that control what passes between them (the “paracellular” route).
- LAMA ratio: the urinary lactulose-to-mannitol excretion ratio, an experimental marker of intestinal permeability that proved too variable to show a drug effect in the outpatient trials.
- GSRS / CeD-GSRS: the Gastrointestinal Symptom Rating Scale and its celiac-specific version, the symptom scores on which larazotide’s low doses showed benefit.
- Gluten challenge: a trial design in which celiac patients on a gluten-free diet deliberately consume a fixed daily amount of gluten (2.4–2.7 g in these trials) to provoke a measurable reaction.
- Futility stop: ending a trial because an interim analysis shows it is unlikely to reach its efficacy endpoint — distinct from a stop for safety.
- MIS-C: multisystem inflammatory syndrome in children, a post-COVID condition in which larazotide was tested as an add-on therapy.
References
- The safety, tolerance, pharmacokinetic and pharmacodynamic effects of single doses of AT-1001 in coeliac disease subjects: a proof of concept study — Aliment Pharmacol Ther, 2007 (PubMed)
- A randomized, double-blind study of larazotide acetate to prevent the activation of celiac disease during gluten challenge — Am J Gastroenterol, 2012 (PubMed)
- Larazotide acetate in patients with coeliac disease undergoing a gluten challenge: a randomised placebo-controlled study — Aliment Pharmacol Ther, 2013 (PubMed)
- Larazotide acetate for persistent symptoms of celiac disease despite a gluten-free diet: a randomized controlled trial — Gastroenterology, 2015 (PubMed)
- Larazotide acetate for treatment of celiac disease: a systematic review and meta-analysis of randomized controlled trials — Clin Res Hepatol Gastroenterol, 2022 (PubMed)
- Viral spike antigen clearance and augmented recovery in children with post-COVID multisystem inflammatory syndrome treated with larazotide — Sci Transl Med, 2025 (PubMed)
- Larazotide acetate: a pharmacological peptide approach to tight junction regulation — Am J Physiol Gastrointest Liver Physiol, 2021 (PubMed)
- Larazotide acetate regulates epithelial tight junctions in vitro and in vivo — Peptides, 2012 (PubMed)
- The therapeutic use of the zonulin inhibitor AT-1001 (larazotide) for a variety of acute and chronic inflammatory diseases — Curr Med Chem, 2021 (PubMed)
- Innovate Biopharmaceuticals completes license for late-stage celiac disease asset from Alba Therapeutics — PR Newswire, 2016
- AT-1001: a high affinity and selective α3β4 nicotinic acetylcholine receptor antagonist — Neuropsychopharmacology, 2012 (PubMed; the unrelated compound sharing the code)
- AT-1001 is a partial agonist with high affinity and selectivity at human and rat α3β4 nicotinic cholinergic receptors — Mol Pharmacol, 2015 (PubMed; the unrelated compound sharing the code)
- CedLara Phase 3 — ClinicalTrials.gov NCT03569007 (terminated by sponsor, 307 enrolled, no results posted)
- AT1001 for the Treatment of Long COVID — ClinicalTrials.gov NCT05747534 (completed June 18, 2026; 107 participants)
- Amicus Therapeutics AT1001 (migalastat) Phase 3 — ClinicalTrials.gov NCT01218659 (the unrelated drug sharing the code)
- 9 Meters Biopharma press release, June 21, 2022 — interim analysis of Phase 3 study of larazotide does not support trial continuation (SEC filing)
- Innovate Biopharmaceuticals press release, August 13, 2019 — first patient dosed in Phase 3 (GlobeNewswire)