Exenatide

A laboratory-made copy of exendin-4, a 39-amino-acid peptide found in the saliva of the Gila monster, a venomous lizard from the deserts of the American southwest. It is not a copy of any human hormone — it happens to resemble the human gut hormone GLP-1 closely enough to switch on the same receptor. When it was approved in the United States on 28 April 2005 it became the first GLP-1 medicine anywhere in the world. It has been sold as Byetta, injected twice a day, and as Bydureon and Bydureon BCise, injected once a week; all three American brand applications are marked Discontinued in the US regulator's own database and a generic twice-daily injection was approved there on 19 November 2024.

Sold as

Injection under the skin.

Legal status in the EU

Approved and still authorised in the EU. Byetta (EMEA/H/C/000698, AstraZeneca AB) was authorised on 20 November 2006, latest renewal 22 July 2016; Bydureon (EMEA/H/C/002020, same holder) on 17 June 2011, latest renewal 18 February 2016. Both EMA medicine pages carried the status "Authorised" when checked on 3 August 2026. Prescription only, for type 2 diabetes; the weekly form is approved from age 10. The picture outside the EU is one of commercial withdrawal, not safety withdrawal: in the United States Drugs@FDA marks BYETTA (NDA 021773), BYDUREON and BYDUREON PEN (NDA 022200) and BYDUREON BCISE (NDA 209210) all as Discontinued, while a generic twice-daily exenatide injection (Amneal, ANDA 206697) was approved on 19 November 2024 and has a current label — an approval that would not follow a withdrawal for safety. In the United Kingdom the Department of Health and Social Care issued Medicine Supply Notification MSN/2025/054 on 15 October 2025 stating that the weekly pens had been discontinued with supplies now exhausted and that other GLP-1 medicines remained available; no safety reason is given anywhere in it. Whether any EU member state still stocks either product could not be established from the regulators' own files. Exenatide does not appear anywhere on the WADA 2026 Prohibited List.

Exenatide switches on the GLP-1 receptor, which makes the pancreas release insulin only when blood sugar is already high, suppresses glucagon (the hormone that pushes blood sugar up), slows the stomach emptying, and reduces food intake through less appetite and more fullness. Because the insulin release is tied to the sugar level, it barely causes low blood sugar on its own. The chain HGEGTFTSDLSKQMEEEAVRLFIEWLKNGGPSSGAPPPS matches residues 48–86 of the Gila monster exendin-4 precursor in UniProt P26349 letter for letter; every residue is a standard L-amino acid and the only chemical modification is an amide cap on the final serine, stated out loud both in UniProt (MOD_RES 86, "Serine amide") and in PubChem's systematic name, which is why the one-letter code can honestly be printed here. Being a lizard peptide rather than a human one, it provokes antibodies in a substantial share of patients, and in about 3 in 100 those antibodies abolish the blood sugar response.

Part of the bodyWhat happensHow it is knownKind of evidence

Blood sugar, and the pancreas

This is what the drug was approved for and the part with the most measurement behind it. It tells the pancreas to release insulin only when blood sugar is already high, and to release less glucagon, the hormone that pushes blood sugar up. Because the insulin release is tied to the sugar level, exenatide on its own almost never drives blood sugar dangerously low. It does when it is combined with a sulfonylurea tablet or with insulin, which push insulin out regardless of what the sugar is doing. The pancreas can also become inflamed, which is the reason the drug is stopped and never restarted if it happens.

Long-term blood sugar measured in 963 adults on the twice-daily form against 483 on dummy injections over 30 weeks: down 0.89 percentage points on the higher amount and 0.59 on the lower, against a rise of 0.08 on the dummy. With a sulfonylurea in the mix, low blood sugar was reported by 23.5% and 25.2% of people on the drug against 12.6% and 3.3% on dummy injections. The European regulator records inflamed pancreas at a frequency it says cannot be estimated from the data it has.

Measured in people

Source [01]

Stomach and gut

The stomach empties more slowly, and for most people that shows up as feeling sick. Nausea is the single most common effect by a wide margin, followed by vomiting and loose stools. It arrives with the first injections and fades for most people as treatment carries on. The slowing is real enough to change how other tablets are absorbed: a standard dose of paracetamol taken at the same time as the injection reached a peak level in the blood between 37% and 56% lower than when it was taken alone, depending on the gap. That is also why the label warns anaesthetists — food can still be sitting in the stomach when someone is put under.

In the 30-week trials, nausea was reported by 44% of adults on the drug against 18% on dummy injections, vomiting by 13% against 4%, and loose stools by 13% against 6%, in 963 people on the drug and 483 on dummy injections. Stomach emptying was measured indirectly, by giving 1,000 mg of paracetamol at set intervals around the injection and tracking it in the blood.

Measured in people

Source [01]Source [13]

Appetite, and body weight

People eat less, and the weight that comes off is modest and arrives early. The product information states plainly that exenatide reduces food intake through lower appetite and a stronger sense of being full. In the trials the weight loss was larger in the people who felt sick than in the people who did not, which is part of why it is hard to separate the appetite effect from the nausea. The effect is much smaller than the newer drugs in this class produce, and exenatide has never been approved for weight loss anywhere.

In 152 adults with obesity and no diabetes given the drug or dummy injections on top of the same supervised diet and exercise programme, weight fell 5.1 kg against 1.6 kg over 24 weeks. In 41 women with obesity and no diabetes, deliberately given no diet or exercise advice at all, weight fell 2.49 kg on the drug and rose 0.43 kg on the dummy. That weight loss is larger in people who get nausea — 2.4 kg against 1.7 kg — is stated in the European product information.

Measured in people

Source [01]Source [14]Source [15]

Heart and blood vessels

Nothing that could be shown. This is the most important negative result in the record and it comes from the biggest trial ever run on the drug. In 14,752 adults with type 2 diabetes, most of whom already had heart disease, heart attacks, strokes and heart deaths happened slightly less often on exenatide than on dummy injections, but the difference was small enough that the trial's own pre-planned test could not call it real. The drug passed the safety half of the trial — it does not make heart problems more likely — and failed the efficacy half.

Measured in 14,752 adults randomised to weekly exenatide or dummy injections and followed for a median of 38.7 months. A heart attack, stroke or heart death happened to 839 of 7,356 people on the drug (11.4%) against 905 of 7,396 on dummy injections (12.2%), hazard ratio 0.91 (95% CI 0.832 to 1.004). The test for being no worse passed, at P<0.001; the test for being better gave P=0.06 and did not pass.

Measured in people

Source [03]Source [07]

The immune system, reacting to the drug itself

Exenatide is a lizard peptide, not a human one, and a good share of people make antibodies against it. For most of them nothing follows: the antibody levels fall over time and blood sugar improves as usual. For a small group the antibodies appear to stop the drug working. This is the clearest practical difference between exenatide and the human-derived GLP-1 medicines that came after it, and it is why the product information for the weekly form tells prescribers to consider a different medicine when blood sugar will not come down despite the injections being taken.

Measured across three placebo-controlled trials in 963 adults on the twice-daily form: 38% had low antibody levels at 30 weeks, with blood sugar control much like people with none. A further 6% had high levels, and about half of those — roughly 3% of everyone given the drug — had no blood sugar response to it at all. In the weekly form, about 45% of adults had low antibody levels at the end of the trials.

Measured in people

Source [01]Source [02]

Kidneys

The kidneys are hit indirectly rather than directly. Days of vomiting or loose stools empty the body of fluid, and the kidneys fail because of that. Most of the reported cases followed exactly that sequence, and some needed dialysis. People whose kidneys are already badly damaged also get more gut trouble from the drug, which is why it is not recommended for them at all.

Reported after the drug went on sale rather than counted in a trial: the European regulator lists altered kidney function, including acute kidney failure, as uncommon — fewer than 1 in 100. Single doses given to people on dialysis produced more frequent and more severe gut reactions than in people with normal kidneys, which is the basis for the recommendation against use below a kidney filtration rate of 30.

Measured in people

Source [01]

Brain and nerves

Nothing measurable, in the one condition where it was properly tested. Laboratory and animal work suggested GLP-1 drugs might protect nerve cells, and a small earlier trial in Parkinson's disease looked encouraging enough to justify a full-sized one. The full-sized one, in 194 people over two years, found no difference at all in how fast Parkinson's disease got worse.

Measured in 194 people with Parkinson's disease at six UK hospitals, randomly given weekly exenatide or dummy injections for 96 weeks. The standard movement score, measured off Parkinson's medication, got worse by 5.7 points on the drug and 4.5 points on the dummy — a difference of 0.92 points (95% CI −1.56 to 3.39), p=0.47. The Lancet published an Expression of Concern about this paper on 27 June 2026; the notice itself could not be opened from this environment, so nothing is reported here about what prompted it.

Measured in people

Source [06]Source [16]

Approved on placebo-controlled trials in adults with type 2 diabetes: pooled across three 30-week studies in 1,446 people, long-term blood sugar fell 0.89 percentage points on the higher twice-daily amount and 0.59 on the lower, against a rise of 0.08 on placebo, and weight fell 1.91 kg and 1.41 kg against 0.65 kg. The one trial designed to show it prevents heart attacks and strokes did not show that: in EXSCEL, 14,752 adults followed a median of 38.7 months, the primary composite happened to 839 of 7,356 on exenatide (11.4%) against 905 of 7,396 on placebo (12.2%), hazard ratio 0.91 (95% CI 0.832–1.004), P<0.001 for non-inferiority but P=0.06 for superiority — the efficacy endpoint was missed. It also lost both of its head-to-head trials: against daily liraglutide (DURATION-6, 911 adults) HbA1c fell 1.28 against 1.48 percentage points, missing the trial's pre-set non-inferiority margin; against weekly semaglutide (SUSTAIN 3, 813 adults, 56 weeks) HbA1c fell 0.9 against 1.5 points and weight 1.9 kg against 5.6 kg. In adults with obesity and no diabetes it took off 5.1 kg against 1.6 kg on placebo over 24 weeks on the same lifestyle programme, but it has never been approved for weight loss anywhere. A phase 3 trial in 194 people with Parkinson's disease found no slowing of the disease over 96 weeks (movement score worse by 5.7 points against 4.5 on placebo, p=0.47); The Lancet published an Expression of Concern about that report on 27 June 2026 and the notice itself could not be opened from this environment.

Blood sugar moves first and fastest. With the weekly form the first measurable fall showed up at the first check, four to six weeks in; with the twice-daily form the regulator puts it at about 12 weeks. Nausea runs the other way — it is worst at the very start and fades. On the weekly form, nausea or vomiting was reported by 2% of people in the first week and 1% by the fourth. Switching from the twice-daily form to the weekly one causes blood sugar to drift up for about two to four weeks before the weekly form takes hold. Lumps under the skin from the weekly injections mostly cleared in four to eight weeks. Over the long run the effect erodes: in the open extension where 258 people carried on with the weekly form, blood sugar was down 2.0 percentage points at week 52, then gradually climbed back, and at seven years was down 1.5 percentage points from where it started — but everybody in that extension was taking the drug, so there is no comparison group behind those two numbers and they cannot be read as what the drug did against nothing. Coming off the weekly form is slow. It takes about ten weeks for the drug to clear, which is why the European product information tells women planning a pregnancy to stop three months in advance.

What was measuredWhat happenedIn whomHow it is known

Heart attacks, strokes and heart deaths in people with type 2 diabetes

The trial's target was to show fewer of them, and it did not. A heart attack, stroke or heart death happened to 839 of 7,356 people on weekly exenatide (11.4%) against 905 of 7,396 on dummy injections (12.2%), hazard ratio 0.91 (95% CI 0.832 to 1.004). The pre-planned test for being better than the dummy gave P=0.06 and failed; the test for being no worse passed at P<0.001. Death from any cause: 507 people (6.9%) against 584 (7.9%). Death from a heart cause: 340 (4.6%) against 383 (5.2%), hazard ratio 0.88 (95% CI 0.76 to 1.02).

14,752 adults with type 2 diabetes at any level of heart risk, 73.1% of whom already had heart disease, followed a median of 38.7 months with a median 27.8 months on treatment

Measured in people

Source [03]Source [07]Source [17]

Long-term blood sugar and weight, twice-daily form

Long-term blood sugar fell 0.89 percentage points on the higher amount and 0.59 on the lower, against a rise of 0.08 on dummy injections. Reaching the treatment target of 7% or below: 33.6% and 25.3% against 7.9%. Weight fell 1.91 kg and 1.41 kg against 0.65 kg.

1,446 adults with type 2 diabetes not controlled on tablets, pooled from three 30-week placebo-controlled trials: 483 on the higher amount, 480 on the lower, 483 on dummy injections

Measured in people

Source [01]

The weekly form against the twice-daily form

The weekly injection worked better on blood sugar than the twice-daily one and no better on weight. In the 24-week trial, long-term blood sugar fell 1.6 percentage points against 0.9, a difference of 0.67 (95% CI 0.94 to 0.39); weight fell 2.3 kg against 1.4 kg, a difference of 0.95 kg (95% CI 1.91 to −0.01), which crosses zero and settles nothing. In the 30-week trial, blood sugar fell 1.9 points against 1.5, a difference of 0.33 (95% CI 0.54 to 0.12); weight fell 3.7 kg against 3.6 kg, a difference of 0.08 kg, which is nothing.

252 adults over 24 weeks (129 weekly, 123 twice daily) and 295 adults over 30 weeks (148 weekly, 147 twice daily), on diet and exercise alone or with tablets

Measured in people

Source [02]

Head to head against semaglutide

Exenatide lost on both counts. Long-term blood sugar fell 0.9 percentage points on weekly exenatide against 1.5 on weekly semaglutide, a difference of 0.62 (95% CI 0.80 to 0.44). Weight fell 1.9 kg against 5.6 kg, a difference of 3.78 kg (95% CI 4.58 to 2.98). Reaching the treatment target: 40% against 67%. Gut side effects went the other way — 33.3% on exenatide against 41.8% on semaglutide — but reactions where the needle goes in were far more common on exenatide, 22.0% against 1.2%.

813 adults with type 2 diabetes not controlled on tablets, randomised one to one, 56 weeks, open-label

Measured in people

Source [04]

Head to head against liraglutide

Exenatide lost again, and by enough that the trial failed even its lower bar of showing the two were equivalent. Long-term blood sugar fell 1.28 percentage points on weekly exenatide against 1.48 on daily liraglutide, a difference of 0.21 (95% CI 0.08 to 0.33) — wider than the 0.25 gap the trial had set as the limit for calling them equivalent. Exenatide was the gentler of the two: nausea in 9% against 21%, loose stools in 6% against 13%, vomiting in 4% against 11%, and 3% stopped because of side effects against 5%.

911 adults with type 2 diabetes on lifestyle measures and tablets, 461 on weekly exenatide and 450 on daily liraglutide, 26 weeks, open-label

Measured in people

Source [05]

Body weight in adults with obesity and no diabetes

Weight fell 5.1 kg on exenatide against 1.6 kg on dummy injections, with both groups on the same supervised diet and exercise programme — a difference of 3.3% of body weight. Nausea was reported by 25% against 4%. A third of each group dropped out: 34% on the drug and 32% on the dummy.

152 adults with obesity and no diabetes, a quarter of them with blood sugar already drifting upwards, 73 on the drug and 79 on dummy injections, 24 weeks

Measured in people

Source [14]

Body weight with no diet or exercise advice at all

Weight fell 2.49 kg on exenatide and rose 0.43 kg on dummy injections. The average hides a wide spread: 30% of the women lost at least 5% of their body weight, 39% lost less than that, and 31% gained weight while on the drug. Nausea was reported during 56% of the exenatide periods against 21% of the dummy periods.

41 women with obesity and no diabetes, each taking both the drug and a dummy for 16 weeks in random order with a 3-week gap between; the researchers deliberately gave no lifestyle advice

Measured in people

Source [15]

How fast Parkinson's disease got worse

No difference. The standard movement score, measured off Parkinson's medication, got worse by 5.7 points on exenatide and by 4.5 points on dummy injections over two years — an adjusted difference of 0.92 points in the wrong direction (95% CI −1.56 to 3.39), p=0.47. Serious side effects happened to 9 people on the drug (9%) and 11 on the dummy (11%). The authors' own conclusion was that they found no evidence to support exenatide as a treatment that slows the disease.

194 people with Parkinson's disease aged 25 to 80 at six UK hospitals, 97 on the drug and 97 on dummy injections, 96 weeks

Measured in people

Source [06]Source [16]

Blood sugar in children and teenagers, twice-daily form

It did not work. Long-term blood sugar was 0.28 percentage points lower on the drug than on dummy injections (95% CI −1.01 to 0.45), a range that runs from a useful benefit to a clear harm and therefore answers nothing. The European regulator records the result as not statistically significant.

120 patients aged 10 to 17 with type 2 diabetes, either on no diabetes medicine or on tablets, randomly given one of two amounts of the drug or dummy injections, 28 weeks

Measured in people

Source [01]

Blood sugar in children and teenagers, weekly form

This one did work, on blood sugar and on nothing else. Long-term blood sugar fell 0.36 percentage points on the drug and rose 0.49 on dummy injections, a difference of 0.85 (95% CI 1.51 to 0.19), p=0.012. Reaching the treatment target: 31.0% against 8.3%. Weight went the drug's way by 1.22 kg (95% CI 3.59 to −1.15) and fasting blood sugar by 1.2 mmol/L (95% CI 2.72 to −0.32), but both of those ranges cross zero and settle nothing.

82 patients aged 10 and over with type 2 diabetes, 58 on the drug and 24 on dummy injections, 24 weeks

Measured in people

Source [02]

Gallstones and an inflamed gallbladder

1.9% of people on weekly exenatide against 1.4% on dummy injections had a gallbladder event inside the 14,752-person heart trial.

The same 14,752 adults with type 2 diabetes

Measured in people

Source [07]

Gut effects dominate. On the twice-daily form, nausea was reported by 44% against 18% on placebo, vomiting by 13% against 4% and loose stools by 13% against 6% (963 on drug, 483 on placebo, 30 weeks); In the controlled trials of 16 weeks or longer, 8% stopped because of side effects against 3% on placebo. The weekly form causes much less nausea and instead produces lumps under the skin at the injection site, recorded as very frequent and mostly clearing in 4–8 weeks; against semaglutide, injection-site reactions were 22.0% versus 1.2%. Blood sugar dropping too low is a real risk only in combination with a sulfonylurea or insulin (23.5% and 25.2% against 12.6% and 3.3% on placebo). Serious but uncommon or unquantified: acute pancreatitis, including the bleeding and tissue-destroying kinds; acute kidney failure after days of vomiting or diarrhoea, sometimes needing dialysis; anaphylaxis; immune destruction of platelets, with reports of fatal bleeding; bowel obstruction; and breathing stomach contents into the lungs under anaesthesia. The US label for the weekly form carries a boxed warning about thyroid C-cell tumours seen in rats and bars anyone with a personal or family history of medullary thyroid cancer or with multiple endocrine neoplasia type 2; the twice-daily form carries no such warning, and Europe's only outright bar for either form is allergy.

EffectWhat it feels likeHow often

Inflammation of the pancreasSerious

Severe pain high in the belly that bores through to the back, often with vomiting. The rarer forms in which pancreas tissue bleeds or dies, sometimes fatally, have been reported. Both regulators instruct that the drug is stopped if it is suspected and never restarted if it is confirmed. The two agencies reviewed this together in 2014 across the whole class of drugs and reported that the assertion of a causal link was inconsistent with the data they had, while stating that a link could not be entirely ruled out.

The European regulator lists it in the column meaning the rate cannot be worked out from the data available. In the 14,752-person heart trial the numbers on the drug and on dummy injections did not differ significantly.

Source [01]Source [03]Source [18]

Kidneys failing after being dried out by vomiting or diarrhoeaSerious

This is a knock-on effect, not a direct one. Days of vomiting or loose stools empty the body of fluid and the kidneys fail because of it. Water tablets, blood-pressure tablets and anti-inflammatory painkillers made it more likely in the reported cases. Kidney function came back in some people once fluids were replaced and the drugs involved were stopped.

Uncommon — fewer than 1 in 100 in the European regulator's table, reported after the drug went on sale rather than counted in a trial. Some cases needed dialysis.

Source [01]

A whole-body allergic reactionSerious

Blood pressure drops, breathing becomes hard and the face or throat can swell. Swelling of the deeper layers of the skin is listed separately, at a frequency the regulator says cannot be estimated. The US label for the weekly form treats a previous serious reaction to exenatide as an outright bar on ever using it again.

Rare — fewer than 1 in 1,000 in the European regulator's table.

Source [01]Source [07]

The immune system destroying the body's own plateletsSerious

Platelets are what makes blood clot. In this reaction the body makes antibodies that attack them, but only while exenatide is present. The US label states that serious bleeding, which may be fatal, has been reported, and that anyone it happens to should never be given exenatide again. It is one of only three outright bars on the US label for the weekly form.

Frequency not known — it appears in the regulator's table only as something reported after the drug went on sale, with no denominator to turn into a rate.

Source [01]Source [07]

Blood sugar dropping too lowSerious

Shaking, sweating, confusion and hunger. On its own exenatide releases insulin only when blood sugar is already up, which is why it barely causes this by itself. Combined with insulin or a sulfonylurea, which push insulin out regardless, the risk becomes real, and both labels say the dose of those medicines may need cutting.

Depends entirely on what else is being taken. With a sulfonylurea tablet in the mix, low blood sugar was reported by 23.5% and 25.2% of people on the drug against 12.6% and 3.3% on dummy injections. With a different class of tablet instead, it was 11% against 7% — much the same. On the weekly form there were no severe episodes at all in the trials, and minor ones were 26.1% among people also taking a sulfonylurea against 0.9% among people who were not.

Source [01]Source [02]

A blocked bowelSerious

A belly that swells, pain, vomiting and no bowel movement. It comes from the same slowing of the gut that causes the nausea, taken to its extreme.

Rare — fewer than 1 in 1,000 in the European regulator's table.

Source [01]

Breathing stomach contents into the lungs during an operationSerious

The drug slows the stomach down, so food can still be sitting there when someone is put under general anaesthetic or deep sedation, even if they fasted as instructed. The US label states plainly that there is not enough information to say whether fasting for longer or pausing the drug would help. The warning is aimed at the anaesthetist rather than the patient.

No rate published. The warning was added to the US label for the weekly form in November 2024 and appears in the European product information as well.

Source [01]Source [07]

Serious reactions at the injection site, with the weekly formSerious

The weekly form works by suspending the drug in tiny polymer beads that release it slowly, and the beads sit under the skin for weeks. Most of what that produces is a harmless small lump. A small number of reports describe the same sites becoming infected or the tissue dying.

No rate published for the serious kind. The US label records abscesses, deep skin infection and tissue death reported after the drug went on sale, with isolated cases needing surgery.

Source [02]Source [07]

Feeling sick

The defining side effect of the twice-daily form and the most common reason people stopped taking it. It appears with the first injections, depends on the amount given, and gets less frequent and less severe for most people as treatment carries on. The weekly form causes far less of it: nausea and vomiting were reported by 2% of people in the first week and 1% by the fourth.

Very common — 44% of people on the twice-daily form against 18% on dummy injections in the 30-week trials, in 963 people on the drug and 483 on dummy injections. The European regulator puts it differently: 36% of everyone given the twice-daily form reported at least one episode.

Source [01]Source [07]Source [13]

Vomiting and loose stools

They arrive with the nausea in the first weeks and usually settle with it. They matter beyond the discomfort, because they are the route by which people get dried out and their kidneys suffer.

Very common on the twice-daily form: vomiting in 13% against 4% on dummy injections, loose stools in 13% against 6%.

Source [13]

Lumps under the skin where the weekly injection goes in

The European product information says small lumps under the skin were seen very frequently and are an expected consequence of how the weekly form is built — the drug is held in tiny polymer beads that dissolve slowly. Most lumps caused no symptoms and disappeared over four to eight weeks. Half a percent of people stopped treatment because of them.

The most common thing recorded with the weekly form. In the two trials behind the US autoinjector, 10.5% of 526 adults reported a lump, ahead of nausea at 8.2% — but those trials compared the drug against other diabetes medicines rather than against a dummy, so those two figures have no dummy arm behind them. The comparison that does have one is against semaglutide, where reactions at the injection site were reported by 22.0% of people on exenatide against 1.2% on semaglutide.

Source [02]Source [04]Source [07]

The drug quietly stopping working

This is not a side effect anyone feels. It shows up as blood sugar that will not come down, and the labels for the weekly form tell prescribers to consider stopping it and using a different diabetes medicine if that happens. Blood sugar going up is also the single most reported term in the US regulator's public side-effect file for exenatide, with 15,473 reports — ahead of weight loss and ahead of nausea.

About 3 in 100 of everyone given the twice-daily form in the controlled trials had no blood sugar response at all, traced to high levels of antibodies against the drug. 38% had low antibody levels and responded normally.

Source [01]Source [07]Source [19]

Feeling jittery, dizzy, or getting a headache

The headache and dizziness figures are worth reading twice. Both are common on the drug and nearly as common on a dummy injection, which is what it looks like when most of a side effect is not the drug. Feeling jittery shows a clearer gap and is at least partly the feeling of blood sugar running lower than the body is used to.

Feeling jittery 9% against 4% on dummy injections; dizziness 9% against 6%; headache 9% against 6%; indigestion 6% against 3%.

Source [13]

Blood thinning more than intended

In people taking warfarin, a blood-thinning tablet, the measure of how thin the blood is has been reported to rise after exenatide is started. The interaction study that was done found no meaningful change in how much warfarin reached the blood; the reports of a rising measure arrived only after the drug went on sale and carry no rate, so both labels ask for that measure to be checked closely when treatment starts and when the amount goes up.

Frequency not known. It appears in the regulator's table only as something reported after the drug went on sale, some reports involving bleeding.

Source [01]

These are the amounts the studies below gave their participants, and they are here as facts about those studies. They are not a recommendation, not a starting point and not a range to pick from. What a person should take, if anything, is a question for a doctor who knows them.

The three 30-week placebo-controlled trials behind the European approval of the twice-daily form, 1,446 adults with type 2 diabetes

5 micrograms twice a day, injected under the skin within the hour before the morning and evening meal, for 4 weeks; then either kept there or raised to 10 micrograms twice a day

30 weeks

Source [01]

The two trials comparing the weekly form against the twice-daily form, 547 adults with type 2 diabetes in total

2 mg once a week, injected under the skin

24 weeks in one trial and 30 weeks in the other, the second followed by an open extension in which everyone took the weekly form for a further 7 years

Source [02]

EXSCEL, the heart trial, 14,752 adults with type 2 diabetes

2 mg once a week, injected under the skin, added to whatever the person's usual diabetes care already was

Median 27.8 months on treatment, median 38.7 months of follow-up

Source [03]

SUSTAIN 3, the head-to-head against weekly semaglutide, 813 adults

2.0 mg once a week, injected under the skin

56 weeks

Source [04]

DURATION-6, the head-to-head against daily liraglutide, 461 adults in the exenatide group

2 mg once a week, injected under the skin

26 weeks

Source [05]

The 24-week weight trial in 152 adults with obesity and no diabetes, run alongside a supervised diet and exercise programme

5 micrograms twice a day before the morning and evening meal for 4 weeks, then 10 micrograms twice a day

24 weeks

Source [14]

The crossover weight study in 41 women with obesity and no diabetes, deliberately run with no diet or exercise advice

5 micrograms twice a day before breakfast and supper for 2 weeks, then 10 micrograms twice a day

16 weeks per period, with a 3-week gap between the two periods

Source [15]

The phase 3 Parkinson's disease trial, 97 people in the exenatide group

2 mg once a week, injected under the skin with a pen

96 weeks

Source [06]

The trial in 82 children and teenagers aged 10 and over with type 2 diabetes

2 mg once a week, injected under the skin

24 weeks

Source [02]

What people report

• Whether it prevents heart attacks and strokes — the trial built to answer that, in 14,752 people, missed its target at P=0.06 and nobody will run another one now that the drug is being withdrawn
• Why the same trial's death figures pointed the drug's way — 6.9% against 7.9% — while the pre-planned order of testing meant that difference could not be claimed as real
• Whether the thyroid tumours seen in rats given the weekly form mean anything for people — the US label says the human relevance has not been determined, and no human study has been long enough to find out
• How much weight it takes off over a year or more in people without diabetes — the two placebo-controlled trials that measured it ran 24 weeks and 16 weeks, in 152 and 41 people
• Why nearly a third of the women in the crossover weight study gained weight while taking it, and whether anything predicts who responds
• What happens to weight and blood sugar after stopping — no published trial of exenatide followed people off the drug
• Whether the antibodies people make against it cause anything besides the loss of blood sugar control — nobody has followed antibody-positive people for years
• How often the weekly form causes injection-site infections or tissue death — the US label records them with no rate at all
• Whether it is safe in pregnancy — the human data are described as inadequate and animal studies showed harm to the offspring
• How it behaves in severe kidney failure or on dialysis — never studied, which is why it is not recommended there, and single doses in people on dialysis produced worse gut reactions
• Whether it does anything at all in children and teenagers taken twice daily — the one trial in 120 of them found a difference of 0.28 percentage points with a range running from benefit to harm
• Whether the weekly form's blood sugar benefit in children translates into anything that matters — the same trial found no clear effect on weight or fasting blood sugar
• What the seven-year figures from the open extension mean — everyone in it was taking the drug, so there is nothing to compare them against
• Whether the Expression of Concern on the Parkinson's trial changes any of its numbers — the notice itself could not be opened from this environment
• How many people in the European Union can still actually get it — the authorisations are live, but whether any member state still stocks either product could not be established from the regulators' own files

Is exenatide still available?

It depends where you are, and it is changing. In the European Union both Byetta and Bydureon still hold marketing authorisations, so the medicine is legally approved there. In the United States both brands are marked Discontinued in the regulator's own database, but a generic twice-daily exenatide injection from Amneal was approved in November 2024 and has a current label. In the United Kingdom the health department announced on 15 October 2025 that the weekly pens had been discontinued and supplies were exhausted.

Source [08]Source [12]Source [20]Source [21]

Was exenatide taken off the market because it was dangerous?

No. Nothing in any regulator's file says a safety problem caused the withdrawals. The manufacturer told the US regulator it was discontinuing the brands, the European authorisations remain in force, and a generic version was approved in the United States after the brand was discontinued — which would not happen with a drug pulled for safety. A medicine can stop being sold simply because it no longer earns enough, and that is what happened here: newer weekly injections beat it clearly in head-to-head trials.

Source [21]Source [22]Source [23]

Does exenatide work for weight loss?

A little, and it has never been approved for it anywhere. In 152 adults with obesity and no diabetes, all of them on the same diet and exercise programme, weight fell 5.1 kg on exenatide against 1.6 kg on dummy injections over 24 weeks. In 41 women given no lifestyle advice at all, weight fell 2.49 kg on the drug and rose 0.43 kg on the dummy — but 31% of them gained weight while taking it. For comparison, in a head-to-head trial weekly exenatide took off 1.9 kg where weekly semaglutide took off 5.6 kg.

Source [04]Source [14]Source [15]

Exenatide vs Ozempic — which is better?

Semaglutide, the drug sold as Ozempic, beat exenatide on both measures when the two were tested head to head in 813 adults over 56 weeks. Long-term blood sugar fell 1.5 percentage points on semaglutide against 0.9 on exenatide, and weight fell 5.6 kg against 1.9 kg. Exenatide had slightly less gut trouble, 33.3% against 41.8%, but far more reactions where the needle goes in, 22.0% against 1.2%.

Source [04]

What are the side effects of exenatide?

The common ones are gut ones. On the twice-daily form, nausea was reported by 44% of people against 18% on dummy injections, vomiting by 13% against 4% and loose stools by 13% against 6%. On the weekly form the commonest thing is a small lump under the skin where the injection goes. The serious ones are rarer: inflammation of the pancreas, kidney failure after days of vomiting or diarrhoea, a whole-body allergic reaction, and — for the weekly form in the United States only — a warning about thyroid tumours seen in rats.

Source [01]Source [07]Source [13]

Is exenatide really made from lizard venom?

The original substance was found in it, yes. In 1992 a research team isolated a 39-piece chain they called exendin-4 from the venom of the Gila monster, a large venomous lizard from the American southwest, and worked out its full sequence. Exenatide is that chain, built in a factory rather than taken from an animal — no lizards are involved in making the medicine. It is not a copy of any human hormone, which is unusual for a drug of this kind and is why some people make antibodies against it.

Source [09]Source [24]

Is exenatide banned in sport?

No. Exenatide does not appear anywhere on the World Anti-Doping Agency's 2026 Prohibited List — not by name, and not through any class on it. The list's section on insulins covers insulins and insulin-mimetics, which is a different thing; GLP-1 medicines are not included. The catch-all section for unapproved substances does not apply either, because exenatide is an approved medicine.

Source [11]

Does exenatide help Parkinson's disease?

The proper test of that said no. 194 people with Parkinson's disease were given weekly exenatide or dummy injections for two years, and their movement scores got worse at the same rate in both groups — by 5.7 points on the drug and 4.5 on the dummy, a difference that could easily be chance. The authors concluded there was no evidence to support it as a treatment that slows the disease. The Lancet published an Expression of Concern about that paper in June 2026; the notice could not be opened from here, so nothing more is claimed about it.

Source [06]Source [16]

Can you buy exenatide online?

Not lawfully without a prescription. Exenatide is a prescription-only medicine everywhere it is approved, and in Sweden advertising an unapproved or prescription medicine to the public is illegal under the Medicinal Products Act. Where the brands have been discontinued the supply chain has gone with them, which is exactly the situation that pushes people towards sellers nobody checks — and what is in a vial from an unregulated source is verified by no one.

Source [12]Source [20]

Each letter represents one amino acid.

1. [01]EMA – Byetta (exenatide), product information: Table 2 (30-week placebo-controlled studies), Table 1 (adverse reactions), sections 4.4, 4.8 and 5.1 including the paediatric result
2. [02]EMA – Bydureon (exenatide), product information: Table 2 (weekly versus twice-daily), Table 7 (paediatric study), section 5.1 EXSCEL and section 4.8 injection site nodules and immunogenicity
3. [03]Holman RR et al. Effects of Once-Weekly Exenatide on Cardiovascular Outcomes in Type 2 Diabetes (EXSCEL; 14,752 adults; 11.4% vs 12.2%; HR 0.91, 95% CI 0.83–1.00; P=0.06 for superiority). N Engl J Med 377:1228–1239 (2017)
4. [04]Ahmann AJ et al. Efficacy and Safety of Once-Weekly Semaglutide Versus Exenatide ER in Subjects With Type 2 Diabetes (SUSTAIN 3; 813 adults, 56 weeks; HbA1c −0.9 vs −1.5; weight −1.9 vs −5.6 kg). Diabetes Care 41:258–266 (2018)
5. [05]Buse JB et al. Exenatide once weekly versus liraglutide once daily in patients with type 2 diabetes (DURATION-6; 911 adults; HbA1c −1.28 vs −1.48, non-inferiority not met). Lancet 381:117–124 (2013)
6. [06]Vijiaratnam N et al. Exenatide once a week versus placebo as a potential disease-modifying treatment for people with Parkinson's disease in the UK: a phase 3 randomised placebo-controlled trial (194 participants, 96 weeks, p=0.47). Lancet 405:627–636 — see also the Expression of Concern, Lancet 407(10548):2588, 27 June 2026 (2025)
7. [07]FDA – BYDUREON BCISE (exenatide extended-release) US prescribing information, revised 05/2025: boxed warning on thyroid C-cell tumours, contraindications, section 14.5 EXSCEL Table 10, Adverse Reactions 6.1 (2025)
8. [08]Drugs@FDA – NDA 209210 BYDUREON BCISE, marketing status Discontinued (NDA 021773 BYETTA and NDA 022200 BYDUREON likewise; ANDA 206697, Amneal generic exenatide, approved 19 November 2024)
9. [09]UniProt P26349 (EXE4_HELSU, exendin-4, Heloderma suspectum) — PEPTIDE 48..86 read HGEGTFTSDLSKQMEEEAVRLFIEWLKNGGPSSGAPPPS, MOD_RES 86 "Serine amide"; protein sequencing by Eng J et al., J Biol Chem 267:7402–7405 (1992)
10. [10]PubChem CID 45588096 – Exenatide, molecular formula C184H282N50O60S, MW 4187; synonym list gives the residue-by-residue name ending "-L-Serinamide" and the note "(Ser-39 = C-terminal amide)"
11. [11]WADA International Standard Prohibited List 2026 (in force 1 January 2026) – exenatide does not appear in any class; S4.4 covers insulins and insulin-mimetics only (2026)
12. [12]Medicine Supply Notification MSN/2025/054 (Department of Health and Social Care, 15 October 2025) – exenatide (Bydureon BCise) 2mg/0.85ml prolonged-release pens discontinued with supplies now exhausted; alternative GLP-1 RAs remain available (2025)
13. [13]DailyMed — BYETTA (exenatide) injection, US prescribing information, Adverse Reactions 6.1
14. [14]Rosenstock J et al. Effects of exenatide and lifestyle modification on body weight and glucose tolerance in obese subjects with and without pre-diabetes (152 adults, 24 weeks). Diabetes Care (2010)
15. [15]Dushay J et al. Short-term exenatide treatment leads to significant weight loss in a subset of obese women without diabetes (41 women, randomised placebo-controlled crossover). Diabetes Care (2012)
16. [16]Expression of Concern: Exenatide once a week versus placebo … for people with Parkinson's disease in the UK. Lancet 407(10548):2588 (2026)
17. [17]EXSCEL trial registration, NCT01144338 (completed 24 April 2017, results posted)
18. [18]Egan AG, Blind E, Dunder K et al. Pancreatic safety of incretin-based drugs — FDA and EMA assessment. N Engl J Med (2014)
19. [19]openFDA drug adverse event counts for exenatide, reaction terms by frequency (data last updated 30 July 2026) (2026)
20. [20]European Medicines Agency — Byetta, status: authorised, EMEA/H/C/000698
21. [21]DailyMed — EXENATIDE injection (Amneal Pharmaceuticals LLC), US prescribing information
22. [22]Drugs@FDA — NDA 021773 BYETTA and NDA 209210 BYDUREON BCISE, marketing status Discontinued
23. [23]European Medicines Agency — Bydureon, status: authorised, EMEA/H/C/002020
24. [24]Eng J, Kleinman WA, Singh L, Singh G, Raufman JP. Isolation and characterization of exendin-4, an exendin-3 analogue, from Heloderma suspectum venom. J Biol Chem (1992)