Does it prevent an actual heart attack?

What this archive is for. Trials of the compound come first, ranked by how many people were actually given it. Studies that only observed people are below them and labelled.

1. 01SemaglutidePeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 12 completed trials, 42,363 people given it.10 reported

   27,254 people given it2 trials1 found nothing

   • This is about as strong as evidence for a drug gets. Everyone taking part already had heart disease and carried extra weight, and none of them had diabetes. Half were given semaglutide, half a dummy injection, and everyone was followed for a bit over three years. Heart attacks, strokes and deaths from heart disease fell by a fifth. In plain counts that is about 15 fewer of those events for every 1,000 people treated.

     The numbers 17,604 adults with cardiovascular disease and BMI 27 or higher but no diabetes: semaglutide 2.4 mg cut major adverse cardiovascular events 20% (6.5% vs 8.0%; HR 0.80) over a mean 39.8 months.

     New England Journal of Medicine, 2023· 17,604 people

   • found nothingThe pill form of semaglutide also cut heart attacks and strokes, by a modest amount. 9,650 people with type 2 diabetes at high risk took a daily 14 mg tablet or a dummy tablet for about four years. Heart attacks, strokes and cardiac deaths hit 12.0% on the drug and 13.8% on the dummy, about 18 fewer such events for every 1,000 people. The kidney and other follow-up measures came out the same in both groups, so this result rests on the heart numbers alone.

     The numbers SOUL, 9,650 participants over a mean 47.5 months: major adverse cardiovascular events in 12.0% versus 13.8% (HR 0.86). The confirmatory secondary outcomes, including the kidney composite, did not differ significantly.

     New England Journal of Medicine, 2025· 9,650 people

2. 02TirzepatidePeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 9 completed trials, 22,802 people given it.11 reported

   16,025 people given it3 trials

   • Tirzepatide guarded the heart about as well as a drug already known to do it. More than 13,000 people with type 2 diabetes and hardened, narrowed arteries took either tirzepatide or dulaglutide. Heart attacks, strokes and deaths from heart disease happened to 12.2% of the tirzepatide group and 13.1% of the dulaglutide group. The trial was built to show tirzepatide was not worse, rather than to show it was better. It was not worse. That is what the 2026 heart licence rests on.

     The numbers 13,299 patients with type 2 diabetes and atherosclerotic disease: primary MACE in 12.2% on tirzepatide vs 13.1% on dulaglutide (HR 0.92), meeting noninferiority against an active comparator with proven benefit; basis of the August 2026 cardiovascular indication.

     New England Journal of Medicine, 2025· 13,299 people

   • Tirzepatide controlled blood sugar better than insulin and caused far fewer dangerous sugar lows. 1,995 people with type 2 diabetes and raised heart risk were treated for a year, everyone knowing which drug they got. Average blood sugar over three months fell about 2.5 points on tirzepatide and 1.44 on insulin glargine. Serious low blood sugar hit 6% to 9% on tirzepatide against 19% on insulin. Heart attacks and strokes leaned tirzepatide's way, on too few events to call.

     The numbers 1,995 treated participants: HbA1c change at 52 weeks -2.43% (10 mg) and -2.58% (15 mg) versus -1.44% on glargine, with severe or clinically significant hypoglycaemia in 6% to 9% versus 19%. Adjudicated four-point MACE hazard ratio 0.74 (95% CI 0.51 to 1.08).

     The Lancet, 2021· 1,995 people

   • Tirzepatide cut heart failure flare-ups in people with a stiff heart and obesity. 731 patients were followed for about two years on the drug or a dummy shot. Death from heart causes or worsening heart failure hit 9.9% on the drug and 15.3% on the dummy, about 54 fewer such events for every 1,000 people. Almost all of that came from fewer heart failure events. Deaths from heart causes alone ran slightly higher on the drug, 2.2% against 1.4%, on numbers too small to read either way.

     The numbers SUMMIT, 731 patients over a median 104 weeks: cardiovascular death or worsening heart failure in 9.9% versus 15.3% (HR 0.62). Cardiovascular deaths alone were 2.2% versus 1.4% (HR 1.58, 95% CI 0.52 to 4.83), so the composite was driven by heart-failure events.

     New England Journal of Medicine, 2025· 731 peoplepoints the wrong way

3. 03ExenatidePeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 15,088 people given it.1 reported

   14,752 people given it1 trial1 found nothing

   • found nothingWeekly exenatide proved safe for the heart but did not clearly protect it. 14,752 people with type 2 diabetes took weekly exenatide or a dummy injection for a bit over three years. 11.4% of the treated group had a heart attack, had a stroke, or died of heart disease. On the dummy, 12.2% did. The gap leaned toward the drug, but it is too small to separate from chance. A trial this large would have shown a clear benefit if there was one to find.

     The numbers 14,752 patients, median 3.2 years: MACE in 11.4% on weekly exenatide vs 12.2% on placebo (HR 0.91, 95% CI 0.83 to 1.00), noninferior for safety but not statistically superior.

     New England Journal of Medicine, 2017· 14,752 people

4. 04LiraglutidePeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 6 completed trials, 16,760 people given it.9 reported

   9,340 people given it1 trial

   • Liraglutide cut heart attacks, strokes and deaths from heart disease in people with diabetes. 9,340 adults with type 2 diabetes and a high risk of heart trouble got either liraglutide or a dummy injection, and were followed for close to four years. About one in eight fewer of those events happened on the drug. In plain counts that is roughly 19 fewer heart attacks, strokes and cardiac deaths for every 1,000 people treated. Everyone here already had diabetes, so it says nothing about healthy people.

     The numbers 9,340 patients with type 2 diabetes at high cardiovascular risk, median 3.8 years: primary MACE in 13.0% on liraglutide vs 14.9% on placebo (HR 0.87, a 13% relative reduction; P=0.01 for superiority).

     New England Journal of Medicine, 2016· 9,340 people

5. 05Relaxin-2PeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 7,706 people given it.5 reported

   7,706 people given it2 trials, 1 watch-only2 found nothing

   • found nothingThe big confirming trial was run and the drug failed it. 6,545 people admitted with sudden heart failure got the same two day drip or a dummy drip. Six months later the same share had died of heart causes in each group. Worsening heart failure in the first five days was no different. Deaths of any cause, readmissions and length of hospital stay were all unchanged. Side effects matched. A trial this large would have found a real benefit if there was one. Funded by Novartis.

     The numbers RELAX-AHF-2, the confirmatory trial, and it failed. 6,545 patients in the intention to treat analysis, same drug, dose, 48 hour infusion and entry criteria as RELAX-AHF. Death from cardiovascular causes at 180 days occurred in 285 of 3,274 patients (8.7 percent) on serelaxin and 290 of 3,271 (8.9 percent) on placebo, hazard ratio 0.98 (0.83 to 1.15), p = 0.77. Worsening heart failure at day 5 occurred in 6.9 percent against 7.7 percent, hazard ratio 0.89 (0.75 to 1.07), p = 0.19. No significant difference in all-cause death at 180 days, in cardiovascular death or rehospitalisation, or in length of the index hospital stay. Adverse event rates were similar. Funded by Novartis. Registered as NCT01870778.

     New England Journal of Medicine, 2019· 6,545 people

   • found nothingSerelaxin eased breathlessness a little in people admitted with sudden heart failure, and changed nothing that matters more. 1,161 people got either a two day drip of the drug or a dummy drip soon after reaching hospital. Breathlessness scored better on one scale and no better on the other. Deaths from heart causes and readmissions at two months were unchanged. A death count at six months leaned the drug's way, but that was a side question, not what the trial was built to prove. Funded by a Novartis company.

     The numbers 1,161 patients hospitalised with acute heart failure randomised 1:1 to a 48 hour infusion of serelaxin at 30 micrograms per kilogram per day or placebo within 16 hours of presentation. Serelaxin improved the visual analogue scale dyspnoea endpoint by 448 mm times hours (95 percent CI 120 to 775, p = 0.007) but had no effect on the co-primary Likert endpoint (27 percent against 26 percent, p = 0.70). There was no effect on cardiovascular death or readmission at 60 days (hazard ratio 1.02, 0.74 to 1.41) or on days alive out of hospital. Among prespecified additional endpoints, 42 deaths occurred by day 180 on serelaxin against 65 on placebo, hazard ratio 0.63 (0.42 to 0.93, p = 0.019). Funded by Corthera, a Novartis affiliate company. Registered as NCT00520806.

     The Lancet, 2013· 1,161 people

   • Investigators went back through every death in the failed trial to work out why the earlier trial had looked promising, and found nothing to rescue it. By six months about one patient in nine had died, and heart failure caused nearly four in ten of those deaths. The drug changed no category of death. People aged 75 and over died more often, and more often of something other than their heart. This explains the failure. It does not undo it.

     The numbers An adjudicated analysis of every death in RELAX-AHF-2, run specifically to understand why the two trials disagreed. By 180 days 11.5 percent of patients had died, 38 percent of those deaths from heart failure. Unlike in RELAX-AHF, there was no apparent effect of serelaxin on any category of cause of death. Patients aged 75 or over died more often (14.2 percent against 8.8 percent) and more often from non-cardiovascular causes. Patients with preserved ejection fraction had less heart failure mortality (30 percent against 40 percent) and more non-cardiovascular mortality (36 percent against 20 percent).

     JACC: Heart Failure, 2020· number of people not givenobserved only, nobody was given it

6. 06BNP and nesiritidePeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 3 completed trials, 7,149 people given it.4 reported

   7,141 people given it1 trial

   • This is the study that settled it, and the drug failed. 7,141 people in hospital with heart failure got nesiritide or a dummy drip on top of usual care. Breathlessness improved a shade more on the drug, not enough to count. Death or return to hospital within 30 days came out the same. Kidneys were no worse, but low blood pressure was more common. The maker paid for the study that ended its own product. A study this size would have found a real benefit.

     The numbers ASCEND-HF: 7,141 people hospitalised with acute heart failure randomised to nesiritide or placebo for 24 to 168 hours on top of standard care. Dyspnoea improvement 44.5 versus 42.1 percent at 6 hours and 68.2 versus 66.1 percent at 24 hours, neither meeting the prespecified threshold. Rehospitalisation or death within 30 days 9.4 versus 10.1 percent (difference -0.7 percentage points, 95 percent confidence interval -2.1 to 0.7). No excess worsening renal function; more hypotension. Funded by Scios, the manufacturer.

     New England Journal of Medicine, 2011· 7,141 peoplepoints the wrong way

7. 07LixisenatidePeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 6,068 people given it.2 reported

   6,068 people given it1 trial

   • This drug was tested for heart protection and gave none. Just over six thousand people with type 2 diabetes who had recently had a heart attack or unstable chest pain took lixisenatide or a dummy injection. Heart attacks, strokes and deaths from heart disease happened at the same rate in both groups. The trial was built to show the drug was not harmful to the heart, and it cleared that bar. It was large enough to have found a benefit, and there was none.

     The numbers 6,068 patients: primary MACE hazard ratio 1.02, showing noninferiority to placebo but no cardiovascular benefit.

     New England Journal of Medicine, 2015· 6,068 people

8. 08InsulinPeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 5 completed trials, 29,691 people given it.4 reported

   5,650 people given it2 trials

   • The benefit of early tight control outlasted the control itself. 4,209 people from the original trial were followed for 10 more years. Within a year of the trial ending, blood sugar in the two groups had evened out. Ten years later the intensively treated group still had fewer diabetes complications, fewer heart attacks and fewer deaths of any cause. Those treated with metformin did better still. Once the trial ended nobody was assigned to anything, so this part is observation rather than a test.

     The numbers UKPDS post-trial monitoring of 4,209 randomised participants. Glycaemic differences disappeared within a year of the trial ending, yet at 10 years the sulfonylurea-insulin group retained reductions in any diabetes-related endpoint (9 percent, p = 0.04), microvascular disease (24 percent, p = 0.001), myocardial infarction (15 percent, p = 0.01) and death from any cause (13 percent, p = 0.007). The metformin group showed larger reductions in myocardial infarction (33 percent) and death (27 percent).

     New England Journal of Medicine, 2008· 4,209 people

   • Tight blood sugar control early on protected hearts for decades afterwards. 1,441 people with type 1 diabetes got either tight or standard insulin treatment for about six and a half years, then were simply followed. Over about 17 years, the tight group had roughly 42 percent fewer heart events of any kind. Heart attacks, strokes and deaths from heart disease together fell by about 57 percent. The tight treatment had ended long before the benefit showed up.

     The numbers DCCT/EDIC: 1,441 people with type 1 diabetes randomised to intensive or conventional insulin therapy for a mean 6.5 years, then followed observationally. Over a mean 17 years, intensive treatment reduced any cardiovascular event by 42 percent (95 percent confidence interval 9 to 63, p = 0.02) and non-fatal myocardial infarction, stroke or cardiovascular death by 57 percent (12 to 79, p = 0.02).

     New England Journal of Medicine, 2005· 1,441 people

9. 09SemaxPeptideModerate. At least one finished study in people, and at least 100 people who took it. 4 completed trials, 354 people given it.9 reported

   327 people given it3 trials

   • The largest semax study in people has nothing to compare itself against. Doctors treated 187 patients with poor blood flow to the brain and followed them over time. The authors report that symptoms improved, that decline slowed, and that fewer strokes and mini strokes happened. Everybody in the study got semax. There was no dummy treatment and no untreated group, so there is no way to know what would have happened anyway. The work is Russian and nobody outside has repeated it.

     The numbers 187 patients with cerebrovascular insufficiency: the authors report clinical improvement, stabilisation of progression and a reduced rate of stroke and transient ischemic attack, with few side effects and good tolerability in older patients. No placebo group is described in the indexed abstract.

     Zhurnal Nevrologii i Psikhiatrii imeni S.S. Korsakova, 2005· 187 people

   • This is the study the stroke claim leans on, and it was not built to prove anything. 110 people recovering from a stroke were given courses of Semax. A brain growth protein in their blood rose, and their scores for everyday tasks improved faster. There was no dummy treatment to compare against, and nobody was kept in the dark about what they were getting. So there is no way to separate the drug from ordinary recovery after a stroke.

     The numbers Russian clinical study of 110 post-stroke patients: Semax courses raised plasma BDNF and were associated with faster Barthel index improvement; not a blinded placebo-controlled trial.

     Zhurnal Nevrologii i Psikhiatrii imeni S.S. Korsakova, 2018· 110 people

   • This is a real study in real stroke patients, and its design cannot size the effect. Thirty people were given semax in the first days after a stroke, alongside their usual hospital care. Eighty others with strokes of similar severity got the usual care alone. Nobody was picked by chance, and nobody was kept in the dark. The authors report that weakness and other deficits faded faster on semax. The work is Russian, and no Western group has repeated it.

     The numbers 30 patients treated in the acute period of hemispheric ischemic stroke against a control group of 80 patients matched for severity and location; faster regression of general cerebral and focal deficits reported, at 12 mg daily for moderate and 18 mg daily for severe strokes. Not randomized, not placebo-controlled.

     Zhurnal Nevrologii i Psikhiatrii imeni S.S. Korsakova, 1997· 30 people

10. 10ANPPeptideModerate. At least one finished study in people, and at least 100 people who took it. 4 completed trials, 188 people given it.4 reported

    184 people given it3 trials, 1 watch-only1 found nothing

    • This one points the other way, and the page keeps both. Two Japanese groups of patients were pooled, 2,435 in all, and sorted by how much carperitide they had been given. Over the following year, deaths from heart causes and from any cause were lower in the low-dose group than in those given none or almost none. This is watching what doctors already did. It cannot tell you what the drug caused.

      The numbers Pooled analysis of two Japanese cohorts, 2,435 patients with acute decompensated heart failure split by carperitide dose. Cardiovascular and all-cause mortality within 1 year were lower in the low-dose group (at or above 0.02 micrograms per kilogram per minute) than in the no-carperitide and very-low-dose groups. This observational result points in the opposite direction to the propensity-matched in-hospital analyses and is reported here alongside them, not instead of them.

      ESC Heart Failure, 2022· 2,435 peopleobserved only, nobody was given it

    • Head to head against a water tablet called tolvaptan, this drug came off slightly worse. 109 people in hospital with heart failure got either tolvaptan or carperitide. Symptoms and a blood test for heart strain improved about equally on both. Blood pressure ran lower on carperitide. More people on carperitide had to stop because their heart failure got worse or their blood pressure dropped. Just over a hundred people is a small study for a comparison like this.

      The numbers 109 hospitalised patients randomised to tolvaptan or carperitide. Symptoms and plasma BNP improved similarly in both groups. Blood pressure was lower after carperitide, and fewer adverse events such as worsening heart failure and hypotension requiring discontinuation occurred with tolvaptan (p = 0.027).

      Journal of Clinical Pharmacology, 2013· 109 peoplepoints the wrong way

    • This looked for who passes more water on the drug, and found two blood tests that predict it. 75 people in hospital with heart failure got a low dose of carperitide through a drip for six hours. Those who started with lower levels of their own natriuretic hormone, and lower vasopressin, produced the most urine. Blood pressure, kidney function and previous water tablets predicted nothing. Everyone here got the drug, so there is no comparison group and no way to know what would have happened anyway.

      The numbers 75 patients with acute heart failure given 0.0125 micrograms per kilogram per minute of carperitide for 6 hours. Lower baseline plasma ANP (r = -0.35, p = 0.002) and lower vasopressin (r = -0.54, p < 0.001) predicted a larger diuretic response; baseline blood pressure, renal function and prior loop diuretic use did not.

      ESC Heart Failure, 2022· 75 people

    • found nothingThis measured what the drug does to salts in the blood rather than whether it helps anyone. 162 people in hospital with heart failure were given carperitide through a drip. Sodium did not move over two days. Potassium drifted down slightly. Potassium falling too low was rare unless a strong water tablet was added on top, and then it was about four times more common. This is a look back at data already collected, so treat it as something to watch for rather than proof.

      The numbers Post-hoc analysis of a multicentre prospective cohort, 162 patients with acute heart failure given intravenous carperitide. Serum sodium was unchanged over 48 hours; serum potassium fell from 4.32 to 4.08 mEq/L with carperitide alone (p = 0.004). Hypokalaemia at 24 hours was uncommon unless 20 mg or more of furosemide was added (12.5 versus 2.8 percent, p = 0.039).

      Journal of Cardiology, 2021· number of people not given

11. 11Semax / SelankBlendModerate. At least one finished study in people, and at least 100 people who took it. 2 completed trials, 172 people given it.9 reported

    110 people given it1 trial

    • A Russian study of Semax after stroke, with a design too loose to trust. A hundred and ten people recovering from a stroke were given courses of the drug. A growth factor in their blood rose, and they regained everyday abilities faster, things like washing, dressing and moving about. Nobody was kept in the dark about what they were getting and there was no dummy group. That matters most for a recovery score, which improves on its own anyway. No Selank was given.

      The numbers Russian clinical study of 110 post-stroke patients. Semax courses raised plasma BDNF and were associated with faster Barthel index improvement. This was not a blinded placebo-controlled trial, and no Selank was given.

      Zhurnal Nevrologii i Psikhiatrii imeni S.S. Korsakova, 2018· 110 people

12. 12ElamipretidePeptideModerate. At least one finished study in people, and at least 100 people who took it. 5 completed trials, 487 people given it.6 reported

    71 people given it1 trial1 found nothing

    • found nothingElamipretide was tested in failing hearts and it changed nothing. Seventy one people with a weak heart pump got a dummy injection or one of two elamipretide doses, daily for four weeks. Ultrasound scans of how much blood the heart held and how much it pushed out looked the same in all three groups. The trial was small and short, so it could only have caught a fairly large effect. It did not find one.

      The numbers 71 patients randomized to placebo, 4 mg or 40 mg daily for 28 days: no significant change in left ventricular end systolic volume or ejection fraction versus placebo (4 mg p = 0.90, 40 mg p = 0.28). Tolerability was similar across the three arms.

      Journal of Cardiac Failure, 2020· 71 people

13. 13ApelinPeptideLimited. People have taken it in a published study, but fewer than 100 of them. 3 completed trials, 68 people given it.3 reported

    44 people given it2 trials

    • The only drug company trial of an apelin receptor activator in patients measured safety, not benefit. Novartis gave 26 people with stable heart failure either CLR325 or a dummy infusion. What it was built to count was side effects, serious side effects and deaths. It finished in January 2019 and posted results in March 2020, and those results are safety and blood level figures. Nothing posted says whether anybody's heart failure improved.

      The numbers Novartis, phase 2, 26 enrolled, APJ receptor agonist CLR325 versus placebo in chronic stable heart failure, primary outcome adverse events, serious adverse events and death. Completed January 2019, results posted March 2020. The only industry trial of an APJ agonist in patients that has reported.

      ClinicalTrials.gov, 2016· 26 people

    • In people with failing hearts, apelin still widened blood vessels and improved pumping while it was being infused. 18 patients with moderate heart failure, 6 more during heart imaging and 26 healthy volunteers took part. Arm vessel widening was preserved in heart failure, while the usual comparison drug had weakened. Apelin into the heart arteries raised blood flow and pumping force and lowered filling pressures. These are short infusions and machine readings, not months of feeling better. Public money paid for it.

      The numbers 18 patients with NYHA II to III heart failure, 6 at coronary angiography and 26 healthy volunteers in randomised double-blind placebo-controlled studies. Forearm vasodilatation to apelin was preserved in heart failure (p 0.3) while the acetylcholine response was attenuated (p 0.01); intracoronary apelin-36 increased coronary flow and left ventricular dP/dt max and lowered filling pressures; systemic (Pyr1)apelin-13 at 30 to 300 nmol/min raised cardiac index and lowered mean arterial pressure and vascular resistance in both groups, raising heart rate only in controls. Non-US government funded.

      Circulation, 2010· 18 people

14. 14ArgirelinePeptideModerate. At least one finished study in people, and at least 100 people who took it. 4 completed trials, 155 people given it.7 reported

    24 people given it1 trial1 found nothing

    • found nothingThis is the one test of the peptide in a medical condition, and it came up short. 24 people with uncontrollable eyelid spasms, already getting botulinum toxin every three months, had either acetyl hexapeptide-8 cream or a dummy cream added. Nobody involved knew which. Relief lasted about 3.7 months with the peptide and 3.0 without, too small a lean to count. Four of twelve on the active cream did hold out longer. No serious side effects appeared. Public money paid for this, not an ingredient maker.

      The numbers Double-blind placebo-controlled randomized trial in 24 patients with blepharospasm on regular three-monthly botulinum toxin. Time to return to baseline Jankovic score was 3.7 months on topical acetyl hexapeptide-8 against 3.0 on placebo, reported as a trend rather than a significant result; 4 of 12 in the active group had symptom control extended to 3.3 to 7.1 months. No significant adverse events. Sponsored by the National Institute of Neurological Disorders and Stroke, not by an ingredient manufacturer.

      European Journal of Neurology, 2013· 24 people

15. 15Angiotensin-(1-7)PeptideLimited. People have taken it in a published study, but fewer than 100 of them. 5 completed trials, 76 people given it.3 reported

    18 people given it1 trial

    • The first test in people was about safety, and it turned up serious harm at the top dose. 18 adults with advanced cancer got the peptide as a shot under the skin, daily for five days, with the dose raised group by group. At the highest dose one person had a severe stroke and another lost nerve function in the head, which recovered. One tumour shrank and three more stopped growing for over three months. These were very ill people and there was no comparison group.

      The numbers 18 adults with advanced solid tumours received escalating subcutaneous doses daily for 5 days on a 3 week cycle. Dose-limiting toxicities at 700 micrograms per kilogram were a grade 4 stroke and a grade 3 reversible cranial neuropathy. One patient had a 19 percent tumour reduction and 3 more had disease stabilisation beyond 3 months. Recommended phase 2 dose 400 micrograms per kilogram.

      Clinical Cancer Research, 2009· 18 peoplepoints the wrong way

16. 16HexarelinPeptideModerate. At least one finished study in people, and at least 100 people who took it. 6 completed trials, 105 people given it.7 reported

    0 people given it1 trial

    • A single dose helped one kind of failing heart and not the other. Patients with badly weakened hearts were given hexarelin into a vein. In those whose damage came from blocked arteries, the heart pumped out more blood with each beat. In those whose heart muscle was enlarged for other reasons, nothing changed. Both groups released the same growth hormone, so the heart effect was not the hormone. This was a small study of one dose, with nothing measured about how people felt.

      The numbers Acute IV hexarelin raised ejection fraction in ischemic cardiomyopathy patients but not in dilated cardiomyopathy, despite similar GH release; small study.

      European Journal of Heart Failure, 2002· number of people not given

17. 17DegarelixPeptideModerate. At least one finished study in people, and at least 100 people who took it. 2 completed trials, 610 people given it.3 reported

    0 people given it1 trial

    • This trial was meant to settle whether degarelix is kinder to the heart than the older drug, and it stopped before it could. 545 men with prostate cancer and known artery disease were enrolled, out of 900 planned. Over a year, 15 men on degarelix and 11 on leuprolide had a heart attack, stroke or cardiac death. Those counts are far too small to separate the two drugs. The authors say the question is still open. Do not read this as proof that the two are equally safe.

      The numbers Randomised, open-label, in men with prostate cancer and known atherosclerotic cardiovascular disease. Enrolment stopped early at 545 of a planned 900 patients across 113 sites in 12 countries. Major adverse cardiovascular events at 12 months: 15 (5.5 percent) on degarelix versus 11 (4.1 percent) on leuprolide, hazard ratio 1.28 (95 percent CI 0.59 to 2.79), P equal to 0.53. The authors conclude that the relative cardiovascular safety of GnRH antagonists and agonists remains unresolved.

      Circulation, 2021· number of people not given

18. 18BradykininPeptideLimited. People have taken it in a published study, but fewer than 100 of them. 2 completed trials, 47 people given it.2 reported

    0 people given it1 trial

    • This showed that bradykinin is part of how blood pressure tablets actually work. Researchers measured blood flow in the forearm while blocking the bradykinin receptor. In 8 healthy volunteers, blocking it stopped the response to injected bradykinin but left resting blood flow alone. In 17 people with heart failure taking ACE inhibitors, blocking it tightened the vessels. That effect vanished when the tablet was stopped and came back when it was restarted. These are small groups, and a forearm is not an outcome.

      The numbers Forearm plethysmography studies: in 8 healthy volunteers the antagonist B9340 blocked vasodilation to infused bradykinin without changing resting blood flow. In 17 patients with NYHA class II to IV heart failure on chronic ACE inhibitors, B9340 caused dose-dependent vasoconstriction (p = 0.01), which disappeared after the ACE inhibitor was withdrawn and returned when it was restarted.

      Circulation, 2001· number of people not given

19. 19RetatrutidePeptideStrong. At least two finished studies in people, and at least 1,000 people who took it. 5 completed trials, 1,345 people given it.8 reported

    nobody was given it1 watch-only study

    • This trial finished and has published nothing either. 1,946 people with severe obesity who already had heart disease took part, and the registry says it completed in May 2026. No results have been posted. Note what it was built to measure, which is weight, in people who happen to have heart disease. It was never designed to show whether retatrutide prevents heart attacks or strokes.

      The numbers Registry record: 1,946 participants with severe obesity and established cardiovascular disease, listed as completed in May 2026 with no results posted. This is a weight trial in a cardiovascular population, not a cardiovascular outcome trial.

      ClinicalTrials.gov NCT05882045 (TRIUMPH-3), Eli Lilly, 2026· number of people not givenobserved only, nobody was given it

Vitamins, minerals, amino acids and botanicals sold over the counter, held to the same standard.

1. 01BetaineSupplementModerate. At least one finished study in people, and at least 100 people who took it. 3 completed trials, 170 people given it.4 reported

   0 people given it1 trial

   • The marker moved and nothing happened to the people. 12,064 heart attack survivors took folic acid and vitamin B12 daily, or a dummy, for 6.7 years, nobody knowing which. Homocysteine fell 28 percent on average. Major vascular events struck 25.5 percent of the treated group and 24.8 percent of the dummy group. A trial this size over nearly seven years would have found a real benefit. This tested B vitamins, not betaine, which is the point. It tests the idea, not the product.

     The numbers 12,064 myocardial infarction survivors randomised double blind to 2 mg folic acid plus 1 mg vitamin B12 daily or placebo, followed 6.7 years. Homocysteine fell by a mean 3.8 micromol/L, 28 percent. Major vascular events occurred in 1,537 of 6,033 on treatment (25.5 percent) against 1,493 of 6,031 on placebo (24.8 percent), risk ratio 1.04 (95 percent CI 0.97 to 1.12, p = 0.28). The marker moved and the outcome did not. This trial used B vitamins rather than betaine, which is the point: it tests the premise, not the product.

     JAMA, 2010· number of people not given

2. 02Omega-3SupplementStrong. At least two finished studies in people, and at least 1,000 people who took it. 3 completed trials, 34,050 people given it.4 reported

   0 people given it1 trial

   • The same idea was tested against a different comparison oil and it did not work. A high dose of omega-3 was given to a similar group of high-risk patients already taking a statin. The comparison group got corn oil instead of mineral oil. There was no reduction in heart attacks, strokes or cardiac deaths. The trial was stopped early because it was clear it would not show a benefit. Put beside the trial above, this is why the fish oil question is still argued over.

     The numbers The null half. A high-dose omega-3 against a corn oil placebo in a comparable high-risk statin-treated population produced no reduction in major adverse cardiovascular events. The trial was stopped for futility.

     JAMA, 2020· number of people not given

3. 03NiacinSupplementStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 29,087 people given it.7 reported

   nobody was given it2 watch-only studies1 found nothing

   • found nothingThis is the study niacin's reputation rests on, and it is weaker than the reputation. 8,341 men who had already had a heart attack were treated between 1966 and 1975, long before statins existed. During the trial, niacin slightly cut repeat heart attacks and did not cut deaths at all. Fifteen years later, almost nine years after everyone had stopped, about 52 in every hundred niacin men had died against 58 on the dummy. That gap was found by watching afterwards, not by the trial itself.

     The numbers The historical result that kept niacin in the guidelines for thirty years, and it is weaker than its reputation. In the Coronary Drug Project, 8,341 men aged 30 to 64 with a previous myocardial infarction were treated between 1966 and 1975. Niacin modestly reduced definite non-fatal recurrent myocardial infarction but did not reduce total mortality during the trial. At a mean follow-up of 15 years, nearly 9 years after the drug was stopped, all-cause mortality was 52.0 percent on niacin versus 58.2 percent on placebo, 11 percent lower, P=0.0004. This is a post-trial observational mortality difference in a pre-statin population, not a result obtained on top of modern therapy, and both AIM-HIGH and HPS2-THRIVE tested niacin on top of modern therapy and found nothing.

     Journal of the American College of Cardiology, 1986· 8,341 peopleobserved only, nobody was given it

   • A waste product of excess niacin is linked to more heart trouble, which would explain why taking more does not help. Researchers measured blood chemicals in over four thousand heart patients across three separate groups. People carrying the most of two niacin waste products had roughly twice the risk of a heart attack, stroke or cardiac death over three years. A gene tied to those chemicals also tracked with a marker of inflamed blood vessels. Nobody was given anything here, so this is a strong suspicion and not proof.

     The numbers Observational metabolomics rather than a trial, and it proposes a mechanism for why more niacin might be actively unhelpful. In a discovery cohort of 1,162 stable cardiac patients and validation cohorts of 2,331 US and 832 European patients, serum levels of the terminal metabolites of excess niacin, 2PY and 4PY, were associated with increased 3-year risk of major adverse cardiovascular events: for 4PY, adjusted hazard ratios of 1.89 (95 percent CI 1.26 to 2.84) and 1.99 (1.26 to 3.14). A genetic variant associated with both metabolites was also associated with soluble VCAM-1 in a meta-analysis of about 106,000 people. This is association plus a mechanistic hypothesis, not causal proof in humans.

     Nature Medicine, 2024· 1,162 peopleobserved only, nobody was given itpoints the wrong way

4. 04CoQ10SupplementModerate. At least one finished study in people, and at least 100 people who took it. 2 completed trials, 162 people given it.4 reported

   nobody was given it1 watch-only study

   • Coenzyme Q10 cut deaths and hospital admissions in people with serious heart failure. Four hundred and twenty patients took it three times a day, or a dummy, for two years. Nobody involved knew which. Major heart events struck 15 percent of the treated group against 26 percent on the dummy, and about half as many died. The first sixteen weeks showed nothing at all, so the benefit only appeared with time. This is one trial of moderate size and it has not been repeated at this scale.

     The numbers 420 patients with moderate to severe chronic heart failure, 100 mg three times daily, two years. Major adverse cardiovascular events 15% against 26%, hazard ratio 0.50 (95% CI 0.32 to 0.80, p = 0.003). Cardiovascular mortality 9% against 16% (p = 0.026) and all-cause mortality 10% against 18% (p = 0.018). The short-term 16-week endpoints were null.

     JACC Heart Failure, 2014· 420 peopleobserved only, nobody was given it

5. 05FolateSupplementPreliminary. Studied in people, but no published study has had people take it. 2 human studies where nobody was given it.2 reported

   nobody was given it2 watch-only studies2 found nothing

   • found nothingLowering homocysteine did not protect the heart. People with heart and blood vessel disease took folic acid, B6 and B12, and their homocysteine dropped. Deaths, heart attacks and strokes did not.

     The numbers HOPE-2. Folic acid plus B6 plus B12 lowered homocysteine. The primary composite of death, myocardial infarction, and stroke did not improve.

     New England Journal of Medicine, 2006· number of people not givenobserved only, nobody was given it

   • found nothingThe same result in people who had just had a heart attack. B vitamins lowered homocysteine but did not cut further heart problems, and they raised a worry about harm.

     The numbers NORVIT. After myocardial infarction, B-vitamin homocysteine lowering did not reduce events and raised concern about harm. Same week as HOPE-2.

     New England Journal of Medicine, 2006· number of people not givenobserved only, nobody was given it

Approved drugs, biologics and hormones that have been measured for this. They are here because leaving them out would misrepresent the evidence, and they are below the peptides because that is not what most people come here to read.

1. 01EmpagliflozinSmall moleculeStrong. At least two finished studies in people, and at least 1,000 people who took it. 3 completed trials, 16,738 people given it.4 reported

   16,738 people given it3 trials, 1 watch-only1 found nothing

   • found nothingA diabetes drug that turned out to keep people alive. 7,020 people with type 2 diabetes and high heart risk took empagliflozin or a dummy for about three years. 12.1 percent of the dummy group had a heart attack, had a stroke, or died of heart disease. On empagliflozin, 10.5 percent did. Deaths from any cause were 8.3 percent against 5.7 percent. Heart attacks and strokes alone did not change, so the gain came from fewer deaths. Genital infections were more common on the drug.

     The numbers EMPA-REG OUTCOME. 7,020 people with type 2 diabetes at high cardiovascular risk, median observation 3.1 years. The primary composite of cardiovascular death, nonfatal myocardial infarction or nonfatal stroke occurred in 10.5 percent on empagliflozin against 12.1 percent on placebo (hazard ratio 0.86, P = 0.04). Death from any cause was 5.7 percent against 8.3 percent, a 32 percent relative risk reduction. There was no significant difference in rates of myocardial infarction or stroke, and genital infection was more common on the drug.

     New England Journal of Medicine, 2015· 7,020 people

   • It slowed kidney disease down, and it did not stop people dying. 6,609 people with chronic kidney disease were followed for two years. Their kidney disease got worse, or they died of heart disease, in 16.9 percent of the dummy group against 13.1 percent on empagliflozin. Deaths from any cause were 5.1 percent against 4.5 percent, a gap too small to separate from chance. Anyone treating this drug as something that makes people live longer should stop at that last sentence.

     The numbers EMPA-KIDNEY. 6,609 people with chronic kidney disease, median 2.0 years. Kidney disease progression or cardiovascular death occurred in 13.1 percent against 16.9 percent (hazard ratio 0.72). Notably for anyone reading the drug as a mortality intervention, death from any cause was 4.5 percent against 5.1 percent and was not a significant difference in this trial.

     New England Journal of Medicine, 2023· 6,609 peopleobserved only, nobody was given it

   • It helped the harder kind of heart failure, mostly by keeping people out of hospital. 5,988 people whose hearts still pumped reasonably well but who had heart failure symptoms were followed for a bit over two years. 17.1 percent of the dummy group died of heart disease or were admitted for heart failure. On empagliflozin, 13.8 percent were. The authors say the difference came mainly from fewer hospital admissions rather than from fewer deaths. That distinction matters to anyone reading this as a life-extending drug.

     The numbers EMPEROR-Preserved. 5,988 people with class II to IV heart failure and an ejection fraction above 40 percent, median 26.2 months. Cardiovascular death or heart failure hospitalisation occurred in 13.8 percent against 17.1 percent (hazard ratio 0.79). The authors state the effect was mainly driven by fewer heart failure hospitalisations rather than fewer deaths.

     New England Journal of Medicine, 2021· 5,988 people

   • It worked in heart failure as well, and it worked whether or not people had diabetes. 3,730 people with heart failure and a weakened pumping heart were followed for about 16 months. 24.7 percent of the dummy group died of heart disease or went into hospital for heart failure. On empagliflozin, 19.4 percent did. Kidney function declined about four times more slowly on the drug. These were people who already had heart failure, not healthy people taking it to prevent anything.

     The numbers EMPEROR-Reduced. 3,730 people with class II to IV heart failure and an ejection fraction of 40 percent or less, median 16 months. Cardiovascular death or heart failure hospitalisation occurred in 19.4 percent against 24.7 percent on placebo (hazard ratio 0.75). Estimated glomerular filtration rate fell by 0.55 against 2.28 ml per minute per 1.73 square metres per year. The effect held regardless of diabetes status.

     New England Journal of Medicine, 2020· 3,730 people

2. 02DulaglutideBiologicStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 10,500 people given it.1 reported

   9,901 people given it1 trial

   • Dulaglutide lowered heart attacks and strokes, and most people in the trial had no heart disease to start with. 9,901 adults with type 2 diabetes took dulaglutide or a dummy injection for about five and a half years. 12.0% on the drug had a heart attack, had a stroke, or died of heart disease, against 13.4% on the dummy. That is roughly 14 fewer such events for every 1,000 people. The benefit per person is small, and it took years to show up.

     The numbers 9,901 participants (only about a third with prior cardiovascular disease), median 5.4 years: MACE in 12.0% on dulaglutide vs 13.4% on placebo (HR 0.88; p=0.026).

     The Lancet, 2019· 9,901 people

3. 03AlbiglutideBiologicStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 10,304 people given it.

   9,463 people given it1 trial

   • Albiglutide cut heart attacks, strokes and cardiac deaths in people with diabetes and heart disease. Nearly ten thousand adults across 28 countries were split by chance between the drug and a dummy injection, and followed for about a year and a half. Nobody involved knew which. Seven in a hundred on the drug had a heart attack, stroke or cardiac death, against nine in a hundred on the dummy. Pancreas and thyroid problems were rare and evenly spread. GlaxoSmithKline funded it.

     The numbers 9,463 adults aged 40 and over with type 2 diabetes and cardiovascular disease, 610 sites, 28 countries, randomised 1:1 to albiglutide 30 to 50 mg weekly or placebo, median follow-up 1.6 years. Primary composite (cardiovascular death, myocardial infarction, stroke): 338 of 4,731 (7 percent) against 428 of 4,732 (9 percent), hazard ratio 0.78 (95 percent CI 0.68 to 0.90), superiority P equals 0.0006. Pancreatitis 10 against 7, pancreatic cancer 6 against 5, medullary thyroid carcinoma 0 against 0. Funded by GlaxoSmithKline.

     Lancet, 2018· 9,463 people

4. 04PioglitazoneSmall moleculeStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 9,114 people given it.5 reported

   9,114 people given it2 trials1 found nothing

   • found nothingThe big heart outcomes trial of pioglitazone in diabetes missed its main target. A wide measure of heart and artery events was not clearly lower. A narrower measure of deaths, heart attacks and strokes was lower. More people on the drug went to hospital with heart failure. These were people with diabetes and existing artery disease.

     The numbers 5,238 patients with type 2 diabetes and macrovascular disease randomised to oral pioglitazone titrated from 15 to 45 mg or matching placebo on top of usual treatment, mean follow-up 34.5 months. Missed its primary endpoint: the broad composite occurred in 514 versus 572 patients (HR 0.90, 95% CI 0.80 to 1.02, p=0.095). The main secondary composite of death, non-fatal myocardial infarction and stroke occurred in 301 versus 358 (HR 0.84, p=0.027). Hospital admission for heart failure 6% versus 4%.

     The Lancet, 2005· 5,238 people

   • In people with insulin resistance but no diabetes who had just had a stroke or a mini-stroke, pioglitazone lowered the chance of another stroke or a heart attack over about five years. It also halved new diabetes. It cost more weight gain, swelling and broken bones. Death rates were about the same.

     The numbers 3,876 patients without diabetes, with insulin resistance and a recent ischaemic stroke or TIA, randomised to pioglitazone (target 45 mg daily) or placebo, median 4.8 years (IRIS). Stroke or myocardial infarction in 9.0% versus 11.8% (HR 0.76, 95% CI 0.62 to 0.93, p=0.007). Diabetes in 3.8% versus 7.7% (HR 0.48). All-cause mortality similar (HR 0.93). Weight gain over 4.5 kg 52.2% versus 33.7%, oedema 35.6% versus 24.9%, fracture needing surgery or hospitalisation 5.1% versus 3.2%.

     The New England Journal of Medicine, 2016· 3,876 people

5. 05Cardiac glycosidesSmall moleculeModerate. At least one finished study in people, and at least 100 people who took it. 1 completed trial, 6,800 people given it.

   6,800 people given it1 trial1 found nothing

   • found nothingDigoxin did not save lives, and it was tested well enough to say so. 6,800 people with weak hearts took digoxin or a dummy pill and were followed about three years on average. About the same share died in each group, roughly 35 percent either way. A study this big would have found a survival benefit if there was one. Digoxin did keep people out of hospital: heart failure stays fell from about 35 percent to 27 percent. This is heart failure proof. It says nothing about ageing.

     The numbers 6,800 patients with heart failure and ejection fraction 0.45 or less randomised to digoxin or placebo, average follow-up 37 months. Mortality unchanged: 1,181 deaths (34.8 percent) versus 1,194 (35.1 percent), risk ratio 0.99, 95 percent confidence interval 0.91 to 1.07. Hospitalisation for worsening heart failure fell from 34.7 to 26.8 percent, risk ratio 0.72. This is cardiology evidence and says nothing about senescence.

     New England Journal of Medicine, 1997· 6,800 people

6. 06TestosteroneHormoneModerate. At least one finished study in people, and at least 100 people who took it. 1 completed trial, 5,246 people given it.10 reported

   5,246 people given it1 trial

   • Testosterone did not raise the risk of heart attacks or strokes in the men most likely to have one. More than 5,000 men aged 45 to 80 took part. All had low testosterone with symptoms, and all were already at risk of heart trouble. Half rubbed on a testosterone gel and half a dummy gel. After about three years, heart deaths, heart attacks and strokes were about as common in each group. This was a gel held to a target blood level, not a home shot.

     The numbers TRAVERSE. 5,246 men aged 45 to 80 with preexisting or high risk of cardiovascular disease, symptoms of hypogonadism and two fasting testosterone levels below 300 ng/dL, randomised double-blind to transdermal testosterone gel titrated to 350 to 750 ng/dL or placebo. Mean treatment 21.7 months, mean follow-up 33.0 months. Primary composite of cardiovascular death, non-fatal myocardial infarction or non-fatal stroke occurred in 182 (7.0%) on testosterone and 190 (7.3%) on placebo, hazard ratio 0.96 (95% CI 0.78 to 1.17), p < 0.001 for non-inferiority.

     New England Journal of Medicine, 2023· 5,246 people

7. 07Testosterone cypionateHormoneModerate. At least one finished study in people, and at least 100 people who took it. 1 completed trial, 5,246 people given it.9 reported

   5,246 people given it1 trial

   • The one large safety study found no extra heart attacks, strokes or heart deaths on testosterone. More than 5,000 men took part, all already at risk of heart trouble. It is here because it is the best safety proof any testosterone product has. Note what it tested. Men rubbed on a gel that doctors adjusted to a set blood level, not a shot like this one.

     The numbers TRAVERSE, 5,246 men, no excess of cardiovascular death, heart attack or stroke against placebo. Included because it is the safety context for any testosterone ester, with the qualifier that TRAVERSE used a transdermal gel titrated to a defined range rather than an injectable.

     New England Journal of Medicine, 2023· 5,246 people

8. 08SimvastatinSmall moleculeStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 24,980 people given it.6 reported

   4,444 people given it1 trial

   • Lowering cholesterol with a statin kept people alive, and this is the trial that showed it. 4,444 people who already had chest pain or a previous heart attack were split by chance into two groups. One took simvastatin, the other a dummy pill, and both were followed for over five years. By the end, 12 percent of the dummy group had died against 8 percent on the drug. Deaths from heart disease fell from 189 to 111. Everyone here already had heart disease, so this says nothing about healthy people.

     The numbers 4,444 people with angina or previous myocardial infarction, randomised double-blind, median follow-up 5.4 years. 256 placebo patients (12 percent) died against 182 on simvastatin (8 percent), relative risk of death 0.70 (95 percent CI 0.58 to 0.85, p equals 0.0003). Coronary deaths fell from 189 to 111. Major coronary events occurred in 28 percent on placebo and 19 percent on simvastatin.

     Lancet, 1994· 4,444 people

   What people using it report

   • The threads split sharply and stay split. One side posts trial evidence that it lowers the risk of heart attack and stroke, the other posts personal accounts of harm and questions the industry behind the data. Neither side moves, and that split describes the discussion better than any conclusion drawn inside it.
9. 09EfpeglenatideBiologicStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 4,076 people given it.3 reported

   4,076 people given it1 trial

   • This one did protect the heart. Just over four thousand people with type 2 diabetes and a high risk of heart trouble took efpeglenatide or a dummy injection. Heart attacks, strokes and deaths from heart disease happened about a quarter less often on the drug. Kidney problems were down by roughly a third. Two results pointing the same way are harder to dismiss than one. Everybody here was picked for being at high risk, so healthier people should not expect the same size of gain.

     The numbers 4,076 high-risk patients: MACE hazard ratio 0.73 and composite renal outcome hazard ratio 0.68 versus placebo.

     New England Journal of Medicine, 2021· 4,076 people

10. 10ErythropoietinBiologicStrong. At least two finished studies in people, and at least 1,000 people who took it. 3 completed trials, 4,641 people given it.6 reported

    4,038 people given it2 trials

    • This drug did not extend life or save kidneys, and it doubled the risk of stroke. 4,038 people with type 2 diabetes, kidney disease and anaemia got darbepoetin alfa or a dummy shot, with rescue treatment if counts dropped too low. Death and heart events came out the same. Kidney failure came out the same. Strokes hit 101 people on the drug against 53 on the dummy. Fewer needed transfusions, and tiredness improved a little. Amgen, which sells it, paid for the study.

      The numbers TREAT: 4,038 people with type 2 diabetes, chronic kidney disease and anaemia randomised to darbepoetin alfa targeting 13 g/dL or to placebo with rescue below 9 g/dL. Death or a cardiovascular event, hazard ratio 1.05 (0.94 to 1.17); death or end-stage renal disease, hazard ratio 1.06 (0.95 to 1.19). Fatal or non-fatal stroke 101 versus 53, hazard ratio 1.92, 95 percent confidence interval 1.38 to 2.68, p < 0.001. Fewer transfusions (297 versus 496) and only a modest improvement in fatigue. Sponsor Amgen.

      New England Journal of Medicine, 2009· 4,038 people

    • Pushing blood counts up to normal in kidney disease harmed people. 1,432 adults with long-term kidney disease were treated towards either a high or a lower blood count target, for about 16 months. Deaths, heart attacks, heart failure hospital stays and strokes were more common in the high target group, 125 events against 97. Quality of life did not improve. Everybody knew which target they were on. The study was stopped, and the company behind the drug ran it.

      The numbers CHOIR: 1,432 people with chronic kidney disease randomised open-label to a haemoglobin target of 13.5 or 11.3 g/dL, median 16 months. The composite of death, myocardial infarction, heart failure hospitalisation and stroke occurred 125 times in the high-target group and 97 times in the low-target group, hazard ratio 1.34, 95 percent confidence interval 1.03 to 1.74, p = 0.03, with no improvement in quality of life. Registry status TERMINATED; sponsor Johnson and Johnson Pharmaceutical Research and Development.

      New England Journal of Medicine, 2006· number of people not given

11. 11AICARSmall moleculeModerate. At least one finished study in people, and at least 100 people who took it. 1 completed trial, 3,080 people given it.5 reported

    3,080 people given it1 trial

    • It was tested properly, in thousands of patients, and it did not work. 3,080 people having heart bypass surgery got acadesine or a dummy at 300 hospitals. The trial was stopped early because it was going nowhere. Death, stroke or serious heart weakness hit about one in twenty people in each group. An earlier pooling of five small trials had looked good. This trial is why the drug went no further.

      The numbers 3,080 patients randomised at 300 sites before the trial was stopped for futility. The composite of death, non-fatal stroke or severe left ventricular dysfunction through day 28 occurred in 5.1% on acadesine against 5.0% on placebo (odds ratio 1.01). An earlier 1997 meta-analysis of five smaller trials had been positive; this trial is why the drug went no further.

      JAMA, 2012· 3,080 people

12. 12EnalaprilSmall moleculeStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 2,822 people given it.5 reported

    2,822 people given it2 trials, 1 watch-only1 found nothing

    • found nothingIn people with no symptoms, enalapril did not clearly save lives. 4,228 people with a weak pumping heart but no symptoms took enalapril or a dummy pill for about three years. 313 died on enalapril against 334 on the dummy pill, a gap small enough to be chance. A study this size would likely have found a real survival gain if there was one. It did delay heart failure. For anyone reading enalapril as a longevity drug, this is the result that matters, and it came up empty.

      The numbers 4,228 asymptomatic people with ejection fraction 0.35 or below, mean follow-up 37.4 months. Total mortality 313 versus 334 deaths, risk reduction 8 percent (95 percent CI -8 to 21, p = 0.30), not significant. Death or development of heart failure 630 versus 818 (29 percent reduction, p < 0.001). The null mortality result in people without symptoms is the relevant one for anyone reading enalapril as a longevity drug.

      New England Journal of Medicine, 1992· 4,228 peopleobserved only, nobody was given it

    • In people with symptoms, enalapril kept more of them alive. 2,569 patients with heart failure and a weak pumping heart took enalapril or a dummy for about three and a half years on average. 452 died on enalapril against 510 on the dummy, roughly a sixth fewer deaths. Hospital admissions for heart failure fell further still. A United States government institute ran it, not a drug company. A large trial, and a real but moderate benefit.

      The numbers 2,569 patients with symptomatic heart failure and ejection fraction 0.35 or below, enalapril 2.5 to 20 mg a day or placebo, mean follow-up 41.4 months. Deaths 452 (35.2 percent) versus 510 (39.7 percent), risk reduction 16 percent (95 percent CI 5 to 26, p = 0.0036); death or heart failure hospitalisation 613 versus 736 (26 percent reduction). Run by the US National Heart, Lung, and Blood Institute.

      New England Journal of Medicine, 1991· 2,569 people

    • This is one of the clearest survival gains ever shown for a pill. 253 patients with the worst kind of heart failure got enalapril or a dummy pill on top of their usual treatment, and nobody involved knew which. At six months, 26% of the enalapril group had died against 44% on the dummy pill. Every life saved came from slowing the heart failure itself, not from preventing sudden death. Seven people stopped for low blood pressure, and none on the dummy pill.

      The numbers 253 patients with NYHA class IV heart failure, double-blind, enalapril 2.5 to 40 mg a day or placebo on top of conventional therapy. Six-month mortality 26 versus 44 percent (40 percent reduction, p = 0.002); one-year 31 percent reduction; 50 versus 68 deaths overall (27 percent). Entire benefit in progressive heart failure deaths, none in sudden death. Hypotension withdrawals 7 versus 0. Funding not stated in the abstract retrieved.

      New England Journal of Medicine, 1987· 253 people

13. 13Zoledronic acidSmall moleculeStrong. At least two finished studies in people, and at least 1,000 people who took it. 4 completed trials, 7,176 people given it.

    2,000 people given it1 trial

    • Fewer heart attacks, cancers and deaths turned up on the bone drip, but this was a side look, not the question the trial asked. It re-counted the illnesses and harms recorded in the same two thousand women. Heart attacks fell from 39 women to 24. Cancers fell by about a third. Deaths leaned lower without being clearly separable from chance. The authors say these hints justify a trial built from the start to measure them, which is a polite way of saying this one cannot settle it.

      The numbers Adverse event analysis of that same 2,000-woman trial. Myocardial infarction in 39 women on placebo against 24 on zoledronate (hazard ratio 0.60, 95 percent CI 0.36 to 1.00). Total cancers reduced, hazard ratio 0.67 (0.51 to 0.89). Hazard ratio for death 0.65 (0.40 to 1.06, p equals 0.08), and 0.51 (0.30 to 0.87) in women without an incident fragility fracture. The authors state these apparent effects justify further appropriately powered trials with these nonskeletal conditions as primary endpoints.

      Journal of Bone and Mineral Research, 2020· 2,000 people

14. 14HydralazineSmall moleculeStrong. At least two finished studies in people, and at least 1,000 people who took it. 3 completed trials, 1,724 people given it.4 reported

    1,724 people given it3 trials1 found nothing

    • In 1,050 Black patients with advanced heart failure, a fixed-dose pill of hydralazine and a nitrate, on top of modern treatment, cut deaths so clearly the trial was stopped early. 6.2% died on the pill against 10.2% on a dummy pill. Again a two-drug pill, in one specific group of people.

      The numbers 1,050 Black patients with NYHA class III or IV heart failure and dilated ventricles randomised to fixed-dose isosorbide dinitrate plus hydralazine or placebo on standard therapy, stopped early. Death from any cause 6.2% versus 10.2% (HR 0.57, a 43% reduction, p=0.01); first heart failure hospitalisation 16.4% versus 22.4% (p=0.001); quality of life score better (p=0.02).

      The New England Journal of Medicine, 2004· 1,050 people

    • Adding hydralazine plus a nitrate to standard heart failure treatment lowered deaths over two years in 642 men. At two years, 25.6% had died on the pair against 34.3% on a dummy pill. It was the first trial of a blood vessel relaxing drug to show people living longer. It is a two-drug result, not hydralazine alone.

      The numbers 642 men with chronic heart failure on digoxin and a diuretic randomised double-blind to placebo, prazosin 20 mg daily, or hydralazine 300 mg plus isosorbide dinitrate 160 mg daily, mean follow-up 2.3 years. Two-year mortality 25.6% on hydralazine-nitrate versus 34.3% on placebo, a 34% risk reduction (p<0.028); 36.2% versus 46.9% at three years. Prazosin no different from placebo. Ejection fraction rose at 8 weeks and 1 year on the combination only.

      The New England Journal of Medicine, 1986· 642 people

    • found nothingHydralazine on its own did not beat a dummy pill in heart failure. Exercise time improved in both groups, which is what a dummy pill can do, and nothing else moved. Thirty-two people, so a small no.

      The numbers 32 patients with class III or IV heart failure on digitalis and diuretics randomised double-blind to hydralazine 200 mg daily or placebo, average follow-up 20 weeks. Exercise duration rose in both groups (259 to 421 seconds on hydralazine, 271 to 339 on placebo) with no significant difference between groups. No change in weight, clinical class, heart size or ejection fraction in either group; clinical events 8 versus 13, not significant.

      American Heart Journal, 1982· 32 people

15. 15TrametinibSmall moleculeStrong. At least two finished studies in people, and at least 1,000 people who took it. 2 completed trials, 1,192 people given it.

    322 people given it1 trial

    • In advanced melanoma with a BRAF mutation, trametinib on its own held the cancer back longer than chemotherapy and more people were alive at six months. Rash, diarrhoea and swelling were the common side effects. This is the trial that led to its approval as a single drug.

      The numbers 322 patients with metastatic melanoma carrying a BRAF V600E or V600K mutation randomised 2:1 to trametinib 2 mg daily or chemotherapy, open-label (METRIC). Median progression-free survival 4.8 versus 1.5 months (HR 0.45, 95% CI 0.33 to 0.63). Overall survival at 6 months 81% versus 67% despite crossover (HR 0.54). Rash, diarrhoea and peripheral oedema were most common; reductions in ejection fraction and eye effects were infrequent.

      The New England Journal of Medicine, 2012· 322 people

16. 16DanazolHormoneLimited. People have taken it in a published study, but fewer than 100 of them. 3 completed trials, 63 people given it.5 reported

    36 people given it1 trial

    • This is what danazol costs, measured in a different illness. 36 patients with acquired aplastic anaemia took danazol on its own, 10 mg per kilogram a day up to a cap of 600 mg. After six months, good cholesterol had fallen 30 percent and bad cholesterol had risen 11 percent. The size of the heart's main pumping chamber changed, as did how well it pumped and relaxed. The authors conclude the drug causes severe blood fat changes and possible heart damage. Different condition, same drug, similar dose.

      The numbers The cost side, in a different indication. 36 patients with acquired aplastic anaemia on danazol monotherapy at 10 mg per kg per day capped at 600 mg. After 6 months HDL cholesterol fell 30 percent and LDL rose 11 percent, with significant changes in left ventricular mass, ejection fraction and diastolic indices. The authors conclude that danazol causes profound dyslipidaemia and potential cardiac dysfunction.

      Blood Cells, Molecules and Diseases, 2025· 36 people

17. 1717-DMAGSmall moleculeLimited. People have taken it in a published study, but fewer than 100 of them. 2 completed trials, 49 people given it.3 reported

    24 people given it1 trial

    • In advanced leukaemia the drug helped a few people and hurt the heart at higher doses. 24 patients were given alvespimycin into a vein twice a week at rising amounts. At the step above the tolerated dose, the harms that set the limit were heart problems, including a troponin rise and a heart attack. Three of 17 patients who could be assessed had their leukaemia clear, though their blood counts did not fully recover. Kosan, the company behind the drug, paid for it.

      The numbers 24 evaluable patients with advanced AML, 8 to 32 mg/m2 twice weekly. Maximum tolerated dose 24 mg/m2; cardiac dose-limiting toxicities at 32 mg/m2 (troponin elevation, myocardial infarction). Complete remission with incomplete count recovery in 3 of 17 evaluable. Company sponsored (Kosan).

      Leukemia, 2010· 24 peoplepoints the wrong way

18. 18Young Plasma InfusionProcedureLimited. People have taken it in a published study, but fewer than 100 of them. 1 completed trial, 18 people given it.6 reported

    18 people given it1 trial

    • This trial asked whether young plasma was safe to give, not whether it helped. 18 people with mild to moderate Alzheimer disease at Stanford took part. Nine had four weekly infusions of plasma from male donors aged 18 to 30, or salt water, in turn, with nobody involved knowing which. Nine more were given plasma openly. Nothing serious happened that was linked to the treatment. One person left with hives, one had an unrelated stroke. No benefit was measured or claimed.

      The numbers 18 patients with probable mild to moderate Alzheimer disease at Stanford, mean age 74.2 years, 12 women. Nine randomised in a double-blind crossover to four weekly infusions of one unit of young fresh frozen plasma from male donors aged 18 to 30 or 250 mL saline; nine received open-label plasma after an amendment. Primary outcomes were safety, tolerability and feasibility. No related serious adverse events. One withdrawal for urticaria, one for an unrelated stroke. Visit adherence 86 percent. NIH funded.

      JAMA Neurology, 2019· 18 people

19. 19ClenbuterolSmall moleculeLimited. People have taken it in a published study, but fewer than 100 of them. 2 completed trials, 11 people given it.5 reported

    nobody was given it1 watch-only study1 found nothing

    • found nothingClenbuterol built leg muscle in heart failure patients and did nothing for their hearts. 7 people supported by a mechanical heart pump took rising oral doses for three months, up to a very high daily amount. Body weight and lean mass rose, and thigh strength improved from about 37 to about 46 kilograms of force. The pumping fraction did not improve and the heart chamber got larger. No serious side effects or rhythm problems appeared, though a muscle enzyme rose in four patients.

      The numbers 7 heart failure patients on a left ventricular assist device given oral clenbuterol up-titrated to 720 micrograms a day for 3 months. No serious adverse events or arrhythmias, creatine phosphokinase rose in 4 patients. Ejection fraction did not improve and end-diastolic dimension increased; body weight and lean mass rose and quadriceps maximal voluntary contraction improved from 37.0 to 45.8 kg. Cardiac function did not improve.

      The Journal of Heart and Lung Transplantation, 2006· 7 peopleobserved only, nobody was given it

20. 20MK-677Small moleculeLimited. People have taken it in a published study, but fewer than 100 of them. 2 completed trials, 65 people given it.11 reported

    nobody was given it1 watch-only study

    • The trial was stopped early over a heart safety signal, and most of what it measured did not improve. 123 older people recovering from a broken hip took MK-677 or a dummy pill. Walking speed improved and IGF-1 rose. Most other measures of function did not move. The study was halted before it finished because of concern about congestive heart failure in treated patients. Stopping early leaves the benefit thin and the harm not properly measured.

      The numbers 123 elderly hip-fracture patients: gait speed improved and IGF-1 rose, but most functional measures did not, and the trial was terminated early over a congestive heart failure safety signal.

      Archives of Gerontology and Geriatrics, 2011· number of people not givenobserved only, nobody was given it

21. 21Azithromycin and roxithromycinSmall moleculeModerate. At least one finished study in people, and at least 100 people who took it. 1 completed trial, 480 people given it.

    nobody was given it1 watch-only study

    • Five days of azithromycin was linked to more heart deaths, though the extra risk to any one person is small. Researchers combed Tennessee Medicaid records and compared people taking the antibiotic against people taking no antibiotic. Death from heart causes was roughly three times as common during the five days of treatment. Amoxicillin showed no such rise. In real counts that is about 47 extra heart deaths for every million courses. Nobody was given anything here. Researchers only watched records.

      The numbers Tennessee Medicaid cohort. During five days of azithromycin, compared with no antibiotics, risk of cardiovascular death rose (hazard ratio 2.88, 95 percent CI 1.79 to 4.63) and of death from any cause (1.85, 1.25 to 2.75). Amoxicillin showed no such increase. Estimated 47 additional cardiovascular deaths per million courses, and 245 per million in the highest cardiovascular risk decile. An observational cohort, not a randomised trial, and the absolute risk is small.

      New England Journal of Medicine, 2012· number of people not givenobserved only, nobody was given it

22. 22AdiponectinBiologicPreliminary. Studied in people, but no published study has had people take it. 3 human studies where nobody was given it.2 reported

    nobody was given it1 watch-only study

    • More adiponectin is not better, and above a point it goes with dying sooner. In older adults free of heart disease, the risk of death fell as adiponectin rose up to a middling level, then climbed again above it. In people who already had heart failure or an irregular heartbeat, higher adiponectin went with clearly higher death rates. Nobody was given anything. This is a group watched over time, and it undercuts the idea of raising this hormone on purpose. The NIH paid for it.

      The numbers Population-based cohort of older adults. In those free of cardiovascular disease, the association with all-cause mortality was U-shaped: hazard ratio 0.81 per SD up to 12.4 mg/L, then 1.19 per SD (95 percent CI 1.12 to 1.27) above it. No association in those with cardiovascular disease alone; in heart failure or atrial fibrillation, higher adiponectin meant higher mortality, hazard ratio 1.31 (1.15 to 1.50) after adjustment. Results were similar for high molecular weight adiponectin and cardiovascular death. NIH funded.

      Circulation, 2012· number of people not givenobserved only, nobody was given it

Peptides and blends

1. 01N-Acetyl Semax AmidatePeptideAnimal or lab only. Only animal or lab studies on this page. No study in people yet. 1 animal study, no human study.8 reported

   1 animal or lab study

   Measured in animals or in the lab

   • In rats. After a lab-made stroke, protein tests showed plain Semax protected the brain. This is the parent peptide too. No study of the capped version has been found.

     The numbers Proteomic evidence of a protective effect for the parent heptapeptide after experimental stroke in rats. Again the parent compound: no published pharmacokinetic or efficacy study of the N-acetyl amidate form has been located.

     International Journal of Molecular Sciences, 2021· animal study

Supplements

1. 01(R/S)-1,3-ButanediolSupplementLimited. People have taken it in a published study, but fewer than 100 of them. 1 completed trial, none stating how many people.2 reported

   1 animal or lab study

   Measured in animals or in the lab

   • In mice. Mice missing a key ketone-making enzyme (HMGCS2) died early, and a 1,3-butanediol diet prevented that. In normal mice, the same diet started early in life raised midlife deaths. Started in old mice, it extended lifespan. It also prevented deaths from clogged arteries in mice bred to get them (ApoE-deficient). A keto diet with no limit on food sharply raised deaths. An independent Japanese group did the work.

     The numbers Mice lacking the ketogenesis enzyme HMGCS2 died early, which a 1,3-butanediol diet prevented. In normal mice the same diet started early in life increased midlife mortality, while started in aged mice it extended lifespan and prevented atherosclerosis deaths in ApoE-deficient mice. An ad libitum ketogenic diet markedly increased mortality. Independent Japanese group.

     Aging Cell, 2023· animal study

Other compounds

1. 01FGF17BiologicPreliminary. Studied in people, but no published study has had people take it. 1 human study where nobody was given it.4 reported

   1 animal or lab study

   Measured in animals or in the lab

   • In animals, FGF17 both kept nerve cells alive and rebuilt the insulating sheath around nerve fibres, restoring what was lost after a stroke. That pushes the finding past normal ageing and into injury.

     The numbers Animal. FGF17 acted on neuronal survival and oligodendrogenesis together to restore deficits in a stroke model, extending the oligodendrocyte finding beyond normal ageing into injury.

     Neurotherapeutics, 2026· animal study

2. 0225-HydroxycholesterolSmall moleculePreliminary. Studied in people, but no published study has had people take it. 1 human study where nobody was given it.2 reported

   1 animal or lab study

   Measured in animals or in the lab

   • In mice. In mice with clogged arteries, 25-hydroxycholesterol made by immune cells made artery swelling and damage worse. It was also measured in human artery tissue, but only measured. This is a harm result in the same part of the body as the artery stiffness study.

     The numbers Mouse atherosclerosis models, with 25-hydroxycholesterol and CH25H expression also measured in human lesions. Macrophage-derived 25-hydroxycholesterol promoted vascular inflammation and lesion remodelling, a harm direction in the same organ system as the aortic stiffness study.

     Circulation, 2023· animal studypoints the wrong way