Ongoing research
Every rule behind this app cites a guideline or a trial, and guidelines change when trials report. This page tracks the specific trials still running, or reported too recently to be in a guideline yet, that bear on a rule the app actually uses — what each one is asking, what's known so far, and what a clear result would change and for whom.
This is information, not advice for you specifically. Nothing below tells you to do anything — for that, see the rules, each of which names the trial or guideline it currently rests on.
Once someone has coronary calcium, does adding a CT coronary angiogram — which shows narrowings and soft plaque, not just calcium — actually change how they're managed, and for the better?
Multicentre randomised trial, recruiting as of August 2026. No results yet.
The app currently tells people with a positive calcium score what a CT angiogram would and wouldn't add, as a judgement call rather than a tested one. A clear result here would turn that into an evidence-based answer for who actually benefits from the extra scan.
In people with obesity but no diabetes and no established heart disease, does tirzepatide prevent heart attacks, strokes and heart failure — or only improve the numbers thought to predict them?
Randomised, placebo-controlled, and event-driven rather than fixed-duration, so it runs until enough events accrue to answer the question rather than stopping on a calendar date. About 15,000 people, age 40 and over, obesity with established cardiovascular disease or several risk factors for it. Primary completion is expected in October 2027.
This is the exact trial the app currently says doesn't exist. A positive result would move obesity-alone GLP-1 use from an approved weight-loss indication with cardiovascular benefit unproven to the same evidence tier SELECT already gives established cardiovascular disease. A negative or null result would mean the risk-factor improvements these drugs produce don't translate into fewer events in this population — a real and consequential possibility, not a formality.
In people with known heart disease and a high Lp(a), does olpasiran — which lowers Lp(a) further than pelacarsen did — actually prevent heart attacks and urgent procedures?
7,297 participants with atherosclerotic cardiovascular disease and elevated Lp(a), randomised to olpasiran or placebo every 12 weeks. Primary completion is expected around the end of 2026, with results plausible in the following months.
Lp(a)HORIZON already reported that pelacarsen lowers Lp(a) without lowering events (see below), which weakened but did not settle the question of whether lowering Lp(a) itself helps — olpasiran lowers it by a larger margin, in a drug class other than antisense oligonucleotides. A positive result would be the first proof that treating Lp(a) directly prevents events, and would very likely change the Lp(a) rules below from 'no approved therapy' to naming one. A negative result would be a second, independent data point against the whole approach, which is a different and more serious conclusion than one drug failing.
Enlicitide, the new once-daily pill version of a PCSK9 inhibitor, lowers LDL by as much as the injectable versions do. Does it actually prevent heart attacks and strokes, the way the injectables have already been shown to?
Phase 3, randomised, placebo-controlled, in people at high cardiovascular risk. Enlicitide itself received FDA approval for lowering LDL in July 2026 on the strength of the CORALreef Lipids and CORALreef HeFH trials (LDL down 57-65%), but approval for lowering a lab value is a lower bar than proof it prevents events. That is what this trial is for.
Injectable PCSK9 inhibitors already have outcomes proof (FOURIER for evolocumab, ODYSSEY OUTCOMES for alirocumab), which is why the app grades enlicitide as a strong but still-surrogate case rather than leaving it out. A positive result here would move it to the same outcome-proven tier as the injectables specifically for the oral form, which matters because nothing about a shared mechanism guarantees a shared outcomes result until it's actually tested in the new formulation.
In people with known heart disease and a high Lp(a), does pelacarsen — which lowers Lp(a) substantially — prevent heart attacks and strokes?
Topline results announced 4 September 2026: pelacarsen lowered Lp(a) substantially but did not reduce the composite of cardiovascular death, heart attack, stroke, or urgent revascularisation, in 8,323 people with established disease and Lp(a) at or above 70 mg/dL. Full results — including whether any subgroup benefited — are due at the AHA Scientific Sessions in November 2026, and this entry will be updated once they're presented.
This already changed what the app says: the Lp(a) pathway now states plainly that the first outcomes trial of an Lp(a)-lowering drug was negative, rather than treating it as pending. What the full data could still change is narrower — whether a subgroup (higher baseline Lp(a), for instance) shows a signal the topline composite result doesn't, which is the kind of detail that moves a guideline recommendation, not a first-line message.
Does deciding who gets a statin by coronary calcium score, instead of the standard risk-factor calculator, change how many people have a heart attack or stroke?
Presented at ESC Congress 2026: 5,772 adults with no known heart disease, diabetes, or prior statin use, randomised to a statin recommendation based on either their calcium score or their calculated risk. After about 4.2 years, major cardiovascular events were close to 3% in both groups, and the trial's own noninferiority criterion — the bar it was designed to clear to call the two approaches equivalent — was not met. That is a narrower result than 'no difference': events were rarer than the trial planned for, which left it underpowered to answer its own question either way. The trial's own investigators framed it as grounds for a better-powered follow-up, not a verdict. One real finding survived the underpowering: people told their calcium score took their prescribed statin far more consistently (62%) than people told their calculated risk (23%), despite being recommended one only a third as often.
This is the first large randomised trial to test what a 2025 systematic review of the same question found missing: adequately powered evidence. It didn't fully supply that either, but it is real movement, not more of the same gap — which is why the app's detection and screening content now names it rather than continuing to say only that the trial hasn't been run.