75% of Statin Users Quit Within Five Years. A Single Injection Just Silenced Their Cholesterol Gene for 18 Months.

Forty-eight dollars. That is what a year of generic atorvastatin costs, roughly the price of two movie tickets, to lower low-density lipoprotein cholesterol and hold off atherosclerotic cardiovascular disease, the condition that kills more than 15 million people per year globally and costs the U.S. healthcare system roughly $378 billion annually. More than 40 million Americans take a statin every day.

The drug works. The patients do not.

Real-world statin adherence falls below 50% within the first year of treatment, according to a 2025 UK Biobank analysis of 76,000 patients. By year five, fewer than one in four are still taking the medication consistently, not because it costs too much and not because it fails to work, but because humans are bad at taking pills every day for decades, and no amount of refill reminders, patient education campaigns, or physician nagging has fixed that in 30 years of trying.

Last month, an Alameda, California biotech called Scribe Therapeutics received clearance from Australia's Therapeutic Goods Administration to begin a first-in-human Phase 1 trial of STX-1150. It is a CRISPR-based therapy, but it does not cut DNA. Instead, it silences the PCSK9 gene through epigenetic modification: chemical tags placed on DNA that dial down a gene's activity without rewriting the underlying sequence: a single injection, no daily pills, no irreversible changes, and the gene can, in principle, wake back up.

What STX-1150 Does, and What It Does Not Do

PCSK9 is a well-validated drug target. Block it, and liver cells pull more LDL cholesterol out of the bloodstream. Two injectable antibodies already exploit this: Amgen's Repatha (evolocumab) and Sanofi's Praluent (alirocumab), both achieving 50–60% LDL-C reductions at approximately $5,850 per year, injected every two weeks. Effective, expensive, and inconvenient.

STX-1150 reaches the same gene through a different mechanism entirely. Rather than blocking the PCSK9 protein after it is produced, Scribe's ELXR platform (Epigenetic Long-Term X-Repressor) adds methyl groups to the gene's regulatory region, preventing transcription into protein in the first place — turning down a dimmer switch rather than smashing the bulb. In preclinical data presented at the 94th European Atherosclerosis Society Congress, a single dose reduced LDL cholesterol by more than 50% in non-human primates, an effect that held for 18 months with a clean tolerability profile. One dose. Eighteen months.

Scribe's platform incorporates an allosteric control loop constraining the methyltransferase enzyme's activity, reducing off-target transcriptional effects while boosting targeted silencing by an average of four-fold. Earlier epigenetic editors struggled with methylating genes they were not supposed to touch, and Scribe claims to have built the safety brake into the molecule itself.

In Phase 1, up to 64 adults with elevated LDL-C and increased cardiovascular risk will receive the drug at sites across Australia and New Zealand, with participants monitored for one year after a single dose. Principal investigator Stephen Nicholls, director of the Victorian Heart Institute at Monash University, will run the trial from Victoria's Victorian Heart Hospital.

The Compliance Calculation Nobody Runs

Cholesterol pharmacoeconomics has a structural blind spot. Cost-effectiveness models compare drug prices and trial-based efficacy, but they almost never model what happens when most patients stop taking the drug, which is exactly what most patients do.

Here is the arithmetic, applied to 40 million Americans currently prescribed statins.

At five years, roughly 75% have stopped, per multiple large cohort analyses that found persistence below 25%. That leaves 30 million people who were handed a cardiovascular-protective prescription and abandoned it.

Apply the Cholesterol Treatment Trialists' meta-analysis: each 1 mmol/L reduction in LDL-C cuts major cardiovascular events by approximately 22%. Using the ACC/AHA 10-year MACE threshold of 7.5% for the statin-eligible population:

30 million non-adherent patients × 7.5% ten-year baseline MACE rate × 22% preventable fraction = approximately 495,000 cardiovascular events per decade that could have been prevented if those patients had kept swallowing a pill costing less than four dollars a month.

At an average acute care cost of $28,000 per major cardiovascular event, that non-compliance costs roughly $1.4 billion per year in avoidable U.S. hospitalizations. Billion, with a B.

A one-shot therapy cannot eliminate cardiovascular risk, but it can eliminate the compliance variable. If 100% of treated patients receive 100% of the intended effect for the full duration, the population-level benefit per dollar spent shifts by multiples, not percentages. A therapy that costs 100 times more than atorvastatin per dose but achieves 100% compliance and lasts five years delivers more aggregate cardiovascular protection than a cheap pill that three-quarters of patients abandon.

Limitations

This is a Phase 1 trial in 64 patients, and STX-1150 has never been injected into a human being.

Non-human primate pharmacokinetics frequently fail to predict human responses, and the 18-month durability observed in primates could shrink to six months, three months, or nothing at all in people, which is precisely the kind of uncertainty that a small first-in-human dose-escalation trial is designed to resolve before anyone stakes a healthcare system's budget on it.

Statins have more than three decades of randomized controlled trial data in hundreds of thousands of patients across dozens of studies, with hard cardiovascular outcomes backing every 1 mmol/L of LDL-C reduction, while STX-1150's evidence base is a surrogate endpoint measured in a small number of animals. Regulators and cardiologists will rightly demand years of outcomes data before this technology displaces established first-line therapy. As they should.

Pricing remains speculative, since Scribe has not disclosed expected costs and Phase 1 is years from commercial launch. If priced like current gene therapies ($500,000 to $2.2 million per dose), the compliance math collapses. Economics require pricing in the range of a cholesterol drug, not a rare-disease cure.

And the compliance argument itself carries a quiet assumption: that STX-1150 works in people at all, which is precisely the question this trial exists to answer.

What You Can Do

If you stopped taking a statin, restart the conversation with your physician today, because the compliance data is a population tragedy that plays out one patient at a time, and alternatives to daily pills already exist: inclisiran (Leqvio), an siRNA injection given twice yearly that also silences PCSK9, and bempedoic acid (Nexletol), which works through a different pathway entirely.

If you are a cardiologist: watch the Scribe readout. Allosteric specificity data from ELXR is the signal to track, because off-target methylation remains the critical safety concern for epigenetic editors in broad preventive use, and Scribe's four-fold specificity improvement will either hold up in human data or it will not.

If you run a health system: model compliance-adjusted cost-effectiveness against your own claims data. Most PBMs can extract five-year PDC (proportion of days covered) metrics in an afternoon, and your board will find your own number more persuasive than the $1.4 billion national estimate.

The Bottom Line

Cardiovascular disease kills more people than any other condition on Earth, and the single largest leak in the treatment pipeline is not drug efficacy but patient persistence. Scribe is betting that three decades of failed adherence campaigns call for a different architecture: one shot that silences a gene for years, without permanent edits, and without asking anyone to remember a pill. It is a Phase 1 bet, not a proven solution, but the compliance arithmetic is relentless, the PCSK9 biology is validated, and $1.4 billion in annual avoidable hospitalizations belongs to nobody and everybody at once.