Oral PCSK9 Inhibitor Approved: LIPFENDRA Price vs Repatha

Oral PCSK9 Inhibitor Approved: LIPFENDRA Lists at $315. Repatha Already Sells for $239.

The first oral PCSK9 inhibitor cleared the FDA on 16 July 2026 with antibody-grade LDL-C data and a manufacturing route worth the Science paper it got. AstraZeneca’s oral will finish its cardiovascular outcomes trial one month earlier, from a start twenty months later. The wall Merck has to clear was never the needle.

Merck priced the first oral PCSK9 inhibitor at $10.50 a tablet. Amgen has been selling Repatha, in cash, for roughly eight dollars a day since October. That is the whole trade in two lines, and most of the coverage will not run it, because the interesting sentence is the one about the pill.

01 — The ApprovalWhat the CORALreef Trials Actually Said

It is an interesting sentence. Enlicitide is a 1,616 Da macrocycle that Merck taught to survive a stomach, and the field had more or less filed that problem under closed. LIPFENDRA (enlicitide decanoate) tablets, 20 mg once daily, are now approved as an adjunct to diet and exercise for adults with hypercholesterolaemia, HeFH included, with no line-of-therapy restriction anywhere on the label.

There is also no outcomes claim on that label. Hold onto that.

CORALreef Lipids randomised 2,904 patients across 168 sites in 14 countries, 1,935 to enlicitide and 969 to placebo, mean age 63, 39% women. Mean baseline LDL-C was 96.1 mg/dL and 97% were already on a statin. Fifty-eight percent had a prior major ASCVD event; the rest were primary prevention at high risk.

CORALreef HeFH ran 303 patients through the same machine and returned 59%. The safety table is quieter than the efficacy table, and it is fair to say so. Adverse events came in at 64% against 62%, discontinuations 3% against 4%, serious adverse events 10% against 12%, new-onset or worsening diabetes 6% in both arms, and mortality 0.7% in both. Serious events were lower on drug. William E. Boden, writing the accompanying NEJM editorial, called the results “impressive”.

Those are evolocumab numbers, produced by a tablet. Dean Y. Li at Merck Research Laboratories called it a “pivotal moment”, and on the chemistry he is not overselling. The chemistry is not what decides the revenue.

02 — FormulationThe Part Nobody Puts in the Headline

Enlicitide binds the LDL-receptor-binding domain of PCSK9 at an IC₅₀ of 2.5 nM. Merck fished the hit out with mRNA display, then published a biocatalytic route in Science using engineered enzymes for fragment coupling and macrocyclisation, which is the only reason a peptide this size is manufacturable at commercial tonnage. That route will outlive this molecule. Genuinely good work.

The molecule is not cell permeable. It never became permeable. Merck co-formulated it with sodium caprate, a medium-chain fatty acid that prises open intestinal tight junctions for long enough to let a peptide slip through paracellularly.

That ratio is a permanent cost-of-goods position, and it is also why the leaflet has to police breakfast. Rybelsus already ran that experiment on the general population. The trial sites were not the problem.

03 — The ConstraintNavar

The lead author on CORALreef Lipids is Ann Marie Navar, UT Southwestern. In 2017, at Duke, Navar was lead author on the JAMA Cardiology paper that pulled 45,029 new PCSK9 inhibitor prescriptions from the class’s first year on the American market and counted what happened to them.

Abandonment tracked out-of-pocket cost, not clinical profile. Patients with LDL-C at or above 190 mg/dL were rejected at a higher rate than patients below it, which is not a clinical algorithm doing that. Mariell Jessup called prior authorisation “the dirty little secret of American medicine” from the discussant’s podium; Navar’s answer was that it had never been a secret to anyone writing the scripts. Myers and colleagues later put a body count on it in Circulation: Cardiovascular Quality and Outcomes: rejected and abandoned patients ran about 10% more cardiovascular events than the ones who got paid for.

None of that was the needle’s doing. Amgen cut Repatha’s list from $14,100 to $5,850 in 2018 and the molecule did not change one atom. Sanofi and Regeneron bought preferred placement at Express Scripts in the same window, same story. The class’s lost decade was an adjudication failure, and the industry solved as much of it as it solved by writing cheques to PBMs.

So Merck has removed a constraint. It removed the one it could engineer against.

04 — PricingThe Arithmetic

$10.50 a tablet. $315 for thirty days. $3,780 a year at list.

Against Repatha’s WAC of about $514 a month, roughly $6,168 a year on the fortnightly 140 mg regimen, that is a 39% undercut. Against Leqvio at roughly $6,497, better still. Every analyst note will use that comparison, and it is the wrong one.

Amgen launched AmgenNow in October 2025 and put Repatha on it at $239 a month — a discount Amgen itself described as around 60% off list — also live through GoodRx.

Read the middle card against the left one. Merck’s oral PCSK9 inhibitor lists 32% above what the incumbent already charges cash, and the incumbent is the one holding FOURIER: 27,564 patients, primary composite down 15%, CV death, MI or stroke down 20%. In a market where US pricing pressure is already rewriting global launch strategy, pricing against a list nobody transacts at is a choice worth interrogating on day one rather than at the first quarterly miss.

Leqvio is the instructive case. It is twice-yearly, priced within a few hundred dollars a year of Repatha, objectively more convenient than a fortnightly injection, and payers have spent four years tiering it behind Repatha anyway. Some restrict it until a patient has failed a monoclonal outright. The reason is the missing outcomes trial, and on that axis LIPFENDRA is standing exactly where Leqvio stands.

The bull case is that an oral is different in kind, and there is precedent: IQVIA’s read on the Wegovy pill argued that an oral format expands a category rather than cannibalising it, by pulling in patients who were never going to start an injection at all. That transfers cleanly to statin-intolerant and needle-averse patients, and it is the strongest thing Merck has. It also assumes those new patients clear prior authorisation, which is where the GLP-1 analogy breaks, because Wegovy was never gated at a 79% day-one rejection rate.

05 — CompetitionAZD0780 Is the Actual Threat, and the Timeline Is Worse Than It Looks

AstraZeneca’s oral PCSK9 inhibitor is a small molecule. No permeation enhancer, no 1% bioavailability, no fasting window, no 360 mg of caprate riding along, and a cost-of-goods floor well underneath a macrocyclic peptide’s. In the PURSUIT Phase 2b, published in JACC, AZD0780 30 mg cut LDL-C by 50.7% on top of statin (95% CI −59.0 to −42.4) and got 84% of participants to their ACC/AHA target. Five points behind enlicitide, from a molecule that costs a fraction to make.

Now the calendar, which is where Merck’s first-mover story stops working.

And the registrational trials that decide when AstraZeneca can actually launch both read out in the first week of January 2027.

Then there is what AstraZeneca has said it intends to do with the molecule: monotherapy or fixed-dose combinations across its portfolio. It has paid CSPC $100 million upfront, with up to $1.92 billion in milestones, for YS2302018, an oral Lp(a) disruptor, explicitly to pair with AZD0780. Consider what a fixed-dose combination requires, and then look at the two tablets side by side.

The oral PCSK9 inhibitor Merck launched cannot be combined. The one AstraZeneca is building was designed to be. Merck needs the franchise regardless, because Keytruda’s US exclusivity ends in 2028 and cardiometabolic is where the replacement is supposed to come from. That is the reason to watch $315 rather than admire it.

06 — IndiaNot the Patient Market. The Supply Chain.

The approval is US-only, and nothing in Merck’s materials addresses India. At $315 a month, roughly what a large share of Indian households spend on everything in a month, it never will address India at list. The Indian PCSK9 story stays what it has been: a cash market, where inclisiran and evolocumab sit above the line for almost everyone and ezetimibe holds the ground. The constraint changes shape without disappearing. Americans lose the drug to a PBM rejection. Indians lose it at the counter.

The opportunity is on the supply side, and it is not generics, because these patents run long. It is that a 1%-bioavailable peptide needs API tonnage per patient-year that no injectable ever did, and that tonnage has to be made somewhere. India’s peptide CDMO base was built on solid-phase synthesis during the GLP-1 wave, and it was built well. Biocatalytic macrocyclisation with engineered enzymes, and permeation-enhanced oral solids with a controlled food effect, are a different capability stack: enzyme engineering, biotransformation at scale, and dosage-form work where the excipient is doing the pharmacokinetics rather than sitting inert beside the API.

Whoever builds that stack in India first is positioned for every oral macrocycle queued behind enlicitide, not just this one. Whoever does not will watch the semaglutide margin trap repeat on a molecule they cannot make, at the wrong end of the value chain, competing on the intermediate instead of the process.

There is a second-order risk worth naming. A capability stack of that kind is built on sub-tier vendors and specialist enzyme suppliers, and US filings do not forgive weak links anywhere in that chain — a point we worked through in the Unicycive CMC case, where a subcontractor nobody had a contract with blocked an application twice.

07 — Evidence IntegrityTwo Things That Should Not Be Waved Through

The pre-specified CORALreef Lipids primary result is 55.8 points. The 60% number now circulating is post-hoc. It came from an analysis run after unblinding, which excluded baseline LDL-C values Merck judged biologically implausible and adjusted for missing data, and it returned 59.7% at week 24. Merck’s release leads with 56% and reports 60% with the caveat attached, which is more discipline than most sponsors manage. The caveat will not survive the news cycle. Witfire has applied this standard to three amyloidosis trials where the effect lived outside the protected statistics. It does not flip because the sponsor is winning.

Then the placebo. In CORALreef AddOn, published in JACC, the enlicitide tablet contained sodium caprate and the matching placebo did not. Merck says blinding held, and there is no reason to doubt it. But caprate is a tight-junction modulator, and the HeFH safety table reads diarrhoea 7% versus 2%, dizziness 9% versus 4%. Whether the Lipids placebo carried caprate is not stated anywhere in the approval materials, and the Lipids safety profile was clean, which is either reassuring or a clue about what the placebo contained. If the enhancer only ever sat in the active arm across the programme, nobody has run an enhancer-controlled tolerability comparison on this class, and the answer will show up in 2027 refill curves rather than a 2026 adverse-event table.

Neither point touches the mechanism. Free plasma PCSK9 fell more than 90% across 14 days at 10 and 20 mg in Phase 1, serious adverse events came in below placebo, and mortality was identical. Enlicitide does what it says. The argument is about the last four percentage points and about who owns the diarrhoea.

08 — ScoringWitfire Risk Score

Five weighted dimensions, scored 1 (low risk) to 10 (very high risk), applied to the LIPFENDRA asset. The composite is the weighted mean.

Regulatory exposure is near the floor, which is unusual for a composite that still lands at Elevated. The weight sits in displacement and execution instead, and that shape is the finding. This is an approval with a low probability of being taken away and a real probability of being outsold.

09 — ForwardWhat to Watch

1. The TrumpRx price for LIPFENDRA. Unannounced as of late July. This is now the nearest catalyst in the file and the one that decides whether section four’s comparison holds or collapses.
2. CORALreef Outcomes on 29 November 2029, and AZURE-Outcomes on 26 October 2029. Read them as one event, because they now are one.
3. Adjudication data from the first two quarters. Not scripts written. Scripts paid. Navar’s 30.9% is the benchmark, and Merck’s own trial leader established it.
4. Amgen’s next move. AmgenNow at $239 is already underneath LIPFENDRA at $315, and Amgen has cut hard twice when it wanted shelf space. There is no reason to think it will not do so again, and Merck has less room to follow, because Merck’s molecule costs more to make.
5. Persistence, quietly. Somewhere in 2027 there is a refill curve that will say what a 30-minute fasting window costs when the patient is not in a trial and nobody is calling them.

10 — VerdictThe Wall Was Never the Needle

Merck cleared an oral bioavailability barrier the field had written off for a molecule this size, and built a biocatalytic route that will outlast the asset. That is real, and it deserves the coverage it is getting.

Here is what it bought. A tablet that is 99% packaging by mechanism, cannot be combined with anything, must be taken on an empty stomach, has no outcomes data until the last month of 2029, and lists at $315 against $239 of cash Repatha that already holds FOURIER. Handed to a pharmacist who works for a benefit manager that rejected four out of five of these prescriptions on day one and has not been given a single reason to stop.

The competitor that has not launched yet will finish its outcomes trial a month earlier, from a start twenty months later, with a molecule that costs a fraction to make and a fixed-dose combination partner already licensed. Merck’s two-year head start is real. It is also all there is.

The one thing that could change this reading is a TrumpRx number that lands near Amgen’s. Until Merck publishes it, the pricing case rests on a list nobody transacts at, against a cash price that everybody can already get.