Resmetirom for Advanced Fatty Liver

Healthcare · §655

For the first time, there's a pill that bends the trajectory of advanced fatty liver disease. Resmetirom — sold as Rezdiffra, FDA-approved in March 2024 — is the first drug to clear the bar of reversing both the liver inflammation and the scarring of MASH on biopsy at one year. The catch: about one in three patients responds, the list price is roughly $47,000 a year, and the proof it actually prevents cirrhosis and liver failure is still being collected. What follows is what the drug does, who it's for, what it costs in money and side effects, and how to think about it next to the alternatives.

Decide · Daily Evidence Moderate Chapter Healthcare

The headline number is longevity: this is the first drug aimed squarely at stopping severe fatty liver disease before it turns into cirrhosis. The taking-it part is easy — one pill a day, with or without food. The paying-for-it part is hard — a list price near a new car each year, mostly absorbed by insurance when it works. And you won't feel much: MASH is mostly silent, and so is its reversal. This is a long game, played with lab numbers and scans, not a drug you notice working.

Fatty liver disease at the F2 or F3 scarring stage is a slow-motion problem with one root cause: the liver has been storing too much fat for too long, the fat-soaked cells become inflamed and injured, and the repair machinery lays down scar tissue. Resmetirom acts on the part of the liver-cell machinery that decides how much fat to burn and how much to keep Karanjia et al. 2024.

The trick is selectivity. Thyroid hormone is a powerful fat-clearing signal, but using it whole-body would speed up your heart, thin your bones, and cause every other thyroid side effect along the way. Resmetirom is shaped to land mostly in the liver, and only on the one thyroid-hormone switch that controls fat handling — not the one in the heart and bones Harrison et al. 2019. So the liver gets the message to burn its own fat stores, clear cholesterol from the blood, and stop laying down new fat. The rest of the body barely notices the drug is there.

Over months, the downstream effect is what you'd expect from less stored fat in liver cells: less inflammation, less injury, and — eventually — less new scar tissue being laid down.

What the trial actually showed

The approval rests on one large, well-run study called MAESTRO-NASH Harrison et al. 2024. It is the first time any drug has cleared the bar that regulators set for fatty liver disease: improve both the inflammation and the scarring, measured by a pathologist reading a liver biopsy, after one year on the drug.

Translated: at the better dose, roughly one in three people on the drug clears the disease at a year, and one in four shows their scarring move back one stage. That is honest, useful, and not a miracle. About two-thirds of patients don't hit those benchmarks at one year — some take longer, some don't respond.

The other side of the trial is what's still unknown. Biopsy improvement is what the FDA calls a surrogate endpoint — a marker we strongly believe predicts the things that actually matter (cirrhosis, liver failure, transplant, death from liver disease) but haven't yet seen the drug change directly. The same study is following the same patients for up to 54 months to answer that question, with results expected around 2027–2028 FDA 2024. The approval is "accelerated" because regulators decided the biopsy result was good enough to start prescribing while that confirmation was being collected.

Why this matters now, not in ten years

Significant fatty liver scarring is a deceptively quiet condition. The liver does not have pain nerves on the inside; you don't feel the inflammation, you don't feel the scarring, and you don't feel the slow drift from "fatty liver" to "scarred liver" to "cirrhosis." A blood test usually catches it accidentally. Most people get diagnosed at the late physical where their primary-care doctor mentions, in passing, that the liver enzymes have been creeping up Younossi et al. 2023.

The forecast without treatment is statistical, not dramatic. A typical patient at F3 scarring faces, on average, about a 5–10% chance per year of progressing to cirrhosis. Once cirrhosis lands, the annual risk of a serious event — internal bleeding, fluid in the abdomen, mental confusion from liver failure, or liver cancer — sits around 3–5%. None of this is felt until it is — and then it is felt all at once, in a hospital. MASH is now the leading reason for liver transplant in U.S. women Younossi et al. 2023.

Stakes for the typical person reading this aren't about today's energy or tomorrow's bloodwork. They're about whether the next FibroScan in two years shows the same number, or a worse one — and what that means for the decade that follows.

How it's taken

The drug is a pill, swallowed once a day, with or without food. The dose is set by your body weight at the start — no titration, no escalation, no timing tricks. You take the same dose every day from day one.

Before the first prescription, your doctor needs proof you actually have the disease the drug is approved for. In practice that means either a liver biopsy or a non-invasive workup — usually a FibroScan plus a blood-based scarring score — pointing at F2 or F3 scarring. The drug is not labelled for milder fatty liver (F0–F1) or for cirrhosis (F4) Madrigal 2024.

One real-world watch-out: resmetirom interacts with some common drugs through liver-handling enzymes. Strong interactions with high-dose statins, the gout/cholesterol drug gemfibrozil, and the antiplatelet drug clopidogrel can either change resmetirom levels or push liver enzymes up. Your doctor will adjust statin doses if needed and avoid the harder interactions Madrigal 2024.

When not to take it

The label's main monitoring concern is the liver itself. Liver enzyme bumps were more common on resmetirom than on placebo, and at low rates the bumps were large enough to count as drug-caused liver injury. The drug is held if enzymes climb past five times the upper end of normal, or if symptoms of liver trouble appear Madrigal 2024. None of this is a reason not to start when the indication fits — it's a reason to keep up with the bloodwork the prescriber asks for.

What else is on the table

Resmetirom is the first drug for MASH-with-scarring, but it is not the only option. The honest ranking, in roughly the order most hepatologists would consider:

Weight loss of 7–10% of body weight. Still the single most effective thing anyone can do for this disease. A landmark biopsy study followed 261 patients who actually lost the weight: roughly nine in ten cleared MASH at that threshold Vilar-Gomez et al. 2015. The catch is in the word "actually" — sustaining a 10% weight loss for a year is hard, and most people don't.
The new obesity drugs. Semaglutide (Wegovy/Ozempic) and tirzepatide (Mounjaro/Zepbound) drive the weight loss the previous bullet describes, and probably reverse MASH directly on top of that. Tirzepatide cleared MASH in roughly half to two-thirds of patients in a mid-stage trial Loomba et al. 2024; semaglutide showed similar resolution rates earlier Newsome et al. 2021. Neither is yet FDA-approved specifically for MASH, but both are widely used in the same patient — most of whom have obesity, diabetes, or both.
Bariatric surgery. The most reliable trigger of MASH and scarring reversal, when the patient qualifies. Up-front cost and risk are real; long-term effect is substantial.
Pioglitazone and vitamin E. Older, cheaper, modestly effective. Recommended by liver-specialty guidelines in specific patients — pioglitazone for biopsy-confirmed MASH in patients with type 2 diabetes, vitamin E for the same indication without diabetes Rinella et al. 2023. Pioglitazone causes weight gain and small bone effects; high-dose vitamin E has a small prostate-cancer signal in men.

The realistic clinical picture: weight loss first, with a GLP-1-class drug helping if needed, and resmetirom layered on for patients who already have significant scarring and need a liver-specific lever — or for patients in whom weight loss is not happening and the scarring is moving.

What this drug is not

It is not a thyroid pill. Despite acting on a thyroid-hormone switch, at the approved doses it does not change your TSH, your free T4, your heart rate, or your weight. If you take levothyroxine, you almost certainly do not need to adjust it Harrison et al. 2024.
It is not a cure. About one in three patients reaches "MASH cleared" on biopsy at one year. That number is real, and it is also the first time any pharmaceutical has reached it — but the other two-thirds didn't, and even the responders need monitoring.
It is not a replacement for losing weight. Every participant in the pivotal trial was counseled on diet and exercise; the drug's effect sits on top of that, not instead of it. The patients who get the most out of resmetirom are usually the ones who haven't been able to sustain the 7–10% weight loss that would otherwise do most of the work.
Approval does not mean long-term benefit is proven. Accelerated approval is conditional on the longer outcomes trial reading out. The biopsy improvement is strongly suggestive of fewer cirrhosis cases over a decade — but "strongly suggestive" is not the same as "proven," and the proof is still being collected FDA 2024.

What it costs and how to get it

The U.S. list price at launch sits around $47,400 a year — roughly $3,950 a month ICER 2024. Almost nobody pays that out of pocket. Most commercially insured patients pay something closer to $10 a month through the manufacturer's copay program, and a patient-assistance program covers some uninsured patients who meet income criteria. Medicare Part D coverage has been the bigger fight; some plans cover, some require steep prior authorisation.

The independent cost-effectiveness watchdog ICER reviewed resmetirom in 2024 and concluded the launch price needed roughly a 30–60% discount to be confidently worth it at standard willingness-to-pay benchmarks — and that the whole cost-effectiveness case is conditional on the long-term outcomes trial showing fewer cirrhosis cases ICER 2024. The drug is not a bad deal — it's the first one ever for the disease — but it is an expensive one, and the price math is contested.

The prescription itself goes through hepatology or gastroenterology in most U.S. systems. A typical workup before the first script: FibroScan (with the elastography number, not just the ultrasound), a blood-based scarring score (FIB-4 or ELF), a lipid panel, basic liver and kidney chemistry, a thyroid blood test, an alcohol history, and a careful medication review. Some payors require a liver biopsy on top of all that, despite the FDA label not asking for one.

Outside the U.S., as of 2026, the drug is not yet approved in Europe, the UK, or Japan. Patients there access it through clinical trials, compassionate-use pathways, or direct private import — all of which involve their own costs and friction.

Where it goes wrong

Quitting in the first month. Diarrhea and nausea hit roughly a quarter of patients in the opening weeks, against a placebo background of about fifteen percent Harrison et al. 2024. It's usually mild and usually fades by week eight, but plenty of people stop before they get there. Warn yourself in advance that the first weeks are the worst weeks and you'll stick.
Adherence on a silent disease. You don't feel the drug working. You don't feel the disease either. A daily pill against nothing you can feel is a hard psychological setup; the people who keep taking it are usually the ones who internalised what the next FibroScan represents.
Drug-drug surprises with statins. The interaction is real, and it can drive liver enzymes up high enough to look like a resmetirom problem when it's actually a statin-dose problem. The fix is a dose adjustment by the prescriber, not stopping the resmetirom.
Insurance denials on biopsy grounds. The label doesn't require a biopsy. Many U.S. payors do. A FibroScan plus the right blood-based score is often enough to win the prior authorisation, but it takes work — and a hepatology office that knows the appeal language.
Treating the drug as the whole plan. Resmetirom plus an unchanged diet, unchanged weight, and unchanged exercise pattern still works on the biopsy — but it works better when the lifestyle pieces are moving too, and the lifestyle pieces affect cardiovascular outcomes the drug can't reach.

What changes if it works

If you are one of the responders, here is what the next few years actually look like.

Weeks one to eight. The most noticeable change is the side-effect curve — looser stools, occasional nausea — flattening out. Nothing about your energy, appearance, or daily life shifts. The liver enzymes on the next blood draw are quietly lower.
Month three to month six. A repeat lipid panel shows your LDL cholesterol down by something like fifteen percent without changing your statin Harrison et al. 2024. Your hepatologist mentions, casually, that the numbers look good. You still don't feel any different.
Year one. A repeat FibroScan (or, in a research setting, a repeat biopsy) shows less liver fat and, in roughly a quarter of responders, less scarring than the year before. This is the data point the drug exists for. From the patient's chair, it is a number on a screen — but it is the first time that number has moved in the helpful direction in most patients' charts.
Year two and beyond. The bet — and it is still a bet, until the long-term trial reads out — is that the FibroScan that would have crept upward into cirrhosis territory instead stays flat or improves. The thing you don't experience is the hospital admission for liver failure or the transplant referral you'd otherwise have eventually had a meaningful chance of getting Younossi et al. 2023.

This is not a drug with a felt payoff. The payoff is a quieter doctor's-appointment cycle, a steadier set of numbers, and — if the trial reads out the way the mechanism predicts — a future that doesn't include the parts of fatty-liver disease that hurt.

Adjacent topics worth knowing about: the new obesity drugs (semaglutide and tirzepatide) and what they do for fatty liver on top of weight; non-invasive liver-scarring measurement (FibroScan, FIB-4, ELF) for catching the disease before it gets to F2 or F3; cardiovascular risk in metabolic disease, which usually kills these patients sooner than the liver does; and bariatric surgery for the subset of patients who qualify and want a one-time intervention rather than a daily pill.

Related in the handbook

Helps

— This is the first pill to reverse the inflammation and scarring of advanced fatty liver — it's the drug for the dangerous end of MASLD.

Alternatives

— The GLP-1 weight-loss drugs also improve fatty liver, and for many patients are the alternative or companion to a liver-specific pill.
— Before this $47k drug, vitamin E is the cheap, older option with biopsy evidence in some MASH patients — worth discussing first.

— Knowing how to read a liver panel makes sense of the lab numbers your doctor watches while you're on this drug.

Substance and claimed effects

Resmetirom (brand name Rezdiffra; developmental code MGL-3196) is an orally bioavailable, liver-directed, selective agonist of the thyroid hormone receptor beta (THR-β) developed by Madrigal Pharmaceuticals. It received accelerated U.S. FDA approval on 14 March 2024 as the first pharmacotherapy specifically indicated for adults with non-cirrhotic metabolic dysfunction-associated steatohepatitis (MASH, formerly NASH) with moderate to advanced liver fibrosis (stage F2 to F3) FDA 2024. The indication explicitly excludes compensated or decompensated cirrhosis (F4). Approved as an adjunct to diet and exercise, dosing is weight-based: 80 mg orally once daily for adults under 100 kg and 100 mg once daily for adults at or above 100 kg Madrigal 2024.

Claimed effects, in order of evidentiary weight:

Resolution of MASH on biopsy (≥2-point reduction in NAFLD Activity Score, with reductions of at least 1 point in both lobular inflammation and hepatocellular ballooning) without worsening of fibrosis.
Improvement of liver fibrosis by ≥1 stage without worsening of MASH activity.
Reduction in hepatic steatosis (liver fat) measured by MRI proton density fat fraction (MRI-PDFF).
Reduction in non-invasive fibrosis biomarkers: ALT, AST, GGT, FibroScan vibration-controlled transient elastography (VCTE), MR elastography (MRE), and proprietary panels (FIB-4, ELF).
Reduction in atherogenic lipids — LDL-C, apolipoprotein B (apoB), triglycerides, lipoprotein(a).

Long-term outcomes — progression to cirrhosis, hepatocellular carcinoma, liver transplant, liver-related mortality, and all-cause mortality — are the subject of the ongoing MAESTRO-NASH outcomes phase (54-month confirmatory readout), which the FDA's accelerated approval is conditioned upon Harrison et al. 2024.

The entry covers the substance and its meaningful consequences across the catalogue's dimensions: short-term health (felt liver-disease reversal is modest in most patients; lab and imaging changes are not "felt"), longevity (the basis for prescribing — interrupting the F2→F3→cirrhosis→HCC/transplant trajectory), cost burden (very high), effort burden (low — one pill daily), and evidence (a single positive pivotal phase 3 trial; histology endpoints are surrogates).

Evidence by addressing question

Mechanism

The thyroid hormone receptor exists in two main isoforms: THR-α (predominant in heart, bone, central nervous system) and THR-β (predominant in liver and anterior pituitary). Native thyroid hormone (T3) activates both, which is why systemic thyromimetics cause tachycardia, atrial fibrillation, bone loss, and HPT-axis suppression at doses that meaningfully affect liver lipid metabolism Karanjia et al. 2024. Resmetirom is engineered for two layers of selectivity: a ~28-fold receptor selectivity for THR-β over THR-α, and hepato-selective uptake driven by structural features that exploit the liver's high first-pass exposure and OATP transporters Harrison et al. 2019.

Mechanistically, THR-β activation in hepatocytes drives a coordinated lipid-clearance program: upregulation of mitochondrial fatty acid β-oxidation, induction of CPT1A and ACADM, increased mitochondrial biogenesis, increased LDL receptor expression on hepatocytes (driving LDL-C clearance), increased apoB secretion that is offset by increased clearance, and downregulation of de novo lipogenesis through reduced SREBP-1c signaling Karanjia et al. 2024 Harrison et al. 2019. Net effect: less fat stored in hepatocytes, less inflammation downstream of lipotoxicity, and over time less stellate-cell activation and less fibrogenesis. The mechanism predicts (and the trials confirm) that liver-fat reduction precedes histological MASH resolution, which precedes measurable fibrosis improvement.

Evidence

The pivotal trial is MAESTRO-NASH, reported by Harrison et al. in NEJM 2024 Harrison et al. 2024. Design: a multinational, double-blind, placebo-controlled phase 3 RCT enrolling 966 adults with biopsy-confirmed MASH and stage F1B, F2, or F3 fibrosis, randomized 1:1:1 to placebo, resmetirom 80 mg, or resmetirom 100 mg, taken orally once daily. Co-primary endpoints assessed by central blinded pathologists on paired biopsies at week 52: (1) MASH resolution with no worsening of fibrosis; (2) ≥1-stage fibrosis improvement with no worsening of NAS.

Results at week 52:

MASH resolution: 25.9% (80 mg) and 29.9% (100 mg) versus 9.7% placebo. Both arms statistically significant (P<0.001).
Fibrosis improvement: 24.2% (80 mg) and 25.9% (100 mg) versus 14.2% placebo. Both statistically significant.
LDL-C: −13.6% (80 mg) and −16.3% (100 mg) versus 0.1% placebo at week 24; effects sustained through week 52.
Liver fat by MRI-PDFF: relative reductions of roughly 35–45% at week 52 in the resmetirom arms versus ~5% on placebo (from substudies and from the parallel MAESTRO-NAFLD-1 safety trial Harrison et al. 2023).
Non-invasive fibrosis measures: VCTE (FibroScan) decreased ~3.7 kPa on 100 mg versus essentially no change on placebo; ELF score modestly improved.

Number needed to treat for MASH resolution: roughly 5 at the 100 mg dose. NNT for fibrosis improvement: roughly 9. These are honest effect sizes for histology — not transformative cure rates, but the first time any pharmacotherapy has met both co-primary biopsy endpoints in MASH.

Supporting evidence:

MAESTRO-NAFLD-1 (phase 3, 1,143 patients), a primarily safety-focused trial of patients with presumed MASH selected non-invasively, confirmed safety over 52 weeks and showed a relative ~50% reduction in MRI-PDFF on the 100 mg dose versus near-zero on placebo, plus reductions in LDL-C, apoB, triglycerides, GGT, ALT, AST Harrison et al. 2023.
MGL-3196-05 (phase 2, 125 patients, NAS ≥4, F1–F3), the earlier 36-week RCT that established hepatic fat reduction by MRI-PDFF as the primary signal and demonstrated lipid effects Harrison et al. 2019.

FDA's accelerated approval rests on the surrogate histology endpoints. Confirmation of clinical benefit (cirrhosis, decompensation, transplant, death) requires the MAESTRO-NASH outcomes phase, which is following ~2,000 patients for up to 54 months and is on track to read out around 2027–2028 FDA 2024 Harrison et al. 2024.

Protocol

Per the FDA label Madrigal 2024: oral, once daily, with or without food. Dose: 80 mg/day for body weight under 100 kg; 100 mg/day for body weight at or above 100 kg. No dose escalation; full dose from day one. Indicated for non-cirrhotic MASH with F2 or F3 fibrosis as an adjunct to diet and exercise — the label does not require a liver biopsy, but in practice many payors do.

Monitoring: liver enzymes (ALT, AST, alkaline phosphatase, total bilirubin) before initiation and as clinically indicated; the label flags drug-induced liver injury as a warning. Free T4 and TSH at baseline only (no requirement for ongoing thyroid monitoring at approved doses, because THR-β selectivity preserves the HPT axis). Lipid panel useful to capture incidental LDL-C reduction.

Drug interactions: resmetirom is a CYP2C8 substrate. Strong CYP2C8 inhibitors (gemfibrozil, clopidogrel at high doses) are contraindicated; moderate inhibitors (cyclosporine, deferasirox) require dose adjustment. Strong OATP1B1/1B3 inhibitors (some statins, including high-dose atorvastatin and rosuvastatin) increase resmetirom exposure; reduce concomitant statin doses or substitute. Resmetirom does not have meaningful CYP3A4 induction at approved doses.

Contraindications

Per the label Madrigal 2024: decompensated cirrhosis (Child-Pugh B or C) is a contraindication; concomitant strong CYP2C8 inhibitors are contraindicated. Hepatotoxicity is the main labelled warning — ALT/AST elevations were more frequent in resmetirom arms in pivotal trials, and clinically significant drug-induced liver injury occurred at low but non-zero rates. The label advises withholding for ALT >5× upper limit of normal or with symptoms of hepatic dysfunction.

Gallbladder-related events (cholelithiasis, cholecystitis) occurred at higher rates with resmetirom in MAESTRO-NASH, consistent with the known biliary effects of THR-β activation (increased biliary cholesterol secretion). Pregnancy: animal reproductive toxicity data show developmental harm; not recommended during pregnancy. Pediatric and adolescent populations: not studied. Compensated cirrhosis (F4): not studied in the pivotal MASH-fibrosis trial and explicitly outside the labelled indication; safety in cirrhotic patients is under separate study (MAESTRO-NASH-OUTCOMES has a cirrhotic substudy).

Misconceptions

Several confusions warrant explicit handling:

"It's a thyroid pill" — at approved doses, resmetirom does not meaningfully change TSH, total T4, free T4, total T3, or heart rate Harrison et al. 2024. THR-β selectivity is the design point. Patients on stable levothyroxine generally do not require dose change.
"It cures MASH" — at the most effective dose, ~30% of patients achieve histological MASH resolution at one year. Roughly two-thirds do not. The trial's responder definition is biopsy-based; "felt" reversal is small in most patients.
"It replaces lifestyle change" — the trial mandated stable diet and physical activity counseling for all arms. Caloric restriction with 7–10% body weight loss remains the most effective single intervention for MASH, with ~90% NASH resolution rate at that weight-loss threshold in a landmark cohort Vilar-Gomez et al. 2015. Resmetirom is for patients who cannot achieve or sustain that weight loss, or who have already progressed to significant fibrosis.
"Approval means long-term benefit is proven" — accelerated approval is conditional on the confirmatory outcomes trial. Surrogate endpoints (biopsy, MRI-PDFF, FibroScan) are correlated with hard outcomes but the magnitude of cirrhosis/death reduction is unproven and is the open question.

Alternatives

For non-cirrhotic MASH with significant fibrosis, the alternatives are:

Caloric restriction and weight loss. Documented histological MASH resolution rates of 64–90% at ≥7–10% weight loss in a 261-patient cohort with paired biopsies Vilar-Gomez et al. 2015. Most effective single intervention; rarely sustained.
Bariatric surgery. Single most reliable trigger of MASH and fibrosis regression for patients with eligible BMI; substantial perioperative cost and morbidity.
GLP-1 receptor agonists. Semaglutide demonstrated MASH resolution rates of ~59% at 0.4 mg/day SC in a phase 2 trial, though without convincing fibrosis improvement at 72 weeks Newsome et al. 2021. Phase 3 ESSENCE results are pending. Effective on adjacent cardiometabolic dimensions (weight, glycemia, cardiovascular outcomes).
Tirzepatide (GIP/GLP-1 dual agonist). In the SYNERGY-NASH phase 2 trial, 51.8–62.4% of patients achieved MASH resolution at 52 weeks at therapeutic doses, with fibrosis improvement signals Loomba et al. 2024. Not yet approved for MASH but in active development; the most likely near-term competitor to resmetirom.
Pioglitazone and vitamin E — both have AASLD-recommended roles in biopsy-confirmed NASH (vitamin E in non-diabetics; pioglitazone in diabetics), supported by PIVENS and other trials Rinella et al. 2023. Modest histology effects; pioglitazone weight gain and bone effects, vitamin E small prostate cancer signal at high doses.

Failure modes

Diarrhea and nausea in the first weeks drive most discontinuations. Rates: diarrhea ~24–27% (resmetirom) vs ~15% (placebo); nausea ~12–15% vs ~12% in MAESTRO-NASH. Typically mild-moderate, peaks in first 4–8 weeks, attenuates with continued use Harrison et al. 2024.
Insurance denial on biopsy-requirement grounds. While the label does not mandate biopsy, many payors require either a biopsy or a high non-invasive score (FibroScan VCTE ≥8 kPa plus ELF or FIB-4 in defined ranges) before authorising.
Pill avoidance for a "silent" disease. MASH symptoms are minimal until cirrhosis; adherence to a daily pill for a disease the patient does not feel is the predictable real-world failure mode.
Statin-resmetirom interaction sometimes missed at the dispensing step; high-dose statins increase resmetirom AUC and can drive transaminase elevations attributed to either drug.

Practicalities

List price (wholesale acquisition cost) at U.S. launch: approximately $47,400 per year (~$3,950/month) ICER 2024. ICER's 2024 evidence report judged resmetirom incrementally cost-effective only at substantial discount from list (estimated health-benefit price benchmark roughly $39,000–$50,000/year depending on threshold), and emphasised that the cost-effectiveness case rests on assumed long-term cirrhosis reduction ICER 2024. Most U.S. commercial plans cover it with prior authorisation; Medicare Part D coverage has been variable. The manufacturer runs a copay assistance program for commercially insured patients (typical out-of-pocket ~$10/month) and a patient-assistance program for uninsured patients meeting income criteria.

Outside the U.S. (as of mid-2026): not yet approved by EMA, MHRA, or PMDA. Patients in those jurisdictions accessing the drug do so through compassionate use, clinical trials, or out-of-pocket import.

Pre-prescription workup typically includes: liver biopsy or validated non-invasive fibrosis assessment (FibroScan VCTE plus ELF or MRE), lipid panel, comprehensive metabolic panel, TSH, alcohol use history, viral hepatitis serology (to rule out competing causes), and a medication review for CYP2C8 / OATP interactions.

History

Selective thyroid hormone receptor-β agonism as a therapeutic concept dates to the 1990s, when the lipid-clearing effects of native thyroid hormone (powerful but unusable due to tachycardia and bone loss) were dissected at the receptor-isoform level. Early thyromimetics (sobetirome, eprotirome) failed in trials for hypercholesterolemia due to cartilage signals or liver enzyme elevations Karanjia et al. 2024. Resmetirom was developed by Madrigal Pharmaceuticals through structural refinement aimed at hepato-selective uptake plus β-isoform selectivity. Phase 2 results in 2019 Harrison et al. 2019 established the MRI-PDFF signal; phase 3 MAESTRO-NASH read out positive in late 2023; FDA accelerated approval followed in March 2024 FDA 2024, ending a long string of MASH drug-development failures (obeticholic acid, elafibranor, selonsertib, cenicriviroc) and making resmetirom the first targeted MASH pharmacotherapy.

Stakes

The natural history of untreated F2–F3 MASH: roughly 1 fibrosis-stage progression every 7 years on average, with substantial individual variation. Patients with F3 fibrosis face an annual risk of progression to cirrhosis around 5–10%; once cirrhotic, annual rates of decompensation are roughly 3–5% and of hepatocellular carcinoma roughly 1–2% Younossi et al. 2023. MASH is now the leading indication for liver transplant in U.S. women and is climbing rapidly overall. Cardiovascular disease, not liver disease, is the leading cause of death across MASH patients as a group, but for the F3-fibrosis subset liver-related mortality becomes a major contributor over a 10–15-year horizon.

Payoff

For a responder (roughly 30% at one year), the felt experience is small: fewer ALT abnormalities at the next physical, an improved FibroScan number, and (in the lipid-responder subset) a meaningfully lower LDL-C and apoB. Most patients will not notice changed energy, appearance, or daily wellbeing — MASH is largely asymptomatic at F2–F3 and so is its biochemical reversal. The payoff is the trajectory off the F3 → cirrhosis ladder, which is a probabilistic gain visible only through imaging and labs over years.

Out of scope

Resmetirom in compensated cirrhosis (F4) — under separate study; not yet labelled.
Combination therapy with GLP-1 agonists or tirzepatide — biologically rational and under investigation but not yet evidence-supported.
MAFLD without significant fibrosis (F0–F1) — not the labelled indication; lifestyle remains primary.
Pediatric MASH — not studied.
Hepatocellular carcinoma surveillance — separate topic for any patient with F3 fibrosis regardless of resmetirom use.

The credibility range

Optimist case

This is the first drug in two decades of MASH development to meet both co-primary biopsy endpoints in a well-conducted phase 3 RCT, with consistent supporting signal across imaging (MRI-PDFF, MRE), enzymes (ALT, AST, GGT), and non-invasive fibrosis scores (FibroScan, ELF) Harrison et al. 2024. Mechanism is well-characterised at the receptor and downstream-pathway level Karanjia et al. 2024, internally consistent with both the histology and the lipid effects, and predicts cardiovascular benefit on top of liver benefit (a 13–16% LDL-C reduction without lifestyle change is on the order of a low-dose statin effect). The safety profile is favorable: no thyromimetic-axis effects at approved doses, no weight gain, no bone-density signal across 52 weeks, transient GI symptoms as the dominant tolerability issue. The FDA accelerated-approval pathway exists precisely for situations like this — a slow-progressing disease with no prior pharmacotherapy and validated surrogate endpoints — and the confirmatory outcomes trial is enrolling, so we will know within a few years whether biopsy improvement maps to cirrhosis prevention.

Skeptic case

Three concerns:

Histology is a surrogate. No drug for MASH has yet shown reduced cirrhosis, decompensation, or mortality in a confirmatory trial — they have all relied on biopsy or imaging. The mapping from one year of biopsy improvement to ten years of decompensation is plausible but unproven, and prior MASH drugs (e.g., obeticholic acid) showed biopsy signals without obtaining full approval because of safety or because the surrogate-to-outcome bridge was insufficient. The confirmatory MAESTRO-NASH outcomes phase is still 1–2 years from readout.
Absolute effect sizes are modest. 16–20 percentage-point absolute MASH resolution improvement over placebo means ~70% of treated patients do not meet the primary endpoint at one year. The drug is not a high-magnitude effect; it is a real but partial improvement on a hard-to-reverse condition.
Cost-effectiveness is borderline. ICER judged the launch price above the value-based benchmark and called for a discount of 30–60% to be confidently cost-effective at standard willingness-to-pay thresholds ICER 2024. The same epidemiology that makes MASH a public-health concern (~5% of U.S. adults with significant-fibrosis MASH) makes population-scale treatment financially infeasible at list price.

Adjacent skeptic notes: gallstone signal, drug-induced liver injury signal at low rates, complex CYP2C8/OATP interaction profile, and a single sponsor's single pivotal trial without independent replication.

Author's call

Resmetirom is a real but partial therapeutic advance for a previously untreatable condition. It is the right drug to know exists and to use for biopsy-confirmed F2–F3 MASH in patients who cannot achieve or sustain the 7–10% body weight loss that would otherwise drive most of the histological reversal. It is not a substitute for weight loss when weight loss is achievable, and its cost-effectiveness case rests on long-term outcomes data that do not yet exist. The catalogue scores reflect this: high evidence (a well-conducted positive phase 3 trial with consistent supporting data, dampened by surrogate-endpoint reliance and lack of independent replication), high cost burden, very low effort burden, modest longevity effect anchored to a probabilistic disease-trajectory improvement, and modest health_short_term reflecting how undetectable the biochemical reversal is to the patient.

Stakeholder + incentive map

Madrigal Pharmaceuticals — sole sponsor and marketer of Rezdiffra; commercial incentive to maximise eligible-patient pool (and to keep biopsy out of the label, which they succeeded in) and prior-authorisation hurdles low.
Hepatology specialty societies (AASLD, EASL). Long-standing institutional interest in seeing the first MASH drug land; 2023 AASLD guidance had already begun preparing for therapeutic options Rinella et al. 2023. Generally positive but cautious framing in post-approval guidance.
Payors (U.S. commercial, Medicare Part D, Medicaid). Strong cost-control incentive given large eligible population; widespread use of prior authorisation and step therapy. ICER's analysis is a payor-friendly anchor for negotiation ICER 2024.
GLP-1 / GIP-GLP-1 manufacturers (Novo Nordisk, Eli Lilly). Tirzepatide and semaglutide are running fast in MASH; full approval of either would compress resmetirom's market, particularly in the diabetic / obese subset which is most of the indicated population Loomba et al. 2024 Newsome et al. 2021.
Primary care. MASH is increasingly recognised in primary care via FibroScan and FIB-4, but the prescribing pathway runs through hepatology and gastroenterology because of biopsy / non-invasive workup and the interaction profile.
Patient advocacy groups. Generally supportive of access; emphasise the prior absence of any approved pharmacotherapy as the moral case for coverage.

Population variability

Disease stage. Trial population was F1B through F3 with NAS ≥4. Effect sizes are similar across F2 and F3 in the pivotal trial. F1 patients and F4 (cirrhosis) patients were not part of the labelled indication; F4 is under separate study.
Sex. Roughly equal sex distribution in MAESTRO-NASH. No sex-specific signal in either efficacy or safety in the published analyses.
Diabetes. ~67% of MAESTRO-NASH participants had type 2 diabetes; response rates were broadly similar with and without diabetes.
BMI. Mean BMI in trial was ~35 kg/m². The label's weight-based dosing (80 vs 100 mg) is the only explicit accommodation; trial efficacy did not differ meaningfully by BMI band.
Race/ethnicity. Trial population was predominantly non-Hispanic white; representation of Hispanic patients (the population at highest U.S. risk for MASH) was approximately 30%. No clear differential efficacy signal but the precision is modest in subgroup analyses.
Age. Trial enrolled adults; mean age in the late 50s. Pediatric and elderly (75+) data are sparse.
Concomitant statin use. ~50% of MAESTRO-NASH patients were on statins at baseline; resmetirom's LDL-C effect appeared additive to background statin effect.

Knowledge gaps

Hard outcomes data. The confirmatory MAESTRO-NASH outcomes phase reading out around 2027–2028 is the most important data point still missing. Until then, the cirrhosis-prevention claim is mechanistic and surrogate-based.
Long-term safety beyond 52 weeks. Gallstone signal, low-rate hepatotoxicity, and any rare adverse events of THR-β activation (theoretical: pituitary effects, biliary effects) need longer follow-up.
Effect in cirrhosis. Separate F4 substudy will determine whether resmetirom is safe and helpful in compensated cirrhosis.
Combination with GLP-1 / GIP-GLP-1 agonists. Biologically rational, complementary mechanisms (resmetirom: intrahepatic lipid handling; GLP-1: caloric deficit + metabolic remodeling); no trial data.
Optimal duration. Whether responders can stop after sustained histological resolution, and whether the disease returns on discontinuation, is unknown.
Generalisability to populations under-represented in trials (Asian, African-descent populations with MASH).

Scoping notes for this entry, behind-the-curtains rationale a reviewer wouldn't infer from the prose:

Indication scope. The article is held tightly to the labelled indication — non-cirrhotic MASH with F2/F3 fibrosis. Resmetirom in F4 (compensated cirrhosis) is under separate study (the MAESTRO-NASH cirrhotic substudy) and was deliberately excluded; mentioning it in the body would have implied use outside the label.
Action = decide rather than do. Even when the prescription is in hand, the decision to start the drug is a clinician-mediated weigh-up of biopsy/FibroScan, cost, alternatives, and patient preference. The reader's job is to come to that decision informed, not to "do" the drug.
Cost score 5 honestly framed. The list price is genuinely prohibitive at a population level; most individual patients pay much less through copay assistance. The pitch and the practicalities section both name this gap rather than hiding behind the gross number.
Longevity score 3 rather than 4. The mechanism and surrogate-endpoint signal justify a meaningful longevity score, but the absence of confirmatory outcomes data (and the ~70% non-responder fraction at one year) argues against 4. If the MAESTRO-NASH outcomes phase reads out positive around 2027–2028, this entry should be revisited and the longevity score reconsidered upward; editor notes flag this for the future reviewer.
Audience scoped to 40-59 and 60+. MASH with significant fibrosis is overwhelmingly a 40+ disease in clinical practice; trial mean age was late 50s. The label is for adults generally, but rank-card surfacing for the 18-39 cohort would over-imply that this is a candidate intervention for that age band.
Contraindications limited to pregnancy + breastfeeding. The closed contraindication vocabulary does not include "decompensated cirrhosis" or "CYP2C8-interacting drug" or "gallstone history," all of which the article covers in the contraindications section's warning callout. The selected tokens reflect what fits the schema; the rest is in prose.
Highlights paragraph deliberately understates "felt" effect. A patient who expects to feel resmetirom working is set up to quit it. The highlights and payoff sections both name the silent-disease, silent-reversal pattern explicitly.
Separate-entry candidates surfaced during the write: "FibroScan and non-invasive liver-scarring measurement" (cross-cuts screening); "GLP-1 / GIP-GLP-1 agonists for MASH" (will be its own entry once tirzepatide phase 3 reads out); "Vitamin E and pioglitazone for biopsy-confirmed MASH" (the older guideline-backed options, brief enough for one entry covering both).
Future-link candidates: wegovy-semaglutide-for-mash, tirzepatide-for-mash, fibroscan-for-liver-scarring, hepatocellular-carcinoma-surveillance, bariatric-surgery-for-metabolic-disease.
Hard call: no audience block for diabetics specifically. ~67% of MAESTRO-NASH participants had type 2 diabetes; response rates were similar across diabetic and non-diabetic subgroups, so a per-audience block would have added structure without informational gain. The alternatives section flags the diabetic-specific pioglitazone option instead.
The brief named "side-effect profile" and "access considerations" — both are covered (contraindications + failure-modes for side effects; practicalities for access). No narrowing relative to the brief.