Fibrotic MASH: Two Approved Drugs, Two Surrogate Endpoints, One Choice
A patient with newly biopsy-confirmed fibrotic MASH is ready to start pharmacotherapy. The disagreement isn't whether to treat — it's whether an approved, purpose-built drug with a real but modest effect size should lose to an agent with bigger trial numbers and no label for this use.
Dana R., a 54-year-old woman, has worked the overnight shift in a hospital's inpatient pharmacy for almost twenty years, a schedule she blames for most of her own health habits — vending-machine dinners, and a gym membership she pays for and rarely uses. She was flagged for a liver workup after a routine physical found elevated transaminases, and a follow-up biopsy confirmed metabolic dysfunction-associated steatohepatitis with stage 2 fibrosis. Her BMI is 34, her hemoglobin A1c is 6.3 percent — prediabetic range, not yet a formal diabetes diagnosis — and she takes no regular medication for either. Now that fibrotic disease rather than simple steatosis is confirmed, the question isn't whether to start pharmacotherapy; both her hepatologist and her endocrinologist agree stage 2 fibrosis clears that bar. It's which drug actually gets her there first.
Resmetirom was the first agent FDA-approved for noncirrhotic MASH with moderate fibrosis, on the strength of MAESTRO-NASH's biopsy-confirmed endpoints — real, but modest: roughly a quarter of treated patients achieved fibrosis improvement without worsening MASH at one year, well ahead of placebo but still leaving most short of that mark. It is no longer the only option. Semaglutide received accelerated approval for the same indication in August 2025 on the strength of ESSENCE (Sanyal et al.), whose entry criteria describe her almost exactly — biopsy-confirmed MASH with stage 2 or 3 fibrosis — so her stage 2 disease places her inside that trial's population rather than adjacent to it, which is what makes its numbers usable for her rather than merely encouraging. ESSENCE reported both co-primary endpoints at higher rates than MAESTRO-NASH did: 62.9 percent versus 34.3 percent for steatohepatitis resolution, 36.8 percent versus 22.4 percent for fibrosis improvement, each against its own placebo arm. Read carefully, though, that is a comparison between two separate trials with separate placebo groups and no head-to-head arm, which is a weaker thing than it looks. And both approvals rest on the same kind of evidence: histology at 72 weeks treated as a surrogate, with the outcome data that would confirm real clinical benefit still accruing in both programs. Her A1c of 6.3 percent matters here for a reason that has nothing to do with her liver — it is the one value in her chart that decides whether the drug with the better histologic numbers is also the drug she can actually be started on today.
Deciding what to start first
Resmetirom is what I would start today. It was built for this target — MAESTRO-NASH (Harrison et al.) backs its approval with real histologic improvement in exactly her picture, noncirrhotic MASH at stage 2 fibrosis, not just a transaminase number. I'll concede straight away that it is no longer the only labeled option and that ESSENCE's numbers are better on their face. My point is narrower: resmetirom acts on the liver directly through THR-beta, its effect doesn't depend on her losing weight, and if she doesn't tolerate a GLP-1 — which a third of ESSENCE's treated patients had some version of — there is no partial credit. Starting with the liver-directed drug is the more defensible first move, not the more timid one.
I'm not disputing that resmetirom is real. But her fibrosis stage and BMI put her squarely inside semaglutide's own trial population, which reported meaningfully higher rates on both co-primary histologic endpoints than resmetirom's trial did — that's not a marginal gap.
'Built for this exact problem' undersells what semaglutide is doing for her. At a BMI of 34 with an A1c already in the prediabetic range, weight loss and glycemic improvement aren't a side benefit — they're treating the upstream driver of her liver disease at the same time the liver drug treats the downstream result.
Both of you are arguing this as though comparative efficacy decides it, and I'd point out that both drugs are sitting on the same kind of approval — accelerated, granted on a histologic surrogate, with confirmatory outcome data still running. Neither of you is holding a proven mortality benefit. What I'd actually weigh is the direction of her A1c. At 6.3 percent she is prediabetic, which means semaglutide is on-label for her MASH and off-label for everything else about her; she does not yet have the diabetes indication or, at a BMI of 34 without a qualifying comorbidity, a clean obesity-treatment route. That narrows her coverage to the MASH pathway alone, and that pathway is new enough that step therapy through resmetirom is a real possibility rather than a hypothetical one. I'd rather we found that out before we promise her the better number.
Agreed: start resmetirom today, given the clean label match and the predictable coverage pathway — no one on the team treated it as a consolation choice.
Not fully settled: whether semaglutide gets substituted or added once, or if, her A1c crosses into a formal diabetes diagnosis, which would give it a second, independent on-label reason to start and widen her coverage beyond the single MASH pathway it now depends on. The endocrinologist views that crossing as likely within a year or two given her trajectory and does not think ESSENCE's margin should wait on it; the hepatologist isn't willing to trade a liver-directed drug she tolerates for a cross-trial comparison, and wants her 72-week histology before revisiting.