Finding at-risk MASH without a liver biopsy

Two medicines now carry approval to treat the liver disease MASH, and both labels cover only patients whose scarring has reached a middle stage. That leaves an awkward question: how do you find those patients without a liver biopsy, a procedure that needs a needle through the ribs and a day of recovery? A study published in Hepatology Communications on 1 September 2026 proposes an answer built on a stiffness scan called magnetic resonance elastography^[1]. Here is what the paper did, what its numbers mean, and what it cannot settle.

MASH and the fibrosis ladder

MASH, short for metabolic dysfunction-associated steatohepatitis, is the inflamed form of fatty liver disease. Fat builds up in the liver, liver cells swell and die, and the repair process lays down scar tissue. That scarring is called fibrosis, and pathologists grade it from F0, meaning no scarring, up to F4, meaning cirrhosis, the stage where the liver's structure is permanently distorted. The middle stages, F2 and F3, are where the risk of progression becomes real but the damage may still be reversible.

Doctors use the phrase at-risk MASH for the patients most likely to progress and most likely to benefit from treatment. The problem is that trial protocols and guidelines have not agreed on one definition. A stricter version requires a biopsy activity score, the NAFLD Activity Score, of 4 or more with points in each of three features: fat, inflammation and cell swelling. The new paper tests a simpler one: MASH plus fibrosis stage 2 or higher^[1].

Why the definition is suddenly urgent

Until recently the definition was an academic argument, because there was no approved treatment to give anyone either way. That has changed. The semaglutide label now includes treatment of noncirrhotic MASH with moderate to advanced fibrosis, stages F2 to F3, in adults, under accelerated approval^[3], and the thyroid hormone analogue resmetirom carries a similar indication. The phase 3 ESSENCE trial behind the semaglutide indication enrolled exactly that population, 1,197 patients with biopsy-confirmed MASH at fibrosis stage 2 or 3, and at the 72-week interim analysis semaglutide 2.4 mg once weekly resolved steatohepatitis without worsening fibrosis in 62.9% of patients against 34.3% on placebo^[2].

So the drug exists and the label names the patient. The bottleneck is identification. ESSENCE required a biopsy to get in^[2], and biopsies are not a realistic screening tool for a disease this common. Every clinic that wants to treat MASH now needs a non-invasive way to answer one question: is this patient at stage F2 or above?

What the study found

The researchers studied 413 patients with suspected or diagnosed fatty liver disease at two medical centres. Every patient had both a liver biopsy and magnetic resonance elastography, so the scan could be checked against the tissue result^[1]. The simplified definition, MASH with fibrosis stage 2 or higher, flagged 122 patients as at-risk, against 77 under the stricter activity-score definition^[1].

The scan's liver stiffness reading, abbreviated LS-MRE, was the single most accurate test for picking out at-risk patients under the simplified definition, with an area under the curve of 0.91^[1]. Area under the curve, or AUC, is a standard accuracy score where 1.0 is perfect and 0.5 is no better than chance, so 0.91 is a high mark for a diagnostic test. The same reading scored 0.93 for cirrhosis^[1]. Both figures beat FIB-4, a common blood-test score, and the individual blood markers AST, ALT and platelet count^[1].

The paper then re-tuned the stiffness cutoffs. Current liver-society guidance works with a band of 3.1 to 4.4 kilopascals, the pressure unit the scanner reports. The study's optimized band of 2.8 to 6.4 kilopascals caught 25 additional biopsy-proven treatment candidates, 28% of everyone eligible, by raising sensitivity from 50% to 80%^[1]. The trade was specificity, which fell from 88% to 73%, meaning more false alarms that would need a follow-up look^[1].

What MR elastography actually is

Magnetic resonance elastography, or MRE, is an add-on to a standard MRI scanner. A small paddle rests on the skin over the liver and sends gentle vibrations through the body. The scanner tracks how the waves move through liver tissue, and software converts the wave pattern into a stiffness number. Scarred liver is stiffer than healthy liver, the same way a callus is firmer than the skin around it. The patient feels a mild buzz and lies still for a few minutes. No needle, no radiation, no sedation.

That matters because the main way to confirm fibrosis stage today is still the biopsy, and blood scores like FIB-4 are screening tools rather than staging tools^[1]. A scan that stages fibrosis accurately enough to guide treatment could sit between the blood test and the needle.

What this study cannot settle

This was a prospective study at two centres with 413 patients, a reasonable proof of concept but not a validation in general clinic populations^[1]. The optimized cutoffs were derived from this same dataset, so they need testing in independent groups before anyone changes practice on them. MRE also requires an MRI scanner fitted with the elastography package, which limits where it can be offered and what it costs. And the simplified definition was built for breadth, catching more candidates, which by design accepts more false positives^[1].

Practical context

If larger studies confirm these cutoffs, the pathway for a patient with fatty liver disease could look very different: a blood test to screen, an MRE scan to stage, and a biopsy reserved for the unclear cases. For the treatment side of that pathway, our semaglutide page tracks the MASH indication and the evidence behind it. For now this paper is a careful argument that the scan can find the patients the new drugs are for. The next step is proving it in clinics that were not part of the study.

This article is educational and is not medical advice. Decisions about testing or treatment for liver disease belong with a qualified healthcare provider who knows your history.