This week in Gastroenterology

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Gastroenterology. This week we're covering 10 notable papers spanning non-invasive assessment and acute complications of chronic liver disease, comparative choices among advanced therapies in inflammatory bowel disease, and less invasive approaches to oesophageal and upper gastrointestinal care. Let's dive in.

Liver disease leads this week, starting with how we stage metabolic liver disease without a needle and ending with two of its most dangerous complications. In The Lancet, Vali and colleagues report LITMUS, a prospective European study that also enrolled in the United States, comparing 12 single markers and 18 multimarker panels against centrally scored biopsy in 1,590 adults with metabolic dysfunction-associated steatotic liver disease [1]. The authors set a demanding bar, an area under the curve above 0.80, as the threshold at which a test might plausibly replace biopsy. For at-risk steatohepatitis, meaning active disease with at least moderate fibrosis, only one test cleared that bar: the SomaSignal proteomic panel, with an area under the curve of about 0.83. No test reached it for steatohepatitis alone. Advanced fibrosis was easier territory, with SomaSignal again on top at about 0.92 and several panels, including FibroMeter combined with transient elastography, Agile 3 plus, and NIS4, also clearing the threshold. For cirrhosis, FibroMeter with elastography performed best at about 0.94. Single markers on their own were variable and limited, and a sequential approach beginning with gamma-glutamyl transferase met the threshold for both advanced fibrosis and at-risk disease. With approved drug therapy now arriving, these data strengthen the case that combined panels can identify fibrosis reliably, while detecting active inflammation without biopsy remains the weak point; proteomic testing is not yet widely available, and guidelines have not absorbed these results. A complementary review in Hepatology by Mironova and colleagues makes the parallel case for imaging [2]. They summarise evidence that MRI proton density fat fraction is the most accurate non-invasive steatosis measure, with a relative reduction of at least thirty percent tracking histologic response, and that magnetic resonance elastography is the most accurate imaging marker of fibrosis, with each one kilopascal rise in stiffness linked to roughly 43 percent higher risk of decompensation. They also note that iron-corrected T1 is sensitive but nonspecific and that its prognostic value remains unproven. Read together, the two papers point toward a future where blood panels and MRI share the work biopsy used to do, though head-to-head comparisons are still needed.

At the sicker end of the spectrum, the Journal of Hepatology publishes a global multisociety consensus on acute-on-chronic liver failure [3]. Competing regional definitions have long made trials in this syndrome hard to compare, and this harmonisation effort offers a unified framework for diagnosis, grading, organ failure assessment, critical care, transplantation, and regulatory-aligned trial endpoints. It is a consensus document rather than new outcome data, but its value lies in giving researchers and clinicians on different continents a common language. In Alimentary Pharmacology and Therapeutics, Zhu and colleagues pooled 44 mostly small, retrospective studies covering 458 patients with bleeding ectopic varices [4]. Duodenal, rectal, and stomal sites were the most common. Sclerosant injection appeared to rebleed far less often than band ligation for rectal varices, and TIPS combined with variceal embolisation appeared to outperform TIPS alone. Mortality was sobering, reaching about two thirds of patients after stomal variceal bleeding over roughly a year of follow-up. The authors are candid that heterogeneity, missing data, and tiny series prevented formal meta-analysis, so these comparisons are hypothesis-generating; their conclusion favours individualised care with early multidisciplinary input.

Inflammatory bowel disease supplies this week's only randomised trial, along with several real-world comparisons of advanced therapies. In Clinical Gastroenterology and Hepatology, Pal and colleagues report BEST-CD, which randomised 106 adults with symptomatic Crohn's strictures shorter than three centimetres to balloon dilation or endoscopic stricturotomy [5]. Clinical recurrence at one year was about 30 percent with stricturotomy versus about 55 percent with dilation, and reintervention was roughly halved. Emergency visits and hospitalisations also fell. Surgery rates did not differ significantly, and adverse events were similar between arms. This is a single trial from an expert setting, and the authors themselves call for multicentre validation, but it is the first randomised evidence that stricturotomy may be more durable than the long-standing standard for short strictures.

Three papers then address which drug, for whom. In Inflammatory Bowel Diseases, Cassinotti and colleagues used propensity matching across 1,406 patients with ulcerative colitis to compare continuous clinical remission at one year, a strict endpoint requiring steroid-free remission without dose escalation or drug-limiting adverse events [6]. Ustekinumab achieved the highest rates overall and in biologic-naive patients, golimumab the lowest, and in biologic-exposed patients upadacitinib was the most effective. Ustekinumab also had the fewest adverse events, while upadacitinib had the most, though withdrawals for them were uncommon. Also in Inflammatory Bowel Diseases, Hradsky and colleagues used the nationwide Czech CREdIT registry to compare risankizumab and upadacitinib in mostly treatment-experienced Crohn's disease [7]. Upadacitinib carried close to double the risk of discontinuation, an effect concentrated in patients with ileal involvement, with no difference in isolated colonic disease. The interaction only narrowly reached significance, so disease location as a selection factor is an intriguing but early signal. Both studies are retrospective, and persistence is an imperfect stand-in for efficacy. Safety, particularly in older patients, is the focus of a meta-analysis of 79 studies in Clinical Gastroenterology and Hepatology from Le Cosquer and colleagues [8]. Among patients starting therapy at fifty or older, JAK inhibitors carried roughly a third higher odds of infection than other advanced therapies, while interleukin-23 and interleukin-12/23 agents carried about a quarter lower risk of serious infection than anti-TNF drugs. There was no overall increase in cardiovascular events, thrombosis, or cancer. A thromboembolic signal in patients over sixty appeared in rheumatology cohorts but not in inflammatory bowel disease cohorts, which hints that the underlying disease may shape JAK inhibitor risk. Taken together, these observational data place ustekinumab-class and interleukin-23 agents favourably on safety, and position JAK inhibitors as effective later-line options carrying an infection trade-off, with the caveat that none of these comparisons is randomised.

Finally, two papers in Alimentary Pharmacology and Therapeutics address upper gastrointestinal care. Kumar and colleagues tested the capsule sponge as a monitoring tool in eosinophilic oesophagitis across three NHS hospitals, with same-day endoscopy as comparator in 136 adults without strictures [9]. Nearly 93 percent of attempts succeeded, overall accuracy was about 85 percent, and sensitivity was about three quarters, meaning roughly one in four active cases would be missed. Just under four in five patients preferred it to endoscopy, and there were no complications. The study supports the sponge as a less invasive monitoring option, with sensitivity the main limit to replacing biopsy outright. Mansk and colleagues pooled nine studies of more than 34,000 patients on antiplatelet therapy comparing potassium-competitive acid blockers with proton pump inhibitors [10]. Gastrointestinal bleeding did not differ significantly. Fewer cardiovascular events and revascularisations with the newer agents came only from observational studies, and the authors stress that these data establish neither non-inferiority nor cardiovascular benefit, so randomised trials are needed.

If you only have time for one paper this week, make it LITMUS in The Lancet [1]. As drug therapy for steatohepatitis becomes reality, it offers the largest biopsy-anchored comparison yet of which non-invasive tests can stand in for histology, and it shows clearly where they still fall short.

Here is what this week's evidence adds up to in Gastroenterology. First, multimarker panels now reliably detect advanced fibrosis and cirrhosis in metabolic liver disease, while identifying active steatohepatitis without biopsy remains largely unsolved outside a single proteomic test. Second, one randomised trial supports endoscopic stricturotomy over balloon dilation for short Crohn's strictures in expert hands, pending multicentre confirmation. Third, observational data favour interleukin-23 and ustekinumab-class agents on safety in older patients and suggest JAK inhibitors perform best in biologic-exposed disease, while disease location as a factor in choosing Crohn's therapy remains a preliminary signal. Fourth, the capsule sponge shows promise for monitoring eosinophilic oesophagitis but misses some active disease. And finally, there is no evidence yet that potassium-competitive acid blockers protect against bleeding better than proton pump inhibitors in patients on antiplatelets.

That's your roundup for This Week in Gastroenterology. The full transcript and references are available on the episode page. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.