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 new targets and therapies in chronic liver disease, the growing push to classify colitis by biology and behaviour rather than by symptoms alone, and quality and patient selection in everyday luminal practice. Let's dive in.

The liver dominates this week, starting with a question that has hung over primary biliary cholangitis for years: is an alkaline phosphatase just under the traditional threshold of 1.67 times the upper limit of normal good enough? Two phase three trials, from two journals, answer from the same direction. In the Journal of Hepatology, Levy and colleagues report the IDEAL trial, which randomised 96 patients on or intolerant of ursodeoxycholic acid, with alkaline phosphatase between normal and 1.67 times normal, to seladelpar 10 milligrams or placebo for a year [1]. About three quarters of patients on seladelpar normalised their alkaline phosphatase with at least a fifteen percent fall, compared with roughly one in six on placebo, and the effect appeared within the first month. Serious adverse events were actually less frequent on the drug. In Hepatology, Kowdley and colleagues report ELSPIRE, a similar design testing elafibranor 80 milligrams in 92 patients [2]. Here about 85 percent of treated patients normalised alkaline phosphatase against roughly a quarter on placebo, and more patients reached complete biochemical response, with no new safety signals. The primary endpoints were defined slightly differently, so the two drugs cannot be ranked from these data, and both trials are small, a year long, and built on a biochemical surrogate rather than transplant or death. Still, together they show that normalisation is achievable in this near-responder group, and they strengthen the argument that the old threshold may be too permissive, particularly for higher-risk patients, though guidelines have not yet formally moved the target.

In metabolic liver disease, The Lancet carries a phase 2b trial led by Lawitz of efinopegdutide, a once-weekly dual GLP-1 and glucagon receptor co-agonist, in 381 adults with biopsy-proven steatohepatitis and stage two or three fibrosis across 26 countries [3]. At one year, roughly six in ten patients on efinopegdutide achieved resolution of steatohepatitis without fibrosis worsening, against about one in eight on placebo. Fibrosis improvement was significant at the 4 and 7 milligram doses, with about half of patients on 7 milligrams improving by at least a stage, but the 10 milligram dose did not reach significance on that endpoint and carried the most gastrointestinal effects, with about one in ten patients stopping treatment. An open-label semaglutide arm landed in a similar range for resolution but lower for fibrosis, although that arm was not blinded and was not a formal head-to-head comparison. This is a dose-finding study, so the signal at the middle dose is promising but awaits phase three confirmation.

Staying with viral and advanced disease, Hepatology also publishes an American Association for the Study of Liver Diseases expert practice opinion on hepatitis delta, written by Lim and colleagues [4]. The group recommends that every person who is hepatitis B surface antigen positive be tested at least once for delta antibody, with RNA testing to confirm active infection, elastography-based fibrosis assessment, and six-monthly surveillance. Its central position is that all adults with viraemic delta infection and compensated disease are candidates for bulevirtide, with added peginterferon as a finite-duration option. This is consensus guidance from a targeted literature review, not a systematic guideline, but it sets a clear benchmark for screening. And for patients already decompensated, a third Hepatology paper, from Calleri and colleagues, examined a remote monitoring programme after hospital discharge, combining transmitted vital signs and symptoms with round-the-clock virtual nursing [5]. Among 430 patients, about two in five enrolled patients were readmitted for liver causes within 90 days, compared with close to six in ten contemporary controls, roughly a forty percent lower risk; the comparison with historical controls pointed the same way but fell short of significance, and mortality did not differ. Benefits looked strongest in sicker patients and in those with alcohol-associated disease. Being observational and from a single programme, this supports remote follow-up as a promising model rather than a proven one.

Our second theme is colitis, and the common thread is an attempt to sort patients by how their disease actually behaves. In Nature Medicine, Schreiber and colleagues used patient-reported outcomes from a head-to-head trial of 768 participants on vedolizumab or adalimumab to define four distinct response trajectories in ulcerative colitis, then linked vedolizumab trajectories to long-term complications in a second trial [6]. Tissue profiling pointed to vascular inflammation, driven by interactions between a macrophage subset and endothelial cells, as a feature of vedolizumab non-response. Strikingly, in two real-world cohorts totalling nearly nine thousand patients, concomitant use of renin-angiotensin system inhibitors was associated with fewer long-term complications on vedolizumab but not on anti-cytokine drugs. The authors are explicit that this taxonomy requires prospective validation, and the blood pressure drug association is hypothesis-generating rather than a basis for combination therapy.

A more pragmatic comparison comes from Inflammatory Bowel Diseases, where Mizuta and colleagues compared infliximab with tofacitinib in a Japanese multicentre cohort, analysing advanced therapy-naive and experienced patients separately [7]. In patients already exposed to advanced therapy, steroid-free remission at one year roughly doubled with tofacitinib, to just under forty percent. In therapy-naive patients the numbers also leaned toward tofacitinib, but that difference was not statistically significant, and three-year persistence did not differ in either group. With fewer than two hundred patients in the primary analysis, the authors themselves describe the findings as exploratory.

Colitis caused by cancer immunotherapy is the focus of a Clinical Gastroenterology and Hepatology study by Ibraheim and colleagues across two United Kingdom centres [8]. In 250 patients, they proposed a simple endoscopic and histological classification: mild inflammation, with or without microscopic colitis, versus moderate-to-severe colitis, which made up a fifth of patients. Bloody stool, anti-CTLA-4 therapy, and PD-1 or PD-L1 blockade combined with chemotherapy independently predicted the severe phenotype, which carried more hospitalisation, steroid-refractory disease and surgery. In the mild group, oral second-generation corticosteroids achieved remission in more than eight in ten patients, and rechallenge with immunotherapy led to recurrence in roughly a third of patients, usually no worse than before. As a retrospective framework, it offers a rationale for steroid-sparing approaches in milder disease, pending prospective testing.

Our final theme is quality and selection in everyday luminal practice. In Clinical Gastroenterology and Hepatology, Chen and colleagues followed nearly 173 thousand patients from the Mass General Brigham system for a median of nine years, identifying 233 post-colonoscopy cancers [9]. Lower quality on almost every indicator tracked with higher risk, and the authors estimate that raising adenoma detection to at least 30 percent in women and 40 percent in men could have prevented close to half of these cancers. Importantly, detection rate and polyp findings acted independently: even patients with no polyps had nearly triple the risk after a low-detection colonoscopy, and those with high-risk polyps and a low-detection endoscopist faced roughly an eightfold higher risk than polyp-free patients scoped by high performers. This is observational, but it adds weight to the view that a clean colonoscopy is only as reassuring as the endoscopist who performed it.

Also in Clinical Gastroenterology and Hepatology, Black and colleagues revisited the ATLANTIS trial of low-dose amitriptyline in irritable bowel syndrome, clustering 401 participants by gastrointestinal and psychological symptoms [10]. Across clusters, the treatment effect did not differ significantly overall. But patients with predominantly psychological symptoms and milder gut symptoms improved more on amitriptyline than placebo, while those with severe symptoms in both domains were more likely to stop medication and reported more side effects. These are post hoc, exploratory analyses, so they hint at who may benefit rather than define it.

If you only have time for one paper this week, make it the IDEAL trial of seladelpar from Levy and colleagues [1]. Read alongside ELSPIRE, it directly challenges the long-accepted 1.67 threshold in primary biliary cholangitis and reopens the question of what an adequate response to first-line therapy really means.

That's your roundup for This Week in Gastroenterology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

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