This Week in Gastroenterology — Sep 22, 2026

Welcome to This Week in Gastroenterology. This week we're covering 10 notable papers spanning liver disease burden and risk stratification, inflammatory bowel disease therapeutics, and the endoscopic management of oesophageal and hereditary neoplasia. Let's dive in.

We start with hepatology, where three papers together map the scale of chronic liver disease and our ability to triage it. In The Lancet Gastroenterology and Hepatology, Liu and colleagues report baseline data from the CHESS-LiverHealth project, a community-based cohort that performed transient elastography on more than twenty-five thousand adults across eighteen sites in China [1]. Steatotic liver disease was present in about 44 percent of participants, overwhelmingly metabolic dysfunction-associated, with much smaller contributions from combined metabolic and alcohol-related disease and from alcohol-associated liver disease alone. Liver stiffness at or above eight kilopascals, their fibrosis threshold, was found in about one in twenty-three participants overall, but rose to roughly one in nine among those with obesity and similarly among those with type 2 diabetes. That gradient is the practical message: community screening is only worth doing where cardiometabolic risk concentrates. The companion question is what to do with those people once you have found them, and that is addressed in JAMA Internal Medicine, where Mezzacappa and colleagues applied nine published risk scores to more than eight hundred and fifty thousand United States veterans with imaging-confirmed steatotic liver disease and no cirrhosis [2]. Over ten years, about 4 percent of the cohort developed cirrhosis and about a third of one percent developed hepatocellular carcinoma. The steatosis-associated fibrosis estimator, the SAFE score, gave the greatest net benefit for predicting cirrhosis and looks reasonable for deciding when to repeat fibrosis assessment. For hepatocellular carcinoma, though, every score performed poorly in decision-curve terms, and the authors conclude that these tools have minimal utility for guiding cancer surveillance decisions in non-cirrhotic steatotic liver disease. So risk-stratify for fibrosis, but do not use these scores to start or withhold cancer screening.

Still in hepatology, two papers refine how we classify patients. In Gut, van Velsen and colleagues report the RADICAL consortium's natural history of strictly defined immune tolerant chronic hepatitis B, following nine hundred and fifty-one patients for a median of thirteen years [3]. This phase is far less stable than the label implies: about half of patients transitioned to immune active disease within five years, and roughly seven in ten had done so by ten years. Patients with high-normal alanine aminotransferase were at substantially greater risk, with values above thirty units per litre associated with around a threefold increase in transition. Reassuringly, progression to significant fibrosis was uncommon at about 6 percent over fifteen years, and hepatocellular carcinoma occurred in about one percent. The implication is not immediate treatment for everyone, but tighter monitoring, and a genuine discussion of pre-emptive therapy where the alanine aminotransferase sits in the upper normal range. And in Clinical Gastroenterology and Hepatology, Kerbert and colleagues describe the modified Hepatic Encephalopathy Staging Tool, developed against the well-known unreliability of West Haven criteria [4]. Across forty-nine clinicians rating video vignettes and then a prospective cohort of one hundred and twenty-eight hospitalised patients with decompensated cirrhosis, inter-rater reliability with the new tool was near-perfect, compared with only fair agreement for West Haven. Plasma ammonia rose stepwise across the new grades but not across West Haven grades, and low grades independently predicted better one-year survival. It has been accepted by the United States Food and Drug Administration as a trial endpoint, so expect to see it in encephalopathy trials, and arguably on the ward.

Turning to inflammatory bowel disease, two papers in Clinical Gastroenterology and Hepatology address how we rescue and how we monitor. Ghimire and colleagues report a multicentre retrospective cohort across thirteen Australian centres of one hundred and fifty adults with acute severe ulcerative colitis treated with a Janus kinase inhibitor, with complete follow-up to fifty-two weeks [5]. Roughly a third of the whole cohort came to colectomy by one year. Colectomy was significantly less frequent with upadacitinib than with tofacitinib, about 23 percent versus about 43 percent, an absolute difference of roughly eighteen percentage points, with upadacitinib also achieving higher rates of clinical, corticosteroid-free, and biochemical remission. Importantly, the difference only emerged at fifty-two weeks, not at earlier timepoints. Sequential salvage therapy, meaning a Janus kinase inhibitor after a failed first rescue agent, produced a numerically higher colectomy rate than first-line rescue, but that difference was not statistically significant, which the authors read as evidence that sequential salvage remains a legitimate option for patients otherwise heading to theatre. This is retrospective data with the confounding that implies, but the safety profile was reassuring and the signal favours upadacitinib. Alongside that, Soliman and colleagues pooled twenty-three studies and nearly six thousand eight hundred patients on vedolizumab trough concentrations [6]. Higher troughs were most robustly associated with endoscopic remission in ulcerative colitis, and with clinical remission in both ulcerative colitis and Crohn's disease, though those clinical estimates were heterogeneous, and there was no association with endoscopic remission in Crohn's disease. The suggested maintenance targets were roughly thirteen to fifteen micrograms per millilitre in ulcerative colitis and ten to twelve in Crohn's. The authors are appropriately cautious: these are cross-sectional associations that may reflect reverse causation, where inflamed patients clear drug faster, and they explicitly say prospective trials are needed before proactive drug monitoring can be recommended.

Our third theme is oesophageal and hereditary neoplasia, where the question is whether surveillance is reaching the right people. In Gut, Boer and colleagues interrogated nationwide Dutch registries covering nearly nine thousand patients treated for Barrett-related dysplasia or oesophageal adenocarcinoma [7]. Ninety percent of them presented de novo, with no prior endoscopy showing non-dysplastic Barrett's. Even among patients who received organ-preserving endoscopic therapy, about two thirds of those patients had never been in a surveillance programme. Among the minority who had been under surveillance, disease was overwhelmingly early stage and missed advanced neoplasia was uncommon, at around 11 percent. So surveillance works for the people in it; the problem is that it captures almost none of the people who go on to need treatment, which is an argument for evaluating targeted screening rather than for intensifying current surveillance intervals. What happens after resection is addressed by Ishihara and colleagues in the American Journal of Gastroenterology, a multicentre prospective cohort of three hundred and eighty-five patients with oesophageal cancer invading muscularis mucosae or superficial submucosa after endoscopic resection [8]. Five-year overall survival was above ninety percent for both squamous cell carcinoma and adenocarcinoma. Among lymphovascular-invasion-negative patients managed with surveillance alone, five-year survival was around 94 percent, and for flat squamous lesions thirty millimetres or smaller the cumulative metastatic recurrence rate was under 3 percent. Notably, second primary oesophageal cancers occurred in about a quarter of squamous patients within five years, so surveillance is reasonable for most lymphovascular-invasion-negative disease, but it must be diligent surveillance. Also in endoscopy, Nabi and colleagues report in Gastrointestinal Endoscopy a randomised trial of one hundred patients undergoing peroral endoscopic myotomy for type one or two achalasia, comparing a gastric myotomy shorter than three centimetres with one of three centimetres or more [9]. At six months, significant reflux oesophagitis, Los Angeles grade B or worse, occurred in about 24 percent with the short myotomy versus about 44 percent with the long one, roughly halving the risk, and upright acid exposure was lower with the short myotomy. There was no difference in reflux symptoms, proton pump inhibitor use, total acid exposure time, or clinical efficacy by Eckardt score. That is a technical change you can act on. Finally, in the American Journal of Gastroenterology, Zare and colleagues analysed nineteen hundred patients with familial adenomatous polyposis from a prospectively maintained registry [10]. Relatives identified through family screening now have mortality comparable to the general population, while patients presenting symptomatically remain at markedly elevated risk. Colorectal disease is still the leading cause of death but fell from about two thirds of deaths before 2001 to under a third after 2013, while duodenal, gastric, and non-polyposis-related deaths all rose. The message for our clinics is that the burden in polyposis has migrated upstream, into the duodenum and stomach, and into general medical care.

If you only have time for one paper this week, make it the Australian Janus kinase inhibitor cohort in acute severe ulcerative colitis [5]. Rescue therapy decisions in this setting are made under pressure with thin evidence, and this is the largest comparative dataset yet on two agents you may already be reaching for.

Here are the key takeaways from this week in Gastroenterology. First, steatotic liver disease is present in over four in ten community-dwelling Chinese adults, but meaningful fibrosis concentrates in obesity and diabetes, so screen where the metabolic risk is. Second, the SAFE score can help you decide when to repeat fibrosis assessment, but no current score reliably guides cancer surveillance in non-cirrhotic steatotic liver disease. Third, so-called immune tolerant hepatitis B is mostly a transient state, and a high-normal alanine aminotransferase should prompt closer follow-up. Fourth, in acute severe colitis, upadacitinib outperformed tofacitinib on one-year colectomy, and sequential salvage remains reasonable before surgery. Fifth, most Barrett-related cancers arrive outside surveillance entirely, and during peroral endoscopic myotomy, a gastric myotomy under three centimetres roughly halves significant reflux oesophagitis without losing efficacy.

That's your roundup for This Week in Gastroenterology. The full transcript and references are available on the episode page in your AudioScholar library. 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 think of one colleague — in any specialty — who never has time to keep up with the literature. Tell them about AudioScholar: a free ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

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