Monthly gastroenterology bulletin — September 28, 2026

Colleagues: this issue covers what was published, presented or approved between August 28 and September 28, 2026. It was a month with more regulatory and industry signals than phase 3 trials published in journals; in East Asia, Russia, the Middle East and Turkey we found no verifiable primary publications from the period, so we did not pad. Thirteen items, each with a critical reading and a link to the primary source.

Hepatology

1. Baveno VIII: portal hypertension management turns toward preventing decompensation

[Guideline] [Europe, Baveno Cooperation / EASL]
The Baveno VIII consensus (272 statements, more than 80 experts, 80% agreement threshold) was published in the Journal of Hepatology and disseminated in September. Concrete changes: clinically significant portal hypertension (CSPH) is ruled out with liver stiffness ≤15 kPa plus platelets ≥150 × 10⁹/L; it is ruled in with stiffness ≥25 kPa (viral, alcohol or non-obese MASLD), spleen stiffness >55 kPa or a ≥75% probability in validated models; carvedilol is recommended to prevent decompensation when that probability exceeds 75%, and "recompensation" is defined (no ascites, encephalopathy or rebleeding for more than six months, with Child A). It adds a pediatric chapter for the first time and consolidates pre-emptive TIPS within 72 hours (ideally 24) of high-risk variceal bleeding.
Why it matters: this is the reference guideline for the next decade, but many of the new statements (recompensation, pediatrics, vascular liver disease) rest on low-level evidence and expert opinion; in the Dominican Republic the bottleneck remains timely access to elastography and TIPS.
https://www.journal-of-hepatology.eu/article/S0168-8278(26)02778-9/fulltext
https://news.gastro.org/issues/2026/september-2026/baveno-viii-shifts-portal-hypertension-care-toward-prevention/

2. Elafibranor in primary biliary cholangitis: two years of controlled data from ELATIVE

[Clinical trial] [Multinational; sponsored by Ipsen]
The Journal of Hepatology published (August 27) the 104-week results of the phase 3 ELATIVE trial (161 patients with PBC and inadequate response to ursodeoxycholic acid, randomized 2:1 to elafibranor 80 mg or placebo). Biochemical response at week 104: 64.3% (18/28) vs 0% (0/11); alkaline phosphatase normalization 10.7% vs 0%; pruritus (numerical rating scale) fell 4.1 points with elafibranor and rose 0.3 with placebo. In the open-label extension, 58.8% (47/80) maintained response at week 104, with no unexpected safety findings.
Why it matters: it confirms the durability of the biochemical and pruritus effect, but with very small numbers at late cutoffs (28 vs 11), descriptive analyses without hypothesis testing and no hard clinical outcomes; access remains limited in our region.
https://www.sciencedirect.com/science/article/pii/S016882782602859X

3. Alcohol and cardiometabolic factors: absolute 10-year cirrhosis risk in 354,624 Danes

[Publication] [Denmark]
Prospective population cohort (Danish National Health Survey linked to registries; median follow-up 9.7 years; 1,211 incident cirrhoses). Absolute 10-year risk was 0.14% at 1–7 drinks per week, 0.68% at 15–28, 2.01% at 29–42 and 4.64% above 42 (HR 22.8; 95% CI 18.5–28.1). Obesity (HR 1.50) and smoking roughly doubled risk at low and intermediate intake but added no risk above 42 drinks. Population attributable fractions: alcohol 59.9%, hypertension 29.3%, smoking 21.2%, obesity 13.8%, diabetes 9.0%.
Why it matters: it provides absolute figures useful in the clinic and reinforces that alcohol dominates risk even in the MASLD era; limitations: 54–60% survey response, single self-reported exposure and an outcome limited to hospital-coded cirrhosis. Published online August 24, right at the edge of this month's window.
https://www.sciencedirect.com/science/article/pii/S0168827826028606

4. Spleen stiffness ≤40 kPa rules out high-risk varices in chronic portal vein thrombosis without cirrhosis

[Publication] [France and 15 VALDIG centres, including Russia and India]
Retrospective diagnostic-accuracy study (derivation at Beaujon, n = 159; validation in 15 centres, n = 187) published online September 1 in the Journal of Hepatology. With spleen stiffness by transient elastography ≤40 kPa, sensitivity was 97% and negative predictive value 97% in derivation (96% in validation), sparing 41–43% of endoscopies with 3–5% of high-risk varices missed.
Why it matters: it extends the Baveno logic to a population without non-invasive criteria, but it is retrospective, used two different FibroScan devices, had few XL-probe cases and only 61–72% of endoscopies within six months of measurement; prospective validation is needed before omitting endoscopy.
https://www.sciencedirect.com/science/article/pii/S0168827826028758

Inflammatory bowel disease and the intestine

5. Anti-IL-15 antibody (TEV '408) protects the duodenal mucosa during gluten challenge in celiac disease

[Industry] [Israel, Teva]
On September 2 Teva reported a randomized, placebo-controlled phase 2a trial (n = 50): a single subcutaneous dose followed by a six-week gluten challenge. Change in villus height-to-crypt depth ratio at week 8: −0.43 with TEV '408 vs −0.88 with placebo (difference 0.45; 95% CI 0.06–0.84; p < 0.05); intraepithelial lymphocyte density: +0.37 vs +27.6 (difference −27.2; 95% CI −39.7 to −14.8). No safety signals reported.
Why it matters: this is the first solid histological signal for an anti-IL-15 in celiac disease, but the data are preliminary press-release results, not peer reviewed, with a small sample and a surrogate endpoint; it does not change practice (the gluten-free diet remains the treatment).
https://www.globenewswire.com/news-release/2026/09/02/3354990/0/en/teva-announces-positive-topline-results-from-phase-2a-study-in-celiac-disease-for-its-anti-il-15-antibody-further-validating-its-pipeline-in-a-product-potential.html

6. SPY003 (long-acting anti-IL-23) in ulcerative colitis: SKYLINE Part A

[Industry] [United States, Spyre Therapeutics]
On September 8 Spyre presented the open-label, uncontrolled Part A of its phase 2 trial (n = 44; 41% previously exposed to advanced therapies). At week 12 the Robarts histological index fell 10.0 points (p < 0.0001), with 20% clinical remission, 30% endoscopic improvement and a single drug-related adverse event (grade 1 pruritus).
Why it matters: it completes proof of concept for the three components the company plans to combine (anti-α4β7, anti-TL1A, anti-IL-23), but without placebo or blinding these figures cannot be compared with the class; the randomized Part B with combinations is not due until 2027.
https://www.globenewswire.com/news-release/2026/09/08/3357613/0/en/spyre-announces-potential-best-in-class-spy003-anti-il-23-part-a-induction-results-from-skyline-trial-completing-proof-of-concept-for-all-three-components-of-its-ibd-combinations.html

7. Glepaglutide (GLP-2 analog) receives a positive CHMP opinion for short bowel syndrome

[EMA regulatory] [Denmark / European Union]
At its September 14–17 meeting the CHMP recommended approval of Zeydovio (glepaglutide, two subcutaneous injections per week) for adults with short bowel syndrome. In the phase 3 EASE-1 trial (n = 106, dependent on parenteral support ≥3 days/week), the twice-weekly regimen reduced weekly parenteral volume by 5.13 L vs 2.85 L with placebo (p = 0.0039); ≥20% response: 65.7% vs 38.9%; nine treated patients reached enteral autonomy vs none on placebo.
Why it matters: it would be the first advance in Europe in more than a decade for a small population with an enormous care burden; the European Commission decision is expected within about two months and U.S. registration still depends on the EASE-5 trial.
https://www.ema.europa.eu/en/news/meeting-highlights-committee-medicinal-products-human-use-chmp-14-17-september-2026
https://www.globenewswire.com/news-release/2026/09/18/3364558/0/en/zealand-pharma-receives-positive-chmp-opinion-for-zeydovio-glepaglutide-for-the-treatment-of-short-bowel-syndrome.html

Endoscopy and digestive oncology

8. BioNTech and Genentech terminate the phase 2 trial of the mRNA vaccine autogene cevumeran in resected colorectal cancer

[Industry] [Germany]
On August 28 BioNTech announced termination of the BNT122-01 trial (NCT04486378), which compared individualized mRNA immunotherapy as monotherapy against watchful waiting in resected, ctDNA-positive high-risk stage II or stage III colorectal cancer. After crossing the futility boundary in October 2025, the data monitoring board now identified a numerical overall survival imbalance between arms and recommended stopping; no figures have been released. No new safety signals.
Why it matters: it is a serious brake on mRNA vaccines as monotherapy in "cold" tumors such as colorectal cancer; the absence of numbers means waiting for the full publication, and the pancreatic cancer program (with chemotherapy and immunotherapy) continues.
https://www.biontech.com/int/en/home/mediaroom/news/statements/2026/08/BioNTech-Provides-Update-on-Phase-2-Clinical-Trial-of-Autogene-Cevumeran-in-Resected-Colorectal-Cancer.html

9. Ivonescimab plus chemotherapy improves overall survival in advanced biliary tract cancer (HARMONI-GI1)

[Industry] [China, Akeso]
Akeso announced on September 21 that its randomized, double-blind phase 3 HARMONi-GI1 trial (ivonescimab, a PD-1/VEGF bispecific, plus chemotherapy vs durvalumab plus chemotherapy in first line) met its overall survival endpoint. Full data will be presented as a late-breaking abstract (LBA8) in the presidential symposium of ESMO 2026 on October 25 in Madrid.
Why it matters: it would be the first trial to beat an anti-PD-L1 immunotherapy (rather than placebo) in this indication, but until hazard ratio, n and population are seen it cannot be judged; the sponsor has every incentive and Chinese results with ivonescimab have not always replicated in Western populations.
https://www.prnewswire.com/news-releases/ivonescimab-first-line-biliary-tract-cancer-phase-iii-overall-survival-data-selected-as-esmo-late-breaking-abstract-nearly-30-akeso-studies-to-be-presented-at-esmo-2026-302885329.html

10. One sugar-sweetened beverage a day is associated with more than double the gastric cancer risk in two U.S. cohorts

[Publication] [United States, Massachusetts General Hospital]
Prospective analysis of the Nurses' Health Study and the Health Professionals Follow-up Study (about 112,000 participants, more than 3.2 million person-years, 278 gastric adenocarcinomas), published in Gastro Hep Advances. More than one sugar-sweetened beverage per day vs never: HR 2.45 (95% CI 1.49–4.04), with no heterogeneity by sex; artificially sweetened beverages were not associated. In a sub-study, intake was unrelated to Helicobacter pylori seropositivity.
Why it matters: it is the first association of this kind in large cohorts and fits the rise of early-onset gastric cancer, but it is observational, with few cases, little adjustment for H. pylori and tumor location, and a mostly White population; in our region, with high H. pylori prevalence, the interaction remains to be studied.
https://www.ghadvances.org/article/S2772-5723(26)00218-9/fulltext

Pancreas, esophagus and stomach

11. FDA approves daraxonrasib, the first broad RAS inhibitor, for metastatic pancreatic cancer

[FDA regulatory] [United States, Revolution Medicines]
On August 26 the FDA approved daraxonrasib (Rasonque, 300 mg orally once daily) for adults with metastatic pancreatic adenocarcinoma after at least one systemic line. Basis: the RASolute 302 trial (n = 500, randomized, open-label, vs physician's choice chemotherapy). Median overall survival 13.2 vs 6.7 months (HR 0.40; 95% CI 0.30–0.53); progression-free survival 7.2 vs 3.6 months (HR 0.49); objective response 30% vs 11%. Warnings: dermatologic toxicity, stomatitis, diarrhea, gastrointestinal perforation and pneumonitis. Approval came six and a half months ahead of the goal date.
Why it matters: it is the largest second-line advance in pancreatic cancer in decades and directly concerns the gastroenterologist who diagnoses and follows these patients; the open-label design and the lack of published RAS G12 subgroup data are limitations, and price and availability in Latin America are unknown.
https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-daraxonrasib-metastatic-pancreatic-adenocarcinoma

12. Very hot drinks and esophageal squamous cell carcinoma: nearly one million British adults

[Publication] [United Kingdom, Oxford Population Health]
Cancer Research UK released on September 9 an analysis of UK Biobank and the Million Women Study (about 980,000 adults, more than ten years of follow-up): those who drink their beverages "hot" had nearly twice the risk of esophageal squamous cell carcinoma compared with "warm" drinkers, and "very hot" drinkers three times the risk. There was no association with adenocarcinoma or with tea or coffee consumption itself.
Why it matters: it confirms in a Western population what IARC already classified as "probably carcinogenic" above 65 °C, with the limitation that temperature was self-reported and no hazard ratios or confidence intervals were published in the release; the practical message is simple and applicable in any clinic.
https://news.cancerresearchuk.org/2026/09/09/link-between-hot-drinks-and-type-of-oesophageal-cancer-in-the-uk/

Cross-cutting

13. XXXIV Dominican Congress of Gastroenterology (SODOGASTRO), Miches, September 17–20

[Meeting] [Dominican Republic]
The annual congress of the Dominican Society of Gastroenterology was held at the Miches Convention Center under the motto "From knowledge to action", dedicated to Dr. Luis Pérez Méndez, together with the II Dominican Congress of Pediatric Gastroenterology, Hepatology and Nutrition and a LASPGHAN pediatric IBD summit. The program included vonoprazan, third-space endoscopy, standardized upper-endoscopy reporting (STAR Project), IBD epidemiology in the country, Rome V criteria, artificial intelligence in endoscopy and a hepatology module (MASLD, hepatitis B, cirrhosis, transplantation).
Why it matters: it is the local reference for what is being adopted in the country; we found no abstracts or registry data published from the congress, so its value this month is agenda and professional networking.
https://diariosalud.do/xxxiv-congreso-dominicano-gastroenterologia-programa

Note of the month: obefazimod (ABTECT, maintenance) circulated again in the specialty press in September, but the data are those of the June 1 press release and there is no new publication, so it is not included. The September CHMP also recommended a tofacitinib generic (Inijaq) with an ulcerative colitis indication and a filgotinib (Jyseleca) extension that concerns axial spondyloarthritis, not Crohn's disease.

Upcoming meetings and dates

• ACG 2026 Annual Scientific Meeting: October 9–14, Nashville (Music City Center).
• UEG Week 2026: October 17–20, Barcelona (Fira Gran Via) and online.
• ESMO Congress 2026: October 23–27, Madrid; HARMONi-GI1 data are presented October 25.
• The Liver Meeting 2026 (AASLD): November 5–9, Denver.

Monthly gastroenterology bulletin — September 28, 2026
13 verified items from the past month: Journal of Hepatology, Gastro Hep Advances, FDA, EMA (CHMP), Cancer Research UK and industry releases.
Three worth your time:
• FDA approved daraxonrasib for metastatic pancreatic cancer: overall survival 13.2 vs 6.7 months (HR 0.40) in 500 patients; open-label trial, no RAS-subgroup data yet.
• Baveno VIII is out: rules out CSPH with LSM ≤15 kPa and platelets ≥150, carvedilol when CSPH probability exceeds 75%, and a formal definition of recompensation; many new statements rest on expert opinion.
• One sugar-sweetened drink a day was tied to HR 2.45 for gastric cancer in 112,000 health professionals; observational, 278 cases, little adjustment for H. pylori.
Full bulletin, with a critical reading of each study and links to primary sources:
https://boletinesmedicos.com/en/gastroenterologia/2026-09-28.html
#Gastroenterology #Hepatology #PancreaticCancer #PortalHypertension

GI, September 2026: FDA approves daraxonrasib for pancreatic cancer (OS 13.2 vs 6.7 months, HR 0.40) and Baveno VIII sets carvedilol when CSPH probability exceeds 75%. Bulletin with critical reading: https://boletinesmedicos.com/en/gastroenterologia/2026-09-28.html