Gastroenterology

Biannual ultrasound plus AFP/AFP-L3/DCP does not improve early HCC detection over ultrasound alone

Adults with cirrhosis or high-risk HBV (n=1208) were randomized to biannual ultrasound alone versus ultrasound plus AFP, AFP-L3, and DCP. Early-stage HCC incidence and detection did not differ between arms (HR 0.81; 95% CI 0.47–1.40; P=.45). Most detected HCCs were early stage in both groups, with similar numbers of late-stage cancers. Elevated AFP-L3 alone performed similarly to using all three biomarkers in the combination arm. The trial suggests routine addition of multiple serum biomarkers to ultrasound does not enhance early HCC surveillance yield in this population. 1

AASLD 2023 HCC surveillance algorithm delivers very high sensitivity and NPV versus prior approaches

In 953 high-risk patients undergoing US and AFP surveillance, HCC prevalence was 5% (50/953). The integrated AASLD 2023 algorithm achieved 94% sensitivity (47/50) and 99.6% NPV (758/761) for HCC detection. Sensitivity for early HCC remained superior to US LI-RADS 2017 and AASLD 2018 algorithms, though specificity fell to 84%. Individual new triggers (VIS-C, rising AFP, lesion growth) were highly specific but variably sensitive when used alone. False positives were more likely with AFP <20 ng/mL and in patients without cirrhosis, informing interpretation of abnormal triggers. 2

ESD-first strategy for Barrett’s cancers >15 mm improves R0 rates and outcomes, especially in T1b disease

This multicenter cohort compared historical selective ESD with a later ESD-first strategy for suspected Barrett’s cancers >15 mm (581 resections). ESD use increased from 21.2% to 77.1%, despite a higher cancer burden in the later period. Basal R0 resection improved from 69.7% to 91.2%, with T1b R0 rates rising from 33.3% to 81.9%. In T1b cancers, curative resection increased from 9.5% to 30.5%, and recurrence fell from 55.6% to 23.6%. Two-year cancer-free survival was higher after ESD than EMR (87.4% vs 50%), with similarly low adverse event rates (2.2%). 3

Baseline fecal calprotectin and high meat intake predict IBD flare risk in long-term PREdiCCt cohort

The PREdiCCt cohort followed 2629 IBD patients in self-reported remission for a median 4.1 years. Baseline fecal calprotectin ≥250 µg/g predicted patient-reported flares (aHR 2.22) and objective flares (aHR 3.25) versus <50 µg/g. Intermediate calprotectin (50–250 µg/g) also increased flare risk, supporting graded risk stratification. In UC, higher total meat intake independently increased objective flare risk (highest vs lowest quartile aHR 1.95). No consistent associations were found for ultraprocessed foods, fiber, or polyunsaturated fats after adjustment. 4