Background/Aims: A high-quality sample allows for next-generation sequencing and the administration of more tailored precision medicine treatments. We aimed to evaluate whether heparinized wet suction can obtain higher quality samples than the standard dry-suction method during endoscopic ultrasound (EUS)-guided biopsy of pancreatic masses. Methods: A prospective randomized crossover study was conducted. Patients with a solid pancreatic mass were randomly allocated to receive either heparinized wet suction first or dry suction first. For each method, two needle passes were made, followed by a switch to the other method for a total of four needle punctures. The primary outcome was the aggregated white tissue length. Histological blood contamination, diagnostic performance and adverse events were analyzed as secondary outcomes. In addition, the correlation between white tissue length and the extracted DNA amount was analyzed. Results: A total of 50 patients were enrolled, and 200 specimens were acquired (100 with heparinized wet suction and 100 with dry suction), with one minor bleeding event. The heparinized wet suction approach yielded specimens with longer aggregated white tissue length (11.07 mm vs 7.96 mm, p=0.001) and less blood contamination (p=0.008). A trend towards decreasing tissue quality was observed for the 2nd pass of the dry-suction method, leading to decreased diagnostic sensitivity and accuracy, although the accumulated diagnostic performance was comparable between the two suction methods. The amount of extracted DNA correlated positively to the white tissue length (p=0.001, Spearman ̕s ρ=0.568). Conclusions Heparinized wet suction for EUS tissue acquisition of solid pancreatic masses can yield longer, bloodless, DNA-rich tissue without increasing the incidence of adverse events (ClinicalTrials.gov. identifier NCT04707560).

Background/Aims: Despite the high efficacy of direct-acting antivirals (DAAs), approximately 1–3% of hepatitis C virus (HCV) patients fail to achieve a sustained virological response. We conducted a nationwide study to investigate risk factors associated with DAA treatment failure. Machine-learning algorithms have been applied to discriminate subjects who may fail to respond to DAA therapy. Methods: We analyzed the Taiwan HCV Registry Program database to explore predictors of DAA failure in HCV patients. Fifty-five host and virological features were assessed using multivariate logistic regression, decision tree, random forest, eXtreme Gradient Boosting (XGBoost), and artificial neural network. The primary outcome was undetectable HCV RNA at 12 weeks after the end of treatment. Results: The training (n=23,955) and validation (n=10,346) datasets had similar baseline demographics, with an overall DAA failure rate of 1.6% (n=538). Multivariate logistic regression analysis revealed that liver cirrhosis, hepatocellular carcinoma, poor DAA adherence, and higher hemoglobin A1c were significantly associated with virological failure. XGBoost outperformed the other algorithms and logistic regression models, with an area under the receiver operating characteristic curve of 1.000 in the training dataset and 0.803 in the validation dataset. The top five predictors of treatment failure were HCV RNA, body mass index, α-fetoprotein, platelets, and FIB-4 index. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the XGBoost model (cutoff value=0.5) were 99.5%, 69.7%, 99.9%, 97.4%, and 99.5%, respectively, for the entire dataset. Conclusions Machine learning algorithms effectively provide risk stratification for DAA failure and additional information on the factors associated with DAA failure.