The aim of this study was to develop a novel deep learning (DL) model without requiring large-annotated training datasets for detecting pancreatic cancer (PC) using computed tomography (CT) images. This retrospective diagnostic study was conducted using CT images collected from 2004 and 2019 from 4287 patients diagnosed with PC. We proposed a self-supervised learning algorithm (pseudo-lesion segmentation (PS)) for PC classification, which was trained with and without PS and validated on randomly divided training and validation sets. We further performed cross-racial external validation using open-access CT images from 361 patients. For internal validation, the accuracy and sensitivity for PC classification were 94.3% (92.8–95.4%) and 92.5% (90.0–94.4%), and 95.7% (94.5–96.7%) and 99.3 (98.4–99.7%) for the convolutional neural network (CNN) and transformer-based DL models (both with PS), respectively. Implementing PS on a small-sized training dataset (randomly sampled 10%) increased accuracy by 20.5% and sensitivity by 37.0%. For external validation, the accuracy and sensitivity were 82.5% (78.3–86.1%) and 81.7% (77.3–85.4%) and 87.8% (84.0–90.8%) and 86.5% (82.3–89.8%) for the CNN and transformer-based DL models (both with PS), respectively. PS self-supervised learning can increase DL-based PC classification performance, reliability, and robustness of the model for unseen, and even small, datasets. The proposed DL model is potentially useful for PC diagnosis.

Jee Hyung Lee, Haeryoung Kim, Sang Hyub Lee, Ja-Lok Ku, Jung Won Chun, Ha Young Seo, Soon Chan Kim, Woo Hyun Paik, Ji Kon Ryu, Sang Kook Lee, Andrew M. Lowy, and Yong-Tae Kim. Gut and Liver 2022;16:625-36. https://doi.org/10.5009/gnl210166

There are limited data on the efficacy of liquid-based cytology (LBC) for EUS-guided FNA specimens. We aimed to evaluate the diagnostic efficacy of LBC for solid pancreatic neoplasms compared with conventional smears (CSs). In this randomized, crossover, noninferiority trial, we randomly assigned (1:1) patients with suspected pancreatic cancer to the LBC group or the CS group. Aspirates from the first needle pass were processed by one method, aspirates from the second pass by the other method, and specimens from the last pass were processed as core biopsy samples. The primary endpoint was the diagnostic efficacy of each method, with the final diagnosis as the gold standard. A noninferiority margin of -10% was assumed. Of 170 randomized patients, 165 were classified as malignant and 5 as benign. Unsatisfactory samples were less frequent in the LBC group (1.78%) compared with the CS group (5.33%). The diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of LBC versus CS were 88.0% versus 83.8% (P = .276), 87.7% versus 83.2% (P = .256), 100% versus 100% (P = .999), 100% versus 100% (P = .999), and 16.7% versus 16.1% (P = .953), respectively. A bloody background was significantly more frequent in the CS group (CS, 85.2%; LBC, 1.8%; P < .001), whereas the nuclear features were similar for both groups. The diagnostic usefulness of LBC was comparable with that of CS. The cytomorphologic features did not differ significantly between the 2 methods, and the reduced bloody backgrounds allowed better visibility in the LBC method. (Clinical trial registration number: NCT03606148.).

Background: Nanoliposomal encapsulation of irinotecan (nal-IRI) with 5-fluorouracil and leucovorin (5-FU/LV) has shown a survival benefit for gemcitabine-pretreated patients with metastatic pancreatic adenocarcinoma (mPAC). The aim of this study was to evaluate the effectiveness and safety of nal-IRI with 5-FU/LV for use beyond second-line treatment after standard frontline therapy for mPAC. Method: This multicenter, retrospective, non-comparative observational study included mPAC patients who received nal-IRI plus 5-FU/LV as third- or later-line therapy after disease progression on first-line FOLFIRINOX (FFX) or gemcitabine plus nab-paclitaxel. Results: In all, 128 patients who received nal-IRI plus 5-FU/LV beyond second-line treatment between October 2017 and July 2021 were analyzed. Most patients (82%) received nal-IRI plus 5-FU/LV as a third-line treatment. The median overall survival (OS) was 4.9 months and the median progression-free survival (PFS) was 2.4 months. Patients with better Eastern Cooperative Oncology Group (ECOG) performance status experienced significantly longer OS (ECOG 0, 8.7 months; ECOG 1, 4.8 months; ECOG 2, 2.9 months; p < 0.001) and PFS (3.9 months; 2.1 months; 1.5 months; p = 0.019). Patients who had not been previously treated with FFX or had a time to progression of 7 months or more on FFX experienced longer OS and PFS than those who did not (6.1 months and 5.6 versus 4.1 months, p = 0.053; 3.6 months and 2.4 versus 2.1 months, p = 0.002). The most common adverse events were neutropenia (56%) and anemia (51%). Conclusion: Our real-world data indicated that nal-IRI plus 5-FU/LV can be effective not only as second-line therapy, but also as third-line or later-line treatment in selected patients. Nal-IRI plus 5-FU/LV may be particularly beneficial for the survival of patients that maintain good general condition or those with favorable prior experience to irinotecan.

Background Fully covered self-expandable metal stents (FCSEMSs) are widely used for endoscopic treatment of distal malignant biliary obstruction (dMBO). We aimed to assess the efficacy of anchoring an external plastic stent to an FCSEMS in dMBO. Methods A multicenter retrospective cohort study was performed in patients with dMBO to compare stent patency between FCSEMSs and FCSEMSs with an externally anchored plastic stent (EPS). For external anchoring, a 7-Fr double-pigtail plastic stent (DPPS) was placed first in the bile duct, then an FCSEMS was deployed side-by-side. Results Among a total of 185 patients, 65 had an FCSEMS alone and 120 had an FCSEMS with an EPS. The median stent patency was significantly longer in the FCSEMS with an EPS group than in the FCSEMS only group (342 vs. 240 days; P = 0.04). The rate of stent migration was significantly lower in the FCSEMS with an EPS group than in the FCSEMS only group (10.8 % vs. 27.7 %; P = 0.01). There were no significant differences in the rates of stent occlusion and adverse events between the two groups. Conclusions A novel and simple technique of anchoring an external plastic stent may decrease the risk of FCSEMS migration and prolong stent patency, without significantly increasing the adverse events rate in dMBO. ^* Joint first authors Received: 09 June 2022 Accepted after revision: 23 November 2022 Article published online: 03 February 2023 © 2023. Thieme. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany

To evaluate the efficacy of proton beam therapy (PBT) as an initial treatment in treatment-naïve hepatocellular carcinoma (HCC) patients and to assess the prognostic significance of albumin-bilirubin (ALBI) grade, 46 treatment-naïve HCC patients treated with PBT were analyzed. The ALBI grade distribution was grade 1 in 11 (23.9%) patients, grade 2 in 34 (73.9%) patients, and grade 3 in 1 (2.2%) patient. The median duration of follow-up was 56.5 months (95% confidence interval [CI], 48.2–64.7). Among the 46 patients, disease progression was observed in 23 (50%) patients: local progression in 3 (6.5%) patients; intrahepatic progression in 22 (47.8%); and extrahepatic progression in 5 (10.9%). The 5-year freedom from local progression (FFLP), progression-free survival (PFS), and overall survival (OS) rates were 92.7% (95% CI, 84.7–100.7), 43.3% (95% CI, 28.2–58.4), and 69.2% (95% CI, 54.9–83.5), respectively. In multivariate analysis, there were no independent factors for FFLP (p > 0.05 each), but tumor stage and ALBI grade were independent factors for PFS and OS (p < 0.05 each). PBT could result in comparable OS in treatment-naïve HCC patients to other recommended first-line treatments, and ALBI grade, in addition to tumor stage, could be useful for predicting OS.