Introduction

Plastic biliary stents that have not been removed for more than 12 months are considered forgotten stents. In this study, we sought to evaluate which factors are associated with forgetting biliary stents and their consequences.

Aims & Methods

Retrospective analysis, in a high-volume ERCP unit, of patients who underwent ERCP between 2016 and 2023 with placement of plastic biliary stents. Procedures in which the patient died or the prosthesis was surgically removed less than 1 year after its placement were excluded.

Results

A total of 555 procedures in 425 patients (49% female, age 68.9±16.8 years) were included. The reasons for placement were: malignant stenosis 29.2% (n=162); choledocholithiasis 23.8% (n=132); benign stenosis 22.7% (n=126); bile leak 7.2% (n=40); acute cholangitis 6.7% (n=37); Mirizzi syndrome 4.7% (n=26); gallstones 4.3% (n=24); post-sphincterectomy bleeding 0.9% (n=5); intrahepatic lithiasis 0.5% (n=3).
Thirty-four procedures (6.1%) with forgotten stents were identified in 34 patients (35.3% female, age 64.4±17.3 years). The most frequent reasons for non-removal were: non-referral after cholecystectomy (n=15) or after hospital discharge (n=11). Among the indications for placement, gallstones (OR 9.71, p < 0.001) and Mirizzi syndrome (OR 3.02, p = 0.049) were associated with an increased risk of forgetting the device.
In 10 patients, extraction was endoscopic, 2 had spontaneous migration, 1 had surgical extraction, and 7 patients died for unrelated reasons. The rest (n=14) were recruited for removal during this study. A total of 9 patients of this group responded to the call and it was confirmed the spontaneous migration of the plastic stent in na abdominal radiography.
8 patients (23.5%) presented complications: acute cholangitis (n=4), jaundice (n=2), abdominal pain (n=1) and liver abscess with septic shock (n=1).

Conclusion

Forgotten stents constitute a rare event but with a significant risk of complications. Following this analysis, we created a database in order to improve the timing until the extraction of stents and reduce related adverse events.

Introduction

The identification and establishment of reliable biomarkers and early diagnostic system for pancreatic ductal adenocarcinoma (PDAC) are urgently needed. Although blood-based liquid biopsy for comprehensive genomic profiling is useful for cancer diagnosis, we have recently shown that driver gene mutations derived from PDAC can be detected at a higher rate in pancreatic juice (PJ) than blood. We have also reported that some RNAs are carried within extracellular vesicles (EVs) and EV-encapsulated RNAs circulating in body fluid are useful as biomarkers for PDAC diagnosis. We recently developed a unique immunoprecipitation method called EViSTEP™ that utilizes a chelating-based reagent and EV surface marker antibodies (CD9/CD63) to isolate EVs with high purity and high yield.

Aims & Methods

In this study, we aimed to evaluate the usefulness of multi-layer liquid biopsy system by DNA mutation and RNA expression in PJ for cancer diagnosis. PJ samples (n=83; 15 benign hepatopancreatobiliary diseases, 33 low-grade (LG) IPMN, 10 high-grade (HG) IPMN and 25 PDAC) were collected during ERCP examination from each patient in a multicenter study. Cell-free (cf) DNA/RNA and EV-derived RNA (EV RNA) were purified from 2 ml of each PJ. Digital PCR was performed for driver gene mutation analysis (4 types of KRAS and 2 types of GNAS), as well as expression analysis of 10 long non-coding RNAs (lncRNA, e.g., HEVEPA, LINC02280) in cfRNA and EV RNA. The diagnostic performance of pancreatic cancer (PDAC+HG-IPMN) was statistically analyzed after normalizing RNA expression with two reference genes (ACTB/B2M).

Results

Logistic regression analysis (objective variables; PDAC+HG IPMN and others) with the KRAS and GNAS variant allele frequency (VAF), and RNA expression as factors was performed. The Area Under Curves (AUC) in Receiver Operating Characteristic (ROC) curve analysis for PDAC+HG IPMN diagnosis were 0.490-0.74 for KRAS mutations, 0.475-0.554 for GNAS mutations, 0.451-0.654 for cfRNA expressions and 0.362-0.639 for EV RNA expressions. Among them, the factors with relatively high AUC were KRASG12V (0.728), KRASG12R (0.687), cfRNA RP11-750H9.5/ACTB(0.654), EV RNA CRNDE/ACTB(0.633). The AUC values of the combinations, [KRASG12V, KRASG12R, cfRNA RP11-750H9.5/ACTB and EV RNA CRNDE/ACTB] and [the aforementioned four factors and CA19-9], were 0.876 and 0.885, respectively in terms of discriminating PDAC+HG IPMN. Moreover, concerning the diagnosis of Stage 0/I PDAC, the AUC values of EV HEVEPA, CA19-9, and the combination [EV HEVEPA and CA19-9] were 0.693, 0.746 and 0.84, respectively, suggesting that EV HEVEPA would be reliable biomarker for early detection of PDAC.

Conclusion

By incorporating cell-free lncRNA and EV lncRNA expression analysis into DNA mutation analysis in PJ samples could be expected to improve the diagnostic performance of pancreatic cancer. Liquid biopsy using PJ would predict the presence of neoplastic lesions with high probability and has a powerful potential for early diagnosis. Establishing the multilayer gene panel using these DNA and RNA has the potential to improve the early diagnostic accuracy for pancreatic cancer, a disease that exhibits significant diversity.

Disclosure

Grants for commissioned/joint research: Hitachi High-Tech Corporation (KT, YM), H.U. Group Research Institute G.K. (KT)