Introduction

PNET is an important determinant of survival in MEN1 patients and the most important cause of MEN1-related death. Early identification of pNETs in MEN1 is important since these represent a common cause of morbidity and mortality of the syndrome. Prognosis is associated with tumour size. Many pNETs are small subcentimetre lesions and only become symptomatic when the tumour is large. Conventional radiology is usually suboptimal in detecting small tumours (< 2cm), thus missing the window to remove them at an earlier stage. We aim to compare EUS detection of pNETs in patients with confirmed MEN1 compared to cross-sectional imaging including CT and/or MRI pancreas and assess the incremental benefit of EUS.

Aims & Methods

Over a span of 7 years, a total of 20 patients with clinically confirmed MEN1 had surveillance of pNET performed with EUS, CT/MRI and biochemical screening. Data were retrospectively retrieved from the hospital electronic records database. EUS and CT/MRI done within 3-6 months of each other were compared to minimize discrepancy in findings due to pNET size progression and new lesions over time. Statistical analysis was performed using SPSS v20 on the size and number of pNET detection by EUS and CT/MRI using the Wilcoxon Signed Rank Test and McNemar Chi-square.

Results

A total of 28 EUS procedures and an equal number of cross sectional imaging (CT/MRI) were performed. pNETs were identified in 95% of all 20 MEN1 cases. Overall median pNET size was 7.1mm on EUS and 14.5mm on CT/MRI (p=0.007). Mean value for smallest PNETs detected by EUS was 4.6mm and 12.7mm on CT/MRI (p=0.001). EUS detected more pancreatic lesions/pNETs compared to CT/MRI (p<0.001) in 25 of the 28 procedures (89.3% more). The remaining 3 procedures showed equal numbers of pNETs detection by both modalities. The interquartile range (IQR) for smallest pNET detected by EUS was 3.0-5.0mm while IQR for CT/MRI lies between 8.3mm-14.8mm. EUS detected all 100% cases of pNETs in our series of MEN1 compared to CT/MRI imaging alone which detected 57.9% cases (p=0.008). 14 of 20 patients had FNA performed with a positive yield of 85.7%. In 50% of patients, pNET measured ≤10 mm. 45% of patients underwent surgery for pNETs while the rest had surveillance.

Conclusion

In MEN1 patients, CT/MRI underestimated the presence of pancreatic lesions/pNETs in 42.1% compared to EUS (p=0.008) and was not able to identify pNETs < 8mm in all but one case.
EUS offers higher sensitivity than cross sectional imaging (CT/MRI) in terms of detecting the number of positive pNET cases as well as a greater number of pancreatic lesions especially subcentimetre lesions in MEN1. EUS should be considered a standard tool in the algorithm for MEN1 workup, instead of an adjunct reserved for diagnostic dilemmas.

References

NCCN Guidelines 2015 version 1.2015 Neuroendocrine Tumours.

Introduction

Currently, most patients with branch duct intraductal papillary mucinous neoplasms (BD-IPMN) are offered indefinite surveillance, resulting in health care costs with questionable benefits regarding cancer prevention.

Aims & Methods

This study sought to identify patients where the risk of cancer is equivalent to an age-matched population, thereby justifying discontinuation of surveillance. Methods included an International multicenter study involving presumed BD-IPMN without worrisome features (WF) or high-risk stigmata (HRS) at diagnosis who underwent surveillance. Clusters of individuals at risk for cancer development were defined according to cyst size and stability for at least 5 years, and age-matched controls were used for comparison using standardized incidence ratios (SIRs) for pancreatic cancer.

Results

Of 3844 patients with presumed BD-IPMN, 775 (20.2%) developed WF and 68 (1.8%) HRS after a median surveillance of 53 (IQR 53) months. Some 164 patients (4.3%) underwent surgery. Of the overall cohort, 1617 patients (42%) remained stable without developing WF or HRS for at least 5 years. In patients 75 years or older, the SIR was 1.12 (95%CI 0.23-3.39), and in patients 65 years or older with stable lesions below 15mm in diameter after 5 years, the SIR was 0.95 (95%CI 0.11-3.42). The all-cause mortality at 10 years of the entire cohort was 9%, while the disease specific mortality was 0.9%.

Conclusion

The risk of developing pancreatic malignancy in presumed BD-IPMN without WF or HRS after 5 years of surveillance is comparable to that of the general population depending on cyst size and patient age. Surveillance discontinuation could be justified after 5 years of stability in patients older than 75 years with cysts < 30 mm, and in patients 65 years or older who have cysts ≤ 15 mm.

Introduction

Mucinous cystic neoplasms (MCN) are rare cystic pancreatic lesions with an evidenced malignant potential. Although the population who bear MCN is well characterized, the natural history of these lesions has not been elucidated as well as the risk features of malignancy. In a recent single centre study (Vullierme et al., 2021), we identified imaging features suggesting malignant MCN on MRI:
1) ≥ 5mm wall thickness,
2) mural nodule size ≥ 9 mm,
3) enhancing intra-cystic septa and/or
4) heterogenous signal on T2-weighted imaging.

Aims & Methods

The aims of this study were to identify clinical malignancy predictive factors and to validate in a multicentre cohort, the imaging malignancy predictive factors identified previously. In this multicentre retrospective study, patients who underwent pancreatic resection between January 2008 and December 2021 for MCN were included. Pathological review confirmed the presence of ovarian-type stroma. The preoperative imaging (CT and/or MRI) had to be available for a centralized expert radiological review. A multivariable logistic regression was performed to identify predictive factors of malignancy.

Results

A total of 198 patients were included from 13 participant centres. The cohort was constituted of 196 female patients (99%) with a median age of 46 (IQR 37-55) years. At diagnosis, 97 patients (49%) were asymptomatic. Low-grade dysplasia (LGD) was observed in 166 (84%) patients and at least high-grade dysplasia (≥HGD) in 32 (16%) patients, from which 14/32 had invasive carcinoma. An age of diagnosis of ≥55 years and the presence of symptoms at diagnosis was significantly associated with ≥HGD (p=0.028 and p=0.01, respectively). Regarding imaging features: ≥ 5mm wall thickness, a mural nodule size of ≥ 9 mm, presence of enhancing intra-cystic septa, and heterogenous cyst signal on T2-weighted imaging, were all significantly associated with ≥HGD (p=0.005, p=0.009, p=0.006, and p=0.003, respectively). In multivariate analysis, symptoms at diagnosis (OR 2.78, CI95% 1.13-6.89; p=0.027), lesion size ≥4cm (OR 3.29, CI95% 1.05-10.4; p=0.041) and presence of mural nodule (OR 6.58, CI95% 2.77-15.64; p<0.001) were independent predictive factors of ≥HGD.

Conclusion

MCN causing symptoms at diagnosis, measuring ≥ 4cm and with a mural nodule in imaging, should be addressed for surgical resection. The imaging features predictive of malignancy, that are confirmed in this cohort, should also be considered in surgical management decisions.

References

Vullierme MP, Gregory J, Rebours V, Cros J, Abelhady-Attia Y, Vilgrain V, Aguilera Munoz L, Laurent L, Levy P, Sauvanet A, Ronot M. MRI is useful to suggest and exclude malignancy in mucinous cystic neoplasms of the pancreas. Eur Radiol. 2021 Aug 10; PMID: 34374801

Disclosure

No conflict of interest to disclose

Introduction

Blood-based liquid biopsy for comprehensive genomic profiling is available for cancer patients; however, given the limited amount of circulating tumor-derived DNA in plasma, it remained challenging to detect pancreatic ductal adenocarcinoma (PDA) at earlier stages. We sought to determine the usefulness of pancreatic juice-containing DNA collected from a site more proximal to the tumor in the early diagnosis of pancreatic cancer.

Aims & Methods

We evaluated the mutation detection in DNA from the duodenal fluid as an alternative biospecimen obtained by a less-invasive approach relative to endoscopic biopsy. In 296 patients diagnosed with PDA (n=175), IPMN (n=61), or other pancreatic diseases (n=60) were recruited in seven institutions. Duodenal fluid was collected from the duodenal lumen through the endoscopic channel and pancreatic juice was obtained directly from the pancreatic duct through a catheter during the ERP examination. Together with plasma cfDNA, both digestive fluid-included DNA was extracted, and molecular barcode sequencing (Ion Amplseq HD) targeting five pancreatic cancer-associated genes, namely, KRAS, GNAS, TP53, CDKN2A, and SMAD4, were performed. Detected mutations were evaluated for concordance with surgically resected tissue or needle biopsy.

Results

DNA extracted from the duodenal fluids (DFDNA) and pancreatic juice (PjDNA) had a much higher yield (avg. 25.4 ng/mL, and 7.7 ng/mL, respectively) than plasma cell-free DNA (avg. 1.3 ng/ml). While most plasma cfDNA had a typical fragmentation pattern with the highest peak at 160 bp, DFDNA and PjDNA showed various fragmentation patterns. Using molecular barcode sequencing, hotspot mutations detected in the DFDNA (n=148) showed 50.7% concordance of the mutations in resected PDA lesions, whereas only 24.1% concordance in the plasma cfDNA (n=114). In contrast to plasma cfDNA, where liquid-tissue matches were limited to advanced stages of cancer, DFDNA assay enabled the higher concordance through a wide range of PDA stages, even Stage 0 (83.3%) or I (45.5%) UICC 8th, or IPMN cases. Furthermore, PjDNA showed a much higher concordance rate (88%, n=25) than DFDNA. In DFDNA and PjDNA, mutations that originated from the primary lesions and from multicentric/concomitant lesions such as IPMN were detected effectively.

Conclusion

Digestive fluid-derived DNA can be a potent biospecimen to detect mutations with higher yield and sensitivity than plasma cfDNA. For practical diagnostics testing, optimized collection and storage of the duodenal fluid and methods for purification of the nucleic acids are needed. Besides as a diagnostic tool, genomic profiling of DFDNA can offer better surveillance for individuals at high-risk or annual checkups for healthy people. PjDNA can predict the presence of neoplastic lesions with extremely high probability and has a powerful potential for further testing before a more definitive diagnosis, even without histological evidence of cancer.

Disclosure

Yusuke Ono, Kenji Takahashi, Yusuke Mizukami: Research grants from Hitachi High-Tech Corporation.