Society: AGA

Abstract
Background and Aims: 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. This study sought to identify patients where the risk of cancer is equivalent to an age-matched population, thereby justifying discontinuation of surveillance.

Methods: 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, 843 (22%) developed a WF or HRS after a median surveillance of 53 (IQR 53) months and 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 2.23 (95%CI 0.45-6.52), and in patients 65 year or older with stable lesions below 15mm in diameter after 5 years, the SIR was 1.77 (95%CI 0.20-6.39).

Conclusions: 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.

<b>Standardized Incidence Ratio (SIR) for pancreatic cancer</b>

Standardized Incidence Ratio (SIR) for pancreatic cancer

Recommendations according to the new evidence provided by the paper

Recommendations according to the new evidence provided by the paper

Surgical management of Pancreatic neuroendocrine tumors (PNETs) has long been debated. Current NCCN guidelines state that non-functioning PNETs >2 centimeters (cm) should undergo primary surgical removal. Tumors < 2cm in size can be observed carefully and surgery is not always indicated. Recent NCDB based analysis suggests that surgical intervention for tumors 1-2 cms is beneficial while observation remains an option for tumors <1cm. In this analysis we seek to understand the influence of surgery on survival for smaller PNETS (</=2 cms).

Methods
The NCDB was used to identify 15,017 patients diagnosed with PNETs from 2004-2018 with known tumor size, surgical status, and survival outcomes. Variables included in this analysis are race, sex, age, rurality, Charlson-Deyo score, grade, surgery status, stage, site, and size of tumor. Variables were summarized with descriptive statistics including counts, percentages, and compared using Chi^2 analysis. Univariate Survival analysis was demonstrated through Kaplan Meyer survival curve analysis. Survival data was fit to an exponential curve and 60-month mortality estimates were reported. Cox-proportional hazards were used to conduct both whole cohort and stratified multivariate analysis. Hazard ratios and confidence intervals (95%) are reported.

Results
Tumor Size was separated into three categories (<1cm, 1-2cm, ≥2 cm). 950 (6.3%) were <1cm, 3043 (20.3%) were 1-2cm, and 11024 (73.4%) were ≥2cm in size. Of the 15017 patients included, 9911(66.1%) underwent surgical resection. Rates of surgical resection across tumor sizes differed with <1cm 680 (71.81%), 1-2 cm 2414 (79.46%) and ≥2 cm 6817 (61.9%). Five-year mortality estimates across all tumor sizes show a significant decrease in mortality with surgical resection. Survival curves show that patients across all tumor sizes that underwent surgical resection had improved survival outcomes. Multivariate analysis showed that patients not undergoing surgical resection, while controlling for effect of tumor size, had worse survival (HR: 2.41). Patients with increased tumor size (≥ 2cm) had worse survival outcomes compared to both smaller tumor size groups; 1-2 cm (HR 1.59) <1cm (1.34) while controlling for surgical status. Stratified Analysis showed that patients not undergoing surgical resection had worse survival across all tumor sizes; >2cm (HR: 2.46) 1-2 cm (HR: 2.22) <1cm (HR: 2.36).

Discussion
Our analysis suggest that surgical intervention across all tumor sizes is beneficial for overall survival. When controlling for other compounding factors, patients that underwent surgical intervention demonstrated increased survival. This analysis suggests that surgical or other ablative interventions should be considered across all PNET presentations regardless of tumor size and that current guidelines for watchful observation are not consistent with emerging survival data.

Background:
Pancreatic cyst biomarkers have been studied to risk-stratify neoplastic cysts for cancer, and previous studies have reported on the utility of measuring glucose and amphiregulin (AREG) to differentiate between mucinous from non-mucinous cysts. Glucose testing is relatively cheap and fast to obtain. Further validation studies with more rigorous study designs are needed.

Methods:
A total of 292 cyst samples with surgical histology, obtained from multiple centers, were sent to a central site for blinding. These blinded samples were then shipped to a separate site to measure cyst glucose and AREG. Glucose was measured using a Verio One Touch IQ Glucometer, and AREG was measured by ELISA. Cysts with glucose levels less than 50 mg/dl were classified as mucinous, and cysts with AREG levels greater than 112 pg/ml were classified as mucinous. The primary goal was to determine if glucose or AREG could reliably differentiate mucinous vs. non mucinous cysts, while secondary aims assessed their diagnostic performance in diagnosing mucinous cysts with advanced neoplasia (high-grade dysplasia or cancer).

Results:
Glucose differentiated mucinous vs. non-mucinous cysts with an AUC of 0.88 and sensitivity of 90% and specificity of 78% at pre-specified threshold. AREG differentiated mucinous vs. non-mucinous with an AUC of 0.60 and sensitivity of 65% and specificity of 47% at pre-specified threshold (Figure 1). Combining glucose and AREG did not improve diagnostic accuracy. CEA was used as the basis for comparison and had an AUC of 0.92 and sensitivity of 60% and specificity of 96%. Neither glucose nor AREG could meaningfully differentiate advanced neoplasia. Combining glucose and AREG did not improve overall diagnostic accuracy.

Conclusion:
In this multi-center blinded validation study, glucose demonstrated high diagnostic utility to differentiate mucinous from non-mucinous cysts. Glucose had greater sensitivity while CEA had greater specificity. AREG had less diagnostic accuracy in this validation study.

Aim of the study: The management of NF-NEPI≤ 2 cm is challenging. Without signs of malignancy, there is no established algorithm to decide between surgery and surveillance. The aim of the study was to assess the incidence of malignancy in sporadic NF-NEPI ≤2 cm. The secondary objective was to identify the 3-years progression rate of non-operated lesions defined as increasing tumor size > 20%, and/or > 2mm/year, and/or the appearance of lymph node and/or distant metastases.
Patients and Methods: This was a multicenter prospective study, included between December 2016 and June 2019 all consecutive patients with NF-NEPI ≤2 cm. The diagnosis must have been proven by a positive EUS-guided FNA/FNB or positive somatostatin receptors imaging (SRI). The exclusion criteria were: grade G3 tumor at the time of inclusion, MEN 1 syndrome. Two groups of patients were defined based on their management: surveillance and surgical. The diagnosis of malignancy was retained if at least one of the following criteria was met: G3 tumor, presence of lymph node and/or distant metastases, doubling in size of the primary tumor during the 36 months of follow-up (FU). Surveillance was performed by imaging and EUS.
Results: All 111 included patients had at least positive EUS-FNA/FNB obtained diagnosis and/or positive SRI. Table 1 summarizes demographics and clinical characteristics of the patients. Due to symptoms (3), suspicion of malignancy (3), by decision of the tumor board (11) and patient request (2), 19 (17%) patients were operated within a median time of 2.76 months (IQR 1.6-4.6) after inclusion, with following final diagnosis: 14 G1 (73.7%) pT1N0 and 5 G2 (26.3%) pT1N0 (3)/N1(2). 92 (83%) patients had active surveillance. The median FU was respectively 38.4 (95% CI, 26, 3-43.6) and 37.8 (36.1-39.7) months in the surgery and surveillance groups. Due to the significant increase in size of the primary tumor, 4 patients from surveillance group were operated in a median time of 17.1 months (IQR 11.7-20.8) after inclusion (3 G1 pT1N0, 1 G2 pT1N1). In total, 4 (3.6%) patients presented malignancy criteria: 3pN+ on the surgical resection specimen; 1 patient from surveillance group developed metastatic disease 12 months after diagnosis. During FU, 2(2%) deaths unrelated to pancreatic NET occurred. No recurrence was observed in operated patients. The incidence of malignancy and the rate of progression at 3 years were 3.6% and 4.5% respectively.
Conclusion: Our study shows a low but notable rate of baseline malignancy among small (≤2cm) NF-NEPI and in terms of progression over a relatively short 3-year timeframe. Our data highlight the need for careful initial evaluation and identification of more accurate predictors of disease progression to better determine disease course and management. Surveillance strategies require multidisciplinary discussion and thorough patient counseling.

Background:
Previous reports showed increasing incidence and improving survival of pancreatic cancer (PC) in the US over the last decades. However, there are limited data on recent incidence and survival rates of different histopathological subtypes of PC. The aim of this study was to conduct a time-trend analysis of recent PC incidence and survival rates using the nationwide USCS and NCDB databases.

Methods:
Incidence data between 2001-2019 were obtained from the USCS database, covers ≈100% of US population, using SEER*Stat software (8.4.0.1, NCI). Incidence rates, age-adjusted to the 2000-US population, were categorized by histopathology using ICD-O-3 codes into pancreatic ductal adenocarcinoma (PDAC) and pancreatic neuroendocrine tumor (PNET), and into the following subtypes: non-specified PDAC (NOS-PDAC), intraductal papillary mucinous neoplasm-related ADC (IPMN-ADC), solid pseudopapillary neoplasm (SPN), adenosquamous carcinoma (ADSC), acinar cell carcinoma (ACC), poorly differentiated neuroendocrine tumor (PD-NET), mucinous cystic neoplasm (MCN), undifferentiated carcinoma (UDC), signet ring cell carcinoma (SRCC), medullary carcinoma (MC), and serous neoplasm (SN). Time-trends, reported as annual percentage change (APC) and average APC (AAPC), were estimated using Joinpoint Regression software (v.4.9.0.1, NCI) utilizing Monte Carlo permutation analysis. Survival rates for each histopathological subtype of PC between 2004-2018 were computed from the NCDB. Overall survival curves were estimated and compared using the Kaplan-Meier method and Log Rank test.

Results:
Between 2001-2019, there were 803,963 patients diagnosed with PC (50.8% males). PC Incidence rates were increasing for PDAC (AAPC=0.79, P<0.01), PNET (AAPC=11.00, P<0.01), NOS-PDAC (AAPC=1.77, P<0.01), IPMN-ADC (AAPC=3.71, P=0.02), SPN (AAPC=13.52, P<0.01), ADSC (AAPC=3.93, P<0.01), and ACC (AAPC=1.76, P<0.01). While PD-NET experienced a stable trend (AAPC=1.18, P=0.21), PC incidence rates were decreasing for MCN (AAPC=-6.39, P<0.01), UDC (AAPC=-5.71, P<0.01), and SRCC (AAPC=-5.14, P<0.01) (Figure 1).
When evaluating PC survival curves, SPN had the best prognosis with >75% of people alive at 16-year follow-up, followed by PNET (median survival of 150.5 months), IPMN-ADC (108 months), PD-NET (56.2 months), SN (47.1 months), MC (32.4 months), ACC (23.1 months), MCN (8.7 months), ADSC (7.6 months), NOS-PDAC (6.6 months), PDAC (6.6 months), SRCC (4.3 months), and lastly UDC (3.3 months), all P<0.01 (Figure 2).

Discussion:
Nationwide USCS data, covering ≈100% of US population, showed a rapid increase in incidence of PNET, IPMN-ADC, and SPN which also tend to have superior survival when compared to other pancreatic cancers. Future studies are warranted to investigate the factors associated with the variation in PC subtypes prognosis and to evaluate PC outcomes in demographic-specific populations.

Figure 1: Age-Adjusted Incidence Rates Per 100,000 Population of Different Histopathological Subtypes of Pancreatic Cancer<br /> A: AAPC of PDAC incidence rates was significantly increasing (P<0.01)<br /> B: AAPC of PNET incidence rates was significantly increasing (P<0.01)<br /> C: AAPC of NOS-PDAC incidence rates was significantly increasing (P<0.01)<br /> D: AAPC of IPMN-ADC incidence rates was significantly increasing (P=0.02)<br /> E: AAPC of SPN incidence rates was significantly increasing (P<0.01)<br /> F: AAPC of ADSC incidence rates was significantly increasing (P<0.01)<br /> G: AAPC of ACC incidence rates was significantly increasing (P<0.01)<br /> H: AAPC of PD-NET incidence rates was stable (P=0.21)<br /> I: AAPC of MCN incidence rates was significantly decreasing (P<0.01)<br /> J: AAPC of UDC incidence rates was significantly decreasing (P<0.01)<br /> K: AAPC of SRCC incidence rates was significantly decreasing (P<0.01)<br /> There were too few cases for MC and SN to be trended

Figure 1: Age-Adjusted Incidence Rates Per 100,000 Population of Different Histopathological Subtypes of Pancreatic Cancer
A: AAPC of PDAC incidence rates was significantly increasing (P<0.01)
B: AAPC of PNET incidence rates was significantly increasing (P<0.01)
C: AAPC of NOS-PDAC incidence rates was significantly increasing (P<0.01)
D: AAPC of IPMN-ADC incidence rates was significantly increasing (P=0.02)
E: AAPC of SPN incidence rates was significantly increasing (P<0.01)
F: AAPC of ADSC incidence rates was significantly increasing (P<0.01)
G: AAPC of ACC incidence rates was significantly increasing (P<0.01)
H: AAPC of PD-NET incidence rates was stable (P=0.21)
I: AAPC of MCN incidence rates was significantly decreasing (P<0.01)
J: AAPC of UDC incidence rates was significantly decreasing (P<0.01)
K: AAPC of SRCC incidence rates was significantly decreasing (P<0.01)
There were too few cases for MC and SN to be trended

Figure 2: Overall Survival Curves for Different Histopathological Subtypes of Pancreatic Cancer.<br /> Solid Pseudopapillary Neoplasm (SPN) had the best prognosis, followed by Pancreatic Neuroendocrine Tumor (PNET), and intraductal papillary mucinous neoplasm-related adenocarcinoma (IPMN-ADC).

Figure 2: Overall Survival Curves for Different Histopathological Subtypes of Pancreatic Cancer.
Solid Pseudopapillary Neoplasm (SPN) had the best prognosis, followed by Pancreatic Neuroendocrine Tumor (PNET), and intraductal papillary mucinous neoplasm-related adenocarcinoma (IPMN-ADC).

Background: Small nonfunctioning pancreas neuroendocrine tumors (pNETS) are thought to be biologically indolent. The National Comprehensive Cancer Network (NCCN) and European Neuroendocrine Tumor Society (ENETS) suggest that incidentally discovered nonfunctioning pNETS <20mm can be selectively observed due to low risk of malignancy. The 2020 North American Neuroendocrine Tumor Society (NANETS) consensus paper recommends that a decision to observe or resect an asymptomatic 11-19mm pNET may be individualized. Routine prognostic marker clinical testing with DAXX/ATRX immunohistochemistry and telomere-specific fluorescence in-situ hybridization (FISH) are not routinely available. In the absence of such biomarkers, we sought to identify tumor size and Ki-67% thresholds associated with disease progression, and the incidence of disease progression within the <10mm and 11-19mm pNET population.
Methods: A single center 20-year retrospective review of 180 consecutive primary pancreas pNET endoscopic ultrasound fine needle aspiration or core biopsy (FNA/FNB) examinations (January 2001-March 2021) were identified using natural language processing via a pathology database and followed over 36.9 (12.2-73.1) months: [65 (54-72) years, 96 (53%) male, 166 (92.2%) solitary, size 16.5 (10-29.5) mm, 86 (47.8%) tail location, 115 (63.9%) FNA and 65 (36%) FNB. Genetic pNET syndromes and non-primary NETS were excluded from evaluation.
Results: Disease progression (25; 14%) was associated with pNET size [34mm (19-51.5) vs. 15mm (10-26); p<0.0001] and predicted with 52% sensitivity and 86.5% specificity using a lesion size ≥32 mm (Table 1). It was also predicted with 71.4% sensitivity and 70.5% specificity using a Ki-67% of ≥3.9% mm (Figure 1). Disease progression was evident in 3 (6.4%), 3 (5.8%) and 19 (23.5%) of ≤10mm, 11-19mm, and ≥20mm pNETs, respectively.
In the potential “observe or resect” 11-19mm cohort, 52 (28.9%) patients were identified and followed over 37.2 (17.7-66.3) months: [65 (53-73) years, 30 (57.7%) male, 48 (92.3%) solitary, and 29 (55.8%) in the pancreas tail. Ki-67 proliferation index was available for 28 (53.8%), and 9 (32.1%) were of intermediate grade (G2; Ki-67 index 3-18%). Disease stage ≥ II was identified in 13 (52%), with disease progression to bone and brain and an overall mortality rate of 13.5%.
Conclusions: In our 20-year experience with non-genetic pNET EUS FNA/FNB lesions, in the absence of DAXX/ATRX immunohistochemistry and telomere-specific FISH, we identified that disease progression was associated with a lesion size ≥32mm and Ki-67% ≥3.9%. Importantly, almost 6% of <20mm pNETS had evidence of disease progression and are a cohort that would greatly benefit from more advanced incremental ancillary prognostic markers to aid future individualized care.

<b>Table 1.</b> Endoscopic Ultrasound Pancreatic Neuroendocrine Tumor Features Among Individuals with and without Disease Progression. pNET=pancreatic neuroendocrine tumor.

Table 1. Endoscopic Ultrasound Pancreatic Neuroendocrine Tumor Features Among Individuals with and without Disease Progression. pNET=pancreatic neuroendocrine tumor.

<b>Figure 1A</b>. ROC Analysis of Pancreatic Neuroendocrine Tumor Size with Disease Progression.<br /> <b>Figure 1B</b>. ROC Analysis of Ki-67 Index with Disease Progression.

Figure 1A. ROC Analysis of Pancreatic Neuroendocrine Tumor Size with Disease Progression.
Figure 1B. ROC Analysis of Ki-67 Index with Disease Progression.

Ferga Gleeson

Mayo Foundation for Medical Education and Research

Appearances