Background: Intraductal papillary mucinous neoplasms (IPMN) are premalignant cystic tumors with variable behavior.Elevated serum CA 19-9 was incorporated as a worrisome feature in the International Association of Pancreatology 2017 guidelines, and may be associated with an increased risk of high-grade dysplasia and invasive carcinoma (HGD/IC).We aimed to compare the independent predictive value of CA 19-9 with other worrisome and highrisk features to better delineate its utility and importance.Methods: We conducted a retrospective analysis of all patients who underwent pancreatic resection for IPMN at a high-volume tertiary center in the United States.Clinicopathologic and radiologic variables were compared by HGD/IC status.Multivariable logistic regression was used to adjust for worrisome features and high-risk stigmata.Results: Of 645 patients who underwent pancreatectomy, 28% had HGD/IC.Patients with HGD/IC were older (71 vs. 68, p 10mm (OR 3.7, p 5mm (OR 3.2, p< 0.001) were significant predictors of HGD/IC.Conclusion: While abnormal serum CA 19-9 levels were associated with HGD/IC, serum CA 19-9 above 100 U/mL performed more similarly to other high-risk features, and was in fact the strongest predictor of HGD/IC.Our results suggest that serum CA 19-9 above 100 U/mL warrants consideration as a high-risk feature 70.

Purpose: Multimodal treatment of clinical stage I-III pancreatic ductal adenocarcinoma (PDAC) consists of curative-intent surgical resection combined with perioperative therapy. Textbook outcome is a major quality control endpoint. We aimed to evaluate surgical and perioperative textbook outcomes in PDAC patients using national registries. Methods: Patients with clinical stage I-III PDAC and surgical resection from 2010-2020 in the United States and Germany were identified using the National Cancer Database (NCDB) and National Cancer Registries data (GCRG/ADT), respectively. Textbook surgical outcome was defined as R0 resection and >12 harvested lymph nodes. The composite endpoint of textbook surgical and perioperative therapy outcome was defined as R0 resection, >12 harvested lymph nodes, and receipt of perioperative therapy. Results: A total of 38,157 patients from the NCDB and 14,589 patients from the GCRG/ADT database were included. In the NCDB, 33,085 (87%) patients had a complete oncologic resection (R0) and 4,839 (13%) had R1 resections. In the GCRG/ADT, 11,595 (79%) patients had R0 resections while 2,674 (18%) had R1 resections. In the NCDB, 9,580 (25%) of the patients and 556 (4%) in the GCRG/ADT had <12 lymph nodes harvested. From the NCDB, 30,036 (79%) of patients underwent perioperative therapy, 8,867 (23%) had neoadjuvant therapy alone, 18,776 (49%) had adjuvant therapy alone and 2,393 (6%) had both. In the GCRG/ADT, 8,333 (57%) patients had perioperative therapy, 2,300 (16%) had neoadjuvant therapy alone, 5992 (41%) had adjuvant therapy alone and 41 (0.2%) underwent both. Textbook surgical outcome was achieved in 24,553 (64%) patients in the NCDB and 11,234 (77%) from the GCRG/ADT. The composite endpoint of textbook surgical and perioperative therapy outcome was achieved in 19,818 (52%) patients in the NCDB and 7878 (54%) patients in the GCRG/ADT. Finally, median OS in patients with composite textbook outcome was 32 months in the NCDB and 27 months in the GCRG/ADT (p < 0.001), while those with non-textbook outcome had a median overall survival of 22 months in the NCDB and 20 months in the GCRG/ADT (p < 0.001). Conclusion: Surgical textbook outcomes were achieved in more than 60% of stage I-III PDAC for both the NCDB and the GCRG/ADT while more than half of the patients achieved the composite surgical and perioperative textbook outcomes. Failure to achieve textbook outcomes was associated with impaired survival across both registries.

Background: Change in tumor size and serum carbohydrate antigen (CA) 19-9 are commonly reported metrics used to assess response to neoadjuvant therapy (NAT) in pancreatic ductal adenocarcinoma (PDAC). We evaluated the impact of the percentual tumor size reduction (TSR) and CA 19-9 on resectability and response to neoadjuvant FOLFIRINOX.

Background: While advances in systemic therapy and optimal treatment sequencing of pancreatic adenocarcinoma (PDAC) have been made in recent years, nihilism towards management of PDAC persists and serves as a barrier to effective treatment delivery. The aim of this study was to examine national trends in the treatment of early-stage PDAC. Methods: Patients diagnosed with stage I and II PDAC from 2007 to 2017 were identified from the SEER-Medicare linked database. Multilevel mixed-effects models were used to assess adjusted overall survival rates and predictors of undergoing therapy. Results: Of 13,289 patients with early-stage PDAC, 3,115 (23.4%) received chemotherapy and underwent surgical resection, 2,595 (19.5%) underwent surgical resection only, 2,703 (20.3%) received chemotherapy only, and 4,876 (36.8%, Figure 1A) received no treatment. Median overall survival (OS) for the chemotherapy and surgical resection, surgical resection only, chemotherapy only, and no treatment groups were 24.5 months, 15.3 months, 10.6 months, and 4.1 months respectively (p< 0.0001, Figure 1B). After adjusting for patient demographics and tumor characteristics, the correlations with OS persisted when compared to the no treatment group (chemotherapy and surgical resection, HR 0.25, 95% CI 0.23-0.26; surgical resection only, HR 0.30, 95% CI 0.29-0.32; chemotherapy only, HR 0.55, 95% CI 0.53-0.58). Factors associated with not receiving chemotherapy or undergoing surgical resection included Asian and Black race (vs. White), older age, higher Charlson comorbidity score, pancreatic head mass (vs. body or tail), and AJCC stage I tumors (vs. stage II). In addition, living in counties with higher proportions of non-English speaking patients as well as living below the poverty level (p< 0.001) were significantly associated with not receiving therapy. Conclusion: A significant proportion of patients with early-stage PDAC do not receive any treatment, with less than a quarter of patients receiving chemotherapy and undergoing surgical resection. National efforts should be made to continue to improve access to multidisciplinary care and effective therapies for patients with early-stage PDAC.

Presenter: Emily Witt MSc | Massachusetts General Hospital Background: Serum markers of inflammation, such as the neutrophil-lymphocyte ratio (NLR), the lymphocyte-monocyte ratio (LMR), and the platelet-lymphocyte ratio (PLR), have previously been associated with poor survival for patients with pancreatic ductal adenocarcinoma (PDAC). However, prior findings have been inconsistent, and it is unclear for which populations, and at what cut-off values, these markers may be helpful. Very few studies have evaluated the prognostic value of these hematologic parameters in patients treated with neoadjuvant therapy. This study sought to use a large institutional sample to evaluate the preoperative prognostic capacity of these inflammatory indices in patients treated with neoadjuvant chemotherapy and/or radiation therapy in PDAC patients prior to resection. Methods: Data were collected on patients from our institution who underwent resection of PDAC by pancreaticoduodenectomy or distal pancreatectomy after receiving neoadjuvant therapy from 2010-2017. Patients with incomplete follow-up data and those who did not have a complete blood count with differential drawn within 31 days of their operation were excluded. Preoperative labs were used to calculate the NLR (absolute neutrophil/lymphocyte count), LMR (absolute lymphocyte/monocyte count), and PLR (platelet/absolute lymphocyte count). The Kaplan-Meier method with log-rank tests was used for initial time-to-event analysis. Optimal cut-off values for hematologic parameters were determined by receiver-operator curve (ROC) analysis of the continuous variables with 1, 2, and 5-year survival time points, and subsequent calculation of the Youden’s J statistic for each. Univariable and multivariable Cox proportional hazards analyses were performed to identify associations between clinical, pathological, and immunological factors and overall survival (OS). Results: 233 patients (53.3% female) were included in the analysis (median age:66; IQR:59-71). The median disease-free survival (DFS) and OS were 12.1 months (5.0-24.6) and 19.9 months (11.1-34.7), respectively. Univariable analysis revealed age, American Society of Anesthesiology (ASA) score, preoperative CA19-9, American Joint Committee on Cancer (AJCC) stage 8th edition, neoadjuvant FOLFIRINOX/FOLFOX, adjuvant therapy, margin positivity, and tumor size to be associated with OS (p<0.05). Preoperative CA19-9, AJCC stage, neoadjuvant FOLFIRINOX/FOLFOX, adjuvant therapy, and margin positivity were identified as independent predictors of OS by multivariable analysis (p<0.05). NLR, LMR, and PLR were not predictive of OS using multiple cut-off values. Conclusion: Preoperative NLR, LMR, and PLR do not predict OS in patients from our institution treated with neoadjuvant therapy prior to PDAC resection. Traditionally predictive clinicopathological factors, such as CA19-9, AJCC stage, and margin status, remain valuable prognostic indicators for PDAC patients treated with neoadjuvant therapy followed by surgery.

AbstrasctBackground An increasing body of evidence suggests that microbiota may promote progression of pancreatic ductal adenocarcinoma (PDAC). It was hypothesized that gammaproteobacteria (such asKlebsiella pneumoniae) influence survival in PDAC, and that quinolone treatment may attenuate this effect. Methods This was a retrospective study of patients from the Massachusetts General Hospital (USA) and Ludwig-Maximilians-University (Germany) who underwent preoperative treatment and pancreatoduodenectomy for locally advanced or borderline resectable PDAC between January 2007 and December 2017, and for whom a bile culture was available. Associations between tumour characteristics, survival data, antibiotic use and results of intraoperative bile cultures were investigated. Survival was analysed using Kaplan-Meier curves and Cox regression analysis. Results Analysis of a total of 211 patients revealed that an increasing number of pathogen species found in intraoperative bile cultures was associated with a decrease in progression-free survival (PFS) (-1 center dot 9 (95 per cent c.i. -3 center dot 3 to -0 center dot 5) months per species;P = 0 center dot 009). Adjuvant treatment with gemcitabine improved PFS in patients who were negative forK. pneumoniae(26 center dot 2versus15 center dot 3 months;P = 0 center dot 039), but not in those who tested positive (19 center dot 5versus13 center dot 2 months;P = 0 center dot 137). Quinolone treatment was associated with improved median overall survival (OS) independent ofK. pneumoniaestatus (48 center dot 8versus26 center dot 2 months;P = 0 center dot 006) and among those who tested positive forK. pneumoniae(median not reachedversus18 center dot 8 months;P = 0 center dot 028). Patients with quinolone-resistantK. pneumoniaehad shorter PFS than those with quinolone-sensitiveK. pneumoniae(9 center dot 1versus18 center dot 8 months;P = 0 center dot 001). Conclusion K. pneumoniaemay promote chemoresistance to adjuvant gemcitabine, and quinolone treatment is associated with improved survival.

Background: SMAD4, a tumor suppressor gene, is inactivated or deleted in 60-90% of pancreatic adenocarcinomas (PDA). Loss of SMAD4 allows tumor progression by limiting cell cycle arrest and apoptosis and increasing metastases. SMAD4 deficient PDA cells are resistant to radiotherapy by upregulation of autophagy, a cell survival mechanism that allows intracellular recycling of macromolecules and organelles. Hydroxychloroquine (HCQ) is a known autophagy inhibitor, suggesting that HCQ treatment in SMAD4 deficient PDA may prevent therapeutic resistance induced by autophagy upregulation.

Background: Current guidelines for IPMN include an elevated serum carbohydrate antigen (CA) 19-9 among the worrisome features. However, the correlation of CA 19-9 with histological malignant features and survival is unclear. Serum CEA is also currently used for preoperative management of IPMN, although its measurement is not evidence-based. Accordingly, we aimed to assess the role of these tumor markers as predictors of malignancy in IPMN. Methods: IPMN resected between 1998 and 2018 at Massachusetts General Hospital were analyzed. Clinical, pathological and survival data were collected and compared to preoperative levels of CA 19-9 and CEA. Receiver operating characteristic (ROC) and Cox regression analyses were performed considering cut-offs of 37 U/ml (CA 19-9) and 5 mu g/l (CEA). Results: Analysis of 594 patients showed that preoperative CA 19-9 levels > 37 U/ml (n = 128) were associated with an increased likelihood of invasive carcinoma when compared to normal levels (45.3% vs. 18.0%, P < 0.001), while there was no difference with respect to high-grade dysplasia (32.9% vs 31.9%, P = 0.88). The proportion of concurrent pancreatic cancer was higher in patients with CA 19-9 > 37 U/ml (17.2% vs 4.9%, P < 0.001). An elevated CA 19-9 was also associated with worse overall and disease-free survival (HR = 1.943, P = 0.007 and HR = 2.484, P < 0.001 respectively). CEA levels did not correlate with malignancy. Conclusion: In patients with IPMN, serum CA19-9 > 37 U/ml is associated with invasive IPMN and concurrent pancreatic cancer as well as worse survival, but not with high-grade dysplasia. Serum CEA appears to have minimal utility in the management of these patients. (C) 2020 IAP and EPC. Published by Elsevier B.V. All rights reserved.

Presenter: Neha Shafique BA | Massachusetts General Hospital Background: Patients undergoing pancreatic surgery at teaching hospitals have been shown to experience a shorter length of stay and lower in-hospital mortality compared to those undergoing surgery at non-teaching hospitals. Academic programs have also been associated with improved oncologic outcomes with a study showing improved 5-year overall survival (OS) following resection for pancreatic ductal adenocarcinoma (PDAC) compared to community programs. With the advent of minimally invasive surgery, it remains unclear whether academic programs continue to be associated with improved outcomes. This study sought to evaluate differences by facility type in postoperative outcomes and OS for patients who underwent minimally invasive resections for PDAC. Methods: The National Cancer Database (NCDB) was used to identify patients diagnosed with stage 1 to 3 PDAC who underwent a pancreatic resection from 2010 to 2014. Patients were included if their operation was performed minimally invasively via either a laparoscopic or robotic approach. Exclusion criteria included stage 4 disease, unknown facility type, or care at integrated networks. Hospitals were characterized as academic or community programs according to American College of Surgeons Commission on Cancer designations. Multivariable logistic regression was used to identify predictors of postoperative clinical and oncologic outcomes, and an adjusted Cox regression was used to compare OS between academic and community programs. Results: Of 2,136 patients who met inclusion criteria, 542 (25.4%) were treated at community hospitals and 1,594 (74.6%) were treated at academic hospitals. The median follow-up interval for the cohort was 18 months. No significant differences in age, sex, race, insurance type, Charlson-Deyo score, clinical stage or tumor grade were identified between the two groups. However, patients treated at academic hospitals were more likely to travel more than 40 miles for treatment (39.1% vs 20.5%, p<0.0001). The median number of NCDB-reported pancreatectomies performed per year at hospitals that also reported at least one minimally invasive pancreatectomy during 2010 to 2014 was 6. Thus, high-volume hospitals were defined as those that reported, on average, more than 6 pancreatectomies per year. High-volume hospitals were more likely to be academic (65.8% academic vs. 34.2% community, p<0.0001). Treatment at an academic hospital was an independent predictor of receiving neoadjuvant therapy (OR 1.69, 95% CI 1.10 – 2.60, p = 0.02) and attaining fewer positive margins following surgery (OR 0.62, 95% CI 0.47 – 0.82, p = 0.001). Surgery at academic hospitals was also independently associated with improved 1-year OS (HR 0.86, 95% CI 0.74 – 0.99, p = 0.04). Conclusion: Patients were more likely to undergo neoadjuvant therapy when receiving treatment at academic hospitals. Minimally invasive pancreatic resections for PDAC performed at academic hospitals were associated with fewer positive margins and improved 1-year OS even after adjusting for facility volume, patient and disease characteristics, and receipt of neoadjuvant therapy.

BACKGROUND:Pancreatic cysts <15 mm without worrisome features have practically no risk of malignancy at the time of diagnosis but this can change over time. Optimal duration of follow-up is a matter of debate. We evaluated predictors of malignancy and attempted to identify a time to safely discontinue surveillance. METHODS:Bi-centric study utilizing prospectively collected databases of patients with pancreatic cysts measuring <15 mm and without worrisome features who underwent surveillance at the Massachusetts General Hospital (1988-2017) and at the University of Verona Hospital Trust (2000-2016). The risk of malignant transformation was assessed using the Kaplan-Meier method and parametric survival models, and predictors of malignancy were evaluated using Cox regression. RESULTS:806 patients were identified. Median follow-up was 58 months (6-347). Over time, 58 (7.2%) cysts were resected and of those, 11 had high grade dysplasia (HGD) or invasive cancer. Three additional patients had unresectable cancer for a total rate of malignancy of 1.7%. Predictors of development of malignancy included an increase in size ≥2.5 mm/year (HR = 29.54, 95% CI: 9.39-92.91, P < 0.001) and the development of worrisome features (HR = 9.17, 95% CI: 2.99-28.10, P = 0.001). Comparison of parametric survival models suggested that the risk of malignancy decreased after three years of surveillance and was lower than 0.2% after five years. CONCLUSIONS:Pancreatic cysts <15  mm at the time of diagnosis have a very low risk of malignant transformation. Our findings indicate the risk decreases over time. Size increase of ≥2.5 mm/year is the strongest predictor of malignancy.

Postoperative major morbidity has been associated with worse survival gastrointestinal tumors. This association remains controversial in pancreatic cancer (PC). We analyzed whether major complications after surgical resection affect long-term survival.

Background: Neoadjuvant therapy (NAT) has been accepted as a strategy to treat borderline resectable and locally-advanced pancreatic adenocarcinoma (PDAC). FOLFIRINOX in addition to radiation therapy allows selected patients the opportunity to undergo a potentially curative operation, with a high rate of margin-negative resection. Given that the vast majority of PDAC patients that undergo curative treatment either recur or develop distant metastases, we sought to evaluate if NAT leads to changes in patterns of progression. Methods: Using the MGH prospectively maintained PDAC database we identified patients with resectable PDAC. First site of progression was designated as either locoregional (LR) or as distant metastasis (DM). DM included metastases to liver, lung, peritoneum, or other sites. If both LR and DM were identified on imaging, patients were categorized as DM. Patients with resectable disease were treated with upfront resection, adjuvant chemotherapy, and individualized radiotherapy. In order to understand if NAT in borderline resectable patients led to a difference in site of first progression compared to resectable disease, we investigated patients from the MGH Phase-II trial evaluating total neoadjuvant therapy (TNT) in borderline-resectable PDAC (Murphy et al. Jama Oncol, 2018), in which site of first progression was collected prospectively. P-value < 0.05 was considered significant. Results: Using the MGH PDAC database from 2011 to 2017, 99 resectable patients progressed during a median follow-up of 60.1 months. The median disease-free survival (mDFS) in this group was 13.0 months. The median overall survival (mOS)was 23.4 months. Patients with resectable PDAC were treated with adjuvant chemotherapy (100%) and radiation (35.5%). Margin-negative resection was achieved in 77 of 99 patients (77.8%). In resectable patients, site of first progression was found to be LR in 18.2% of patients and was DM in 81.8% of patients. We compared resectable patients to patients from the MGH Phase-II trial (NCT01591733) evaluating TNT in borderline resectable PDAC. Out of 48 evaluable patients, 36 recurred during the median follow-up period of 36 months. The mDFS was 17.7 months and the mOS was 37.7 months. All patients in this trial received neoadjuvant chemotherapy with FOLIFIRINOX, followed by chemoradiation. 31 of 48 patients (65%) achieved an R0 resection. Out of these patients, 50% had site of first progression locoregionally and 50% had site of first progression as DM. The distribution of site of first progression was different between the two cohorts (p < 0.0001) and demonstrated an increase in the early failure of locoregional control in patients with borderline resectable PDAC treated with NAT. Conclusion: Upfront resectable patients with PDAC who progress after curative-intent resection and adjuvant therapy do so more often distantly than locoregionally. However, in patients with borderline resectable PDAC treated with neoadjuvant chemotherapy and radiation, there is an increase in local failure as site of first progression, versus distant control in comparison to resectable patients treated with adjuvant therapy. This may be a reflection of altered disease biology between borderline and resectable patients, or better distant control with NAT.