Background: Frailty is associated with increased postoperative complications. Despite evidence-based modalities to assess frailty, surgeons often make subjective assessments of patients' ability to tolerate surgery, and preoperative interventions are not commonplace to improve functional status. Objective assessments of frailty in the Surgical Oncology population may improve patient selection for major resections.

Background: Adjuvant hepatic artery infusion (HAI) chemotherapy in patients with colorectal liver metastasis (CRLM) has demonstrated survival benefit over systemic chemotherapy alone. However, this surgical approach has been offered only at a few centers as of recently. As new HAI centers emerge, it is important to better understand patient outcomes and survival associated with this surgical approach.

Introduction: Major postoperative complications (MPC) after pancreatoduodenectomy limit adjuvant chemotherapy (AC) administration, which may factor in increasing use of neoadjuvant chemotherapy (NC) for resectable pancreatic ductal adenocarcinoma (PDAC). Here we examined effects of NC on outcomes and MPC on AC administration. Methods: We performed a retrospective chart review of patients who underwent pancreatoduodenectomy for PDAC at our institution, 2015-2020. Demographic, clinicopathologic, and postoperative variables were analyzed. Results: We identified 146 patients, 68 (47%) received NC (Table). NC patients had smaller tumors [median 2.4cm vs. 2.8cm, p<0.002]; higher R0 resection rates (94% vs. 85%, p=0.048); lower vascular invasion rates (35% vs. 71%, p<0.001); lower LODDS [log of positive to negative lymph nodes; -1.2 (-1.5-(-0.9) vs. (-0.8 (-1.3-(-0.4), p<0.001]. NC was not associated with MPC or AC administration. With 21 months median follow-up, there was a trend towards improved recurrence free survival (RFS) (median 40 vs. 24 months, p=0.327) in NC patients. Of 78 (53%) patients who did not receive NC, 13 (17%) experienced MPC, and AC receipt was lower in this group (80% vs 39% p=0.006), with a trend toward lower RFS (median 25 vs 16 months p=0.456). On univariate regression analysis, MPC were associated with lower AC administration rates, both in patients that received and did not receive NC (OR 0.091, p=0.002; and OR 0.156, p=0.006, respectively). Conclusions: NC before pancreatoduodenectomy for PDAC was associated with favorable pathology and a trend toward improved RFS. As MPC limits AC administration, NC should be considered for all patients with resectable PDAC.

Introduction: Systemic chemotherapy (SYS) is recommended in the management of pancreatic ductal adenocarcinoma (PDAC) and distal cholangiocarcinoma (dCCA). Postoperative complications after pancreatoduodenectomy may limit adjuvant chemotherapy (AC) administration, therefore neoadjuvant chemotherapy (NC) is utilized increasingly for PDAC, but not dCCA. Here we examined the impact of SYS timing and postoperative complications on the receipt of SYS in patients undergoing pancreatoduodenectomy for PDAC and dCCA. Methods: We retrospectively reviewed records of patients who underwent pancreatoduodenectomy at our institution between 2015-2020 for dCCA or PDAC, comparing receipt of SYS through analysis of demographic, clinicopathologic, and postoperative variables. Results: We identified 187 patients (dCCA N=41, PDAC N=146). SYS administration was significantly lower in the dCCA group [66% vs. 86%, p=0.001], due to lower NC administration rate [10% vs. 47%, p<0.001] (Table). Rates of major postoperative complications and AC administration were similar between groups. In dCCA patients, postoperative major complications were associated with lower rate of AC receipt [30% (3/10) vs. 74% (23/31), p=0.022]. On univariate logistic regression major complications were associated with reduced probability of AC receipt (OR 0.15, p=0.018). Conclusions: In patients undergoing pancreatoduodenectomy for dCCA or PDAC, postoperative complications and AC administration rates were similar between groups. In the dCCA group, major postoperative complications were associated with a lower rate of AC administration. Any SYS receipt was higher in PDAC patients, due to increased rates of NC administration. These findings suggest a reconsideration in timing SYS administration for patients with dCCA to the neoadjuvant or perioperative setting.

Presenter: Yael Berger MD | Icahn School of Medicine at Mount Sinai Background: Gallbladder cancer accounts for 1.2% of total global cancer diagnoses, with an increasing incidence in the developed world. The literature on clinicopathologic characteristics and prognosis of biliary type adenocarcinoma and specifically of carcinoma arising from intracholecystic papillary-tubular neoplasms of the gallbladder is limited. This study describes a single institution experience in New York City. Methods: A retrospective review was performed of patients who underwent cholecystectomy for a malignant neoplasm of the gallbladder between 2007 and 2017. Demographic, clinicopathologic, and operative variables, and survival outcomes were analyzed. All available pathologic specimens were re-reviewed by a hepatobiliary pathologist. Results: A total of 145 patients were identified. The most common histological subtype was biliary-type adenocarcinoma, accounting for 93 (64%) cases. Compared to non-biliary adenocarcinoma, biliary-type adenocarcinomas were diagnosed at a lower AJCC pathologic stage (p = 0.045) [stage I- 18% versus 8% respectively, p = 0.2; stage II- 33% versus 19% respectively, p = 0.2; stage III- 42% versus 50% respectively, p = 0.5; stage IV- 7% versus 23% respectively, p = 0.023] and demonstrated a trend towards higher margin-negative (R0) resection rate (81% versus 62%, p = 0.06). Furthermore, median recurrence free survival was significantly longer for biliary-type adenocarcinoma when compared to other adenocarcinoma subtypes [38 months versus 16 months respectively, p = 0.014; median follow-up 36 months;]. Tumors arising from intracholecystic papillary-tubular neoplasms were found in 18 (12%) patients and were more commonly associated with biliary-type adenocarcinoma (14 patients). Compared to 29 cases of biliary-type adenocarcinoma that were not associated with intracholecystic papillary-tubular neoplasm, biliary-type adenocarcinoma associated with intracholecystic papillary-tubular neoplasm demonstrated lower pathologic stage (p = 0.006) [stage I- 57% versus 14% respectively, p = 0.009; stage II- 36% versus 28% respectively, p = 0.7; stage III- 7% versus 55% respectively, p = 0.003; stage IV- 0% versus 3% respectively, p = 1] and lower rates of liver invasion (0% versus 49%, p<0.001) and peri-neural invasion (14% versus 48%, p = 0.04). Overall 5-year survival rate was higher for patients with gallbladder neoplasm of any subtype associated with intracholecystic papillary-tubular neoplasms compared to those that were not associated with intracholecystic papillary-tubular neoplasms (54% versus 41%, p = 0.019; Median follow-up 23 months); this difference was also found to be significant when comparing between biliary-type adenocarcinoma alone (63% for tumors arising from intracholecystic papillary-tubular neoplasms versus 52% for tumors that did not, p = 0.005) [Figure]. Conclusion: This study demonstrates unique pathological and prognostic features of biliary-type adenocarcinoma and of carcinoma arising from intracholecystic papillary-tubular neoplasms. Histopathological variance may implicate prognosis and be used to better guide clinical decision making in treatment of these patients.

Background In the United States, mortality after a diagnosis of hepatocellular carcinoma (HCC) is higher in patients who are Black than in patients of other racial groups. The objective of this study was to clarify factors contributing to this disparity by analyzing liver and tumor characteristics in patients with HCC who have a history of hepatitis C virus (HCV) infection. Methods Records of patients with HCV and HCC at the authors' institution from 2003 to 2018 were retrospectively reviewed. Race and ethnicity were self-identified. Imaging, laboratory, and pathologic features were compared between Black and non-Black cohorts. Results Among 1195 individuals with HCC, 390 identified as Black. At the time of HCC diagnosis, Black patients had better liver function, as measured by Child-Pugh score, Model of End-Stage Liver Disease score, histology of nontumor tissue, and fibrosis-4 (FIB-4) score (all P < .05). FIB-4 scores were <3.25 in 31% of Black patients. In addition, Black patients had less early stage HCC (20.2% vs 32.3%; P < .05), larger tumors (median [interquartile range]: 3.5 cm [2.2-6.2 cm] vs 3.1 cm [2.1-5.1 cm]; P < .01), more multiple tumors (median, [interquartile range]: 1 tumor [1-3 tumors] vs 1 tumor [1-2 tumors]; P = .03), more poorly differentiated tumors (30.3% vs 20.5%; P < .05), and more microvascular invasion (67.2% vs 56.5%; P < .05). Conclusions Black patients with HCV exposure develop HCC at earlier stages of liver disease than members of other racial groups. Nearly one-third would not qualify for HCC screening using the common FIB-4 cirrhosis threshold. Practice guidelines that stress HCC surveillance for cirrhotic patients with HCV may need to be revised to be more inclusive for Black patients. In addition, tumors in Black patients carry worse prognostic features, and molecular studies are needed to characterize their biologic properties.

However, it is unclear if the technical advantages of the robotic platform result in improved clinical outcomes compared to those following laparoscopic hepatic resection.Methods: Patients undergoing laparoscopic or robotic hepatectomy between 2009 and 2018 were included.Ninety-day morbidity and mortality, as well as length of stay, readmission, operative time, conversion rate and estimated blood loss were compared between the two approaches using bivariate and multivariate logistic regression analysis.Propensity score analysis was also performed to adjust for differences between the two groups.Results: Out of 203 minimally invasive cases, 58 (28.6%) robotic procedures were performed.Resection of 3 segments (9.7% vs. 1.7%, p=0.05) and tumor size 5 cm (23.2% vs. 7.7%, p=0.015) were more common in the laparoscopic group.Superior segment resections (12.1% vs. 3.5%, p=0.019), biliary malignancy (22.4% vs. 8.3%, p=0.05) and hepatic pedicle lymph node dissection (19% vs. 1.4%,P < 0.001) were more common in the robotic group.The use of hand-assistance (2.8% vs. 0%), conversion to open (12.4% vs. 10.3%),estimated blood loss (mean, 260 ml vs. 237 ml), length of hospital stay (mean 3.7 day vs. 3.3 day), readmission (8.3% vs. 10.3%),Clavien-Dindo Class III complications (11.7% vs. 10.3%) and mortality (1.7% vs. 1.4%) were not different between the two groups (P>0.05).Robotic procedures were 50 min longer on average (mean, 265 min vs. 215 min, p=0.028).These findings were persistent on a multivariate logistic regression analysis that adjusted for potential confounders. Ina propensity score-matched cohort with balanced intraoperative and preoperative characteristics (N=37 in each arm) , hand-assistance (2.7% vs. 0%), estimated blood loss (314 vs. 213 ml), length of hospital stay (mean, 3.9 vs. 3.0 days), readmission (5.4% vs. 8.1%), operative time (226 vs. 253 min), Clavien Class III complications (16.2% vs. 10.8%) and mortality (2.7% vs. 2.7%) were not different between laparoscopic and robotic procedures; however, conversion rate was higher in laparoscopic procedures (18.9% vs. 0%, p=0.005).Conclusion: Robotic hepatectomy is associated with longer operative time and lower conversion rate compared to laparoscopic hepatectomy, but overall is not associated with improved clinical outcomes.The robotic platform is associated with increased use for hepatic pedicle lymph node dissection and resection of superior segment liver lesions.

The Chicago Consensus Working Group provides multidisciplinary recommendations for the management of peritoneal surface malignancies of various causes. These guidelines are developed with input from leading experts including surgical oncologists, medical oncologists, pathologists, radiologists, palliative care physicians, and pharmacists. These guidelines recognize and address the emerging need for increased awareness of the appropriate management of peritoneal surface disease. They are not intended to replace the quest for higher levels of evidence. This article serves as an introduction to this effort.

Background: Although gallbladder adenocarcinoma (GBA) is potentially curable with surgical resection, the influence of the squamous histological subtype upon outcomes is not well described. Methods: We retrospectively analyzed all patients with gallbladder cancer (GBC) who underwent surgical resection with curative intent in our health system from 2007 to 2017. We compared clinicopathologic, perioperative, and oncologic outcomes of adenocarcinoma (GBA) to histologic subtypes of adenosquamous (GBAS) and pure squamous cell (GBS). Results: 91 patients met criteria; 3 GBS, 12 GBAS, 76 GBA. The three cohorts had similar preoperative profiles. Patients with any squamous histology had larger primary tumors (60mm GBS, 48mm GBAS vs. 28mm GBA, p < 0.001), requiring adjacent organ resection (33%, 42% vs. 7%, p = 0.003), and were higher stage (stage 3/4 66%, 75% vs. 52%, p = 0.039) with more frequent LVI (67%, 58% vs. 36%, p = 0.200) and G3 disease (67%, 75% vs. 43%, p = 0.382). R0 resection rates and use of adjuvant therapy were similar in all cohorts. Median OS was similar between GBA and GBAS (26 and 23 months), but significantly worse in pure GBS (<5 months) with no patients alive at 1 year (vs. 68% GBAS 71% GBA, p = 0.502). PFS was significantly worse in squamous variants (GBAS 8 months, GBS 1 month) than GBA (15 months), p = 0.001. On multivariate analysis, R1 resection, higher grade disease and recurrence were significant independent predictors of shorter OS (p < 0.05). Although R0 improved OS in GBAS/GBS (24 vs. 14 months R1), it was still extremely poor. Conclusion: Squamous cell variants of gallbladder cancer confer more aggressive disease. Even with R0 resection and adjuvant therapy, surgical resection is not superior to best supportive care. Our data suggest that preoperative identification of a squamous variant may identify patients who could be spared from unnecessary surgery.

Background: Gallbladder cancer (GBC) is a rare aggressive malignancy and there is a paucity of data regarding patterns of recurrence after curative resection. Our goal was to attempt to identify any clinicopathologic features that could be predictive of patterns of recurrence after curative resection. Methods: We retrospectively analyzed all patients who underwent surgical resection with curative intent for GBC in our health system from 2007 to 2017. We compared clinicopathologic, perioperative, and oncologic outcomes of patients with postoperative recurrence (PR) and with no postoperative recurrence (NPR) and performed a logistic regression analysis to determine factors predictive of recurrence location. Results: 143 patients met criteria. At a median follow-up of 18 months, 48% developed recurrence (PR) and 52% (NPR) did not. While the two cohorts underwent similarly extensive surgeries [portal lymphadenectomy (65% vs. 51%), CBD resection (21% vs. 20%), extended hepatectomy (13% vs. 11%), adjacent organ resection (10% vs. 5%), p > 0.05], the PR cohort had significantly larger tumors (median 42 vs. 29 mm, p = 0.017), a higher incidence of LVI (46% vs. 25%, p = 0.017), and more advanced stage disease (Stage 3/4 64% vs. 35%, p = 0.004). The rates of neoadjuvant therapy (9% vs. 3%), adjuvant therapy (50% vs. 21%) and R0 resection (71% vs. 79%) were similar, p>0.05. The PR cohort had significantly worse OS (24 vs. >109 months, p = 0.001). In the PR cohort, median PFS was 8 months; 28% exhibited locoregional recurrence (resection bed/regional lymph nodes), 4% developed peritoneal recurrence, 18% developed distant metastases, and 50% developed both locoregional and distant metastases. Median time to recurrence was 6, 15, 16 and 6 months, respectively. Larger tumors, PNI, LVI, positive lymph nodes, stage 3/4 and lack of adjuvant treatment were all predictors of local recurrence, while only stage 3/4 was predictive of distant metastases. Conclusion: In patients undergoing resection with curative intent for GBC, nearly half will develop postoperative recurrence; the majority in a distant pattern (distant, peritoneal, or locoregional with distant spread). Although the data are insufficient to make definitive conclusions, they suggest that stage and histological features are relevant in considering adjuvant therapies. Large tumors with aggressive histological features are at risk for locoregional recurrence and adjuvant local therapy (i.e. radiation) may be appropriate. Early stage tumors with good histological features may not require adjuvant treatment, however, late stage tumors will likely benefit from systemic chemotherapy to prevent distant recurrence.

Background: The role of adjuvant treatment has been well established in cholangiocarcinoma, Gallbladder cancer (GC) is a rare and less studied tumor. Methods: We retrospectively analyzed all patients with gallbladder malignancy that underwent surgical removal of the gallbladder in our health system from 2007 to 2017. We compared clinical and oncological outcomes for two groups of patients: a group that received adjuvant treatment and a group that did not receive adjuvant treatment. The adjuvant treatment group was composed of patients who received either chemotherapy alone (19 patients) or chemo-radiation (21 patients). Results: Overall 69 patients with advanced (TNM stage III/IVa) gallbladder cancer (GBC) underwent resection for curative purposes. Among them, 40 patients (58%) were in the adjuvant treatment group and 29 patients (42%) were in the NO adjuvant treatment group. Gender (female, 58% vs 60%), ASA score (%) (I – 22 vs. 24, II - 39 vs. 37, III- 32 vs. 39, IV, 7 vs. 0), TNM stage IV (%, 17 vs. 10), and neoadjuvant chemotherapy administration (10% vs. 12%) were comparable (p = NS). The patients in theadjuvant treatment group were significantly younger (mean, years, 63 vs. 70, p = 0.03). Intraoperatively, surgical approach (laparoscopic, %, 28 vs. 25), incidental finding (%, 52 vs. 48), and tumor max dimension (mean, mm, 44 vs. 45) were all similar. Significantly more patients in the adjuvant treatment group underwent radical cholecystectomy (%, 69 vs. 98, p = 0.01). At median follow-up of 20 months, 43 patients recurred. Patients in adjuvant treatment group had significantly better survival (months, 25 vs. 10, p = 0.01). Similarly, the median disease-free survival was improved for patients in adjuvant treatment group (months, 18 vs. 8), although this didn’t reach statistical significance (p = 0.16). Conclusion: Adjuvant treatment has significant survival benefits for patients with advanced GBC and should be offered to patients with acceptable performance status. Despite these data, the application of adjuvant therapy after resection is not universally applied.

Background: Malignant neoplasms of the gallbladder are often lethal with the vast majority represented by adenocarcinomas. While the number of patients is low, previous reports support that approximately 10% of patients are noted to have non-adenocarcinoma histology on pathologic analysis. Little is known about how these rare histological variants affect presentation and survival. Methods: 145 patients with pathologically confirmed neoplasms of the gallbladder underwent surgery at our institution from 2007–2017 and were included from a prospectively maintained database. Pathological analysis revealed 8 histologic types distinct from adenocarcinoma. Presenting symptoms, pathologic stage, overall survival (OS), and progression free survival (PFS) were evaluated for each histology. Results: Out of 145 patients, 79 patients had adenocarcinoma (AC, 54%) and 66 had non-adenocarcinoma (non-AC, 46%) histologies. Of the 8 non-AC variants, the histologies included papillary (n = 20), adenosquamous (AS) (n = 12), mucinous (n = 8), squamous (n = 3), signet ring (n = 3), lymphoma (n = 3), neuroendocrine (NE) (n = 3), and sarcomatoid (n = 1). The majority of histologic types presented in the 6th decade (median 65 years), although neuroendocrine and sarcomatoid presented earlier and later (57 years, 93 years), respectively. The majority of patients were symptomatic at presentation regardless of histology (75-100%). While 29% of AC patients presented with jaundice, it was not seen in any patients with AS (0/11), squamous (0/3), or NE (0/3) variants and all patients with signet ring (3/3) and sarcomatoid variants (n = 1). Lymph node involvement was 54% in AC patients and 46% in non-AC patients (p = 0.896). The majority of AS (75%) and signet ring (100%) tumors were poorly differentiated versus 47% of AC tumors. Mean overall survival was 23 months in AC patients versus 26 months in non-AC patients (p = 0.159) while progression-free survival was 20 months and 22 months, respectively (p = 0.951). Conclusion: While non-adenocarcinoma variants of gallbladder cancer are rare, our data suggest they are more common than previously reported in the literature, representing 46% of our cohort. Retrospective reviews and meta-analysis could provide enough data to determine if differences exist between various subtypes, and thus guide future studies in an effort to optimize management of a very challenging disease.

Background: Preoperative jaundice (POJ) in gallbladder cancer (GBC) suggests advanced disease with involvement of the porta hepatis and historically has implied inoperability. However, some studies have postulated that select patients may benefit from surgical resection.

and portal vein resections.The number and severity of complications such as delayed gastric emptying and pancreatic leak were not statistically different between the two groups.Additionally, the 30-day reoperation, readmission, and mortality rates were not statistically different.The total number of deaths in the octogenarian group was 2 (4.2%) vs. 6 (1.7%) in the non-octogenarian group (p = 0.25).The length of stay was 1.6 days shorter in the octogenarians undergoing distal pancreatic resection (p = 0.02).The length of stay after a pancreaticoduodenectomy was similar among the two age groups.All of the octogenarians undergoing distal pancreatectomy and 57% of those undergoing pancreaticoduodenectomy were discharged to home.Discharge to a facility was not predicted by a calculated modified frailty index.Conclusion: At a large-volume academic community cancer center with a dedicated surgical oncology team, octogenarians can undergo complex abdominal surgeries safely with similar outcomes to their younger counterparts.

margins (18.4% vs. 24.2%,p = 0.001).Thirty-day readmission rate was significantly higher in the PD group (6.8% vs. 3.9%, p = 0.004) likely representing the morbidity of POPFs.There was no significant difference in long-term survival identified.Conclusion: Total pancreatectomy has a lower short-term morbidity and similar long-term survival compared to pancreaticoduodenectomy for patients with pancreatic head adenocarcinoma.Therefore, total pancreatectomy is a reasonable alternative to pancreaticoduodenectomy that should be utilized more liberally if clinically indicated.