Intraductal papillary mucinous neoplasms of the pancreas: an updated experience.

OBJECTIVE: To update the authors' experience with intraductal papillary mucinous neoplasms (IPMNs) of the pancreas. BACKGROUND DATA: IPMNs are intraductal mucin-producing cystic neoplasms of the pancreas with clear malignant potential. Since the authors' 2001 report, the number of IPMNs resected at our institution has more than doubled, providing an opportunity to define the clinical features of this distinct neoplasm. METHODS: All patients undergoing pancreatic resection for an IPMN at the Johns Hopkins Hospital between January 1987 and March 2003 were evaluated. Noninvasive IPMNs were classified as "adenoma," "borderline," or "carcinoma-in situ" (CIS) depending on the degree of dysplasia within the specimen. Invasive cancers were classified as tubular, colloid, mixed, or anaplastic types. Pathology was retrospectively reviewed to identify main-duct or branch-duct origin of the tumors. Long-term overall survival for patients having IPMNs with invasive cancer was compared with those patients having IPMNs without an invasive component. RESULTS: Between January 1987 and March 2003, inclusive, 136 pancreatic resections were performed for patients with IPMNs, with 78 resections performed since January 2001. The mean age of the patients was 66.8 +/- 1.1 years, with 57% being male and 89% white. Pancreaticoduodenectomy was performed in 71% of patients, total pancreatectomy in 15%, distal pancreatectomy in 12%, and central pancreatic resection in 2%. IPMNs without evidence of invasive cancer were identified in 62% (n = 84) of patients (17% adenoma, 28% borderline, or 55% CIS). The remaining 38% (n = 52) of patients had IPMNs with associated invasive cancer (60% tubular, 27% colloid, 7% mixed, and 6% anaplastic). The mean age of patients with IPMN adenoma was 63.2 years, 66.7 years for those with borderline/CIS IPMNs, and 68.1 years for those with invasive cancer (P = 0.08, adenomas vs. invasive cancer). In those patients with invasive cancers, 15% had invasive cancer at the final surgical margin, 23% had IPMN without invasive cancer at the margin, and 54% had lymph node metastases. Residual IPMN was identified at the neck or uncinate margin in 24% of patients with noninvasive IPMNs. The overall 5-year survival for patients having IPMNs without invasive cancer was 77% (several deaths secondary to metachronous invasive cancer), compared with 43% in those patients with an invasive component (P < 0.0001). There were no differences in survival when comparing adenomas, borderline neoplasms, and CIS. Similarly, there were no statistically significant differences in survival when comparing branch-duct, main-duct, and combined variants; however, the branch-duct variants were more often noninvasive. For those patients with invasive IPMNs, 2-year survival was 40% when margins were positive for invasive cancer or for IPMN without invasive cancer, and 60% when margins were tumor-free (P = 0.15). Those patients with colloid carcinomas (n = 14) had improved survival compared with those with tubular carcinomas (n = 31), with 5-year survival rates of 83% and 24%, respectively. IPMN recurrences and deaths from cancer occurred in patients with both invasive and noninvasive IPMNs at initial resection. CONCLUSIONS: IPMNs continue to be recognized with increasing frequency. Five-year survival for those patients following resection of IPMNs with invasive cancer (43%) is improved compared with those patients with resected pancreatic ductal adenocarcinoma in the absence of IPMN (averages 15%-25%). Survival following resection of IPMNs without invasive cancer (regardless of degree of dyplasia) is good, but recurrent disease in the residual pancreas suggests that long-term surveillance is critical. Based on the age at resection data, there appears to be a 5-year lag time from IPMN adenoma (63.2 years) to invasive cancer (68.1 years).

p16 Inactivation in pancreatic intraepithelial neoplasias (PanINs) arising in patients with chronic pancreatitis.

Patients with long-standing chronic pancreatitis are thought to be at increased risk of developing pancreatic ductal adenocarcinoma, but the mechanism for this increased risk is unknown. Since increasing evidence supports the notion that infiltrating pancreatic ductal adenocarcinomas arise from pancreatic intraepithelial lesions (PanINs), we sought to determine if patients with chronic pancreatitis harbor PanINs with alterations in tumor suppressor genes that are associated with infiltrating pancreatic ductal adenocarcinoma. We identified 122 patients with a diagnosis of chronic pancreatitis and 29 patients with a well-differentiated pancreatic endocrine tumor that underwent pancreatic surgery at the Johns Hopkins Hospital from 1985 to 1999. PanINs from each resection specimen were identified, graded, counted, and correlated with smoking and alcohol history. The expression patterns of p16 and Smad4 were determined in a subset of PanINs by immunohistochemistry, and the pattern of labeling compared with that seen in PanINs associated with infiltrating adenocarcinoma of the pancreas as identified in prior studies, and to PanINs associated with pancreatic endocrine tumor. Duct lesions were present in 80 of the 122 pancreata with chronic pancreatitis (66%). Of 405 duct lesions identified in the chronic pancreatitis group, 7.6% were reactive changes, 65.5% were PanIN-1A, 18% were PanIN-1B, 7.4% were PanIN-2, and 1.5% were PanIN-3. Within the pancreatic endocrine tumor group, 22 PanINs were identified: 15 PanIN-1A, 4 PanIN-1B, and 3 PanIN-2. There were significantly fewer high-grade PanINs in the pancreata with chronic pancreatitis than in pancreata with pancreatic adenocarcinoma (P < 0.0001). Within the chronic pancreatitis group, the 80 patients with PanINs were significantly older than the 42 patients without PanINs (mean age 57.0 +/- 14.1 years vs. 50.9 +/- 14.7 years, P = 0.01). Smoking history was not associated with PanIN prevalence or grade, but patients who reported a history of excessive alcohol consumption had fewer PanINs (25 of 44 harbored PanINs, 57%) than those who did not (54 of 74, 73%, P = 0.07). In the chronic pancreatitis group, 0% of PanIN-1A, 11% of the PanIN-1B, 16% of the PanIN-2, and 40% of the PanIN-3 lesions showed loss of p16 expression, whereas all of the PanINs from patients with an pancreatic endocrine tumor retained p16 expression. All of the PanINs analyzed from patients with chronic pancreatitis retained normal Smad4 expression. We conclude that a significant minority of PanINs arising in patients with chronic pancreatitis show loss of p16 expression. This alteration, common to pancreatic cancer-associated PanINs, may contribute to the predisposition of patients with chronic pancreatitis to develop pancreatic ductal adenocarcinoma.

Results of pancreaticoduodenectomy for lymphoplasmacytic sclerosing pancreatitis.

OBJECTIVE: To compare the presentation and short-term results of pancreaticoduodenectomy for lymphoplasmacytic sclerosing pancreatitis (LPSP) and pancreatic adenocarcinoma (PA) and to provide long-term follow-up on patients undergoing resection for LPSP. SUMMARY BACKGROUND DATA: LPSP is a rare form of chronic pancreatitis characterized by a mixed inflammatory infiltrate centered around pancreatic ducts and ductules, combined with obliterative phlebitis. Its presentation may mimic that of PA. METHODS: Among 1,648 patients undergoing pancreaticoduodenectomy at the authors' institution from January 1992 to May 2002, 37 with LPSP were identified. The demographics, clinical features, and short- and long-term outcomes of these patients were analyzed. Where applicable, comparisons were made to a consecutive group of 45 patients who underwent pancreaticoduodenectomy for PA from July 2001 to December 2001. RESULTS: Twenty-four percent (9/37) of the LPSP patients and none of the PA patients had a history of either atopic or autoimmune disease. The LPSP patients had a similar clinical presentation to the PA patients, but the LPSP patients were less likely to have a discrete pancreatic mass on CT and more likely to have CT findings suggesting a diffusely enlarged pancreas. The LPSP patients were also less likely to have a discrete pancreatic mass at operation, more likely to have a diffusely firm or hard gland, and more likely to have a difficult portal vein/superior mesenteric vein dissection when compared to the PA patients. Transfusion requirement, operative time, postoperative length of stay, and overall complication rate were similar between groups; however, the LPSP patients had a greater operative blood loss. The median length of follow-up for LPSP patients was 33 months. No patients had recurrent jaundice. One patient with LPSP had clinically evident recurrent pancreatitis. Among LPSP patients available for current telephone interview, 68% subjectively rated their quality of life as better, 18% reported no change, and 14% reported diminished quality of life compared to before surgery. CONCLUSION: LPSP mimics PA in clinical presentation, though CT findings of a diffusely enlarged pancreas without a discrete mass may suggest a diagnosis of LPSP. Nevertheless, differentiation from pancreatic neoplasia remains difficult. Patients undergoing pancreaticoduodenectomy for LPSP have durable relief of symptoms and a subjectively improved quality of life.

Pancreaticoduodenectomy: role of interventional radiologists in managing patients and complications.

Although the mortality rate after pancreaticoduodenectomy has decreased, the morbidity rate remains high. Major morbidity is often managed with the aid of interventional radiologists. The objective of this study was to evaluate the cooperative roles of interventional radiologists and pancreatic surgeons in complex pancreatic surgery, specifically pancreaticoduodenectomy. Our pancreaticoduodenectomy database was reviewed for all patients undergoing pancreaticoduodenectomy between January 1, 1995 and December 31, 2000. The interventional radiologic procedures for each patient were evaluated. A total of 1061 patients underwent pancreaticoduodenectomy. The overall mortality and morbidity rates were 2.3% and 35%, respectively. Five hundred ninety patients (56%) had no interventional radiologic procedures, whereas 471 patients (44%) had interventional radiologic procedures. Of those, 342 (32%) had preoperative biliary drainage (PBD) and 129 (12%) required postoperative interventional radiologic procedures. Percutaneous aspiration/catheter drainage was required in 84 patients for intra-abdominal abscess, biloma, or lymphocele, with 24 requiring two or more abscess drains. Thirty-nine patients underwent postoperative PBD for bile leaks due to anastomotic disruption, undrained biliary segments, or T-tube/bile stent dislodgment. Eighteen patients had hemobilia/gastrointestinal bleeding treated by angiography with embolization. The reoperation rate for the entire cohort of 1061 patients was 4.1% (n = 43). Nineteen of the 129 patients (15%) requiring postoperative radiologic intervention required reoperation. Although 4 of 18 patients who required embolization for bleeding subsequently required surgical intervention for the same reason, only 4 of 84 patients undergoing abscess drainage later required operation for anastomotic disruption or unsuccessful percutaneous drainage. As would be expected, the patients who required postoperative radiologic intervention (n = 129) had a higher incidence of postoperative complications including pancreatic fistula (20% vs. 6%, P < 0.01), bile leakage (22% vs. 1%, P < 0.01), and wound infection (16% vs. 8%, P < 0.01). With the complications in these 129 patients, the postoperative mortality rate was only 6.2% compared to 1.7% in patients who did not require radiologic intervention (n = 932, P < 0.01). The median postoperative length of stay was 15 days in those patients requiring postoperative radiologic intervention, 10 days in those not requiring intervention (P < 0.01; postoperative interventional radiology vs. no postoperative interventional radiology), and 29.5 days for patients needing reoperation. Interventional radiologists play a critical role in the management of some patients undergoing pancreaticoduodenectomy. Although complications such as anastomotic leaks, abscess formation, and bleeding can result in increased mortality and a longer hospital stay, the skills of the interventional radiology team provide expert management of some life-threatening complications, thus avoiding reoperation, speeding recovery times, and minimizing morbidity.

Standard vs. radical pancreaticoduodenectomy for periampullary adenocarcinoma: a prospective, randomized trial evaluating quality of life in pancreaticoduodenectomy survivors.

This study was designed to assess the health-related quality of life (QOL) of patients who had been randomly assigned to either standard or radical pancreaticoduodenectomy for periampullary adenocarcinoma. Pancreaticoduodenectomy has been performed in increasing numbers for periampullary adenocarcinoma. The appropriate extent of resection (standard vs. radical [extended]) remains controversial, particularly as concerns survival benefit. Past reports comparing standard vs. radical resection have suggested that the more extensive resection is attended by negative functional outcomes (diarrhea and weight loss) and poorer QOL, diminishing the impact of any possible survival advantage of the radical resection. A prospective, randomized single-institution trial comparing standard pancreaticoduodenectomy (pylorus preservation preferred) to radical pancreaticoduodenectomy (including distal gastrectomy and retroperitoneal lymphadenectomy) evaluated 299 patients with periampullary adenocarcinoma between April 1996 and June 2001. A standard Functional Assessment of Cancer Therapy-Hepatobiliary (FACT-Hep) QOL survey designed for hepatobiliary cancer was sent to 150 of these patients surviving pancreaticoduodenectomy. QOL and functional status were assessed via a series of subscale scores for physical, social, emotional, and functional well-being. A total of 105 QOL surveys (70%) were returned and analyzed, with 55 of the patients having been randomized to the standard group and 50 to the radical group. The patients were evaluated at a mean of 2.2 years after pancreaticoduodenectomy. The two groups were statistically similar with regard to multiple parameters including age at operation (64.6 years), race, intraoperative blood transfusions, pathologic diagnosis and staging, and perioperative complications. The radical group had a significantly higher percentage of men (66% vs. 44%; P = 0.02), a longer operative time (369 minutes vs. 327 minutes; P < 0.001), and a longer postoperative length of hospital stay (13.6 days vs. 10.1 days; P < 0.01). The FACT-Hep total QOL scores were similar between the standard and radical groups: 143.5 vs. 147.3, respectively. Additionally, the individual FACT-G subscale scores evaluating physical (22.1 vs. 23.3), social (24.5 vs. 24.4), emotional (19.2 vs. 19.6), and functional well-being (20.6 vs. 22.4) were comparable between the standard and radical groups. Subgroup analyses based on pathologic diagnosis (pancreatic, ampullary, distal bile duct, etc.) failed to reveal any differences in QOL assessment between the standard and radical pancreaticoduodenectomy groups. Finally, QOL measures were similar when comparing time since operation (<2 years' follow-up vs. >2 years' follow-up) and age (< or =65 years vs. >65 years). This is the largest report comparing QOL assessment in survivors of pancreaticoduodenectomy randomized between standard and radical resection. These data demonstrate no differences in long-term QOL between standard and radical resection. These results imply that no negative long-term QOL measures are associated with radical pancreaticoduodenectomy (as performed in this study) for periampullary adenocarcinoma.

Pancreaticoduodenectomy (Whipple resections) in patients without malignancy: are they all 'chronic pancreatitis'?

Pancreaticoduodenectomy (Whipple resection) has evolved into a safe procedure in major high-volume medical centers for the treatment of pancreatic adenocarcinoma and refractory chronic pancreatitis. However, some Whipple resections performed for a clinical suspicion of malignancy reveal only benign disease on pathologic examination. We evaluated the frequency of such Whipple resections without tumor in a large series of pancreaticoduodenectomies and classified the diverse pancreatic and biliary tract diseases present in these specimens. Of 442 Whipple resections performed during 1999-2001, 47 (10.6%) were negative for neoplastic disease and, in 40 cases, had been performed for a clinical suspicion of malignancy. Most Whipple resections revealed benign pancreatic disease, including 8 (17%) alcohol-associated chronic pancreatitis, 4 (8.5%) gallstone-associated pancreatitis, 1 (2.1%) pancreas divisum, 6 (12.8%) "ordinary" chronic pancreatitis of unknown etiology, and 11 (23.4%) lymphoplasmacytic sclerosing pancreatitis. In particular, patients with lymphoplasmacytic sclerosing pancreatitis were all thought to harbor malignancy, whereas only 13 of 19 (68.4%) of Whipple resections showing histologically "ordinary" forms of chronic pancreatitis were performed for a clinical suspicion of malignancy. Benign biliary tract disease, including three cases of primary sclerosing cholangitis, two cases of choledocholithiasis-associated chronic biliary tract disease, and four fibroinflammatory strictures isolated to the intrapancreatic common bile duct, was a common etiology for clinically suspicious Whipple resections (22.5% of cases). Pancreatic intraepithelial neoplasia (PanIN) was a common finding among all pancreata, whether involved by pancreatitis or histologically normal. Overall, PanIN 1A/1B was present in 68.1%, PanIN 2 in 40.4%, and PanIN 3 in just 2.1%. These findings indicate that "benign but clinically suspicious" Whipple resections are relatively common in high-volume centers (9.2%) and reveal a diverse group of clinicopathologically distinctive pancreatic and biliary tract disease.

Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma, part 2: randomized controlled trial evaluating survival, morbidity, and mortality.

OBJECTIVE: To evaluate, in a prospective, randomized single-institution trial, the end points of operative morbidity, operative mortality, and survival in patients undergoing standard versus radical (extended) pancreaticoduodenectomy. SUMMARY BACKGROUND DATA: Numerous retrospective reports and a few prospective randomized trials have suggested that the performance of an extended lymphadenectomy in association with a pancreaticoduodenal resection may improve survival for patients with pancreatic and other periampullary adenocarcinomas. METHODS: Between April 1996 and June 2001, 299 patients with periampullary adenocarcinoma were enrolled in a prospective, randomized single-institution trial. After intraoperative verification (by frozen section) of margin-negative resected periampullary adenocarcinoma, patients were randomized to either a standard pancreaticoduodenectomy (removing only the peripancreatic lymph nodes en bloc with the specimen) or a radical (extended) pancreaticoduodenectomy (standard resection plus distal gastrectomy and retroperitoneal lymphadenectomy). All pathology specimens were reviewed, fully categorized, and staged. The postoperative morbidity, mortality, and survival data were analyzed. RESULTS: Of the 299 patients randomized, 5 (1.7%) were subsequently excluded because their final pathology failed to reveal periampullary adenocarcinoma, leaving 294 patients for analysis (146 standard vs. 148 radical). The two groups were statistically similar with regard to age (median 67 years) and gender (54% male). All the patients in the radical group underwent distal gastric resection, while 86% of the patients in the standard group underwent pylorus preservation ( <.0001). The mean operative time in the radical group was 6.4 hours, compared to 5.9 hours in the standard group ( =.002). There were no significant differences between the two groups with respect to intraoperative blood loss, transfusion requirements (median zero units), location of primary tumor (57% pancreatic, 22% ampullary, 17% distal bile duct, 3% duodenal), mean tumor size (2.6 cm), positive lymph node status (74%), or positive margin status on final permanent section (10%). The mean total number of lymph nodes resected was significantly higher in the radical group. Of the 148 patients in the radical group, only 15% (n = 22) had metastatic adenocarcinoma in the resected retroperitoneal lymph nodes, and none had retroperitoneal nodes as the only site of lymph node involvement. One patient in the radical group with negative pancreaticoduodenectomy specimen lymph nodes had a micrometastasis to one perigastric lymph node. There were six perioperative deaths (4%) in the standard group versus three perioperative deaths (2%) in the radical group ( = NS). The overall complication rates were 29% for the standard group versus 43% for the radical group ( =.01), with patients in the radical group having significantly higher rates of early delayed gastric emptying and pancreatic fistula and a significantly longer mean postoperative stay. With a mean patient follow-up of 24 months, there were no significant differences in 1-, 3-, or 5-year and median survival when comparing the standard and radical groups. CONCLUSIONS: Radical (extended) pancreaticoduodenectomy can be performed with similar mortality but some increased morbidity compared to standard pancreaticoduodenectomy. The data to date fail to indicate that a survival benefit is derived from the addition of a distal gastrectomy and retroperitoneal lymphadenectomy to a pylorus-preserving pancreaticoduodenectomy.

Cyclooxygenase 2 expression in pancreatic adenocarcinoma and pancreatic intraepithelial neoplasia: an immunohistochemical analysis with automated cellular imaging.

We immunohistochemically examined material from 36 pancreata (adenocarcinomas, 30 lesions; pancreatic intraepithelial neoplasia [PanIN], 65; normal pancreatic ducts, 30) for cyclooxygenase 2 (COX-2) with an automated platform. We analyzed 7 to 10 discrete foci and generated an average percentage of positive cells and average staining intensity for each lesion. These 2 values were then multiplied to create an overall "HistoScore" for each lesion. COX-2 demonstrated considerable heterogeneity of expression between and within cases. The overall average percentage of positive cells in adenocarcinomas was 47.3%; in PanINs, 36.3%; and in normal ducts, 19.2%. COX-2 was expressed in more than 20% of cells in 23 adenocarcinomas (77%), 42 PanINs (65%), and 12 normal ducts (40%). The overall average HistoScore for adenocarcinomas was 6.1; for PanINs, 5.4; and for normal ducts, 3.5. Significant differences in COX-2 expression were demonstrable in adenocarcinomas vs normal ducts, PanINs vs normal ducts, and PanIN 2/3 vs PanIN 1a/1b. In general, the pattern of COX-2 expression increased from normal to PanIN to adenocarcinoma. The up-regulation of COX-2 in a subset of noninvasive precursor lesions makes it a potential target for chemoprevention with selective COX-2 inhibitors.

Immunohistochemical validation of a novel epithelial and a novel stromal marker of pancreatic ductal adenocarcinoma identified by global expression microarrays: sea urchin fascin homolog and heat shock protein 47.

We extended the results of a previous microarray analysis by immunohistochemical validation of differential protein expression in a series of 57 surgically resected infiltrating ductal pancreatic adenocarcinomas. Two representative genes were examined: sea urchin fascin homolog (overexpressed in both cell lines and primary tumors) and heat shock protein 47 (HSP47; overexpressed in primary tumors only). Protein expression also was evaluated in the precursor lesions of pancreatic cancer pancreatic intraepithelial neoplasia (PanIN), and normal ductal epithelium. Fascin expression was seen in the neoplastic cells of 54 (95%) of 57 ductal adenocarcinomas but not in 49 (94%) of 52 adjacent nonneoplastic epithelium. In the multistep pathogenesis of ductal adenocarcinomas, fascin expression seemed to be a late event, usually present in PanINs 2 and 3. HSP47 expression was almost universal and most intense in the ductal adenocarcinoma-associated stromal desmoplasia (57/57), although 37 cases (65%) also expressed HSP47 in the neoplastic epithelium. HSP47 expression was absent in the majority of nonneoplastic pancreata (46 [88%]). Fascin and HSP47 are novel tumor markers with potential diagnostic and therapeutic implications for pancreatic carcinoma. These results establish the usefulness of global expression platforms to identify novel tumor markers.

High-throughput measurement of the Tp53 response to anticancer drugs and random compounds using a stably integrated Tp53-responsive luciferase reporter.

Human Tp53 is normally a short-lived protein. Tp53 protein is stabilized and levels are increased in response to a variety of cellular stresses, including those induced by genotoxic anticancer drugs and environmental exposures. To engineer an efficient assay based on this property, we constructed and integrated a Tp53-specific reporter system into human cancer cells, termed p53R cells. We tested a range of conventional chemotherapeutic agents as well as over 16 000 diverse small compounds. Ionizing radiation and two-thirds of conventional chemotherapeutic agents, but only 0.2% of diverse compounds activated Tp53 activity by two-fold or greater, consistent with the presumptive genotoxic activation of Tp53 function. Cytotoxicity was independent of TP53 genetic status when paired, syngeneic wild-type TP53 and TP53-null cells in culture were treated with compounds that activated Tp53. From the unbiased survey of random compounds, Tp53 activation was strongly induced by an analog of AMSA, an investigational anti-cancer agent. Tp53 was also strongly induced by an N-oxide of quinoline and by dabequine, an experimental antimalarial evaluated in humans; dabequine was reported to be negative in other screens of mutagenicity and clastogenicity but carcinogenic in animal studies. Further exploration of antimalarial compounds identified the common medicinals chloroquine, quinacrine, and amodiaquine as Tp53-inducers. Flavonoids are known to have DNA topoisomerase activity, a Tp53-inducing activity that is confirmed in the assay. A reported clinical association of Tp53 immunopositive colorectal cancers with use of the antihypertensive agents was extended by the demonstration of hydralazine and nifedipine as Tp53-inducers. p53R cells represent an efficient Tp53 functional assay to identify chemicals and other agents with interesting biologic properties, including genotoxicity. This assay may have utility in the identification of novel chemotherapeutic agents, as an adjunct in the pharmaceutical optimization of lead compounds, in the exploration of environmental exposures, and in chemical probing of the Tp53 pathway.

Diagnostic laparoscopy for periampullary and pancreatic cancer: what is the true benefit?

The role of diagnostic laparoscopy in patients with periampullary and pancreatic malignancies is controversial. A retrospective review was performed including all patients (n = 188) with a periampullary or pancreatic malignancy who underwent both CT and laparotomy at our institution between January 1997 and December 1999. The overall resectability rate for all periampullary cancers was 67.3% (115 of 171 patients). This compared favorably with the resectability rate for cancers of the pancreatic body and tail (3 of 17 patients, 17.6%; P < 0.01 vs. periampullary cancers). Fifty percent of patients with periampullary cancers were unresectable because of metastatic disease, whereas metastatic disease precluded resection in 64.3% of patients with cancers of the pancreatic body and tail. After patients undergoing operative palliation were eliminated, a nontherapeutic laparotomy would have been precluded by the use of diagnostic laparoscopy in only 2.3% of patients with periampullary cancers (4 of 171 patients). In contrast, 6 (35.3%) of 17 patients with cancers of the pancreatic body and tail underwent a nontherapeutic laparotomy (P < 0.01 vs. periampullary cancers). One hundred fifty-eight (84%) of the 188 CT reports reviewed could be definitively categorized as either "likely to be resectable" or "likely to be unresectable." The remaining 16% were equivocal. Of the 107 patients categorized as likely to be resectable, 89 were actually resected (83.2%). In contrast, only 10 of the 51 patients categorized as likely to be unresectable could be resected (19.6%).

Almost all infiltrating colloid carcinomas of the pancreas and periampullary region arise from in situ papillary neoplasms: a study of 39 cases.

Colloid carcinomas of organs such as the breast, colon, and prostate have been well characterized. However, up until now there have been only a few studies of colloid carcinomas of the pancreas and periampullary region, and the number of colloid carcinomas in these studies has been limited. A search of our files revealed 39 resections for pancreatic and periampullary carcinomas with colloid differentiation. All neoplasms were extensively sampled. "Carcinomas with colloid differentiation" were defined as tumors associated with abundant extracellular mucin containing free-floating mucinous epithelial cells. Cases with >50% colloid differentiation were classified as "colloid carcinomas," whereas those with less were termed "carcinomas with focal colloid features." Cases with no colloid differentiation at all were designated "carcinomas without colloid differentiation." Of the 39 carcinomas, 31 were colloid carcinomas, and eight were carcinomas with focal colloid features. Twenty-seven were centered in the pancreas, seven were in the duodenum, and five were in the ampulla of Vater. Remarkably, 38 of the 39 carcinomas (97%) arose in association with an intraductal papillary mucinous neoplasm or a tubular/tubulovillous adenoma. Of the patients with colloid carcinomas, the 2-and 5-year actuarial survival rates were 69% and 29%, respectively. There was no significant difference in survival rates between patients with colloid carcinomas and patients with adenocarcinomas without colloid differentiation, whether or not the latter arose in association with intraductal papillary mucinous neoplasms or tubular/tubulovillous adenomas. In a multivariate model colloid differentiation was not an independent predictor of patient survival, while other factors such as tumor location, perineural invasion, vascular invasion, and margin status after resection independently influenced patient survival. Most colloid carcinomas of the pancreas and periampullary region arise in association with a well-defined in situ papillary neoplasm. The diagnosis of a pancreatic or periampullary colloid carcinoma should encourage the pathologist to search for an associated low-grade in situ component. In addition, colloid carcinomas of the pancreas and periampullary region do not necessarily have a better prognosis than carcinomas without colloid differentiation. Instead, other factors such as tumor location, perineural invasion, vascular invasion, and margin status after resection are far more important.