Pancreatic Cystic Neoplasms - CBC...Pancreatic Cystic Neoplasms Jennifer E. Verbesey, MDa,*, J....

Pancreatic CysticNeoplasms

Jennifer E. Verbesey, MDa,*, J. Lawrence Munson, MDb

KEYWORDS

� Pancreatic cystic neoplasms � Serous cystadenoma� Mucinous cystic neoplasms� Intraductal papillary mucinous neoplasm

Cystic neoplasms of the pancreas have been recognized for almost two centuries. In1830, Becourt’s first description noted a tumor ‘‘the size of a child’s head andcomposed of very strong fibrous walls.’’1 Soon after, Gross published the first reportof a cystic neoplasm in the United States. It took 50 years for Bozeman to understandin 1872 that aspiration alone would not cure this type of tumor, and he accomplishedthe first successful resection of a tumor with survival of the patient. In 1900, ReginaldFitz acknowledged the malignant potential of such tumors. Much progress in terms ofthe understanding of etiology, diagnosis, treatment, and prognosis has been accom-plished since then. In 1978, Compagno and Oertel2 divided cystic tumors into 2 types:benign serous neoplasms and mucinous cystic tumors that harbored a malignancy. In1982, Ohashi described what one would now call an intraductal papillary mucinousneoplasm (IPMN), which he initially described as a ‘‘mucin-producing cancer.’’3

Over 70% of cystic neoplasms are found in asymptomatic patients, usually as anincidental finding on scans obtained for another reason. In an autopsy study publishedin 1995, the investigators found that almost half of the 300 deceased patients studiedhad small cystic lesions.4 Older age was directly related to the probability of findinga lesion. Most published studies agree that the ratio of benign to malignant lesionsis approximately 2:1. Cystic neoplasms are found in females at a ratio of 2:1 to 3:1compared with males. Ferrone and colleagues5 looked at 401 patients with pancreaticcystic lesions and found that 61% were women with a median age of 62 years.

Given the malignant potential of cystic neoplasms, it is important that this entity bedifferentiated from pancreatic pseudocysts. Postinflammatory cystic collections,especially pseudocysts, represent the vast majority of cystic lesions (as much as90%). Pancreatic pseudocysts lack an epithelial lining and will, many times, followan illness, particularly a bout of pancreatitis. Radiologic imaging plays a large role in

a Department of Transplantation, Lahey Clinic Medical Center, 41 Mall Road, Burlington, MA01805, USAb Department of General Surgery, Lahey Clinic Medical Center, 41 Mall Road, Burlington, MA01805, USA* Corresponding author.E-mail address: [email protected]

Surg Clin N Am 90 (2010) 411–425doi:10.1016/j.suc.2009.12.006 surgical.theclinics.com0039-6109/10/$ – see front matter ª 2010 Elsevier Inc. All rights reserved.

mailto:[email protected]://surgical.theclinics.com

Verbesey & Munson412

diagnosis. Evidence of chronic pancreatitis make a pseudocyst more likely. On theother hand, calcification in the wall of the cyst, septations, or solid components canpoint one more in the direction of a cystic neoplasm.6 Communication with the pancre-atic duct can be seen in both pseudocysts and some forms of mucinous cystic lesions,so this does not confirm a diagnosis. Abdominal and endoscopic ultrasound hasproven very helpful in differentiation between these 2 entities. The type of cyst wallseen, presence of a multilocular versus unilocular cyst, appearance of septations orsolid components, and simultaneous lymphadenopathy can all help make an accuratediagnosis.7,8 A recent study showed that presence of internal dependent debris wasa highly specific magnetic resonance imaging (MRI) finding indicating the diagnosisof pseudocyst.9 In cases that remain difficult to diagnose despite radiologic studies,cyst fluid analysis can help. Analysis includes cyst epithelial cell staining, stains formucin, amylase levels, and tumor markers carcinoembryonic antigen (CEA) andcarbohydrate antigen 19-9 (CA19-9).6 The positive predictive value of such tests isvery high, but negative results are unreliable and cannot be used to rule out malig-nancy. In a minority of cases, differentiating between a pseudocyst and pancreaticcystic neoplasm proves impossible before going to the operating room. However, ingeneral a combination of the patient’s history, radiologic studies, and laboratory testswill allow the physician to feel confident in choosing the differential diagnosis.

WORLD HEALTH ORGANIZATION CLASSIFICATION OF CYSTIC NEOPLASMS

The World Health Organization (WHO) first published a classification system for cysticneoplasms in 1996, and a revision in 2000. This classification system divided pancre-atic cystic neoplasms into 3 categories: malignant tumors (in situ or invasive), border-line (uncertain malignant potential), and benign (adenomas).10,11 Microcystic,glycogen-rich, serous tumors are almost universally benign, and macrocystic,mucinous tumors accepted as either malignant or premalignant.

Malignant tumors include serous cystadenocarcinomas, mucinous cystadenocarci-nomas, intraductal papillary mucinous carcinomas, invasive papillary mucinous carci-nomas, and acinar cell cystadenocarcinomas. Borderline tumors consist of mucinouscystic tumors with moderate dysplasia, IPMN with moderate dysplasia, and solidpseudopapillary tumors. Benign or adenomatous tumors are serous cystadenomas,mucinous cystadenomas, and certain IPMNs.

SEROUS TUMORS: SEROUS CYSTADENOMA AND SEROUSCYSTADENOCARCINOMAS

Serous tumors represent approximately 1% to 2% of pancreatic neoplasms andalmost 25% of all cystic tumors.12 Serous neoplasms are distributed evenlythroughout the pancreas. These tumors are almost universally found incidentallywhen a computed tomography (CT) scan is obtained for another reason. The tumorscause symptoms only if they are very large (>15 cm), usually compressive symptomsfrom a mass effect. Although extremely rare, the most common symptoms would beeither jaundice or gastric outlet obstruction, caused by extrinsic pressure on adjacentstructures. Serous tumors tend to be found more often in older women (average age61 years). A mean diameter of 5 to 8 cm is most common.

Serous neoplasms will stain negative for mucin. It is generally believed that theycarry no risk of invasive cancer; however, there are several case reports in the litera-ture describing serous cystadenomas that presented with metastases, now morecorrectly termed a serous cystadenocarcinoma.13–16 Serous neoplasms haveglycogen-rich clear cytoplasm that stains positive with a PAS test (periodic acid-Schiff

Pancreatic Cystic Neoplasms 413

stain) and uniform, cuboidal epithelium. Serous neoplasms are microcystic and canhave thin-walled septae that produce the typical ‘‘honeycomb appearance.’’ On CTscan, these neoplasms will have a thin wall and a typical sunburst pattern of calcifica-tion with a central, late-enhancing scar (Figs. 1 and 2). The tumor will appear grosslyto be grapelike, with 6 or more uniformly sized cysts 2 cm or smaller. Small septae andinternal debris may be seen in the individual cysts (Figs. 3–5). The capsule is usuallypoorly developed, and there is often a poor distinction between tumor andsurrounding pancreatic parenchyma. Serous neoplasms never communicate withthe main pancreatic duct.

MUCINOUS CYSTIC TUMORS: MUCINOUS CYSTADENOMAS AND MUCINOUSCYSTADENOCARCINOMAS

Mucinous tumors represent approximately 2% of all pancreatic neoplasms and one-third of all cystic neoplasms.12 Mucinous tumors are almost all found in the body or tailof the pancreas, most frequently in women in their fifth or sixth decade of life.Mucinous tumors range in size from 6 to 35 cm. The contents of the tumor are usuallythick, but may be hemorrhagic, watery, or necrotic.17 The tumors are circumscribedunilocular or multilocular cysts with no communication with the pancreatic duct(with rare exceptions) (Fig. 6). CT scan will show a cyst with a thick wall and possiblecalcification within the wall (Fig. 7). Mucinous neoplasms rarely, if every, recur aftercomplete resection.18

Mucinous neoplasms are thought to behave in a way similar to other cancers, in thatthere seems to a progression from atypia to frank malignancy. Malignancy risk isdirectly related to an increase in size and duration of existence. Approximately 65%of the epithelium is composed of mucin-producing cells, and it is in these cells thatmalignant deterioration occurs. In a study conducted at Massachusetts GeneralHospital, 64% of all mucinous tumors found proved to be malignant, 33% had metas-tases, and only 63% were resectable. Of those resected 76% were cured, which rep-resented 48% of the total. All of the patients who were unresectable died of thedisease.19

Ovarian-like stroma that usually stains positive histologically for estrogen andprogesterone receptors is found.20 It is suspected that mucinous neoplasms may arise

Fig. 1. Contrast-enhanced CT scan demonstrating a large cystic mass in the tail of thepancreas consistent with a serous cystadenoma.

Fig. 2. Contrast-enhanced CT scan showing 4 axial sections of cystic mass in body and tail ofpancreas. Pathology confirmed the presence of a serous cystadenoma.


from ovarian rests in the pancreas. Tanaka and colleagues11 evaluated the hypothesisthat ovarian-type stroma is a histologic requirement for diagnosing a mucinous cysticneoplasm (MCN). Their hypothesis was confirmed when they found that women olderthan 60 years and also men had ovarian-type stroma, and they concluded thatovarian-type stroma is a prerequisite for the diagnosis of a MCN.

Mucinous ductal ectasia is a premalignant finding. Pathology demonstrates papil-lary hyperplasia and mucin overproduction along the pancreatic duct. The pancreaticduct can become obstructed by filling with mucin, and can lead to pancreatitis. Attimes, mucus can be seen emanating from the ampulla.19

Fig. 3. Resected pancreatic head and duodenum (Whipple procedure) for serouscystadenoma.

Fig. 4. Freshly removed and opened pancreatic serous cystadenoma. Septations and ‘‘honey-combing’’ evident.


INTRADUCTAL PAPILLARY MUCINOUS NEOPLASMS

The WHO defined an IPMN as a papillary mucin-producing neoplasm that arises in themain pancreatic duct or major branches. This entity was first described by Ohashi andcolleagues3 in 1982. IPMNs represent approximately 1% of all pancreatic neoplasmsand 25% of cystic neoplasms.12 IPMNs have direct communication with either themain pancreatic duct or the smaller branch ducts, and are most commonly found inthe head or proximal region of the pancreas (Fig. 8). It was traditionally thought thatIPMNs were more common in men older than 70 years.21 However, recent studiesby Salvia, Sohn, D’Angelica and colleagues22–24 showed a more equal distributionbetween sexes. IPMNs are divided into main duct IPMNs or branch duct IPMNs;mixed type (which is present approximately 40% of the time) are considered mainduct for the purpose of prognosis and treatment. Branch duct tumors are morecommonly found in younger patients and have a lower malignant potential. Noninva-sive IPMNs are divided into low-grade dysplasia (adenoma), moderate dysplasia(borderline IPMN), and high-grade dysplasia or carcinoma in situ. Malignancy is foundin a reported 58% to 92% of main duct tumors. Malignancy is much less common inbranch duct tumors, with a reported presence of 6% to 46%.25 Fujino and

Fig. 5. Pathologic specimen after resection of pancreatic serous cystadenoma. Microcysticnature, ‘‘grapelike’’ form, and small septations can be seen.

Fig. 6. Resected pancreatic mucinous cystic neoplasms. Both of these tumors were multilo-culated, macrocystic tumors.


colleagues26 found main duct type tumors in 55% of their patients, and 8.6% wereadenomas whereas 91.4% were carcinomas. Consistent with past studies, 45%were branch type tumors, of which 62% were adenomas and only 38% carcinomas.Woo and colleagues25 studied 190 patients and found malignancy was more commonin diabetic patients, and showed an even lower prevalence of invasive cancer inbranch duct tumors (6.3%).

Multiple studies have shown an average lag time of approximately 5 years from ageat presentation of IPMN with low-grade dysplasia to the point at which it becomes aninvasive carcinoma. In their published study, Sohn and colleagues23 showed that theaverage age of patients with low-grade dysplasia was 63.2 years and that of thosewith invasive cancer was 68.2 years. In addition, several studies have showed thatpatients with IPMNs have a higher risk of synchronous or metachronous primary ex-trapancreatic tumors, including 25% of patients that had cancers of the stomach,colon, rectum, lung, breast, or liver.27,28

On histologic examination, IPMNs are grossly visible. Tall, columnar, mucin-con-taining epithelium with cytoarchitectural atypia is found. There are 2 subtypes: intes-tinal subtype, which exhibits abundant extracellular mucin and typically has a highsurvival rate after resection, and pancreatobiliary subtype, which is the form that typi-cally evolves into ductal adenocarcinoma and leads to a very poor survival rate. A thirdsubtype can be found in branch type tumors that is described as a gastric foveolarsubtype, which rarely progresses to malignancy.17 The spectrum of atypia ranges

Fig. 7. CT scans from 2 different patients showing mucinous cystic neoplasms in the tail andbody of the pancreas.

Fig. 8. Contrast-enhanced CT scans of pancreatic IPMNs. Patients can often have multipleIPMNs along the length of the pancreas.


from hyperplasia (adenoma) to low-grade dysplasia to invasive carcinoma. Mucinousductal ectasia is sometimes present. Mucin accumulation leads to ductal dilatation.

A summary of common characteristics of pancreatic pseudocysts, serous neo-plasms, MCNs, and IPMNs is found in Table 1.

CLINICAL PRESENTATION

In most cases of pancreatic cystic neoplasm, symptoms are due to mass effect of thetumor on adjacent structures. Looking at all patients with a cystic neoplasm, mosthave a history of no trauma or alcoholic pancreatitis. The vast majority feel, at somepoint, epigastric fullness or ache. Fifty percent will have palpable mass. A patientrarely will complain of weight loss, steatorrhea, and back pain, or show signs of jaun-dice. Serous neoplasms are most commonly found incidentally, so are therefore mostoften asymptomatic.

With MCNs, some patients may complain of vague abdominal pain or discomfort.Weight loss or anorexia is associated with malignancy. In contrast to most serousand mucinous lesions, IPMNs are most often discovered in association with symp-toms (75%). The clinical scenario most often mimics that of pancreatitis, with vagueepigastric pain and occasional steatorrhea. For the most part, patients have no historyof trauma or alcohol abuse. Laboratory values are normal outside of an acute attack ofpancreatitis. Jaundice is usually only present if invasive cancer exists.

DIAGNOSIS

Multiple radiologic modalities can be used to make the diagnosis of a pancreaticcystic neoplasm including ultrasound, CT, MRI/magnetic resonance cholangiopan-creatography (MRCP), endoscopic retrograde cholangiopancreatography (ERCP),arteriography, and endoscopic ultrasound. CT scanning is widely accepted as thebest initial diagnostic tool. CT can be used to evaluate the important qualities of thecystic lesion, such as size, multi- or unilocular composition, solid components, andpresence of calcifications. In addition, it can be used to evaluate the relationship ofthe mass to the celiac axis, superior mesenteric artery, superior mesenteric vein,and portal vein, and assess the resectability of the tumor. Septations can be visualizedin mucinous neoplasms, ‘‘sunburst’’ calcifications in serous cystadenomas, andperipheral calcification in cystadenocarcinomas. Finally, CT can be used to differen-tiate main duct from branch duct type IPMNs. Nodules in the main duct system or dila-tation of the main duct to greater than 1 cm is suggestive of a main duct IPMN,

Table 1Comparison of common characteristics of various pancreatic cystic lesions

Pseudocyst Serous Neoplasm Mucinous Cystic Neoplasm IPMN

Demographics Any age FemalesAverage age 61

Females 20:1Average age 45

Males or 1:1Average age 50–60s

Symptoms Early satietyPain

Asymptomatic unless verylarge

Usually asymptomatic Occasional pain

Location Throughout More often in body/tail Body/tail (>90%) Head (2/3)

CT Findings Thick Wall Thin wallSunburst/honeycomb

Thick wall Cystic 1/� solidcomponents

ERCP/MRCP 1/� Communication withpancreatic duct

No communication withpancreatic duct

No communication withpancreatic duct

Connection with main orbranch pancreatic ducts

Aspiration quality Thin, dark, opaque,nonmucinous

Serous fluidNo mucin

Clear, thick, mucin-richfluid

Mucinous fluid

Tumor markers in cyst fluid Not elevated Not elevated Possibly elevated Possibly elevated

Cancer risk None None 6%–36% Main Duct 70%Branch Duct 25%

Gross appearance Orangelike Grapelike

Surgery If symptomatic or large If >4 cm or symptoms Resect all Resect all main ductneoplasms

Resect branch duct if largeor suspicious findings(see text)

Recurrence risk innoninvasive tumors

Zero Near zero Continued risk in remainingpancreas

Recurrence risk if invasive Significant risk Significant risk

Abbreviations: CT, computed tomography; ERCP, endoscopic retrograde cholangiopancreatography; IPMN, intraductal papillary mucinous neoplasm; MRCP,magnetic resonance cholangiopancreatography.

Verb

ese

y&

Mu

nso

n418


whereas a mucinous cyst that communicates with the ductal system in the context ofa main duct with a normal diameter is probably a branch duct IPMN.

MRI/MRCP is a useful adjunct in that it can identify the cyst and show its relationshipto the ductal anatomy (Fig. 9), which is particularly useful in identifying IPMNs andvisualizing small connections to the ductal system. MRCP may also be a good wayto follow IPMNs postoperatively. ERCP is the true gold standard for evaluating pancre-atic ductal anatomy and communication between cyst and duct, and in addition thereis the added advantage of being able to obtain cytology (Fig. 10). ERCP is the mostdefinitive test for IMPNs in that intraductal mucin or mucin leakage from a patulousor dilated ampulla is pathognomonic for an IPMN.25 However, of importance is thatonce the tumor has become invasive, mucin can block the communication so thatmost IPMNs will not have mucin emanating from the papilla. A positive finding ofmucin at the ampulla is highly sensitive for the presence of an IPMN; however, a nega-tive finding in no way rules out its presence. ERCP can also be helpful in revealingductal stricturing, displacement, or obstruction.

Least invasive of all tests is an abdominal ultrasound that can be used to evaluateseptations and debris in the cyst. In addition, mural nodularity and vascular flowcan be evaluated. Endoscopic ultrasound (EUS) is increasingly viewed as the bestmodality to stage tumors for resection, identify nodal involvement, and show presenceor absence of vascular invasion; it is also one of the best ways to obtain cytologicsampling. EUS can be used to follow up on vague findings on other studies or tohelp differentiate benign from malignant. Kubo and colleagues29 found that a dilatedmain pancreatic duct, branch type tumors that are greater than 30 mm with irregularsepta, or mural nodules indicate malignancy. Multiple other studies have demon-strated that predictors of malignancy include Type III or IV mural nodules or irregular-ities of the pancreatic duct.30,31

Arteriography is rarely used at the present time during the workup of cystic lesions.In the past, if performed, one could find a hypervascular blush with serous cysts anddraping or displacement of vessels by mucinous tumors.

CYTOLOGY

CEA is a glycoprotein found in embryonic endodermal epithelium. Mucinous cysts arelined by endoderm-derived columnar epithelium that secretes CEA. Nonmucinouscysts are lined by simple cuboidal epithelium that should not have any CEA.32 This

Fig. 9. MRI showing presence of IPMN in body of pancreas.

Fig. 10. ERCP of IPMN. Communication with the main pancreatic duct is visualized.


situation makes CEA, along with other tumor markers such as CA19-9 and CA-125,helpful in diagnosis. Pseudocysts will often have an elevated amylase and lipase,but negligible tumor markers. Fluid aspirated from MCNs will often exhibit low amylaseand lipase levels, but high CEA, CA19-9, and CA-125 levels. On the other hand, IPMNsshow variable levels of amylase and lipase but also higher levels of CEA and CA19-9.However, because these values are often not consistent, the true benefit of fine-nee-dle aspiration is controversial.33 Sawhney and colleagues32 showed very poor agree-ment between CEA levels and molecular analysis (DNA quantity, k-ras 2 pointmutation or 2 or more allelic imbalance mutations, and pathologic analysis of cyst)for diagnosis of mucinous neoplasms. Their study showed a sensitivity of 82% forCEA and 77% for molecular analysis, but 100% combined sensitivity.

Cytologic contribution to diagnosis and prognosis should certainly be a major focusof future research. Recent studies such as the one published by Fritz and colleagues33

showed how using array comparative genomic hybridization (CGH) to look for recur-rent mutations in IPMNs to help differentiate these from pancreatic adenocarcinomascan be helpful with diagnosis and prognosis. Future studies will need to focus on theusefulness of molecular tumor markers and mutations in the care of these patients.

PREDICTORS OF MALIGNANCY

One would optimally wish to be able to identify tumors with an extremely low chance ofmalignancy so that those patients can pursue more conservative or observationaltherapy and avoid aggressive surgical intervention. Given the vastly improving abilityof radiologic examinations to pick up even the smallest abnormality, and theincreasing number of incidental findings, it is of paramount importance that there becriteria to determine which lesions are most likely to be malignant.

A large number of studies have been published that detail numerous factors foundto be independent predictors of malignancy. In general, irregularity of borders, mixedsolid and cystic components, or symptomatic lesions are all ominous signs. Nearly alltumors with radiologic calcification are malignant. Many MCNs that are malignantdemonstrate peripheral calcification, a thickened cyst wall, papillary proliferations,and a hypervascular pattern, and may show vascular involvement.17 In addition, Gar-cea and colleagues demonstrated that the earliest change in malignant cells ofa mucinous neoplasm is a mutation of the Kras2 oncogene on chromosome 12p.


This mutation is found in 89% of MCNs with carcinoma in situ (compared with 20% ofadenomas).17

Multiple studies have confirmed predictors of malignancy for IPMNs. These factorsinclude the type of IPMN (main duct type has higher malignant potential), larger tumorsize (tumor diameter >30 mm), proximal location, involvement of a dilated mainpancreatic duct larger than 7 mm, presence of mural nodules, protruding lesions indilated branch ducts, thick cyst wall, a patulous papilla with mucin leakage from theampulla of Vater, and increased CA19-9 level. Other independent risk factors are olderage, presence of jaundice, diabetes, and pancreatitis.17,22,25,26,30,34 In partial conflictwith other reports, Fujino and colleagues26 showed that the following factors were notsignificant predictors of malignancy: gender, location, mural nodules, abdominal pain,the presence of other tumors, history of pancreatitis, cyst size, serum CEA, and serumamylase. A low level of CA19-9 does not distinguish between invasive and noninvasiveIPMNs. IPMN malignancies have very different biologic behavior to invasive ductaladenocarcinomas. IPMNs tend to grow less aggressively, have a lower incidence ofnodal positivity, and have lower rates of perineural and vascular invasion.18,23

TREATMENT

There are certain basic principles for the surgical treatment of cystic neoplasms. Whilenot all tumors will be amenable to resection, surgery can be curative for many. Growthby displacement, which is usually the case, instead of invasion into adjacent struc-tures, favors resection. A distal pancreatectomy is the traditional treatment for bodyand tail lesions. A Whipple procedure (pancreaticoduodenectomy) is usually neededfor head lesions. Total pancreatectomy can be considered for IMPNs with diffusedysplasia. Central pancreatectomy can be performed for body lesions, particularlyat the neck, with pancreaticoenteric reconstruction. A variety of limited pancreatec-tomies have been proposed, but there are no long-term data verifying the efficacyof these procedures. In past years, it was recommended that all cystic neoplasmsbe resected due to the difficulty in predicting which lesions were malignant. At present,given the large number of tumors that are incidental findings and studies that havedemonstrated that many tumors have a very low malignant potential, other treatmentstrategies need to be considered at times.

SEROUS NEOPLASMS

Surgical treatment is not indicated for serous neoplasms unless the patient hasobstructive symptoms or symptoms from local compression of surrounding struc-tures. Acceptable growth of serous neoplasms is approximately 0.6 cm per year.Larger tumors that are more than 4 cm in size may increase as much as 2 cm peryear and start to cause local symptoms. Therefore, in excellent surgical candidateswithout significant comorbidities, surgical resection is a reasonable option for tumorsthat are 4 cm or greater.12

MUCINOUS NEOPLASMS

All MCNs must be considered premalignant and can undergo malignant transforma-tion at any time. Their progression toward malignancy is thought to mirror that ofpancreatic intraepithelial neoplasia, albeit at a much more indolent rate. Therefore,the general recommendation is to resect all mucinous tumors, given the patient isan acceptable surgical risk. Because more than 90% of mucinous neoplasms arefound in the body or tail of the pancreas, a distal pancreatectomy is the usual


treatment, with intraoperative frozen section of the pancreatic margin. Cameron andcolleagues34 report that a laparoscopic pancreatectomy is appropriate if there isa low risk of invasive cancer, the cyst is small (


recurrence and requires no follow-up. A resected, malignant MCN has a much higherrisk of recurrence and should therefore be followed every 6 months using CT or MRI.

Resected, benign IPMNs carry a small risk of recurrence and should be followedyearly with CT or MRI. This follow-up interval can be spaced out to longer periods ifnothing is found over a time span of several years. On the other hand, resected, malig-nant IPMNs carry a very significant risk of recurrence and need to be reevaluated every6 months with radiologic imaging. CEA and CA19-9 levels have had no proven value inthe follow-up of IPMNs.

Branch duct IPMNs that have not been resected should be followed very closely.For lesions between 10 and 20 mm, CT or MRI should be obtained every 6 to12 months. For cystic lesions larger than 20 mm, radiologic imaging should be evalu-ated every 3 to 6 months.11 In addition, on each occasion patients should be reas-sessed for any symptoms that can be attributed to the tumor.

Given the high rate of synchronous extrapancreatic tumors in patients with IPMNs,care should be taken to screen these patients for other malignant neoplasms on a peri-odic basis.

PROGNOSIS

The prognosis for cystic pancreatic neoplasms is much better than ductal pancreaticadenocarcinoma, for which 5-year survival rates are 20% to 25% in the most opti-mistic studies. Serous neoplasms are not malignant and therefore have a 100% 5-year survival rate. Benign MCNs and benign IPMNs have a 95% to 100% 5-yearsurvival rate, with benign MCNs having a recurrence risk of near zero. Multiple inves-tigators have agreed that survival rate for a malignant IPMN is between 60% and70%.10,22,34,37–42 The worst prognosis is for malignant MCNs, which have a 5-yearpredicted survival rate of 50% to 60%.43,44

ACKNOWLEDGMENTS

All CT and MRI scans are courtesy of Dr Francis Scholz, Lahey Clinic.

REFERENCES

1. Becourt PJ. Recherches sur le pancreas: ses foncions et ses alterations organi-ques. Strasbourg, France: Levrault; 1830.

2. Compagno J, Oertel JE. Mucinous cystic neoplasms of the pancreas with overtand latent malignancy (cystadenocarcinoma and cystadenoma). A clinicopatho-logic study of 41 cases. Am J Clin Pathol 1980;74(1):1–11.

3. Ohashi K, Mirukami Y, Muruyama M, et al. Four cases of mucus secretingpancreas cancer. Prog Dig Endosc 1982;20:348–51.

4. Kimura W, Nagai H, Kuroda A, et al. Analysis of small cystic lesions of thepancreas. Int J Pancreatol 1995;18:197–206.

5. Ferrone CR, Correa-Gallego C, Warshaw AL, et al. Current trends in pancreaticcystic neoplasms. Arch Surg 2009;144(5):448–54.

6. Sand J, Nordback I. The differentiation between pancreatic neoplastic cysts andpancreatic pseudocyst. Scand J Surg 2005;94:161–4.

7. Hernandez LV, Mishra G, Forsmark C, et al. Role of endoscopic ultrasound (EUS)and EUS-guided fine needle aspiration in the diagnosis and treatment of cysticlesions of the pancreas. Pancreas 2002;25:222–8.


8. Ahmad NA, Kochman ML, Lewis JD, et al. Can EUS alone differentiate betweenmalignant and benign cystic lesion of the pancreas? Am J Gastroenterol 2001;96:3295–300.

9. Macari M, Finn ME, Bennett GL, et al. Differentiating pancreatic cystic neoplasmsfrom pancreatic pseudocysts at MR imaging: value of perceived internal debris.Radiology 2009;251:77–84.

10. Kloppel G, Solcia E, Longnecker DS, et al. Histological typing of tumors of theexocrine pancreas. In: World Health Organization. International histological clas-sification of tumors. 2nd edition. Berlin: Springer; 1996. p. 15–21.

11. Tanaka M, Chari S, Adsay V, et al. International consensus guidelines for manage-ment of intraductal papillary mucinous neoplasms and mucinous cysticneoplasms of the pancreas. Pancreatology 2006;6:17–32.

12. Phillips K, Fleming JB, Tamm EP, et al. Unusual pancreatic tumors. In:Cameron JL, editor. Current surgical therapy. 9th edition. Philadelphia: MobyElsevier; 2008. p. 524–9.

13. Klug JC, Ng TT, White SC, et al. Pancreatic serous cystadenocarcinoma: a casereport and review of the literature. J Gastrointest Surg 2009;13:1864–8.

14. George DH, Murphy F, Michalski R, et al. Serous cystadenocarcinoma of thepancreas: a new entity? Am J Surg Pathol 1989;13:61–6.

15. Kamei K, Funabiki T, Ochiai M, et al. Multifocal pancreatic serous cystadenomawith atypical cells and focal perineural invasion. Int J Pancreatol 1991;10:161–72.

16. Eriguchi N, Aoyagi S, Nakayama T, et al. Serous cystadenocarcinoma of thepancreas with liver metastases. J Hepatobiliary Pancreat Surg 1998;383:56–61.

17. Garcea G, Ong SL, Rajesh A, et al. Cystic lesions of the pancreas: a diagnosticand management dilemma. Pancreatology 2008;8:236–51.

18. Fritz S, Warshaw AL, Thayer S. Management of mucin-producing cysticneoplasms of the pancreas. Oncologist 2009;14(2):125–36.

19. Warshaw AL, Compton CC, Lewandrowski K, et al. Cystic tumors of thepancreas—new clinical, radiologic, and pathologic observations in 67 patients.Ann Surg 1990;212(4):432–43.

20. Volkan Adsay N. Cystic lesions of the pancreas. Mod Pathol 2007;20(Suppl 1):S71–93.

21. Woodside KJ, Riall TS. Intraductal papillary mucinous neoplasms of thepancreas. In: Cameron JL, editor. Current surgical therapy. 9th edition. Philadel-phia: Moby Elsevier; 2008. p. 530–4.

22. Salvia R, Fernandez-del Castillo C, Bassi C, et al. Main-duct intraductal papillarymucinous neoplasms of the pancreas: clinical predictors of malignancy andlong-term survival following resection. Ann Surg 2004;239:678–85 [discussion:685–7].

23. Sohn TA, Yeo CH, Cameron JL, et al. Intraductal papillary mucinous neoplasms ofthe pancreas: an updated experience. Ann Surg 2004;239:788–97 [discussion:797–9].

24. D’Angelica M, Brennan MF, Suriawinata AA, et al. Intraductal papillary mucinousneoplasms of the pancreas: an analysis of clinicopahtologic features andoutcome. Ann Surg 2004;239:400–8.

25. Woo SM, Ryu JK, Lee SH, et al. Branch duct intraductal papillary mucinousneoplasms in a retrospective series of 190 patients. Br J Surg 2009;96:405–11.

26. Fujino Y, Matsumoto I, Ueda T, et al. Proposed new score predicting malignancyof intraductal papillary mucinous neoplasms of the pancreas. Am J Surg 2007;194:304–7.


27. Yamaguchi K, Yokohata K, Noshiro H, et al. Mucinous cystic neoplasm of thepancreas or intraductal papillary-mucinous tumour of the pancreas. Eur J Surg2000;166:141–8.

28. Eguchi H, Ishikawa O, Ohigashi H, et al. Patients with pancreatic intraductalpapillary mucinous neoplasms are at high risk of colorectal cancer development.Surgery 2006;139:749–54.

29. Kubo H, Chijiiwa Y, Akahoshi K, et al. Intraductal papillary-mucinous tumors of thepancreas: differential diagnosis between benign and malignant tumors by endo-scopic ultrasonography. Am J Gastroenterol 2001;96:1429–34.

30. Sugiyama M, Atomi Y. Intraductal papillary mucinous tumors of the pancreas:imaging studies and treatment strategies. Ann Surg 1998;228:684–91.

31. Ohno E, Hirooka Y, Itoh A, et al. Intraductal papillary mucinous neoplasms of thepancreas: differentiation of malignant and benign tumors by endoscopic ultra-sound findings of mural nodules. Ann Surg 2009;249:628–34.

32. Sawhney MS, Devarajan S, O’Farrel P, et al. Comparison of carcinoembryonicantigen and molecular analysis in pancreatic cyst fluid. Gastrointest Endosc2009;69:1106–10.

33. Fritz S, Fernandez-del Castillo C, Mino-Kenudson M, et al. Global genomic anal-ysis of intraductal papillary mucinous neoplasms of the pancreas reveals signif-icant molecular differences compared with ductal adenocarcinoma. Ann Surg2009;249:440–7.

34. Cameron JL, Riall TS, Coleman J, et al. One thousand consecutive pancreatico-duodenectomies. Ann Surg 2006;244:10–5.

35. Jang JY, Kim SW, Ahn YJ, et al. Multicenter analysis of clinicopathologic featuresof intraductal papillary mucinous tumor of the pancreas: is it possible to predictthe malignancy before surgery? Ann Surg Oncol 2005;12:124–32.

36. Maire F, Hammel P, Terris B, et al. Prognosis of malignant intraductal papillarymucinous tumours of the pancreas after surgical resection. Comparison withpancreatic ductal adenocarcinoma. Gut 2002;51:717–22.

37. Yamao K, Nakamura T, Suzuki T, et al. Endoscopic diagnosis and staging ofmucinous cystic neoplasms and intraductal papillary-mucinous tumors. J Hepa-tobiliary Pancreat Surg 2003;10:142–6.

38. Brugge WR, Lewandrowski K, Lee-Lewandrowski E, et al. Diagnosis of pancre-atic cystic neoplasms: a report of the cooperative pancreatic cyst study. Gastro-enterology 2004;126:1330–6.

39. Allen PJ, Jaques DP, D’Angelica M, et al. Cystic lesions of the pancreas: selectioncriteria for operative and nonoperative management in 109 patients. J Gastroint-est Surg 2003;7:970–7.

40. Izumo A,Yamaguchi K,Eguchi T, et al. Mucinous cystic tumor of the pancreas: immu-nohistochemical assessment of ‘‘ovarian-type stroma’’. Oncol Rep 2003;10:515–25.

41. Sarr MG, Carpenter HA, Praghaka LP, et al. Clinical and pathologic correlation of 84mucinous cystic neoplasms of the pancreas: can one reliably differentiate benignfrom malignant (or premalignant) neoplasms? Ann Surg 2000;231:205–12.

42. Wilentz RE, Albores-Saavedra J, Zahurak M, et al. Pathologic examination accu-rately predicts prognosis in mucinous cystic neoplasms of the pancreas. AmJ Surg Pathol 1999;23:1320–7.

43. Hruban RH, Takaori K, Klimstra DS, et al. An illustrated consensus on the classi-fication of pancreatic intraepithelial neoplasia and intraductal papillary mucinousneoplasms. Am J Surg Pathol 2004;28:977–87.

44. Wilentz RE, Albores-Saavedra J, Hruban RH. Mucinous cystic neoplasms of thepancreas. Semin Diagn Pathol 2000;17:31–42.

Pancreatic Cystic NeoplasmsWorld Health Organization classification of cystic neoplasmsSerous tumors: serous cystadenoma and serous cystadenocarcinomasMucinous cystic tumors: mucinous cystadenomas and mucinous cystadenocarcinomasIntraductal papillary mucinous neoplasmsClinical presentationDiagnosisCytologyPredictors of malignancyTreatmentSerous NeoplasmsMucinous NeoplasmsIntraductal Papillary Mucinous NeoplasmsFrozen sectionsFollow-upPrognosisAcknowledgmentsReferences

Pancreatic Cystic Neoplasms - CBC...Pancreatic Cystic Neoplasms Jennifer E. Verbesey, MDa,*, J....

Documents

Transcript of Pancreatic Cystic Neoplasms - CBC...Pancreatic Cystic Neoplasms Jennifer E. Verbesey, MDa,*, J....