Miscellaneous Nonpancreatic

Nonendocrine Tumors

Heather A. Lillemoe, John D. Abad, and Keith D. Lillemoe

Contents
Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Pathology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Ampullary Neoplasms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Distal Common Bile Duct Neoplasms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
Duodenal Neoplasms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
Rare Periampullary Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Clinical Presentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
Diagnostic Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Laboratory Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Imaging Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Preoperative Staging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Surgical Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
Endoscopic Resection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
Local Excision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Pancreaticoduodenectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Segmental Resection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Palliative Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
Adjuvant and Neoadjuvant Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
Survival . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

H. A. Lillemoe (*)
Vanderbilt University Medical Center, Nashville, TN, USA
e-mail: [email protected]
J. D. Abad
Feinberg School of Medicine, Northwestern University, Warrenville, IL, USA
e-mail: [email protected]
K. D. Lillemoe
Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
e-mail: [email protected]

# Springer Science+Business Media, LLC, part of Springer Nature 2018 283

J. P. Neoptolemos et al. (eds.), Pancreatic Cancer,
https://doi.org/10.1007/978-1-4939-7193-0_10
284 H. A. Lillemoe et al.

Key Practice Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308

Future Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
Cross-References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

Abstract
Nonendocrine, nonpancreatic periampullary tumors are generally classified as
arising from the ampulla of Vater, distal common bile duct, or duodenum. The
most common clinical finding on presentation is obstructive jaundice. These lesions
may occur spontaneously or as part of a hereditary syndrome (familial adenoma-
tous polyposis, Gardner’s syndrome, and inflammatory bowel disease). The most
effective diagnostic strategies for determining extent of disease and resectability of
periampullary tumors include dual-phase computed tomography and endoscopic
ultrasound. Small, benign periampullary lesions may be amenable to endoscopic
resection. For benign lesions <3 cm that are unable to be completely removed
endoscopically, transduodenal local resection should be considered. Appropriate
surgical candidates with larger lesions >3 cm or suspicion of invasive carcinoma
should undergo a pancreaticoduodenectomy. Five-year survival for duodenal,
ampullary, and distal common bile duct carcinomas are 51–59, 37–39, and
23–27%, respectively. For each of these tumors, both lymph node status and
negative margins are significant predictors of outcome. At this point, neoadjuvant
and adjuvant therapies have not clearly demonstrated a survival benefit for non-
pancreatic periampullary cancers. The future success in treating these cancers likely
rests in the development of novel biological and targeted therapies in the setting of
well-designed multi-institutional clinical trials. This chapter will focus on benign
and malignant nonpancreatic and nonneuroendocrine periampullary tumors and
will include the pathology, clinical presentation, diagnostic workup, and manage-
ment strategies to approach these neoplasms.

Keywords
Cholangiocarcinoma · Bile duct cancer · Ampullary adenoma · Ampullary
cancer · Duodenal neoplams

Epidemiology

The majority of periampullary tumors are malignant, with pancreatic adenocarci-

noma being the most common followed by cancers of the ampulla of Vater, distal
common bile duct, and duodenum, respectively. Periampullary adenocarcinoma has
a yearly incidence in the United States of approximately 35,000 cases which has
remained stable over the last few decades [1]. Pancreatic adenocarcinoma likely
accounts for up to 90% of these cases, although without surgical resection and
pathologic analysis, the specific organ of origin can be difficult to determine. The
relative frequency of malignant periampullary neoplasms in resected specimens is
Miscellaneous Nonpancreatic Nonendocrine Tumors 285

Table 1 Relative frequency periampullary neoplasms in resected specimens

Location Percentage (%)
Head of pancreas 56
Ampulla of Vater 21
Distal common bile duct 17
Duodenum 3

shown in Table 1 [2]. The percentage of nonpancreatic malignancies is likely higher

in this surgical resection series given such tumors have a higher rate of resection
compared to primary pancreatic cancers.

Pathology

Ampullary Neoplasms

Tumors of the ampulla can be either benign or malignant, both of which are rare.
Autopsy studies demonstrate that the overall prevalence of ampullary adenomas is
approximately 0.04–0.12% [3]. Among all malignancies of the gastrointestinal tract,
ampullary neoplasms account for only 0.5% [4].

Benign Ampullary Tumors

Benign adenomas are generally defined as adenomatous lesions arising on or within
2 cm of the ampulla of Vater. They are classified by their microscopic findings as
either intestinal type or biliary type. Ampullary adenomas are seen both sporadically
and in association with familial syndromes such as familial adenomatous polyposis
(FAP). The incidence of sporadic adenomas appears to be increasing, which is likely
the direct result of increased detection due to the increased utilization of upper
endoscopy. Sporadic ampullary adenomas usually occur during the sixth decade of
life and are an average diameter of 2 cm [5, 6] (Fig. 1).
Ampullary adenomas are frequently identified in patients with FAP, who have a
cumulative lifetime risk near 100% [3]. The median age at presentation for familial
adenomas is earlier than the sporadic cases, presenting at 30–40 years of age. The
diagnosis of periampullary adenomas associated with FAP usually occurs well after
the diagnosis of colonic polyps typically at a mean follow-up of 17 years after
colectomy [7]. At presentation, these lesions can often be multiple and involve both
the ampulla and duodenal mucosal surface simultaneously.
Similar to the well-defined transformation of colonic adenomas into adenocarci-
noma, ampullary adenomas have the potential for malignant degeneration. Patients
with FAP have a 100–200-fold higher risk of periampullary cancer compared to the
general population and a prevalence of ampullary cancer of 3–12% [4]. Thus, close
screening and follow-up is extremely important in this population.
286 H. A. Lillemoe et al.

Fig. 1 Endoscopic
appearance of benign villous
adenoma

Fig. 2 Surgical specimen

demonstrating an ulcerated
ampullary carcinoma. The
papilla is replaced by an
exophytic papillary and
ulcerated tumor (Reprinted
from Mino and Lauwers [8])

Malignant Ampullary Tumors

Ampullary carcinoma is classified as four types based on macroscopic features:
intra-ampullary (24%), periampullary duodenal (6%), mixed exophytic (31%), and
mixed ulcerated (39%) (Fig. 2). Overall, intra-ampullary cancers have a better
prognosis than other subtypes as these tumors usually present smaller lesions with
less angiolymphatic invasion, fewer lymph node metastases, and less direct invasion
of the pancreas.
Adenocarcinomas of the ampulla are further divided into intestinal-type (50%)
and pancreaticobiliary type (20%) based on histologic features The most prevalent
type of ampullary adenocarcinomas are intestinal-type which resemble primary
adenocarcinomas of the colon pathologically with simple or cribriform glands
lined by atypical cells with features of intraluminal necrosis and inflammation
(Fig. 3). Pancreaticobiliary-type resembles primary pancreatic and biliary
Miscellaneous Nonpancreatic Nonendocrine Tumors 287

Fig. 3 Adenocarcinoma,
intestinal type. The tumor is
composed of complexed
glands lined by atypical cells.
Note the typical luminal
inflammation (Reprinted from
Mino and Lauwers [8])

Fig. 4 Adenocarcinoma,
pancreatobiliary type. The
tumor is composed of simple
malignant glands lined by low
columnar cells. Note the
markedly atypical nuclei and
the surrounding desmoplasia
(Reprinted from Mino and
Lauwers [8])

adenocarcinomas. These tumors are composed of simple glands lined with low
columnar cells with features of atypical nuclei and surrounding desmoplastic stroma
(Fig. 4). Compared to the intestinal type, the pancreaticobiliary type more often
demonstrates perineural invasion, but angiolymphatic invasion is less common. In
instances where both microscopic features of intestinal and pancreaticobiliary are
present, these tumors are classified as intestinal, unless there is a predominant
pancreaticobiliary phenotype. A 2008 series from the University of California San
Francisco of 118 patients with ampullary adenocarcinomas noted patients with
pancreaticobiliary type presented with jaundice more frequently and had signifi-
cantly worse survival compared to those with intestinal type [9].
There are several unusual subtypes of ampullary cancers including papillary,
mucinous, and signet-ring carcinomas. Papillary carcinomas are uncommon and
are reported in 6% of ampullary carcinomas. They are classified as either invasive
or noninvasive. Invasive papillary carcinomas appear as complex branching papil-
lary structures with fibrovascular cores and/or micropapillary structures without
fibrovascular cores. These are lined by either intestinal or pancreatobiliary-type
cells. In contrast, noninvasive papillary carcinomas are exophytic tumors arising in
288 H. A. Lillemoe et al.

the intra-ampullary mucosa and lined by pancreatobiliary-type epithelium. The

neoplasms are similar to noninvasive papillary carcinomas of the extrahepatic
bile ducts and noninvasive intraductal papillary mucinous neoplasms of the
pancreas.
Mucinous or colloid carcinomas represent only 4–7% of ampullary carcinomas.
These neoplasms demonstrate two particular morphologies both with greater than
50% containing extracellular mucin. These carcinomas are composed of columnar
epithelium with nuclear atypia or contain clusters of neoplastic cells.
Signet-ring cell carcinomas of the ampulla are extremely rare. These neoplasms
contain cells with nuclei forced to the periphery by intracytoplasmic mucin. In order
to diagnose these tumors, greater than 50% of the tumors must contain signet-ring
cells with a diffuse growth pattern, and a primary from another site must be
excluded.

Distal Common Bile Duct Neoplasms

The common bile duct (CBD) is divided into four parts: (1) supraduodenal,
(2) retroduodenal, (3) intrapancreatic, and (4) intraduodenal. The periampullary
distal CBD is considered to include the intrapancreatic and intraduodenal segments.
Tumors of epithelial, nonepithelial, and mesenchymal origin can arise from the
distal CBD.

Benign Distal Bile Duct Tumors

Adenomas are extremely rare lesions of the distal common bile duct and are less
common than carcinomas. These lesions are usually small, often single, and may
appear as pedunculated or sessile polyps. They are histologically classified similarly
to adenomas of the colon: tubular, tubulovillous, and villous. Reports of distal
common bile duct adenomas have been reported in familial adenomatous polyposis
and Gardner’s syndrome [10].
Cystadenomas are mucinous cystic tumors that can arise from various structures
in the upper gastrointestinal tract, most commonly seen in the liver, pancreas, and
extrahepatic bile ducts. These tumors occur in the biliary tree of middle-aged females
and may grow as large as 20 cm. Malignant transformation is rare, although
dysplasia is seen in 13% of these tumors [11]. Complete local excision for symp-
tomatic lesions is necessary due to a high rate of recurrence if incompletely resected.
Cystadenomas can be differentiated from intraductal papillary mucinous neoplasms
(IPMN) of the biliary tract by the presence of mesenchymal stroma in the former.
Biliary IPMN, also seen originating from the pancreas, has emerged as a unique
entity and may account for >7% of biliary neoplasms. These lesions are considered
precursors to cholangiocarcinoma with risk of malignant transformation [12].
Biliary papillomatosis is a rare phenomenon of multicentric complex papillary
neoplasms which involve the extra- and intrahepatic biliary systems and gallbladder,
and may extend into the pancreatic ducts. It affects both males and females equally
Miscellaneous Nonpancreatic Nonendocrine Tumors 289

during the sixth decade of life. Surgical resection is difficult and recurrence is
common. The treatment of choice is total hepatectomy and liver transplantation.
Granular cell tumors are neoplasms of the extrahepatic biliary system usually
involving the common bile duct. These tumors typically occur in young women
(median age 34 years). Patients typically present with jaundice and abdominal pain.
Granular cell tumors are occasionally multicentric with lesions in the gallbladder,
skin, omentum, esophagus, and stomach. Within the common bile duct, these lesions
appear as small (<2 cm), firm, submucosal nodules that invade the lumen. These
tumors are not malignant, however may invade into periductal tissue and adjacent
pancreas. Diagnosis usually occurs by ultrasound with subsequent MRCP/ERCP.
These lesions are clinically similar to malignant distal common bile duct tumors and
often require operative resection to make the diagnosis.

Malignant Distal Bile Duct Tumors

The incidence of extrahepatic cholangiocarcinoma in the United States is low, with
approximately 6,000 new cases diagnosed annually [13]. Using the classification
system proposed by Nakeeb and colleagues, cholangiocarcinoma lesions are divided
into intrahepatic, perihilar, and distal subgroups [14]. Up to 45% of these malignan-
cies are classified as extrahepatic [13]. Although the etiology is unknown, there are
several well-documented risk factors. The incidence of all types of cholangio-
carcinoma increases with age and is higher in males. Patients with ulcerative colitis
and sclerosing cholangitis have a significantly increased risk of developing both
intra- and extrahepatic cholangiocarcinoma at 65%, which is 4x higher than that of
the general population. Although patients with Crohn’s disease are at an increased
risk for developing cholangiocarcinoma, the risk is approximately half of that for
ulcerative colitis patients [15]. Biliary pathology such as cholangitis, choledocho-
lithiasis, cholecystitis, and choledochal cysts are also independent risk factors for
development of cholangiocarcinoma. Hepatolithiasis and biliary parasitic infestation
(Clonorchis sinensis or Opisthorchis viverrini), both prevalent in parts of Asia, also
increase the risk of cholangiocarcinoma.
Adenocarcinoma is the primary histologic subtype in the distal common bile duct
malignancies. The three macroscopic classifications of cholangiocarcinoma are
sclerosing, nodular, and papillary. Sclerosing lesions are the most common and
appear as thickening of the bile duct with diffuse infiltration of adjacent tissues.
Nodular tumors are irregular nodules that invade into the lumen of the bile duct.
Nodular-sclerosing lesions, as implied, have characteristics of both. The papillary
subtype represents only 10% of cholangiocarcinomas and is more common in the
distal bile duct than the hepatic bifurcation [16]. These tumors are soft polypoid
lesions with little or no invasive component and generally have a more favorable
prognosis compared with the sclerosing subtype [17, 18]. These tumors spread
longitudinally along the duct wall beneath the epithelial lining. As a result, preop-
erative imaging and intraoperative examination may not appreciate the extent of
submucosal spread, highlighting the importance of intraoperative frozen section to
determine adequate margins for resection (Table 2).
290 H. A. Lillemoe et al.

Table 2 Tumor classification table

Miscellaneous nonpancreatic nonneuroendocrine tumors
Periampullary tumor classification
1. Ampullary neoplasms 3. Duodenal neoplasms
A. Benign adenomas A. Benign adenomas (tubular, villous, Brunner gland)
B. Adenocarcinomas (intestinal B. Lipomas
type, pancreaticobiliary type)
C. Papillary carcinoma (invasive, C. Hamartomas
noninvasive)
D. Mucinous or colloid carcinomas D. Hemangiomas
E. Signet-ring carcinomas E. Primary duodenal adenocarcinomas
2. Distal common bile duct 4. Mesenchymal neoplasms
neoplasms
A. Benign adenomas A. Leiomyomas, lipomas
B. Cystadenomas B. Neurogenic tumors (neurofibromas, ganglioneuromas)
C. Biliary papillomatosis C. Vascular tumors (hemangiomas, lymphangiomas)
D. Granular cell tumors D. Granular cell tumors
E. Cholangiocarcinoma (sclerosing, E. Schwann cell tumors
nodular, papillary)
F. Gastrointestinal stromal tumors (GIST)
5. Lymphomas (B cell lymphomas)
6. Metastatic tumors (renal cell carcinoma, melanoma,
breast cancer, squamous cell carcinoma, endometrioid
adenocarcinoma, osteosarcoma)
7. Pseudotumors (myoepithelial hamartoma, Brunner
gland hyperplasia)

Duodenal Neoplasms

Benign Duodenal Tumors

Small bowel tumors are rare and represent only 1–1.5% of all gastrointestinal
neoplasms. Depending on the series, the proportion of benign small bowel tumors
ranges from 14% to 52% [19]. Familial syndromes such as Gardner’s syndrome and
familial adenomatous polyposis are often associated with duodenal adenomas.
Adenomas are comprised of three types: (1) tubular, (2) villous, and (3) Brunner
gland. Tubular adenomas are usually pedunculated and generally have low risk for
invasive carcinoma. Villous adenomas have a higher malignant potential, especially
when greater than 2 cm. Brunner gland adenomas originate from hyperplastic
exocrine glands in the proximal duodenum and carry no malignant risk.
Lipomas are rare tumors of the duodenum and are usually identified as incidental
findings on CT as circumscribed tumors of fat density in the bowel wall. If symp-
tomatic, they present as bleeding or obstruction. If small (<2 cm) and asymptomatic,
they do not require resection. However, symptomatic, large or increasing size on
serial CT requires endoscopic or segmental resection to rule out the possibility of
liposarcoma.
Miscellaneous Nonpancreatic Nonendocrine Tumors 291

Hamartomas are lesions seen almost exclusively in Peutz-Jeghers syndrome, an

autosomal dominant condition characterized by multiple GI hamartomas throughout
the bowel with mucocutaneous pigmentation. Rarely, these tumors cause obstruction
or bleeding. Malignant transformation is rare but requires that these patients have
close surveillance. Surgical intervention should be considered for symptomatic
lesions or concern for the development of malignancy.
Hemangiomas are rare congenital lesions that present as acute or chronic bleeding
during midlife. They are usually single and have no malignant potential. If these
tumors are symptomatic, treatment consists of endoscopic or segmental resection.
Additional treatment modalities including endoscopic sclerotherapy or angiographic
embolization have also been described.

Malignant Duodenal Tumors

The incidence of small bowel cancer in the United States is approximately 10,000
cases per year with approximately 1300 deaths per year as a result [20]. The majority
of small bowel adenocarcinomas arise in the duodenum and up to half of primary
duodenal adenocarcinomas occur in the periampullary region [21, 22]. The inci-
dence is higher in older patients and males more than females. Most cancers of the
duodenum are sporadic. Familial adenomatous polyposis is the most prominent
genetic predisposing factor with a relative risk of over 300 times that of the normal
population. Hereditary nonpolyposis colorectal cancer, celiac sprue, and Crohn’s
disease are also associated with duodenal cancer.
Most of these tumors are solitary, sessile lesions, which often appear in associ-
ation with adenomas. They are usually moderately well-differentiated. These lesions
are similar to the malignant transformation of adenocarcinomas found in the colon
with similar pathologic features.

Rare Periampullary Tumors

Mesenchymal Neoplasms
Benign and malignant periampullary mesenchymal tumors are extremely uncom-
mon. The most common benign neoplasms are leiomyomas or lipomas. Other rare
benign lesions consist of neurogenic tumors (neurofibromas, ganglioneuroma),
vascular tumors (hemangiomas, lymphangioma), or granular cell tumors of Schwann
cell origin. Neurogenic tumors involving the ampulla may arise in patients with
neurofibromatosis.
Malignant mesenchymal tumors mostly consist of gastrointestinal stromal tumors
(GISTs). Duodenal and periampullary stromal tumors compose about 3–5% of all GI
stromal tumors [23]. The sub-proliferation in the majority of gastrointestinal stromal
tumors is thought to be driven by gain-of-function mutations of the KIT gene, which
encodes a type of tyrosine kinase receptor. Activating mutations of KIT can be found
in most periampullary stromal tumors. These tumors can occur at any age and
usually present with gastrointestinal bleeding associated with a growth of a large
size with central necrosis. Complete surgical excision is the treatment of choice.
292 H. A. Lillemoe et al.

Because lymph node metastasis is rare, local resection can be employed selectively.
Larger tumors, however, may require pancreaticoduodenectomy.

Lymphomas, Metastatic Tumors, and Pseudotumors

Other rare periampullary tumors include lymphomas and metastatic tumors. Most
reports of lymphoma involving the ampulla of Vater involve high-grade B cell
lymphoma and marginal zone B cell lymphoma. Metastatic disease involving the
periampullary region is often from direct extension from an adjacent locally
advanced tumor. Hematogenous spread from a primary neoplasm is extremely rare
but most commonly reported with renal cell carcinoma. Other malignant tumors
reported to metastasize to the periampullary region include melanoma, breast cancer,
squamous cell carcinoma of the larynx, endometrioid adenocarcinoma, and osteo-
sarcoma. Pseudotumors are recognized as 23% of tumors identified in the ampullary
region [24]. They include myoepithelial hamartoma and Brunner gland hyperplasia,
which collectively are more common than adenomas. It can be challenging to discern
pseudotumors from neoplastic lesions, and often result to unnecessary surgery.

Clinical Presentation

Benign periampullary and duodenal adenomas are often asymptomatic and discov-
ered incidentally or during surveillance for familial syndromes. Presenting symp-
toms depend on location and size of the tumors but can include jaundice, bleeding,
and obstruction such as are seen with malignant periampullary tumors.
Generally, periampullary and pancreatic carcinomas are difficult to diagnose in
their early stages. Symptoms tend to be nonspecific and often the diagnosis is not
made until patients develop jaundice. Compared to pancreatic primaries however,
tumors of the ampulla of Vater, distal common bile duct, and periampullary duode-
num tend to present at an earlier stage due to higher propensity for biliary obstruction
leading to jaundice. The mean diameter in one series of 149 patients diagnosed with
ampullary cancer (2.7 cm) was significantly smaller, compared to pancreatic head
cancer (3.5 cm) [25]. This generally translates to higher resectability rates than
pancreatic cancers. Usually jaundice is progressive and relentless and may be
associated with significant pruritus. Occasionally however, ampullary carcinomas
may present with intermittent jaundice due to the “ball valve” effect of a polypoid
tumor or necrosis during the growth phase leading to extrahepatic biliary obstruc-
tion. The development of jaundice is more commonly associated with a peri-
ampullary carcinoma (70%) than a benign tumor (20–30%) [26–30].
Periampullary neoplasms may also present with abdominal pain, anorexia, nau-
sea, weight loss, and gastrointestinal bleeding. Partial biliary or pancreatic duct
obstruction may result in complaints of abdominal pain prior to the development
of jaundice. This pain is usually dull, moderate intensity, located in either the
epigastrium or right upper quadrant, possibly radiating to the back, and aggravated
by eating. Vomiting secondary to duodenal obstruction is usually a late manifestation
of periampullary cancers in general, but may occur earlier in bulky duodenal cancers.
Miscellaneous Nonpancreatic Nonendocrine Tumors 293

Ampullary or duodenal cancers may present with chronic or intermittent gastroin-

testinal bleeding. An episode of acute pancreatitis of unclear etiology should raise
suspicion for an underlying periampullary neoplasm and initiate a thorough evalu-
ation once the acute episode has resolved. In a report by Rattner et al., acute
pancreatitis was the presenting symptom in 25% of patients diagnosed with ampul-
lary neoplasms [31].
Duodenal adenocarcinomas not immediately adjacent to the ampulla of Vater may
present with vague complaints of abdominal pain, weight loss, symptoms of bowel
obstruction, or bleeding. These lesions tend to represent more advanced disease than
periampullary adenocarcinomas.
Past medical and family history may be significant in evaluating a patient for a
possible periampullary neoplasm. Patients with Gardner’s syndrome and familial
polyposis may carry a 200-fold increased risk for ampullary and duodenal carcino-
mas [32]. These patients will often have multiple polyps involving a significant
portion of the duodenal mucosa.
Aside from jaundice, physical examination findings are commonly absent in
patients with periampullary tumors. Hepatomegaly may be present and usually
reflects hepatic congestion from biliary obstruction, not necessarily the presence of
metastatic disease. Ascites, however, may represent advanced disease. A palpable
gallbladder may be present in approximately 25% of patients. Occult fecal blood
may be seen in those with bleeding periampullary cancers as well.

Diagnostic Evaluation

Laboratory Data

Nearly all patients with periampullary cancers present with abnormal liver function
tests which includes increased plasma bilirubin and alkaline phosphatase, character-
istic of extrahepatic obstruction. Transaminase levels may also be increased but
usually not as significantly as alkaline phosphatase levels. In cases of longstanding
extrahepatic obstruction, the prothrombin time may be prolonged. Anemia may be
present with any periampullary cancers arising from the ampulla or duodenum
secondary to gastrointestinal bleeding. Tumor markers, such as CEA and CA19-9,
are generally not of diagnostic value as they are not specific for malignancy and may
be elevated in benign causes of extrahepatic obstruction.

Imaging Studies

Early diagnosis of periampullary cancers is dependent on prompt evaluation of the

jaundiced patient. Current imaging modalities provide detailed information regard-
ing the level and etiology of biliary obstruction. Once these lesions are identified, a
focused surgical approach gives these patients the best chance for long-term
survival.
294 H. A. Lillemoe et al.

Ultrasonography
Transabdominal ultrasound (US) is often used in the initial evaluation of patients
presenting with abdominal pain or obstructive jaundice, as it documents the presence
of biliary obstruction with a dilated biliary tree and can define the level of biliary
obstruction thereby narrowing the differential diagnosis. Other important findings
that can be visualized with US include gallstones, ascites, and liver metastases. A
major limitation of US is the frequent inability to identify a periampullary tumor and
the high rates technically inadequate studies, which can result from patient body
habitus, the presence of intervening bowel gas, or technical limitations of the
operator. Conversely, the lack of radiation exposure and its relatively low cost are
some of the advantages offered by US.

Computed Tomography
Despite the advantages of US, the high accuracy and reproducibility of computed
tomography (CT) and its widespread availability, make it the most useful, and often
the most cost-effective test in the initial evaluation of a patient with a suspected
periampullary malignancy [33]. CT can detect the presence of a periampullary mass
of at least 2 cm in size and also provides important information about the level of
biliary obstruction with respect to the pancreatic parenchyma, if no mass is seen
(Fig. 5). Pancreatic duct dilatation may also be seen. The optimal technique for
evaluation of the periampullary region involves administration of both intravenous
and oral contrast and obtaining 1- to 3-mm slices within a single breathhold during
both the arterial and portal venous phase of intravenous contrast enhancement [34,
35]. Scans obtained during the rapid intravenous injection of an iodinated contrast
agent result in an increase in the pancreatic parenchymal attenuation, as well as
excellent contrast enhancement of the major peripancreatic blood vessels. This
technique not only results in clear delineation of the tumor but may also demonstrate
involvement of adjacent major visceral vessels, such as the portal/superior

Fig. 5 Computed tomography scan of a patient with obstructive jaundice due to ampullary
carcinoma: (a) Scan demonstrated a 3-cm ampullary mass (black arrow) and (b) scan at higher
level demonstrating bile duct dilation within pancreatic parenchyma indicating distal duct obstruc-
tion (white arrow)
Miscellaneous Nonpancreatic Nonendocrine Tumors 295

mesenteric vein complex or superior mesenteric or hepatic arteries, suggesting

unresectability. The value of CT lies in the virtual absence of technically unsatis-
factory examinations and in its high accuracy in both the detection and staging of
periampullary carcinoma. The positive predictive value associated with
CT-determination of unresectability is greater than 90% [34]. Magnetic resonance
imaging (MRI) is equivalent to, but not superior to, CT for detection and staging of
periampullary tumors and has a higher cost [36]. However, it does offer the advan-
tages of avoiding exposure to radiation or ionic contrast and so is a more suitable test
for patients with contrast allergies or renal insufficiency.

Magnetic Resonance Cholangiopancreatography

Magnetic resonance cholangiopancreatography (MRCP) has emerged as a noninva-
sive method to determine the most likely etiology of a pancreaticobiliary abnormal-
ity. It is most helpful in evaluating abnormalities of the proximal bile ducts and liver.
In periampullary lesions, the thick slab MR images will delineate the biliary and
pancreatic ductal anatomy with detail that is similar to the more invasive techniques
of endoscopic retrograde cholangiopancreatography (ERCP). The other MR
sequences will define the presence or absence of a mass, the level of the obstruction
and the location of any given abnormality relative to the regional vessels.
The pattern on cholangiopancreatography can be characteristic for ampullary, bile
duct, and pancreatic carcinomas. Cancers of the ampulla or duodenum will obstruct
both the pancreatic and bile duct at the ampulla whereas pancreatic cancer will show
the classic “double duct” sign. Distal bile duct cancers show a characteristic “apple
core” appearance, with a normal appearing pancreatic duct.

Endoscopy/Endoscopic Ultrasound
Simple upper endoscopy can define the extent, size, and gross appearance of a
periampullary lesion suspected of being malignant and allows for simultaneous
performance of an endoscopic biopsy and cytologic brushings. The endoscopic
appearance of an ampullary lesion, however, is often similar for benign and malig-
nant tumors. Furthermore, endoscopic biopsies can reveal false negative results due
to sampling error, with accuracy rates ranging from 62% to 79% in various series
[37–40]. The demonstration of malignancy on biopsy specimens is definitive and
will in most cases indicate the need for pancreaticoduodenectomy. However, a
diagnosis of a benign adenoma does not rule out the presence of an adenocarcinoma
elsewhere in the adenoma. Finally, an important consideration is that ampullary
adenomas are a premalignant condition since they tend to progress to carcinoma.
Therefore, regardless of whether the biopsy shows a malignant or benign histology,
complete resection (either operative or endoscopic) is warranted.
Endoscopic ultrasonography (EUS) is a very useful modality in diagnosis of
periampullary disease, which combines and modifies the techniques of gastrointes-
tinal endoscopy and US. This combination decreases the distance between the
ultrasonic source and the organ of interest, thereby markedly improving the resolu-
tion and imaging of the surrounding structures. Real-time EUS enables one to
evaluate and integrate, on the same examination, mucosal, vascular, ductal, and
296 H. A. Lillemoe et al.

Fig. 6 Endoscopic
ultrasonography scan of
ampullary tumor, represented
by the hypoechoic area on the
right. An endoprosthesis
(small black arrows) can be
seen running through the
center of the tumor. The tumor
infiltrates beyond the
muscularis propria (open
arrows) into the pancreas.

parenchymal abnormalities. It allows detection of periampullary tumors, evaluation

of their size and depth of invasion, as well as assessment of regional lymph nodes.
EUS appears to be superior to CT and MRI for the detection of small pancreatic
tumors (<2 cm) [41]. However, the sensitivity of EUS decreases in the setting of
chronic pancreatitis [34]. EUS is able to demonstrate depth of invasion (T stage) of
mucosal-based ampullary and duodenal tumors with an accuracy rate of 73–84%,
increasing accuracy with higher T stages [42–44] (Fig. 6). This feature is of
importance in detecting noninvasive benign periampullary neoplasms from malig-
nant tumors with invasion through the bowel wall. Although results are not conclu-
sive, several reports have also indicated that EUS has greater sensitivity and
accuracy in detecting vascular invasion than CT [41, 43, 45].
The value of defining a benign versus a malignant periampullary mucosal-based
tumor is the opportunity to locally excise a benign lesion as opposed to offer
pancreaticoduodenectomy for malignant tumors. However, since frozen section
analysis of resected specimens can fail to detect malignancy in 14% of patients
[46], the surgeon always risks the possibility of a final diagnosis of cancer following
local excision. Furthermore, with a sensitivity of approximately 75% in predicting
T1 lesions, this modality is not necessarily optimal for predicting endoscopic
resectability. Underestimating the depth of the tumor penetration seldom occurs,
while overestimation is more common and is often due to edema of the submucosa
from associated pancreatitis or from peritumoral inflammation in ampullary carci-
noma. Similarly, indwelling transpapillary stents can cause inaccuracies and
overstaging [47].
Finally, EUS can determine the presence or absence of enlarged regional lymph
nodes. Reported accuracies of EUS-assessment of lymph node status have ranged
from 63% to 84%, which is at least equivalent to CT [34, 41, 43, 48]. Furthermore,
EUS offers the ability to perform fine-needle aspiration (FNA) of both the lesion and
suspicious regional lymph nodes.
Limitations of EUS include its complexity and in operator variability in both
performing and interpreting, its invasive nature, and its limited view (2–4 cm depth),
Miscellaneous Nonpancreatic Nonendocrine Tumors 297

which does not allow evaluation for distant sites of metastases. The combination of
CT and EUS is better than either alone in detecting resectability in patients with
periampullary cancers. The strategy of obtaining a CT for all patients with suspected
periampullary malignancies, followed by EUS in those patients in whom CT does
not clearly demonstrate unresectability has been shown to be the most cost-effective
strategy for preoperative staging of and determination of resectability in these tumors
[45, 48].

Endoscopic Retrograde Cholangiopancreatography

With advances in cross-sectional imaging and the introduction of endoscopic ultra-
sound and MRCP, the role of endoscopic retrograde cholangiopancreatography
(ERCP) in the diagnosis of periampullary lesions has become limited (Fig. 7). The

Fig. 7 (a) Magnetic resonance cholangiopancreaticogram (MRCP) showing an ampullary carci-

noma obstructing the distal common bile duct, (b) MRCP with distal common bile duct carcinoma.
Note the normal appearance of the main pancreatic duct, indicating a bile duct origin for the tumor,
and (c) MRCP of a pancreatic carcinoma, with partial obstruction of both the main pancreatic duct
and the common bile duct (“double-duct” sign)
298 H. A. Lillemoe et al.

Fig. 8 Endoscopic photo of

biliary stent placed through an
obstructing ampullary
carcinoma

most common current indication for ERCP in patients with periampullary tumors is
for placement of a temporary stent in the common bile duct to relieve biliary
obstruction preoperatively or as palliation (Fig. 8). Although stent placement will
lead to colonization of the biliary tree and a higher perioperative infection rate in
resected patients, it is appropriate in a number of clinical circumstances: (1) patients
who present with symptoms of cholangitis requiring immediate intervention to treat
the biliary infection, (2) patients presenting with intractable pruritus that can be
relieved during the period of preoperative evaluation, and (3) patients with hyper-
bilirubinemia associated with renal insufficiency, which will correct with relief of the
biliary obstruction. Under these circumstances, at least 2–3 weeks should be allowed
prior to definitive resection to allow the metabolic derangements to normalize and to
ensure the absence of active infection after instrumentation. Endoscopic stenting can
provide relief of jaundice in patients in whom delay in surgery may be necessary to
allow referral to a high-volume institution or for planned neoadjuvant therapy.
A final role for ERCP is to provide a tissue diagnosis of malignancy by either
histology or cytology through direct brushings. A tissue diagnosis is valuable
primarily for isolated bile duct strictures in clinical settings in which a benign
etiology of the stricture would alter management (resection vs. stenting or bypass).
In selected cases, Spyglass technology, essentially a smaller scope inserted the
working channel of a standard ERCP scope, can allow directed visualization of the
stricture for biopsy [49].
Miscellaneous Nonpancreatic Nonendocrine Tumors 299

Percutaneous Transhepatic Cholangiography

As with ERCP, percutaneous transhepatic cholangiography (PTC) for diagnosis of a
periampullary malignancy is seldom indicated due to the availability of noninvasive
techniques. It does remain a therapeutic option in some patients, particularly when
the endoscopic route is unsuccessful due to complete obstruction of the ampulla by
tumor or the ampulla is not accessible due to a prior surgical procedure such as
gastric bypass. PTC is often technically easier with a dilated biliary tree and is useful
in defining the proximal biliary system, which is critical in the decision-making
process concerning biliary reconstruction. Percutaneously placed catheters can be
helpful during operative management for either resection or palliation, especially in
reoperative cases or in the early postoperative period to allow biliary decompression
in order to protect the biliary anastomosis. In most patients, however, PTC offers
little advantage over ERCP, has a greater morbidity, and should be considered only if
ERCP is technically not possible.

Preoperative Staging

The goal of preoperative staging is to determine which tumors are potentially

resectable and have not already metastasized to distant sites or directly invaded the
major peripancreatic vessels. This is more important in patients with pancreatic
periampullary neoplasms because of the lower rate of resectability in this group. In
the past, laparotomy was required in all patients to establish the diagnosis and,
thereafter, resection or operative palliation was performed. Today, modalities includ-
ing dual-phase CT, EUS, and diagnostic laparoscopy allow us to clinically stage
patients preoperatively. The dynamic spiral CT scan is currently the most valuable of
these studies, playing a role in both diagnosis and staging of periampullary neo-
plasms. Its primary advantages are the lower cost and noninvasive nature of the
technique. Computed tomography can detect liver metastases (>1.0 cm) or larger
peritoneal implants [34]. EUS has high accuracy for evaluating T stage and defining
malignancy by demonstrating invasion. The technique can also be used to perform
an FNA for histologic evaluation of suspicious lymph nodes. However, EUS cannot
be used as the sole modality for staging. Given its inability to adequately rule out
peritoneal or hepatic metastases, it should be combined with CT or laparoscopy for
complete staging.
One of the limitations of CT is its poor sensitivity for detecting lesions in the liver,
omentum, or peritoneal surface that are less than 1 cm in size. In an attempt to
identify such metastases in a minimally invasive manner, laparoscopy has been
suggested as a method for further staging. A recently published Cochrane review
of 16 studies and a total of 1,146 patients with pancreatic or periampullary cancer
suggests that the addition of diagnostic laparoscopy decreases the rate of unneces-
sary laparotomy in those patients deemed resectable on CT scan by 20% [50]. How-
ever, this review includes studies with dates ranging from 1986 to 2014, and CT
has become more effective at picking up suspicious small volume metastases with
300 H. A. Lillemoe et al.

dual-phase imaging. Furthermore, the yield of diagnostic laparoscopy is likely lower

for patients with ampullary and duodenal tumors than those with pancreatic cancer,
leading many surgeon to avoid this step in patients with these tumors [51, 52].
The decision to stage patients with periampullary neoplasms via laparoscopy is
largely dependent on the treatment algorithms of the surgeon. Those surgeons
favoring surgical palliation as opposed to nonoperative palliation of unresectable
tumors consider laparoscopy unnecessary. Whereas those surgeons who feel
endoscopic palliation is adequate for most patients suggest that laparoscopy can
save a substantial number of patients from the morbidity of a noncurative lapa-
rotomy. Those centers currently investigating neoadjuvant chemotherapeutic and
radiation protocols also feel that laparoscopy is important in order to document
the absence of liver or peritoneal metastases. Improvements in preoperative
imaging and the addition of EUS to our clinical armamentarium has allowed for
better selection of patients for operation with fewer patients being found to be
unresectable at the time of operation, thereby minimizing unnecessary morbidity.
Nonoperative techniques for the management of obstructive jaundice secondary
to a periampullary tumor have also improved and can provide adequate palliation
for most patients with unresectable neoplasms. However, as mentioned, patients
with nonpancreatic periampullary neoplasms typically present earlier in the pro-
gression of their disease and have a much higher rate of resectability, thus in many
cases preoperative staging with the currently available imaging modalities is
sufficient.

Surgical Management

Endoscopic Resection

Benign periampullary tumors and small, ampullary tubular adenomas with very low
malignant potential may be endoscopically resected. Small, pedunculated adenomas
of the distal common bile duct can also be successfully treated and excised endo-
scopically. For tubular duodenal and Brunner gland adenomas, endoscopic excision
is the most suitable option. With villous duodenal adenomas, transduodenal local
excision should be considered depending on the size of the lesion. Endoscopic
resection of a villous adenoma may be performed only if the entire lesion can be
safely removed (Fig. 9). Close follow-up with repeat endoscopy is indicated in such
cases, as recurrence rates can be seen in 10–25% of cases [53]. Finally, it is
reasonable to consider endoscopic resection as a palliative option with patients
that cannot tolerate general anesthesia to perform even a local excision for peri-
ampullary cancers.
Complications following endoscopic resection of ampullary tumors includes
pancreatitis (5–15%), bleeding (4–15%), perforation (<2%), and cholangitis
(<2%). Mortality, however, remains very uncommon.
Miscellaneous Nonpancreatic Nonendocrine Tumors 301

Fig. 9 (a) Endoscopic appearance of a benign periampullary adenoma, (b and c) endoscopic

cautery excision of lesion, and (d) final appearance after complete endoscopic excision

Local Excision

Local resection of an ampullary tumor with reimplantation of the pancreatic and

common bile ducts was first described by Halsted in 1899. Initially, this procedure
was associated with high operative mortality and low long-term survival; however,
with improvements in technique and preoperative staging, transduodenal ampullary
resection has regained popularity. Local resection of the ampulla of Vater has been
suggested for benign ampullary tumors or low grade ampullary carcinomas. Histo-
logic confirmation of malignancy, large size, or extension into the common bile duct
or pancreatic duct precludes local excision. Furthermore, the false negative rate of
endoscopic biopsy (up to 25%) or even intraoperative frozen section (up to 14%)
requires that complete histologic diagnosis of the entire resected specimen be
completed [37, 46]. If invasive cancer is found in permanent sections, subsequent
resection with pancreaticoduodenectomy is necessary.
The operation begins with an exploration of the abdomen through a right sub-
costal or upper midline incision to rule out metastatic disease. An extended Kocher
maneuver is performed to mobilize the duodenum. A longitudinal duodenotomy is
made over the junction of the second and third portions of the duodenum. Stay
sutures are placed to expose the ampullary lesion, and the common bile duct is
302 H. A. Lillemoe et al.

Fig. 10 The ampulla is

exposed via a longitudinal
duodenotomy, and the
common bile duct is
cannulated (Reprinted from
Clary et al. [54])

cannulated through the center of the mass. If the common bile duct cannot be directly
entered, passage of a biliary Fogarty catheter from above via cannulation through the
cystic duct following cholecystectomy is advisable. Next, a resection margin of
0.5–1.0 cm of normal tissue is created by scoring the mucosal surface with electro-
cautery (Fig. 10). The lesion is excised by dissecting lateral to medial in the
submucosal plane. In this approach, the common bile duct located at 11 o’ clock,
is transected prior to the pancreatic duct and located at 5 o’ clock. The specimen is
sent to pathology for frozen-section analysis. If a negative margin is not accom-
plished or an invasive component is identified, then a pancreaticoduodenectomy
should be performed. In a series of 39 patients undergoing ampullectomy at Duke
University Medical Center, the negative predictive values of frozen-section analysis
was 94% [29]. If the lesion is benign and negative margins are achieved, then the
common channel between the common bile duct and pancreatic duct is reconstructed
by dividing the intervening septum with scissors. Next, the circumferential anasto-
mosis between the duodenal mucosa to the common channel is performed with 5-0
Vicryl interrupted sutures. Lastly, the duodenum is closed transversely in two layers.
Recurrence rates after local excision in patients with sporadic adenomas are
0–26% [28–30, 54, 55]. Significantly, increased rates of recurrences are seen in
patients with polyposis syndromes and approximately 25% of all recurrences are
invasive carcinomas [30]. This highlights the importance of surveillance endoscopy
following ampullectomy. Most series demonstrated complication rates of 20–25%
which included delayed gastric emptying, duodenal leak, pancreatitis, cholangitis,
and common bile duct stricture.
Miscellaneous Nonpancreatic Nonendocrine Tumors 303

Pancreaticoduodenectomy

Since its introduction by Whipple et al. in 1935, pancreaticoduodenectomy has been

the most effective treatment for periampullary carcinomas [56]. Either classic or
pylorus-preserving pancreaticodenectomy is appropriate for most periampullary
cancers, with the exception of patients with extensive duodenal polyposis associated
with FAP. In such cases all duodenal mucosa should be removed, and therefore the
total duodenectomy approach of the classic resection is appropriate. A prospective
randomized study by Yeo and colleagues, showed no advantage to an extended
retroperitoneal lymphadenectomy when performing a pancreaticoduodenectomy for
periampullary adenocarcinomas including ampullary and distal common bile duct
primaries [57].
Perioperative morbidity and mortality rates have continued to improve over the
past decade with mortality rates of 2% or less and morbidity rates of 30–40%
expected in patients treated at high volume centers [58–60]. One of the complica-
tions of pancreaticoduodenectomy that may be slightly increased in nonpancreatic
tumors is the rate of pancreatic anastomotic leak due to the normal, soft texture of the
pancreas. On the other hand, since local vascular invasion by periampullary non-
pancreatic tumors is uncommon, the procedures are often technically easier.
To date, no study has directly compared local ampullary resection with pancreati-
coduodenectomy for small ampullary cancers. There are several series including
subsets of patients with T1 lesions for whom local resection was performed, usually
in high-risk patients that were poor candidates for the more radical resection [28, 31,
61, 62]. Although these subsets are not prospectively randomized, patients that
underwent pancreaticoduodenectomy for T1 tumors generally experienced both
higher disease-free and overall survival rates [27, 28, 63, 64]. As a result, local
excision is only acceptable for patients with small ampullary cancers that are unable
to tolerate a pancreaticoduodenectomy. Refer to the clinical algorithm in Figure 11
for the evaluation and management of periampullary tumors (Fig. 11).

Segmental Resection

Surgery options for duodenal adenocarcinomas include segmental resection and

pancreaticoduodenectomy. For lesions involving the proximal first and second
portions of the duodenum, the treatment of choice is a Whipple procedure. Patients
with more distal tumors involving the third and fourth portions of the duodenum, an
en bloc segmental resection of the distal duodenum and proximal jejunum with
lymphadenectomy is appropriate. Past studies have demonstrated that pancreatico-
duodenectomy has an improved disease free interval and overall survival compared
to segmental resections. This difference was most likely due to the earlier detection
of periampullary duodenal adenocarcinomas than more distal tumors. More recent
research suggests that although radical resection with pancreatiocduodenectomy is
associated with a greater number of lymph nodes sampled, the overall survival is the
same as with segmental resection [65, 66].
304 H. A. Lillemoe et al.

Fig. 11 Clinical algorithm for the evaluation and management of a periampullary lesion. Note: If
invasive carcinoma discovered on endoscopic or local resection, proceed with
pancreaticoduodenectomy

Palliative Procedures

In patients with unresectable or metastatic disease found at exploration, palliative

operative gastric or biliary bypass should be strongly considered and performed
especially if patient is symptomatic. For those with recurrent disease or known
metastatic disease prior to exploration, palliative biliary stents and duodenal wall
stents placed endoscopically may be the most appropriate local therapy to relieve
symptoms and avoid delaying any additional systemic therapies being considered. In
patients with bulky bleeding tumors, gastrojejunostomy (potentially performed
laparoscopically), and radiation therapy can usually control symptoms.

Adjuvant and Neoadjuvant Therapy

The use of adjuvant and neoadjuvant therapies for nonpancreatic periampullary

cancers has been reported. Due to the rarity of these lesions, most series remain
low-powered and nonrandomized. Nevertheless, the use of neoadjuvant strategies
for treatment of periampullary malignancies is becoming more popular. These
approaches are mostly observed with pancreatic adenocarcinoma as very little
Miscellaneous Nonpancreatic Nonendocrine Tumors 305

published data exists at this point regarding nonpancreatic periampullary primaries.

The theoretical advantages include the delivery of a systemic therapy to well-
oxygenated tissues, and the potential for down-staging unresectable and borderline
resectable lesions. In multiple series, neoadjuvant chemoradiation did not increase
the mortality or morbidity of pancreaticoduodenectomy for periampullary cancers,
and interestingly yielded fewer pancreatic leaks and leak-associated morbidity and
mortality compared to those not receiving neoadjuvant therapy [67, 68]. Critics of
neoadjuvant protocols for potentially resectable periampullary cancers point to
selection biases based on favorable biology in those that proceed on to resection
following chemoradiation treatment.
The role of adjuvant therapy in ampullary cancer has been assessed in numerous
small studies. In a series from Stanford, 12 patients with resected ampullary cancers
having lymph node metastases, positive margins, tumor size >2 cm, poorly differ-
entiated, or neurovascular invasion were given adjuvant chemoradiation resulting
in an 89% actuarial 1-year survival [69]. In another series from Johns Hopkins,
17 of 106 patients with a resected ampullary cancer received adjuvant therapies
without any survival benefit [26]. In the European Organization for Research and
Treatment of Cancer (EORTC) Trial 40,891, there was no benefit of adjuvant
chemoradiation over observation for nonpancreatic periampullary malignancies
[70]. Finally, a recent meta-analysis of ten retrospective studies, including 3,361
patients, has demonstrated adjuvant chemoradiation therapy improved overall
survival [71]. Due to these mixed results, chemotherapy with regimens similar to
those used for colon cancer rather than the more aggressive chemoradiation
protocols.
Adjuvant therapies for cholangiocarcinoma are also not well defined. A Japanese
randomized, multi-institutional trial of 139 patients with bile duct cancer showed no
difference in 5-year survival for patients receiving adjuvant chemotherapy [72]. In
contrast, there is some data to support its use from a recent retrospective study. A
Johns Hopkins study from 1994 to 2003, treated 34 patients with distal bile duct
adenocarcinomas with pancreaticoduodenectomy followed by adjuvant
chemoradiation and compared with historical controls from the same institution.
For both lymph node positive and negative patients, overall survival was improved
in patients that received surgery plus adjuvant chemoradiation [73]. There are a few
prospective, randomized trials ongoing to determine the role of adjuvant
chemoradiation for biliary tract cancer [74].
Due to the relatively rare incidence of primary duodenal adenocarcinomas,
current data regarding its utility has not been able to identify a role for adjuvant
therapy. The group at Johns Hopkins published a small retrospective series of
14 patients with stage III/IV periampullary adenocarcinoma of the duodenum that
were treated with pancreaticoduodenectomy and adjuvant chemoradiotherapy. Com-
paring their results with historic controls, there was no difference in overall 5-year
survival between surgery plus adjuvant chemoradiation versus surgery alone
[75]. Despite the lack of data to justify adjuvant therapies for primary duodenal
adenocarcinoma at this time, most medical oncologists would recommend its use for
advanced stage disease.
306 H. A. Lillemoe et al.

Finally, the European Study Group for Pancreatic Cancer (ESPAC) – 3 trial was
reported in 2012 [76]. This open-label phase 3 randomized controlled trial involving
100 centers included 428 patients with resected nonpancreatic periampullary cancers
and compared adjuvant therapy via multiple regimens with observation. The results
were mixed with adjuvant chemotherapy not showing significant survival benefit
over observation (43.1 months vs. 35.2 months). However, multivariable analysis
adjusted for prognostic variables did show significant survival benefit with adjuvant
chemotherapy.

Survival

Overall, the survival following surgical resection for nonpancreatic periampullary

cancers are substantially better than periampullary pancreatic cancer (Fig. 12). In the
series from Johns Hopkins, duodenal and ampullary cancers demonstrate the 5-year
survival rates of 51–59 and 37–39%, respectively [2, 77]. In the same series, distal
cholangiocarcinomas and pancreatic cancers have the lowest 5-year survival rates, at
23–27 and 15–17%, respectively. In the Memorial Sloan Kettering experience,
ampullary carcinomas had the highest overall survival rates (median 43.6 months)
and resectability (82.1%) for periampullary tumors [64]. Beger et al. reviewed
171 cases of consecutive ampullary cancer treated by local or radical resection.
The 5-year survival rates by stage in that series were 84% (stage I), 70% (stage II),
and 27% (stage III) and 0% (stage IV) [28]. Poor prognostic indicators for recurrence

Fig. 12 The tumor-specific actual 5-year survival curves for the cohort of 242 patients treated by
pancreaticoduodenectomy for periampullary adenocarcinoma (Reprinted from [77])
Miscellaneous Nonpancreatic Nonendocrine Tumors 307

after resection of ampullary adenocarcinoma are advanced T stage, lymph node

involvement, positive margins, neural invasion, and poor differentiation [26, 27,
63, 64]. The two most important factors commonly found among different series are
T stage and nodal status, where the rate of lymph node involvement is a reflection of
the T stage progression. For T1–T2 and T3–T4 tumors the percent of lymph node
positivity is approximately 20 and 50%, respectively [28]. However, as previously
stated, there is no advantage to an extended retroperitoneal lymphadenectomy when
performing a pancreaticoduodenectomy for any periampullary adenocarcinoma [57].
With complete resection of distal cholangiocarcinoma, 5-year survivals range
from 21% to 54% [78–80]. Resection rates are generally between 40% and 85%. In
the series from DeOliveira et al., reviewing 564 patients with bile duct cancer
undergoing surgery, the 239 patients with distal cholangiocarcinoma had an overall
5-year survival for all patients and those after R0 resection were 23 and 27%,
respectively. The significant predictors of survival for patients with distal cholangio-
carcinoma included negative margins, lymph node involvement, size >2 cm, and
degree of differentiation [81]. The Japanese have compiled their extensive experi-
ence of distal cholangiocarcinomas into a national registry demonstrating a similar
5-year survival of 26% [82].
The most significant predictors of long-term survival for primary duodenal
carcinoma include margin negative resection and lymph node involvement. For
node-negative patients, overall 5-year survival following resection varies from
38% to 83%. For node-positive patients, the 5-year survival drops to 15–56%. In a
2000 series from the Mayo Clinic of 101 consecutive patients undergoing surgery for
adenocarcinoma of the duodenum, lymph node involvement, stage III or greater,
positive margin, and weight loss each carried a significantly negative impact on
survival [83]. In the same series, the tumor grade, size, and location within the
duodenum had no impact on survival (5-year survival, 54%). In the Memorial series,
the survival benefit between node-positive (5-year survival, 56%) and node-negative
(5-year survival, 83%) tumors demonstrated in patients with 15 nodes sampled did
not carry a similar positive prognostic impact on survival when <15 lymph nodes
were sampled [84]. The Hopkins group published their retrospective experience of
55 patients surgically treated primary adenocarcinoma of the duodenum [85]. Similar
to other series, the 5-year survival was 53%. In this series, negative margins,
pancreaticoduodenectomy, and tumors involving the first and second portions of
the duodenum were favorable predictors of long-term survival. Nodal status, tumor
diameter and grade did not influence survival in this study.

Conclusion

Nonendocrine, nonpancreatic periampullary tumors are rare lesions that encompass

a large array of pathology, originating most commonly from the ampulla of Vater,
distal common bile duct, and duodenum. These tumors are often asymptomatic and
have a tendency to be malignant. Treatment options depend on size and malignant
potential, ranging from endoscopic resection to pancreaticoduodenectomy. Further
308 H. A. Lillemoe et al.

research is required both to better understand the molecular biology of periampullary

tumors and the role of perioperative chemotherapy.

Key Practice Points

• Nonpancreatic periampullary malignancies originate from the distal common bile

duct, ampulla of Vater, and duodenum.
• Clinical findings of biliary obstruction require cross-sectional imaging and eval-
uation of the biliary system to exclude periampullary malignancies.
• Patients with familial syndromes (FAP, Gardner’s syndrome, inflammatory bowel
disease) must undergo close surveillance for periampullary cancers.
• Preoperative staging with CT and endoscopic ultrasound are the most cost-
effective diagnostic strategies for determining resectability.
• In general, due to the higher rate of resectability of nonpancreatic periampullary
neoplasms, preoperative staging and laparoscopic exploration are less important
than pancreatic primary tumors.
• Small, benign periampullary lesions may undergo endoscopic resection.
• Transduodenal resection should be considered for small (<3 cm) ampullary
tumors or low grade ampullary malignancies in patients unable to tolerate a
pancreaticoduodenectomy.
• For large periampullary lesions (>3 cm) and invasive periampullary malignan-
cies, pancreaticoduodenectomy remains the standard treatment.
• Adjuvant and neoadjuvant treatments for nonpancreatic periampullary tumors
have been investigated; however, no clear survival benefit has been identified.
• Actual 5-year survival rates following surgical resection for nonpancreatic peri-
ampullary cancers as 51% for duodenal cancer, 37% for ampullary cancer, and
23% for distal common bile duct cancer.
• Poor prognostic indicators for ampullary adenocarcinoma include advanced T
stage, lymph node involvement, positive margins, neural invasion, and poor
differentiation.
• Significant predictors of survival for distal cholangiocarcinoma include negative
margins, lymph node involvement, size >2 cm, and degree of differentiation.
• Significant predictors of survival for duodenal carcinoma include margin negative
resection and lymph node involvement.

Future Research Directions

• Translational investigations to better understand the molecular biology of peri-

ampullary tumors to improve early detection and targeted therapies.
• Technologic advances to improve local endoscopic diagnosis, staging, and
management.
• Multi-institutional clinical trials to investigate adjuvant and neoadjuvant therapies
for resectable periampullary cancers.
Miscellaneous Nonpancreatic Nonendocrine Tumors 309

Cross-References

▶ Chemotherapy for Advanced Pancreatic Cancer

▶ Clinical Decision-Making in Pancreatic Cancer
▶ EUS and Its Role in Pancreatic Cancer
▶ Laparoscopic Surgery for Pancreatic Neoplasms
▶ MRI and MRCP for Diagnosis and Staging of Pancreatic Cancer
▶ Surgical Resection for Pancreatic Cancer Using the International Study Group of
Pancreatic Surgery (ISGPS) Classifications
▶ Therapeutic Endoscopy in the Management of Pancreatic Cancer

References
1. Riall TS, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended
retroperitoneal lymphadenectomy for periampullary adenocarcinoma – part 3: update on 5-year
survival. J Gastrointest Surg. 2005;9(9):1191–204.
2. Riall TS, et al. Resected periampullary adenocarcinoma: 5-year survivors and their 6- to 10-year
follow-up. Surgery. 2006;140(5):764–72.
3. Chini P, Draganov PV. Diagnosis and management of ampullary adenoma: the expanding role
of endoscopy. World J Gastrointest Endosc. 2011;3(12):241–7.
4. Panzeri F, Crippa S, Castelli P, et al. Management of ampullary neoplasms: a tailored approach
between endoscopy and surgery. World J Gastroenterol. 2015;21(26):7970–87.
5. Blackman E, Nash SV. Diagnosis of duodenal and ampullary epithelial neoplasms by endo-
scopic biopsy: a clinicopathologic and immunohistochemical study. Hum Pathol. 1985;16(9):
901–10.
6. Yamaguchi K, Enjoji M. Carcinoma of the ampulla of Vater. A clinicopathologic study and
pathologic staging of 109 cases of carcinoma and 5 cases of adenoma. Cancer. 1987;59(3):
506–15.
7. Galandiuk S, et al. Villous tumors of the duodenum. Ann Surg. 1988;207(3):234–9.
8. Mino M, Lauwers GY. Pathology of periampullary tumors. In: Von Hoff DD, Evans DB,
Hruban RH, editors. Pancreatic cancer. Boston: Jones and Bartlett; 2005. p. 686–702.
9. Carter JT, et al. Tumors of the ampulla of Vater: histopathologic classification and predictors of
survival. J Am Coll Surg. 2008;207(2):210–8.
10. Komorowski RA, et al. Assessment of ampulla of Vater pathology. An endoscopic approach.
Am J Surg Pathol. 1991;15(12):1188–96.
11. Devaney K, Goodman ZD, Ishak KG. Hepatobiliary cystadenoma and cystadenocarcinoma.
A light microscopic and immunohistochemical study of 70 patients. Am J Surg Pathol. 1994;
18(11):1078–91.
12. Barton JG, Barrett DA, Maricevich MA, et al. Intraductal papillary mucinous neoplasm of the
biliary tract: a real disease? HPB (Oxford). 2009;11(8):684–91.
13. Everhart JE, Ruhl CE. Burden of digestive diseases in the United States part III: liver, biliary
tract, and pancreas. Gastroenterology. 2009;136(4):1134–44.
14. Nakeeb A, et al. Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors.
Ann Surg. 1996;224(4):463–73.
15. Cardinale V, Semeraro R, Torrice A, et al. Intra-hepatic and extra-hepatic cholangiocarcinoma:
new insight into epidemiology and risk factors. World J Gastrointest Oncol. 2010;2(11):407–16.
16. Weinbren K, Mutum SS. Pathological aspects of cholangiocarcinoma. J Pathol. 1983;139(2):
217–38.
17. Pitt HA, et al. Malignancies of the biliary tree. Curr Probl Surg. 1995;32(1):1–90.
310 H. A. Lillemoe et al.

18. Jarnagin WR, et al. Papillary phenotype confers improved survival after resection of hilar
cholangiocarcinoma. Ann Surg. 2005;241(5):703–12.
19. Howe JR, et al. The American College of Surgeons Commission on Cancer and the American
Cancer Society. Adenocarcinoma of the small bowel: review of the National Cancer Data Base,
1985–1995. Cancer. 1999;86(12):2693–706.
20. Surveillance, Epidemiology, and End Results (SEER) Program. Research data (1973–2010),
www.seer.cancer.gov. Accessed 18 Mar 2017.
21. Bilimoria KY, Bentrem DJ, Wayne JD, Ko CY, Bennett CL, Talamonti MS. Small bowel cancer
in the United States: changes in epidemiology, treatment, and survival over the last 20 years.
Ann Surg. 2009;249(1):63–71.
22. Joesting DR, et al. Improving survival in adenocarcinoma of the duodenum. Am J Surg.
1981;141(2):228–31.
23. Joensuu H, et al. Gastrointestinal stromal tumor (GIST). Ann Oncol. 2006;17(Suppl):280–6.
24. Leese T, et al. Tumours and pseudotumours of the region of the ampulla of Vater: an endoscopic,
clinical and pathological study. Gut. 1986;27(10):1186–92.
25. Yamaguchi K, Enjoji M. Tsuneyoshi, pancreatoduodenal carcinoma: a clinicopathologic study
of 304 patients and immunohistyochemical observation for CEA and CA19-9. J Surg Oncol.
1991;47(3):148–54.
26. Talamini MA, et al. Adenocarcinoma of the ampulla of Vater. A 28-year experience. Ann Surg.
1997;225(5):590–9.
27. Klempnauer J, et al. Carcinoma of the ampulla of Vater: determinants of long-term survival in
94 resected patients. HPB Surg. 1998;11(1):1–11.
28. Beger HG, et al. Tumor of the ampulla of Vater: experience with local or radical resection in
171 consecutively treated patients. Arch Surg. 1999;134(5):526–32.
29. Clary BM, et al. Local ampullary resection with careful intraoperative frozen section evaluation
for presumed benign ampullary neoplasms. Surgery. 2000;127(6):628–33.
30. Farnell MB, et al. Villous tumors of the duodenum: reappraisal of local vs. extended resection.
J Gastrointest Surg. 2000;4(1):13–21, discussion 22–3.
31. Rattner DW, et al. Defining the criteria for local resection of ampullary neoplasms. Arch Surg.
1996;131(4):366–71.
32. Iwama T, et al. Indications for local excision of ampullary lesions associated with familial
adenomatous polyposis. J Am Coll Surg. 1994;179(4):462–4.
33. Balthazar EJ, Chako AC. Computed tomography of pancreatic masses. Am J Gastroenterol.
1990;85(4):343–9.
34. Walsh RM, Connelly M, Baker M. Imaging for the diagnosis and staging of periampullary
carcinomas. Surg Endosc. 2003;17(10):1514–20.
35. Wyatt SH, Fishman EK. Spiral CT of the pancreas. Semin Ultrasound CT MR. 1994;15(2):
122–32.
36. Steiner E, et al. Imaging of pancreatic neoplasms: comparison of MR and CT. AJR Am J
Roentgenol. 1989;152(3):487–91.
37. Roggin KK, Yeh JJ, Ferrone CR, Riedel E, Gerdes H, Klimstra DS, Jaques DP, Brennan
MF. Limitations of ampullectomy in the treatment of nonfamilial ampullary neoplasms. Ann
Surg Oncol. 2005;12:971–80.
38. Menzel J, Poremba C, Dietl KH, Böcker W, Domschke W. Tumors of the papilla of Vater –
inadequate diagnostic impact of endoscopic forceps biopsies taken prior to and following
sphincterotomy. Ann Oncol. 1999;10:1227–31.
39. Elek G, Gyôri S, Tóth B, Pap A. Histological evaluation of preoperative biopsies from ampulla
vateri. Pathol Oncol Res. 2003;9:32–41.
40. Grobmyer SR, Stasik CN, Draganov P, Hemming AW, Dixon LR, Vogel SB, Hochwald
SN. Contemporary results with ampullectomy for 29 “benign” neoplasms of the ampulla.
J Am Coll Surg. 2008;206:466–71.
41. Rivadeneira DE, et al. Comparison of linear array endoscopic ultrasound and helical computed
tomography for the staging of periampullary malignancies. Ann Surg Oncol. 2003;10(8):890–7.
Miscellaneous Nonpancreatic Nonendocrine Tumors 311

42. Chen CH, Tseng LJ, Yang CC, Yeh YH. Preoperative evaluation of periampullary tumors by
endoscopic sonography, transabdominal sonography, and computed tomography. J Clin Ultra-
sound. 2001;29:313–21.
43. Kubo H, et al. Pre-operative staging of ampullary tumours by endoscopic ultrasound. Br J
Radiol. 1999;72(857):443–7.
44. Trikudanathan G, Njei B, Attam R, Arain M, Shaukat A. Staging accuracy of ampullary tumors
by endoscopic ultrasound: meta-analysis and systematic review. Dig Endosc. 2014;26(5):
617–26.
45. Tierney WM, et al. The accuracy of EUS and helical CT in the assessment of vascular invasion
by peripapillary malignancy. Gastrointest Endosc. 2001;53(2):182–8.
46. Sharp KW, Brandes JL. Local resection of tumors of the ampulla of Vater. Am Surg. 1990;
56(4):214–7.
47. Chen CH, Yang CC, Yeh YH, et al. Reappraisal of endosonography of ampullary tumors:
correlation with transabdominal sonography, CT, and MRI. J Clin Ultrasound. 2009;37:18–25.
48. Soriano A, et al. Preoperative staging and tumor resectability assessment of pancreatic cancer:
prospective study comparing endoscopic ultrasonography, helical computed tomography, mag-
netic resonance imaging, and angiography. Am J Gastroenterol. 2004;99(3):492–501.
49. Sun Y, et al. Single-operator cholangioscopy in the diagnosis of indeterminate biliary strictures.
Gastrointest Endosc. 2015;82(6):1136–7.
50. Allen VB, Gurusamy KS, Takwoingi Y, Kalia A, Davidson BR. Diagnostic accuracy of
laparoscopy following computed tomography (CT) scanning for assessing the resectability
with curative intent in pancreatic and periampullary cancer. Cochrane Database Syst Rev.
2016;7:CD009323.
51. Brooks AD, et al. The value of laparoscopy in the management of ampullary, duodenal, and
distal bile duct tumors. J Gastrointest Surg. 2002;6(2):139–45.
52. Beenen E, van Roest MHG, Sieders E, Peeters P, Porte RJ, de Boer MT, et al. Staging
laparoscopy in patients scheduled for pancreaticoduodenectomy minimizes hospitalization in
the remaining life time when metastatic carcinoma is found. Eur J Surg Oncol. 2014;40(8):
989–94.
53. Ridtitid W, et al. Endoscopic paillectomy: risk factors for incomplete resection and recurrence
during long-term follow-up. Gastrointest Endosc. 2014;79(2):289–96.
54. Clary BM, Pappas TN, Tyler DS. Transduodenal local resection for periampullary neoplasms.
In: Evans DB, Abbruzzese JL, Pisters PW, editors. Pancreatic cancer, M.D. Anderson solid
tumor oncology series. New York: Springer; 2007. p. 181–91.
55. Cahen DL, et al. Local resection or pancreaticoduodenectomy for villous adenoma of the
ampulla of Vater diagnosed before operation. Br J Surg. 1997;84(7):948–51.
56. Whipple AO, Parsons WB, Mullins CR. Treatment of carcinoma of the ampulla of Vater. Ann
Surg. 1935;102(4):763–79.
57. Yeo CJ, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended
retroperitoneal lymphadenectomy for periampullary adenocarcinoma, part 2: randomized con-
trolled trial evaluating survival, morbidity, and mortality. Ann Surg. 2002;236(3):355–66.
58. Wray CJ, et al. Surgery for pancreatic cancer: recent controversies and current practice.
Gastroenterology. 2005;128(6):1626–41.
59. Schmidt CM, et al. Pancreaticoduodenectomy: a 20-year experience in 516 patients. Arch Surg.
2004;139(7):718–25.
60. Stephens J, et al. Surgical morbidity, mortality, and long-term survival in patients with peri-
pancreatic cancer following pancreaticoduodenectomy. Am J Surg. 1997;174(6):600–3.
61. Klein P, et al. Is local excision of pT1-ampullary carcinomas justified? Eur J Surg Oncol.
1996;22(4):366–71.
62. Branum GD, Pappas TN, Meyers WC. The management of tumors of the ampulla of Vater by
local resection. Ann Surg. 1996;224(5):621–7.
63. Su CH, et al. Factors affecting morbidity, mortality and survival after pancreaticoduodenectomy
for carcinoma of the ampulla of Vater. Hepato-Gastroenterology. 1999;46(27):1973–9.
312 H. A. Lillemoe et al.

64. Howe JR, et al. Factors predictive of survival in ampullary carcinoma. Ann Surg. 1998;228(1):
87–94.
65. Cloyd JM, Norton JA, Visser BC, Poultsides GA. Does the extent of resection impact survival
for duodenal adenocarcinoma? Analysis of 1,611 cases. Ann Surg Oncol. 2015;22:573–80.
66. Onkendi EO, Boostrom SY, Sarr MG, et al. 15-year experience with surgical treatment of
duodenal carcinoma: a comparison of periampullary and extra-ampullary duodenal carcinomas.
J Gastrointest Surg. 2012;16(4):682–91.
67. Cheng TY, et al. Effect of neoadjuvant chemoradiation on operative mortality and morbidity for
pancreaticoduodenectomy. Ann Surg Oncol. 2006;13(1):66–74.
68. Ferrone CR, et al. Tadiological and surgical implications of neoadjuvant treatment with
FOLFIRINOX for locally advanced and borderline resectable pancreatic cancer. Ann Surg.
2015;261(1):12–7.
69. Mehta VK, et al. Adjuvant chemoradiotherapy for “unfavorable” carcinoma of the ampulla of
Vater: preliminary report. Arch Surg. 2001;136(1):65–9.
70. Smeenk HG, et al. Long-term survival and metastatic pattern of pancreatic and periampullary
cancer after adjuvant chemoradiation or observation: long-term results of EORTC trial 40891.
Ann Surg. 2007;246(5):734–40.
71. Kwon J, Kim BH, Kim K, et al. Survival benefit of adjuvant chemoradiotherapy in patients with
ampulla of Vater cancer. Ann Surg. 2015;262(1):47–52.
72. Takada T, et al. Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A
phase III multicenter prospective randomized controlled trial in patients with resected pancreati-
cobiliary carcinoma. Cancer. 2002;95(8):1685–95.
73. Hughes MA, et al. Adjuvant concurrent chemoradiation for adenocarcinoma of the distal
common bile duct. Int J Radiat Oncol Biol Phys. 2007;68(1):178–82.
74. Ben-Josef E, Guthrie KA, El-Khoueiry AB, et al. SWOG S0809: a phase II intergroup trial of
adjuvant capecitabine and gemcitabine followed by radiotherapy and concurrent capecitabine in
extrahepatic cholangiocarcinoma and gallbladder carcinoma. J Clin Oncol. 2015;33(24):
2617–22.
75. Swartz MJ, et al. Adjuvant concurrent chemoradiation for node-positive adenocarcinoma of the
duodenum. Arch Surg. 2007;142(3):285–8.
76. Neoptolemos JP, et al. European Study Group for Pancreatic Cancer. Effect of adjuvant
chemotherapy with flourouracil plus folinic acid or gemcitabine vs observation on survival in
patients with resected periampullary adenocarcinoma. The ESPAC-3 Periampullary Cancer
Randomized Trial. JAMA. 2012;308(2):147–56.
77. Yeo CJ, et al. Periampullary adenocarcinoma. Analysis of 5-year survivors. Ann Surg. 1998;
227(6):821–31.
78. Fong Y, et al. Outcome of treatment for distal bile duct cancer. Br J Surg. 1996;83(12):1712–5.
79. Wade TP, et al. Experience with distal bile duct cancers in U.S. Veterans Affairs hospitals:
1987–1991. J Surg Oncol. 1997;64(3):242–5.
80. Jang J-Y, Kim S-W, Park DJ, et al. Actual long-term outcome of extrahepatic bile duct cancer
after surgical resection. Ann Surg. 2005;241(1):77–84.
81. DeOliveira ML, et al. Cholangiocarcinoma: thirty-one-year experience with 564 patients at a
single institution. Ann Surg. 2007;245(5):755–62.
82. Nagakawa T, et al. Biliary tract cancer treatment: results from the Biliary Tract Cancer Statistics
Registry in Japan. J Hepato-Biliary-Pancreat Surg. 2002;9(5):569–75.
83. Bakaeen FG, et al. What prognostic factors are important in duodenal adenocarcinoma? Arch
Surg. 2000;135(6):635–41.
84. Sarela AI, et al. Adenocarcinoma of the duodenum: importance of accurate lymph node staging
and similarity in outcome to gastric cancer. Ann Surg Oncol. 2004;11(4):380–6.
85. Sohn TA, et al. Adenocarcinoma of the duodenum: factors influencing long-term survival.
J Gastrointest Surg. 1998;2(1):79–87.