AJR 2003; 180:1311-1323
© American Roentgen Ray Society
Imaging in Oncology from The University of Texas M. D. Anderson
Cancer Center |
Diagnosis, Staging, and Surveillance of Pancreatic Cancer
Eric P. Tamm1,
Paul M. Silverman1,
Chusilp Charnsangavej1 and
Douglas B. Evans2
1 Department of Diagnostic Radiology, The University of Texas M. D. Anderson
Cancer Center, 1515 Holcombe Blvd., Box 57, Houston, TX 77030.
2 Department of Surgery, The University of Texas M. D. Anderson Cancer Center,
Houston, TX 77030.
Received May 20, 2002;
accepted after revision September 3, 2002.
Address correspondence to E. P. Tamm.
Introduction
Pancreatic ductal adenocarcinoma is the fifth leading cause of cancer death
in the Western hemisphere. The age adjusted incidence rates range from 3.1 to
20.8 deaths per 100,000 in developed countries
[1,
2,
3], with a peak incidence in
patients between 60 and 80 years old. Because of the very low rate of
survival, the incidence and mortality rates are similar.
Factors associated with an increased risk of pancreatic cancer include
smoking, chronic pancreatitis, diabetes, prior gastric surgery, and exposure
to radiation or chemicals such as chlorinated hydrocarbon solvents
[4,
5,
6]. A number of syndromes are
identified with an increased incidence of pancreatic cancer, including
familial atypical multiple-mole melanoma syndrome, hereditary nonpolyposis
colorectal cancer, hereditary pancreatitis, Peutz-Jeghers syndrome, and
hereditary breast–ovarian cancer syndrome
[7]. Genetic point mutations,
amplifications, or overexpression of oncogenes, such as KRAS, HER/2-neu,
[8,
9,
10,
11], and alterations of tumor
suppressor genes, such as p16 and TP53
[12,
13], have also been associated
as risk factors.
Pathology
On macroscopic examination, ductal adenocarcinomas are firm, poorly defined
masses. Histopathologically, these tumors range from poor-to
well-differentiated, with glandular structures embedded in desmoplastic stroma
[14]. The desmoplastic stroma
and the variable presence of mucin likely contribute to the typical hypodense
appearance on CT. These tumors do not stain with endocrine markers
[15], and serum levels of
pancreatic enzymes such as chymotrypsin, trypsin, and lipase are usually not
elevated [16].
Tumor Markers
No tumor marker has been identified for screening the population. The most
widely used serum marker is CA 19-9, which is also elevated in patients with
other malignancies (stomach, colon, and biliary tree) and can be elevated in
benign conditions that include pancreatitis, hepatitis, acute cholangitis,
biliary obstruction, and cirrhosis. Sensitivity and specificity for pancreatic
ductal adenocarcinoma (using a cutoff of 37 U/mL) have been reported to be
81–85% and 85–90%, respectively
[17,
18]. Other markers, such as
carcinoembryonic antigen (CEA), CA 242, CA 72-4, and telomerase are of limited
clinical usefulness [17,
19,
20].
Staging
The TNM staging system is used for staging of pancreatic cancer
[21,
22]
(Table 1 and Figs.
1A,
1B,
2A,
2B,
3A,
3B,
4,
5A,
5B,
6A,
6B). The resectability of a
local tumor is determined by whether the tumor extends to major arterial
structures (i.e., celiac axis, superior mesenteric structures) and whether
long segment involvement, or occlusion, of major venous structures is present
(Table 2). Surgery to remove
the primary tumor is not considered appropriate in the setting of distant
metastases. Recent changes have been made in the T system of classification
and stage grouping to provide a clearer distinction between potentially
resectable and locally advanced tumors; these changes are reflected in the
tables provided [21,
22]. The prior TNM
classification described T4 disease, typically representing unresectable
disease, as tumor extension into the stomach, spleen, colon, or large adjacent
vessels (arteries or veins). With recent advances in surgical techniques,
particularly in venous interposition grafts, the tumors of patients with
limited superior mesenteric vein involvement are now considered to be
resectable at many institutions. Nevertheless, these patients would have been
considered to have T4 disease under the old system. The newer TNM system
describes such isolated venous involvement as T3 disease.
View larger version (71K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —T1 and T2 tumors. Drawing shows T1 tumor, which is defined as
being equal to or smaller than 2 cm in maximum diameter and confined to
pancreas, and T2 tumor, larger than 2 cm and confined to pancreas.
|
|
View larger version (150K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —T3 tumor. 68-year-old man with history of T3 N0 M0 pancreatic
ductal adenocarcinoma. Contrast-enhanced axial CT image shows T3 tumor
(medium-length arrows) that has involved common bile duct, requiring
a stent (curved arrow), and that extends medially beyond confines of
pancreatic head. Tumor is separated from superior mesenteric vein (long
arrow) and superior mesenteric artery (short arrow) by fat plane
(type A relationship). Note that tumor involves duodenum
(arrowhead).
|
|
View larger version (162K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —T4 tumor. 69-year-old woman with T4 NX M0 disease.
Contrast-enhanced axial CT image shows pancreatic tumor (white
arrows) engulfing celiac axis. Short black arrow = splenic artery, long
black arrow = common hepatic artery.
|
|
View larger version (77K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4. —Drawing shows lymph node groups that can be involved by local
spread of tumor: 1, superior to pancreatic head; 2, superior to pancreatic
body; 3, near pancreatic tail; 4, splenic hilum; 5, anterior
pancreaticoduodenal; 6, inferior to pancreatic head and body; 7, near common
bile duct; 8, near pancreaticoduodenal groove and pylorus. Proximal mesenteric
nodes (9) are hidden posteriorly.
|
|
View larger version (160K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A. —60-year-old man with history of T3 N1 MO pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image shows pancreatic carcinoma
(thin arrows) in uncinate process that involves duodenal wall.
Gastrocolic trunk (thick arrow) is enlarged.
|
|
View larger version (178K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B. —60-year-old man with history of T3 N1 MO pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image acquired at level superior
relative to A shows portacaval nodal disease (arrows),
identified as separate from primary mass.
|
|
View larger version (127K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A. —54-year-old woman with history of T4 N1 M1 pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image shows ductal carcinoma
(white arrows) of pancreatic body and tail. Liver metastasis is
present (black arrow). Prominent local varices are seen
(arrowheads). Superior mesenteric vein (S) is involved by tumor.
|
|
View larger version (116K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B. —54-year-old woman with history of T4 N1 M1 pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image obtained in portal venous
phase shows multiple liver metastases (black arrows) and left gastric
adenopathy (white arrow).
|
|
The location of the tumor in the pancreas determines its route of spread
and the nodal groups involved (Fig.
4). A tumor in the anterior portion of the pancreatic head grows
along the anterior pancreaticoduodenal arcades towards the gastroduodenal
artery and the proper hepatic artery, whereas a tumor in the posterior
pancreatic head typically extends along the posterior pancreaticoduodenal vein
towards the inferior portal vein surface. A tumor in the cranial aspect of the
pancreatic head grows towards the common hepatic artery. A tumor in the
pancreatic head near the gastrocolic trunk confluence may infiltrate the base
of the transverse mesocolon. In contrast, uncinate tumors typically grow along
the inferior pancreaticoduodenal arcade along the posterior surface of the
superior mesenteric artery or into the jejunal mesentery. Tumors in the
pancreatic body and tail generally infiltrate along the splenic artery and
vein to either the celiac axis or portal vein, with potential for direct
invasion of the spleen, peritoneum, stomach, colon, and left adrenal gland
[14].
Treatment
Currently, surgery remains the only option for cure. The surgical procedure
for resection of a tumor in the pancreatic head or uncinate process is
pancreaticoduodenectomy. In this procedure, the gallbladder is removed; the
common hepatic duct, the gastroduodenal artery, and the pancreatic neck are
divided; and the gastric antrum (or duodenum) and the proximal jejunum are
transected. The following anastomoses are then made: end-to-end or end-to-side
pancreaticojejunostomy, hepaticojejunostomy, and gastrojejunostomy
[23]. In the
pylorus-preserving variant of the Whipple procedure
[24], the antrum and pylorus
of the stomach are spared in an attempt to more closely preserve normal
gastrointestinal physiology. The five-year survival after
pancreaticoduodenectomy (Whipple procedure) is 18–21%
[25,
26,
27,
28]. Tumors in the pancreatic
body and tail typically require distal pancreatectomy, usually combined with
splenectomy.
Local involvement of venous structures by tumor tissue is seen in nearly
33% of patients with pancreatic cancer
[29]. Primary reconstruction
with an end-to-end anastomosis, or an interposition graft, will be attempted
in these patients in many institutions
[29,
30]. However, long segment
involvement or thrombosis of either the superior mesenteric vein or the portal
vein is usually a contraindication to resection.
Because of the poor prognosis seen with surgery alone, investigations have
been made into the use of adjuvant or neoadjuvant therapies. The most
frequently studied chemotherapeutic agent has been 5-FU. Two studies using
5-FU in combination with radiotherapy as adjuvant therapy showed an
improvement of median survival to 20 months over observation alone
(11–14 months) [31,
32]. Subsequently,
gemcitabine, a pyrimidine antimetabolite, was shown to provide a survival
advantage over 5-FU in patients with advanced and metastatic pancreatic cancer
[33]. However, the combined
use of gemcitabine with 5-FU has not resulted in improvement in survival
compared with that of patients treated with gemcitabine alone
[34]. Early data from small
studies evaluating the use of gemcitabine in combination with other agents
have been mixed [35]. The
Radiation Therapy Oncology Group is currently evaluating gemcitabine versus
5-FU for adjuvant therapy.
Unfortunately, the postoperative morbidity of many patients prevents them
from receiving systemic therapy. Therefore, some institutions, such as our
own, have advocated the preoperative (neoadjuvant) use of chemoradiation
therapy. The limited preliminary data available on the use of gemcitabine,
typically combined with radiation therapy, have been promising
[36,
37].
Preliminary investigations are underway into novel systemic therapies based
on new knowledge of the mechanisms of this disease; these include the use of
farnesyl transferase inhibitors, tyrosine kinase inhibitors, antiangiogenic
agents, and gene therapy [38,
39]. Imaging plays an
important role in following the results of neoadjuvant therapy, both for
assessing the efficacy of treatment and for determining whether the tumor has
progressed to a point that would contraindicate surgery (e.g., liver
metastases).
Radiologic Evaluation: Imaging Approaches
Imaging often begins with transabdominal sonography to identify a cause of
abdominal pain or jaundice. Sonography can screen for gallstones, signs of
cholecystitis, and for the presence and level (intrahepatic, suprapancreatic,
or intrapancreatic) of common bile duct obstruction. However, the presence of
obscuring overlying bowel gas and the variable skill of the operator limit the
sensitivity of this technique for identification and staging of pancreatic
tumors.
After sonography, CT is the modality most used as the primary modality for
diagnosis and staging. Optimal imaging by helical CT (single-or multidetector)
for pancreatic cancer is obtained after the rapid injection of iodinated
contrast material (140–150 mL at 4–5 mL/sec). The relatively
hypovascular tumor is best detected during the pancreatic parenchymal phase of
enhancement, approximately 35–50 sec after the beginning of contrast
medium injection [40,
41]. On the other hand, liver
metastases are best imaged during the portal venous phase of liver
enhancement, approximately 60–70 sec after the beginning of contrast
medium injection. A "dual-phase" technique is therefore often used
to obtain information regarding staging and metastases. Unfortunately,
circulation times vary between patients, and simple rules for timing may not
always be effective. Currently available computer-based automated scanning
techniques can be used to compensate; however, they incur an additional
expense and are dependent on equipment and software
[42].
No study has compared the use of water to a positive oral contrast agent in
the diagnosis and staging of pancreatic ductal adenocarcinoma. The use of
water as an oral contrast agent has been shown to be useful in detecting
hypervascular islet cell tumors that may be located in the duodenal wall,
because it enhances the difference in contrast between the tumor and the
duodenal lumen [43]. A study
of 211 consecutive patients by Richter et al.
[44] using water and IV
N-butylscopo-laminium bromide showed an accuracy of 95% for detecting
pancreatic neoplasms. At our institution, patients being imaged either before
or after surgery typically drink a barium sulfate suspension; our anecdotal
experience is that this practice aids in the detection of metastatic
implants.
Thin-section imaging is vital for optimizing lesion detection; thin-section
imaging diminishes the impact of volume averaging on obscuring small lesions.
Slices are typically obtained at a slice thickness of 3–5 mm on
single-detector helical CT units
[45,
46,
47,
48]. To our knowlege, no
reports have been published regarding slice thickness and multidetector CT. At
our institution, images are obtained during the pancreatic parenchymal phase
of imaging at a slice thickness of 2.5 mm and during the portal venous phase,
at a slice thickness of 5.0 mm and for each phase are reconstructed to half
that thickness. Current optimized multidetector CT protocols allow for
advanced image reformatting to show vascular and biliary anatomy
[49,
50] (Figs.
7A,
7B).
View larger version (152K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7A. —59-year-old man with history of T4 NX M0 pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image shows tumor in pancreatic
head (long white arrows) that has concave point of contact with
superior mesenteric artery (short white arrow), type D relationship.
Inferior pancreaticoduodenal artery (arrowhead) is encased by tumor.
Stent (black arrow) is present.
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7B. —59-year-old man with history of T4 NX M0 pancreatic ductal
adenocarcinoma. Coronal oblique reformatted CT image shows tumor (white
arrows) involving superior mesenteric artery (arrowhead). Stent
(black arrow) is also seen.
|
|
A study of 57 patients compared the use of single-detector helical CT
versus dynamic MR imaging in the evaluation of pancreatic carcinoma. The
findings of the two reviewers in that study showed CT had sensitivities of 83%
and 96% and specificities of 81% and 89%
[47]. The sensitivity and
specificity for multidetector CT has yet to be determined. In the same study,
the two reviewers found that MR imaging had sensitivities of 80% and 95% and
specificities of 71% and 78%
[47].
MR imaging offers several benefits for imaging of the pancreas. It
inherently offers better soft-tissue contrast than CT before the
administration of an IV contrast agent, and images can be obtained in multiple
planes. MR imaging can be performed in patients with a history of allergy to
iodinated contrast agents and in those with renal insufficiency. However, CT
offers higher spatial resolution. MR imaging protocols typically include
T1-weighted spin-echo or fast spoiled-gradient breath-hold sequences with or
without fat suppression, T2-weighted fast spin-echo with fat suppression
sequences, and dynamically enhanced T1-weighted spoiled-gradient breath-hold
with or without fat suppression sequences (Figs.
8A,
8B and
9).
View larger version (162K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8A. —59-year-old man with history of T3 NX M0 pancreatic ductal
adenocarcinoma. T1-weighted axial MR image shows pancreatic carcinoma
(black arrows) involving superior mesenteric vein (white
arrow).
|
|
View larger version (134K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8B. —59-year-old man with history of T3 NX M0 pancreatic ductal
adenocarcinoma. T2-weighted axial MR image acquired at same level as
T1-weighted image (A) shows diffuse increased signal in pancreatic head
(arrows). Differentiation between normal pancreatic parenchyma and
tumor is limited. Patient was placed on neoadjuvant therapy in preparation for
surgery; subsequent imaging (not shown) revealed development of liver
metastases.
|
|
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9. —54-year-old woman with history of T4 NX M0 pancreatic ductal
adenocarcinoma. Fat suppressed spin-echo T1-weighted MR image of pancreas
shows tumor (thin white arrows) surrounding and narrowing superior
mesenteric vein (black arrow) and in contact with superior mesenteric
artery (thick white arrow).
|
|
T2-weighted MR images obtained with long echo times have been used to
create cholangiographic images (MR cholangiopancreatography). This technique
can be used to acquire images in any plane to provide additional information
on the level of obstruction of the biliary or pancreatic ductal systems, with
a sensitivity and specificity that rivals that of endoscopic retrograde
cholangiopancreatography (Figs.
10A,
10B,
10C,
11A,
11B,
11C,
12)
[51].
View larger version (157K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10A. —68-year-old man with history of T4 NX M0 pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image shows pancreatic head mass
(medium-sized arrows) involving superior mesenteric vein (large
arrow). Superior mesenteric artery (small arrow) was not
definitely shown to be involved on this image, but other images (not shown)
revealed involvement.
|
|
View larger version (152K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10B. —68-year-old man with history of T4 NX M0 pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image acquired at level superior to
A shows dilated biliary tree (arrows) associated with
obstruction by mass in pancreatic head.
|
|
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10C. —68-year-old man with history of T4 NX M0 pancreatic ductal
adenocarcinoma. ERCP image shows dilated biliary tree (white arrows)
and obstruction of common bile duct (black arrow) associated with
tumor in pancreatic head.
|
|
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11A. —64-year-old man with T3 NX M1 pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image obtained in portal venous
phase shows tumor (arrowheads) surrounding stent within common bile
duct (thick white arrow). Dilated pancreatic duct (thin white
arrow) is partially seen. Liver metastasis is present (black
arrow). Superior mesenteric vein (S) is involved by tumor.
|
|
View larger version (105K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11C. —64-year-old man with T3 NX M1 pancreatic ductal
adenocarcinoma. Left posterior oblique image from ERCP shows common bile duct
narrowed by tumor and crossed by stent (arrowheads). Proximal common
bile duct (thick arrow) is dilated, and dilated pancreatic duct
(thin arrows) is partially opacified.
|
|
View larger version (82K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12. —72-year-old man with T4 NX M0 pancreatic ductal
adenocarcinoma. Coronal thick-slab (50-mm) image from MR
cholangiopancreatography shows double-duct sign caused by obstruction by
tumor. Dilated common bile duct (thick arrows) and dilated pancreatic
duct (thin arrows) are seen proximal to abrupt cutoff
(arrowheads).
|
|
Endoscopic sonography uses a high-frequency (7.5- to 12-MHz) sonographic
transducer that is introduced into the gastrointestinal tract via a
side-viewing endoscope. The operator is thereby provided real-time
cross-sectional images of the gastrointestinal wall and adjacent soft-tissue
structures. Studies that have compared endoscopic sonography and CT have shown
endoscopic sonography to be more sensitive for the detection of pancreatic and
periampullary tumors, particularly small lesions (< 20 mm), with a
sensitivity of 93–100%
[52,
53,
54,
55,
56]. Fine-needle aspiration of
suspected pancreatic lesions can then be performed with real-time imaging
guidance, often in a safer manner than by percutaneous CT or transabdominal
sonographically guided biopsy. The use of fine-needle aspiration reportedly
increases specificity for tumor—in some series, to as high as 100%
[56,
57].
Positron emission tomography (PET) traditionally uses FDG labeled with 18F.
Images are obtained anywhere between 40 and 180 min after injection
[58,
59]. Attenuation correction of
the emission scan is performed, and standardized uptake values may be
determined (Figs. 13A,
13B).
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13A. —48-year-old woman with T3 N0 M0 pancreatic ductal
adenocarcinoma. Positron emission tomography (PET) scan obtained in coronal
plane shows activity in pancreatic head (arrow) and in right lobe of
thyroid gland (arrowhead). Sonogram of thyroid (not shown) revealed
multinodular goiter.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13B. —48-year-old woman with T3 N0 M0 pancreatic ductal
adenocarcinoma. Follow-up PET scan obtained 3 months later than A,
after patient had undergone chemotherapy and radiation therapy, no longer
shows pancreatic head activity. Thyroid gland activity (arrowhead)
persists.
|
|
Imaging Findings
Primary Tumor
Approximately 60% of pancreatic cancers originate in the pancreatic head,
15% in the body, and 5% in the tail; approximately 20% of pancreatic cancers
diffusely involve the pancreas
[60]. A tumor typically
appears hypodense in comparison with the normal pancreatic parenchyma on
contrast-enhanced CT and hypointense, on T1-weighted unenhanced, dynamic, and
contrast-enhanced MR imaging sequences. Pancreatic tumor has a variable
appearance on T2-weighted MR imaging.
Unfortunately, tumor can appear similar to normal parenchyma on delayed
contrast-enhanced images for both modalities. Pitfalls include tumors that are
small and isodense or isointense to normal pancreatic parenchyma on all phases
of contrast enhancement, small uncinate tumors that may not obstruct the
pancreatic duct or the common bile duct, and chronic pancreatitis
(Fig. 14). In their evaluation
of a small series of patients with chronic pancreatitis who presented for
evaluation for a pancreatic mass, Kim et al.
[61] concluded that chronic
pancreatitis can show many of the features of pancreatic ductal adenocarcinoma
on both CT and MR imaging, including having the appearance of a focal mass,
appearing isodense (isointense on all sequences) or hypodense (hypointense on
T1-weighted MR images or gadolinium-enhanced sequences) to the remaining
pancreatic parenchyma, proximal pancreatic duct dilatation, and atrophy of the
proximal pancreas.
View larger version (102K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 14. —Contrast-enhanced axial CT image shows low-attenuation mass
(arrow) in pancreatic head in 45-year-old man. Endoscopic
sonographically guided fine-needle aspiration biopsy (not shown) revealed
chronic pancreatitis.
|
|
ERCP and MR cholangiopancreatography visualize the effect of tumor on the
bile and pancreatic ducts. Typically, tumor encases, narrows, and can
completely obstruct these ducts. However, approximately 20% of patients with
pancreatic cancer have a normal caliber pancreatic duct
[62], particularly those who
have small uncinate tumors.
The typical features of pancreatic cancer seen by endoscopic sonography
include an in-homogeneous solid mass with irregular borders that appears
hypoechoic to normal pancreatic parenchyma
[63]. However, a study of 115
patients found that using morphologic features alone in endoscopic sonography
yielded a specificity of only 53%
[63]. Entities in this study
that were mistaken for pancreatic cancer included focal changes from chronic
pancreatitis, neuroendocrine tumors, and metastases to the pancreas. However,
endoscopic sonography offers an excellent means to biopsy pancreatic lesions:
specificities as high as 100% have been reported
[57].
Endoscopic sonography–guided fine-needle aspiration is dependent on
the skill and experience of the user in the performance of both endoscopic
sonography and the biopsy itself, and it may not be possible to obtain
sufficient biopsy material
[55,
57]. In a study of 166
patients, 162 of whom underwent endoscopic sonography–guided biopsy of
the pancreas, Shin et al. [57]
reported that the false-negative rate, which was attributed to inadequate
sampling, was 13%, decreasing sensitivity to 81.7% when those patients with
nondiagnostic aspirates or inadequate cellularity were included. Harewood et
al. [64] studied the impact of
training and histopathologic interpretation on the success of endoscopic
sonography–guided fine-needle aspiration biopsy techniques for
evaluation of the pancreas. Those investigators noted that significant
improvements in accuracy were achievable with mentored training, and that
errors during the initial phase primarily resulted from inadequate specimens
[64].
On PET, pancreatic cancer, like most cancers, shows a region of intense,
typically welldefined, FDG radiotracer uptake. Changes from pancreatitis can
mimic tumor, and false-negative findings can occur in patients with
hyperglycemia or small, well-differentiated tumors
[65,
66]. Specific uptake values
have been used to help discriminate between tumor and inflammation. Opinions
on how specific uptake values should be calculated and on the criteria to use
in assessing specific uptake values vary among researchers. A recent
literature review by Zimny and Schumpelick
[65] reported that PET has
sensitivities of 71–100%, and specificities of 64–100% with a
median specificity of 82%. Much of this difference in findings could be
attributed to differences in technique, populations, study sizes, used in the
various studies. A study by Nitzsche et al.
[67] showed 100% accuracy when
FDG kinetics were used to discriminate inflammation from tumor; however, an
earlier study by Nakamoto [68]
that used a similar characteristic showed a great overlap between these two
entities. Although a dominant technique has yet to emerge, newer techniques
with improving accuracies may be useful in assessing early treatment response
and unexpected sites of metastases.
Vascular Invasion
Techniques for determining whether a tumor is locally advanced (i.e., local
extension of tumor to adjacent organs such as the stomach, colon, spleen, and
vascular structures) include CT, MR imaging, and endoscopic sonography. CT
criteria that have been developed to indicate the probability of vascular
involvement use the relationship of tumor to adjacent vessels
[67,
69,
70]
(Table 3 and Figs.
2B,
7A,
15,
16). Typical reports in the
literature regarding the accuracy of single-detector helical CT for predicting
vascular invasion range from 62% to 92%
[44,
45,
46,
47,
71,
72,
73]. MR imaging can provide
similar information regarding vascular invasion. A recent study of 48 patients
that compared MR imaging, MR imaging with MR angiography, and dual-phase CT
found comparable accuracies (87%, 90%, and 90%, respectively), sensitivities,
and specificities for the different modalities
[74]. Endoscopic sonography
offers real-time imaging of vascular structures. A variety of criteria have
been suggested for determining whether vascular invasion is present on
endoscopic sonography. A study that looked specifically at the accuracy of
various signs for tumor involvement of vascular structures showed the
following accuracies: proximity of mass to vessel (73%), loss of interface
between mass and vessel (78%), and irregularity of the venous wall (87%)
[75]. Although the last
criterion showed the highest accuracy in this study, it had a very low
sensitivity (47%) because of the limited ability of endoscopic sonography to
detect superior mesenteric vein invasion (sensitivity, 17%)
[75]. A study by Tio et al.
[76] reported an overall
accuracy of 83.6% for endoscopic sonography for local tumor staging (tumor
size, invasion of local organs, invasion of major vessels); limitations
included the inability to detect microinvasion of the splenoportal confluence
and mesocolon (5/52), and misinterpretation of changes from pancreatitis as
possible involvement of the splenoportal confluence (5/52).
View larger version (138K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 15. —68-year-old man with history of T3 N1 M0 pancreatic ductal
adenocarcinoma. Pancreatic tumor (thin white arrows) is separated
from superior mesenteric vein (V) by normal pancreatic parenchyma (type B
relationship). Tumor (thick white arrow) has convex point of contact
with superior mesenteric artery (A) in type C relationship. Stent is also seen
(black arrow). Given this relationship, tumor involvement of superior
mesenteric artery cannot be reliably predicted. At time of surgery, superior
mesenteric artery was found not to be involved by tumor.
|
|
View larger version (155K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 16. —76-year-old man with history of T3 NX M0 pancreatic ductal
adenocarcinoma. Contrast-enhanced axial CT image shows type F relationship of
tumor to vasculature. Portal vein (medium-sized arrow) is thrombosed.
On more inferior axial images (not shown) portal and superior mesenteric veins
were occluded by tumor. Numerous collaterals (small arrow) are seen
between pancreatic head and duodenum. Although this image shows stranding
surrounding common hepatic artery (large arrow), with anomalous
origin from the aorta, artery is not definitely involved by tumor.
|
|
Metastases
Unfortunately, 60% of patients who present with pancreatic ductal
adenocarcinoma have advanced disease
[27]. CT has a reported
sensitivity of 75–87% for liver metastases
[44,
73,
77]. In a study comparing MR
imaging with helical CT, MR imaging had an accuracy of 93.5% for detection of
liver metastases compared with 87% for CT
[73]. PET has been reported to
have a sensitivity of 70% and a specificity of 95% for liver metastases, with
sensitivity decreasing as lesion size decreased
[66].
The identification of nodal and peritoneal disease is difficult with all
imaging modalities. On cross-sectional imaging, size (> 1 cm) is the
criterion for identifying nodal metastases. The accuracy of both CT and MR
imaging is limited [45,
78]; PET could potentially
provide greater specificity in the evaluation of lymph nodes, but it may also
be of limited use in evaluating small metastases to lymph nodes of normal
size. For this reason, laparoscopic surgical evaluation, with or without
laparoscopic sonography, has been suggested to increase the accuracy of
staging by improving detection of peritoneal, nodal and liver metastases.
Recent studies have confirmed an increased sensitivity of laparoscopic
evaluation compared with CT for detection of peritoneal metastases
[79,
80]. Jimenez et al.
[81] reported on their
institution's experience from 1994 to 1998 with 125 patients whose tumors were
identified as resectable on CT; on laparoscopic evaluation, 25% of these
patients were found to have liver or peritoneal metastases. However, a recent
review of the surgical literature by Pisters et al.
[82] concluded that as few as
4–13% of patients with tumors judged resectable on high-quality CT may
actually benefit from laparoscopic evaluation. Pisters et al. noted that
several studies of laparoscopic staging had included patients with M0 staging
but locally advanced disease or those who had CT of limited quality compared
with current capabilities. The review by Pisters et al. also noted that
because laparoscopic sonography did improve sensitivity for metastatic disease
that it might be of use in patients found to have marginally resectable
disease on CT. However, they concluded that, given the limitations of several
studies of this technique, larger studies were necessary, and stated that at
the current time "laparoscopic ultrasonography cannot be considered as a
recommended component of the laparoscopic staging of pancreatic cancer"
[82].
Role of Imaging
Assessing Treatment Response
Imaging has an important role in assessing treatment response. This is
particularly true for patients who are undergoing neoadjuvant therapy for
presumed resectable disease. If disease progresses, patients may be started on
a different treatment protocol. Imaging can also reveal whether disease has
progressed to the point that it is no longer resectable. Disease response can
be depicted by CT, MR imaging, or PET (Figs.
13A,
13B and
17A,
17B). CT is the modality most
typically used at our institution because of its usefulness in detecting
distant metastases and in assessing the response of local disease to
therapy.
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 17A. —63-year-old man with history of T4 NX M0 pancreatic ductal
adenocarcinoma confirmed by biopsy. Contrast-enhanced axial CT image shows
large mass (thick arrow) in pancreatic head involving superior
mesenteric artery (short thin arrow) and superior mesenteric vein
(arrowhead). Stent (long thin arrow) is present in common
bile duct.
|
|
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 17B. —63-year-old man with history of T4 NX M0 pancreatic ductal
adenocarcinoma confirmed by biopsy. Contrast-enhanced axial CT image obtained
6 months later than A, after patient had undergone treatment with
radiation therapy and gemcitabine, shows residual soft tissue representing
site of tumor (large arrows). Tumor is posterior relative to both
superior mesenteric artery and superior mesenteric vein (small
arrows).
|
|
Detecting Recurrence
The reasons for imaging the patient who has undergone
treatment—whether surgery or treatment with radiation therapy and
chemotherapy—are to detect postoperative complications, local recurrence
(Fig. 18), local disease
progression (Figs. 19A,
19B), or distant metastases.
Sites of local recurrence after surgery depend on the original site of tumor
within the pancreas and the likely nodal drainage routes. Because local
changes as a result of therapy (Figs.
17A,
17B) can be indistinguishable
from recurrent tumor, it is important that close follow-up imaging be obtained
and that each study be compared with an initial baseline postoperative
examination when available. We advocate imaging during the portal venous
phase, to better detect liver metastases, and the use of thin-section (2.5-mm)
imaging to enhance detection of recurrent disease in the surgical bed. The
role of PET in the evaluation of possible sites of recurrence has yet to be
determined. A small study of 20 patients, including patients who had undergone
the Whipple procedure, suggested that PET may be useful in clarifying
equivocal CT examinations for recurrent disease
[83].
View larger version (140K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 18. —50-year-old woman with history of T3 N1 M0 pancreatic ductal
adenocarcinoma. Patient had undergone pancreaticoduodenectomy 15 months
earlier for ductal adenocarcinoma. Contrast-enhanced axial CT image shows
tumor (medium-length arrows) has recurred near superior mesenteric
artery (large arrow). Surgical clip (small arrow) is
seen.
|
|
View larger version (154K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 19A. —77-year-old woman with history of T4 N0 M0 pancreatic ductal
adenocarcinoma that has been treated with chemotherapy and radiation therapy.
Surveillance imaging obtained during a 6-year period revealed stable residual
disease involving superior mesenteric artery. Contrast-enhanced axial CT image
obtained 7 years after initial diagnosis shows new irregular narrowing of
superior mesenteric artery (thick arrow) by tumor (thin
arrows).
|
|
View larger version (154K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 19B. —77-year-old woman with history of T4 N0 M0 pancreatic ductal
adenocarcinoma that has been treated with chemotherapy and radiation therapy.
Surveillance imaging obtained during a 6-year period revealed stable residual
disease involving superior mesenteric artery. Contrast-enhanced axial CT image
acquired at level superior relative to A shows new fistula
(medium-length arrows) between stomach (thick arrow) and
recurrent tumor (small arrow).
|
|
References
- Jemal A, Thomas A, Murray T, Thun M. Cancer statistics, 2002.
CA Cancer J Clin
2002;52:23
–47[Abstract/Free Full Text]
- Simon B, Printz H. Epidemiological trends in pancreatic neoplasias.
Dig Dis
2001;19:6
–14[Medline]
- Parkin DM, Muir CS. Cancer incidence in five continents:
comparability and quality of data. IARC Sci Publ
1992;120:45
–173
- Weiderpass E, Partanen T, Kaaks R, et al. Occurrence, trends and
environment etiology of pancreatic cancer. Scand J Work Environ
Health 1998;24:165
–174[Medline]
- Lowenfels AB, Maisonneuve P, Cavallini G, et al. Pancreatitis and
the risk of pancreatic cancer: International Pancreatitis Study Group.
N Engl J Med
1993;328:1433
–1437[Abstract/Free Full Text]
- Gold EB, Goldin SB. Epidemiology of and risk factors for pancreatic
cancer. Surg Oncol Clin N Am
1998;7:67
–91[Medline]
- Klein AP, Hruban RH, Brune KA, Petersen GM, Goggins M. Familial
pancreatic cancer. Cancer J
2001;7:266
–273[Medline]
- Almoguera C, Shibata D, Forrester K, Martin J, Arnheim N, Perucho
M. Most human carcinomas of the exocrine pancreas contain mutant c-K-rass
genes. Cell
1988;53:549
–554[Medline]
- Day JD, Digiuseppe JA, Yeo C, et al. Immunohistochemical evaluation
of HER-2/neu expression in pancreatic adenocarcinoma and pancreatic
intraepithelial neoplasms. Hum Pathol
1996;27:119
–124[Medline]
- Ghadimi BM, Schrock E, Walker RL, et al. Specific chromosomal
aberrations and amplification of the AIB1 nuclear receptor coactivator gene in
pancreatic carcinomas. Am J Pathol
1999;154:525
–536[Abstract/Free Full Text]
- Poremba C, Yandell DW, Huang Q, et al. Frequency and spectrum of
p53 mutations in gastric cancer: a molecular genetic and immunohistochemical
study. Virchows Arch
1995;426:447
–455[Medline]
- Redston MS, Caldas C, Seymour AB, et al. p53 mutations in
pancreatic carcinoma and evidence of common involvement of homocopolymer
tracts in DNA microdeletions. Cancer Res
1994;54:3025
–3033[Abstract/Free Full Text]
- Rozenblum E, Schutte M, Goggins M, et al. Tumor-suppressive
pathways in pancreatic carcinoma. Cancer Res
1997;57:1731
–1734[Abstract/Free Full Text]
- Kloppel G, Hruban RH, Longnecker DS. Ductal adenocarcinoma of the
pancreas. In: Hamilton SR, Aaltonen LA, eds. Pathology and
genetics: tumours of the digestive system. Lyon: IARC,
2000: 221–230
- Bommer G, Friedel U, Heitz PU, Koppel G. Pancreatic PP cell
distribution and hyperplasia: immunocytochemical morphology in normal
pancreas, chronic pancreatitis, and pancreatic carcinoma. Virchows
Arch A Pathol Anat Histopathol
1980;387:319
–331
- Hoorens A, Lemoine NR, McLellan E, et al. Pancreatic acinar cell
carcinoma. An analysis of cell lineage markers, p53 expression, and Ki-ras
mutation. Am J Pathol
1993;143:685
–698[Abstract]
- Niederau C, Grendell JH. Diagnosis of pancreatic carcinoma: imaging
techniques and tumor markers. Pancreas
1992;7:66
–86[Medline]
- Steinberg W. The clinical utility of the CA 19-9 tumor-associated
antigen. Am J Gastroenterol
1990;85:350
–355[Medline]
- Lundin J, Roberts PJ, Kuusela P, Haglund C. Prognostic significance
of serum CA 242 in pancreatic cancer. A comparison with CA 19-9.
Anticancer Res
1995;15:2181
–2186[Medline]
- Iwao T, Hiyama E, Yok T, et al. Telomerase activity is detected in
pancreatic cancer but not in benign tumors. J Natl Cancer
Inst 1997;89:1621
–1623[Free Full Text]
- Sobin LH, Wittekind C, eds. TNM classification of
malignant tumours, 6th ed. Baltimore: Wiley-Liss,
2002
- [No authors listed] Exocrine pancreas. In: Freene FL, Page DL,
Felming ID, Fritz AG, Balch CM, Haller DG, Morrow M. AJCC cancer
staging manual, 6th ed. New York: Springer-Verlag,
2002: 157–164
- Evans DB, Lee JE, Pisters PWT. Pancreaticoduodenectomy (Whipple
operation) and total pancreatectomy for cancer. In: Baker RJ, Fischer JF, eds.
Mastery of Surgery, 4th ed. Philadelphia: Lippincott
Williams & Wilkins, 2001:1299
–1318
- Traverso LW, Longmire WP. Preservation of the pylorus in
pancreaticoduodenectomy. Surg Gynecol Obstet
1978;146:959
–962[Medline]
- Trede M, Schwall G, Saeger HD. Survival after
pancreatoduodenectomy. Ann Surg
1990;211:447
–458[Medline]
- Lowenfels AB, Maisonneuve P, Boyle P. Epidemiology of pancreatic
cancer. In: Howard J, Idezuki Y, Ihse I, et al., eds. Surgical
disease of the pancreas, 3rd ed. Baltimore: Williams and Wilkins,
1998: 433–437
- Douglass HJ, Kim S, Meropol N. Neoplasms of the exocrine pancreas.
In: Holland J, Frei EI, Bast RJ, Kufe DW, Morton DL, Weichselbaum RR, eds.
Cancer medicine, 4th ed. Baltimore: Williams and
Wilkins, 1997:1989
–2018
- Evans DB, Abbruzzese JL, Willett CG. Cancer of the pancreas. In:
DeVita VT, Hellman S, Rosenberg SA, eds. Cancer: Principles and
Practice of Oncology, 6th ed. Philadelphia: Lippincott Williams
& Wilkins, 2001:1126
–1161
- Bold RJ, Charnsangavej C, Cleary KR, et al. Major vascular
resection as part of pancreaticoduodenectomy for cancer: radiologic,
intraoperative, and pathologic analysis. J Gastrointest
Surg 1999;3:233
–243[Medline]
- Harrison LE, Klimstra D, Brennan MF. Isolated portal vein
involvement in pancreatic adenocarcinoma: a contraindication for resection?
Ann Surg
1996;224:342
–349[Medline]
- Kalser MH, Ellenberg SS. Pancreatic cancer: adjuvant combined
radiation and chemotherapy following curative resection. Arch
Surg 1985;120:899
–903[Abstract]
- Neoptolemos JP, Dunn JA, Stocken DD, et al. Adjuvant
chemoradiotherapy and chemotherapy in resectable pancreatic cancer: a
randomised controlled trial. Lancet
2001;358:1576
–1585[Medline]
- Burris HA, 3rd, Moore MJ, Andersen J, et al. Improvements in
survival and clinical benefit with gemcitabine as first-line therapy for
patients with advanced pancreas cancer: a randomized trial. J Clin
Oncol 1997;15:2403
–2413[Abstract/Free Full Text]
- Di Costanzo F, Sdrobolini A, Carlini P, et al. Gemcitabine (GEM)
alone or in combination with 5-FU continuous infusion (CI) in the treatment of
advanced pancreatic cancer (APC): a GOIRC randomized phase II trial.
37th Annual scientific meeting of Clinical Oncology.
San Francisco: Society of Clinical Oncology, 2001:154a
- Fung MC, Sakata T. What's new in pancreatic cancer treatment?
J Hepatobiliary Pancreat Surg
2002;9:61
–75[Medline]
- Sasson AR, Wetherington RW, Hoffman JP, et al. Neoadjuvant therapy
for adenocarcinoma of the pancreas: histopathologic effect and analysis of
outcome. 37th Annual scientific meeting of Clinical
Oncology. San Francisco: Society of Clinical Oncology,
2001:156a
- Hietaniemi K, Karkkainen P, Kiviluoto T, Kivisaari L, Elomaa I,
Joensuu T. Phase I–II trial of twice-weekly gemcitabine (GEM) and
concomitant irradiation in patients (Pts) with locally advanced pancreatic
cancer. 37th Annual scientific meeting of Clinical
Oncology. San Francisco: Society of Clinical Oncology,
2001: 142b
- Wolff RA, Chiao P, Lenzi R, et al. Current approaches and future
strategies for pancreatic carcin
Top
Introduction
Pathology
