Editor: Ian Penman
3. EUS for staging gastrointestinal and pancreatic cancer
Thomas J. Savides
Endoscopic ultrasonography (EUS) has gained a pivotal role in cancer staging for both gastrointestinal and bronchogenic or
mediastinal malignancies, because of its high resolution imaging and ability to sample lesions that are often difficult to
access by other methods. As advances in radiologic imaging by multidetector CT, MRI, and positron emission tomography (PET)
scanning develop the role of EUS constantly needs to be evaluated, but at present it provides important complementary staging
information in many patients with these malignancies. The rationale for, and roles of EUS for cancer staging are discussed
in this chapter.
EUS for cancer staging
Accurate cancer staging is needed because it:
- provides prognostic information;
- guides treatment;
- allows researchers to compare similar patient populations.
The most commonly accepted reference sources for cancer staging are the American Joint Commission on Cancer (AJCC) Cancer
Staging Manual  or the very similar International Union Against Cancer (UICC) TNM Classification of Malignant Tumours . These use the tumornodemetastasis (TNM) staging system and are revised regularly to reflect advances in clinically relevant staging.
Endoscopic ultrasound (EUS is ideally suited for local staging of luminal gastrointestinal tract cancers and pancreatic cancer,
because of its ability to image both the layers of the intestinal wall and peri-intestinal lymph nodes and organs. EUS should
be used after CT or MRI scan has shown no distant metastatic disease. Additionally, EUS should only be used if the information
it provides will have an impact on patient care. Usually this requires close communication with the surgeon and medical oncologist
caring for the patient, to determine what information will impact management. EUS staging accuracy is similar among all luminal
GI tract and pancreatic cancers, with an approximate T-stage accuracy of 85% and N-stage accuracy of 75%.
Esophageal cancer staging with EUS
Approximately 14 250 cases of esophageal cancer are diagnosed annually in the US, of which 13 300 die of their disease . Survival for esophageal cancer depends on cancer stage (Fig. 1), but overall 5-year survival is only 1020%.
Staging is generally first performed using CT scan to look for distant metastatic disease. EUS is performed after there is
no obvious evidence of distant metastases. Positron emission tomography (PET scan is also being increasingly used to evaluate
for distant metastatic disease. Laparoscopy can also be used to evaluate intra-abdominal metastatic disease, but is an invasive
Esophageal cancer TNM staging
The TNM staging classification and stage groupings for esophageal cancer are shown in Fig. 2. There is no serosal lining of the oesophagus, and therefore invasion occurs into peri-esophageal fat. Figure 3 shows how the regional lymph nodes for esophageal cancer are defined by the anatomic location of the tumor. Metastatic disease
is separated into M1a or M1b based on anatomic location of the esophageal cancer, with M1a disease associated with cervical
lymph nodes for upper thoracic tumors, and celiac lymph nodes for distal esophageal cancer (Fig. 4), and M1b being any other distant metastases. For cancers of the gastroesophageal junction, lymph nodes in the celiac and
left gastric nodal sites are considered regional N1 disease, and not metastatic disease.
More advanced T-stage disease is associated with greater risk of malignant lymph nodes, with 77% of patients with T3 disease
having N1 disease (Fig. 5). Most patients with dysphagia have T3 disease Figs 6 and 7).
It has been suggested that the current esophageal cancer staging system be modified to more precisely reflect prognosis for
survival, by dividing stage T1 into T1a (limited to mucosa) and T1b (involves into the submucosa), and also to define lymph
node stage based on the number of malignant nodes, with N0, N1 (12 LNs), N2 (>= 3 LNs) rather than anatomic location of the nodes [4,5]. To date this has not been validated or agreed internationally.
Technique for performing EUS staging of esophageal cancer
Upper endoscopy with a forward-viewing endoscope is usually performed first to localize the tumor precisely and to determine
whether dilation is needed prior to passing the echoendoscope. Once the echoendoscope is passed into the stomach, ultrasound
imaging is performed on withdrawal of the scope. Specific attention is given to determining the extent of tumor wall invasion,
aortic involvement, peri-tumor lymph nodes, celiac lymph nodes, and liver metastases.
EUS of stenotic esophageal tumors
Approximately one-third of tumors will have esophageal narrowing which prevents passage of the echoendoscope. The reason for
passing the scope distal to the stricture is to evaluate for celiac lymphadenopathy, as well as staging the remainder of the
tumor. Early studies using EUS to stage esophageal cancer found up to a 24% risk of esophageal perforation if esophageal dilation
was performed prior to EUS . However, this may have been related to the larger diameter and less tapered tips of the original EUS scopes. More recent
studies suggest careful dilation may be safe. When dilation is performed, it is usually performed using either through-the-scope
balloons or Savary dilators, the goal being to dilate gently the esophageal stricture up to approximately 45 Fr to allow passage
of the echoendoscope. Great care must still be taken when dilating esophageal cancers to prevent perforation. If the echoendoscope
cannot pass after dilation, then EUS can still partially assess the esophageal tumor, by either wedging the echoendoscope
into the tumor, by using a thin wire-guided esophageal echoendoscope that does not have fiberoptic or videoendoscopy capability,
or by using a catheter ultrasound probe passed through the scope.
EUS evaluation of superficial tumors
Standard echoendoscopes (7.5 MHz, 12 MHz) can be used to evaluate superficial esophageal tumors (Fig. 8), but the low frequency may limit resolution. Higher frequency catheter probes (12, 20, and 30 MHz) provide better resolution to define mucosal vs. submucosal involvement (Fig. 7). However, the greater the frequency the less the depth of penetration, which limits visualization of adjacent lymph nodes
and sometimes the true depth of penetration of the tumor.
EUS evaluation of lymph nodes
The general EUS criteria for malignant lymph nodes include a short axis size > 10 mm, round shape, sharply demarcated borders, and echo-poor (hypoechoic) sonographic features (Fig. 9). If all four of these criteria are met, the sensitivity is 80100%, however, only 25% of malignant lymph nodes have all these criteria [7,8]. Additionally, many endosonographers consider any round hypoechoic lymph node immediately adjacent to a tumor to be malignant.
The number of nodes identified at EUS also correlates with prognosis and should be documented.
EUS-FNA of peri-esophageal lymph nodes
Trans-esophageal or transgastric EUS fine-needle aspiration (FNA) (Fig. 10) can be used to biopsy any malignant-appearing lymph nodes. Transintestinal FNA is not performed if the needle would need
to pass through luminal tumor first, in order to avoid false positive EUS-FNA cytology because of sampling of the primary
tumor, and also to avoid inadvertent seeding of a benign lymph node with tumor. Another problem with EUS-FNA of lymph nodes
is that a negative FNA cytology could be sampling error. The greatest utility of EUS-FNA is for diagnosing celiac lymph nodes,
which might change managementceliac nodal involvement is staged as M1a disease and signals unresectability and a poor prognosis in esophageal cancer.
Accuracy and limitations of EUS staging of esophageal cancer
The overall staging accuracy of esophageal cancer for T-stage is 85% and for N-stage 75% using diagnostic EUS . The addition of EUS-FNA slightly increases the N-stage accuracy to 85% .
Limitations of T-stage accuracy result from overstaging or understaging. EUS tends to overstage tumors more often than understage.
This is because EUS images cannot distinguish tumor from peri-tumoral inflammation. Distortion of the imaging plane due to
tumor compression can also result in tangential imaging which causes the lesion to appear to have greater depth of invasion
than it does. This problem is mostly seen in staging pathologic T1 and T2 tumors. Understaging may occur if there is microscopic
tumor invasion, or if there is an esophageal stricture that prevents visualization of more advanced disease.
Limitations of N-stage accuracy for EUS include limitations of the morphologic appearance of the lymph nodes. Additionally,
if there is a stricture, peri-esophageal or celiac lymph nodes distal to the stricture may not be visualized. Limitations
of N-stage accuracy for EUS-FNA staging are that a negative FNA cytology does not completely exclude malignancy. Additionally,
there is the possibility of false positive EUS-FNA cytology, as one well-designed study reported a false positive rate of
7% for EUS-FNA cytology of peri-esophageal cancer lymph nodes .
EUS re-staging of esophageal cancer after chemoradiation
The accuracy of EUS staging after chemoradiation of esophageal cancer is markedly decreased because of inflammation and fibrosis,
with an accuracy rate of only 40% [11,12]. The utility of re-staging cancer after chemoradiation depends on whether it can give prognostic information, and more importantly,
whether the oncologist and surgeon managing the patient will use this information to decide further treatment. Upper endoscopy
alone, however, may be the most useful means of re-staging patients, because this will define which 25% of patients who have
undergone preoperative chemoradiation have had a complete pathologic response . There are some preliminary EUS studies which suggest that a reduction in maximal cross-sectional area of the tumor is more
reliable than standard TNM determination by radial EUS, but it is presently unknown if this will translate into altered clinical
care [11,13]. Future studies evaluating 3-D tumor volumes by EUS are awaited.
A very promising possible use of EUS for re-staging esophageal cancer involves patients with esophageal adenocarcinoma who
have a malignant celiac lymph node. Some expert esophageal cancer surgeons have proposed that if repeat EUS-FNA of the celiac
node does not reveal any residual cancer, then the patient may proceed on to surgical resection, but if there is any residual
tumor, then surgery should not be undertaken because it would not improve survival.
Impact of EUS staging on esophageal cancer management
The impact of EUS staging of esophageal cancer depends somewhat on local institutional biases. The exact role of preoperative
chemoradiation is still debated, although it is commonly given for suspected stage T3 or N1 tumors. Tumors with adjacent organ
invasion (T4) (Fig. 7) or distant metastases are generally considered unresectable. Figure 12 shows the usual impact of EUS staging of esophageal cancer on patient management.
Gastric cancer staging with EUS
Approximately 90% of all gastric neoplasms are adenocarcinomas [1,2]. The overall 5-year survival is approximately 1520%, and increases to approximately 55% if the tumor is limited to the stomach [1,2]. In the USA in 2004 there were approximately 22 710 new cases diagnosed and 11 780 deaths from gastric cancer .
Gastric cancer TNM staging
TNM staging for gastric cancer differs slightly from esophageal and rectal luminal tumors, in that stage T2 is subdivided,
and the number of regional lymph nodes is assessed (rather than their anatomic location). The TNM staging system and the tumor stage groupings
for gastric cancer are shown in Fig. 13. Note that early stage gastric cancers can have malignant regional lymph nodes.
EUS staging of advanced gastric adenocarcinoma
Most patients with gastric cancer in Western countries present with symptoms of abdominal pain, gastric obstruction, or bleeding.
These tumors are generally bulky and locally advanced at the time of diagnosis. Because most patients with advanced gastric
adenocarcinoma will undergo surgical resection without preoperative chemoradiation for palliation, the main role of EUS is
to identify patients with tumors that invade into adjacent organs (Fig. 14) and who would not be operative candidates. Similarly, EUS can identify small volume ascites not detectable by other methods
and which can be sampled by EUS-FNA (Fig. 15).
EUS staging of early gastric adenocarcinoma
Early gastric cancer is most commonly detected during endoscopic screening programs in high risk parts of the world such as
Asia. Lesions which are limited to the mucosal layer are amenable to attempted endoscopic mucosal resection (EMR). It is quite
useful to perform EUS before attempting EMR in order to make sure there is no submucosal invasion, and to make sure there
are no malignant-appearing peri-gastric lymph nodes, which would preclude attempted endoscopic resection. Imaging with higher
frequency 20 MHz or 30 MHz catheter probes may be more useful for determining submucosal invasion than a dedicated 7.5/12 MHz echoendoscope. EUS has a T-stage accuracy of 8090% for determination of early gastric cancer .
EUS staging of gastric MALT lymphoma
Mucosal-associated lymphoid tissue (MALT) lymphoma of the stomach is unique in that it is almost always associated with Helicobacter pylori infection. Antibiotic treatment to eradicate H. pylori infection can lead to regression of the MALT lymphoma. EUS is useful in that patients with MALT limited to the mucosal/submucosal layers are likely to have complete response to antibiotic therapy, while those with invasion deeper than the submucosa
are much less likely to respond to antibiotics [15,16].
Rectal cancer staging with EUS
There are 41 000 new cases of rectal cancer diagnosed in the United States each year . Management of rectal cancer differs from colon cancer in that accurate preoperative staging determines the type of surgical
approach to the rectal tumors and possible preoperative chemoradiation.
Rectal cancer TNM staging
The TNM classification for rectal cancer is shown in Fig. 16. Note that N-status depends on the number of malignant regional lymph nodes. Also note that iliac nodes are considered regional
lymph nodes in rectal cancer.
Pathologic staging of rectal cancer
Most rectal cancers are at an advanced stage at diagnosis, with 75% of patients with rectal cancer presenting with stage T3
or greater, and/or N1 disease . Lymph node metastases increase with T-stage, with approximate prevalence rates for T1 = 10%, T2 = 25%, and T3/4 = 50% [18,19].
Surgical management of rectal cancer
Rectal cancer is managed differently to colonic cancer because of the limited ability of surgery to achieve clear distal margins due to the end of the rectum, and the limited ability to obtain clear radial margins because of space restraints in the pelvis.
Tumors more than 5 cm from the dentate line can be removed with a transabdominal low anterior resection, which removes the tumor, provides good
radial and longitudinal margins, and allows for lymph node sampling. Tumors located within 5 cm of the dentate line usually require a complete abdominoperineal resection (APR) with permanent colostomy in order to obtain
radial and longitudinal margins, as well as lymph node sampling. Transanal excision can be attempted for tumors located within
5 cm of the dentate line, and which are suspected to be pathologic stage T1, N0 or superficial T2, N0, in an effort to avoid
APR and colostomy.
The main problem with traditional rectal cancer surgery is local recurrence, with rates of up to 30% reported. Recurrence
rates have been reported to decrease to 20% after chemoradiation, 10% after either total mesorectal excision or preoperative
radiation and traditional surgery, and 2% after preoperative radiation and total mesorectal excision .
Management algorithm for rectal cancer (Fig. 17)
A general algorithm for management of rectal cancer is that if the patient is suspected to have pathologic T3 and/or N1 stage tumor, then preoperative chemoradiation is given, followed by surgery. If the lesion is staged as a T1 tumor,
then either transanal or low anterior resection is performed. If the tumor is stage T2, therapy is individualized in terms
of either preoperative chemoradiation or proceeding directly to surgery.
Technique for performing EUS rectal cancer staging
Local staging with rectal EUS is usually performed after CT scan has excluded distant metastatic disease. Ultrasound staging
can be performed either using a rigid ultrasound probe or a flexible echoendoscope. The advantage of the flexible echoendoscopes
is direct endoscopic visualization of the tumor, in addition to ultrasound visualization.
Patient preparation is usually either with two enemas administered a few hours before the procedure as is done for flexible
sigmoidoscopy, or possibly with full bowel preparation as for colonoscopy. Conscious sedation is generally not used, except
in the occasional situation where the patient has significant pain from the tumor. A standard video flexible sigmoidoscope
is generally used first in order to localize the lesion, assess the adequacy of the bowel prep, and to position the patient
such that the lesion is in a dependent position so that it can be covered with water for good transmission of the ultrasound
waves. The sigmoidoscope is withdrawn and replaced with the radial echoendoscope.
In order to assess T-stage, the scope is advanced to the tumor, water is instilled in the rectum, air in the lumen is removed
with suction, and the echoendoscope balloon is partially filled with water. The transducer is placed perpendicular to the
long axis of the tumor, and a few centimeters away from the tumor for the optimal focal length for the ultrasound. The scope
is moved across the length of the tumor, and attention focused on depth of wall invasion, invasion into peri-rectal fat, and
involvement of adjacent organs such as the bladder, prostate, seminal vesicles, vagina, or anal sphincters.
In order to assess N-stage, the scope is initially advanced under endoscopic guidance to approximately 25 cm from the dentate line. Water is instilled into the balloon, and air is removed with suction from the lumen. Initially attention
is focused on visualizing the iliac vessels, which are present at 25 cm, to look for any iliac nodes. The scope is then slowly withdrawn, constantly suctioning air from the lumen, in order to
evaluate for any peri-rectal lymph nodes.
EUS staging of rectal cancer
Tumors generally appear as hypoechoic masses. Often it is not possible to tell with certainty if a tumor invades a certain
level, or just abuts the level. Given this difficulty, it is acceptable to give a more descriptive staging, such as stating
that the tumor invades the muscularis propria and may barely invade into the peri-rectal fat (either T2 or superficial T3,
(Fig. 18). Malignant-appearing lymph nodes are generally round, hypoechoic, well-demarcated, and > 10 mm diameter. However, because peri-rectal lymph nodes are rarely seen in non-malignant transrectal ultrasound, even small
(< 10 mm) peri-tumoral lymph nodes are often considered likely to be malignant by EUS.
Accuracy of EUS in staging rectal cancer
Summaries of published literature on rectal cancer staging suggest that the T-stage accuracy is approximately 85%, and the
N-stage accuracy approximately 75% .
When EUS staging is incorrect for depth of penetration, it is generally because of overstaging rather than understaging. This
overstaging occurs because ultrasound cannot distinguish tumor from peri-tumor inflammation, and because of tangential imaging
that can increase the appearance of a tumor. Overstaging is most common for stage T2 tumors, which can appear by EUS to be
T3 tumors. Lymph node accuracy is limited by the ability to distinguish malignant from inflammatory lymph nodes.
EUS vs. CT and MRI for rectal cancer staging
Most comparative studies show a greater T-stage and N-stage accuracy for EUS vs. CT scan . EUS and MRI are more similar in staging accuracy.
A costeffectiveness analysis of three different staging strategies (abdominal and pelvic CT vs. abdominal CT plus EUS vs. abdominal
CT plus pelvic MRI) found that abdominal CT plus EUS is the most cost-effective approach for non-metastatic proximal rectal
EUS/FNA for rectal cancer lymph node staging
It is possible that transrectal EUS-guided FNA of peri-rectal lymph nodes could improve nodal staging accuracy. However, most
peri-rectal lymph nodes are usually adjacent to the primary tumor, which precludes EUS-FNA because the needle would traverse
the primary tumor to reach the node, which risks either a false positive cytology or possibly seeding a benign node with tumor.
Additionally, most malignant peri-rectal lymph nodes are associated with stage T3 or greater stage, and so if there is T3
disease then it is likely the visualized nodes are malignant. One prospective study did not find any increase in sensitivity
or change in management with the addition of EUS-FNA compared to EUS alone for diagnosing malignant peri-rectal lymph nodes
. EUS-FNA of peri-rectal lymph nodes may be important in occasional cases where the T-stage is T1 or T2, but there are lymph
nodes located in a position which can undergo FNA without traversing the primary tumor (Fig. 19).
Stenotic rectal tumors
Approximately 15% of rectal cancers cannot be completely evaluated due to tumor stenosis which prevents passage of the ultrasound
probe. If this occurs, the echoendoscope can usually be wedged into the distal margin of the tumor, which provides partial
staging information about T and N stage. Usually these are at least stage T3 tumors.
Rectal EUS staging after radiation therapy
The accuracy of EUS for staging rectal cancer after radiation therapy is markedly decreased due to postradiation edema, inflammation,
fibrosis, and necrosis. Staging accuracy after radiation is only 50%, with a 40% overstaging rate [23,24].
Colon cancer staging with EUS
Colon cancers are not staged with EUS because the treatment here involves surgical resection, which will provide definitive
anatomic staging. There are no treatment algorithms which incorporate EUS staging for colon cancer.
Anal cancer staging with EUS
There were approximately 4010 new cases of anal cancer diagnosed in the USA in 2004, with 580 deaths from this malignancy
. Anal canal cancer is usually squamous cell carcinoma. Staging plays a somewhat less important role in management of anal
cancer compared to other GI tract cancers, because most patients receive radiation therapy alone as definitive treatment,
unless there is an extremely large tumor or obvious metastatic disease. Staging of anal cancer also differs from rectal cancer
staging in that the actual dimensions of the tumor define T-stage, rather than depth of invasion. Anal cancer TNM staging
is shown in Fig. 20. EUS can be used to help stage anal cancer and may help in determining the extent of radiation, especially if lymph nodes
are involved .
There were estimated to be 31 860 new cases of pancreatic cancer and 31 270 deaths from pancreatic cancer in the United States in 2004. The 5-year survival rate after surgery is 3%, but if surgery
achieves clear margins and negative lymph nodes, the 5-year survival rate is nearly 25% .
Staging of pancreatic cancer
The currently used staging system for pancreatic cancer based on the 2002 American Joint Commission on Cancer (AJCC) Cancer staging manual, 6th edition  is shown in Fig. 21.
EUS staging of pancreatic cancer (Figs 12,13)
It is important to note that most of the published literature on pancreatic cancer staging with EUS is based on the 5th edition
of the AJCC Cancer staging manual. In the 5th edition staging system, stage T4 tumors involved any of the major vascular structures (portal vein, superior
mesenteric vein (SMV), superior mesenteric artery (SMA), or celiac artery). However, the 2002 6th edition staging system classifies
stage T3 tumors as involving either the portal vein or SMV, while T4 tumors are reserved for invasion of the SMA or celiac
artery. The impetus for this change relates to the possibility of performing pancreatic resection with portal vein reconstruction
in selected cases with minimal portal vein or SMV involvement, and reserving stage T4 designation for technically inoperable
tumors. The implication of this is that the published TNM staging accuracy for pancreatic cancer is not based on the current
staging system. It is unknown how this will change the reported staging accuracy of EUS.
Most studies show that the EUS T-stage accuracy for pancreatic cancer is approximately 85%, and the N-stage accuracy is 75%
. However, more recent studies suggest the staging accuracy may be lower. A study from the University of Pennsylvania found
the T-stage accuracy to be 69%, and the N-stage accuracy to be 54% . Another study from Germany reviewed videotapes of EUS cases of pancreatic tumors who had surgical or angiographic documentation
of vascular invasion, and found that sensitivity and specificity for detecting vascular invasion were 43% and 91%, respectively
Combination of EUS and CT/MRI for pancreatic cancer staging and determining resectability
Data is emerging that, because neither EUS nor CT nor MRI is highly accurate in determining resectability, a combination of
the tests may be the best staging strategy. A study comparing EUS and MRI found that the positive predictive value (PPV) of
EUS for predicting resectability was 69% compared to 77% for MRI, but the combined PPV of EUS + MRI was 89%, and the PPV for unresectability if both agreed was 76% . Another study comparing EUS, CT, and MRI found that TNM accuracy for EUS, CT, and MRI was 40, 46, and 36%, respectively
. Accuracy for predicting tumor resectability for EUS, CT, and MRI was 67, 83, and 75%, respectively, with the combination
of CT + EUS having an accuracy of 87%. A cost minimization analysis favored the sequential strategy of first performing helical CT
scan, followed by EUS as a confirmatory test in those patients with potentially resectable tumors.
Nearly all of the comparative data between EUS and CT for pancreatic cancer evaluation have involved older generation CT scanners.
Image resolution and speed have improved with the introduction of multidetector CT. Preliminary results from a study from
the Mayo Clinic comparing multidetector CT(MDCT) scan to EUS and MRI found that the accuracy for predicting resectability
was 83% for MDCT, 81% for MRI, and 78% for EUS, and that the combination of EUS + MDCT had an accuracy of 95% for predicting resectability . In contrast, a well-designed study from Indiana University revealed that both EUS and MDCT have similar rates of determining
resectability of pancreatic cancer, and that the combination of the two was no better than either alone. This would suggest
that MDCT might be used as the primary modality for determining resectability, especially as it would be anticipated to have
improved accuracy with further advances in CT technology.
EUS-FNA for staging pancreatic cancer
Trans-duodenal or transgastric EUS-FNA has a 90% sensitivity for obtaining a cytologic diagnosis of carcinoma in pancreatic
cancer. The utility of EUS-FNA in affecting management depends on the local institutional approach to pancreatic cancer. If
involved lymph nodes are not felt to preclude surgical resection, because the nodes will be removed in the surgical specimen,
then FNA of the nodes will not change management and is not needed. EUS-FNA of undetected liver or mediastinal metastatic
lesions is safe and potentially quite important, as this would preclude patients from undergoing attempted surgical resection
due to metastatic disease [36,37]. EUS-FNA to establish a tissue diagnosis is often helpful and/or necessary to allow patients to undergo adjuvant therapy or to enter clinical trials.
Recommendations for EUS staging of pancreatic cancer
The most prudent approach to using EUS for pancreatic cancer seems to be in conjunction with a good CT (i.e. multidetector
CT using pancreatic protocol) or MRI. EUS should only be performed after CT or MRI show no obvious evidence of distant metastatic
disease. If both tests suggest the patient is unresectable, then the patient does not go to surgery. If both tests suggest
the patient is resectable, then the patient proceeds to surgery. If there is disagreement between the two tests, then the
decision to operate should be on a case-by-case basis after informed discussion with the patient.
It can sometimes be difficult to distinguish ampullary cancer from pancreatic, duodenal, or bile duct cancer. Accurate local
staging is potentially important in centers where ampullary resection is performed . However, in most cases locally staging is usually not critical, as most patients will undergo a pancreatic head resection.
The staging criteria are shown in Fig. 24. A recent study found that EUS had a staging accuracy of 82% for T-stage and 71% for N-stage . EUS can often detect small ampullary neoplasms when other imaging modalities reveal biliary or pancreatic ductal dilatation
but no evidence of a mass [Fig. 25]. It is also often possible to tell whether small lesions have penetrated the duodenal muscularis and invaded the pancreas,
in which case Whipple's resection is necessary.
Extrahepatic bile duct cancer
The staging system for extrahepatic bile duct cancer is shown in Fig. 26. Transduodenal EUS has been reported to have staging accuracy of 85% . Transpapillary intraductal catheter probe EUS may also be helpful in staging these tumors [41,42].
Future trends and outstanding issues
The TNM staging classification of cancers is regularly updated to take account of new information about tumor behavior and
endosonographers need to be constantly aware of this if they are to provide accurate staging information. In esophageal cancer,
problems with the classification and accuracy of staging of junctional tumors exist and need to be clarified. Which, if any,
patients with superficial lesions need EUS prior to endoscopic mucosal resection is also a subject of debate at present. The
number of identified lymph nodes, not just their location, may predict prognosis and this may need to be incorporated into
staging systems. Accuracy for lymph node staging remains problematic and further improvements in this area are needed. Whether
or not analysis of molecular markers in EUS-FNA aspirates from lymph nodes will help in this regard requires study. Larger
studies of the utility of EUS re-staging after neoadjuvant therapy are also awaited with interest.
In pancreatic cancer, other imaging modalities such as multidetector CT and MRI are improving all the time and new comparative
studies of EUS with these modalities are necessary. Molecular studies of pancreatic FNA samples are underway and in the near
future results of microarray and proteomic analysis on pancreatic samples obtained by EUS will be reported, hopefully pointing
the way to a new role for EUS in diagnosis and staging of this malignancy.
In lung cancer, similar studies have already been reported and hopefully further studies will corroborate these promising
results. Endobronchial ultrasound (EBUS) with FNA, discussed in the chapter by Dr Eloubeidi, is showing great promise and
is likely to become widely used in the next few years. How it fits into the diagnostic and staging algorithm remains to be
decided but it may be possible to combine EUS and EBUS to allow almost complete mediastinal staging without resort to more
invasive techniques, and this is already under study.
In summary, EUS has a firmly established role in cancer staging but this role may continue to evolve as non-invasive imaging
techniques improve, with a greater emphasis on obtaining tissue for cytology, histology, and molecular marker analysis.
1 AJCC (2002). Cancer Staging Manual, 6th edn. Springer, New York.
2 Sobin, LH & Wittekind, C (2002). TNM: classification of malignant tumours, 6th edn. UICC Publications, Geneva.
3 Jemal, A, Tiwari, RC & Murray, T et al. Cancer statistics, 2004. CA Cancer J Clin 2004; 54 (1): 829. PubMed
4 Rice, TW, Blackstone, EH & Rybicki, LA et al. Refining esophageal cancer staging. J Thorac Cardiovasc Surg 2003; 125 (5): 110313. PubMed CrossRef
5 Rusch, VW. Should the esophageal cancer staging system be revised? J Thorac Cardiovasc Surg 2003; 125 (5): 9923. PubMed CrossRef
6 Catalano, MF, Alcocer, E & Chak, A et al. Evaluation of metastatic celiac axis lymph nodes in patients with esophageal carcinoma: accuracy of EUS. Gastrointest Endosc 1999; 50 (3): 3526. PubMed
7 Catalano, MF, Sivak, MV Jr, Rice, T, Gragg, LA & Van Dam, J. Endosonographic features predictive of lymph node metastasis. Gastrointest Endosc 1994; 40 (4): 4426. PubMed
8 Bhutani, MS, Hawes, RH & Hoffman, BJ. A comparison of the accuracy of echo features during endoscopic ultrasound (EUS) and EUS-guided fine-needle aspiration for
diagnosis of malignant lymph node invasion. Gastrointest Endosc 1997; 45 (6): 4749. PubMed
9 Rosch, T. Endosonographic staging of esophageal cancer: a review of literature results. Gastrointest Endosc Clin N Am 1995; 5 (3): 53747. PubMed
10 Vazquez-Sequeiros, E, Wiersema, MJ & Clain, JE et al. Impact of lymph node staging on therapy of esophageal cancer. Gastroenterology 2003; 125: 162635. PubMed CrossRef
11 Isenberg, G, Chak, A & Canto, MI et al. Endoscopic ultrasound in restaging of esophageal cancer after neoadjuvant chemoradiation. Gastrointest Endosc 1998; 48 (2): 15863. PubMed CrossRef
12 Zuccaro, G Jr, Rice, TW & Goldblum, J et al. Endoscopic ultrasound cannot determine suitability for esophagectomy after aggressive chemoradiotherapy for esophageal cancer. Am J Gastroenterol 1999; 94 (4): 90612. PubMed
13 Willis, J, Cooper, GS & Isenberg, G et al. Correlation of EUS measurement with pathologic assessment of neoadjuvant therapy response in esophageal carcinoma. Gastrointest Endosc 2002; 55 (6): 65561. PubMed CrossRef
14 Yasuda, K. EUS in the detection of early gastric cancer. Gastrointest Endosc 2002; 56 (4 Suppl.): S68S75. PubMed CrossRef
15 Sackmann, M, Morgner, A & Rudolph, B et al. Regression of gastric MALT lymphoma after eradication of Helicobacter pylori is predicted by endosonographic staging: MALT
Lymphoma Study Group. Gastroenterology 1997; 113 (4): 108790. PubMed CrossRef
16 Caletti, G, Fusaroli, P & Togliani, TE. US in MALT lymphoma. Gastrointest Endosc 2002; 56 (4 Suppl.): S21S26. PubMed
17 Murty, M, Enker, WE & Martz, J. Current status of total mesorectal excision and autonomic nerve preservation in rectal cancer. Semin Surg Oncol 2000; 19 (4): 3218. PubMed CrossRef
18 Hu, KS & Harrison, LB. Adjuvant therapy for resectable rectal adenocarcinoma. Semin Surg Oncol 2000; 19 (4): 33649. PubMed CrossRef
19 Kim, HJ & Wong, WD. Role of endorectal ultrasound in the conservative management of rectal cancers. Semin Surg Oncol 2000; 19 (4): 35866. PubMed CrossRef
20 Savides, TJ & Master, SS. EUS in rectal cancer. Gastrointest Endosc 2002; 56 (4 Suppl.): S12S18. PubMed CrossRef
21 Harewood, GC & Wiersema, MJ. Cost-effectiveness of endoscopic ultrasonography in the evaluation of proximal rectal cancer. Am J Gastroenterol 2002; 97 (4): 87482. PubMed CrossRef
22 Harewood, GC, Wiersema, MJ & Nelson, H et al. A prospective, blinded assessment of the impact of preoperative staging on the management of rectal cancer. Gastroenterology 2002; 123 (1): 2432. PubMed CrossRef
23 Napoleon, B, Pujol, B, Berger, F, Valette, PJ, Gerard, JP & Souquet, JC. Accuracy of endosonography in the staging of rectal cancer treated by radiotherapy. Br J Surg 1991; 78 (7): 7858. PubMed
24 Rau, B, Hunerbein, M & Barth, C et al. Accuracy of endorectal ultrasound after preoperative radiochemotherapy in locally advanced rectal cancer. Surg Endosc 1999; 13 (10): 9804. PubMed CrossRef
25 Magdeburg, B, Fried, M & Meyenberger, C. Endoscopic ultrasonography in the diagnosis, staging, and follow-up of anal carcinomas. Endoscopy 1999; 31 (5): 35964. PubMed CrossRef
26 Stephens, J, Kuhn, J & O'Brien, J et al. Surgical morbidity, mortality, and long-term survival in patients with peripancreatic cancer following pancreaticoduodenectomy. Am J Surg 1997; 174 (6): 6003. PubMed CrossRef
27 Gress, F, Savides, T & Cummings, O et al. Radial scanning and linear array endosonography for staging pancreatic cancer: a prospective randomized comparison. Gastrointest Endosc 1997; 45 (2): 13842. PubMed CrossRef
28 Palazzo, L, Roseau, G & Gayet, B et al. Endoscopic ultrasonography in the diagnosis and staging of pancreatic adenocarcinoma: results of a prospective study with
comparison to ultrasonography and CT scan. Endoscopy 1993; 25 (2): 14350. PubMed
29 Yasuda, K, Mukai, H, Nakajima, M & Kawai, K. Staging of pancreatic carcinoma by endoscopic ultrasonography. Endoscopy 1993; 25 (2): 1515. PubMed
30 Hunt, GC & Faigel, DO. Assessment of EUS for diagnosing, staging, and determining resectability of pancreatic cancer: a review. Gastrointest Endosc 2002; 55 (2): 2327. PubMed CrossRef
31 Ahmad, NA, Lewis, JD, Ginsberg, GG, Rosato, EF, Morris, JB & Kochman, ML. EUS in preoperative staging of pancreatic cancer. Gastrointest Endosc 2000; 52 (4): 4638. PubMed CrossRef
32 Rosch, T, Dittler, HJ & Strobel, K et al. Endoscopic ultrasound criteria for vascular invasion in the staging of cancer of the head of the pancreas: a blind reevaluation
of videotapes. Gastrointest Endosc 2000; 52 (4): 46977. PubMed CrossRef
33 Ahmad, NA, Lewis, JD, Siegelman, ES, Rosato, EF, Ginsberg, GG & Kochman, ML. Role of endoscopic ultrasound and magnetic resonance imaging in the preoperative staging of pancreatic adenocarcinoma. Am J Gastroenterol 2000; 95 (8): 192631. PubMed CrossRef
34 Soriano, A, Castells, A & Ayuso, C et al. Preoperative staging and tumor resectability assessment of pancreatic cancer: prospective study comparing endoscopic ultrasonography,
helical computed tomography, magnetic resonance imaging, and angiography. Am J Gastroenterol 2004; 99 (3): 492501. PubMed CrossRef
35 Wiersema, MJ, Fletcher, JG, Burgart, LJ, Farrell, MA, Stephens, DH & Johnson, CD. Prospective evaluation of potentially resectable pancreatic ductal adenocarcinoma: performance of triple-phase helical CT,
gadolinium-enhanced MR and endoscopic ultrasound. Radiology 2001; 221: 250.
36 tenBerge, J, Hoffman, BJ & Hawes, RH et al. EUS-guided fine needle aspiration of the liver: indications, yield, and safety based on an international survey of 167 cases. Gastrointest Endosc 2002; 55 (7): 85962. PubMed CrossRef
37 Hahn, M & Faigel, DO. Frequency of mediastinal lymph node metastases in patients undergoing EUS evaluation of pancreaticobiliary masses. Gastrointest Endosc 2001; 54 (3): 3315. PubMed CrossRef
38 Quirk, DM, Rattner, DW, Fernandez-del Castillo, C, Warshaw, AL & Brugge, WR. The use of endoscopic ultrasonography to reduce the cost of treating ampullary tumors. Gastrointest Endosc 1997; 46 (4): 3347. PubMed CrossRef
39 Skordilis, P, Mouzas, IA, Dimoulios, PD, Alexandrakis, G, Moschandrea, J & Kouroumalis, E. Is endosonography an effective method for detection and local staging of the ampullary carcinoma? A prospective study. BMC Surg 2002; 2 (1): 1. PubMed CrossRef
40 Tio, TL, Cheng, J, Wijers, OB, Sars, PR & Tytgat, GN. Endosonographic TNM staging of extrahepatic bile duct cancer: comparison with pathological staging. Gastroenterology 1991; 100 (5 Part 1): 135161. PubMed
41 Fujita, N, Noda, Y, Kobayashi, G, Kimura, K & Yago, A. Staging of bile duct carcinoma by EUS and IDUS. Endoscopy 1998; 30 (Suppl. 1): A132A134. PubMed
42 Tamada, K, Nagai, H & Yasuda, Y et al. Transpapillary intraductal US prior to biliary drainage in the assessment of longitudinal spread of extrahepatic bile duct
carcinoma. Gastrointest Endosc 2001; 53 (3): 3007. PubMed
43 Rice, TW, Zuccaro, G Jr, Adelstein, DJ, Rybicki, LA, Blackstone, EH & Goldblum, JR. Esophageal carcinoma: depth of tumor invasion is predictive of regional lymph node status. Ann Thorac Surg 1998; 65 (3): 78792. PubMed CrossRef
Copyright © Blackwell Publishing, 2005