2. Pancreatic Malignancies—PDAC (Pancreatic Ductal Adenocarcinoma)
EUS was initially developed in the late 1980s and 1990s for pancreas tumor diagnosis and for the identification of papilla Vateri tumors and submucosal lesions in the upper gastrointestinal tract. It soon became clear that, in specific cases, this examination technique was superior to transabdominal ultrasound
[2]. Right from the start, EUS exhibited a sensitivity of over 90% in the detection of malignant pancreatic tumors
[3]. Up until then, pancreatic tumors were identified on the basis of abnormal lab results and abdominal ultrasound as well as CTs. However, the tumors identified using these procedures were usually in an advanced stage. For pancreatic cancer, contemporary studies suggest that EUS has a sensitivity of up to 99 percent, with malignant pancreatic tumors measuring between 2 and 3 cm. This clearly demonstrates its overwhelming superiority over all other tomography devices, including CT, transabdominal ultrasound and magnetic resonance
[4][5]. This is primarily due to the possibility of moving the endoscopic-sonographic transducer to the immediate proximity of the lesion. Of course, EUS is not free of errors (for example, regarding the differentiation between early malignant and post-inflammatory lesions in the pancreas) and has its limitations regarding specificity in the diagnosis of pancreatic carcinoma. Diagnosis becomes more difficult when, in addition to pancreatitis, the pancreatic tissue displays heterogeneous morphology, so even seasoned examiners may overlook pancreatic cancer. On the other hand, there are studies that imply that EUS provides higher diagnostic accuracy than CT for pancreatic malignancy in chronic pancreatitis. There are criteria and predicting factors (hypoechoic pattern, pancreatic duct dilation, distal pancreatic atrophy) that help to differentiate benign from malignant lesions in patients with chronic pancreatitis. New developments in MR technology have produced a valuable supplementary diagnostic technique specifically for cases where malignancies are suspected, and difficult examination conditions reduce diagnostic reliability
[6][7]. An MR provides greater connective tissue contrast than CT images, which, in the final analysis, allows varying pancreatic tissue masses to be differentiated more accurately
[2][8]. Eleven percent of all PDACs are isodense, meaning they take up the same amount of contrast agent as the surrounding tissue, and this share increases to as much as 27 percent in the case of small-diameter tumors (≤2 cm). In this case, a combination of MR, dual-energy CT (with enhanced contrast) and EUS is helpful. EUS provides a differentiated view of pancreatic tissue morphology and remains one of the most accurate tools for the identification of unclear lesions. On that account, EUS is the method of choice for diagnosing and staging pancreatic adenocarcinoma
[9]. It does not only have high specificity in detecting malignancies but (reversely) also serves as the primary means to exclude pancreatic cancer
[5]. After a 24-month observation period, a group of patients who were referred to EUS due to abnormal CT findings in the pancreas and suspected adenocarcinoma did not develop any pancreatic malignancy, as predicted by EUS. This ultimately yields a negative predictive value (NPV) of 100 percent
[5].
3. Cytology and Histology
Nowadays, combining EUS and fine needle aspiration cytology (FNA) or fine needle biopsy (FNB) is the standard everyday procedure. To be able to initiate adequate treatment, a lesion needs to be cytologically or histologically confirmed. In particular, when primary surgery is impossible, locally advanced tumors are unresectable a priori (locally advanced pancreatic ductal adenocarcinoma, LAPDAC), or neoadjuvant therapy is indicated in the case of a ‘borderline-resectable’ tumor. In retrospective analyses of EUS databases, EUS-FNB has been associated with a diagnostic precision of 89 percent for solid pancreatic tumors
[8][10][11][12]. The possibility of obtaining biopsies of a suspected malignant pancreatic lesion directly during an imaging procedure has an immediate bearing on the medical management of the affected patients. Since only a minority of patients are eligible for curative therapy right away, obtaining cytological and histological material for diagnostic confirmation is absolutely necessary so that (possibly neoadjuvant) chemotherapy can be started or continued
[10][11][12].
Lately, reports on cases of intra-abdominal tumor seeding have emerged, especially after EUS-FNA occurred. Although these complications seem to be extremely rare, the impact on the further faith of the patients may be tremendous. Safety measures such as shortening the puncture path, decreasing the number of punctures and adding needle sleeves could minimize the risk of this unwanted complication. Nevertheless, prospective controlled studies must be carried out to objectify the frequency of the phenomenon and its impact on survival
[13].
4. On-Site Pathology
The likelihood of arriving at a diagnosis with high specificity is markedly improved by the attendance of on-site pathologists, especially when conventional biopsy needles are used for the purpose of aspiration cytology (FNA). A prompt assessment reduces the risk of completing an examination, although the tissue samples taken are unusable
[14][15][16][17]. With the support of on-site pathologists, EUS-FNA biopsy can facilitate earlier diagnosis and possibly suggest an alternative diagnosis, thereby reducing patient mortality. Apparently, diagnosis is improved on account of the direct communication between pathologists and endoscopists and the possibility of sharing an endoscopic impression. Ever since new types of 19-, 22- and 25-gauge needles with ultra-sharp blades have found their way into routine practice as a means of obtaining histological and cytological material, the presence of on-site pathologists, whose attendance has often been requested (but very rarely approved in practice), has become even more expedient. As the technique is more refined, not only conventional cell aspiration (FNA) but also the collection of tissue using fine needle biopsy (FNB) and thus a more accurate diagnosis have become possible
[8][9].
EUS is not only superior to conventional ultrasound or MR in certain cases when it comes to identifying and excluding pancreatic malignancies but also serves as a supplement to these non-invasive procedures in a meaningful way
[4][5].
Even in cases of papilla Vateri adenoma and carcinoma, EUS can be used to correctly assess the invasiveness of the lesion and the endoscopic resectability where necessary.
5. Cystic Neoplasms of the Pancreas
Pancreatic Cystic Lesion
The first step when confronted with cystic pancreatic lesions is to distinguish them from pseudocysts by differential diagnosis. In histopathological terms, pseudocysts frequently turn out to be ductal ectasia rather than a cyst or an actual cystic neoplasm of the pancreas. Pseudocyst diagnosis is usually assumed on the basis of anamnestic records, and a patient history consistent with pancreatitis usually points the way
[18]. Biochemical analysis of the cyst content and imaging usually provide additional information. However, patients with cystic neoplasms occasionally also have concomitant pancreatitis or pancreatitis developing as a result of congestion induced by the neoplasm. Furthermore, patients with a pseudocyst do not necessarily always have a history that indicates previously occurring pancreatitis. Once pseudocysts have been excluded as a diagnosis, the type of cystic neoplasm of the pancreas must be determined.
Broadly speaking, there are two types of cystic neoplasms: mucinous cystic neoplasms, which include intraductal papillary mucinous neoplasms (IPMN) and mucinous cystic neoplasms (MCN) and serous cystic neoplasms (SCN). The paper of Sahani et al.
[19] contains an overview of cystic lesions and neoplasms of the pancreas.
Patients with mucinous cysts, meaning patients with main duct (MD) IPMN and mixed type (MT) IPMN as well as MCN, undergo surgery if operability is possible, i.e., the cyst is resected.
After exclusion of any ‘worrisome’ or ‘high-risk features’ (which constitute a relative or absolute indication for surgery), imaging is performed in most branch duct (BD-)IPMN cases. Generally, the consensus policy should be adhered to
[20].
MR and EUS imaging is usually performed on patients with SCN unless the diameter of the cyst exceeds 4 cm, tumor-associated symptoms present themselves or the cyst exhibits a high growth rate.
In cases of solid pseudopapillary tumors (SPNs), surgical resection is absolutely necessary
[19][21][22].
6. EUS-FNB
There is no generally valid consensus policy for EUS-FNB indication in cases of cystic neoplasms of the pancreas.
EUS with or without FNB is not indicated if tomography provides a clear, reliable diagnosis. However, in case of doubt, an additional referral to EUS is key to ensuring a final diagnosis and personalized therapy because of the specific endosonomorphologic imaging provided and the cystic fluid analysis
[23].
The high resolution and better image characteristics compared to CTs are a clear benefit. Endoscopic ultrasound makes sense if the CT-/MR-based diagnosis is uncertain, if cysts exhibit what are referred to as ‘worrisome features’ and if the malignancy diagnosis needs to be verified for high-risk patients ahead of surgery (comorbidity, age). Cyst fluid can be aspirated and analyzed, and grape-like nodes and small intracystic tumors can be biopsied and histologically/cytologically examined. Lymph node metastasis can be identified, biopsied and their relation to the primary tumor can be verified. Frequently (but not always reliably), vascular invasion can be established or excluded. A growing number of new biomarkers are being identified in cyst fluid that can be used to predict malignancy with increasing certainty
[24]. In addition to analyzing cyst fluid to determine its biochemical and cytological composition and analyzing DNA in the cells found, the macroscopic assessment of cysts fluid can provide initial indicators early on: a highly viscous fluid, distinguishable on account of its behavior during aspiration into the biopsy syringe, is very likely to have a high mucin content and thus indicate IPMN or MCN. Even with a sensitivity of 73 percent and a specificity of 84 percent, the high CEA concentration in the cyst fluid is also indicative of a mucinous, unlike a non-mucinous, cystic neoplasm of the pancreas; a cut-off of 190 ng/mL is frequently cited in such cases
[25]. Conversely, the CEA level in the cyst fluid is not predictive of the lesion’s malignancy.