Key points

Introduction

Endoscopic ultrasound (EUS) has revolutionized the ability to diagnose and stage cancers of the gastrointestinal tract and assess the pancreas. EUS fine-needle aspiration (FNA) is now routinely used to acquire tissue from different tissue structures located in the vicinity of the gastrointestinal tract. The pancreas and lymph nodes are the most common organs targeted in EUS-FNA with excellent sensitivity and specificity. For the diagnosis of solid pancreatic masses a recent medical literature review showed a 78% to 95% sensitivity, 75% to 100% specificity, and 78% to 95% accuracy confirming that EUS-FNA is an effective and safe method to obtain a cytologic diagnosis of pancreatic tumors. EUS-FNA has also been proved to increase the accuracy of lymph node staging and thereby reduce the number of unnecessary surgical interventions. A meta-analysis from 76 studies on mediastinal lymph nodes demonstrated that FNA improved the sensitivity from 84% to 88% and specificity from 88% to 96% compared with EUS imaging alone.

Nevertheless, the negative predictive value of EUS-FNA is low, especially for the diagnosis of pancreatic neoplasms. The same is true for other targets, such as lung cancer during endoscopic bronchial US. Pathologists are well aware of the limited amount of tumor that may actually be present in diagnostic samples used for lung cancer diagnosis. A recent study demonstrated that the median percentage of tumor present in malignant samples was little more than 20% when squamous cell or adenocarcinoma was diagnosed, whereas this figure was only 10% in non–small cell lung cancer cases.

The emergence of selective, targeted therapies, directed toward a particular molecular characteristic of an individual patient’s tumor, is now driving the need for biomarker identification and testing in several cancer types. This means that the technique needs improvement to provide more material, in fewer passes, with more flexible, sharp, and clearly echovisible needles and with a similar safety. Future needs and expectancies are as follows:

• •
  

  Either acquisition of more material

  
  
  
  
  • ○
    

    To allow for cytology, histology, immunocytochemistry or histochemistry, and molecular biology

    
    
  • ○
    

    With fewer passes

    
    
  • ○
    

    With needles that are flexible, sharp, echovisible, and cost effective

    
    
  • ○
    

    With auxiliary devices inserted in the needle (brush, forceps, and so forth)

  
  
  
  
• •
  

  Or avoidance of tissue acquisition or limited tissue acquisition with the help of ancillary techniques, such as

  
  
  
  
  • ○
    

    Optical biopsies

    
    
  • ○
    

    Contrast-enhanced US

    
    
  • ○
    

    Elastography

    
    
  • ○
    

    Other tissue analysis techniques.

  
  

These issues may be combined or selectively used for specific lesions. Future trends include the following:

• •
  

  Improvement of the sampling technique

  
  
• •
  

  Development of new needles

  
  
• •
  

  Widespread use of liquid phase cytology and cell blocks

  
  
• •
  

  Combination of cytology and histology with FNA biopsy

  
  
• •
  

  Use of immunocytochemistry and histochemistry

  
  
• •
  

  Use of polymerase chain reaction, flow cytometry

  
  
• •
  

  Use and validation of molecular markers

  
  
• •
  

  More routine use of ancillary techniques, such as elastography, contrast-enhanced EUS

  
  
• •
  

  Optical biopsies obtained through the needle

Introduction

Endoscopic ultrasound (EUS) has revolutionized the ability to diagnose and stage cancers of the gastrointestinal tract and assess the pancreas. EUS fine-needle aspiration (FNA) is now routinely used to acquire tissue from different tissue structures located in the vicinity of the gastrointestinal tract. The pancreas and lymph nodes are the most common organs targeted in EUS-FNA with excellent sensitivity and specificity. For the diagnosis of solid pancreatic masses a recent medical literature review showed a 78% to 95% sensitivity, 75% to 100% specificity, and 78% to 95% accuracy confirming that EUS-FNA is an effective and safe method to obtain a cytologic diagnosis of pancreatic tumors. EUS-FNA has also been proved to increase the accuracy of lymph node staging and thereby reduce the number of unnecessary surgical interventions. A meta-analysis from 76 studies on mediastinal lymph nodes demonstrated that FNA improved the sensitivity from 84% to 88% and specificity from 88% to 96% compared with EUS imaging alone.

Nevertheless, the negative predictive value of EUS-FNA is low, especially for the diagnosis of pancreatic neoplasms. The same is true for other targets, such as lung cancer during endoscopic bronchial US. Pathologists are well aware of the limited amount of tumor that may actually be present in diagnostic samples used for lung cancer diagnosis. A recent study demonstrated that the median percentage of tumor present in malignant samples was little more than 20% when squamous cell or adenocarcinoma was diagnosed, whereas this figure was only 10% in non–small cell lung cancer cases.

The emergence of selective, targeted therapies, directed toward a particular molecular characteristic of an individual patient’s tumor, is now driving the need for biomarker identification and testing in several cancer types. This means that the technique needs improvement to provide more material, in fewer passes, with more flexible, sharp, and clearly echovisible needles and with a similar safety. Future needs and expectancies are as follows:

• •
  

  Either acquisition of more material

  
  
  
  
  • ○
    

    To allow for cytology, histology, immunocytochemistry or histochemistry, and molecular biology

    
    
  • ○
    

    With fewer passes

    
    
  • ○
    

    With needles that are flexible, sharp, echovisible, and cost effective

    
    
  • ○
    

    With auxiliary devices inserted in the needle (brush, forceps, and so forth)

  
  
  
  
• •
  

  Or avoidance of tissue acquisition or limited tissue acquisition with the help of ancillary techniques, such as

  
  
  
  
  • ○
    

    Optical biopsies

    
    
  • ○
    

    Contrast-enhanced US

    
    
  • ○
    

    Elastography

    
    
  • ○
    

    Other tissue analysis techniques.

  
  

These issues may be combined or selectively used for specific lesions. Future trends include the following:

• •
  

  Improvement of the sampling technique

  
  
• •
  

  Development of new needles

  
  
• •
  

  Widespread use of liquid phase cytology and cell blocks

  
  
• •
  

  Combination of cytology and histology with FNA biopsy

  
  
• •
  

  Use of immunocytochemistry and histochemistry

  
  
• •
  

  Use of polymerase chain reaction, flow cytometry

  
  
• •
  

  Use and validation of molecular markers

  
  
• •
  

  More routine use of ancillary techniques, such as elastography, contrast-enhanced EUS

  
  
• •
  

  Optical biopsies obtained through the needle

Improvement of sampling and sample processing

Previous papers in this issue address the key points to improve tissue acquisition. The first step is appropriate training. The American Society for Gastrointestinal Endoscopy recommends a minimum of 150 supervised cases, of which 75 should be pancreaticobiliary, and 50 EUS-FNA should be performed. The next step is a puncture technique adapted to the target: choice of needle caliber; with or without suction depending on tissue hardness and vascularization; number of passes (presence of a cytopathologist in the room or not); and most importantly use of the fanning technique (at individual passes, the needle should be positioned at different areas within the mass and then moved back and forth two to three times in each area to procure tissue). The technique may also be adapted to the needle used, histology needles needing less suction and passes through the targeted tissue.

Getting a sample in the appropriate target is only the first step. Next steps depend on the quality and amount of the material and on its adequate processing. A case is made in many expert centers for a more widespread use of immunocytochemistry/histochemistry, even in cases where the pathologist is more confident of the diagnosis on morphologic grounds. Indeed, a lower threshold for the use of immunochemistry may be appropriate and improve diagnostic accuracy. The use of immunochemistry to augment morphologic diagnosis on small tissue samples consumes much more tissue than conventional morphologic diagnosis alone. The routine preparation of cell blocks from fluid and needle aspiration cytology samples will also assist in providing extradiagnostic material for other tests. The importance of tissue preparation and preservation is emphasized by the emergence of predictive molecular marker testing.