Advertisement

Through-the-needle biopsy of pancreas cysts: Who, what, where, when, and mostly… why?

      Abbreviation:

      TTNB (through-the-needle biopsy)
      For those who find value in sampling pancreatic lesions, through-the-needle biopsy (TTNB) is surely an exciting new modality because cyst fluid analysis and/or cyst wall sampling with FNA needles are far from perfect.
      • de Jong K.
      • Poley J.W.
      • van Hooft J.E.
      • et al.
      Endoscopic ultrasound-guided fine-needle aspiration of pancreatic cystic lesions provides inadequate material for cytology and laboratory analysis: initial results from a prospective study.
      ,
      • Sahai A.V.
      • Chua T.S.
      • Paquin S.
      • et al.
      Analysis of variables associated with surgery versus observation in patients with pancreatic cystic lesions referred for endoscopic ultrasound.
      For others, like me, who find cyst puncture for any reason truly useful in very few (<10%) cases, the available data on TTNB (including the meta-analysis presented here by Facciorusso et al
      • Facciorusso A.
      • Del Prete V.
      • Antonino M.
      • et al.
      Diagnostic yield of EUS-guided through-the-needle biopsy in pancreatic cysts: a meta-analysis.
      ) do not change much.
      The meta-analysis format of the present study cannot overcome the overall, arguably extreme, lack of high-quality prospective comparative data in a demonstrably clinically appropriate spectrum of patients. As usual, the studies included also fail to make any attempt to assess the decisional incremental value of adding TTNB to the clinical picture (eg, presence or absence of symptoms, history of pancreatitis), imaging (presence and severity of signs of aggressive behavior), and cyst fluid appearance (particularly the presence or absence of the string sign). After all, if the cyst fluid is clearly mucinous, isn’t the diagnosis of a mucinous lesion already obvious? Or, for example, if the lesion has huge mural nodules and main pancreatic duct obstruction, do we really need cyst wall sampling to decide whether to operate?
      That being said, it is not hard to believe the authors’ conclusion that TTNB gets more tissue than FNA (because FNA is so weak) and that it is relatively safe (no major adverse events). But the risk of adverse events must always be balanced with the clinical value of the information obtained, because no prospective studies have shown that any form of cyst puncture improves clinical outcomes such as disease-free survival. Cyst puncture may change management, but there are not conclusive data showing that these changes in management actually improve outcomes.
      • Dumonceau J.M.
      • Deprez P.H.
      • Jenssen C.
      • et al.
      Indications, results, and clinical impact of endoscopic ultrasound (EUS)-guided sampling in gastroenterology: European Society of Gastrointestinal Endoscopy (ESGE) clinical guideline - Updated January 2017.
      What is somewhat harder to accept is the external validity of the reported diagnostic accuracy of 88.3%, because it was obtained after a subgroup analysis that excluded almost 90% of the patients in whom surgical correlation was unavailable. We do not know why patients did (or did not) undergo surgery. It seems most logical to assume that patients who underwent surgery had more highly suggestive lesions to begin with. If this is the case, this spectrum bias could limit the utility of this information in less suggestive cases. In other words, the true accuracy in more indeterminate cases (cases in which the indication for surgery is less clear) could be significantly lower than that reported.
      But let’s assume that TTNB is fairly accurate and safe, and better than FNA and cyst fluid analysis. Many questions remain. Who needs it? What is the goal? When should it be performed? Where should we sample? If these answers are not clear, then why should it be used at all?
      Who needs cyst puncture? For solid pancreatic lesions, the value of EUS-guided biopsy is clear. It helps exclude surgery for medically treatable pathology such as autoimmune pancreatitis or can provide the histological proof of malignancy required for preoperative chemotherapy. However, for cystic lesions, surgery is often performed preventatively, to avoid malignant transformation; or for cyst-related symptoms such as pain. Features such as rapid progression, significant nodules or wall thickening, or pancreatic duct or bile duct obstruction are worrisome. Therefore, it is difficult to understand how cyst puncture (whether suspicious, positive, or negative for malignancy) would change management in such cases.
      What is the goal of cyst puncture? It would appear indicated only in cases where it would clearly change management on the decision to perform preventive surgery for asymptomatic lesions. In our institution, this is reserved primarily for indeterminate macrocystic head lesions, in younger surgical candidates, to distinguish serous from mucinous lesions. If cyst puncture shows mucinous fluid, more serious consideration is given to closer follow-up or preventive surgery. In comparison with body lesions, the threshold for surgery for head lesions is higher because of the more elevated mortality and morbidity of the Whipple procedure. If the fluid is nonmucinous (no string sign) and fluid carcinoembryonic antigen is <5 ng/mL, then the cyst is considered serous, and conservative management is favored.
      It is noteworthy that the above strategy actually does not require cyst wall sampling (ie, TTNB or FNA).
      Where and when should the cyst wall be sampled? Presumably, it could help distinguish a mucus ball from an actual mural nodule. However, this could probably also be done with an FNA needle (mucus balls can be mobilized and separated from the wall, and would be acellular). Contrast-enhanced imaging could also distinguish a vascular nodule from avascular mucus. Perhaps it could be used to diagnose dysplasia or cancer in nodules or other suggestive intracystic lesions. However, dysplasia and cancer can be focal, so if the result is negative, would this really prevent surgery, or at least further close follow-up? Maybe it could be helpful in trying to distinguish mucinous from serous lesions when the string sign is negative but the carcinoembryonic antigen is >5 ng/mL. This is possible; but the accuracy of TTNB for this very unusual specific indication, especially in cysts that show no worrisome features, is really unknown, and it could be quite poor.
      Until we can clearly answer these questions, I must continue to wonder why endosonographers would want to aspirate or perform wall biopsy in more than a small minority of cystic lesions.

      Disclosure

      The author disclosed no financial relationships.

      References

        • de Jong K.
        • Poley J.W.
        • van Hooft J.E.
        • et al.
        Endoscopic ultrasound-guided fine-needle aspiration of pancreatic cystic lesions provides inadequate material for cytology and laboratory analysis: initial results from a prospective study.
        Endoscopy. 2011; 43: 585-590
        • Sahai A.V.
        • Chua T.S.
        • Paquin S.
        • et al.
        Analysis of variables associated with surgery versus observation in patients with pancreatic cystic lesions referred for endoscopic ultrasound.
        Endoscopy. 2011; 43: 591-595
        • Facciorusso A.
        • Del Prete V.
        • Antonino M.
        • et al.
        Diagnostic yield of EUS-guided through-the-needle biopsy in pancreatic cysts: a meta-analysis.
        Gastrointest Endosc. 2020; 92: 1-8.e3
        • Dumonceau J.M.
        • Deprez P.H.
        • Jenssen C.
        • et al.
        Indications, results, and clinical impact of endoscopic ultrasound (EUS)-guided sampling in gastroenterology: European Society of Gastrointestinal Endoscopy (ESGE) clinical guideline - Updated January 2017.
        Endoscopy. 2017; 49: 695-714

      Linked Article