Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Commentary

Among the various types of cystic neoplasms of the pancreas, the management of solid pseudopapillary neoplasm (SPN) does not have any controversy. Mucinous cystic neoplasm (MCN) described in this chapter also poses little controversy to the clinician who happens to diagnose them by some imaging modalities. Most of the MCNs found incidentally are still benign and represent currently a very good indication for laparoscopic distal pancreatectomy or local resection (enucleation) when feasible. Serous cystic neoplasms (SCNs) also presented in this chapter do not need resection unless they are indistinguishable from other types of cystic neoplasm detailed in the chapter, such as intraductal papillary mucinous neoplasm (IPMN) or MCN, when SCN takes on a macrocystic or oligocystic appearance as opposed to its typical microcystic appearance.

In contrast to SPN, MCN, and SCN, IPMN of the pancreas, especially the branch duct type (BD-IPMN), excites a lot of controversies in regard to differentiation from other pancreatic cysts, diagnosis of malignancy, and need for and type of operative/non-operative management. BD-IPMNs must be differentiated from MCNs, macrocystic or oligocystic SCNs, epidermoid cysts, lymphoepithelial cysts, and cystic variants of other neoplasms. Even with complete understanding of the imaging characteristics of each entity (Tanaka et al. 2006), it is sometimes difficult to differentiate BD-IPMNs confidently from MCNs, macrocystic or oligocystic SCNs, and lymphoepithelial cysts preoperatively.

Because it is especially important to differentiate non-mucinous cysts from IPMNs and MCNs with malignant potential, endoscopic ultrasonography-guided fine-needle aspiration (EUS-FNA) is receiving enthusiastic interest lately. A high level of carcinoembryonic antigen (CEA) in the cystic fluid is characteristic of mucinous cysts. Although the cut-off concentration that provides a confident diagnosis of mucinous epithelium varies from report to report (>367 (Lewandrowski et al. 1993), >800 (van der Waaij et al. 2005), ≥480 (Linder et al. 2006), >800 (Attasaranya et al. 2007), and >192 ng/ml (Brugge et al. 2004)), an increased value of even >5 ng/ml is highly suggestive of a mucinous neoplasm. Although the CEA levels are not necessarily consistent with levels of other molecular markers, including a glycan variant of MUC-5AC (Haab et al. 2010), mucin-like carcinoma-­associated antigen (Khalid et al. 2009), KRAS mutations (Bernard et al. 2002) and CA72-4 (Jang et al. 2005), the diagnostic sensitivity was reported to improve when combined (Haab et al. 2010; Sawhney et al. 2009).

The diagnosis of malignant transformation of BD-IPMN remains controversial at present. The original Sendai guidelines recommend resection of BD-IPMN with one or more of five criteria for suspected malignancy, i.e., positive pancreatic juice cytology, the presence of mural nodules, cyst size >3 cm, dilation of the main pancreatic duct, and abdominal pain (Tanaka et al. 2006). Nevertheless, 80–85 % of all BD-IPMNs resected according to these guidelines are benign. Therefore, we need to identify other diagnostic aids that would avoid or at least minimize a “false positive” resection. Based on the understanding of histologic subtypes, i.e., gastric, intestinal, pancreatobiliary, and oncocytic, and the recent observation that the intestinal subtype is more likely to de-differentiate into malignancy, selection of the intestinal subtype may be helpful to distinguish BD-IPMNs with a greater tendency for malignant transformation. A few such attempts have been reported by immunohistochemical or molecular analysis of cells contained in the pancreatic juice (Hibi et al. 2007; Nakata et al. 2009).

On the contrary, the authors of this chapter have suggested that even small BD-IPMNs ≤3 cm without malignant stigmata (“Sendai-negative”) have a relatively high risk of malignancy. Among their 69 patients with “Sendai-negative” BD-IPMNs, 25 % had in situ or invasive carcinoma. Lee et al. (2008) claimed that one of 30 BD-IPMNs resected with no Sendai criteria had carcinoma in situ; however, the absence of mural nodules was judged by CT, MR, or EUS in both of these studies. It is well accepted that EUS is the most sensitive modality to evaluate the presence or absence of a mural nodule and not CT or MR. In a collective series of 349 patients who underwent EUS initially to prove the absence of mural nodule, there were 7 patients who underwent resection without any of the Sendai criteria during a median follow-up of 3.5 years, and none of them had carcinoma (Maguchi et al. 2011). There have been four series describing clearly the relationship of malignancy to the size and the presence/absence of mural nodules. In 124 BD-IPMNs <3 cm without mural nodules, there was no single case of malignancy (Tanaka 2011).

If expertise in EUS-FNA and cytologic interpretation of “high grade atypia” in the cyst fluid are available, the cytologic analysis of the cyst fluid obtained by EUS-FNA might add diagnostic value, although the sensitivity is often limited by scant cellularity of the aspirate and contamination by gastrointestinal mucosal cells (van der Waaij et al. 2005; Pitman and Deshpande 2007; Pitman et al. 2010; Frossard et al. 2003; Belsley et al. 2008; Recine et al. 2004; Michaels et al. 2006; Layfield and Cramer 2005; Emerson et al. 2006; Maire et al. 2003, 2008). Cells with “high-grade atypia” in mucinous cyst fluid obtained by EUS-FNA indicated the presence of malignancy with a sensitivity of 72 % and an accuracy of 80 % (Pitman et al. 2010). The same group claimed that “high-grade atypia” was the most sensitive predictor of malignancy even in small (≤30 mm) BD-IPMNs (67 %), compared to mural nodules and a dilated main pancreatic duct which were highly specific (>90 %) but insensitive (39–44 %) (Genevay et al. 2011).

Follow-up surveillance of BD-IPMNs without malignant signs is an especially challenging problem in the management of IPMNs. EUS seems to be the best modality but has the drawbacks of increased cost, invasiveness, and intraobserver and interobserver variability. In reality, we cannot subject all patients to routine surveillance by EUS. How and how often to detect malignant changes of BD-IPMNs and to survey the development of distinct ductal adenocarcinoma remain very important controversies. Since we reported the occurrence of in situ or invasive ductal carcinoma concomitant with a benign BD-IPMN (Tanaka et al. 1997; Yamaguchi et al. 1997, 2002), this phenomenon has attracted increasing attention. Several reports have suggested that 3–9 % of patients with BD-IPMNs had or developed pancreatic ductal carcinoma distinct from IPMN-related invasive cancer (Tanaka 2011). During a median follow-up of 87 months, in 60 patients with BD-IPMNs, even when <1 cm in size, developed 5 ductal carcinomas (8 %) (Uehara et al. 2008). Worsening diabetes and high or increasing levels of serum CA19-9 predicted the presence of ductal carcinoma (Ingkakul et al. 2010; Kanno et al. 2010). Also, older age, smaller size of BD-IPMN, and smaller caliber of the main pancreatic duct were reported to be associated with the development of ductal carcinoma compared with the patients who did not develop ductal carcinoma (Tanno et al. 2010). The appropriate method and interval of surveillance of BD-IPMNs remain to be further investigated.