IPMN of the pancreas – does histological subtyping allow for improved stratification and follow-up?

Intraductal papillary mucinous neoplasms (IPMNs) are mucin-producing cystic lesions that involve the pancreatic ductal system. A subset of IPMNs will progress to invasive carcinoma. IPMNs can be classified based on anatomy as well as histology. IPMNs engaging the main pancreatic duct are termed main duct type, those engaging branch ducts are termed branch duct type, and those communicating with both ducts are termed mixed type. IPMNs that involve the main pancreatic duct have a higher risk of malignancy compared to IPMNs with exclusively branch duct engagement [1]. Histologically, IPMNs are divided into three subtypes: intestinal, gastric, and pancreatobiliary. These subtypes have been suggested to develop via different signaling pathways and appear to have different biological behavior [2].

Histological subtyping of IPMNs has gained interest in recent years to enhance risk stratification. However, it is still not known whether histological subtyping actually contributes to better clinical management. What evidence is available for histological stratification of patients with IPMN?

The classification of IPMNs into histological subtypes is based on morphological and immunohistochemical criteria [3] (Figure 1). The intestinal-type IPMN shows similarity to intestinal villous neoplasms with tall columnar epithelial cells. It expresses MUC2, MUC5AC, and CDX2, but is negative for MUC1. The gastric-type IPMN is composed of cells showing similarity to gastric foveolar epithelium. The cells express MUC5AC and MUC6, but are negative for MUC1 and MUC2. The pancreatobiliary-type IPMN displays fern-like complex papillae with enlarged hyperchromatic nuclei and moderate amphophilic cytoplasm. The cells are positive for MUC1, MUC5AC, and MUC6 (Table 1).

Figure 1. Histological appearance of the subtypes of intraductal papillary mucinous neoplasm of the pancreas. (A) The intestinal type; (B) the gastric type; (C) the pancreatobiliary type; (D) normal pancreas. Image courtesy of Dr Agata Sasor. Printed with permission [19].

Table 1. Morphological subtypes of IPMN.

	Intestinal	Gastric	Pancreatobiliary
Papillae	Villous papillae	Thick finger-like papillae	Fern-like
MUC1	−	−	+
MUC2	+	−	−
MUC5AC	+	+	+
MUC6	−	+	+
CDX2	+	−	−

Different types of invasive carcinoma can arise from IPMNs. Invasive carcinoma arising from the intestinal-type IPMN corresponds to mucinous (colloid) carcinoma. When derived from pancreatobiliary-type and gastric-type IPMNs, it corresponds to conventional ductal (tubular) adenocarcinoma.

Emerging evidence supports the predictive value of histological subtype in terms of malignant transformation and prognosis [4–8]. Pancreatobiliary-type IPMNs have the highest risk for malignant transformation. Survival for patients with tubular adenocarcinoma is as poor as for conventional pancreatic ductal adenocarcinoma, while the colloid type generally has a better prognosis.

Adjuvant therapy is a cornerstone in the surgical management of pancreatic ductal adenocarcinoma [9]. However, less is known regarding the clinical benefit of adjuvant therapy for IPMN. It has been shown that adjuvant therapy may not be beneficial to all patients with IPMN, but only in specific subsets of patients. Adjuvant therapy has found to prolong survival only in patients with tubular carcinomas [10].

Another controversial issue in the management of IPMNs includes postoperative surveillance. Histological subtype appears to affect the risk of metachronous high-risk lesions in the remnant pancreas after partial pancreatectomy. Patients with the pancreatobiliary or intestinal subtype have the highest risk for metachronous development and should undergo intensified, long-term surveillance of the remnant pancreas after surgery as long as the patient is fit for surgery [11].

However, despite the importance of histological subtyping, a large portion of IPMNs cannot be correctly classified. MUC staining has been suggested to be effective for determining the histological subtype [12]. However, almost 25% of tumors are reported to be non-classifiable due to discordant immunohistochemical results or overlapping histological characteristics [13]. Interobserver variability is another challenge [14]. Molecular techniques can be used in conjunction with endoscopic ultrasound-fine needle aspiration (EUS-FNA) in the diagnosis of IPMNs. Different histologic subtypes of IPMNs display different frequencies of specific genetic mutations [2,15,16]. GNAS is more prevalent in intestinal subtype precursor lesions and colloid carcinomas, while KRAS is more frequent in pancreatobiliary precursor lesions and tubular carcinomas. Dysplasia in IPMN is two-tiered (low and high grade). Certain molecular alterations are more often seen in high-grade IPMNs (e.g., TP53, PIK3CA, and/or PTEN) and these alterations may help to stratify the risk regardless of the subtype [17,18]. Thus, preoperative cyst fluid analysis with both MUC staining and mutational status may potentially assist with risk stratification.

In conclusion, recent evidence supports the clinical value of histological subtyping of IPMNs. The different subtypes represent distinct cancer pathways and have different malignancy risks and prognosis. However, accurate diagnosis of IPMN subtypes is still a challenge and complementary genetic analysis may be necessary. The pancreatobiliary-type IPMN has the highest risk for malignant transformation. Prognosis is worse for patients with tubular adenocarcinoma as compared to colloid carcinoma. Adjuvant chemotherapy seems to be beneficial in patients with tubular carcinomas. Furthermore, increased focus should be directed to the remnant pancreas in case of a pancreatobiliary or intestinal subtype.

Disclosure statement

No potential conflict of interest was reported by the author(s).