Intrahepatic bile duct mixed adenoneuroendocrine carcinoma: a case report and review of the literature

Malignancies affecting the extrahepatic bile duct are rare, accounting for around 0.1–0.2 % of all cancer diagnoses [1]. Of these cholangiocarcinoma is the commonest cause, with adenocarcinoma making up around 80 % of cases and other subtypes occurring much less frequently [2]. Rarely biliary ductal tumours with neuroendocrine differentiation arise within the extrahepatic bile ducts [3, 4]. Initial presentation may be associated with features of local disease (biliary colic, painless jaundice) or, occasionally, as a result of production of active hormones.

According to the latest World Health Organisation (WHO) classification system [5] biliary tract neuroendocrine tumours (BNET) are classified into neuroendocrine tumours (NET), neuroendocrine carcinomas (NEC), and mixed adeno-neuroendocrine carcinomas (MANEC). MANEC is a subtype that shows characteristics of both glandular and endocrine differentiation. MANEC of the biliary ductal system are extremely rare, with only a handful of published cases to-date [6, 7]. To date there have been no published cases of MANEC derived from the intrahepatic biliary ductal system. As a result little is known about their pathogenesis and outcomes.

We report a case of MANEC of the intrahepatic bile duct. Our aims are to describe its clinical presentation, and the detailed immunophenotypic characteristics of this tumour. Despite the paucity of cases in the literature, another objective is to compare the clinicopathologic characteristics of various biliary tumours, and predict outcomes of MANEC in biliary tumours.

We present a rare case of intrahepatic biliary MANEC and highlight the difficulty in establishing a pre-operative diagnosis [8]. The differential diagnoses in this case include intrahepatic cholangiocarcinoma, hepatocellular carcinoma and intrahepatic neuroendocrine carcinoma (Table 1 for clinicopathological comparison).

Cholangiocarcinoma occur most frequently in the hilar region, at the confluence of the right and left hepatic ducts. Those involving only the intrahepatic ducts are less frequent, 8 % of all cholangiocarcinomas in one large retrospective study [9]. The pattern of tumour growth and its anatomical location mean that obstructive jaundice is a late feature and tumour size tends to be larger with local infiltration, as occurred in our case [9‐11]. Blood or bile testing for tumour markers such as CA19-9 and carcinoembryonic antigen (CEA), are often of limited diagnostic utility in isolation as both markers lack sensitivity and specificity, as was evident in this case [12‐14].

Mass-forming intrahepatic cholangiocarcinoma may be difficult to differentiate from HCC on imaging. Dual-phase contrasted CT is frequently a useful tool in differentiating these pathologies with their characteristic radiological findings. Cholangiocarcinomas are typically hypo- or iso-attenuating relative to normal liver parenchyma at both arterial and portal venous phases, with enhancement only in the delayed phase [8, 15]. This finding reflects the hypovascular desmoplastic composition of cholangiocarcinoma. HCC on the other hand are more hypervascular and tend to show arterial phase enhancement with portal venous phase washout. The lesion in this case was unusual as contrast CT imaging showed arterial enhancement with mild washout at portal venous phase suggestive of HCC, yet the MRI liver demonstrated peripheral biliary obstruction, which is more in keeping with a bile duct neoplasm. One large retrospective study demonstrated that a significant proportion of mass-forming intrahepatic cholangiocarcinomas had arterial enhancement on dual-phase CT [16].

BNET are very rare and represent less than 0.3 % of all NET [3, 7]. They were first described in 1959, with only 150 cases published since then. The majority of these tumours are located in the extrahepatic biliary tracts, with smaller numbers originating at the bifurcation of the extrahepatic biliary tree [3, 7, 17]. In clinical practice, these lesions are difficult to diagnose and to distinguish from cholangiocarcinoma preoperatively. Certain clinical features can be useful in distinguishing one from the others. For instance, BNET tend to affect younger patients, whilst aggressive local invasion and distant metastatic disease is seen more frequently in adenocarcinoma. A pre-operative diagnosis can be made by examining brush cytology [18] or detecting raised serum tumour markers [7], though these methods have high false negative rates. A more precise pre-operative diagnostic technique is the use of spyglass endoscopy to obtain tissue biopsy from the lesion of the biliary ductal system, though availability of this investigative modality is often limited. Most of the time, surgical resection is the treatment of choice in the first instance, particularly when the lesion is deemed resectable on imaging, with a final diagnosis made following histopathological examination of the resected specimen. Pure BNET show typical histological features of NET and may show production of hormones, including somatostatin, serotonin and gastrin [7]. Somatostatin analogues such as octreotide, have been shown in clinical trials to be efficacious in improving symptoms and reducing tumour growth [19]. Prognostic data for BNET are limited due to their rarity. According to the most recent WHO classification, they should be graded as conventional NET taking into account in particular mitotic activity and Ki67 proliferative index [5, 20]: G1: mitotic count <2 HPF and/or Ki67 index <2 %; G2: mitotic count 2–20 per 10 HPF and/or Ki67 index 3–20 % and G3: mitotic count >20 per 10 HPF and/or Ki67 index >20 %. Overall outcomes appear to be better than in other types of biliary tract malignancies.

Our case differs from conventional BNET and cholangiocarcinoma due to the mixed but tightly intermingled neoplastic components on histological examination. The bulk of the mass is represented by a moderately differentiated NET, which stains diffusely for chromogranin and synaptophysin. A minor cell population scattered throughout the tumor and more prominently in its deep aspect closer to the nearby liver parenchyma is composed of cells with signet ring morphology. These cells highlight well with Alcian blue/periodic acid-Schiff (ALCIAN-PAS) and MUC1 staining, demonstrating acid mucus production by these cells and providing support to divergent glandular differentiation. This dual phenotype fits with the category of biliary MANEC and resembles the case described by Harada [6] with the adenocarcinomatous component found at the peripheral.

Biliary MANEC is exceedingly rare with only a few cases published in the literature [21‐26]. In one series of hepatobiliary MANEC, only 9 cases were identified as MANEC [6]. Two of these cases were categorised as hepatic hilar MANEC, with the rest originating from gallbladder and extrahepatic bile ducts. To date and to our knowledge there have been no published cases of MANEC originating from intrahepatic bile ducts. Biliary MANEC have a phenotype that morphologically resembles both adenocarcinoma and NEC. According to the updated WHO classification, it was suggested that to be classified as MANEC, at least 30 % of the main lesion excluding any broad invasive area had to be made up of each component [27]. Harada et al. found that the adenocarcinoma components were predominantly located on the surface of the tumours, whilst the neuroendocrine components were found deeper and associated more often with stromal, vascular and lymphatic invasion [6]. Our case is similar because the glandular component was more pronounced at the peripheral intraductal part of the tumour; whereas the infiltrating component was mainly moderately differentiated (G2) neuroendocrine (Ki67 up to 8 %). Nonetheless, the definition of malignancy in this peculiar category of tumours is still not well established [5].

As with Harada et al. [6], the proliferative index in our case was higher in the neuroendocrine component compared with the adenocarcinomatous population, suggesting that neuroendocrine is the population upon which prognosis depends.

Additional histopathological diagnoses to be considered in mixed differentiated intrahepatic tumours include intraductal papillary neoplasms (IPMN) and HCC with neuroendocrine differentiation. IPMN are a precursor lesion that may rarely show neuroendocrine differentiation [7], but are characterized by protruding papillae covered by dysplastic epithelium with mucin secretion. In our case, there was an abrupt passage from the normal bile duct epithelium to the neoplastic mass without any evidence of preneoplastic lesion. Moreover, the epithelial marker CA19.9 and MUC1 stained only part of the tumour suggesting that there are two separate and discrete albeit mixed components. MANEC is therefore more likely to be derived from the adenocarcinomatous cells, rather than from neuroendocrine populations [21]. HCC with neuroendocrine differentiation has been anecdotally reported and remains unrecognised in the most recent WHO classification system. Our tumour did not resemble HCC morphologically, and Hep-Par1 staining was negative [28].

With only a paucity of cases, it is difficult to fully ascertain the long-term prognosis of biliary MANEC. There had been a report of 45-month survival in one patient who underwent surgical treatment alone [29]. However, that patient had a composite adenocarcinoma (40 %) and NEC (60 %) of the common bile duct. Although the location of the tumour is distinguishably different from our case, both cases had low proliferative fractions (<10 % Ki-67-positive tumour cells), which have been shown to provide significant prognostic information [20].

In summary, we report an intraductal invasive neoplasm with divergent predominantly neuroendocrine differentiation arising from the right intrahepatic biliary duct in keeping with a biliary MANEC. To our knowledge, this is the first reported case of an intrahepatic biliary MANEC, with the few previously reported cases affecting the extrahepatic biliary ductal system. On-going research is necessary to increase our understanding of the pathogenesis of this rare disease.

Written informed consent was obtained from the patient for publication of this Case Report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.