Goblet cell adenocarcinoma of the gallbladder: Report of two cases and a review

Abstract

Malignant tumors of the gallbladder are unusual.1 A set of malignant tumors have been reported to arise from the gallbladder, such as biliary-type adenocarcinoma, intestinal-type adenocarcinoma, mucinous adenocarcinoma, squamous cell carcinoma, sarcoma, and neuroendocrine tumor, etc. Adenocarcinoma is the most encountered histological subtype, representing almost 90% of all gallbladder malignancies, whereas neuroendocrine carcinoma (NEC) represents only 2.1%.2, 3 More than one-third of the NECs have an adenocarcinoma component (mixed endocrine non-endocrine neoplasm [MiNEN]).4 However, amphicrine tumors, known as goblet cell adenocarcinoma (GCA) according to the latest World Health Organization (WHO) classification of appendiceal tumors, are extremely rare in the gallbladder.5, 6 The appendix is almost the exclusive site of occurrence of this unique type of mixed endocrine–exocrine neoplasm.6 Its terminology and classification have gone through changes over the years. And its histogenesis and management remain controversial. However, several reports of extra-appendiceal GCA have been published. Two cases of GCA of the gallbladder with various morphological and immunohistochemical features are herein reported. A literature review of similar cases, terminology, and histogenesis of this tumor are studied. A 52-year-old female patient presented to the emergency room complaining of pain in her upper right abdomen. All routine laboratory tests upon admission were within normal limits, with the exception of an elevation of erythrocyte sedimentation rate (ESR) to 76 mm/h (normal range 0–20 mm/h). Abdominal ultrasonography showed stones in the gallbladder. Cholecystectomy was then performed, during which gallbladder wall thickening was noticed, with no liver or lymph node involvement. On gross examination, the size of the gallbladder was 8 cm × 3 cm with a thick wall and multiple polypoid formation of the inner wall measuring 2–3 mm in diameter; small gallstones were also noticed. Microscopic examination revealed a largely eroded gallbladder mucosa and foci of high-grade biliary intraepithelial neoplasia (BillN) (Figure 1A) and cholecystitis. A massive neoplastic proliferation invaded the gallbladder. Tumor cells were arranged in small tight clusters, nests, cords, and aggregates, exhibiting a goblet cell morphology (Figure 1B). The clusters, representing more than 75% of cell proliferation, were solid or showed lumen formation. The goblet cells presented prominent intracytoplasmic mucin and a tiny, compressed nucleus. These cells were significantly stained with Periodic acid-Schiff (PAS) and alcian blue (AB) (Figure 1C). There were minimal nuclear atypia and rare mitosis, and no Paneth cell metaplasia was observed. The stromal reaction was inconspicuous, and extracellular mucin pools were observed. The tumor cells extended from the mucosal layer to the muscular layer and the subserosa. Perineural invasion and emboli (Figure 1D) made up of goblet cell carcinomatous cells were seen. The resection margin was not negative for tumor invasion. Chromogranin-A (Figure 2A), synaptophysin (Figure 2B), mucin 1 (MUC1), mucin 2 (MUC2), cytokeratin-7 (CK7), and caudal-related homeobox gene 2 (CDX-2), were positive for tumor cells. However, CD56 staining was negative, while CK20 labeled some of the tumor cells. According to the 2019 WHO classification for appendiceal tumors, the patient was diagnosed with grade 1 GCA. Appendectomy was performed to look for a possible appendiceal origin for this tumor; however, the entire appendix was completely free of any tumor cell proliferation. The patient was finally diagnosed with a primary grade 1 GCA of the gallbladder. Laparoscopic cholecystectomy was performed in an 86-year-old male who had symptomatic cholelithiasis. During the surgery, gallbladder wall thickening was observed. There was no liver or lymph node involvement. On gross examination, the gallbladder was 7 cm × 4 cm in size with wall thickening. The mucosa showed a granular surface. On microscopic examination, the epithelium presented small foci of intestinal metaplasia and high-grade BillN lesions. The mucosa showed florid follicular cholecystitis. It was largely ulcerated by large extracellular mucin pools and dyshesive mucinous cells (Figure 3A). The tumor proliferation extended through the entire gallbladder wall to the serosa, and arranged in large confluent solid sheets, irregular small round, and oval clusters of goblet-like cells (Figure 3B). Less than 25% of the tumor cells presented as infiltrative small nests and clusters of goblet-like cells. There was focal loss of intracytoplasmic mucin and nuclear atypia (Figure 3C). The stroma was desmoplastic with large extracellular mucin pools. These cells were deeply stained with PAS and AB. Paneth cell metaplasia was absent and perineural invasions and emboli were seen (Figure 3D). Tumor cells expressed CK7 and CDX-2, and were negative for chromogranin-A, synaptophysin, CK20, CD56, and MUC2. However, some epithelial cells at the neck of the gallbladder expressed chromogranin-A. The cystic lymph node was largely metastatic. The diagnosis was compatible with a grade 3 GCA according to the WHO classification for appendiceal tumors. The patient was lost to follow-up after surgery; therefore, gallbladder presentation of an occult primary appendiceal GCA could not be ruled out. It remained the first diagnosis, although the CK7/CK20 profile was in favor of a primary gallbladder tumor. GCA is a special type of mixed neuroendocrine–exocrine neoplasm that is almost exclusively found in the appendix.1 This tumor was seen in 0.3%–0.9% of appendectomies, accounting for approximately 14% of all malignancies of the appendix and 35%–58% of all appendiceal neoplasms.6 Extra-appendiceal GCAs, including that of the gallbladder, are extremely rare. They remain poorly studied and understood. Subbuswamy et al first introduced in 1974 the term “goblet cell carcinoid” to describe the primary cell type that morphologically resembled intestinal goblet cells with an admixed population of argentaffin cells.7 In 1978, Warkel et al proposed the term “adenocarcinoid”.8 Histologically, different terms have been used to describe this tumor, including goblet cell carcinoma, goblet cell adenocarcinoma, combined carcinoma, amphicrine neoplasm, amphicrine carcinoma, and crypt cell adenocarcinoma, causing additional confusion regarding its staging and treatment.6 As the survival rate of these tumors has been found to be correlated with the carcinomatous component, they are currently regarded as variants of adenocarcinoma.9, 10 The WHO Classification of Tumours in 2019 named these tumors as GCAs.4 This terminology was proposed to avoid confusion among pathologists in staging these tumors as well-differentiated neuroendocrine tumors and among oncologists in treating them with inappropriate adjuvant chemotherapeutic regimen.9, 11 For extra-appendiceal GCA, the terminology remains controversial. Regarding both of our cases, the same terminology adopted by the 2019 WHO classification for appendiceal GCA is used. Real extra-appendiceal GCAs are rare; GCA diagnosed in locations other than the appendix is an extra-appendiceal presentation of an occult primary appendiceal tumor.6 Among the 16 cases of putative extra-appendiceal GCAs reported by Gui et al, primary appendiceal GCA was confirmed in only four cases.12 In 10 cases, though the appendix was surgically removed, it was not available for review. The appendix was found to be tumor-free in only two cases, but the entire appendix was available for histological analysis in only one patient and is the only case of true extra-appendiceal GCA in that study.12 Amphicrine tumor of the gallbladder has been reported in three cases, including the first case by Muto et al.13 Amphicrine tumor of the gallbladder refers to as a composite tumor composing of goblet cell adenocarcinoid, typical carcinoid tumor, and ordinary adenocarcinoma.13 The second case, refers to as a signet-ring cell carcinoid, was reported by Papotti et al.5 Whereas the third case was reported by Sakaki et al and was identified as gallbladder adenocarcinoma with extensive Paneth cell metaplasia and florid neuroendocrine cell nests.14 However, in these three reports the appendix was not histologically examined. The appendix was entirely examined in our first case, which was found to be tumor-free. Thus, this case was diagnosed to be a true extra-appendiceal GCA. However, in the second case the appendix could not be examined since the patient was lost to follow-up. Therefore, despite the presence of intestinal metaplasia and high-grade BillN, which suggested a primitive gallbladder GCA in that case, metastasis from the appendiceal malignancy could not be excluded and remained the most probable diagnosis. Regarding the histogenesis of GCA, it is thought to arise from the pluripotent intestinal stem cells at the base of crypts that are able to undergo both mucinous and neuroendocrine cell differentiation.15 In the gallbladder, two hypotheses might clarify the histogenesis of extra-appendiceal GCA. The first suggests that amphicrine GCA might develop from the few argentaffin cells present in the mucinous glands of the gallbladder neck. These cells express chromogranin-A.16 The second hypothesis states that different gallbladder carcinoma cell phenotypes are the consequences of intestinal metaplastic differentiation in the presence of chronic inflammation. This hypothesis is supported by the lineage-plasticity model.17 In our study, inflammation and epithelial metaplasia were present in both cases. Ultrasounographic and CT features of GCA of the gallbladder are nonspecific and may mimic those of gallbladder carcinoma or cholecystitis. Wall thickening of the gallbladder wall and polypoid mass are the most encountered features.2, 14 While in the appendix, GCA appeared as an irregular firm nodular thickening of the appendix; in the gallbladder GCA appeared either as a polypoid mass as in the first case we presented and in the case reported by Sakaki et al,14 or as a surface granulation and wall thickening due to diffuse infiltration into the wall as described in the second case of the present study and the one reported by Papotti et al.5 Determining the size of the tumor along its longest axis in extra-appendiceal GCA remains difficult, which results from the diffuse pattern of infiltration of these tumors and might require the examination of the entire gallbladder.6 To be classified as GCA, a tumor must include at least one component of classic low-grade goblet cell adenocarcinoma, which were observed in both of our cases. Low-grade tumor presents as small clusters, nests, or cords of tumor cells, exhibiting a goblet cell morphology with a tiny crescentic nucleus and prominent intracytoplasmic mucin.6 Positivity for mucicarmine, PAS, PAS with diastase, and AB is seen in goblet-like cells.18 A distinct component of small to intermediate cells, with eosinophilic finely vacuolar or granular cytoplasm, vesicular nuclei, prominent nucleoli, and mild cytologic atypia, characteristic of typical carcinoids, as well as scattered Paneth cells can be seen.15 Tubular fusion and architectural disarray may be encountered. Nuclear atypia is mild and mitoses are uncommon. Extracellular mucin is often present and sometimes abundant, as in Case 2. High-grade histological features comprise infiltrating, dyshesive tumor cells. These cells may be disposed in confluent sheets, cribriform masses, and large aggregates of goblet-like or signet ring-like cells.4 Prominent desmoplastic stromal response, high-grade cytological features, elevated mitotic rate with atypical mitotic figures, and necrosis may be present in high-grade areas. Vascular and perineural invasion are frequent. Case 2 presented features of grade 3 GCA with numerous vascular and perineural invasions. Many grading systems have been proposed for the GCA tumor.9, 10, 19, 20 Tang et al19 proposed a three-tier grading system for GCA based exclusively on morphological features of the tumors, in which GCA was subdivided into the following three groups: group A, well-defined goblet cells arranged in clusters or cohesive linear pattern, with minimal architectural distortion and cytologic atypia, and minimal to no stromal desmoplasia; group B, goblet cells or signet ring cells arranged in irregular clusters with significant cytologic atypia, stromal desmoplasia, and destruction of the appendiceal wall; and group C, with the presence of at least focal signs of goblet cell morphology with a component (>1 low-power field or 1 mm2) identical to poorly differentiated adenocarcinoma.19 However, this grading system had shortcomings and significant interobserver variability. Therefore, Lee et al introduced 7 years later a two-tier grading system based on the presence of cytologic atypia, stromal desmoplasia, and solid growth pattern.20 Currently, the 2019 WHO classification adopted the three-tier system suggested by Yozu et al.9 The tubular and clustered components were quantified as low-grade components under this grading scheme. Tumors with low-grade components of 75% or higher, 50%–75%, and <50% were categorized as low-grade (grade 1), intermediate-grade (grade 2), and high-grade (grade 3), respectively. In the appendix, this grading system predicts overall survival regardless of the tumor stage.10 Unlike the classic well-differentiated neuroendocrine tumor (WDNET) showing strong and diffuse immunoreactivity for neuroendocrine markers, GCA often shows scattered positive cells only, as in our first case. Expressions of neuroendocrine markers are necessary for the diagnosis of GCA.4 Therefore, in Case 2 the diagnosis was made based on morphological criteria only regardless of the negative neuroendocrine markers. Similar to appendiceal adenocarcinomas, the positivity of CK20 and CK7 was 100% and 70.5% of GCA, respectively. CEA and CDX-2 expressions in these tumors are similar to that of colonic adenocarcinoma.15 Grades 1 and 2 GCAs express MUC2. However, in grade 3 GCA, MUC2 expression is lost and MUC1 is overexpressed. MUC2 was expressed in both of our cases. In Case 1, tumor cells were positive for CK7, and CK20 labeled some tumor cells. While in Case 2, tumor cells were positive for CK7 and negative for CK20. The role of Ki-67 as a prognostic factor for GCA of the appendix was recently studied by Liu et al, which included 12 patients and revealed no prognostic significance for goblet cell carcinoid tumors using Ki-67.21 Thus, Ki-67 is not mandatory for a diagnosis of GCA.6 However, it ranged from 10% to 70% in 12 cases with extra-appendiceal GCA, as reported by Lin et al.22 In our cases, the Ki-67 index was not assessed. Currently, it is well established that GCAs are distinct entities different from adenocarcinomas and carcinoid tumors, since these tumors have various molecular pathways. Mutations in typical colorectal cancer-related genes (eg, KRAS, APC, and TP53) were not found in appendiceal GCA. Besides, Lin et al concluded that gene expression patterns in extra-appendiceal GCA were different from those of colorectal and small intestinal adenocarcinoma with cohesive of a signet ring cell component.22 This indicates that GCA are genetically distinct from colorectal-type adenocarcinomas. However, GCA had mutations in genes associated with the Wnt-signaling pathway (CTNNA1, CTNNB1, and NOTCH1).23 No molecular study has been performed for the three previous cases of gallbladder GCA.5, 13, 14 Given the absence of the required technical platform, the molecular study could not be carried out in our cases. But it might be useful to compare the molecular profile of gallbladder GCAs with appendicular GCAs, gallbladder adenocarcinomas, and WDNET. We propose that primary GCA of the gallbladder should be staged using the TNM staging system designed for gallbladder adenocarcinoma as for appendiceal GCA, which, according to the American Joint Committee on Cancer (AJCC) Cancer Staging Manual 8th Edition,24 are staged as appendiceal adenocarcinoma. However, it may be difficult to make evidence-based decisions regarding the classification and staging system due to the rarity of the primary GCA of the gallbladder. Extra-appendiceal GCA, including GCA of the gallbladder, are unique and independent tumors with distinct histogenesis and molecular pathways. Unfortunately, they are underdiagnosed due to the lack of knowledge of these entities in both clinicians and pathologists. In fact, GCA was wrongly considered as exclusively appendiceal tumors and extra-appendiceal GCA were taken for classic adenocarcinomas or carcinoid tumors. Extra-appendiceal GCA must meet strict histopathological criteria, and the appendix must be entirely examined in these cases to eliminate extra-appendiceal metastases developing from appendiceal GCA and to confirm the primary origin of extra-appendiceal GCA. A better knowledge of these tumors and their diagnostic criteria by pathologists is mandatory to not miss the diagnosis. In conclusion, GCA is an uncommon, distinct entity of gallbladder tumor, which is diagnosed essentially based on morphological features made of goblet-like cell clusters. Expressions of neuroendocrine markers are not mandatory for its diagnosis. Metastases from appendiceal GCA should be ruled out by examining the entire appendix histologically. It is essential to accurately diagnose primary GCA of the gallbladder, as it is more aggressive than WDNET, and should be managed as an adenocarcinoma. Primary GCA of the gallbladder deserves to be recognized and properly diagnosed by pathologists and to be introduced in the WHO classification of gallbladder tumors. All authors declare that there are no competing interests. Generative artificial intelligence technology (available from: https://www.deepl.com/fr/translator; https://quillbot.com/grammar-check) was used in translation, paraphrasing, and English polishment in this work.