The present cross-sectional study was conducted over a period of six years from 2007 to 2012 in the Department of Pathology, Shrimati Kashibai Navale Medical College and General Hospital, Pune, Maharashtra, India. Ethical clearance was obtained from the Institute’s Ethical Committee. All cases of cholecystectomies received for histopathological examination were included. The relevant clinical details, investigation findings, histomorphological features and final diagnosis were noted. One section each from body, fundus and neck of the GB and additional section from grossly abnormal appearing areas was taken. Section from the cystic duct and lymph node was taken if the GB contained the tumour. Lymph nodes were searched along the GB neck. Sections were further processed as routine surgical specimen. They were then stained with Haematoxylin and Eosin (H&E) and representative slides were studied. All the lesions in the surrounding mucosal lining epithelium of non neoplastic and neoplastic GB were documented.

A technique of manual tissue array confirming sufficient representation of different areas in the cores was used for IHC [9]. The primary antibodies used were MUC1 (Clone Ma 695, Novocastra) and MUC5AC (Clone CLH2, Novocastra). Sections from lung and muscle tissue were used as positive and negative controls respectively for MUC1. Sections from gastric mucosa and tonsils were used as positive and negative controls respectively for MUC5AC. MUC1 was expressed in the luminal border and cytoplasm of cancer cells. MUC5AC was mainly expressed in the cytoplasm of cancer cells.

Staining results were estimated semi quantitatively by agreement of two pathologists on the basis of the percentage of positive cells: negative-<10%, 1-10%–25%, 2-26%–50%, 3->50%. For statistical analysis, cases with staining in 10% or more of the cells were considered.

A correlation was attempted between clinical, immunohistopathological findings using Pearson’s Chi-square test and Fisher-exact test. Software program “The Primer of Biostatistics 5.0” (manufactured by McGraw-Hill) was used for analysis of MUC1 or MUC5AC expression. The results were considered to be significant when the p<0.05.

Out of 16072 histopathological specimens received, 629 (3.91%) were GBLs. Percentage of calculous cholecystectomies (71.45%) was higher than acalculous cholecystectomies (28.64%). Gall stones were present more commonly in females (63%) in the age group of 41-50 years.

Age and sex distribution of non neoplastic and neoplastic GBLs is given in [Table/Fig-1].

Out of 605 of non neoplastic GBLs, 32 (5.29%) expressed MUC1 while 515 (85.12%) cases expressed MUC5AC [Table/Fig-2]. All cases of Acalculous Cholecystitis (ACC) (160) were negative for MUC1 [Table/Fig-3]. Expression rate for MUC5AC was significantly lower in ACC (73.12%, 117/160) as compared to calculous cholecystitis (93.26%, 346/371) (χ²=38.811; p<0.001) [Table/Fig-3].

Out of 24 cases of neoplastic GBLs, 20 cases (83.33%) showed positivity for MUC1 and nine cases (37.5%) were positive for MUC5AC [Table/Fig-4].

a) MUC1

Positive expression rate of MUC1 was significantly higher in neoplastic lesions (83.33%, 20/24) as compared to non neoplastic GBLs (5.29%, 32/605) (χ²=175.255; p<0.001). The rate of MUC1 expression was significantly higher in GBC (18GBC+3CIS) (85.71%, 18/21) than chronic cholecystitis (4.71%, 25/531) (χ²=173.440; p<0.001). Expression rate for MUC1 was significantly lower in calculous chronic cholecystitis (CCC) (6.47%, 24/371) as compared to GBC (18GBC+3 CIS) (85.71%, 18/21) (χ²=122.357; p<0.001)

b) MUC5AC

Rate of MUC5AC expression was significantly higher in non-neoplastic lesions (85.12%, 515/605) as compared to neoplastic lesions of gallbladder (37.5%, 9/24) (χ²=34.302; p<0.001). The positive rate of MUC5AC expression was significantly lower in GBC (18GBC+3CIS) (28.57%, 6/21) than chronic cholecystitis (463/531) (χ²=49.850; p<0.001). Positive expression rate of MUC5AC was significantly higher in CCC (93.26, 346/371) as compared to GBC (18GBC+3CIS) (28.57%, 6/21) (χ²=83.849; p<0.001).

Expressions of MUC1 and MUC5AC and their correlation with clinicopathologic parameters of malignant GB tumours are depicted in [Table/Fig-5].

Meticulous examination of the surrounding mucosal lining epithelia of the non neoplastic and surrounding lining epithelia of the neoplastic gallbladder mucosa showed presence of hyperplasia, metaplasia, dysplasia and CIS [Table/Fig-6]. Combination of various lesions was also noted in some cases [Table/Fig-7]. MUC1 and MUC5AC expression in various GBLs is shown in [Table/Fig-8,9] respectively.

[Table/Fig-8]:

MUC1 expression in GBL: a) Hyperplasia showing +1 positivity (40X); b) Metaplasia showing +1 positivity (40X); c) Dysplasia showing +2 positivity (10X); d) CIS showing +3 positivity (40X); e) Invasive carcinoma showing +3 positivity (40X). Arrow denotes the +1 and +2 positivity in the same photo photomicrograph.

[Table/Fig-9]:

MUC5AC expression in GBL: a) Hyperplasia showing +2 positivity (10X); b) Metaplasia showing +3 positivity (40X); c) Dysplasia showing +1 positivity (40X); d) Invasive carcinoma showing +1 positivity (40X).

The positive rate of MUC1 expression was significantly higher in GBC (83.33%, 15/18) than that in hyperplasia (4.35%%, 11/253) (χ²⁼111.935, p<0.001). The positive rate of MUC5AC expression was significantly higher in hyperplasia (91.70%, 232/253) than that in GBC (16.67%, 3/18) (χ²⁼75.744, p<0.001). Pyloric metaplasia was more common (29.38%, 156/531) type of metaplasia seen in chronic cholecystitis than intestinal metaplasia (13.37%, 71/531). There was significant difference between rate of expression of MUC1 in pyloric metaplasia (33.20%, 84/253) and GBC (83.33%, 15/18) (χ²⁼16.117, p<0.001). No case of intestinal metaplasia showed expression for MUC1. Rate of expression for MUC5AC was significantly higher in metaplastic change (83.64%, 276/330) as compared to GBC (16.67%, 3/18) (χ²⁼44.037, p<0.001). Both pyloric and intestinal metaplasia showed positivity for MUC5AC with no significant difference in the expression rates between them (χ²⁼0.258, p=0.611). Rate of expression for MUC1 was higher in dysplasia (85.71%, 18/21) as compared to GBC (83.33%, 15/18) though not statistically significant (χ²⁼0.057, p=0.811). MUC5AC expression was higher in dysplasia (57.14%, 12/21) than that in GBC (16.67%, 3/18) but this was not statistically significant (χ²⁼5.108, p=0.024). All cases of CIS showed MUC1 expression. MUC5AC expression was higher in CIS (25%, 3/12) as compared to GBC (16.67%, 3/18) though not statistically significant (χ²⁼0.009, p=0.926). The percentage of cases showing MUC1 expression increased as the severity of disease progressed from hyperplasia to CIS (χ²⁼150.494, p<0.001). The percentage of cases showing MUC5AC expression decreased as the severity of disease progressed from hyperplasia to CIS (χ²⁼55.431, p<0.001).

Only a small fraction of GBL encountered is neoplastic. However, chronic cholecystitis if left alone may progress through a series of premalignant changes to invasive cancer. This has led to recent studies insisting on the importance of precursor lesions and apomucin expression in cholecystectomy specimens [4,5,10]. Frequently calculous and acalculous cholecystitis present a large range of associated lesions such as hyperplasia, metaplasia, and dysplasia which have cancerous potential [10]. Our study included 84.42% (531/629) cases of chronic cholecystitis.

Gallstones are considered to be the most important risk factor for developing GBC, being reported in 70-98% cases of GBC in literature [11]. However, autopsy studies show that only 1-4% of cholelithiasis cases develop GBC as compared to 0.2% of acalculous cholecystitis [11]. In our study, 3.38% (21/621) of GB lesions showed presence of GBC (18GBC+3CIS). Also, 33.33% (6/18) of GBC cases were associated with gall stones. Recent studies provide quintessential evidence that chronic irritation either due to gall stones or due to changes in the bile owing to the reflux of pancreatic juice into the common bile duct as the basis for gall bladder carcinogenesis [3].

The results of our study strongly suggest that chronic cholecystitis produces a series of mucosal pathological changes which represents the precursor lesions of GBC [12]. We examined the MUC1 and MUC5AC expression in various GBL immunohistochemically. Few studies can be found describing the expressions of MUC1 and MUC5AC and their clinicopathologic significance in GBL in literature [4,13,14]. In our study, the percentage of cases showing MUC1 expression increased and percentage of cases showing MUC5AC expression decreased as the severity of disease progressed from hyperplasia to CIS [5]. Our study gives some evidence to support the pathogenesis of malignant GB lesions which progresses through epithelial hyperplasia, dysplasia, CIS and invasive cancer as seen in literature [5]. The positive expression rate of MUC1 was significantly higher and that of MUC5AC was significantly lower in GB adenoca than chronic cholecystitis in our study in accordance with findings of Xiong L et al., [5].

In present study, 18 cases were of carcinoma of which one case was poorly differentiated, eight cases were moderately differentiated and nine cases were well differentiated. Histomorphologically, 13 cases were of biliary type and two cases were of intestinal type. Two cases were of papillary adenocarcinoma, both of which were well differentiated. These tumours were named as intracystic (GB) papillary neoplasm with associated invasive ca according to 2010 WHO classification [15]. One case was carcinosarcoma with heterologous differentiation type which accounts for less than 1% of GBC [16]. Other histological variants of GBC were not encountered in present study. Also, reported were three cases of biliary intraepithelial neoplasia or CIS. There were three cases of adenomas of which one presented with multiple adenomatous polyps in the gall bladder as a part of familial adenomatous polyposis syndrome. A total of 2/3 adenomas showed low expression of MUC1 and all adenomas (3/3) showed MUC5AC expression [5]. This is similar to a study done by H. Mohan H et al., who reported 12 cases of GBC and two cases of adenoma out of 1100 cholecystectomies [1]. Sharma SP et al., recorded high incidence of 17.8% which could be related to geographical, ethnic or dietary factors [17]. Kim S-M et al., examined 54 cases of GBC of which 28, 21 and 5 were well, moderately and poorly differentiated carcinomas respectively [18].

In the present study GBC cases (2) with lymph node metastasis and T3 stage ca (7) cases were positive for MUC1. Moderately differentiated ca (8) and all T1 stage ca (6) cases were negative for MUC5AC expression. One case of carcinosarcoma of GB showed MUC1 positivity and lack of MUC5AC expression in epithelial component. There was no significant statistical correlation between age, sex or presence of calculi with expressive rates of MUC1 or MUC5AC. Earlier literature documents varying results of MUC1 expression in GB [5,18]. One study, mentions MUC1 expression of 56.8%, 38.6% and 4.6% in 25 well-differentiated 17 moderately and two poorly differentiated GBC respectively. However, there was no correlation between MUC1 expression and other clinicopathological features such as age, sex, T-stage and nodal status [18]. Another study reports significantly lower positive rates of MUC1 in cases with no metastasis of lymph node as compared to metastasis of lymph node, and invasion of regional tissues (p<0.01). The positive rates of MUC5AC were significantly higher in the well-differentiated adenoca cases than those in poorly differentiated adenoca cases (p<0.05) [5].

MUC1 was seen in 4.64% cases of secondary hyperplasia and 3.39% cases of adenomatous hyperplasia. 85.71% of dysplasia and 83.33% of GBC cases showed high MUC1 expression especially in invasive areas as seen in literature [18].

Pyloric metaplasia and intestinal metaplasia was seen in 40.22% (253/629) and 12.24% (77/629) cases of GBLs as seen in literature [19]. Pyloric metaplasia was seen in 224 cases of CCC, 15 cases of ACC and 14 cases of GB neoplastic lesions. Intestinal metaplasia was seen in 71 cases of ACC, five cases of ACC and one case of papillary adenocarcinoma of GB. In present study, no case of intestinal metaplasia showed MUC1 expression. However, 33.20% cases of pyloric metaplasia showed MUC1 positivity pointing towards similar characteristics of both lesions for tendency to express MUC1. Similar findings is noted in literature [3]. Duarte I et al., in his study of 162 cholecystectomy specimens found pyloric and intestinal metaplasia in 95.1% and 58.1% of cases [19]. Chronic inflammation owing to gall stones is mentioned in literature as being more responsible for causing both intestinal and pyloric metaplasia which subsequently leads to dysplasia and carcinoma transformation [3].

One study documented in literature showed lower MUC1 expression in T1 stage tumours than in higher stages. No correlation was observed between MUC1 expression and histopathologic type and grade in their study. However, they found angio-lymphatic invasion to be associated with depolarized MUC1expression [7]. In the present study, high MUC1 expression was correlated with less differentiated tumours. Xiong L et al., previously examined the correlation of MUC1 and MUC5AC in GB adenoca, finding that the expression of MUC1 is linked with T-stage (p<0.01) [5]. Additionally, increased MUC1 expression or decreased MUC5AC expression was associated with decreased overall survival. Furthermore decreased MUC5AC expression was found to be an independent prognostic predictor [5]. MUC1 expression in GB carcinoma may signify histological dedifferentiation, increased proliferative activity and invasiveness [13].

Above findings re enforce the fact that though GB specimens are commonly encountered in surgical pathology, GBC cases are uncommon. It can be stated that the expression of MUC1 and MUC5AC might be closely related to the pathogenesis of neoplastic and non-neoplastic GBLs.

Smaller sample size of the neoplastic GB lesions owing to its rarity.

In this study, 96.18% cases were non neoplastic GBLs of which chronic cholecystitis (87.77%) were predominant. 3.81% of the GBLs constituted for neoplastic lesions of which 75% were gall bladder carcinomas. GBL have significant female preponderance with non-neoplastic lesions having peak incidence during 5^th decade of life and neoplastic lesions during 6^th decade. MUC1 shows higher rates of expression in neoplastic GBLs. MUC5AC shows higher rates of expression in non neoplastic GBLs. Studying MUC1 and MUC5AC profile in every gall bladder neoplasm and its surrounding mucosa may help to understand the biology of the disease and tumourigenesis pathways enabling us to identify cancer risk patients which could aid in better management.