Expression of Female Sex Hormone Receptors, Connective Tissue Growth Factor and HER2 in Gallbladder Cancer

Gallbladder cancer (GBC) is a highly malignant tumor with poorly understood etiology. An insight into phenotypic features of this malignancy may add to the knowledge of its carcinogenesis and pave the way to new therapeutic approaches. We assessed the expression of female sex hormone receptors (ERα, ERβ, PR), connective tissue growth factor (CTGF) and HER2 in GBC, and adjacent normal tissue (NT), and determined their prognostic impact. Immunohistochemical (IHC) expression of all biomarkers was performed in formalin-fixed, paraffin-embedded specimens in 60 Caucasian GBC patients (51 women and 9 men). ERβ, cytoPR and CTGF expression were found in 89%, 27%, 91% of GBC, and in 63%, 87%, 100% of NT, respectively. No ERα expression was found in GBC and NT. Strong (3+) HER2 expression by IHC or HER2 amplification was seen in five GBC (10.4%). A positive correlation was found between HER2 and CTGF and ERβ expression in GBC and matched NT. In the multivariate analysis, patient age >70 years, tumor size and ERβ expression in GBC was highly predictive for OS (p = 0.003). The correlation between HER2, CTGF and ERβ expression in GBC and NT may indicate the interaction of these pathways in physiological processes and gallbladder pathology.

Results patient characteristics. The study group included 60 GBC Caucasian patients (51 women and 9 men).

Discussion
We have performed a comprehensive analysis of female sex hormone receptors, HER2 and CTGF in GBC and in adjacent NT in a relatively large group of Caucasian patients. The lack of ERα expression in GBC and in adjacent NT confirms the results of studies published after 2007 [12][13][14] . ER expression demonstrated in some earlier reports was likely due to nonspecific staining with antibodies involving both ERα and ERβ receptors [6][7][8][9][10][11] . In these studies ER expression was demonstrated in both metaplasia, and GBC, irrespective of its differentiation 6,7,9 . The prognostic value of ER expression in GBC is contradictory. Two studies from India suggested that the expression of ER and PR in GBC does not have an impact on the prognosis 12,13 . In our study ERβ expression in GBC was correlated with a shorter OS, and in NT with a longer OS. This finding is intriguing but may be incidental owing to multiple comparisons. In some malignancies, including breast, ovarian and prostate cancers, ERβ plays a suppressive and anti-proliferative roles 29,30 . In contrast to ERβ, expression of cytoPR was higher in NT compared to GBC, and was associated with the co-existence of gallstones. Additionally, cytoPR expression was negatively correlated with CTGF expression. A few studies demonstrated PR expression (including cytoPR) in GBC 6,11,12 . In the Baskaran et al. study 11 PR was more often expressed in neoplastic compared to benign lesions, and in the Nakamura et al. study 6 , PR expression was lower in the metaplasia and high-grade GBC. Some studies did not demonstrate PR expression in GBC, which may be related to the use of antibodies detecting only nuclear PR expression (PR-A isoform), whereas isoform B (PR-B) is also expresses in the cell cytoplasm 13,14,31 . The different functions of three PR isoforms (A, B, and C) are well recognized in breast cancer [32][33][34] . PR-B activates expression of progesterone-dependent genes by palindromic-progesterone-response DNA elements related to the metabolism of sex hormones [32][33][34] .
High expression of ERβ GBC found in this study supports a possible role of anti-estrogen therapy in GBC, probably with a different approach than in breast cancer though. In obese postmenopausal women, adipose tissue is the main source of estrogen biosynthesis, and this hormone has been shown to increase cholesterol level in bile and decrease gallbladder contractility 35 . Likewise, physiological processes in premenopausal women, such as menstrual cycle phase and pregnancy, or contraceptive use, are accompanied by changes of gallbladder functions and increased gallstone formation 36,37 . A recent study postulated that cholesterol gallstones in women are related to differences in liver cholesterol metabolism in response to estrogen, a process mediated by up-regulating of the ESR1 (ERα coding gene) expression 38 . In other study ESR1 polymorphic variants: IVS1-397C > T, ESR1 IVS1-351A > G and ESR2-789 A > C correlated with GBC risk, mediated through gallstone dependent pathway 39 .
Tamoxifen is the oldest and most-prescribed selective estrogen receptor modulator in breast cancer patients. A Turkish study demonstrated increased risk of gallstone formation in postmenopausal breast cancer patients administered tamoxifen 40     www.nature.com/scientificreports www.nature.com/scientificreports/ patients with ESR1-mutated breast cancer showed better response to fulvestrant, a hormonal protein degrader, compared to aromatase inhibitors 42 . Further research is warranted to assess the fulvestrant activity in GBC.
The present study did not show different CTGF expression in GBC and adjacent NT, or the prognostic value of this biomarker. The role of CTGF in GBC progression and its favorable prognostic impact was earlier reported in a Chilean study 28 . This effect may be attributed to the stromal response to the neoplastic process in an autocrine or paracrine manner 28 . Inconsistent results of both studies may be due to different etiology of GBC in the Latin American and Caucasian populations 3,4,6 .
HER2 overexpression or gene amplification occur in 12-15% of GBC 15,20,[22][23][24][25] . In our study, using the breast cancer criteria, IHC 2+ or 3+ expression was found in 15% of cases, more than a half of which (8.3%) showed a true HER2 positivity (IHC 3+ expression or FISH amplification). Aberrant HER family signaling may be important in the development and progression of GBC 15,20 . Some studies reported adverse prognostic impact of HER2 expression 16,[23][24][25][26]43 , but others, including ours, did not show such correlation 27,44 . A few studies have investigated anti-HER2 therapy in advanced GBC [45][46][47][48][49] . This approach was also attempted in bilary tract cancer (BTC) patients. Two phase II studies in unselected BTC patients did not show lapatinib activity 45,46 . In the MyPathway trial including seven HER2-positive BTC patients treated with the combination of the anti-HER2 antibodies, trastuzumab and pertuzumab, the objective response rate was 29% 47 . In the NCT02675829 clinical trial, the response rate for ado-trastuzumab emtansine in HER2 amplified BTC patients was 17% 48 . Recently, a basket trial showed the activity of a pan-HER tyrosine kinase inhibitor neratinib in HER2-mutant BTC patients 49 . In our study expression of HER2 and CTGF in GBC was positively correlated with their expression in surrounding NT. This finding may suggest a connection between these pathways in both physiological and pathological processes of the gallbladder. For example, in breast cancer, there is a progestin-independent relationship between the pathways for steroid receptors and growth factor receptors 50 .
Similarly to other GBC studies, age over 70 years and higher pT stage adversely impacted OS 1-5 .
Our study contributes to the current knowledge on the biology of GBC, but owing to its retrospective nature and the relatively small group of patients, should be interpreted cautiously. In recent years, somatic profiling with next-generation sequencing has identified several genes, including TP53, SMAD4 and KRAS, which seem to play a role in GBC carcinogenesis [51][52][53] . At present, a number of agents targeting new pathways are being investigated in clinical trials in GBC patients. Our tissue material and clinical database may be exploited in future scientific projects.

Materials and Methods
Study population. This study was approved by the Institutional Review Board of the coordinating center, the Military Institute of Medicine in Warsaw, Poland. The patients were diagnosed and underwent surgery between 2004 and 2016 in four oncology centers in Poland. Demographic, clinicopathologic, and clinical follow-up data were extracted from medical records. All data were coded to secure full protection of personal information, therefore, patient consent was not sought. All research was performed in accordance with relevant guidelines and regulations. immunohistochemical analysis. The starting material from each patient was an archival formalin-fixed, paraffin-embedded (FFPE) surgical specimen of the primary GBC. The pathologic diagnosis was confirmed by a board-certified pathologist (RP) who reviewed FFPE tissue sections stained with hematoxylin and eosin. A representative paraffin block from each specimen was chosen for immunohistochemical analysis (IHC). The two biopsy specimens of GBC and surrounding NT ("tissue core") were placed on the previously prepared tissue-free paraffin blocks ("recipients"). Tissue microarrays were constructed using Manual Tissue Arrayer I by Beecher Instruments (MTAI, K7 BioSystems). IHC was performed on 4 μm thick tissue microarray sections. The staining was conducted according to the manufacturers' protocols (Table 3). ERα, ERβ and PR were evaluated in the cell nuclei or/and cytoplasm. The occurrence of nuclear and/or cytoplasmatic ERα, ERβ, PR reaction in at least 1% cells was considered a positive reaction. CTGF expression was evaluated in the cytoplasm and cell membrane, and HER2 expression in the cell membrane. For all biomarkers the intensity of staining was defined as weak (1), moderate (2), or strong (3). The H-score was calculated for each biomarker by the formula: 3 × % strong cellular staining (cytoplasmic, nuclear and/or membranous) + 2 × % moderate staining + 1 × % weak staining. This made a range of 0-300. Additionally, ERα, ERβ and PR were scored by the Allred method. This system is graded on a scale of 0 to 8, with 0 indicating a completely negative result and 2 to 8 used as a means of semiquantifying the immunoreactivity 54 . Based on breast cancer criteria for HER2-positivity only samples showing strong  Table 3. Antibodies, dilutions and methods of evaluation. RU: ready to use; HIER: Heat-Induced Epitope Retrieval; SQ: semiquantitative; BR: breast cancer; SM: smooth muscle.