Trail Overexpression Inversely Correlates with Histological Differentiation in Intestinal-Type Sinonasal Adenocarcinoma

Introduction. Despite their histological resemblance to colorectal adenocarcinoma, there is some information about the molecular events involved in the pathogenesis of intestinal-type sinonasal adenocarcinomas (ITACs). To evaluate the possible role of TNF-related apoptosis-inducing ligand (TRAIL) gene defects in ITAC, by investigating the immunohistochemical expression of TRAIL gene product in a group of ethmoidal ITACs associated with occupational exposure. Material and Methods. Retrospective study on 23 patients with pathological diagnosis of primary ethmoidal ITAC. Representative formalin-fixed, paraffin-embedded block from each case was selected for immunohistochemical studies using the antibody against TRAIL. Clinicopathological data were also correlated with the staining results. Results. The immunohistochemical examination demonstrated that poorly differentiated cases showed a higher percentage of TRAIL expressing cells compared to well-differentiated cases. No correlation was found with other clinicopathological parameters, including T, stage and relapses. Conclusion. The relationship between upregulation of TRAIL and poorly differentiated ethmoidal adenocarcinomas suggests that the mutation of this gene, in combination with additional genetic events, could play a role in the pathogenesis of ITAC.


Introduction
Malignant tumors of the nasal cavity and paranasal sinuses account for 0.2% of all human primary malignant neoplasms, with an incidence of 0.1-1.4 new cases/year/100,000 inhabitants [1][2][3].
Adenocarcinomas account for 10-20% of all primary malignant neoplasms of the sinonasal tract [4,5]. Many of these have salivary gland origin, while others have histologic patterns resembling those of colon adenocarcinoma. This second type of sinonasal adenocarcinoma has been named intestinal-type adenocarcinoma (ITAC) and is responsible for less than 4% of the total malignancies of this region [6].
ITACs of the nasal cavity and paranasal sinuses can occur sporadically or are associated with occupational exposure to hardwood and leather dusts [7]. Exposure to wood and leather dusts increases the risk of adenocarcinoma by 500fold [8,9]. Findings from several studies have suggested clinical differences between ITAC arising in individuals with occupational dust exposure and ITAC arising sporadically. In fact tumors related to occupational exposure affect men in 85-90% of cases, showing a strong tendency to arise in the ethmoid sinuses [10][11][12].
ITACs are aggressive tumors characterized by frequent local recurrences, low incidence of distant metastases, and an overall mortality of approximately 53% [12]. Histopathological grading appears to be a significant prognostic indicator [11][12][13]. Surgery is considered the standard treatment.
ITAC seems to be preceded by intestinal metaplasia of the respiratory mucosa, induced by hardwood dust, leather dust, and other unknown agents, which is accompanied by a switch to an intestinal phenotype. The molecular mechanisms involved in metaplastic transformation of terminally differentiated epithelium to a phenotypically different epithelium are largely unknown. The morphological appearance of these tumors is variable, and they may resemble conventional colorectal adenocarcinoma. The similarities between ITAC and intestinal adenocarcinoma involve their ultrastructural and immunohistochemical aspects [14][15][16][17].
Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) is a TNF-family member, found in a variety of tissues [33] and with a conditional expression in several immune effector cells [34,35]. To date, five different receptors have been identified to interact with TRAIL: TRAILR1 (DR4), TRAIL-R2 (DR5), TRAIL-R3 (DcR1), TRAIL-R4 (DcR2), and osteoprotegerin. DR4 and DR5 are two different death domain-containing membrane receptors, whereas DcR1 and DcR2 are two decoy receptors that compete for TRAIL binding with DR4 and DR5 [36][37][38][39]. The final effect of the TRAIL signaling is the induction of apoptosis by the intrinsic death pathway, recruiting the inactive form of caspase-8 [40]. Expression of TRAIL and its receptors has been detected in various human tumors, suggesting that TRAIL signaling pathway is involved in endogenous tumor surveillance [40], but the mechanism of how TRAIL and its receptors contribute to carcinogenesis remains unknown.
To gain further insight into the phenotype and possible mechanisms of ethmoidal ITAC, we investigated the expression of TRAIL, correlating with clinicopathological data.

Patients Selection.
The samples of 23 primary ethmoidal intestinal-type adenocarcinomas were retrospectively retrieved from the archives of the Institute of Pathology, Marche Polytechnic University, Ancona, Italy. All the tumors in which the diagnosis of ITAC was confirmed were subsequently subtyped, as papillary, colonic, solid, mixed, or mucinous types, as described by Barnes [12].
The medical records of these patients were reviewed. Inclusion criteria were complete clinical data, uniformity of histological differentiation throughout the tumor sample, and the availability of sufficient material from the primary tumor for investigations. Patients with previous or synchronous second malignancies or with previous radiation therapy or chemotherapy were excluded from the study. The follow-up time ranged from 2 to 10 years (mean 4.8 years).

Immunohistochemistry.
Four-micrometer serial sections from formalin-fixed, paraffin-embedded blocks of tumour representative areas were cut for each case. Immunohistochemical staining for TRAIL was performed using the following antigen retrieval system. Sections were deparaffinized in two changes of xylene for 10 minutes each, then, were rehydrated through graded alcohols, and immersed in 0,3% hydrogen peroxide in methanol for 30 minutes to block endogenous peroxidase activity. Sections were then washed in PBS. The tissue sections were placed in a microwave oven (Philips, Cooktyronic M720, 700 W) in a plastic Coplin jar filled with 10 mM sodium citrate buffer (pH 6,0) at 5minute interval, the fluid level in the Coplin jar was removed from the microwave oven and allowed to cool. Slides were incubated overnight with a 1 : 50 dilution of the primary mouse antihuman TRAIL monoclonal antibody (Dako DO-7, Glostrup, Denmark). A biotin-streptavidin detection system was used with diaminobenzidine as the chromogen. Slides were washed twice with PBS and incubated with the linking reagent (biotinylated anti-immunoglobulins) for 15 minutes, at room temperature. After rinsing in PBS, the slides were incubated with the peroxidase-conjugated streptavidin label for 15 minutes at room temperature. The sections were again rinsed in PBS and incubated with diaminobenzidine for 10 minutes, in the dark. After chromogen development, slides were washed in two changes of water and counterstained with a 1 : 10 dilution of hematoxylin. The sections were then dehydrated, cleared in xylene, and mounted.
TRAIL expression and location were evaluated on histological section using a Leitz Orthoplan microscope equipped with a X 400 objective. The percentage of TRAIL positive cells was evaluated from a minimum of 1,000 cells in each case. Only nuclear staining of epithelial cells was observed, and the nuclei with a clear brown color, regardless of staining intensity, were regarded as TRAIL positive.
A negative control for TRAIL immunostaining was performed in all cases by omitting the primary antibody, which, in all instances, resulted in negative immunoreactivity. The positive control consisted of staining a human normal mammary gland epithelium.

Scoring of Preparations.
Both the histological diagnoses and evaluation of the positivity for TRAIL were carried out indipendently by two of the authors (C.R. and A.Z.). Immunohistochemical labeling for TRAIL was classified as positive, when more than 5% of nuclei or cells were stained.

Statistical Analysis.
The following pathological and clinical parameters were analyzed: tumour grade of differentiation (G1,G2,G3), tumour extension (T1,T2,T3,T4), tumor stage (I, II, III, IV), and TRAIL immunoreactivity. Differences in TRAIL immunoreactivity between the different groups were compared by the nonparametric Kruskal-Wallis test. Statistical significance was set at < 0.05. All statistical analyses were performed using the SPSS statistical package (SPPS Inc., Chicago, IL).

Clinical
Findings. There were 21 males and 2 females, with a mean age of 66.3 years (range 54-77). There were 8 grade I, 8 grade II, and 7 grade III adenocarcinomas. 3 patients were in I stage, 7 patients were in II stage, 11 patients were in III stage, All the patients had a known history of occupational exposure to hardwood dust, and intestinal-type adenocarcinoma was localized in all cases in the ethmoid region as confirmed by endoscopic and imaging (TC and/or MR) evaluation.
The main clinicopathologic features of the patients and the oncoprotein expression are summarized in Table 1.

3.2.
Immunohistochemistry. The cases classified as ITAC showed a variable cellular appearance and were composed of a mixture of tall columnar absorptive cells, atypical stratified cylindrical cells similar to the cells seen in conventional colorectal adenocarcinoma, goblet cells, and large round to polygonal nondescriptive epithelial cells.

TRAIL Expression.
The percentage of neoplastic cells immunoreactive for TRAIL ranged from 2% to 90%. In all cases, TRAIL expression was nuclear. There was a relationship between TRAIL overexpression and the histological grading. Indeed, poorly differentiated cases (G2 and G3) showed a higher percentage of TRAIL expressing cells in comparison to well differentiated (G1) cases. No correlation between TRAIL expression and other clinicopathological parameters, including T, stage and relapses was found (Figures 1 and 2).

Discussion
Despite their histological similarity to colorectal carcinomas, there is very little information about the molecular events involved in the pathogenesis of ITAC. Several crucial pathways of tumorigenesis have been identified in colorectal adenocarcinomas [18,19]. These pathways involve the mutation and inactivation of multiple oncogenes, tumor suppressor genes, and DNA mismatch repair genes including K-ras, APC, p53, MLH1, and MSH2 [18][19][20].
Working on the hypothesis that morphological similarities to colorectal adenocarcinomas might reflect equivalent genetic alterations, several authors have investigated the presence of activating mutations of Ras oncogenes and TP53 mutations in ITAC [21][22][23][24][25]. TP53 mutations were found in 18-44% of mostly occupational ITACs, whereas, K-Ras mutations were found in 10-15% of ITACs [21][22][23][24][25]. The results of these studies suggest that mutations of K-Ras and other Ras genes are relatively uncommon in ITAC, and similarly, TP53 mutations in ITACs have not been widely demonstrated.
Other studies have shown that K-Ras mutation and Cerb-2 expression could be associated with more aggressive ITACs [16,26].
Licitra et al. found the existence of two genetic ITACs subgroups, defined by differences in TP53 mutational status or protein functionality, that strongly influence pathologic response to primary chemotherapy and, ultimately, prognosis [41]. Kennedy et al. found that sinonasal ITACs have a distinctive phenotype, with all cases expressing CK20, CDX-2, and villin and most ITACs also expressing CK7, so that the expression pattern of CK7, CK20, CDX-2, and villin positive may be useful in separating these tumors from other non-ITAC adenocarcinomas of the sinonasal tract [43].
Given these findings, to explore other pathways involved in the molecular pathogenesis of ITACs, we immunohistochemically investigated the expression of the apoptosisregulating protein TRAIL in a group of 23 ethmoidal ITAC associated with occupational exposure. This immunohistochemical expression was also retrospectively correlated with the patient outcome to evaluate their independent prognostic relevance. To the best of our knowledge, there are no previous reports on the expression of TRAIL protein in ITACs.
TRAIL is an apoptosis-inducing protein and a molecule important in inhibiting cellular immunity [44][45][46]. Similarly, cancer cells may use TRAIL to evade the antitumor immune response [47]. In literature, there are many works that have detected the expression of TRAIL and its receptors in several tumors, correlating this expression with clinicopathological analysis.
Trail-R1 was identified as an independent prognostic factor for disease-free survival in 128 patients with colon cancer [48]. McCarthy et al. showed that high TRAIL-R2 expression was significantly associated with decreased survival and lymph node involvement in patients with primary breast cancer [49]. Yoldas et al. revealed that TRAIL and DR4 expression were correlated with the pathological grading in patients with laryngeal squamous cell carcinoma, while the alteration in DR5 expression was correlated with the clinical staging [50]. Despite the several works, the specific mechanism of how TRAIL and its receptors contribute to carcinogenesis remains unknown. The results of our immunohistochemical examination showed a relationship between TRAIL upregulation and the increase of the histological tumor grading. No correlation was found with other clinicopathological parameters, including T, stage and relapses.

Conclusion
Our results suggest that mutations of TRAIL expression, in combination with additional genetic events, could play a role in the pathogenesis of ITAC. However, this study showed that TRAIL expression cannot be considered as an independent prognostic factor in patients with ITACs, because there was not sufficient statistical power to detect significant associations between immunohistochemical expression of this protein and the clinicopathologic parameters of the tumors, therefore, further investigations with a larger sample are needed.