Abstract
Background:
Despite having a dramatically larger surface area than the large intestine, the small intestine is an infrequent site for the development of adenocarcinoma. To better understand the molecular abnormalities in small bowel adenocarcinoma [SBA], we characterised a number of candidate oncogenic pathways and the immunophenotype of this rare cancer.
Methods:
Tissue microarrays were constructed from tumour samples from 54 patients with all stages of the disease. Immunohistochemistry and microsatellite instability [MSI] testing were conducted.
Results:
The profile of cytokeratin 20 and 7 coexpression was variable, but expression of caudal type homeobox transcription factor 2 [CDX2] was present in 70% of cases. In this young population [median age 54 years], loss of mismatch repair [MMR] proteins occurred in 35% of patients, with confirmed MSI in 100% of tested cases. Expression of vascular endothelial growth factor-A [VEGF-A] and epidermal growth factor receptor [EGFR] was common, occurring in 96 and 71% of patients, respectively. Only one case showed HER2 expression and none showed loss of phosphatase and tensin homologue mutated on chromosome 10 [PTEN].
Conclusions:
These results suggest that alterations in DNA MMR pathways are common in SBAs, similar to what is observed in large bowel adenocarcinomas. Furthermore, the high percentage of tumours expressing both EGFR and VEGF suggests that patients with this rare cancer may benefit from therapeutic strategies targeting EGFR and VEGF receptor [VEGFR].
Main
Adenocarcinoma of the small bowel is a rare, aggressive malignancy with poor overall outcome. The majority of patients with small bowel adenocarcinoma [SBA] present with advanced disease, with 5-year disease-specific survival rates of 35% for patients with lymph node involvement and 4% for patients with metastatic disease [Howe et al, 1999]. Although the small intestine makes up approximately 75% of the length and 90% of the mucosal surface of the gastrointestinal tract, SBA occurs 50 times less frequently than colorectal adenocarcinoma [O'Riordan et al, 1996; DeSesso and Jacobson, 2001]. Despite this intriguing biological difference in the incidence of SBA and colorectal adenocarcinoma, few investigations into the mechanisms of small bowel carcinogenesis have been conducted. A number of theories have been postulated to explain the relative protection of the small intestine from the development of carcinoma; however, none have been proven. Proposed protective factors have generally centred around two concepts. First, the rapid turnover of small intestinal cells results in epithelial cell shedding before the accumulation of the genetic damage critical to carcinogenesis. Second, exposure of the small intestine to the carcinogenic components of our diet is limited because of the small intestine's rapid transit time, lack of bacterial degradation activity, and relatively dilute alkaline environment.
As in colorectal cancer, a subset of SBAs are characterised by a defect in DNA mismatch repair [MMR], which results in DNA microsatellite instability [MSI] [Planck et al, 2003]. Microsatellite instability is characterised by the accumulation of changes in the length of simple repeated nucleotide sequences known as microsatellites, caused by mutations in MMR genes such as MutS homologue 2 [hMSH2], MutL homologue 1 [hMLH1], post meiotic segregation increased 1 [hPMS1], hPMS2, and hMSH6. Although MSI is a hallmark of hereditary nonpolyposis colorectal cancer syndrome, in which mutations of one or more MMR genes are found in >90% of the cases, it has also been reported in approximately 10% of sporadic colorectal adenocarcinomas [Liu et al, 1995]. In colorectal cancer, defects in DNA MMR are characterised by a younger age of onset, poorly differentiated mucinous tumours, and improved outcomes [Gryfe et al, 2000; Jenkins et al, 2007]. However, for patients with SBA, no comprehensive study has been conducted to compare the clinical and pathological differences between patients with intact and deficient DNA MMR.
Other oncogenic signalling pathways that are active in colorectal cancer include the epidermal growth factor receptor [EGFR], vascular endothelial growth factor receptor [VEGFR], and phosphatidylinositol 3-kinase [PI3K]/AKT pathways. Epidermal growth factor receptor is a member of the ErbB family of transmembrane receptor tyrosine kinases that have a crucial role in tumour cell proliferation, survival, adhesion, migration, and differentiation [Yarden and Sliwkowski, 2001]. Although the expression of EGFR has not been evaluated in SBA, one study has evaluated the expression of the ErbB family member, HER2. In this study, 9 of 16 patients showed HER2 expression and such expression was associated with a worse overall survival [OS; Zhu et al, 1996]. Vascular endothelial growth factor signalling has a key role in tumour-associated neo-angiogenesis and the PI3K/AKT pathway is a critical pro-survival and pro-proliferation pathway that is frequently activated in a variety of cancers [Stambolic et al, 1998]. The phosphatase and tensin homologue mutated on chromosome 10 [PTEN] protein acts as a potent tumour suppressor of this pathway and loss of PTEN expression is observed in approximately 5% of colorectal cancer [Zhou et al, 2002; Goel et al, 2004]. Currently, no studies have evaluated the expression of EGFR, VEGF-A, or PTEN in SBA.
In addition, the published data on the immunophenotype of SBA are limited and conflicting [Lee et al, 2003; Chen and Wang, 2004]. To date, no study has evaluated the expression of caudal type homeobox transcription factor 2 [CDX2] in a large number of patients with SBA. It is critical for the development and differentiation of the intestines, and highly expressed in cancers of the large intestine [Moskaluk et al, 2003; Werling et al, 2003].
To characterise the immunophenotype and the expression of a number of candidate oncogenic proteins, we performed immunohistochemical analysis of tumour samples from a large data set of SBA patients treated at The University of Texas M.D. Anderson Cancer Center. As a number of the proteins analysed represent therapeutic targets for drugs currently in clinical use, the results of this study have direct clinical relevance for the potential treatment of this rare cancer.
Materials and methods
Study population and tumour samples
We searched the M.D. Anderson Cancer Center pathology database from 1981 to 2007 for cases of SBA with accompanying pathologic materials. A total of 83 cases were identified and 54 cases were included. In all, 23 cases with inadequate pathologic materials and six cases of ampullary adenocarcinoma were excluded. The haematoxylin and eosin [H&E]-stained slides from all 54 cases were reviewed by a gastrointestinal pathologist [HW] to confirm the diagnosis and tumour grades. Normal small bowel mucosa samples were present in all but 12 of the cases. We retrospectively collected clinical and follow-up data, including demographics, cancer treatment history, stage, and survival for each patient from patient records. Tumour samples were from the primary site of resection in 43 cases and from a metastatic site in 11 cases. A total of 10 patients received anticancer therapy before collection of the pathologic material, with therapy consisting of chemoradiation in three cases and systemic chemotherapy in seven cases.
This study was carried out in accordance with the guidelines of the institutional review board.
Tissue microarray
To construct the tissue microarray, the formalin-fixed, paraffin-embedded archival tissue blocks and their matching H&E-stained slides were retrieved, reviewed, and screened for representative tumour regions and normal small bowel mucosa by a gastrointestinal pathologist [HW]. For each patient, three cores of tumour and when available, two cores of paired normal small bowel mucosa, were sampled from representative areas using a 1.0-mm punch. The tissue microarray was constructed with a tissue microarrayer [Beecher Instruments, Sun Prairie, WI, USA] as described previously [Wang et al, 2002]. The constructed TMA blocks were sealed with paraffin, and 5-μm-thick slides were cut from the TMA blocks for immunohistochemical staining.
Immunohistochemistry
Immunohistochemical stains were performed on 5 μm unstained sections from the tissue microarray blocks using the antibodies listed in Table 1. To retrieve the antigenicity, the tissue sections were treated at 100 °C in a steamer containing 10 mmol citrate buffer [pH 6.0] for 60 min. The sections were then immersed in methanol containing 0.3% hydrogen peroxidase for 20 min to block the endogenous peroxidase activity and were incubated in 2.5% blocking serum to reduce nonspecific binding. Sections were incubated for 90 min at 37 °C with primary antibodies at the dilutions specified in Table 1. Standard avidin–biotin immunohistochemical analysis of the sections was performed according to the manufacturer's recommendations [Vector Laboratories, Burlingame, CA, USA]. Diaminobenzidine tetrahydrochloride was used as a chromogen, and haematoxylin was used for counterstaining.
Table 1 Antibodies used
Full size table
Two observers independently analysed the immunostaining results and their intensities [JVP and HW]. Positive and negative controls were used for each antibody on each slide. A third investigator [MJO] re-evaluated any cases with discrepancies between the two interpretations. Expression of PTEN, PMS2, hMLH-1, hMSH-2, hMSH-6, CK20, CK7, and CDX2 was considered positive if >10% of the tumour cells show immunoreactivity. The staining for HER2 was graded according to the guideline for HER2 testing in breast cancer with 2+ or 3+ membranous staining in 10% of the tumour cells as positive [Wolff et al, 2007]. For VEGF-A and EGFR, both the percentage of positive tumour cells and the intensity of positive staining were graded. The percentage of staining was graded as follows: 0, staining in