Gains and Amplifications of c-myc, EGFR, and 20.q13 Loci in the No Dysplasia–Dysplasia–Adenocarcinoma Sequence of Barrett's Esophagus

Barrett's esophagus is a condition that involves the replacement of normal squamous epithelium with a columnar type of mucosa (1, 2). Compared with the general population, patients with Barrett's esophagus have an increased risk to develop esophageal adenocarcinoma (3-5). The transformation of Barrett's esophagus to invasive esophageal adenocarcinoma may follow a predictable progression of histologic changes known as dysplasia, but adenocarcinoma develops in only a minority of patients with Barrett's esophagus (5). Because long-term survival of patients with esophageal adenocarcinoma is highly dependent on early diagnosis, the identification of patients with Barrett's esophagus at high risk for malignant progression is important (6-8). The present endoscopic and histopathologic evaluations of Barrett's esophagus do not allow the effective identification of high-risk patients at an early stage (9, 10). Therefore, it is important to search for specific markers that can identify patients with Barrett's esophagus at a higher risk for esophageal adenocarcinoma and serve as a supplement to the histopathologic staging of dysplasia (11, 12).

Numerous studies have showed the accumulation of genetic abnormalities from Barrett's esophagus cells to invasive malignant cells (11, 13-15). Among the most important genetic changes contributing to Barrett's esophagus progression are mutations and allelic losses involving p16 and p53 tumor suppressor genes and DNA ploidy changes (13, 16-18). In addition to chromosomal deletions, amplification of oncogenes and growth factors also may play an important role in promoting neoplastic progression (19-21). A variety of gains and amplifications of chromosomal regions have been detected in esophageal adenocarcinoma specimens and adjacent mucosa by Quantitative-PCR, traditional cytogenetic, and metaphase-based comparative genomic hybridization techniques (19, 22-24). Among the genomic amplifications suggested to play a role in esophageal adenocarcinoma are c-myc, epidermal growth factor receptor (EGFR), and amplification of the 20q13.2 locus (19, 23, 25).

One of the methods to detect the cytogenetic abnormalities is fluorescence in situ hybridization (FISH), which uses fluorescently labeled DNA probes to pericentromeric chromosomal regions or unique chromosomal loci. The DNA-FISH technique gives an accurate quantification of the signals and has advantages over comparative genomic hybridization methodologies, particularly in analyzing the alteration of specific genes. DNA-FISH allows also simultaneous evaluation of genetic abnormalities with cellular morphology and can be successfully applied to both tissue section (14, 23, 26) and brush cytology specimens of Barrett's esophagus (27-29). DNA-FISH on brush cytology specimens offers many advantages as a method to detect genetic markers, including simplicity, lower cost, and the potential to sample a larger area of the Barrett's esophagus epithelium when compared with taking random biopsies (30, 31). Therefore, the purpose of the present study was to assess the frequency of c-myc, EGFR, and 20q.13 loci gains and amplifications in the sequence of no dysplasia–dysplasia–adenocarcinoma in patients with Barrett's esophagus using DNA-FISH on brush cytology specimens. DNA probes for the centromeric region of chromosome (CEP) 7 and probes for the locus-specific regions (LSI) of 7p12 (EGFR), 8q24.12 (c-myc), and 20q13.2 were applied on 99 brush cytology specimens of patients with Barrett's esophagus with different stages of dysplasia or esophageal adenocarcinoma. Here, FISH outcomes were compared with those of histopathology.

In this prospective study, a cohort of 99 patients with Barrett's esophagus and Barrett's esophagus-associated adenocarcinoma who underwent endoscopy at the Academic Medical Center in Amsterdam between 2002 and 2007 were included. The ethics committee of the Academic Medical Center approved the study. All patients signed informed consent for the use of their biopsy and brush cytology material. Only patients that had proven (incomplete) intestinal type of metaplasia in biopsies taken during endoscopy were included. All patients were on long-term proton pump inhibition of 40 to 80 mg daily to prevent reflux esophagitis. During endoscopy, the brush cytology specimens were taken before biopsies. The brushes of the normal squamous epithelium were taken from patients without dysplasia at least 3 cm above the Barrett's esophagus segment and were used for control purposes. Biopsies for routine histologic examination were taken at least per 2 cm in 4 quadrants and of all suspected visible lesions using the protocol of Reid et al. (9).

All biopsy specimens were routinely processed and stained with H&E according to standard procedures. Subsequently, the biopsies were evaluated by two pathologists: a junior staff member supervised by a senior gastrointestinal pathologist with extensive experience in the evaluation of Barrett's neoplasia (F.t.K.). The histologic outcome was classified into no dysplasia, indefinite for dysplasia, low-grade dysplasia, high-grade dysplasia, and esophageal adenocarcinoma.

Cytologic brush material was sampled using the Wilson-Cook (Winston-Salem) brush type LCB-220-3-1.5-S. Directly before brushing, the mucosal surface was sprayed with acetylcysteine (50 mg/mL) for dissolving the mucus layer. After the procedure, the brushes were inserted in a vial with 20 mL of 5% acetylcysteine in 0.9% of NaCl, mixed gently to obtain a homogeneous cell suspension. The cell suspension from the brush was poured into a 50-mL conical tube and centrifuged at 2,100 rpm for 10 minutes at 4°C. Most of the supernatant was discarded, leaving the pellet in 5 mL of solution that was subsequently agitated to generate a cell suspension. A cytospin (Shandon Cytospin 4 Cytocentrifuge, Thermo) was used to generate a single layer of the cells on a glass slide. First, 50 μL of PBS was loaded to the cytospin chambers and was centrifuged for 1 minute at 550 rpm at room temperature. Subsequently, up to 150 μL of cells suspension were loaded into the cytospin chambers and were centrifuged 2 minutes at 550 rpm at room temperature. The cytospin slides were dried overnight at room temperature and then stored at -80°C until FISH was done.

We used directly labeled fluorescent chromosomal centromeric probes (CEP) for chromosome 7 and the locus-specific probes (LSI) for regions of 7p12 (EGFR), 8q24.12-q.13 (c-myc), and 20q13.2, obtained from Vysis (Downers Grove). Dual-color probes were used: CEP 7 SpectrumGreen with LSI EGFR (7p12) SpectrumOrange and LSI 20q13.2 SpectrumOrange with LSI c-myc (8q24.12-q13) SpectrumGreen. DNA-FISH was done according to the manufacturer's instructions provided by Vysis as described previously (29).

After the FISH procedure, 100 to 200 interphase nuclei of Barrett's esophagus cells were scored per slide by an experienced scorer (A.M.R.) using an Olympus BX61 fluorescent microscope. The cases were evaluated without previous knowledge of histologic findings. Damaged cells and cells with indistinct and blurry signals were excluded from the analysis. The following categories of abnormalities were distinguished: 3 to 4 signals of CEP 7 (green) accompanied by the same number of signals for EGFR locus (7p12; orange) was considered as a gain (polysomy) of chromosome 7, signals between 4 and 10 of EGFR locus corrected for chromosome 17 copy numbers (green) were considered as low amplification, and 10 or more EGFR locus signals corrected for chromosome 17 copy numbers or a tight cluster were considered as high amplification. Similarly, for the 8q24 (c-myc) and 20q13.2 locus, 3 or 4 signals were considered as gain, signals between 4 and 10 were considered as low amplification, and number of signals 10 or more or a tight cluster was considered as high amplification. The cutoff value for gains and amplifications of the loci was 4% or more of abnormal nuclei. The cutoff values were obtained from counts in the normal squamous epithelium (100 nuclei evaluated) of 20 patients with Barrett's esophagus without dysplasia and calculated as the mean percentage of squamous nuclei with signal gain plus thrice the SD.

Differences in frequencies of abnormalities were compared using Fisher's exact test, and statistical significance was set at a P of less than 0.05. The statistical analyses were conducted using SPSS software (version 12.0, SPSS, Inc.).

Of the 99 patients with Barrett's esophagus, 80 were men and 19 were women. Median age was 62 (range, 31-87) years, and median length of Barrett's esophagus segment was 4 cm (range, 1-13 cm). The study included 36 patients with no dysplasia, 11 patients with low-grade dysplasia, 14 patients with indefinite for dysplasia, 22 patients with high-grade dysplasia, and 16 patients with esophageal adenocarcinoma. The population of patients with esophageal adenocarcinoma population included 5 patients with T₁/T₂N₀M₀, 3 patients with T₃N₀M₀, and 4 patients with T₃N₁M₀ stage. The tumor-node-metastasis classification data were not available for 4 of 16 patients with esophageal adenocarcinoma.

Detailed FISH results together with histologic diagnosis are presented in Table 1. Gains of chromosome 17, c-myc, and 20q.13 loci were detected already in nondysplastic cases in a low frequency of 8%, 14%, and 8%, respectively (Fig. 1). Their combined frequencies were 22% (8/36) in no dysplasia and 24% (6/25) in indefinite for dysplasia/low-grade dysplasia cases, and significantly increased to 72% (16/22) in high-grade dysplasia and 94% (15/16) in esophageal adenocarcinoma cases (P < 0.0001). The amplification of any of the loci was found in none of the patients with no dysplasia and indefinite for dysplasia/with low-grade dysplasia, but it was detected in 14% (3/22) of high-grade dysplasia and 50% (8/16) of patients with esophageal adenocarcinoma (P = 0.015). In patients with high-grade dysplasia, the amplification mostly concerned the c-myc locus, seen in 14% (3/22) of the cases, and EGFR and 20q13 loci amplification was detected in a lower frequency of 5% (1/22 each; Fig. 1). In esophageal adenocarcinoma cases, the amplifications of c-myc, 20q13 loci, and EGFR loci were observed in the frequency of 25%, 25% (4/16), and 19% (3/16), respectively (Fig. 1). Although the frequencies of the amplification of c-myc, 20q13, and EGFR loci were higher in esophageal adenocarcinoma compared with high-grade dysplasia (25% versus 14%, 25% versus 5%, and 19% versus 5%, respectively), these differences did not reach statistical significance (P = 0.38, P = 0.06, and P = 0.15, respectively).

In total, we detected 16 amplification events of c-myc, EGFR, or 20q13 in 11 cases (3 cases of high-grade dysplasia and 8 cases of esophageal adenocarcinoma), including low and high levels of amplification of these loci (Table 1). High-amplification events had significantly lower frequency in high-grade dysplasia cases (20%, 1/5) compared with esophageal adenocarcinoma cases (73%, 8/11; P = 0.049). Figure 2 depicts examples of Barrett's esophagus nuclei with the oncogenic loci copy number changes.

In the present study, we assessed the frequencies of c-myc, EGFR, and 20q.13 loci gains and amplifications in the sequence of no dysplasia–dysplasia–adenocarcinoma of Barrett's esophagus. DNA-FISH on brush cytology samples revealed that gains, which presumably are the precursor changes before true amplifications of the loci occur, seem early in nondysplastic Barrett's esophagus. We found that low- and high-level amplifications of the investigated loci occur late only in high-grade dysplasia and esophageal adenocarcinoma cases.

We showed that gains (3-4 copies) of chromosome 17, c-myc, and 20q.13 loci can already be seen at low frequencies of 8% to 14% in nondysplastic Barrett's esophagus. Their combined frequencies significantly increase with increasing stage of dysplasia and have high incidence in esophageal adenocarcinoma. These findings correspond well with comparative genomic hybridization and FISH studies examining the same chromosomal regions (23, 28, 32). The detected gains in our study may reflect aneuploidy. Aneuploidy is indicative for Barrett's esophagus malignant progression as shown by DNA flow cytometry studies (33-35). We further observed that amplification of at least 1 of the loci (c-myc, EGFR, and 20q.13) is seen in 14% of high-grade dysplasia and increases significantly to 50% in patients with esophageal adenocarcinoma. This is concordant with other reports showing gene amplification as a late event with higher frequencies in esophageal adenocarcinoma compared with high-grade dysplasia cases (25, 36). In addition, we found low levels (>4 and <10 copies) and high levels (>10 copies or clusters) of the loci amplifications in both high-grade dysplasia and esophageal adenocarcinoma, indicating extensive genetic heterogeneity of these cases. We also showed that high-amplification events have significantly higher frequency in esophageal adenocarcinoma cases (72%) compared with high-grade dysplasia cases (20%). This finding corresponds well with the study by Miller et al. (19) using Quantitative-PCR (Q-PCR) that showed much higher levels of oncogene loci amplifications in adenocarcinoma tissues compared with high-grade dysplasia. To date, there is a very limited number of DNA-FISH studies investigating the different levels of the oncogene loci amplification in Barrett's esophagus. Previously, different levels of HER-2 oncogene amplifications have been described in esophageal adenocarcinoma cases and adjacent high-grade dysplasia samples by Walch et al. (23, 36). In general, our results suggest that the low and high levels of oncogenic loci amplification initiate in high-grade dysplasia and impel further malignant transformation and invasive growth. Therefore, low and high levels of these loci amplification may be considered as late markers for Barrett's esophagus malignant progression. Importantly, our results indicate the EGFR, c-myc, and 20q.13 amplifications may be preceded by early gains of these loci, which are detectable already in nondysplastic cases, and therefore may be good candidates for early markers of Barrett's esophagus malignant progression.

The c-myc proto-oncogene encodes a transcriptional factor involved in the regulation of normal cellular proliferation, differentiation, and apoptosis (37). c-Myc amplification and overexpression of the protein are found in a variety of human tumors (38). In this study, c-myc locus (8q24) amplification was the most frequent one, detected at comparable frequency in high-grade dysplasia (14%) and esophageal adenocarcinoma (25%) cases. Previously, c-myc locus amplification has been found in 14% to 25% of esophageal cancers, which corresponds well with our results (21). The c-myc amplification has been also documented in patients with Barrett's esophagus with high-grade dysplasia (19, 20, 36).

It has been shown that an increased copy number of 20q is associated with cellular immortalization, and amplification of 20q13.2 has been correlated with genomic instability (39, 40). Interestingly, many human tumor types, for example, breast cancer (41), ovarian cancer (42), and head-and-neck cancer (43), display gain or amplification of this region, suggesting that the genes on 20q play an important role in carcinogenesis. In this study, we found the 20q13 locus amplification in a much lower frequency in high-grade dysplasia (5%) compared with esophageal adenocarcinoma (25%). However, the frequency of 20q13 amplification in the esophageal adenocarcinoma cases is lower than that detected in the study by Falk et al. (27), who showed this abnormality in 62% of esophageal adenocarcinoma cases (5/8 patients). However, in that study, a smaller number of patients were analyzed and there was no discrimination between gains and amplification that may explain this discrepancy. It is important that the 20q13.2 probe used in the present study lies at the center of the region within 20q. Several candidate genes have been proposed as potential target genes in this region, for example, NABCI, BTAK, ZNF217, and BCASI, and it is likely that more than one putative oncogene is involved in the overrepresentation of 20q in Barrett's esophagus (44-46).

EGFR, another proto-oncogene investigated in this study, plays an important role in tumor cell survival and proliferation. EGFR is amplified and overexpressed in many epithelial cancers, including lung non–small cell carcinoma and colorectal adenocarcinoma (47, 48). Here, we found EGFR locus amplification in 19% of esophageal adenocarcinoma that fits with the 8% to 30% range of the locus amplification reported in studies using PCR techniques (19, 49). In these studies, the EGFR amplification was not detected in high-grade dysplasia cases. In contrast, we clearly found low amplification of the EGFR locus in one of the high-grade dysplasia case. In a recent FISH study, gains of the EGFR locus were reported, but amplification was not described in high-grade dysplasia cases (28).

To our knowledge, this is the first study discriminating between gains and low and high levels of amplification of the c-myc, EGFR, and the 20q13.2 loci in the sequence of metaplasia-dysplasia-adenocarcinoma of Barrett's esophagus using FISH on brush cytology specimens. Our results indicate that the amplification of these loci, observed in esophageal adenocarcinoma and, to a lesser extent, in high-grade dysplasia, initiates in high-grade dysplasia and impel subsequent malignant growth. Detection of these amplifications may help identify patients with high-grade dysplasia who have a higher risk for developing cancer. In addition, this study shows that gains of the investigated chromosomal loci, which are highly prevalent in esophageal adenocarcinoma and high-grade dysplasia, occur already at the stage of no dysplasia. These gains may be a result of early polyploidy, preceding the later genetic event of amplification as observed in high-grade dysplasia and esophageal adenocarcinoma. Prospective follow-up of the Barrett's esophagus cohort displaying these gains will determine whether any of them are predictive as early markers for progression to malignancy.

No potential conflicts of interest were disclosed.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.