p53 protein overexpression in low grade dysplasia (LGD) in barrett’s esophagus: Immunohistochemical marker predictive of progression

THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2001 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.

Vol. 96, No. 5, 2001 ISSN 0002-9270/01/$20.00 PII S0002-9270(01)02414-5

ORIGINAL CONTRIBUTIONS

p53 Protein Overexpression in Low Grade Dysplasia (LGD) in Barrett’s Esophagus: Immunohistochemical Marker Predictive of Progression Allan P. Weston, M.D., Sushanta K. Banerjee, Ph.D., Prateek Sharma, M.D., Trang M. Tran, B.S., Robert Richards, M.D., M.S., and Rachel Cherian, M.D. Cancer Research Unit, Research Division, Veterans Administration Medical Center, Kansas City, Missouri; and Departments of Medicine and Pathology, University of Kansas Medical Center, Kansas City, Kansas

OBJECTIVES: The presence of low grade dysplasia (LGD) within Barrett’s esophagus (BE) has a multitude of ramifications. Identification of markers that could risk stratify LGD would be of great clinical benefit. We aimed to prospectively evaluate the prognosis of the immunohistochemical overexpression of p53 protein in BE colocalized to LGD. METHODS: Consecutive BE patients in whom LGD was found had a repeat esophagogastroduodenoscopy within 8 –12 wk per an ongoing prospective study. At each esophagogastroduodenoscopy, a therapeutic scope was used in conjunction with the Seattle Biopsy Protocol. Patients were observed until development of multifocal high grade dysplasia (mHGD), presence of an HGD dysplasia-associated lesion or mass (DALM) lesion, or frank adenocarcinoma. p53 protein overexpression was determined by computerized immunoquantitation using image analysis software on step serial-sectioned specimens of BE segment(s) harboring LGD. Kaplan-Meier survival curves were made on the ability of p53 staining colocalized to areas of LGD to predict progression to mHGD, HGD DALM, or cancer during prospective follow-up. RESULTS: Forty-eight BE patients with LGD were observed for a mean of 41.2 ⫾ 22.5 months. During this period, five of 48 patients progressed to mHGD with a focus in which intramucosal cancer could not be excluded (one), mHGD/ DALM with one or more foci in which intramucosal cancer could not be excluded (two), cancer (one), or mHGD (one). Twelve had persistent LGD and 31 had regressed to no dysplasia. p53 staining was positive and colocalized to areas of LGD in 4/31 of patients that regressed, 3/12 that persisted, and 3/5 that progressed. Kaplan-Meier curves differed significantly between p53 positive and negative patients for outcome defined as progression of LGD. CONCLUSIONS: p53 colocalization with LGD at index LGD diagnosis is a risk factor for progression of LGD. This can potentially be used to risk stratify BE LGD patients in terms A.P.W. and S.K.B. contributed equally to this work.

of surveillance intervals or enrollment into secondary prevention studies. (Am J Gastroenterol 2001;96:1355–1362. © 2001 by Am. Coll. of Gastroenterology)

INTRODUCTION Barrett’s esophagus (BE), acquired secondary to long-standing gastroesophageal reflux, is the only known precursor of esophageal and esophagogastric junction adenocarcinomas. Barrett’s adenocarcinoma develops via a multistep process recognized phenotypically as the histological sequence of metaplasia–low grade dysplasia (LGD)– high grade dysplasia (HGD)–adenocarcinoma (1– 4). Consequently, the detection of LGD within BE has a multitude of ramifications, including an increased concern for eventual progression to cancer and intensification of endoscopy surveillance (5, 6) with resultant increase in health care expenditures. However, the majority of BE patients harboring LGD do not progress (7–12). Furthermore, despite standardization of histological criteria for the grading of dysplasia, intraobserver and interobserver disagreement in the diagnosis of LGD exists (13). Because of the difficulties in assessing LGD and its variable natural history (progression, persistence, or regression), identification of a biomarker that is technically simple, economical, quick, and widely available and that could risk stratify Barrett’s patients harboring LGD would be of great clinical benefit. Alterations in the p53 gene are one of the most commonly identified genetic abnormalities found in human cancers (14 –20). The p53 gene encodes for a protein responsible for the regulating transcription of genetic sequences for other proteins that control the entry of proliferating cells into the S phase of the cell cycle (21). p53 protein functions to prevent the propagation of DNA errors by inducing cell death (21, 22). As one advances up the multistep and sequential histological spectrum of Barrett’s mucosa negative for dysplasia–LGD–HGD–adenocarcinoma, p53 protein overexpression has been shown to progressively increase (3, 23–31). The prospective endoscopic surveillance outcome of BE patients exhibiting p53 protein overexpression in the

1356

Weston et al.

absence of HGD or adenocarcinoma has not been thoroughly examined. The aim of our study was to determine the status of p53 protein colocalized to BE segments harboring LGD and also to evaluate the impact of any p53 expression during follow-up—that is, whether or not p53 protein overexpression could serve as an objective biomarker for BE complicated by LGD. Specifically, we prospectively evaluated the risk of progression to HGD or adenocarcinoma in patients with BE complicated by LGD using p53 protein overexpression determined immunohistochemically.

MATERIALS AND METHODS Consecutive Barrett’s patients in whom LGD was detected among a large cohort of Barrett’s subjects were eligible for this study. Patients with LGD had their medical therapy intensified by utilization of single, double, or higher dose proton pump inhibitors (PPIs) with or without adjuvant use of a promotility agent. PPI dosing was guided by clinical and endoscopic findings. Prokinetics were used as adjuvants to PPIs in the absence of absolute contraindications in the setting of clinical or endoscopic evidence of delayed gastric emptying or persistent regurgitation. At follow-up upper endoscopy esophagogastroduodenoscopy (EGD) in 8 –12 wk, careful visual examination of the Barrett’s mucosa for any mucosal irregularities was undertaken (i.e., ulcer(s), erosion(s), friability, nodules). If any Barrett’s mucosal irregularities were noted they were “target” biopsied, as they could potentially represent dysplasia-associated lesion or mass (DALM) lesions, followed by four quadrant surveillance biopsies every 1–2 cm throughout the entire length of BE. If LGD persisted, the patient continued medical treatment, and endoscopic surveillance every 12 months was undertaken until LGD “regressed” to no dysplasia or progressed to HGD, HGD DALM, or frank adenocarcinoma. If LGD regressed to no dysplasia at follow-up, EGD with biopsy was repeated at 12 months. If LGD progressed to HGD, then management was based upon whether the HGD was detected within a DALM lesion and whether the HGD was unifocal or multifocal. For patients with unifocal HGD (uHGD), EGD was repeated in 4 – 8 wk and four quadrant surveillance biopsies every 1 cm were obtained. If uHGD persisted, the patient continued medical treatment, and endoscopic surveillance every 3– 6 months was undertaken until uHGD either “regressed” to LGD or progressed to multifocal HGD (mHGD), HGD DALM, or frank adenocarcinoma. If uHGD regressed to LGD or no dysplasia at follow-up, EGD with biopsy was repeated at 3– 6 months. If LGD or no dysplasia persisted, then surveillance intervals were broadened to every 6 –12 months. DALM lesions were considered focal Barrett’s mucosal irregularities such as erythematous, granular, friable areas, nodules, erosions, strictures, or fixed/immobile wall regions in conjunction with biopsies (target biopsies) showing HGD. A conscious decision to distinguish uHGD from mHGD was made in this study as part of a prospective effort to gain further insight

AJG – Vol. 96, No. 5, 2001

into Barrett’s histogenesis and risk stratification based upon observations concerning genomic instability and clonal evolution of metaplastic columnar epithelial cells suggesting that chronological and biological differences exist between focal and widely distributed HGD (32–34). Unifocal HGD was not considered an endpoint biomarker, as prospective, long term follow-up of this lesion (uHGD) has been shown to have variable outcomes (35). HGD DALM lesions were considered as an endpoint based in part upon the observation that a higher incidence of actual cancer invading into the esophageal wall (submucosa and even deeper) and greater chance of regional lymph node metastasis exist in HGD and cancerous BE patients harboring subtle mucosal irregularities relative to those patients who have normal BE mucosal surfaces (36). Esophageal biopsy specimens were obtained utilizing direct endoscopic vision and the turn and suction method using a large, fenestrated, spiked 9-mm open span jumbo biopsy forceps (FK-13K-1 [Olympus, Lake Success, NY] or Radial Jaw III Max Capacity 1589 [Boston Scientific, Natick, MA]). The endpoint of the study was development of mHGD, DALM in conjunction with HGD, or adenocarcinoma. Barrett’s esophagus patients with LGD were excluded from the study for the following reasons: 1) past history of HGD or if dysplasia was detected within DALM, 2) further surveillance examinations were declined by the patient, or 3) surveillance examinations could not be performed because of patient relocation or death, or they were not indicated due to presence of terminal illness(es) and/or multiple severe comorbid illnesses. In addition, patients in whom the BE histological changes were inconclusive, between reactive atypia and low grade or indefinite dysplasia (IND), and in whom follow-up EGD exams failed to confirm definite dysplasia were excluded. Histology Biopsy specimens were fixed in 10% buffered formalin, embedded in paraffin, step serially sectioned, and then stained with hematoxylin and eosin. The diagnoses of LGD, IND, and HGD were according to the previously established criteria (1, 37, 38). Criteria for distinguishing reactive changes from IND and LGD outlined by Haggitt (1) were followed. Patients with changes described as IND were grouped into the category of LGD because of the high rate of interobserver variation in these diagnoses (13). All cases of HGD required confirmation by independent review. All cases of LGD and IND also required independent confirmation as either LGD or IND. Histological changes inconclusive between reactive atypia and IND or LGD were not considered as dysplasia and were insufficient to gain inclusion into this study. Unifocal HGD was considered as a single Barrett’s mucosal segment or biopsy specimen harboring HGD, whereas multifocal HGD was considered as multiple biopsy specimens from different Barrett’s segments harboring HGD. The term invasive carcinoma cannot

AJG – May, 2001

be excluded was applied when one or more foci within HGD were noted to have glands growing in a cribriform, back to back pattern or when they were closely packed together, or when ill-defined abortive glands were present in the lamina propria (39). For patients exhibiting multiple foci of dysplasia, only the highest grade of dysplasia was included in the analysis. p53 Immunohistochemistry The immunohistochemical staining procedure was carried out according to the manufacturer’s instructions (Zymed Laboratories, South San Francisco, CA) and our previously described modified method (40, 41). Briefly, 5-␮m sections were cut from tissue samples embedded in paraffin that on hematoxylin and eosin staining had displayed LGD. Additionally, in those BE patients who had showed progression of LGD during prospective follow-up, tissue sections were cut from those specimens demonstrating mHGD, DALM/ HGD, or cancer. Next, the tissue was dewaxed in xylene and rehydrated in 1⫻ phosphate-buffered saline (PBS) through different concentrations of ethanol. Endogenous peroxidase activity was blocked by incubating the slides in 3% hydrogen peroxide (vol/vol) in methanol for 5 min at room temperature. The slides were then incubated in Citrate buffer (Zymed Laboratories) in a microwave oven at high power for 5 min and allowed to cool for 15 min. After washing the slides in water followed by 1⫻ PBS, sections were incubated in ready to use tissue blocker for 15 min at room temperature. Tissue blocker was replaced with primary antibodies (p53 monoclonal antibodies, Zymed Laboratories) or preimmune IgG as negative controls and incubated overnight at 4°C. After incubation with primary antibodies, slides were rinsed with 1⫻ PBS (3 ⫻ 5 min), and were incubated for 10 min at room temperature in biotinylated rabbit-antigoat IgG (Zymed Laboratories). After rinsing with PBS (3 ⫻ 5 min), sections were incubated in peroxidase-conjugated linker for 10 min at room temperature. Antibodies were detected by incubation with 3,3⬙-diaminobenzidine tetrahydrochloride solution, provided by the manufacturer with its kit, for a period of time sufficient to yield a dark brown color, usually 5 min. Nuclear staining of Barrett’s mucosa tissue was evaluated by three independent observers (A.P.W., S.K.B., T.M.T.) using a Leica (Heerbrugg, Switzerland) microscope and Sigma Pro-Scan 5 computer program (SPSS, Chicago, IL). After blinded independent assessment of p53 staining by the investigators, discrepancies were resolved at a joint multihead microscopic conference. p53 measurements for computerized immunoquantitation were performed with a digitized interactive Pro-Scan imaging analysis software system. All glandular epithelium was examined and p53 quantitated by measuring the glandular area percentage of positive nuclei. Statistical Analysis The object of the analysis was to determine if persons with p53 immunohistochemical overexpression colocalized to

p53 Protein, Low Grade Dysplasia, and Barrett’s Esophagus

1357

Barrett’s mucosa harboring LGD were more likely to be associated with progression of LGD to mHGD, HGD DALM, or cancer. Because subjects had different lengths of follow-up (right censoring), a survival analysis was performed. The LIFETEST procedure in SAS (version 8, SAS Institute, Cary, NC) was used to compute nonparametric estimates of the survivor function by the Kaplan-Meier method. The log rank test was used to compute the KaplanMeier curves of those who were p53 positive versus those who were p53 negative. The significance level was set at ␣ ⫽ 0.05. Student’s t test was performed on the percentage of cells positive for p53 at index diagnosis of LGD relative to those at endpoint for those patients in whom LGD progressed during follow-up. The Cochran-Armitage linear trend test using SAS software was used to determine whether p53 protein overexpression increased among the three LGD outcome subgroups (those that regressed, those that persisted, and those that progressed). The study was approved by the Human Studies Committee as part of a long term, prospective endoscopic, histological, and molecular study of BE.

RESULTS Patients’ Clinical Data A total of 283 BE patients were diagnosed and observed over the course of the study period; in 54 of them LGD was detected. Forty-eight of these 54 patients met all inclusion criteria. Six patients were excluded because of death soon after diagnosis of LGD (one patient), refusal to undergo follow-up EGD (one patient), inability to perform follow-up because of patient relocation (two patients), or presence of severe comorbid illnesses (two patients). LGD was detected at initial diagnosis of Barrett’s in 26 cases (prevalence, in two patients the diagnosis was IND rather than LGD), whereas 22 cases were detected during surveillance (incidence, in one case it was IND rather than LGD). The demographic and endoscopic features of the 48 LGD Barrett’s patients are given in Table 1. LGD changes were confined to a single Barrett’s mucosa segment in 25 patients, whereas in the remaining 23 patients LGD was noted in multiple Barrett’s mucosal segments (two to four). For the 22 incidence cases, Barrett’s diagnosis had been made on average 19.7 ⫾ 15.5 months earlier. The 48 BE patients were prospectively observed for up to 82 months for a mean ⫾ SD of 41.2 ⫾ 22.5 months (range ⫽ 8 – 82 months, median ⫽ 43 months, and mode ⫽ 54 months). Figure 1 provides a schematic flow diagram of the histological outcome of the study patients. Two hundred fourteen surveillance EGDs were performed (average ⫽ 4.6 ⫾ 2.6 EGD/patient) during a period of 1978 patient months of follow-up. Over the course of follow-up, 5/48 LGD patients (10.4%) progressed— one to mHGD with a focus in which intramucosal carcinoma could not be excluded, two to mHGD/DALM with one or more foci in which intramucosal cancer could not be excluded, one to

1358

Weston et al.

AJG – Vol. 96, No. 5, 2001

Table 1. Demographic and Endoscopic Features of Barrett’s Low Grade Dysplasia Patients Age Range Mean ⫾ SD Gender (male:female) Race (white:other) Endoscopic features Barrett’s length (cm) Range Mean ⫾ SD Esophagitis grade* 0 1 2 Barrett’s ulcer Hiatal hernia None Yes Size (cm) ⱕ2 3–5 ⱖ6

30.9–79.3 61.9 ⫾ 12.0 48:0 48:0 1–18 5.5 ⫾ 3.8 34 1 11† 2‡ 9 39 14 19 6

* Hetzel-Dent endoscopic esophagitis grading system. † Eight of 11, the esophagitis was within the squamous epithelium above Barrett’s mucosa. ‡ Barrett’s ulcer separate from area harboring low grade dysplasia.

cancer, and one to mHGD. Staging of four cases by radial echoendosonography and chest CT scan was uT0/TisN0M0 (no ultrasonographic evidence of tumor); in one case by surgical resection it was pTisN0M0. In contrast, in only one case out of 200 BE patients who did not have LGD at index diagnosis or anytime during follow-up was there progression directly to cancer in the absence of preceding endoscopic surveillance biopsies showing LGD (pathological

stage of this tumor was pT1mN0M0). Thirty-one LGD patients (64.6%) regressed over the course of 8 –75 months of follow-up (mean ⫾ SD ⫽ 44.5 ⫾ 21.9). Nine of the 12 patients with short segment BE with LGD regressed during follow-up. Twelve LGD patients (25.0%) have had persistent LGD throughout the course of their follow-up, ranging from 15 to 82 months (mean ⫾ SD of 39.2 ⫾ 24.1 months). Immunohistochemical Demonstration of p53 Protein Status in Barrett’s Esophagus LGD To study the association between the p53 protein expression in LGD and to determine the prognostic impact of p53 expression during the follow-up of these LGD patients, p53 immunohistochemical analysis was performed on biopsy specimens harboring index LGD. p53 immunopositive cells were present in 10 of 48 patients (20.8%, 4/31 that regressed, 3/12 in which LGD persisted, and 3/5 in whom LGD progressed. The percentage of columnar epithelial cells positive for p53 at index diagnosis of LGD in the Barrett’s biopsy sample(s) harboring LGD ranged from 41.8% to 61.9% for the four patients who underwent regression, 17.4 – 65.9% for the three patients in whom LGD persisted, and 8.3–58.1% for the three patients who progressed. Figure 2 shows examples of p53 overexpression. The Kaplan-Meier survival curves using outcome defined as progression of LGD versus persistence of LGD or regression of LGD are shown in Figure 3. The difference was significant (p ⬍ 0.002 by log rank test). Cochran-Armitage linear trend test confirmed a significant progression of p53 positivity (p ⫽ 0.01), demonstrating a sequential order of p53 positivity as progressive LGD ⬎ persistent LGD ⬎ regressive LGD. For the five patients in whom BE LGD progressed, p53 staining was also examined

Figure 1. Flow diagram of the histological outcome of Barrett’s low grade dysplasia. Ca? ⫽ cannot exclude intramucosal cancer; Dys ⫽ dysplasia.

AJG – May, 2001

p53 Protein, Low Grade Dysplasia, and Barrett’s Esophagus

1359

Figure 2. p53 overexpression in Barrett’s. (A) Negative control (preimmune serum only, no primary antibody) (⫻200). (B) Barrett’s mucosa with low grade dysplasia, entire sample negative for p53 (⫻400). (C) Barrett’s mucosa with low grade dysplasia, positive for p53 (⫻400). (D) Barrett’s mucosa with high grade dysplasia, positive for p53 (⫻400).

Figure 3. Kaplan-Meier estimates of progression of low grade dysplasia to a more advanced lesion (multifocal high grade dysplasia, high grade dysplasia within a dysplasia-associated lesion or mass, or adenocarcinoma). p53 positivity was associated with a higher rate of progression.

1360

Weston et al.

AJG – Vol. 96, No. 5, 2001

Table 2. Computer Immunoquantitation of p53 Overexpression at Index Diagnosis and Endpoint for Those Low Grade Dysplasia Barrett’s Esophagus Patients Who Progressed % Glandular Cells Patient Positive at No. Index Diagnosis 1 2 3 4 5

8.3 13.5 58.1 0.0 0.0

% Glandular Cell Positive at Endpoint 52.0 (HGD/DALM/intramucosal Ca?) 72.9 (HGD/DALM/intramucosal Ca?) 75.8 (HGD/intramucosal Ca?) 84.2 (cancer) 79.7 (mHGD)

Ca? ⫽ cannot exclude intramucosal cancer.

in the tissue specimens that had exhibited mHGD, HGD DALM, or cancer (Table 2). Not only was p53 detected in all cases, but the percentage of cells staining for p53 had markedly progressed relative to that noted at index diagnosis of LGD (p ⬍ 0.005). In the two patients in our study who were p53 negative at index diagnosis of LGD but who eventually progressed to mHGD/HGD DALM/cancer, tissue samples from follow-up surveillance exams harboring LGD were found to be p53 positive.

DISCUSSION Our observations indicate that the fate of Barrett’s esophagus complicated by LGD was highly variable, with 64.6% regressing, 25.0% persisting, and only 10.4% progressing to mHGD, HGD DALM, or cancer over a period of follow-up spanning up to 82 months. This variable fate with a predominance of regression has also been noted by others (8, 42, 43). However, the impact of pathophysiological factors and/or genetic alterations on the variability of fate of BE during follow-up has not yet been precisely elucidated. Several studies have demonstrated the possible role of the p53 tumor suppressor gene in BE complicated by LGD (22, 29, 44, 45), although some studies have questioned the prognostic significance of p53 protein overexpression in BE (29 –31). Limitations of the immunohistochemical detection of p53 protein overexpression include 1) not all p53 gene mutations result in p53 overexpression as some mutations lead to no expression of protein or expression of truncated protein that does not react with antibody, and 2) p53 overexpression can occur in the absence of gene mutation (46, 47). However, the present immunohistochemical studies demonstrated that the number of p53 positive patients was significantly higher in persistent and progressive groups, which comprised 27% and 60% of the LGD population, respectively, as compared to the regressed group. Our studies showed that the p53 positivity colocalized to the BE mucosal region harboring LGD was a positive predictor of eventual histological progression of BE to HGD, DALM/ HGD, or frank adenocarcinoma during prospective followup. The present observation is consistent with the recent work of Younes et al. (44), who reported that of 25 BE patients with LGD, nine were positive for p53, and of these

five eventually developed HGD/cancer, whereas none of the 16 p53 negative patients developed HGD/cancer. The importance of the p53 tumor suppressor gene in the histological progression of BE noted by the Younes et al. study was restrained by the fact that specific details regarding the endoscopic and biopsy surveillance protocol were not provided and the chronological and histological relationship between p53 positive samples and LGD was not clear. This is distinct from our study, in which these details are provided. Interestingly, the two patients in our study who were p53 negative at index diagnosis of LGD but who eventually progressed had on follow-up surveillance exams additional LGD tissue samples that were found to be p53 positive. Although the prevalence of p53 positivity was significantly greater in the subgroup of BE LGD patients who progressed to mHGD/HGD DALM/cancer, it was far from absolute. Seventy percent (7/10) of the BE LGD p53 positive patients did not progress. However, the final outcome of these seven BE patients will require continued endoscopic follow-up, as their BE histology may change given further observation, especially the three patients in whom LGD persists. Identification of additional endoscopic features, such as hiatal hernia (12) and length of BE (12, 48, 49), and/or use of other biomarkers will be necessary to improve the risk stratification of Barrett’s complicated by LGD. Furthermore, our results imply that, despite the presence of molecular alterations responsible for generating the overexpression of p53, the Barrett’s mucosa can undergo histological improvement during follow-up. This indicates that p53 mutation is an early preneoplastic insult, and for malignant transformation to occur from this preneoplastic stage, further genetic alterations are required. Based upon the results of this study, p53 protein overexpression could play a significant role in the management of BE complicated by LGD. Specifically, the high risk subgroup of LGD patients exhibiting p53 positivity could be advised to enter a proactive secondary prevention–interventional program as opposed to continued close observation (i.e., surveillance approach). For those patients with LGD in the absence of p53 protein overexpression, endoscopic surveillance per American College of Gastroenterology guidelines (5) would be warranted because of the high likelihood of dysplasia regression and low risk of cancer progression. Because our study involved a relatively small number of cases, confirmation of our findings in a larger study population with longer follow-up is warranted. Furthermore, our results are applicable for LGD diagnosis made upon expert pathology review and may not be generalized to pathologists’ interpretations in community practice, as the diagnosis of Barrett’s and dysplasia has recently been shown to be highly variable (50).

ACKNOWLEDGMENTS Supported by the Department of Veterans Affairs and Midwest Biomedical Research Foundation. Histological

AJG – May, 2001

technical assistance by Inga Barringer was also greatly appreciated. Reprint requests and correspondence: Allan P. Weston, M.D., Gastroenterology Section Chief–VAMC, VA Medical Center, 4801 Linwood Boulevard, Kansas City, MO 64128-2226. Received Mar. 9, 2000; accepted Oct. 5, 2000.

p53 Protein, Low Grade Dysplasia, and Barrett’s Esophagus

18. 19. 20. 21.

REFERENCES 1. Haggitt RC. Barrett’s esophagus, dysplasia, and adenocarcinoma. Hum Pathol 1994;25:982–93. 2. Haggitt RC, Reid BJ, Rabinovitch PS, Rubin CE. Barrett’s esophagus. Correlation between mucin histochemistry, flow cytometry, and histologic diagnosis for predicting increased cancer risk. Am J Pathol 1988;131:53– 61. 3. Levine DS. Barrett’s oesophagus and p53. Lancet 1994; 344(8917):212–3. 4. Neshat K, Sanchez CA, Galipeau PC, et al. Barrett’s esophagus: A model of human neoplastic progression. Cold Spring Harb Symp Quant Biol 1994;59:577– 83. 5. Sampliner RE. Practice guidelines on the diagnosis, surveillance, and therapy of Barrett’s esophagus. The Practice Parameters Committee of the American College of Gastroenterology. Am J Gastroenterol 1998;93:1028 –32. 6. Garewal H, Ramsey L, Sharma P, et al. Biomarker studies in reversed Barrett’s esophagus. Am J Gastroenterol 1999;94: 2829 –33. 7. Reid BJ. Barrett’s esophagus and esophageal adenocarcinoma. Gastroenterol Clin North Am 1991;20:817–34. 8. Reid BJ, Blount PL, Rubin CE, et al. Flow-cytometric and histological progression to malignancy in Barrett’s esophagus: Prospective endoscopic surveillance of a cohort. Gastroenterology 1992;102:1212–9. 9. Blount PL, Rabinovitch PS, Reid BJ. DNA content flow cytometry and neoplastic progression in the gastrointestinal tract. In: Eastwood GL, ed. Premalignant conditions of the gastrointestinal tract. Pathogenesis, diagnosis, and management. New York: Elsevier, 1990:55–77. 10. Weston AP, Krmpotich PT, Cherian R, et al. Prospective long-term endoscopic and histological follow-up of short segment Barrett’s esophagus: Comparison with traditional long segment Barrett’s esophagus. Am J Gastroenterol 1997;92: 407–13. 11. Weston AP, Krmpotich PT, Cherian R, et al. Prospective evaluation of intestinal metaplasia and dysplasia within the cardia of patients with Barrett’s esophagus. Dig Dis Sci 1997; 42:597– 602. 12. Weston AP, Badr AS, Hassanein RS. Prospective multivariate analysis of clinical, endoscopic, and histological factors predictive of the development of Barrett’s multifocal high-grade dysplasia or adenocarcinoma. Am J Gastroenterol 1999;94: 3413–9. 13. Reid BJ, Haggitt RC, Rubin CE, et al. Observer variation in the diagnosis of dysplasia in Barrett’s esophagus. Hum Pathol 1988;19:166 –78. 14. Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in human cancers. Science 1991;253(5015):49 –53. 15. Kastan MB, Onyekwere O, Sidransky D, et al. Participation of p53 protein in the cellular response to DNA damage. Cancer Res 1991;51:6304 –11. 16. Culotta E, Koshland DE Jr. p53 sweeps through cancer research. Science 1993;262(5142):1958 – 61. 17. McKay BC, Ljungman M, Rainbow AJ. Potential roles for p53

22. 23.

24. 25. 26. 27.

28.

29. 30. 31. 32. 33. 34.

35. 36. 37. 38.

1361

in nucleotide excision repair. Carcinogenesis 1999;20:1389 – 96. Arrowsmith CH. Structure and function in the p53 family. Cell Death Differ 1999;6:1169 –73. Lohrum MA, Vousden KH. Regulation and activation of p53 and its family members. Cell Death Differ 1999;6:1162– 8. Ikawa S, Nakagawara A, Ikawa Y. p53 family genes: Structural comparison, expression and mutation. Cell Death Differ 1999;6:1154 – 61. el Deiry WS, Harper JW, O’Connor PM, et al. WAF1/CIP1 is induced in p53-mediated G1 arrest and apoptosis. Cancer Res 1994;54:1169 –74. Krishnadath KK, Tilanus HW, van Blankenstein M, et al. Accumulation of p53 protein in normal, dysplastic, and neoplastic Barrett’s oesophagus. J Pathol 1995;175:175– 80. Gimenez A, Minguela A, Parrilla P, et al. Flow cytometric DNA analysis and p53 protein expression show a good correlation with histologic findings in patients with Barrett’s esophagus. Cancer 1998;83:641–51. Ramel S, Reid BJ, Sanchez CA, et al. Evaluation of p53 protein expression in Barrett’s esophagus by two-parameter flow cytometry. Gastroenterology 1992;102:1220 – 8. Jones DR, Davidson AG, Summers CL, et al. Potential application of p53 as an intermediate biomarker in Barrett’s esophagus. Ann Thorac Surg 1994;57:598 – 603. Rice TW, Goldblum JR, Falk GW, et al. p53 immunoreactivity in Barrett’s metaplasia, dysplasia, and carcinoma. J Thorac Cardiovasc Surg 1994;108:1132–7. Hanas JS, Lerner MR, Lightfoot SA, et al. Expression of the cyclin-dependent kinase inhibitor p21(WAF1/CIP1) and p53 tumor suppressor in dysplastic progression and adenocarcinoma in Barrett esophagus. Cancer 1999;86:756 – 63. Polkowski W, van Lanschot JJ, Ten Kate FJ, et al. The value of p53 and Ki67 as markers for tumour progression in the Barrett’s dysplasia-carcinoma sequence. Surg Oncol 1995;4: 163–71. Ireland AP, Clark GW, DeMeester TR. Barrett’s esophagus. The significance of p53 in clinical practice. Ann Surg 1997; 225:17–30. Kubba AK, Poole NA, Watson A. Role of p53 assessment in management of Barrett’s esophagus. Dig Dis Sci 1999;44: 659 – 67. Fitzgerald RC, Triadafilopoulos G. Recent developments in the molecular characterization of Barrett’s esophagus. Dig Dis 1998;16:63– 80. Levine DS. Management of dysplasia in the columnar-lined esophagus. Gastroenterol Clin North Am 1997;26:613–34. Rabinovitch PS, Reid BJ, Haggitt RC, et al. Progression to cancer in Barrett’s esophagus is associated with genomic instability. Lab Invest 1989;60:65–71. Reid BJ, Rabinovitch PS. Barrett’s esophagus. A human model of genomic instability in neoplastic progression. In: Hirschowitz BI, ed. Accomplishments in oncology. Vol. 3. Philadelphia: Lippincott, 1989:32– 44. Weston AP, Sharma P, Topalovski M, et al. Long-term follow-up of Barrett’s high-grade dysplasia. Am J Gastroenterol 2000;95:1887–93. Nigro JJ, Hagen JA, DeMeester TR, et al. Occult esophageal adenocarcinoma. Extent of disease and implications for effective therapy. Ann Surg 1999;230:433– 40. Schmidt HG, Riddell RH, Walther B, et al. Dysplasia in Barrett’s esophagus. J Cancer Res Clin Oncol 1985;110:145– 52. Riddell RH, Goldman H, Ransohoff DF, et al. Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol 1983;14:931– 68.

1362

Weston et al.

39. Bronner MP. Columnar lined epithelium and dysplasia— confirming the diagnosis. Presented at Esophagology for Clinicians: Pathophysiology, Diagnosis & Therapy; December 2– 4, 1999; Orlando. 40. Zoubine MN, Weston AP, Johnson DC, et al. 2-Methoxyestradiol-induced growth suppression and lethality in estrogenresponsive MCF-7 cells may be mediated by down regulation of p34cdc2 and cyclin B1 expression. Int J Oncol 1999;15: 639 – 46. 41. Banerjee SK, Weston AP, Zoubine MN, et al. Expression of cdc2 and cyclin B1 in helicobacter pylori-associated gastric MALT and MALT lymphoma: Relationship to cell death, proliferation, and transformation. Am J Pathol 2000;156:217– 25. 42. Menke-Pluymers MB, Mulder AH, Hop WC, et al. Dysplasia and aneuploidy as markers of malignant degeneration in Barrett’s oesophagus. The Rotterdam Oesophageal Tumour Study Group. Gut 1994;35:1348 –51. 43. Miros M, Kerlin P, Walker N. Only patients with dysplasia progress to adenocarcinoma in Barrett’s oesophagus. Gut 1991;32:1441– 6. 44. Younes M, Ertan A, Lechago LV, et al. p53 protein accumu-

AJG – Vol. 96, No. 5, 2001

45. 46. 47.

48. 49.

50.

lation is a specific marker of malignant potential in Barrett’s metaplasia. Dig Dis Sci 1997;42:697–701. Krishnadath KK, van Blankenstein M, Tilanus HW. Prognostic value of p53 in Barrett’s oesophagus. Eur J Gastroenterol Hepatol 1995;7:81– 4. Hamelin R, Flejou JF, Muzeau F, et al. TP53 gene mutations and p53 protein immunoreactivity in malignant and premalignant Barrett’s esophagus. Gastroenterology 1994;107:1012– 8. Muzeau F, Flejou JF, Potet F, et al. Profile of p53 mutations and abnormal expression of p53 protein in 2 forms of esophageal cancer. Gastroenterol Clin Biol 1996;20:430 –7 (in French). Iftikhar SY, James PD, Steele RJ, et al. Length of Barrett’s oesophagus: An important factor in the development of dysplasia and adenocarcinoma. Gut 1992;33:1155– 8. O’Connor JB, Falk GW, Richter JE. The incidence of adenocarcinoma and dysplasia in Barrett’s esophagus: Report on the Cleveland Clinic Barrett’s Esophagus Registry. Am J Gastroenterol 1999;94:2037– 42. Alikhan M, Rex D, Khan A, et al. Variable pathologic interpretation of columnar lined esophagus by general pathologists in community practice. Am J Gastroenterol 1999;50:23– 6.