Embed Size (px)
Citation preview
(CANCER RESEARCH 52, 2946-2950, May 15, 1992]
Persistent Clonal Areas and Clonal Expansion in Barrett's Esophagus1
Wendy H. Raskind,2 Thomas Norwood, Douglas S. Levine, Rodger C. Haggitt, Peter S. Rabinovitch,
and Brian J. ReidDepartments of Medicine [W. H. R., D. S. L., R. C. H., B. J. R.J and Pathology IT. N., R. C. H., P. S. RJ, university of Washington School of Medicine, Seattle,Washington 98195
ABSTRACT
Three patients with Barrett's esophagus who had cytogenetic abnor
malities detected in their metaplastic epithelium developed high-gradedysplasia or adenocarcinoma during prospective surveillance over aperiod of 1.5 to 6 years. In the 3 cases, cytogenetic abnormalities thatwere associated with the most advanced histológica!lesions were presentin samples obtained 11, 25, and 48 months prior to the diagnosis of high-grade dysplasia or carcinoma. In a fourth patient, marker chromosomesfound in a Barrett's adenocarcinoma were also present in an esophageal
region spatially removed from the tumor. In all four patients, clonalcytogenetic abnormalities were present in samples obtained at widespreadlocations in the Barrett's segment. These observations suggest that insome patients with Barrett's esophagus clonal proliferations arise inregions of benign histology and spread to involve large areas of Barrett'smucosa. These clones persisted when the disease progressed to high-grade dysplasia or adenocarcinoma.
INTRODUCTION
Barrett's esophagus is a condition in which the normal strat
ified squamous mucosa is replaced by a metaplastic columnarmucosa. It develops as a complication of chronic gastroesoph-ageal reflux in 10-12% of patients and predisposes to thedevelopment of esophageal adenocarcinoma (1,2). In the 1970sand 1980s, the incidence of Barrett's associated adenocarcino-
mas, once rare, increased more rapidly than that of any othercancer in the United States (3, 4). Unfortunately, over 90% ofesophageal adenocarcinomas are detected at an advanced, incurable stage (5-7).
Several studies have concluded that endoscopie surveillanceof patients with Barrett's esophagus is warranted for the early
detection of adenocarcinoma (8, 9), but the factors responsiblefor neoplastic progression in this disease are poorly understood.However, it is clear that in Barrett's esophagus evolution from
the normal cell to the malignant one can be associated withlarge changes in DNA content (10-12). For example, four flowcytometric studies of Barrett's esophagus have reported an
increased prevalence of aneuploidy with increasing histologicalrisk of malignancy (10, 13-15). Furthermore, each of thesestudies detected patients whose metaplastic epithelium hadaneuploidy in the absence of high-grade dysplasia or adenocarcinoma, and one prospective study has shown that aneuploidyand increased G2/tetraploid populations identify a subset ofpatients at high risk for progression to high-grade dysplasiaand adenocarcinoma (16).
DNA content flow cytometry cannot, however, identify withcertainty clonal relationships that exist between different regions of the Barrett's segment or that develop serially over time.
Although clonal relationships can be recognized by cytogeneticinvestigations, evaluation of human solid tissue neoplasia isnotoriously difficult and relatively few studies of premalignanttissues have been reported. Barrett's esophagus represents an
ideal condition in which to look for clonal changes that develop
Received 12/10/91; accepted 3/1/92.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by N1H Grant POI DK32971 and Grant PDT316B
from the American Cancer Society.2To whom requests for reprints should be addressed, at Department of
Medicine RG20, University of Washington, Seattle. WA 98195.
with progression because biopsies can be taken from the samepatient during serial surveillance endoscopies. One previousstudy reported karyotypic abnormalities in biopsies of Barrett's
metaplasia that were cultured in vitro for prolonged periods oftime (17). However, neither the spatial extent of the clone inthe Barrett's segment nor its temporal persistence was evalu
ated. Furthermore, because these biopsies were cultured forprolonged periods, it is possible that the observed abnormalitiesrepresented a minority population in the esophagus that had aselective growth advantage in vitro. To investigate the biologyof Barrett's esophagus in individual patients, serial flow cyto
metric, cytogenetic, and histological evaluations were performed as part of a prospective cancer surveillance program.We report findings in three patients whose Barrett's esophagus
had both cytogenetic and flow cytometric abnormalities andwho progressed to high-grade dysplasia or adenocarcinomaduring follow-up. We also report cytogenetic findings in apatient in whom adenocarcinoma was present at entry into thecohort.
MATERIALS AND METHODS
Patients. Case 1, a 72-year-old Caucasian male who presented withan upper gastrointestinal hemorrhage in November 1984 and was foundon endoscopy to have an 8-cm length of Barrett's esophagus (11); Case2, a 58-year-old Caucasian male who presented in January 1985 with along history of reflux esophagitis who had two episodes of uppergastrointestinal hemorrhage secondary to ulcers in a 9-cm segment ofBarrett's esophagus; Case 3, a 67-year-old Caucasian male who pre
sented in April 1985 with a long history of reflux esophagitis and an11-cm segment of Barrett's esophagus who developed recurrent esoph
ageal strictures after failure of a Nissen fundoplication (10); Case 4, a68-year-old Caucasian male with a 20-year history of reflux esophagitiswho presented in November 1985 with an adenocarcinoma arising inBarrett's esophagus.
These four patients are among the 44 patients with Barrett's esoph
agus evaluated by histology and flow cytometry at initial endoscopy aspart of a cancer surveillance study at the University of Washingtonbetween November 1984 and March 1986 (10).
Endoscopy and Biopsy. Experiments were approved by the HumanSubjects Committee at the University of Washington. All patientsparticipated in this study voluntarily and informed consent was obtainedprior to each procedure. Endoscopies and biopsies were performedaccording to a previously published protocol (10). At each endoscopy,the length of the Barrett's segment was recorded as the distance between
the region of the lower esophageal sphincter and the squamocolumnarjunction. The site from which each endoscopie biopsy was obtained wasrecorded as the distance in cm from the incisors. Four quadrant biopsieswere taken at 2-cm levels in the Barrett's segment; one biopsy at each
level was divided in half for histology and flow cytometry. Samples forcytogenetic evaluation were taken from defined levels so that karyotypescould be related to histological and flow cytometric results. Esophagec-tomy specimens were mapped by coordinates on a grid as describedpreviously (11). Endoscopie biopsies or tissue samples obtained atesophagectomy were placed in supplemented L-15 medium (18) fortransport to the tissue culture laboratory. The samples were mincedwith scalpels and disaggregated by incubation with collagenase II(Sigma Chemical Co., St. Louis, MO), 0.4 mg/ml in Hanks' balancedsalt solution for 2 h at 37°C.Samples were cultured overnight in
minimal essential medium (GIBCO, Grand Island, NY), supplementedwith 15% fetal calf serum, L-glutamine (0.3 mg/ml), penicillin (100units/ml) and streptomycin (100 ¿tg/ml)or in Chang medium andserum (10% v/v; Irvine Scientific, Santa Ana, CA), supplemented withinsulin (5 Mg/ml)/transferrin (5 Mg/ml)/selenium (5 ng/ml) (Collaborative Research, Lexington, MA) and antibiotics as above.
Cytogenetics. Colcemid, 0.05 Mg/ml, was added 4 to 12 h prior to
2946
on June 20, 2015. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
CLONAL EVOLUTION IN BARRETT'S ESOPHAGUS
harvest for cytogenetic evaluation. After incubation for 20-30 min in0.075 M KC1, preparations were fixed 3 times in methanol:glacial aceticacid (3:1). Slides were prepared and banded by standard methods.Because the endoscopie biopsies from different levels of the Barrett's
esophagus were processed individually, only 3 or 4 slides could beprepared from each sample.
Flow Cytometry. This was performed as described previously (10-
12,19).Histology. Biopsies were processed for histological evaluation as
described previously (10). Barrett's specialized metaplastic epitheliumwas graded negative for dysplasia, indefinite/low-grade dysplasia, high-grade dysplasia, or carcinoma. The categories of indefinite for dysplasiaand low-grade dysplasia were combined because of the results of a studyof observer variation in the diagnosis of dysplasia in Barrett's esophagus
(20).
RESULTS
The four cases described here were among the group reportedpreviously (10). We performed cytogenetic studies on otherpatients in the University of Washington Barrett's cohort, but
we found that biopsies that contained only metaplastic tissueyielded either no metaphase cells or only cells with normalkaryotypes. The exception, Case 3, was the only patient in ourprevious series with DNA content aneuploidy in metaplasia(Table 1; Ref. 10). A summary of the histological, flow cyto-metric, and cytogenetic evaluations performed on samples obtained simultaneously from the same level of the esophagusduring endoscopie surveillance of the four cases is presented inTable 1.
Case 1. In November 1984, the patient's first endoscopiebiopsy evaluation revealed high-grade dysplasia extending overapproximately 8 cm of Barrett's esophagus and five aneuploid
populations. Endoscopie samples obtained at multiple levels ofthe Barrett's segment in April 1985 and November 1985 revealed high-grade dysplasia and residual foci of Barrett's spe
cialized metaplasia, but no intramucosal carcinoma. The twolargest aneuploid populations in the earlier sample were againdetected by flow cytometry and extended over 10 cm of esoph-ageal mucosa. Biopsies taken in May 1986 documented intramucosal carcinoma. Esophagectomy was performed in July1986 and the specimen contained 13 different aneuploid populations that could be resolved by flow cytometry (11).
Cytogenetic studies were done on biopsies taken in April1985 and November 1985 and from the esophagectomy specimen in July 1986. The adenocarcinoma in the esophagectomyspecimen was so small that it was not visible grossly and, forthis reason, was not sampled for cytogenetic evaluation. However, 2 of the 5 samples submitted for cytogenetic study camefrom the region of a 3.2-3.45 N aneuploid population fromwhich the cancer appeared to arise (11). In all endoscopie andsurgical specimens evaluated by cytogenetics a minority ofmetaphase cells were diploid; the majority of metaphase cellswere aneuploid and contained multiple rearranged chromosomes. Two of the marker chromosomes had distinct configurations and could be identified in metaphase cells even thoughthe exact nature of the rearrangements could not be determined.These two marker chromosomes were documented in April1985 and November 1985 in cells from endoscopically obtainedbiopsies from the Barrett's segment and 15 months later in 4
regions of the surgical specimen.Case 2. This patient presented in January 1985 with 9 cm of
Barrett's metaplasia containing abnormalities in the indefinite/low-grade dysplasia range and increased S phase and G2 fractions. Serial endoscopie biopsies revealed progression to high-grade dysplasia (January 1987) and intramucosal adenocarcinoma (June 1987) (Table 1). Flow cytometry performed onsuccessive endoscopie biopsies between January 1985 and June1987 showed persistently increased S and G: fractions; in somesamples the DNA content histogram suggested the presence ofa hypodiploid aneuploidy that was not consistently resolved.
Esophagectomy was performed in June 1987.Cytogenetic studies were performed on endoscopie biopsies
taken in May 1985, December 1985, June 1986, and June 1987and in samples from the esophagectomy specimen in June 1987.A hypodiploid population with a modal number of 41-42 wasevident in all samples studied and several structurally rearranged and marker chromosomes defining a clone were presentin the samples that were technically évaluable(Fig. 1). Theclone was identified at the first evaluation in May 1985, wasdetected at 7 levels of the Barrett's metaplasia, spanning at
least 4 cm, and persisted for the next 25 months. The adenocarcinoma was not grossly visible in the surgical specimen andwas entirely contained within the portion sent for histologicalexamination. Therefore, it was not sampled for cytogenetics orflow cytometry. However, the hypodiploid clone persisted andwas present in samples taken from two widely separated regionsof the esophagectomy specimen (approximately 5 cm apart).
Case 3. At endoscopy in April 1985, this patient had 12 cmof Barrett's metaplasia negative for dysplasia, but flow cyto-
metric evaluation detected a major aneuploid population of 2.3N (47-76% of cells). A cell with the karyotype 48,XY,+3,+8was detected in a biopsy taken at the same endoscopy. In March1986, flow cytometry again documented an aneuploid population with a DNA content of approximately 2.2 N at multiplelevels of the Barrett's segment. Biopsies submitted for histological evaluation showed Barrett's metaplasia with focal abnormalities in the indefinite/low-grade dysplasia range. At thesame endoscopy, a clone with trisomies for chromosomes 3 and8 was identified and was found at multiple levels of the Barrett's
segment. The 2.2 N aneuploidy and the cytogenetically abnormal clone were both detected in some biopsies from subsequentendoscopies. During the interval between 1989 and 1991, thearea covered by Barrett's metaplasia partially receded and squa-
mous islands appeared distal to the ora serrata.Case 4. At entry into the Barrett's cohort this patient was
found to have an esophageal adenocarcinoma. Samples weretaken from the esophagectomy specimen for cytogenetic analysis. Metaphase cells from the Barrett's segment on the wall
opposite the tumor were found to have the same marker chromosomes seen in the cancer. Flow cytometry showed a majoraneuploid population with a DNA content of 3.4 N.
DISCUSSION
In general, studies of the pathogenesis of human solid tumorsare hampered by problems in identifying patients at high riskof developing malignancy, infeasibility of obtaining serial samples, and difficulty in recovering adequate cytogenetic preparations. Because the development of malignancy in Barrett's
esophagus follows a progression of metaplasia to dysplasia tocarcinoma and endoscopie surveillance allows sequential sampling of epithelium that is at risk of neoplastic progression, thisdisease provides an excellent system for studying the eventsinvolved in this process.
It is not known whether Barrett's specialized metaplasia
develops as a clonal proliferation or results from polyclonalregeneration of the acid-injured mucosa. Aneuploid cell populations have been documented by flow cytometry in endoscopiebiopsies from patients with Barrett's dysplasia and metaplasiawithout dysplasia (10, 13-15). Although DNA content aneuploidy suggests the presence of a clonal proliferation, proof ofclonality requires a more specific marker. There is only oneprevious report of cytogenetic results in Barrett's specializedmetaplasia without high-grade dysplasia or adenocarcinoma(17). The investigators reported that cultured Barrett's epithe
lial cells often exhibited loss of the Y chromosome. In addition,one culture contained extra copies of chromosome 7 and another had a translocation involving chromosomes 3 and 6.These findings suggest that clonal genomic alterations can be
2947
on June 20, 2015. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
CLONAL EVOLUTION IN BARRETT'S ESOPHAGUS
Table 1 Histológica!,DNA content flow cytometric, and cytogenetic evaluations in 4 patients with Barrett's esophagus
Cytogenetic findings
CaseStudydate- Level* Grade'
DNAcontent Clonal marker
No. withabnormalkaryotype
No. withmarker Total
cells
4/85
11/85
7/86°
2 5/85
12/85
6/86
6/87
6/87°
25-2729
31-33
2729313437
1,2*
345
29
293133
32
HGDHGDHGD
HGDHGDHGDHGDHGDCa
HGDHGDHGDHGD
I/LGD
I/LGDI/LGDI/LGD
I/LGD
3.3 N3.3 N3.3 N, 3.4 N3.1 N, 2.9 N
3.3 N3.2 N3.3 N3.3 N3.3 N, 2.9 N13 aneuplo-
idies^
2N2N2N
2N28 IMCa30 LGD
34-35 HGD 2 N
Ca'
2 marker chromosomes with distinct morphology
Hyperdiploid'
Same marker chromosomes
Hyperdiploid'
59-103 chromosomes with same markersHyperdiploid'59-103 chromosomes with same markers
41-42, -Y,der(15)t(9;15)(ql2;pl3),8q+,t(14q21q)*
Same clone
i
Same clone
Hypodiploid'
i
10
2
2
4
4241
13
284
15
461
21
13
261
14
13
2
2
4
4251
13
2106
IS
462
34/853/865/884/894/914
11/85°1229313335283032343634323638IMMMMMIIIMI/LCDI/LGDHGDICa'HGD2.3
N2.3N2.3N2.3N2N2.2
N2.2N2.2
N2.2
N2N3.4
NSame
clonei+3,+8Hyperdiploid'Same
cloneSame
clone1Same
cloneSamecloneSame
clone60-77
chromosomes, no mode, Ml,M2"168-77chromosomes, no mode. Ml, M2311371224114123
114122*41873110>37132441412
°Surgery; all other entries represent endoscopies.b Distance in cm from the incisors at endoscopy or region of the esophagectomy specimen.' Highest histológica! grade at that level: M, metaplasia; I, indefinite for dysplasia; LGD, low-grade dysplasia; HGD, high-grade dysplasia; IMCa, intramucosal
carcinoma; Ca, carcinoma.'' Metaphase cells in which the clonal marker could definitely be identified.' Only chromosome counts were possible.^Given the complex pattern of 13 overlapping aneuploid populations with varying DNA contents detected by flow cytometry, it is not surprising that no distinct
modal number was determined in the cytogenetic samples. Details of the flow cytometry are given by Rabinovitch et al. (11).* The total length of the Barrett's segment was 8 cm. Regions 1-3 were close to each other and to the ora serrata; Regions 4 and 5 were not adjacent to each other
but were both 6 cm from the ora; the carcinoma was 1-2 cm below Region 4.* Less frequent recurrent alterations included trisomy 16, 2q+, 6p+, 6q+, 5q—,and one marker chromosome.' Regions 1 and 2 were 1.5 and 4.0 cm, respectively, from the ora serrata; 5 cm from each other; and at least 3 cm from the focus of cancer.' Nine other cells appeared to have a diploid chromosome content but could not be analyzed in detail.* Subclone of the originally detected clone with +20 as well as +3 and +8.' Region 1 was on the opposite esophageal wall from the carcinoma."Ml, isochromosome of 9q; M 2, 4q+; other abnormal chromosomes were also present.
present in the mucosa of some patients with Barrett's esophagus
who do not have dysplasia or cancer. However, multiple regionswere not sampled from a single patient; therefore the extent ofthe abnormalities could not be determined, nor were patientsreevaluated to document the persistence of the cytogeneticabnormalities. To investigate whether development and progression of Barrett's esophagus involve a persistent clonal proc
ess we repeatedly evaluated patients enrolled in a surveillanceprogram. To decrease the possible artifact of in vitro selectionof minority populations, we harvested samples for cytogenetic
analysis after overnight incubation.Our findings suggest that neoplastic progression in Barrett's
esophagus involves the development of abnormal clones thatcan expand to occupy large regions of esophageal mucosa andpersist for prolonged periods. The presence of abnormal cloneswas suspected in three of the four patients (Cases 1, 3, and 4)in whom aneuploid populations were identified by flow cytometry. In three patients who had or developed adenocarcinoma(Cases 1, 2, and 4), cytogenetically abnormal clones were documented to exist in regions of the Barrett's segment that were
2948
on June 20, 2015. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
CLONAL EVOLUTION IN BARRETT'S ESOPHAGUS
«ttouAft. IR2 ' 3
M10 11 12
•¿�H"A*13 14 15
AS« IM A16 17 18
•¿�4 »»19 20 21
ft22
ïîfSK NA Aj
8 10 11 12
13 14 16 17 18
XX19
^20 21 22 XY
Fig. l . R-banded karyotypes from endoscopie biopsies from Case 2. Arrows, recurrent marker chromosomes. All analyzable abnormal metaphase cells at all dateswere lacking the Y chromosome, a, obtained May 8, 1985: 41 chromosomes with trisomy 16, der(2)t(2;?)(q35;?), del(5)(ql4), der(8)t(8;?)(q24;?), rob(14;21),der(15)t(9;15)(ql I;pl3), and small pale metacentric marker. The 2q+ (3 of 10 cells) and trisomy 16 (5 of 10 cells) were not identified in later samples, b, obtainedDecember 4, 1985: 38 chromosomes with der(6)t(6;?)(qter;?), der(8)t(8;?)(q24;?), rob(14;21), der(15)t(9;15)(qll;pl3), and del(17)(pll). The 17p abnormality wasseen only in this cell.
2949
on June 20, 2015. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
CLONAL EVOLUTION IN BARRETTS ESOPHAGUS
distinct from the adenocarcinoma. In these 3 patients as wellas in Case 2, in whom the DNA content was diploid, cytogeneticevaluations documented the presence of a clone. Furthermore,three of the patients developed karyotypic abnormalities priorto the detection of high-grade dysplasia (Case 3), adenocarcinoma (Case 1), or both high-grade dysplasia and cancer (Case2).
Persistence of a cytogenetically abnormal clone was found inCases 2 and 3. In Case 2, prospective endoscopie biopsy surveillance documented histological progression from Barrett'smetaplasia with indefinite/low-grade dysplasia abnormalities tohigh-grade dysplasia. A hypodiploid clone was initially detectedin regions that had metaplasia or abnormalities in the indefinite/low-grade dysplasia range and was subsequently detectedin a region of high-grade dysplasia that developed during surveillance. The cancer eventually developed within the high-grade dysplasia. During this time, multiple subclonal karyotypicchanges were detected within the predominant hypodiploidclone (Fig. 1).
The presence of an aneuploid clone may not predict inevitableprogression of disease and development of cancer. Despite thepersistence of a hyperdiploid clone for at least 6 years and thedevelopment of high-grade dysplasia, the extent of Barrett's
metaplasia in Case 3 has been observed to decrease. Continuedsurveillance will be required to determine whether or not thispatient develops cancer in the future.
These cases demonstrate the complementary nature of cyto-genetics and flow cytometry. DNA content flow cytometry candetermine the relative proportions of populations of differentploidies but may not detect near-diploid aneuploidies or minority populations. Cytogenetics, on the other hand, is not quantitative. This method is able to identify a clone, even when itcomprises a small number of cells, but cannot estimate itsprevalence. Cytogenetics can prove that an aneuploidy detectedby flow cytometry represents a clone and can identify subclonalrelationships.
One patient (Case 4) was found to have adenocarcinoma onentry to our study. An abnormal clone was detected by cytoge-netics in the noncancerous mucosa as well as in the adenocarcinoma. A theoretical possibility in this case is that the abnormal metaphase cells recovered from the region of the Barrett's
segment that was spatially separate from the cancer were contributed by an unidentified microscopic focus of carcinoma,because these specimens were obtained simultaneously. However, it is unlikely that this explanation accounts for the abnormalities in Cases 1-3, which persisted for several years beforehigh-grade dysplasia (Case 3) or cancer (Cases 1 and 2) wasdiagnosed. The possibility that the abnormal metaphase cellsin Cases 1-3 came from undetected regions of high-gradedysplasia or adenocarcinoma is made even less likely becauseof the intensive endoscopie biopsy surveillance in these patients(10). An average of 6 endoscopies (range, 5-8) and 173 totalbiopsies (range, 154-199) were performed per patient.
We have provided evidence that Barrett's metaplasia can
contain cytogenetically abnormal clones that occupy extensiveregions of the Barrett's segment, persist for several years, andprogress to high-grade dysplasia and adenocarcinoma. Takentogether, these findings are consistent with Nowell's hypothesis
(21 ) and suggest that neoplastic progression in Barrett's esoph
agus occurs by a multistep pathway of clonal evolution.
ACKNOWLEDGMENTS
We are grateful to Steven Forbes, Susan Irvine, Catherine Morgan,Carissa Sanchez, and John Wolff for excellent technical assistance andto Dr. Harinder Garewal for helpful suggestions.
REFERENCES1. Sjogren, R. W., and Johnson, L. F. Barrett's esophagus: a review. Am. J.
Med., 74:313-321, 1983.2. Spechler, S. J., and Goyal, R. K. Barrett's esophagus. N. Engl. J. Med., 315:
362-371, 1987.3. Blot, J. W., Devesa, S. S., Kneller, R. W., and Fraumeni, J. F. Rising
incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA,265:1287-1289, 1991.
4. Yang, P. C., and Davis, S. Incidence of cancer of the esophagus in the US byhistologie type. Cancer (Phila.), 61: 612-617, 1988.
5. Skinner, D. B., Walther, B. C., Riddell, R. H., Schmidt, H., lascone, C, andDeMeester, T. R. Barrett's esophagus. Comparison of benign and malignantcases. Ann. Surg., 198: 554-566, 1983.
6. Witt, T. R., Bains, M. S., Zaman, M. B., and Martini, N. Adenocarcinomain Barrett's esophagus. J. Thorac. Cardiovasc. Surg., SS: 337-345, 1983.
7. Silverberg, E., Boring, C. C., and Squires, T. S. Cancer statistics, 1990.Cancer (Phila.), 40:9-26, 1990.
8. Spechler, S. J. Endoscopie surveillance for patients with Barrett's esophagus:does the cancer risk justify the practice? Ann. Intern. Med., 106: 902-904,1987.
9. Robertson, C. S., Mayberry, J. F., Nicholson, D. A., James, P. D., andAtkinson, M. Value of endoscopie surveillance in the detection of neoplasticchange in Barrett's oesophagus. Br. J. Surg., 75: 760-763, 1988.
10. Reid, B. J., Haggitt, R. C., Rubin, C. E., and Rabinovitch, P. S. Barrett's
esophagus. Correlation between flow cytometry and histology in detection ofpatients at risk for adenocarcinoma. Gastroenterology, 93:1-11, 1987.
11. Rabinovitch, P. S., Reid, B. J., Haggitt, R. C., Norwood, T. H., and Rubin,C. E. Progression to cancer in Barrett's esophagus is associated with genomicinstability. Lab. Invest., 60:65-71. 1989.
12. Blount, P. L., Rabinovitch, P. S., and Reid, B. J. DNA content flow cytometryand neoplastic progression in the gastrointestinal tract. /;/: G. L. Eastwood(ed.), Premalignant Conditions of the Gastrointestinal Tract, pp. 55-77. NewYork: Elsevier Science Publishers, 1991.
13. McKinley, M. J., Budman. D. R., Grueneberg, M. S., Bronzo, R. L., Weiss-man, G. S., and Kahn, E. DNA content in Barrett's esophagus and esophagealmalignancy. Am J. Gastroenterol., 82: 1012-1015, 1987.
14. Fennerty, M. B., Sampliner, R. E., Way, D., Riddell, R., Steinbronn, K., andGarewal, H. S. Discordance between (low cytometric abnormalities anddysplasia in Barrett's esophagus. Gastroenterology, 97: 815-820, 1989.
15. Ahnen, D. J., Bunn, P. A., and Doos. W. G. Aneuploidy and Barrett's
esophagus: data from VA cooperative study #277. Gastroenterology, 98:A15, 1990.
16. Reid, B. J., Blount, P. L., Rubin, C. E., Levine, D. S., Haggitt, R. C., andRabinovitch, P. S. Predictors of progression to malignancy in Barrett's
esophagus: endoscopie, histologie and flow cytometric followup of a cohort.Gastroenterology, 98: A305, 1991.
17. Garewal, H. S., Sampliner, R., Liu, Y., and Trent, J. M. Chromosomalrearrangements in Barrett's esophagus. Cancer Genet. Cytogenet., 42: 281-
296, 1989.18. Leibovitz, A. Development of tumor cell lines. Cancer Genet. Cytogenet.,
19: 11-19, 1986.19. Haggitt, R. C., Reid, B. J., Rabinovitch. P. S., and Rubin, C. E. Barrett's
esophagus. Correlation between mucin histochemistry, flow cytometry, andhistologie diagnosis for predicting increased cancer risk. Am. J. Pathol., 131:53-61, 1988.
20. Reid, B. J., Haggitt. R. C., Rubin, C. E., Roth, G., Surawicz, C. M., VanBelle, G., Lewin, K., Weinstein, W. M., Antoni,ili, D. A., Goldman, H.,MacDonald, W., and Owen, D. Observer variation in the diagnosis ofdysplasia in Barrett's esophagus. Hum. Pathol., 19: 166-178, 1988.
21. Nowell, P. J. The clonal evolution of tumor cell populations. Science (Washington DC), 194: 23-28, 1976.
2950
on June 20, 2015. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
1992;52:2946-2950. Cancer Res Wendy H. Raskind, Thomas Norwood, Douglas S. Levine, et al. EsophagusPersistent Clonal Areas and Clonal Expansion in Barrett's
Updated version
http://cancerres.aacrjournals.org/content/52/10/2946
Access the most recent version of this article at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
To request permission to re-use all or part of this article, contact the AACR Publications
on June 20, 2015. © 1992 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
