Distinct Chromosomal Imbalances in Nonpolypoid and Polypoid Colorectal Adenomas Indicate Different Genetic Pathways in the Development of Colorectal Neoplasms
2003; Elsevier BV; Volume: 163; Issue: 1 Linguagem: Inglês
10.1016/s0002-9440(10)63652-8
ISSN1525-2191
AutoresHedwig Richter, Premysl Slezak, Axel Walch, Martin Werner, Herbert Braselmann, Edgar Jaramillo, Åke Öst, Ichiro Hirata, Kazuya Takahama, Horst Zitzelsberger,
Tópico(s)Colorectal and Anal Carcinomas
ResumoCytogenetic changes are widely unknown for nonpolypoid (synonymously termed as “flat” or “depressed”) colorectal adenomas. A comparison with polypoid adenomas will contribute to the discussion whether different genetic pathways for colorectal tumorigenesis depending on its origin from nonpolypoid or polypoid adenomas exist. Tissue samples of nonpolypoid (n = 22), polypoid (n = 28) adenomas, carcinomas ex-nonpolypoid adenomas (n = 9), carcinomas ex-polypoid adenomas (n = 14), and normal colonic mucosa (n = 9) were investigated by comparative genomic hybridization of whole genomic DNA. Chromosomal imbalances were detected from average comparative genomic hybridization profiles for each entity. Nonpolypoid adenomas show recurrent chromosomal losses on chromosomes 16, 17p, 18, 20, and 22 and gains on chromosomes 2q, 4q, 5, 6, 8q, 12q, and 13q. In polypoid adenomas losses of whole chromosomes 16, 18, and 22 and gains of chromosomes 7q and 13 were detected. The frequency of copy number changes was higher in nonpolypoid compared to polypoid adenomas and early onset of chromosomal changes became apparent in low-grade dysplasias of nonpolypoid adenomas. Gains on chromosomes 2q, 5, 6, 8q, and 12q and losses on chromosomes 17p and 20 occurred exclusively in nonpolypoid adenomas, whereas 16p deletions are significantly more frequent in nonpolypoid than in polypoid adenomas. Carcinomas ex-nonpolypoid adenomas are characterized by more complex aberration patterns compared to nonpolypoid adenomas exhibiting frequent losses on chromosomes 8p, 12q, 14, 15q, 16, 17p, 18, and 22 and gains on 3q, 5, 6, 7, 8q, 12q, and 13, respectively. Normal colonic mucosa showed no chromosomal imbalances. Distinct differences of chromosomal imbalances between nonpolypoid and polypoid colorectal adenomas have been characterized that support the hypothesis that different genetic pathways may exist in the development of colorectal adenomas exhibiting nonpolypoid and polypoid phenotype. Cytogenetic changes are widely unknown for nonpolypoid (synonymously termed as “flat” or “depressed”) colorectal adenomas. A comparison with polypoid adenomas will contribute to the discussion whether different genetic pathways for colorectal tumorigenesis depending on its origin from nonpolypoid or polypoid adenomas exist. Tissue samples of nonpolypoid (n = 22), polypoid (n = 28) adenomas, carcinomas ex-nonpolypoid adenomas (n = 9), carcinomas ex-polypoid adenomas (n = 14), and normal colonic mucosa (n = 9) were investigated by comparative genomic hybridization of whole genomic DNA. Chromosomal imbalances were detected from average comparative genomic hybridization profiles for each entity. Nonpolypoid adenomas show recurrent chromosomal losses on chromosomes 16, 17p, 18, 20, and 22 and gains on chromosomes 2q, 4q, 5, 6, 8q, 12q, and 13q. In polypoid adenomas losses of whole chromosomes 16, 18, and 22 and gains of chromosomes 7q and 13 were detected. The frequency of copy number changes was higher in nonpolypoid compared to polypoid adenomas and early onset of chromosomal changes became apparent in low-grade dysplasias of nonpolypoid adenomas. Gains on chromosomes 2q, 5, 6, 8q, and 12q and losses on chromosomes 17p and 20 occurred exclusively in nonpolypoid adenomas, whereas 16p deletions are significantly more frequent in nonpolypoid than in polypoid adenomas. Carcinomas ex-nonpolypoid adenomas are characterized by more complex aberration patterns compared to nonpolypoid adenomas exhibiting frequent losses on chromosomes 8p, 12q, 14, 15q, 16, 17p, 18, and 22 and gains on 3q, 5, 6, 7, 8q, 12q, and 13, respectively. Normal colonic mucosa showed no chromosomal imbalances. Distinct differences of chromosomal imbalances between nonpolypoid and polypoid colorectal adenomas have been characterized that support the hypothesis that different genetic pathways may exist in the development of colorectal adenomas exhibiting nonpolypoid and polypoid phenotype. Besides polypoid adenomas, another type of colorectal neoplasia has endoscopically been identified as a nonpolypoid adenoma.1Muto T Kamiya J Sawada T Konishi F Sugihara K Kubota Y Adachi M Agawa S Saito Y Morioka Y Tanprayoon T Small “non-polypoid adenoma” of the large bowel with special reference to its clinicopathologic features.Dis Colon Rectum. 1985; 28: 847-851Crossref PubMed Scopus (448) Google Scholar The terms superficial, flat, and depressed adenoma are all used synonymously to describe this entity. Because flat or depressed adenomas display little or no mucosal elevation, endoscopical diagnosis can be very difficult. This type of nonpolypoid adenoma is associated with a higher potential of malignancy.2Adachi M Muto T Okinaga K Morioka Y Clinicopathologic features of the flat adenoma.Dis Colon Rectum. 1991; 34: 981-986Crossref PubMed Scopus (134) Google Scholar, 3Kasumi A Kratzer GL Fast-growing cancer of the colon and rectum.Am Surg. 1992; 58: 383-386PubMed Google Scholar, 4Watanabe T Sawada T Kubota Y Adachi M Saito Y Masaki T Muto T Malignant potential in non-polypoid elevations.Dis Colon Rectum. 1993; 36: 548-553Crossref PubMed Scopus (54) Google Scholar Nonpolypoid colorectal neoplasias exhibit histological features that differ from those of polypoid adenomas, being mainly tubular structures that may show high-grade dysplasia at an early stage.5Kuramoto S Ihara O Sakai S Shimazu R Kaminishi M Oohara T Depressed adenoma in the large intestine. Endoscopic features.Dis Colon Rectum. 1990; 33: 108-112Crossref PubMed Scopus (69) Google Scholar, 6Matsumoto T Iida M Kuwano Y Tada S Yao T Fujishima M Minute non-polypoid adenoma of the colon detected by colonoscopy: correlation between endoscopic and histologic findings.Gastrointest Endosc. 1992; 38: 645-650Abstract Full Text PDF PubMed Scopus (53) Google Scholar, 7Jaramillo E Watanabe M Slezak P Rubio C Non-polypoid neoplastic lesions of the colon and rectum detected by high-resolution video endoscopy and chromoscopy.Gastointest Endosc. 1995; 42: 114-122Abstract Full Text Full Text PDF Scopus (256) Google Scholar The grade of dysplasia in nonpolypoid colorectal neoplasias may not necessarily be related to size or to a villous component. During the past few years, colorectal tumorigenesis has been associated with the progressive acquisition of a variety of genomic alterations by neoplastic cells, some of these have been linked to early stages of evolution. Two groups have been distinguished according to their genetic alterations.8Kinzler KW Vogelstein B Lessons from hereditary colorectal cancer.Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4260) Google Scholar The major group, called LOH+ (loss of heterozygosity) or microsatellite stability (MSS), is characterized by frequent allelic losses associated with mutations that inactivate tumor suppressor genes (APC, p53) and account for at least two-thirds of tumors; the second group of tumors exhibit a high frequency of replication error at microsatellite loci, thus termed RER+ (replication error) or microsatellite instability (MSI), and were shown to be impaired in the DNA mismatch repair pathway. Furthermore, genes of the ras family have been found activated by missense mutations in 45% of colorectal neoplasias. It has been reported that there may be a different clinical behavior and histopathological character between nonpolypoid and polypoid adenomas,2Adachi M Muto T Okinaga K Morioka Y Clinicopathologic features of the flat adenoma.Dis Colon Rectum. 1991; 34: 981-986Crossref PubMed Scopus (134) Google Scholar, 3Kasumi A Kratzer GL Fast-growing cancer of the colon and rectum.Am Surg. 1992; 58: 383-386PubMed Google Scholar, 4Watanabe T Sawada T Kubota Y Adachi M Saito Y Masaki T Muto T Malignant potential in non-polypoid elevations.Dis Colon Rectum. 1993; 36: 548-553Crossref PubMed Scopus (54) Google Scholar, 5Kuramoto S Ihara O Sakai S Shimazu R Kaminishi M Oohara T Depressed adenoma in the large intestine. Endoscopic features.Dis Colon Rectum. 1990; 33: 108-112Crossref PubMed Scopus (69) Google Scholar, 6Matsumoto T Iida M Kuwano Y Tada S Yao T Fujishima M Minute non-polypoid adenoma of the colon detected by colonoscopy: correlation between endoscopic and histologic findings.Gastrointest Endosc. 1992; 38: 645-650Abstract Full Text PDF PubMed Scopus (53) Google Scholar, 7Jaramillo E Watanabe M Slezak P Rubio C Non-polypoid neoplastic lesions of the colon and rectum detected by high-resolution video endoscopy and chromoscopy.Gastointest Endosc. 1995; 42: 114-122Abstract Full Text Full Text PDF Scopus (256) Google Scholar suggesting an alternative pathway in the genesis of colorectal cancer.9Yamagata S Muto T Uchida Y Masaki T Sawada T Tsuno N Hirooka T Lower incidence of K-ras mutation in non-polypoid colorectal adenomas than in polypoid adenomas.Jpn J Cancer Res. 1994; 85: 147-151Crossref PubMed Scopus (130) Google Scholar, 10Hasegawa H Ueda M Watanabe M Teramoto T Mukai M Kitajima M K-ras gene mutations in early colorectal cancer: flat elevated vs poly-forming cancer.Oncogene. 1995; 10: 1413-1416PubMed Google Scholar However, the clinical and molecular genetic dignity of nonpolypoid neoplastic lesions still remain rather unclear. The molecular genetic characteristics of nonpolypoid colorectal adenomas has been described in relatively few publications. K-RAS mutations were significantly associated with the polypoid phenotype of the adenomas. Further, loss of heterozygosity at chromosomes 3p, 2p, 5q, 17p, and 18q have been reported11Yashiro M Carethers JM Laghi L Saito K Slezak P Jaramillo E Rubio C Koizumi K Hirakawa K Boland CR Genetic pathways in the evolution of morphologically distinct colorectal neoplasms.Cancer Res. 2001; 61: 2676-2683PubMed Google Scholar and the frequency of somatic mutations of the APC and p53 genes in nonpolypoid colorectal adenomas12Wyk van R Slezak P Hayes VM Buys CHHCM Kotze M de Jong G Rubio C Dolk A Jaramillo E Koizumi K Grobbelaar J Somatic mutations of the APC, KRAS, and TP53 genes in nonpolypoid colorectal adenomas.Genes Chromosom Cancer. 2000; 27: 202-208Crossref PubMed Scopus (26) Google Scholar has been described. Other studies have focused on the replication error status and transforming growth factor-βRII mutations paying attention also to mutations of APC and K-RAS.13Olschwang S Slezak P Roze M Jaramillo E Nakano H Koizumi K Rubio CA Larent Puig P Thomas G Somatically acquired genetic alterations in non-polypoid colorectal neoplasias.Int J Cancer. 1998; 77: 366-369Crossref PubMed Scopus (40) Google Scholar A recent publication reports on the aberrant expression of G1 phase cell-cycle regulators in nonpolypoid and polypoid colorectal adenomas.14Bartkova J Thullberg M Slezak P Jaramillo E Rubio C Thomassen LH Bartek J Aberrant expression of G1-phase cell cycle regulators in non-polypoid and exophytic adenomas of the human colon.Gastroenterology. 2001; 120: 1680-1688Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar To date these sparse molecular genetic investigations, with the exception of low K-RAS mutation rate in nonpolypoid adenomas, indicate that nonpolypoid adenomas seem to follow the tumorigenesis pathways very similar to those identified in polypoid adenomas and carcinomas. All of the molecular genetic results mentioned above are based mainly on the analysis of individual genes that are known to be involved in the colorectal tumorigenesis pathway and they do not satisfactorily explain the differences that seem to exist between colorectal adenomas with polypoid and nonpolypoid phenotype. The aim of this study is to investigate the entire genome of nonpolypoid adenomas for chromosomal imbalances by means of comparative genomic hybridization (CGH) profiling and to examine whether there is a specific genotype for nonpolypoid adenomas, which could explain the different clinical behavior and histopathological character between both phenotypes of colorectal adenomas. In this study we investigated 22 nonpolypoid adenomas, 9 carcinomas ex-nonpolypoid adenomas, and 28 polypoid adenomas by CGH and compared resulting chromosomal gains and losses to obtain any clues for differences in the tumorigenic pathway. The tissue samples selected for CGH analysis in this study have been either removed at colonoscopy (polypoid and nonpolypoid adenomas, carcinomas ex-nonpolypoid adenomas) or by surgical resection (carcinomas ex-polypoid carcinomas). A total of 22 nonpolypoid adenomas and 9 carcinomas ex-nonpolypoid adenomas from 27 patients (5 females, 22 males) were investigated in the present study. Additionally, 28 polypoid adenomas from 27 patients (13 female, 14 male) as well as 14 polypoid carcinomas from 14 patients were studied for comparison. Nine normal mucosa samples were examined for control. Histopathological data of the patients are summarized in Table 1, Table 2. For nonpolypoid neoplasias tumor size ranges from 3 to 35 mm (mean, 11.3 mm), tumors were located for 14 cases in the right colon and for 17 cases in the left colon and patients had an age range of 30 to 91 years (mean, 69 years). For polypoid neoplasias the tumor size was determined between 5 and 40 mm (mean, 10.6 mm), 5 tumors showed a location in the right and 23 tumors in the left colon and patients (14 males, 13 females) had an age range of 56 to 84 years (mean, 69 years).Table 1Summary of Clinical and Histopathological Characteristics of Nonpolypoid and Polypoid Colorectal AdenomasCase no.Size (mm)LocalizationAge*Age at diagnosis in years; M, male; F, female./genderHistology†TA, Tubulary adenoma; TVA, tubulovillous adenoma; SA, serrated adenoma; VA, villous adenoma; Intramuc. C, intramucosal carcinoma; LGD, low-grade dysplasia; HGD, high-grade dysplasia.Nonpolypoid adenomas 120Coecum82/FLGD/TA 2‡Samples of different loci from one patient.6Coecum71/MLGD/TA 3‡Samples of different loci from one patient.12Coecum71/MLGD/TA 4‡Samples of different loci from one patient.8Coecum71/MLGD/TA 57C. descendens79/MLGD/TA 66C. transversum73/MLGD/TA 75C. descendens63/MLGD/TA 87C. sigmoideum60/MLGD/TA 910C. descendens73/MLGD/TA 1010C. transversum30/MLGD/TA 115C. ascendens81/MLGD/TA 1220C. ascendens70/FLGD/TA 1310C. sigmoideum63/MLGD/TA 145C. sigmoideum68/FLGD/TA 153C. transversum57/MLGD/TA 1610C. ascendens73/MLGD/TA 174C. sigmoideum56/MLGD/TA 185C. sigmoideum56/MLGD/SA 196C. transversum39/MLGD/SA 205C. sigmoideum91/FHGD/TA 2110C. descendens45/MHGD/TA 224C. descendens44/MHGD/VAPolypoid adenomas 325C. sigmoideum63/MLGD/TA 3310C. sigmoideum76/FLGD/TA 3410C. transversum75/FLGD/TA 355C. sigmoideum64/FLGD/TA 36‡Samples of different loci from one patient.10C. sigmoideum54/FLGD/TA 37‡Samples of different loci from one patient.10C. sigmoideum54/FLGD/TA 387Rectum62/MLGD/TA 397Rectum66/MLGD/TA 4010C. descendens67/FLGD/TA 4140C. transversum71/FLGD/TVA 4210Rectum56/MLGD/TVA 435C. sigmoideum56/MLGD/TVA 448C. sigmoideum80/MLGD/TVA 4510C. descendens84/MLGD/TVA 467C. transversum54/MLGD/TVA 4715Rectum83/FLGD/TVA 487C. sigmoideum61/FLGD/SA 495Rectum56/FLGD/SA 5015Coecum56/MHGD/TA 517C. sigmoideum64/FHGD/TA 5210Rectum61/MHGD/TA 535C. sigmoideum55/MHGD/TA 5415Rectum62/MHGD/TA 5515Coecum56/MHGD/TA 5610C. sigmoideum68/FHGD/TA 5710C. sigmoideum79/MHGD/TA 5820C. descendens78/FHGD/TVA 5910C. sigmoideum67/FHGD/TVA* Age at diagnosis in years; M, male; F, female.† TA, Tubulary adenoma; TVA, tubulovillous adenoma; SA, serrated adenoma; VA, villous adenoma; Intramuc. C, intramucosal carcinoma; LGD, low-grade dysplasia; HGD, high-grade dysplasia.‡ Samples of different loci from one patient. Open table in a new tab Table 2DNA Copy Number Changes (≥2 Alterations) in Nonpolypoid Neoplasias (N = 31) and Polypoid Neoplasias (n = 42), Total n = 73Pathological diagnosisNonpolypoid neoplasiasPolypoid neoplasiasLow-grade adenoma10/193/18High-grade adenoma1/34/10Carcinoma9/914/14Total20/3121/42 Open table in a new tab Formalin-fixed and paraffin-embedded tissue specimens have been cut and stained with hematoxylin and eosin (H&E) according to standard procedures for routine histopathological examination. For CGH analyses representative areas of the neoplasms were microdissected from tissue sections. The histopathological classification of each sample was re-evaluated independently by two pathologists. Predominantly tubulary adenomas were present (n = 19 for nonpolypoid and n = 17 for polypoid adenomas). Only a minority of cases showed serrated (n = 2 for nonpolypoid and polypoid adenomas each), tubulovillous (n = 9 for polypoid adenomas), or villous (n = 1 for nonpolypoid adenomas) subtypes. Nonpolypoid (ie, nonprotruding) adenomas were defined as lesions lacking an exophytic, polypoid configuration at colonoscopy, usually consisting of slightly elevated dysplastic mucosal plaques never more than two times the thickness of the adjacent nondysplastic segments at histopathological examination. Polypoid (ie, exophytic) adenomas were defined as protruding lesions either with (pedunculated) or without (sessile) stalks. Histopathological classification was performed according to World Health Organization.15Hamilton SR Aaltonen LA World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Digestive System. IARC Press, Lyon2000: 112-113Google Scholar In particular, criteria for defining nonpolypoid (flat) adenomas include the presence of slightly elevated lesions with flat or depressed surface lacking an exophytic polypoid configuration. Those lesions that were classified as carcinomas ex-nonpolypoid (flat) adenomas demonstrated in addition to the above described criteria the presence of invasion, containing invasive carcinoma. Lesions that were classified as carcinomas ex-polypoid adenomas are exophytic cancers with presence of residual (polypoid) adenoma. The analyses were performed on archival material from formalin-fixed tissues embedded in paraffin. Serial sections were cut for tissue microdissection (5 μm). The first and the last section were stained with H&E. Microdissection was performed on hematoxylin-stained sections under an inverted microscope yielding at least 80 to 90% pure neoplastic cells for DNA extraction. At least four to six serial sections were sampled, depending on the size of area. The cells were transferred into a sterile microcentrifuge tube and the DNA was extracted with NucleoSpin Tissue Kit (Clontech Laboratories, Inc., Palo Alto, CA) as specified by the manufacturer. For amplification, MseI restriction endonuclease digest followed by ligation-mediated polymerase chain reaction (PCR) were performed on DNA extracts according to Klein and colleagues.16Klein CA Schmidt-Kittler O Schardt JA Pantel K Speicher MR Riethmüller G Comparative genomic hybridization, loss of heterozygosity, and DNA sequence analysis of single cells.Proc Natl Acad Sci USA. 1999; 96: 4494-4499Crossref PubMed Scopus (364) Google Scholar Briefly, the MseI-digested DNA was ligated by annealing of the primers MseLig 21 and MseLig 12. At 15°C the T4-DNA-ligase (Roche Diagnostics, Mannheim, Germany) was added and primers and DNA fragments were ligated overnight. After primary PCR amplification the size of DNA fragments was checked by agarose gel electrophoresis and should range between 500 and 1500 bp. The amplified DNA was indirectly labeled by PCR using biotin-16-dUTP (Roche Diagnostics) for the tumor DNA and digoxigenin-11-dUTP (Roche Diagnostics) for the female or male reference DNA. The amplification products were then MseI digested for 3 hours at 37°C. After purification the DNAs were used for subsequent CGH analysis. For each CGH hybridization 1 μg of tumor DNA and 800 ng of normal female or male DNA, plus 25 μg of CotI DNA (Invitrogen, Karlsruhe, Germany) and 20 μg of herring sperm DNA (Sigma-Aldrich, Taufleivchen, Germany) were co-hybridized to denatured metaphases for 72 hours at 37°C. After hybridization, biotin-labeled tumor DNA was detected with avidin-FITC DCS (Vector Laboratories, Wertheim-Bettingen, Germany), the digoxigenin-labeled control DNA with anti-digoxigenin-rhodamine (Roche Diagnostics). Slides were counterstained with 4,6-diamidino-2-phenylindole in anti-fade solution and mounted in Vectashield (Vector Laboratories, Burlingame, CA). For CGH analysis, at least 10 metaphases were captured and karyotyped after visualization with a Zeiss Axioplan 2 fluorescence microscope (Zeiss, Oberkochen, Germany) equipped with filter sets (single-band excitation filters for 4′-6-diamidino-2-phenylindole, Cy2, and Texas Red). Averaged profiles were generated by CGH analysis software (ISIS 3, V2.84; MetaSystems, Altlussheim, Germany) from at least 10 to 15 homologous chromosomes and interpreted according to published criteria17Kallioniemi OP Kallioniemi A Piper J Isola J Waldman FM Gray JW Pinkel D Optimizing comparative genomic hybridization for analysis of DNA sequence copy number changes in solid tumors.Genes Chromosom Cancer. 1994; 10: 231-243Crossref PubMed Scopus (880) Google Scholar, 18Solinas-Toldo S Wallrapp C Muller-Pillasch F Bentz M Gress T Lichter P Mapping of chromosomal imbalances in pancreatic carcinoma by comparative genomic hybridization.Cancer Res. 1996; 56: 3803-3807PubMed Google Scholar using statistical confidence limits based on t-statistics. A chromosomal gain was classified as high-level amplification when the CGH ratio exceeded a value of 1.5, or if the Cy2 fluorescence showed a strong, distinct signal by visual inspection and the corresponding ratio profile was diagnostic of overrepresentation. Telomeric regions, heterochromatic regions, and the Y chromosome were all excluded from analyses, because of interindividual variations within these regions. Ligation-mediated PCR-amplified DNA obtained from morphologically normal mucosa DNA was co-hybridized with male or female reference DNA to metaphase preparations. In these experiments no chromosomal changes were detected except for chromosome region 1p34-36 and chromosome 19. These regions are known to show artifactual results by CGH.19Weber RG Scheer M Born IA Joos S Cobbers JM Hofele C Reifenberger G Zoller JE Lichter P Recurrent chromosomal imbalances detected in biopsy material from oral premalignant and malignant lesions by combined tissue microdissection, universal DNA amplification, and comparative genomic hybridization.Am J Pathol. 1998; 153: 295-303Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar, 20Kallioniemi OP Kallioniemi A Piper J Isola J Waldman FM Gray JW Pinkel D Optimizing comparative genomic hybridization for analysis of DNA sequence copy number changes in solid tumors.Genes Chromosom Cancer. 1994; 10: 231-243Crossref PubMed Scopus (944) Google Scholar, 21Lichter P Bentz M du Manoir S Joos S Comparative genomic hybridization.in: Verma RS Babu A Human Chromosomes. McGraw Hill, New York1995: 191-210Google Scholar The study cohort consists of patients from a hospital-based case control study. For control cases (polypoid adenomas, normal mucosa) patients have been selected randomly from the patient cohort. For comparison of average aberration frequencies in each entity, Fisher's exact test, binomial test, and pairwise comparison analysis were applied. CGH was used to identify DNA copy number changes that occur in nonpolypoid adenomas in the course of colorectal tumorigenesis. The histopathological data for the individual adenoma cases are summarized in Table 1. The summary of the CGH results including DNA gains and losses for each of the 82 samples investigated are presented in Table 2, Table 3. Chromosomal imbalances are additionally compared between nonpolypoid and polypoid adenomas in Figure 1 and examples are shown in Figure 2.Table 3Frequent Chromosomal Imbalances from CGH Analysis of Nonpolypoid and Polypoid NeoplasiasNonpolypoid colorectal neoplasiasPolypoid colorectal neoplasiasChromosomal changes*Only chromosomal imbalances occurring in more than two cases are shown.†Minimal common regions are given in brackets.LGDHGDCaLGDHGDCaGains 2q[2q32–33]3/19–2/9––– 3q[3q26.1–26.3]––3/9––– 4q[4q21–26]2/19–2/9––– 52/19–5/9––– 6[6q22–26]2/19–3/9––– 7[7q31]––7/9–2/10– 8q[8q13–24]2/19–3/9––– 12q[12q14–21]2/19–3/9––– 13[13q21–22]3/19–8/9–2/10– 20[20q12–13]––2/9––3/14Losses 2p[2p22–24]–––––2/14 3p[3p21]–––––2/14 4p––2/9––2/14 5[5q31qter]––2/9––2/14 7[7q32-qter]––2/9––– 8p[8p21–22]––3/9––2/14 8q[8q22]–––––2/14 10q[10q25]–––––4/14 12q[12q23-qter]––4/9––3/14 14[14q31-qter]––3/9––4/14 15q[15q24-qter]––3/9––4/14 16[16p13.1–13.3]6/192/33/92/182/104/14 16q[16q22–23]–––––2/14 17p[17p12–13]5/19–6/9––2/14 17q[17q21–24]–––––3/14 18[18q]3/19–6/9–2/108/14 20[20q]3/19––––– 22[22q]8/19–5/9–3/104/14–, No aberrations were detected.* Only chromosomal imbalances occurring in more than two cases are shown.† Minimal common regions are given in brackets. Open table in a new tab Figure 2CGH profiles of colorectal neoplasias showing exemplary gains on chromosomes 5q, 6p/q, 8q, 12, and 13q, and losses on chromosomes 16, 17p, 18q, and 22, which are typical changes in nonpolypoid neoplasms. Red and green lines represent thresholds for chromosome losses and gains, respectively. Chromosome numbers and numbers of chromosomes calculated are indicated below each profile.View Large Image Figure ViewerDownload Hi-res image Download (PPT) –, No aberrations were detected. No chromosomal aberrations were detected in this series of normal mucosa. The DNAs from normal specimens were amplified by PCR and labeled as test DNAs. The ratios for the sex chromosomes were consistent with the gender of the patients in all cases. Recurrent chromosomal aberrations (at least two alterations) were observed as DNA copy number losses on chromosomes 16, 17p, 18, 20, and 22 as well as gains of chromosomal regions on chromosomes 2q, 4q, 5q, 6, 8q, 12q, and 13q (Figure 1). These chromosomal imbalances were detected in 11 low-grade adenomas (11 of 19) as well as in 3 high-grade adenomas (3 of 3). Eighteen low-grade adenomas and 10 high-grade adenomas were investigated. Recurrent DNA losses were detected on chromosomes 16, 18, and 22 and gains were identified on chromosomes 7q and 13 (Figure 1). Chromosomal changes were preferentially detected in high-grade adenomas, whereas low-grade adenomas showed in single cases only copy number changes for whole chromosomes indicating aneuploidy. DNA losses were frequently detected on chromosomes 8p, 12q, 14, 15q, 16, 17p, 18, and 22 as well as DNA gains on chromosomes 3q, 5, 6, 7, 8q, 12q, and 13. Very complex aberrations were detected in some cases. A summary of aberration frequencies is shown in Table 2. Frequent DNA losses were detected on chromosomes 10q, 12q, 14q, 15q, 16p, 17q, 18, and 22 as well as DNA gain on chromosome 20. Also typical deletions on chromosomes 5q and 8p became apparent in two cases each. Some cases showed very complex aberrations similar than in carcinomas ex-nonpolypoid adenomas. Aberration frequencies are also demonstrated in Table 2. A comparison of frequent chromosomal alterations (≥2 alterations) between nonpolypoid and polypoid adenomas and carcinomas is summarized in Table 3. Statistical analysis revealed that cases with more than two aberrations are significantly more frequent for nonpolypoid adenomas (Fisher's exact test, P = 0.035). In contrast to polypoid adenomas, nonpolypoid adenomas exhibit frequent chromosomal aberrations already in low-grade lesions. A pairwise comparison test of single chromosomal aberrations revealed a significant higher frequency of 16p deletions in nonpolypoid adenomas compared to polypoid adenomas (Fisher's exact test, P = 0.0026). This aberration type is significantly more frequent than any other chromosomal aberration in nonpolypoid adenomas (P < 0.001, binominal test). There is an accumulation of chromosomal imbalances in the adenoma-carcinoma sequence for both entities. As demonstrated in Table 3 the aberration pattern for DNA losses is very similar between carcinomas ex-nonpolypoid adenomas and carcinomas ex-polypoid adenomas, however, differences exist for DNA gains. It is a generally accepted hypothesis that most of colorectal cancers develop from a pre-existing adenomatous polyp. Such benign lesions are mostly of polypoid origin, while a small proportion of adenomas are described as nonpolypoid lesions. The latter morphological category has therefore been proposed to exhibit specific genetic alterations. To confirm this assumption we have investigated 82 tissue specimens for chromosomal imbalances consisting of 22 nonpolypoid and 28 polypoid adenomas. Nine carcinomas ex-nonpolypoid adenomas, 14 polypoid carcinomas, and 9 specimens of normal epithelium were analyzed for control. This is the first comprehensive CGH study to detect chromosomal imbalances between nonpolypoid and polypoid adenomas of the colon. Our results show that recurrent chromosomal aberrations in nonpolypoid adenomas of the colon are found as DNA copy number losses on chromosomes 16, 17p, 18q, 20, and 22. Gains of chromosomal regions were observed on chromosomes 2q, 4q, 5q, 6, 8q, 12q, and 13q (Table 3). Carcinomas ex-nonpolypoid adenomas that represent intramucosal carcinomas are characterized by complex aberration patterns showing frequently chromosomal losses on 8p, 12q, 14, 15q, 16, 17p, 18, and 22, gains on 3q, 5, 6, 7, 8q, 12q, and 13 (Table 3). Although more amplifications were detected in carcinomas ex-nonpolypoid adenomas than in carcinomas ex-polypoid adenomas; many of the deletions detected occurred in both tumor types (eg, deletions on 17p and 18q). So far, there are no cytogenetic data on nonpolypoid adenomas of the colon. For comparison, polypoid adenomas were additionally investigated in this study. Our subset of polypoid adenomas exhibited frequent losses of whole chromosomes 16, 18, and 22 as well as gains of chromosomes 7q and 13, however, a lower frequency of chromosomal aberrations was detected compared to nonpolypoid adenomas. Moreover, loss of whole chromosomes is the most frequent finding indicating that aneuploidy is a very early event in polypoid adenomas, which is in agreement with an earlier study of Ried and colleagues22Ried T Knutzen R Steinbeck R Blegen H Schröck E Heselmeyer K du Manoir S Auer G Comparative genomic hybridization reveals a specific pattern of chromosomal gains and losses during the genesis of colorectal tumors.Genes Chromosom Cancer. 1996; 15: 234-245Crossref PubMed Scopus (344) Google Scholar reporting on crude aneuploidy in low- and high-grade polypoid adenomas. The most frequent affected chromosomal regions in nonpolypoid and polypoid adenomas from this study are summarized in Table 3 and Figure 1. Nonpolypoid adenomas show chromosomal changes very early in low-grade dysplasias, which is different to the observation in polypoid adenomas, although an accumulation of chromosomal imbalances became apparent in the adenoma-carcinoma sequence comparable to the situation in polypoid adenomas. Moreover, cases with more than two aberrations are significantly more frequent in nonpolypoid than in polypoid adenomas (P = 0.035, Fisher's exact test). This represents another indication for a different carcinogenic pathway in both lesions. More striking evidence for this hypothesis comes from a detailed analysis of single aberrations. It is obvious from Figure 1 that gains on chromosomes 2q, 5q, 6, 8q, and 12q occurred exclusively in nonpolypoid adenomas. In addition loss on chromosome 17p became apparent in very early lesions (low-grade dysplasia) of nonpolypoid adenomas, which cannot be confirmed for polypoid adenomas. It is of special interest that 16p deletions are significantly more frequent than any other aberration in nonpolypoid adenomas (8 of 36 aberrations, P < 0.001, binominal test) and that pairwise comparison using Fisher's exact test reveals also a significant difference for this aberration between nonpolypoid and polypoid adenomas (P = 0.0026). The successive molecular changes proposed to occur at different stages in the adenoma-carcinoma sequence were primarily based on DNA studies of polypoid adenomas. An accumulation of genetic changes in tumor suppressor genes and oncogenes such as adenomatous polyposis coli (APC), K-RAS, and TP53 has been reported in this type of colorectal carcinogenesis.23Ilyas M Straub J Tomlinson IPM Bodmer WF Genetic pathways in colorectal and other cancers.Eur J Cancer. 1999; 35: 1986-2002Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar, 24Maltzman T Knoll K Martinez ME Byers T Stevens BR Marshall JR Reid ME Einspahr J Hart N Bhattacheryya AK Kramer CB Sampliner R Alberts DS Ahnen DJ Ki-ras proto-oncogene mutations in sporadic colorectal adenomas: relationship to histologic and clinical characteristics.Gastroenterology. 2001; 121: 302-309Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar For nonpolypoid adenomas and carcinomas ex-nonpolypoid adenomas such genetic changes are widely unknown for the adenoma-carcinoma sequence. The low incidence of KRAS mutations reported for nonpolypoid colorectal adenomas10Hasegawa H Ueda M Watanabe M Teramoto T Mukai M Kitajima M K-ras gene mutations in early colorectal cancer: flat elevated vs poly-forming cancer.Oncogene. 1995; 10: 1413-1416PubMed Google Scholar, 11Yashiro M Carethers JM Laghi L Saito K Slezak P Jaramillo E Rubio C Koizumi K Hirakawa K Boland CR Genetic pathways in the evolution of morphologically distinct colorectal neoplasms.Cancer Res. 2001; 61: 2676-2683PubMed Google Scholar, 13Olschwang S Slezak P Roze M Jaramillo E Nakano H Koizumi K Rubio CA Larent Puig P Thomas G Somatically acquired genetic alterations in non-polypoid colorectal neoplasias.Int J Cancer. 1998; 77: 366-369Crossref PubMed Scopus (40) Google Scholar suggested a different genetic basis for the transformation process in these lesions. Also RAS mutations, a frequent genetic change in early colorectal tumor development of polypoid origin8Kinzler KW Vogelstein B Lessons from hereditary colorectal cancer.Cell. 1996; 87: 159-170Abstract Full Text Full Text PDF PubMed Scopus (4260) Google Scholar, 25Andreyev HJ Norman AR Cunningham D Oates J Dix BR Iacopetta BJ Young J Walsh T Ward R Hawkins N Beranek M Jandik P Benamouzig R Jullian E Laurent-Puig P Olschwang S Muller O Hoffmann I Rabes HM Zietz C Troungos C Valavanis C Yuen ST Ho JW Croke CT O'Donoghue DP Giaretti W Rapallo A Russo A Bazan V Tanaka M Omura K Azuma T Ohkusa T Fujimori T Ono Y Pauly M Faber C Glaesener R de Goeij AF Arends JW Andersen SN Lovig T Breivik J Gaudernack G Clausen OP De Angelis PD Meling GI Rognum TO Smith R Goh HS Font A Rosell R Sun XF Zhang H Benhattar J Losi L Lee JQ Wang ST Clarke PA Bell S Quirke P Bubb VJ Piris J Cruickshank NR Morton D Fox JC Al-Mulla F Lees N Hall CN Snary D Wilkinson K Dillon D Costa J Pricolo VE Finkelstein SD Thebo JS Senagore AJ Halter SA Wadler S Malik S Krtolica K Urosevic N Kirsten ras mutations in patients with colorectal cancer: the ‘RASCAL II’ study.Br J Cancer. 2001; 85: 692-696Crossref PubMed Scopus (734) Google Scholar have been reported to be not involved in the carcinogenesis of ex-nonpolypoid adenoma.26Fujimori T Satonaka K Yamamura-Idei Y Nagasako K Maeda S Non-involvement of ras mutations in non-polypoid colorectal adenomas and carcinomas.Int J Cancer. 1994; 57: 51-55Crossref PubMed Scopus (134) Google Scholar Similar conclusions about different avenues for colorectal cancer formation have been drawn from recent loss of heterozygosity studies reporting about differences in the KRAS locus and on microsatellite loci on chromosome 3p between polypoid and nonpolypoid lesions.27Smith G Carey FA Beattie J Wilkie MJ Lightfoot TJ Coxhead J Garner RC Steele RJ Wolf CR Mutations in APC, Kirsten-ras, and p53-alternative genetic pathways to colorectal cancer.Proc Natl Acad Sci USA. 2002; 99: 9433-9438Crossref PubMed Scopus (378) Google Scholar It became also apparent from a very recent study of Hermsen and colleagues28Hermsen M Postma C Baak J Weiss M Rapallo A Sciutto A Roemen G Arends J-W Williams R Giaretti W de Goeij A Mijer G Colorectal adenoma to carcinoma progression follows multiple pathways of chromosomal instability.Gastroenterology. 2002; 123: 1109-1119Abstract Full Text Full Text PDF PubMed Scopus (264) Google Scholar demonstrating the presence of varying combinations of few chromosomal abnormalities during adenoma-to-carcinoma progression, which indicate also the existence of multiple chromosomal instability pathways. As indicated in the same study28Hermsen M Postma C Baak J Weiss M Rapallo A Sciutto A Roemen G Arends J-W Williams R Giaretti W de Goeij A Mijer G Colorectal adenoma to carcinoma progression follows multiple pathways of chromosomal instability.Gastroenterology. 2002; 123: 1109-1119Abstract Full Text Full Text PDF PubMed Scopus (264) Google Scholar gains on chromosomes 8q23 and 13q21-22 and losses on chromosomes 8p21, 17p12-13, and 18q12-21 are associated with tumor progression present in advanced lesions (adenomas containing carcinoma foci or developed carcinomas). Interestingly, most of these alterations could be detected in our tissue samples preferentially in nonpolypoid adenomas showing low-grade dysplasia and in nonpolypoid intramucosal carcinomas (Table 3). No single polypoid adenoma with low-grade dysplasia in our subset of cases has shown chromosomal alterations in the critical loci indicated by Hermsen and colleagues.28Hermsen M Postma C Baak J Weiss M Rapallo A Sciutto A Roemen G Arends J-W Williams R Giaretti W de Goeij A Mijer G Colorectal adenoma to carcinoma progression follows multiple pathways of chromosomal instability.Gastroenterology. 2002; 123: 1109-1119Abstract Full Text Full Text PDF PubMed Scopus (264) Google Scholar In our cohort of low-grade dysplasias only nonpolypoid adenomas revealed chromosomal imbalances as mentioned above. Therefore, this phenomenon may reflect a higher malignant potential of nonpolypoid adenomas, which also show a higher extent of genetic instability than polypoid adenomas. Assuming a higher malignant potential for nonpolypoid adenomas one may also speculate that a genetic link between the specific growing pattern of nonpolypoid neoplasms and the malignant potential may exist. A puzzling finding in colorectal neoplasms is the overexpression of pRB and amplification of the RB1 gene,14Bartkova J Thullberg M Slezak P Jaramillo E Rubio C Thomassen LH Bartek J Aberrant expression of G1-phase cell cycle regulators in non-polypoid and exophytic adenomas of the human colon.Gastroenterology. 2001; 120: 1680-1688Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar which is paradoxical to the common inactivation of the RB gene in most types of human cancers.29Weinberg RA The retinoblastoma protein and cell cycle control.Cell. 1995; 81: 323-330Abstract Full Text PDF PubMed Scopus (4303) Google Scholar, 30Sherr CJ Cancer cell cycle.Science. 1996; 274: 1672-1677Crossref PubMed Scopus (4954) Google Scholar The finding of frequent 13q gain for nonpolypoid and polypoid adenomas in this study accounts for this puzzling situation and may contribute to support one of several hypotheses from elevated pRB being a secondary neutral change resulting from chromosome 13 gains, to a potential oncogenic role of pRB.14Bartkova J Thullberg M Slezak P Jaramillo E Rubio C Thomassen LH Bartek J Aberrant expression of G1-phase cell cycle regulators in non-polypoid and exophytic adenomas of the human colon.Gastroenterology. 2001; 120: 1680-1688Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 31Yamamoto H Soh J-W Monden T Klein MG Zhang LM Shirin H Arber N Tomita N Schieren I Stein CA Weinstein IB Paradoxical increase in retinoblastoma protein in colorectal carcinomas may protect cells from apoptosis.Clin Cancer Res. 1999; 5: 1805-1815PubMed Google Scholar, 32Wildrick DM Boman BM Does the human retinoblastoma gene have a role in colon cancer?.Mol Carcinog. 1994; 10: 1-7Crossref PubMed Scopus (20) Google Scholar The present study summarizes chromosomal changes detectable in nonpolypoid adenomas of the colon. It has been demonstrated that distinct differences to polypoid adenomas exist. This observation supports the assumption from earlier genetic and microsatellite studies10Hasegawa H Ueda M Watanabe M Teramoto T Mukai M Kitajima M K-ras gene mutations in early colorectal cancer: flat elevated vs poly-forming cancer.Oncogene. 1995; 10: 1413-1416PubMed Google Scholar, 11Yashiro M Carethers JM Laghi L Saito K Slezak P Jaramillo E Rubio C Koizumi K Hirakawa K Boland CR Genetic pathways in the evolution of morphologically distinct colorectal neoplasms.Cancer Res. 2001; 61: 2676-2683PubMed Google Scholar, 26Fujimori T Satonaka K Yamamura-Idei Y Nagasako K Maeda S Non-involvement of ras mutations in non-polypoid colorectal adenomas and carcinomas.Int J Cancer. 1994; 57: 51-55Crossref PubMed Scopus (134) Google Scholar that different genetic pathways for tumor progression may exist for colorectal neoplasias of polypoid and nonpolypoid phenotype. Moreover, this study provides a basis for search of new gene alterations in nonpolypoid adenomas starting at frequent chromosomal imbalances.
Referência(s)