Chromosomes 8, 12, and 17 Copy Number in Astler-Coller Stage C Colon Cancer in Relation to Proliferative Activity and DNA Ploidy1

Home , Chromosome 12

[CANCER RESEARCH 5.1.681-686, February 1. 1993] Chromosomes 8, 12, and 17 Copy Number in Astler-Coller Stage C Colon Cancer in Relation to Proliferative Activity and DNA Ploidy1

Melissa G. Steiner,2 Seth P. Harlow, Edoardo Colombo,3 and Kenneth D. Bauer4

Department of Pathology and the Cancer Center, Northwestern university, McGaw Medical Center, Chicago. Illinois 60611

ABSTRACT to glass slides; evaluation of the number of signals in each nucleus reflects the number of copies of that chromosome in that nucleus. Fluorescence in situ hybridization using centromere-specific DNA Most ISH studies have been applied to hematological cancers (22-24); probes to chromosomes 8,12, and 17 was applied to 23 archival paraffin- however, a few have addressed the chromosomal anomalies in bladder embedded stage C colonie cancer specimens. Chromosome copy number tumors (4, 25-27) and in primary breast tumors (28-30). Furthermore, was related to flow cytometric determinations of S-phase fraction and techniques of ISH can be applied to archival paraffin-embedded spec DNA ploidy. Three to eight copies of chromosomes 8, 12, and 17 were observed at mean frequencies of 28.7%, 37.8%, and 20.9%, respectively. imens (26, 30). The mean frequency of multiple copies of chromosome 12 was signifi Combining FCM and ISH on the same specimen is generally fea cantly greater than that for chromosome 17 (/' < 0.0025). The mean sible, because only thousands of cells are necessary to analyze the frequency of single copies of chromosome 17 was significantly greater than sample in a statistically Ã©valuablemanner using either method. FCM that for chromosomes 8 and 12 (/' < 0.0025 and /' < 0.0005, respectively). is a rapid and objective screening method for DNA content in malig Regarding the fourth quartile of cases, defined on the basis of the fre nancies and has been shown to provide prognostic information based quency of multiple chromosome copies, the proportion demonstrating on DNA ploidy (31-34) and proliferative activity (31, 34). Tumor moderate to high proliferative activity greatly exceeded the proportion proliferative activity, based on cell cycle phase distribution analysis, displaying low proliferative activity. The same cases (most chromosomally aberrant) also generally demonstrated DNA aneuploidy. The results indi has been shown to correlate with tumor aggressiveness in colon cancer cate a substantial degree of karyotypic instability in advanced colon can (35) and in other tumors (36, 37). Although FCM can provide useful cer, particularly in cases with high proliferative activity and DNA aneu information regarding DNA content, cytogenetics provides more spe ploidy. cific information regarding DNA abnormalities and better sensitivity for the determination of small variations in DNA content (4, 30). The INTRODUCTION few studies combining FCM and ISH have so far been limited to bladder (4, 25, 27) and breast cancer (28, 29). The combination of Although solid tumors comprise more than 90% of human cancers FCM and ISH has started to address the relationship between DNA (1,2), more than 80% of current cytogenetic data on human neoplasia ploidy and karyotypic abnormality; however, there is a paucity of are based on hematopoietic neoplasms ( 1). What is known about solid studies (29, 30) combining these techniques to investigate the rela tumors has often been obtained from highly advanced tumors which tionship between chromosomal anomalies and proliferative activity. are at a cytogenetically late stage; however, chromosomal anomalies In the present study, rhodamine-labeled DNA probes, recognizing have also been noted in diagnostic biopsies and in adenomas or the highly repetitive region of the centromeres of chromosomes 8, 12, precancerous lesions (1, 3). Considering the relatively high incidence and 17, were applied to 23 archival paraffin-embedded colonie can of colorectal cancers, both benign and malignant, the number of cases cer specimens. The tumors analyzed were of a homogeneous stage studied cytogenetically is surprisingly small. (Astler-Coller Dukes stage C), in an attempt to define the frequency of Recurrent abnormalities that have been identified in colorectal can abnormalities in cancers of comparatively similar age. Chromosome cer include several chromosomes (1-11). The best documented chro copy number data were related to tumor proliferative activity (S-phase mosomal aberrations in spontaneous colonie tumors are deletions of fraction) and DNA ploidy to address possible interrelationships be 17p and 18q, which have been shown to be late stage events (7-9, 12). tween these biological features. Chromosomal anomalies may reflect oncogenic changes ( 13-21 ). Fur thermore, aberrations in chromosome copy number reflect karyotypic instability and may relate to tumor aggressiveness and progression MATERIALS AND METHODS (4,9, 17,21). Interphase cytogenetics, the specific application of ISH5 techniques Pathologic Material. Pathologic specimens for this study were obtained from patients who underwent surgery at Northwestern Memorial Hospital from to interphase cells or nuclei, allows for the investigation of numerical 1978 to 1985. Archival paraffin blocks were analyzed from 23 patients who abnormalities in target chromosomes without requiring the culture of had resections for primary untreated adenocarcinoma of the colon that, on neoplastic cells or preparation of metaphase spreads. Fluorescence pathological staging, were all classified as Astler-Coller stage C (38). Corre labeled DNA probes can be applied to interphase cells (FISH) affixed sponding slides from all specimens were reviewed by a pathologist, and careful diagrams of the location of tumor cells were made so that areas containing the highest yield of neoplastic cells without debris or nonneoplastic cells could be Received 7/27/92; accepted 11/19/92. dissected free when sectioning the blocks. Paraffin-embedded tissues used in The costs of publication of this article were defrayed in part by the payment of page this investigation had been fixed in 10% neutral buffered formalin and pro charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. cessed by standard histological embedding techniques. 1Supported in part by Grant PDT-285 awarded by the American Cancer Society Preparation of Nuclear Suspensions. Paraffin blocks were sectioned into (National Office). 50-nm sections, and areas of representative tumor cells were hand dissected to 2 Present address: Genentech, Inc., 460 Point San Bruno Blvd., South San Francisco, obtain an optimal sample. The tissue was then deparaffinized and dissociated CA 94080. 3 Present address: Department of Pathology, Valduce Hospital, 22100 Como, Italy. according to the literature (36, 39). Briefly, the tumor slices were washed for 4 Present address: Genentech, Inc., 460 Point San Bruno Blvd., South San Francisco, IO min each in successive solutions of xylene, xylene, 100% ethanol, 100% CA 94080. To whom requests for reprints should be addressed. ethanol, 90% ethanol, 70% ethanol, and 50% ethanol and then placed in 5 The abbreviations used are: ISH, in situ hybridization; FCM, flow cytometry; FISH, fluorescence in situ hybridization; HBSS, Hanks' balanced salt solution; HEPES, 4-(2- distilled water. After at least 24 h, the tissue was washed in distilled water, hydroxyethyl)-l-piperazineelhanesulfonic acid; DAPI, 4',6'-diamidino-2-phenylindole finely minced with dissecting scissors, suspended in 0.5% pepsin solution dihydrochloride; LQ, first quartile; HQ, fourth quartile. (Sigma, St. Louis, MO) (in 0.9% NaCl and adjusted to pH 1.5 with HC1) and 681

incubated at 37Â°C.The suspension was vigorously mixed with a Vortex mixer histograms had a background of cellular debris contributing to the DNA at 5-min intervals for 30 min or until the solution became turbid. The digested histogram. This was subtracted using a "sliced nuclei" curve fitted to the material was then filtered through 37-um monofilament mesh (Tetko, background to the left of the G0G, peak and to the right of the G2M peak (41). Briarcliff, NY). HBSS-HEPES was added to the filtrate, which was then Tumor proliferative activity was expressed in terms of the calculated percent centrifuged at 4Â°Cand 1000 rpm for 10 min; the pellet was resuspended in age of cells in S-phase. The S-phase data were divided into three levels based HBSS-HEPES and the nuclei were counted. on a statistical definition: Preparation of Slides for ISH. The resulting nuclear suspensions were Mean % S-phase Â±0.5 SD = moderate proliferative activity. fixed in methanohacetic acid (3:1) for 20 min: at this point the suspensions could be stored at -20Â°C or centrifuged at 4Â°Cand 1000 rpm for 10 min and An S-phase fraction below this range (i.e., <13.7%) was defined as low then washed 2-3 times in the fixative before finally being resuspended at 1-2 x IO6 nuclei/ml. The fixed nuclei were dropped onto poly-Mysine (Sigma)- proliferative activity, whereas S-phase values greater than this range (i.e., >21%) were defined as high proliferative activity. -coated slides, already wet with fixative and warmed over a humid bath. Two Normal colonie epithelium from the same block as the tumor was used as a to three rounds of dropping and brief drying were carried out to achieve control to determine which DNA ploidy peak represented the diploid G0G, sufficient nuclei per slide. The slides were stored at least 12 h in slide boxes in sealed plastic bags at -20Â°C, but they may be stored for several weeks in this population. An aneuploid population was considered to be present only when two or more distinct G0G, populations were present. The DNA index was manner. calculated by dividing the mean fluorescence channel of the aneuploid G0G, Prehybridization Steps. Using a diamond-tipped scribe, the areas for hy peak by the mean channel number of the diploid G()G] peak. bridization were marked on each slide. The slides were subjected to a series of Data Analysis. The chromosome copy number data were stratified to in prehybridization incubations as follows: 20 min of 0.2 s HC1; phosphate- buffered saline rinse; 0-10 min of 80Â°Csodium thiocyanate (l M in water; dicate relative aberrancy based on the statistical definition of quartiles. LQ is Sigma); 5-min water rinse (twice); 0-10 min of 37Â°Cproteinase K (50 ug/ml defined as the lowest 25% of the cases based on the frequency of either multiple copies or single copies of the chromosomes; similarly, HQ is defined in phosphate-buffered saline; Sigma); 30 s of 5 mw EDTA; phosphate-buffered as the uppermost 25% of the cases based on the frequency of multiple or single saline rinse; and 2 min each of 70%, 85% and 100% ethanol. copies. For the purpose of relating chromosomal aberrancies to biological Variable durations of exposure to sodium thiocyanate and proteinase K were features of advanced malignancies, we selected the cases falling in the first and performed because these steps often proved too harsh for the archival material. fourth quartiles with respect to the frequency of either multiple copies or single As a standard, a 5-min exposure to each compound in the described order was copies of the chromosomes specified (LQ5ing|C,LQmul,,pie,HQ5ingie,HQmU|tiple). performed; the duration and combination of the two steps were varied to achieve good hybridization efficiency (>85% nuclei having Ã©valuableISH signals) while maintaining acceptable nuclear morphology (26). RESULTS Following the ethanol dehydration, the slides were air dried, and 10 ul of probe mix (containing rhodamine-labeled DNA probes, recognizing the highly Fig. 1 provides an example of FISH using a centromere-specific chromosome 12 probe applied to a deparaffmized colonie adenocar- repetitive region of the centromeres of chromosomes 8, 12, and 17; provided by Imagenetics, Naperville, IL) were applied to each hybridization area. The cinoma specimen. Multiple copies of chromosome 12 and heteroge slides were coverslipped and placed in a tightly closed plastic box, which was neity with respect to copy number are evident. Fig. 2 illustrates floated in a 76Â°Cwater bath to denature the target and probe DNA together. frequency distributions of signals per nucleus in two primary colonie After 10 min, the box was removed, a damp blotter was added, and the resealed adenocarcinomas. One case (Fig. 2A) illustrates the prevalence of 2 box was placed in a 42Â°Cincubator overnight. signals/nucleus; also note that very few nuclei are unhybridized. The Posthybridization Washes. Subsequent to the overnight hybridization, other case (Fig. 25) demonstrates extreme chromosomal anomalies; coverslips were removed, and the slides were washed for 10 min each in a series of 45Â°Cbaths to remove any unhybridized probe; the washes were as up to six copies of chromosome 8 are noted per nucleus, and nearly 75% of the nuclei have 3 or 4 copies of chromosome 12. While the follows: 3 successive 50% formamide washes; one sodium citrate solution (300 frequency distributions from individual tumors often demonstrate the mm sodium chloride, 30 ITIMsodium citrate; pH 7.0) wash; and one wash in 0.1% Nonidet P-40 in standard sodium citrate solution. Following these unique abnormalities in copy number of a particular chromosome, the washes, 7 ul of counterstain (1-2 ug/ml DAPI; Boehringer Mannheim, India napolis, IN) in antifade solution (p-phenylenediamine dihydrochloride, 1 mg/ml in glycerol, pH 8; Sigma) were applied, and the slides were carefully coverslipped. The slides were stored in dark slide boxes at -20Â°C until eval uation. Evaluation of ISH Signals. To visualize the direct-labeled rhodamine probe simultaneously with the DAPI counterstain, we used a DAPI/F4 filter setup (430ejl/468em and 532â€ž/625cmwith a dual-band dichroic filter: Omega Optical, Brattleboro, VT) on a Leitz Laborlux D model fluorescent microscope, equipped with a water immersion X100 objective. Two hundred nuclei were evaluated, and the nuclei were classified on the basis of the number of signals observed per nucleus. The evaluation criteria follow those detailed previously (25). Briefly, 85% of nuclei are required to have some ISH signal; otherwise, the hybridization is repeated. Evaluable nuclei do not overlap and are not asymmetrically covered by cytoplasm, and only ISH signals of similar fluo rescence intensity and size are counted, omitting minor signals; occasionally, split signals were observed and were counted as one, provided it was clear that the signals were derived from one (barely a perceptible distance between them). Cell Cycle and DNA Ploidy Analysis. Flow cytometry results presented in this study were from cases reported in a previous study (35) from this labo ratory which examined the prognostic significance of DNA ploidy and proliferative activity in Dukes C colonie cancer. Cell cycle analysis was performed Fig. 1. Example of FISH on a paraffin-embedded specimen of colon cancer using a on an IBM PS2 model 80 computer (IBM Corp., Boca Raton, FL). Multicycle rhodamine-labeled centromere-specific probe to chromosome 12. The nuclei are counter- software (Phoenix Flow Systems, San Diego, Ã‡A),a program using the math stained with DAPI. Chromosome heterogeneity is evident based on 2, 3, and 4 centromeric ematical method of Dean and Jett (40), was used for cell cycle analysis. All copies of chromosome 12, respectively, in the three nuclei in the field. 682

1UU908070605040- copies are plotted with respect to the specific chromosomes (Fig. 3A), it is apparent that, on average, aberrancy is most frequent for chro mosome 12 compared to the other chromosomes studied. Three to eight copies of chromosomes 8, 12, and 17 were observed at mean frequencies of 28.7%, 37.8%, and 20.9%, respectively (Table 1).

Table 1 Frequency of chromosomal aberrations in 23 cases of colon cancer" 30S 20E- 828.7 1237.8Â±4.2fe12.6Â±2.8<'62.21720.9

10 K3 NucleicopiesOverall having >2 ^â€¢, & *""(J<1-----1r.TL_- nA---â€¢..â€¢ .T\3 mean(Â± Â±4.27.8 Â±2.95.7 SEM)First 1 WWu quartilemean(Â± Â±1.454.0 Â±0.840.2 SEM)Fourth 901 Â±3.8*9.9 8070605040302010gBâ€¢â€¢â€¢fiâ€¢.-i1, quartilemean(Â± Â±8.610.3 Â±4.215.7 SEM)Nuclei

copiesOverallhaving single mean(Â± Â±1.14.7 Â±0.85.7 Â±1.3rf8.2 SEM)First quartilemean(Â± Â±0.517.2 Â±0.814.9 Â±0.8'22.0 SEM)Fourth quartilemean(Â± Â±1.2Chromosome Â±1.2Chromosome Â±\.lf SEM)Chromosome -ICHROMOSOME8 a Chromosome copy number was determined by FISH using centromere-specific Jffi. probes to the specified chromosomes; 200 nuclei were evaluated per specimen. Kii Â¿--..i h Significantly greater than chromosome 17 (P < 0.0025. Student's / test). c Significantly greater than chromosome 17 (P < 0.05). 12 17 ''Significantly greater than chromosome 8 (/" < 0.0025) and chromosome 12 Fig. 2. Frequency distributions of signals/nucleus in colon cancer case 1 (A ) and case (P < 0.0005). 8 (B) using centromere-specific DNAprobes to three chromosomes. Notice the prevalence ' Significantly greater than chromosome 8 (P < 0.005) and chromosome 12 (P < 0.05). of 2 signals/nucleus for all three chromosomes in A in contrast to the multiple copies of ^Significantly greater than chromosome 8 (P < 0.01) and chromosome 12 (P < 0.001). chromosome 8 and predominance of 3 and 4 copies of chromosome 12 per nucleus in B. â€¢¿,0signals; O 1 signal; D, 2 signals; Â£5.3 signals; B. 4 signals; B. 5 signals; 0. 6 signals. UU90807060504030 8

UU908070605040302010A- ,19Â§0:

Â§0.m: o ! 8:-Ã›-- 18,17.7.12.15Qy23.6.3.4.2.1CHROMOSOME _Â¡hâ€”O,-^ZLuUerLiJCL1 2010QCHROMOSOME1â€¢ ~B"HHP5 so908070605040302010o16..1411.108,21.13.22.20.5 â€¢¿ri 8 i S1009080706050403020O 12OsB-iO1

Jh-O>â€”

ZLUOKLu0.1 21.1920.2.11.17.1018,15.79.51.3.4CHROMOSOME

10Bâ€¢â€¢â€¢â€¢â€¢O:-i 4- 1:CHROMOSOME8 nnlUU9080706050403020100823.12.1613.6,22.14 17'

12 17 Fig. 3. Frequency of abnormal chromosome copy number with respect to chromosomes 8, 12, and 17 in 23 specimens of colon cancer. A. nuclei having more than 2 copies of the 8oiio..12191 chromosomes per nucleus; B, nuclei having single copies of the chromosomes per nucleus. The dashed line marks the mean observed frequency for each chromosome. Significant 48.22.20.13.11.1516.7.5.17.18.9 differences in copy number were noted (*) for chromosome 12 versus chromosome 17 (A, P < 0.0025; Student's r test) and for chromosome 17 versus chromosomes 8 and 12 (B, P < 0.0025 and P < 0.0005, respectively). 21,23.6,310.1.2.4 average frequency distribution of all 23 cases shows 2 copies/nucleus as the most common observation for each chromosome (data not Fig. 4. The frequency of nuclei having greater than 2 copies/nucleus for chromosome shown). 8 (top), chromosome 12 (middle), and chromosome 17 (bottom) with case numbers assigned. This figure illustrates the heterogeneity of any particular case with respect to the When the data for the percentage of nuclei having more than 2 different chromosomes. 683

Table 2 Colonie adenocarcinoma: chromosome copy number in relation to tumor chromosomes, the mean S-phase fraction for cases in HQmii|tipie is proliferante activity" greater than for those cases falling in LQmU|tip,e.However, those data Mean SPF Â±SEM did not achieve statistical significance. When analyzing only the most 815.2 1214.4 1716.5 aberrant cases (HQmultiple),this association remained evident. Specif NucleibLQHQNuclei with multiple copies ically, the proportion of cases demonstrating moderate to high prolif Â±2.918.9 Â±1.818.9 Â±3.518.6 erative activity greatly exceeded those cases displaying low prolifer Â±2.319.6 Â±2.116.7 Â±2.418.5 ative activity (5 of 6 versus 1 of 6, 5 of 6 versus 1 of 6, 6 of 6 versus copiesLQHQChromosomewith single 0 of 6) for chromosomes 8, 12, and 17, respectively (data not shown). Â±2.514.9 Â±1.513.8 Â±2.613.5 Â±3.1Chromosome Â±2.5Chromosome Â±2.4 There appears to be a positive correlation between higher proliferative " Six cases constituting either LQ or HQ of the colonie cancer cases, denned by the activity and increased frequency of additional chromosome copies. frequency of multiple copies or single copies of chromosomes 8, 12, and 17. Tumor DNA ploidy findings for the cases having additional copies of the proliferative activity (S-phase fraction) was calculated by DNA flow cytometry and cell cycle analysis (see "Materials and Methods" for details). chromosomes are shown in Table 3. Regarding only the most aberrant Chromosome copy number was determined by FISH using centromere-specific cases (HQmultiple), the proportion of cases having DNA aneuploid probes to the specified chromosomes. tumors is higher than those cases with DNA diploid tumors (5 of 6 versus 1 of 6, 4 of 6 versus 2 of 6, 6 of 6 versus 0 of 6) for Regarding these observations, the mean frequency of additional cop chromosomes 8, 12, and 17, respectively; statistically significant dif ies of chromosome 8 is not significantly different from that for chro ferences, however, were noted only for chromosome 17 (P < 0.025; mosomes 12 and 17, but the mean frequency of additional copies of X2). There appears to be a correlation between DNA aneuploidy and chromosome 12 is significantly higher than that for chromosome 17 (P < 0.0025; Student's / test). This difference is maintained when increased frequency of additional copies of the chromosomes. This trend is not limited to HQmuWple,since it also holds true for chromo examining only those cases with either a low or a high frequency somes 8 and 17 for cases in LQmulliple. of multiple copies of the chromosomes (first and fourth quartiles; Similar comparisons were made with the cases having single copies Table 1). of chromosomes 8, 12, and 17 (Tables 2 and 3). For the three chro In addition, cases having a high proportion of nuclei with more than mosomes, the mean S-phase fraction for the most aberrant cases 2 signals do not necessarily tend to demonstrate similar abnormalities (HQsingie) is 'ess than for those cases falling in LQsÂ¡ngle.Nostatisti for all 3 chromosomes (Fig. 4). For example, case 19 has a high cally significant differences were found. For the most aberrant cases frequency of multiple copies for chromosomes 8 and 17 but only (HQsingie). the proportion of cases having moderate to high prolifer moderate abnormality for chromosome 12; similarly, case 8 shows a ative activity is higher than those cases having low proliferative ac high frequency of additional copies of chromosome 12 but only mod tivity (4 of 6 versus 2 of 6) for chromosomes 8 and 12 and is erate abnormality for chromosomes 8 and 17. On the other hand, case equivalent for chromosome 17 (data not shown). There appears to be 14 is among the 5 cases with the greatest frequency of additional a stronger, although not statistically significant, trend between mod copies for all 3 chromosomes. Thus, considerable chromosome het erogeneity is evident. erate to high proliferative activity and the least aberrant (LQsingle) for When the percentage of nuclei having a single copy of chromo all three chromosomes (5 of 6 versus 1 of 6). DNA aneuploid tumors are more prevalent than diploid/near-diploid tumors, regarding the somes 8, 12, and 17 is plotted for individual cases (Fig. 3B), it is apparent that chromosome 17 is most frequently involved. The mean lower frequency (LQsingle) of single copies for all three chromosomes frequency of nuclei having a single copy is 10.3%, 9.9%, and 15.7% (4 of 6 versus 1 of 6). These data, however, did not achieve statistical for chromosomes 8, 12, and 17, respectively (Table 1). There was no significance. For the most aberrant cases (HQsjngle), the proportion of significant difference for the mean frequencies for chromosomes 8 cases having DNA aneuploid tumors is higher than for those cases and 12, but the mean frequency of single copies for chromosome 17 having DNA diploid tumors for chromosomes 8 and 17 (5 of 6 versus is significantly higher than that for chromosome 8 (P < 0.0025; 1 of 6) and is Ã©quivalentfor chromosome 12. Student's / test) and for chromosome 12 (P < 0.0005; Student's / test).

These differences are maintained when the data are subclassified into DISCUSSION relative levels of aberrancy (first and fourth quartiles; Table 1). Proliferative activity data for the cases having multiple copies of The results of this study demonstrate anomalies in copy number of chromosomes 8, 12, and 17 are displayed in Table 2. For the three chromosomes 8, 12, and 17 and their relationship to proliferative

Table 3 Colonie adenocarcinoma: relationship between chromosome copy number and DNA ploidy0

Chromosome 8 Chromosome 12 Chromosome 17

NucleibDiploid/near-diploidAneuploidNuclei with multiple copies c(67%)2/6 c(33%)4/6 c(33%)4/6''(67%)2/6 d(0%)6/6

r(67%)2/6 c(83%)1/6'(17%)5/6 c(33%)2/6c(33%)4/6 c(67%)3/6 "(100%)1/6

copiesDiploid/near-diploidAneuploidLQ2/6c(33%)4/6with single c(33%)4/6 c(50%)3/6 c(33%)4/6 c(17%)5/6

r(67%)HQ1/6C(17%)5/6 f(83%)LQ4/6 c(67%)HQ2/6 c(50%)LQ2/6 c(67%)HQ0/6 c(83%)

" Six cases constituting either LQ or HQ of colonie cancer cases, defined on the basis of frequency of multiple copies or single copies of chromosomes 8, 12, and 17. DNA ploidy measurements were determined by FCM (see "Materials and Methods" for details); diploid/near-diploid is defined as a DNA index of 1.0-1.2, and DNA aneuploid is defined as a DNA index of >1.2. Values in parentheses are the percentages of cases falling in the given categories. h Chromosome copy number was determined by FISH using centromere-specific probes to the specified chromosomes. c No significant difference from expected (P < 0.50; x2). d Significantly different from expected (P < 0.025; *2).

684

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1993 American Association for Cancer Research. CHROMOSOME ABERRATIONS RELATE TO S-PHASE AND DNA PLOIDY activity and DNA ploidy in Astler-Coller Dukes stage C adenocarci- abnormalities in the present investigation, we cannot comment on the nomas of the colon. Our data show the most frequent chromosomal presence of other regions on these chromosomes which encode genes anomalies in these 23 cases to be additional copies of chromosomes 8 regulating or influencing tumor growth. Telomeric or specific loci and 12 and frequent losses of chromosome 17. In fact, chromosome 12 probes of sufficient size would certainly clarify this point, but such was found to have a significantly increased frequency of additional probes were not available to us at the time of the study. copies per nucleus compared to chromosome 17, and chromosome 17 In conclusion, this study demonstrates a high degree of aberrancy was found to have a significantly increased frequency of a single copy regarding copy number of chromosomes 8, 12, and 17 in colon cancer. per nucleus compared to the other chromosomes studied. The integration of FISH with DNA flow cytometry proved to be useful Nearly 30% of all scored nuclei in this investigation had additional in terms of directly evaluating the relationship between tumor prolif copies of chromosome 8 and nearly 40% for chromosome 12; trisomy erative activity and numerical chromosomal abnormality. Despite the was the predominant observation for those nuclei, similar to published modest sample size, which is likely to be responsible for the lack of reports on cytogenetic aberrations in colon cancer (1, 3, 11-13). While statistical significance, the trends are clear and suggest a relationship the present study demonstrated chromosome 8 gains similar to those between aberrancy in chromosome copy number and DNA ploidy and in other investigations, they were not statistically significant with between proliferative activity and abnormal chromosome copy num respect to the other chromosomes studied. Perhaps one explanation for ber. A larger study involving cases with both early and advanced the lack of significance regarding chromosome 8 is the advanced disease appears necessary to confirm these findings as well as the degree of tumor progression. Chromosome 8 gains reflect earlier possible adjunctive prognostic value of chromosome copy number in stages, being one of the most frequent observations in adenomas of the colorectal cancer. colon (1, 3). All the specimens in this study, however, were lymph node metastatic adenocarcinomas. One might expect to see more ACKNOWLEDGMENTS involvement of chromosomes 12 and 17, as anomalies in these chro mosomes have been reported to be associated with more advanced We gratefully acknowledge the generous gift of the chromosome probes from Imagenetics and specifically thank Dr. Mike Bittner, Dr. Doug Taron. and disease (1, 3). Lucy Stohls at Imagenetics for their technical advice and initial training with Chromosomal gains, even as high as 10 copies, have also been the chromosome-specific probes. We also thank Kevin Becker of Phoenix Flow reported by FISH analyses in other solid tumors (25, 26, 28). Because Systems for providing the Multicycle Cell Cycle analysis software used in this such observations are common and because duplicate preparations of investigation and Joan S. Chmiel at Northwestern University for her helpful several cases in our laboratory revealed reproducible frequencies of comments on the use of appropriate statistics. chromosome copy number, we are confident that the multiple copies observed in the present study were not artifactual. Nonetheless, it REFERENCES should be noted that the DNA probes utilized in this study recognize 1. Sandberg, A. A. The Chromosomes in Human Cancer and Leukemia, Eid. 2. New only the centromeric region of the chromosomes, rather than the entire York: Elsevier Science Publishing Co.. 1990. chromosome. 2. Cancer Facts and Figures. Allanta: American Cancer Society, 1991. 3. Heim, S., and Mitelman, F. Cancer Cytogenetics. New York: Alan R. Liss. 1987. Reports on allelic losses and structural rearrangements of chromo 4. Hopman, A. H. N., Moesker, O., Smeets, A. W. G. B., Pauwels, R. P. E.. Vooijs, G. some 17 in colon cancer are prevalent in the literature (7, 8, 12, 13, 17, P.. and Ramaekers, F. C. S. Numerical chromosome I. 7. 9, and II aberrations in 21). The structural rearrangements often lead to deletions of the short bladder cancer detected by in situ hybridization. Cancer Res., 51: 644-651. 1991. 5. Mitelman, F. Catalog of Chromosome Aberrations in Cancer. Ed. 4. New York: arm or the total chromosome. In the present study there is statistical Wiley-Liss, 1991. support for an increased frequency of a single copy of chromosome 17 6. Reichmann. A., Martin. P., and Levin, B., Chromosomes in human large bowel tumors: a study of chromosome #1. Cancer Genet. Cytogenet.. 12: 295-301, 1984. per nucleus, suggesting a true deletion ofthat chromosome. Additional 7. Vogelstein. B.. Fearon. E. R., Hamilton. S. R.. Kern. S. E., Preisinger, A. C.. Leppert. factors contributing to an observation of a single copy are poor probe M., Nakamura. Y., White, R.. Smils. A. M. M.. and Bos, J.L. Genetic alterations penetration, loss of DNA, incomplete denaturation of target DNA, and during colorectal-tumor development. N. Engl. J. Med.. 319: 525-532. 1988. 8. Fearon. E. R.. and Vogelstein. B. A genetic model for colorectal tumorigenesis. Cell. the inability to resolve overlapping signals using 2-dimensional anal 61: 759-767. 1990. ysis. We believe most of these contributing factors were minimized by 9. Kern, S. E., Fearon. E. R.. Tersmette. K. W. F., Enterline. J. P., Leppert. M., Naka mura, Y, White, R., Vogelstein, B., and Hamilton, S. R. Allelic loss in colorectal the experimental design, specifically requiring greater than 85% hy carcinoma. JAMA. 261: 3099-3103, 1989. bridization efficiency, maintenance of nuclear morphology, and ade 10. Becker, R., Gibas. Z., and Sandberg. A. A. Involvement of chromosomes 7 and 12 in quate section thickness for the preparation of nuclear suspensions (26, large bowel cancer: trisomy 7 and 12q-. Cancer Genet. Cytogenet.. 9: 329-332, 1983. 11. Ochi. H., Takeuchi, J., Holyoke, D., and Sandberg, A. A. Possible specific chromo 42). Therefore, the increased frequency of single copies of chromo some changes in large bowel cancer. Cancer Genet. Cytogenet., 10: 121-122, 1983. some 17 is considered to be due to true deletions ofthat chromosome. 12. Muleris, M., Salmon. R. J.. Zafrani. B.. Girodet. J.. and Dutrillaux. B. Consistent Tumor cells having an increased number of genetic aberrations may deficiencies of chromosome 18 and the short arm of chromosome 17 in eleven cases of human large bowel cancer: a possible recessive determinism. Ann. Genet., 28: have some growth advantage (43) and contribute to malignant pro 206-213. 1985. gression (9, 21). As the cells grow and divide, the probability of 13. Konstantinova, L. N., Fleischman. E. W.. Knisch, V. I., Perevozchikov, A. G., and Kopnin, B. P. Karyotype peculiarities of human colorectal adenocarcinoma. Hum. genetic changes increases and the existing changes are propagated. Genet., 86: 491^Â»96, 1991. This concept directly supports the observed overall tendency for high 14. Bos. J. L., Fearon, E. R.. Hamilton, S. R., deVries, M. V., van Boom, J. H.. van der proliferative activity and DNA aneuploidy to be correlated with in Eb, A. J., and Vogelslein. B. Prevalence of rax gene mutations in human colorectal cancers. Nature (Lond.). 327: 293-297, 1987. creased frequencies of chromosomal abnormalities. Previous cytoge 15. Forrester. K., Almoguera, C., Han. K., Grizzle, W. E.. and Perucho, M. Detection of netic studies of bladder cancer (27) and breast cancer (29) have also high incidence of K-ro.s oncogenes during human colon tumorigenesis. Nature demonstrated a positive correlation between chromosomal aberrations (Lond.), 327: 298-303, 1987. 16. Gulman, M.. Ravia, Y, Assaf, D., Yamamoto. T., Rozin. R.. and Shiloh, Y. Amplifi and proliferative activity and DNA ploidy. cation of c-myc and c-erbB-2 proto-oncogene in human solid tumors: frequency and The inverse relationship between frequency of single copies and clinical significance. Int. J. Cancer, 44: 802-805, 1990. proliferative activity suggests that the presence of single copies of the 17. Baker. S. J.. Fearon. E. R.. Nigro. J. M.. Hamilton. S. R.. Preisinger. A. C.. Jessup, J. M.. van Tuinen. P., Ledbetter, D. H.. Barker. D. F.. Nakamura. Y. White. R.. and chromosomes did not confer any growth advantage on the tumors, Vogelstein, B. Chromosome 17 deletions and p53 gene mutations in coloreclal car although it should be recognized that all cases presented in this study cinomas. Science (Washington DC), 244: 217-221, 1989. 18. Levine, A. J., Momand, J., and Finlay, C. A., The p53 tumour suppressor gene. Nature were advanced (lymph node metastatic) disease. Because centromeric (Lond.), 351: 453-456, 1991. probes to these chromosomes were utilized to define chromosomal 19. Hollstein, M., Sidransky. D.. Vogelstein, B., and Harris. C. C. p53 mutations in human 685

cancers. Science (Washington DC), 253: 49-53, 1991. Gray, J., and Waldman, F. M. Deletion of chromosome 17p loci in breast cancer cells 20. Rodrigues, N. R., Rowan. A., Smith, M. E. F., Kerr, I. B., Bodmer, W. F.. Gannon, J. detected by fluorescence in situ hybridization. Cancer Res.. 52: 3474-3477. 1992. V, and Lane, D. P. p53 mutations in colorÃ©ela!cancers. Proc. Nat. Acad. Sci. USA, 31. Bauer, K. D., Lincoln. S. T., Vera-Roman, J. M.. Wallemark, C. B., Chmiel, J. S., 87: 7555-7559. 1990. Madurski. M. L., Murad, T. M., and Scarpelli. D. G. Prognostic implications of 21. Baker, S. J.. Preisinger, A. C. Jessup, J. M. Paraskeva, C. Markowitz, S.. Willson, J. proliferative activity and DNA aneuploidy in colonie adenocarcinomas. Lab. Invest., K. V., Hamilton, S., and Vogelstein, B. p53 gene mutations occur in combination with 57: 329-335, 1987. I7p allelic deletions as late events in colorectal tumorigenesis. Cancer Res., 50: 32. Kallioniemi, O. P., Blanco. G.. Alavaikko. M.. Hietamen. T., Manilla, J.. Lanslahti, K.. 7717-7722, 1990. and Koivula, T. Tumor DNA ploidy as an independent prognostic factor in breast 22. Poddighe, P. J., Moesker, O.. Smeets. D., Awwad. B. H., Ramaekers, F. C. S.. and cancer. Br. J. Cancer, 56: 637-642. 1987. Hopman, A. H. N. Interphase cytogenetics of hematological cancer: comparison of 33. Frankfurt, O. S.. Arbuck. S. G., Greco. W. R.. Slocum. H. H., Pavelic. Z. P.. Lawrence, classical karyotyping and in situ hybridization using a panel of eleven chromosome D. D., Petrelli, N.. Mittleman, A., and Rustum, Y. M. Prognostic significance of DNA specific DNA probes. Cancer Res., 51: 1959-1967, 1991. aneuploidy in human colorectal cancer. Proc. Am. Soc. Clin. Oncol., 3: 17. 1984. 23. Anastasi, J., LeBeau, M. M., Vardiman. J. W., and Westbrook, C. A. Detection of 34. Tribukait, B., Gustafson. H.. and Esposti, P. L. The significance of ploidy and numerical chromosomal abnormalities in neoplastic hematopoietic cells by in situ proliferation in the clinical and biological evaluation of bladder tumors: a study of 100 hybridization with a chromosome-specific probe. Am. J. Pathol.. 136: 131-139, 1990. untreated cases. Br. J. Urol., 54: 130-135, 1982. 24. Anastasi, J., Thangavelu, M., Vardiman, J. W., Hooberman, A. L., Bian, M. L.. Larson, 35. Harlow, S. P., Erikson, B. L., Poggensee. L., Chmiel, J. S., Scarpelli, D. G., Murad. R. A., and LeBeau, M. M. Interphase cytogenetic analysis detects minima] residual T.. and Bauer, K. D. Prognostic implications of proliferative activity and DNA disease in a case of acute lymphoblastic leukemia and resolves the question of origin aneuploidy in Astler-Coller Dukes stage C colonie adenocarcinoma. Cancer Res., 5/: of relapse after allogeneic bone marrow transplantation. Blood, 77: 1087-1091. 1991. 2403-2409, 1991. 25. Hopman, A. H. N., Ramaekers, F. C. S., Raap, A. K.. Beck. J. L. M., Devilee, P.. van 36. Bauer, K. D.. Merkel. D. E., Winter. J. N.. Marder. R. J., Hauck, W. W. , Wallemark, der Ploeg, M., and Vooijs, G. P. In situ hybridization as a tool to study numerical C. B., Williams, T. J., and Variakojis, D. Prognostic implications of ploidy and chromosome aberrations in solid bladder tumors. Histochemistry, 89: 307-316, 1988. proliferative activity in diffuse large cell lymphomas. Cancer Res., 46: 3173-3178, 26. Hopman. A. H. N.. van Hooren, E., van de Kaa, C. A., Vooijs, G. P.. and Ramaekers, 1986. F. C. S. Detection of numerical chromosome aberrations using in situ hybridization in 37. Clark, G. M.. Dressier. L. G., Owens, M. A., Pounds. G., Oldaker. T.. and McQuire, paraffin sections of routinely processed bladder cancers. Mod. Pathol., 4: 503-513, W. L. Predictions of relapse or survival in patients with node-negative breast cancer 1991. by DNA now cytometry. N. Engl. J. Med., 320: 627-633, 1989. 27. Waldman. F. M.. Carroll, P. R.. Kerschmann. R., Cohen, M. B., Field, F. G.. and 38. Astler, V. B., and Coller, F. A. The prognostic significance of direct extension of Mayall, B. H. Centromeric copy number of chromosome 7 is strongly correlated with carcinoma of the colon and rectum. Ann. Surg.. 139: 846-851, 1954. tumor grade and labeling index in human bladder cancer. Cancer Res., 51: 3807- 39. Hedley. D. W., Friedlander, M. L., Taylor, I. W., Rugg. C. A., and Musgrove, E. A. 3813, 1991. Method for analysis of cellular DNA content of paraffin-embedded pathologic ma 28. Devilee. P.. Thierry, R. F., Kievits, T., Kolluri, R., Hopman, A. H. N., Willard, H. F., terial using flow cytometry. J. Histochem. Cytochem., 31: 1333-1335, 1983. Pearson, P. L., and Comelisse, C. J. Detection of chromosome aneuploidy in inter- 40. Dean. P. N.. and Jett. J. H. Mathematical analysis of DNA distributions derived from phase nuclei from human primary breast tumors using chromosome-specific repetitive flow microfluorometry. J. Cell Biol.. 60: 523-527, 1974. DNA probes. Cancer Res., 48: 5825-5830. 1988. 41. Rabinovitch, P.S. Numerical compensation for the effects of cell clumping on DNA 29. Chen. L-C., Neubauer. A., Kurisu. W., Waldman. F. M.. Ljung, B.-M.. Goodson, W., histograms. Cytometry.

686

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1993 American Association for Cancer Research. Chromosomes 8, 12, and 17 Copy Number in Astler-Coller Stage C Colon Cancer in Relation to Proliferative Activity and DNA Ploidy

Melissa G. Steiner, Seth P. Harlow, Edoardo Colombo, et al.

Cancer Res 1993;53:681-686.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/53/3/681

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/53/3/681. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.