| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Department of Clinical Genetics, University Hospital, SE-221 85 Lund, Sweden [M. H., F. M., N. M., I. F-E.]; Department of Genetics, Lund University, SE-223 62 Lund, Sweden [T. S.]; and Department of Cancer Genetics, The Norwegian Radium Hospital, NO-0310 Oslo, Norway [S. H.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
BC,3 mostly TCC, is a common malignancy, being the fourth most common cancer among men in Europe and the United States (7) . The clinical course of BC is very difficult to predict based on prognostic factors such as tumor grade and stage, multifocality, tumor shape, location, and the presence of carcinoma in situ (8) . The identification of novel progression markers associated with aggressive disease behavior might help design new therapeutic strategies to be tried out in progression-prone cases, and the acquired genetic aberrations of TCC would seem to be prime candidates for such a role. Although >200 cases of BC with clonal karyotypic abnormalities have been reported,4 knowledge about their cytogenetic profile remains meager. The aberrations are clearly nonrandom, but no completely specific primary or secondary karyotypic abnormality has been identified, and the chronological order of how chromosomal changes are acquired during disease progression is not well understood. In the present study, we constructed a genomic imbalance map from all of the reported BC cases with abnormal karyotypes4 and used this map to identify the most frequent imbalances. Tumors were then classified with respect to the presence or absence of these imbalances and statistically analyzed to identify and assess karyotypic profiles, correlations between imbalances and the stage and grade of the tumor, and the chronological order of appearance of the chromosomal imbalances.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10% of the cases were selected for the present study (Table 1)
. When two or more neighboring bands exceeded the 10% limit, they were combined to form a contiguous region of imbalance. Each case was then evaluated with regard to involvement of each of the selected regions, treating each imbalance as a binary variable, i.e., present or absent. The initial analyses included calculations of basic statistics, such as relative frequencies, correlation coefficients, and distributions of the NIPT.
|
PCA.
To analyze for possible patterns of correlated imbalances, PCA was performed using the SAS package (SAS Institute Inc.) or the Statistica software (Statsoft, Tulsa, OH). PCA is a standard multivariate method frequently used to search for underlying structures in data sets (10)
. In short, principal components are linear combinations of the original variables, orthogonal, and ordered with respect to their variance so that the first principal component has the largest variance. To analyze imbalances, these were used as variables and the individual tumors as the observations; this will group imbalances seen frequently in the same tumors. To analyze the tumor population, the tumors were used as variables and the imbalances as observations; this will group tumors with similar sets of imbalances. To organize the tumors in three dimensions with respect to complexity, the NIPT for each individual tumor was used as an observation in addition to the imbalances. Because this parameter shows the greatest variation, from 1 to 12 compared with 0 or 1 for the individual imbalances, the first principal component will be dominated by this factor, and the position of the tumors along the first principal component will be almost identical to karyotypic complexity. The PCA results were analyzed both in three and two dimensions.
Clinical Correlations.
Grade (G1, G2, and G3) and stage (Ta, T1, T2, T3, and T4) were recorded when reported. Because only a few stage T4 tumors were present in the material, these were grouped together with T3 tumors. Correlations between chromosomal imbalances and the stage and grade of tumors were analyzed by Pearson correlation coefficient, 
2, and Fisher’s exact test using the Statistica software.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The most frequent imbalance was -9 (42% of the cases) followed by -11p (27%) and 8p- (23%). The distribution of the NIPT showed a geometrical distribution with very few tumors having >10 imbalances (Fig. 1A)
. The same type of distribution was obtained when all of the aberrations, including also imbalances less frequent than 10%, balanced translocations, and marker chromosomes, were included (Fig. 1A)
. The temporal analysis revealed loss of chromosome 9 to be the earliest imbalance followed by gain of chromosome 7 (Fig. 2)
. Both these imbalances appeared earlier (TO = 1) than expected from the simulation. Among the moderately early imbalances (TO = 2–4), 1q+ appeared earlier than expected from the simulation, whereas of the late imbalances (TO > 4), -5q, +5p, -6q, -15, -18, and -22p appeared later than expected from the simulation.
|
|
, one can discern two imbalance clusters, a large one consisting of 1p-, -2p, -4p, -10, -15, -16 -18, and -22p, and a minor one including 3p-, +8q, 8p-, and -17p. Separated from these clusters, on the right side, are located the early imbalances -9, 1q+, and +7, and also -11p. The late imbalances -5q, +5p, and -6q were located between the two clusters on the opposite side of the early imbalances. Furthermore, the early imbalance -9 was negatively correlated with +7 (P = 0.002), 3p- (P = 0.022), 8p- (P = 0.025), and +8q (P = 0.046) but positively correlated with -11p (P = 0.002), whereas +7 was positively correlated with 3p- (P = 0.013). The first principal component correlated well with the TA (r = 0.84; P < 0.001). By combining the data obtained from the TA and the PCA, two cytogenetic pathways emerge. One is characterized by -9 as the initiating imbalance, has -11p and/or 1q+ as frequent secondary events, and 1p-, -4p, -6q -5q, +5p, -10, -15, -16, -17p, and -18 as late changes. In the second pathway, the early imbalance +7 is followed by 3p-, 8p-, +8q, and -6q, -5q, and +5p as late changes.
|
revealed geometrically shaped distributions with increasing averages, which in G3 tumors produced an almost flat distribution. Thus, although stage/grade is correlated with complex karyotypes, many of the high stage and grade tumors have simple karyotypes.
The cases were divided into Ta, T1, T2, and T3 tumors, and the frequency for each imbalance was calculated and plotted (Fig. 4)
. The frequencies for 11 of the 19 selected imbalances showed significant dependence on stage. The early imbalance -9 was most frequent in Ta and T1 tumors, 49% and 54%, respectively, but showed decreased frequencies in T2 (25%) and T3 (10%) tumors. Loss of chromosome 9 was positively correlated with T1 but negatively correlated with T3 tumors. Gain of chromosome 7, the second early imbalance, was seen in 7% of Ta but in 26% of T1 tumors. The frequencies declined slightly in T2 (17%) and T3 (21%). Gain of chromosome 7 was positively correlated with T1 but negatively correlated with Ta tumors.
|
As the imbalances +5p, -5q, and -6q were well separated from the remaining imbalances in the PCA of the imbalances, tumors containing these three imbalances were analyzed separately. This group of tumors was found to have +7 in 20% and -9 in 40% of the cases and consisted only of T2 and T3 tumors. However, when these tumors were grouped according to stages T2 (n = 8) and T3 (n = 9), +7 was found to be absent from the T2 but present in 3/9 of the T3 tumors, whereas -9 was seen in 4/8 of the T2 and in none of the T3 tumors. Thus, although -5q, +5p, and -6q are late imbalances, they appear at different tumor stages depending on the previous imbalances.
Analysis of the Tumor Population.
A PCA was performed in which the tumors were the variables and the imbalances the cases. The number of imbalances present in each individual tumor was also included as a parameter, in addition to the presence of specific imbalances; this procedure groups tumors with similar complements of imbalances and number of imbalances close together in the resulting PCA diagram. In Fig. 5A
, six clusters of tumors are seen in which each tumor is organized in order of karyotypic complexity, i.e., less complex tumors in the periphery and more complex tumors toward the center. The obtained structure in Fig. 5A
was systematically analyzed for imbalances that organized the individual clusters, i.e., they were present in all of the tumors encompassing one or more of the identified subgroups. It was found that all of the subgroups could be characterized by the presence or absence of +7, -9, -11p, or combinations thereof. In Fig. 5B
, seven clusters were defined with respect to the presence or absence of +7, -9, or -11p. The tumors located between the +7 and -11p tumors, brown in Fig. 5B
, were characterized by the absence of all three imbalances.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The TA revealed a distinct order of appearance of the common imbalances. Loss of chromosome 9 and gain of chromosome 7 both appeared earlier than expected from the simulations indicating that they are important in the first stages of tumorigenesis. Of the nine imbalances having TOs of 5 or more, six (-5q, +5p, -6q, -15, -18 and -22p) appeared later than expected from the simulations, indicating that they are important in tumor progression. Their selective value may be dependent on the presence of other earlier imbalances. The TA showed a good correspondence to tumor stage. The two early imbalances, -9 and +7, were positively correlated with T1. Of the four imbalances (-10, -11p, -15, and -16) showing positive correlation to T2, only two (-15 and -16) had TOs larger than 5, and only one (-15) appeared later than expected from the simulation, whereas six (3p-, -4p, -5q, +5p, -6q, and -22p) of the nine imbalances showing positive correlation to T3 had TOs of 5 or more, and four (-5q, -6q, +5p, and -22p) appeared later than expected from the simulations. Furthermore, many of the later imbalances, which showed a positive correlation with stages T2 and T3, showed a negative correlation to Ta. Thus, the temporal appearance of imbalances in BC is highly ordered and correlated with stage.
Two Cytogenetic Pathways in TCC.
In the PCA of the imbalances, the first principal component was found to correspond well to the TA and, thus, approximated a time axis within the representation. The two early imbalances -9 and +7 were well separated in the PCA and also showed negative correlation. This shows that -9 and +7 represent alternative early steps in the karyotypic evolution of BC. Three imbalances, 3p-, 8p-, and +8q, were positioned close to +7 in the PCA diagram and were also negatively correlated with -9, which suggests a cytogenetic pathway with +7 as an early imbalance followed by 8p-, +8q, and 3p-. Loss of chromosome 9, on the other hand, was located close to the imbalances -11p and 1q+, suggesting that these, particularly -11p, are secondary changes to the loss of chromosome 9.
To expand the analysis of the cytogenetic events, the tumors were subjected to PCA. To resolve the complex PCA pattern obtained, the NIPT was added as an observation for each individual tumor. This procedure produced a characteristic structure with seven clusters of tumors. By systematically analyzing these clusters for imbalances that were present in all of the tumors in a given subgroup, three cluster-determining imbalances were identified: +7, -9, and -11p. Either the presence or absence of these three imbalances could describe all of the potential subgroups. This adds weight to the interpretation that -9, -11p, and +7 are important in the development of BC.
In the analysis of the frequencies of imbalances with respect to tumor stage, -9 behaved in an extraordinary way, being very frequent in Ta and T1 tumors, 49% and 52%, respectively, but rare in T3 tumors (10%). Furthermore, -9 showed a positive correlation with T1 tumors but a negative one to T3 tumors. Thus, the low frequency of -9 in high-stage tumors is highly significant. The imbalances -10, -11p, -15, and -16, shown to be associated with -9 in the PCA, all showed a positive correlation with T2 tumors and a decreased frequency in T3 compared with T2 tumors. This suggests the existence of a cytogenetic pathway through -9 with -11p and 1q+ as secondary changes followed later by -15, -16, and -10. The -9 tumors seem to progress from stage Ta to T1 and possibly to T2, but very rarely to T3. Although the frequency of 9p (9%) and 9q (6%) losses did not reach the inclusion criteria, the importance of -9 in BC (11) motivated an analysis also of these partial imbalances. Loss of 9q was found to appear relatively early (TO = 3) and was seen earlier than expected from the simulation. Furthermore, the frequency of -9q declined with stage, being 10% in Ta and absent in T3 tumors. Losses of 9p material, on the other hand, increased in frequency with stage, from 5% in Ta to 11% in T3, and appeared later than -9q (data not shown). Thus, loss of 9q material behaved similarly to loss of the entire chromosome 9, being an early imbalance predominantly present in low-stage tumors, whereas -9p appeared to be a rather late event.
Gain of chromosome 7 was seen in 7% of the Ta tumors but in 26% of the T1 tumors. Gain of chromosome 7 was also negatively correlated with Ta tumors and positively correlated with T1 tumors. Furthermore, the frequency of +7 was maintained (
20%) in T2 and T3 tumors. This, together with the previous results, suggests the existence of a second cytogenetic pathway occurring in more aggressive tumors than those of the -9 pathway characterized by the changes +7, 8p-, +8q, and 3p-. The specificity of the -9 pathway for Ta-T2 tumors and the +7 pathway for T1-T3 tumors was particularly apparent in tumors with -6q, -5q, and +5p. The frequencies of -9 and +7 in this advanced group of tumors, which were either T2 or T3, were 40% and 20%, respectively. However, when this group was divided into T2 and T3 tumors, half of the T2 tumors contained -9 and none +7, whereas one-third of the T3 tumors contained +7 and none -9.
A large subgroup of TCCs showed absence of the imbalances -9, +7, and -11p. These tumors also demonstrated a lower NIPT. Despite this, a higher frequency of advanced tumors, invasive versus noninvasive and G3 versus G2, was seen in this class of tumors. As the TA showed a highly ordered temporal appearance of genetic events in TCCs, leading to the more advanced stages, this suggests that these tumors may have cytogenetically cryptic changes equivalent to the early events and, therefore, show a lower NIPT. Thus, given a multistep model of tumorigenesis, the initiating events of TCC may not always be detectable as cytogenetic changes.
Final Conclusions.
The presented data indicate the presence of at least two cytogenetic pathways in BC, the first characterized by the loss of chromosome 9, the second by the gain of chromosome 7 (Fig. 6)
. Loss of chromosome 9 was predominantly followed by -11p and 1q+ and then by -17p. Of the later imbalances, -10, -15, and -16 were predominantly associated with -9. Gain of 7 was followed by 8p- and +8q, both of which were highly specific for the +7 pathway, and by -17p. Furthermore, 3p- was a specific late event in the +7 pathway. Although these events, except for -17p, had the character of being pathway-specific, the two well-separated cytogenetic routes converged at subsequent stages as both accrued the imbalances -2p, -4p, -5q, +5p -6q, -18, and -22p. Richter et al. (12)
has presented a similar but not converging pathway of chromosomal events. One interesting feature of the identified pathways is that the -9 pathway predominately leads to Ta-T2 tumors, whereas the +7 pathway leads to T1-T3 tumors. The fact that these two pathways are associated with different levels of tumor aggressiveness but still have many of the very late imbalances in common, indicates that the progression to an aggressive tumor may already be predetermined by the early events in the pathway, something that could possibly be used as a prognostic tool in the clinical management of BC patients.
|
|
FOOTNOTES |
|---|
1 Supported by the Swedish Cancer Society, the Norwegian Cancer Society, Swedish Match AB, the Crafoord Foundations, the Erik-Philip Sörensen Foundation, and the Nilsson Family Foundation. 
2 To whom requests for reprints should be addressed, at Department of Clinical Genetics, University Hospital, SE-221 85 Lund, Sweden. Phone: 46-46-173739; Fax: 46-46-131061; E-mail: mattias.hoglund{at}klingen.lu.se
3 The abbreviations used are: BC, bladder cancer; TCC, transitional cell carcinoma; TA, temporal analysis; NIPT, number of imbalances per tumor; TO, time of occurrence; PCA, principal component analysis.
4 Internet address: http://cgap.nci.nih.gov/Chromosomes/Mitelman.
Received 5/21/01. Accepted 9/19/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Saletta, G. Matullo, M. Manuguerra, S. Arena, A. Bardelli, and P. Vineis Exposure to the Tobacco Smoke Constituent 4-Aminobiphenyl Induces Chromosomal Instability in Human Cancer Cells Cancer Res., August 1, 2007; 67(15): 7088 - 7094. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Knowles Molecular subtypes of bladder cancer: Jekyll and Hyde or chalk and cheese? Carcinogenesis, March 1, 2006; 27(3): 361 - 373. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Hiller, J. Bradtke, H. Balz, and H. Rieder CyDAS: a cytogenetic data analysis system Bioinformatics, April 1, 2005; 21(7): 1282 - 1283. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Heidenblad, E. F. P. M. Schoenmakers, T. Jonson, L. Gorunova, J. A. Veltman, A. G. van Kessel, and M. Hoglund Genome-Wide Array-Based Comparative Genomic Hybridization Reveals Multiple Amplification Targets and Novel Homozygous Deletions in Pancreatic Carcinoma Cell Lines Cancer Res., May 1, 2004; 64(9): 3052 - 3059. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Isaka, A. L. Nestor, T. Takada, and D. C. Allison Chromosomal Variations Within Aneuploid Cancer Lines J. Histochem. Cytochem., October 1, 2003; 51(10): 1343 - 1353. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hoglund, D. Gisselsson, G. B. Hansen, T. Sall, and F. Mitelman Ovarian Carcinoma Develops through Multiple Modes of Chromosomal Evolution Cancer Res., June 15, 2003; 63(12): 3378 - 3385. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hoglund, D. Gisselsson, G. B. Hansen, T. Sall, F. Mitelman, and M. Nilbert Dissecting Karyotypic Patterns in Colorectal Tumors: Two Distinct but Overlapping Pathways in the Adenoma-Carcinoma Transition Cancer Res., October 15, 2002; 62(20): 5939 - 5946. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Fabarius, A. Willer, G. Yerganian, R. Hehlmann, and P. Duesberg Specific aneusomies in Chinese hamster cells at different stages of neoplastic transformation, initiated by nitrosomethylurea PNAS, May 14, 2002; 99(10): 6778 - 6783. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hoglund, D. Gisselsson, G. B. Hansen, T. Sall, and F. Mitelman Multivariate Analysis of Chromosomal Imbalances in Breast Cancer Delineates Cytogenetic Pathways and Reveals Complex Relationships among Imbalances Cancer Res., May 1, 2002; 62(9): 2675 - 2680. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
