Accumulation of genetic changes is associated with poor prognosis in ...

American Journal ofPathology, Vol. 151, No. 6, December 1997 Copynight C) American Societyfor Investigative Pathology

Accumulation of Genetic Changes Is Associated with Poor Prognosis in Grade 11 Astrocytomas

Satu-Leena Sallinen,*t Pauli Sallinen,* Hannu Haapasalo,* Juha Kononen,* Ritva Karhu,t Pauli Helen,§ and Jorma Isolat From the Department of Pathologe and Neurosurgery,s and the Laboratory of Cancer Genetics,t Tampere University Hospital, University of Tampere, Finland, and the Laboratory of Cancer Genetics,* National Center for Human Genome Research, National Institutes of Health, Bethesda, Maryland

and, subsequently, subtyping of the most malignant astrocytomas, glioblastomas.34 As recent molecular genetic and cytogenetic studies have demonstrated a wide spectrum of genetic aberrations across all astrocytoma malignancy scale,5-'0 it has become increasingly evident that genetic characterization of astrocytomas could be used for predicting aggressive tumor behavior before of its histopathological appearance. Glioblastomas represent the culmination of the astrocytoma malignant transformation process and very poor patient prognosis. Therefore, clinically and biologically more interesting are the grade 11 astrocytomas that give rise to glioblastomas. Usually grade 11 astrocytomas grow slowly, and complete surgical resection results in 5-year survival rates of nearly 80%.11 However, some grade 11 astrocytomas possess a strong capacity for rapid progression, creating a diagnostic and therapeutic dilemma. Currently, there is no consensus on the optimal treatment for grade 11 astrocytomas. Postoperative radiotherapy has been a controversial issue, and it is not routinely used. Patient age of more than 30 or 40 years, incomplete tumor resection, and contrast enhancement of the original tumor on computed tomography (CT) have been associated with shortened survival.1 1-3 In terms of diagnostic histopathology, gemistocytic differentiation and, recently, increased cell proliferation activity determined by the Ki-67MIB-l labeling index have been interpreted as signs of aggressive tumor behavior and poor prognoSiS.'3,'4 Nevertheless, alternative approaches for the characterization of individual tumors are highly warranted. One possibility is to determine the accumulation of genetic alterations that have already been used for subdividing glioblastomas. Clinically, this would have even greater significance in case of grade 11 astrocyto-

Unexpectedly aggressive clinical course of some grade II astrocytomas is a diagnostic dilemma for routine histopathology. Because increasing tumor malignancy is a consequence of progressive accumulation of chromosomal alterations, we investigated whether aggressive behavior of grade I astrocytomas could be predicted by the number and type of gross chromosomal aberrations. We used comparative genomic hybridization to analyze 11 grade H astrocytomas with typical (good, n = 7) or poor (n = 4) prognosis. The results were also compared with a reference material of 13 grade m-Iv astrocytomas and nine established cell lines. We found a median of two aberrations (range 0 to 4) in tumors with good prognosis and of 15.5 changes (range 8 to 28) in tumors with poor prognosis. Chromosomal gains were present in both groups, whereas chromosomal losses were frequent in tumors with poor prognosis (median 9.5, range 3 to 14) but rare in tumors with good prognosis (range 0 to 2). AR chromosomal gains were also found in the high-grade astrocytoma group and the majority of them in cell lines. Chromosomal losses in grade II astrocytomas with poor prognosis were very similar to those in grade m-IV astrocytomas and cell lines. We conclude that an early accumulation of genetic changes in grade H astrocytomas is closely associated with poor patient prognosis, suggesting diagnostic use for comparative genomic hybridization in characterization of grade II astrocytomas. (AmJ Pathol 1997, 151:1799-1807)

The development of comparative genomic hybridization (CGH) has provided a convenient method for genome-wide surveys of genetic aberrations in solid tumors. We performed CGH analysis in search for the chromosomal regions that are either amplified or deleted in a series of 11 diffusely infiltrating grade 11 astrocyto-

Pathogenesis of malignant gliomas is a consequence of multiple genetic changes that have accumulated either during stepwise progression from a low-grade tumor or more rapidly without a clinical history of a previous astrocytic neoplasm.'-' According to the well established scheme by the World Health Organization,' the genetic heterogeneity can be used for detailed characterization

Supported by grants from the Medical Research Fund of Tampere University Hospital, the Finnish Medical Foundation, Pirkanmaa Cancer Society, Pirkanmaa Cultural Society, Ida Montin Foundation, and the Finnish Cancer Society. Accepted for publication September 11, 1997. Address reprint requests to Dr. Satu-Leena Sallinen, Laboratory of Cancer Genetics, Institute of Medical Technology, University of Tampere, PO Box 607, FIN-33101 Tampere, Finland. E-mail: [email protected]

mas.

1799

1800

Sallinen et al

AJP December 1997, Vol. 151, No. 6

Table 1. Clinical Characterization of 11 Diffusely Infiltrating Grade II Astrocytomas Directed for CGH Analysis

Case number

Age/sex

1

43/Female

Histology/location

Tumor size* Contrast (cm) enhancement

Fibrillary grade 11 fronto-temporal

4

+

Extent of

surgeryt Subtotal

Postoperative adjuvant therapy None

Course of disease Recurrence at 70 months

2

29/Male

Fibrillary grade 11 parieto-occipital

5

-

Gross total

None

3

38/Male

11

3

-

Gross total

None

4

13/Female

11

3.5

+

Gross total

None

Alive at 61 months

5

3/Male

11

4

+

Gross total

None

Alive at 73 months

6

26/Male

Fibrillary grade temporal Fibrillary grade temporal Fibrillary grade cerebellar Fibrillary grade temporal

(II) Alive at 72 months Recurrence at 49 months (AA) Recurrence at 85 months (GBM) Died at 93 months Alive at 72 months

11

7

-

Subtotal

Radiotherapy

7

42/Male

Gemistocytic grade 11 frontal

7

+

Gross total

Radiotherapy

8

29/Male

Fibrillary grade 11 fronto-basal

5

-

Gross total

None

9

47/Male

2.5

-

Gross total

Radiotherapy

Recurrence at 64 months (AA) Alive at 79 months Recurrence at 46 months (GBM) Died at 50 months Recurrence at 26 months (GBM) Died at 29 months Died at 19 months

4

-

Gross total

Radiotherapy

3

-

Subtotal

Radiotherapy

10

45/Female

Fibrillary grade 11 temporal Gemistocytic grade 11 temporal

11

56/Male

Fibrillaryt grade 11

Recurrence at 18 months (GBM) Died at 28 months Died at 19 months

temporal The greatest dimension by CT or MRI; t, Based on macroscopic evaluation and postoperative CT; *, Minor oligoastrocytic focus, CGH done on the pure fibrillary differentiation; AA, anaplastic astrocytoma; GBM, glioblastoma. mas. We investigated genetic aberrations in the tumors associated with good and poor prognosis. The results were also compared with those of five anaplastic astrocytomas (grade 111), eight glioblastomas (grade IV), as well as nine established highly malignant astrocytoma cell lines.

Materials and Methods Tumor Samples Grade 11 Astrocytomas

Formalin-fixed, paraffin-embedded diffusely infiltrating grade 11 astrocytomas (n = 21) were collected from our previously reported material of 102 astrocytomas operated on at Tampere University Hospital (Tampere, Finland) between 1988 and 1992.16 Five cases were excluded from the study because of additional oligodendrocytic differentiation and five cases because of poor quality of DNA or lack of sufficient amount of starting material for the CGH analysis. The remaining eleven patients were operated on through open craniotomy for extensive or radical tumor resection. The clinical follow-up included brain imaging by CT or magnetic resonance. The clinical characterization of the tumors is presented in Table 1. The classification and grading of the tumors were done by two neuropathologists according to the criteria presented by the latest World Health Organi-

zation classification system.1 Glial fibrillary immunoacidic protein staining was used for the differential diagnosis. The location, size (greatest dimension), and contrast enhancement of the tumors were determined by preoperative CT scans. The extent of the surgical removal of the tumors was estimated on basis of macroscopic evaluation during the surgery and postoperative CT or magnetic resonance. Five patients received postoperative radiotherapy. Five tumors progressed to anaplastic astrocytomas (grade 111) or glioblastomas (grade IV). The primary cause of death in all cases was determined to be because of the brain tumor. The very short recurrence free-survival, quick malignant transformation into a glioblastoma and/or rapid deterioration leading to death of the four patients enabled us to divide the patients into categories of good and poor prognosis, the cut-off point being set to 2.5 years. Table 2 shows the immunohistochemical p53 and Ki-67MIB-1 data available from our previous studies.16,17 The mean nuclear size (tLm2) of tumor cells was estimated using computer-based image analysis on 5-,um histological sections (CAS200TM Cell Measurement Software, Becton Dickinson, Mountain View, CA). Tumor ploidy was determined by image cytometric DNA measurement (CAS200TM Quantitative DNA Analysis Software) from 50-,um cytospinned and Feulgen-stained tissue sections. Three ploidy categories were used: diploid, nondiploid (near diploid), and true aneuploid (more than 5% of the neoplastic cells exceeded the 5 c value (DNA content more than 5 sets of 23 chromosomes)).

CGH in Grade II Astrocytomas

1801


Table 2. Characterization of 11 Diffusely Infiltrating Grade II Astrocytomas by Immunohistochemistry and Quantitative Histopathology

Ki-67MIBl labeling

Case number 2 3 4 5 6 7 8 9 10 11

index17 (%)

p53 immuno-

staining16

Mean nuclear size (IIm2)

DNA ploidy*

57.7 56.5 59.0 56.1 56.6 56.4 55.8 56.5 55.8 56.6 55.4

Nondiploid True aneuploid Nondiploid Diploid Diploid Diploid True aneuploid True aneuploid True aneuploid True aneuploid True aneuploid

8.2 5.4 9.3 8.1 4, median value) with

Table 3. Genetic Aberrations in 11 Grade II Astrocytomas by CGH Number of chromosomal aberrations

Case number

Status at 30 months

1 2 3 4 5 6 7 8

Alive Alive Alive Alive Alive Alive Alive Deceased

(losses/gains) 0 (0/0) 1 (0/1) 2 (0/2) 2 (0/2) 3(0/3) 4 (2/2) 4(0/4) 8 (3/5)

9

Deceased

15 (10/5)

10

Deceased

16 (11/5)

11

Deceased

28 (14/14)

Chromosomal aberrations detected by CGH No gains or losses

+Xcen-q13 +7q32-qter, +X + 1 p32-pter, +11q13 +1p32-pter, +8q21, +X +7q21-q32, -16pter-q12, -22q, +X +3p24-pter, +llql4-qter, +12p, +Xq - 1 p32-pter, +1q24-q25, +6q15-q22, +8q12-q21, +9q22-q33,

-16pter-q12, -22, +X +1p31-pter, -1p22-q24, -2q22-q34, -3p14-q13, -4p15-qter, -5p14-q34, -9pter-q22, +11q13, -12q14-q22, -13q14-qter, 15q23, -18q, +19, -21q, +Xq -1p33-q23, -2q22-q33, -3p13-q13, -3q23-q28, -4q, -5p, +8p, -9p, -10pll-q22, +11q13, +11q23-q24, +12q24, -13q21-qter, -14cen-q31, -18pter-q21, +Xpter-q25 +1p34-pter, -1p31-q23, -2q22-q36, +3p21, +3p25, -3p14-q13, -3q24-qter, -4pl 5-qter, -5pl 4-q23, -6cen-q25, -7cen-q32, -9p13-pter, -10cen-q21, +10q26, +11q13, +11q23-24, -12p12q22, +12q24, -13, -14cen-q22, +14q32, +15q22-q25, +16p, +17, -18, 19, +20q, +22q

=


1803


I

3 I

6

4

7

:1 1 11

lIi

I I

I'I

~I

8

9

10

11

10

12

5

2 III III III III III III IllII III

I:3

:

1

1i

17

19

20

21

III

IIII IIII IIII IIII IIII IIII IIII

22

IIII

15

14

13

'16

3 i1 1

1Ii x

Figure 2. Summary of chromosomal aberrations detected by CGH in 11 grade II astrocytomas. Discontinuous lines represent the tumors with good prognosis ( n =

7) and continuous lines represent the tumors with poor prognosis (n = 4). Losses are indicated by lines on the left side of the chromosome idiogram, and lines on the right side represent gains.

recurrence-free (P = 0.0004) and overall survival (P = 0.0004, Mantel-Cox tests). The gains on chromosomes 1 p, 3p, 8q,11 q, and X and losses on chromosomes 16p and 22q were similar chromosomal aberrations in both of the prognostic subgroups, whereas most of the losses (-1p, -2q, -3, -4q, -5, -6q, -7q, -9p, -10q, -12, -13q, -14q, -18, -21) were found only in the tumors with poor prognosis.

30

r 0

0

.01

c .0

20, t

es 0

.c) .2 .0

Q

r.

1

-0-

.

a,

z0

m

Good Prognosis Poor Prognosis Grade III-IV Grade 11 Astrocytomas Astrocytomas

High-Grade Astrocytoma Cell Lines

Figure 3. The total number of chromosomal aberrations per tumor detected by CGH in grade II astrocytomas with good prognosis (n = 7) and poor prognosis (n = 4), grade III-IV astrocytomas (n = 13), and grade III-IV astrocytoma cell lines (n = 9). Horizontal lines indicate median values.

The Comparison of Grade 11 Astrocytomas with Grade Ill-lVAstrocytomas and Cell Lines To compare the genetic aberrations of the prognostic subgroups of the grade 11 astrocytomas with those of highly malignant astrocytomas, we analyzed 13 grade III-IV astrocytomas and nine cell lines (grade III-IV) by CGH. The results are summarized in Figure 4. The median number of chromosomal aberrations was eight (range 4 to 13) in grade III-IV astrocytomas. The most common aberrations were gains on chromosomes 7q (9 of 13) and 8q (7 of 13). Other common gains, in order of frequency, were found on chromosomes 10p (5 of 13), 17q (5 of 13), X (4 of 13), 1p (3 of 13), 3q (3 of 13), 9q (3 of 13), 11q24-qter (3 of 13), 19 (3 of 13), and 20q (3 of 13). Chromosomal loss was most often seen on chromosome 13q (3 of 8 glioblastomas and 2 of 5 anaplastic astrocytomas). Other common losses were detected on chromosomes 9p (4 of 13) and 10q (4 of 13). In cell lines the most common alterations were gains on chromosomes 7 (8 of 9), 19 (6 of 9), 20/20q (6 of 9), and 17 (4 of 9) and losses on chromosomes 13q (8 of 9), 14q (7of9),3(6of9),4/4q(6of9),6q(6of9),9p(6of9),10 (6 of 9), 1q (5 of 9), 11q (5 of 9), 12q (5 of 9), 18q (5 of 9), 2q (4 of 9), 5q (4 of 9), and Xq (4 of 9). In Table 4, the grade 11 astrocytomas are compared with grade III-IV astrocytomas of the current study and

1804 Sallinen et al AJP December 199 7, Vol. 151, No. 6

3

1

~~

2

8

6

01 11W 8lllri 71g 14

10

12

5 ~~~~~4

I oucs1cD

13

9

o t1~~~~~~~~CD o 19

2

221

22

15

0

J11 i12

3 6 8 10 2~~~~~~~~~~~~~~~~~~~~~~~~~~1 5 4

1

~ I ~~~~

~

~

19~~ ~ ~

20

21

22

17 14

15

Figure 4. Summary of the chromosomal aberrations detected by CGH in 13 grade III and IV astrocytomas (A), and nine Losses are indicated by lines on the left side of the chromosome idiogram, and lines on the right side represent gains.

X

high-grade astrocytoma cell lines (B).


1805


Table 4. Grade II Astrocytomas Compared with Grade III-IV Astrocytomas

Chromosomal aberration -1 p32-qter + 1 p32-pter

-2q -3 -4 -5 -6q +7q +8q -9p -10q +11q -12q +12q24 -13q -14q -18 +19 +X

Current study good prognosis (n = 7)

Grade II astrocytoma Current study poor prognosis (n = 4)

Grade III-IV astrocytoma Literature (n = 28)9.10

Current study (n = 13)

Literature (n = 93)5 810

3 (75) 3 (75) 3 (75) 3 (75)

2 (7) 1 (4)

1 (8) 3 (23)

-

-

3(75)

-

3 (75) 1 (25)

3(11) 2 (7)

1 (8) 1 (8)

4 (4) 7 (8) 4 (4) 6 (6)

2 (28) -

-

2 (28) 1 (14) -

-

2(28) -

-

-

5 (71)

-

6 (21)

-

-

2(15) 9 (69) 7 (54) 4 (31) 3 (23)

2(50) 3 (75) 2 (50) 3(75) 2 (50) 2 (50)

4(14) 1 (4) -

4(31) 2 (15) 1 (8)

3 (75) 2 (50)

-

5 (38) -

3(75) 2 (50) 3 (75)

1 (4) 3 (11)

4(31) 4 (31) 4 (31)

1 (4) -

-

9(10) 5 (5) 11 (12) 47 (51) 11 (12) 33 (35) 51 (55) 3(3) 3 (3) 2 (2) 28 (30) 11 (12) 4(4) 19 (20) 2 (2)

The comparison of the prevalence of the genetic aberrations in the 11 grade 11 astrocytomas and 13 grade III-IV astrocytomas studied here with 28 grade 11 astrocytomas and 93 grade l1I-IV astrocytomas collected from previously published studies of CGH changes on astrocytomas. The percentage of observations is presented in the parentheses.

astrocytoma cases presented in the earlier literature.510 The comparison is based on most frequent chromosomal aberrations detected in the grade 11 tumors with poor prognosis. It is shown that the chromosomal changes of grade 11 tumors with poor prognosis are very similar to those changes detected in anaplastic astrocytomas or

glioblastomas.

Discussion The majority of the diffusely infiltrating grade 11 astrocytomas exhibit a relatively good prognosis. However, some of the tumors show very aggressive behavior leading to early patient death. This creates a dilemma both for

the neuropathologist and the clinician. Currently, patient age and preoperative Karnofsky performance score are the most significant factors that provide help for prognostication and therapeutic decision making in case of grade 11 astrocytomas.11 Novel methods such as tumor proliferation analysis using Ki-67MIBl-1immunostaining have provided some help,13 but the efficiency of the methods is not sufficient for avoiding misinterpretation. We evaluated the CGH for its applicability to aid in prognostication of grade 11 astrocytomas and showed that an apparent grade 11 histology may hide an accumulation of genetic changes, the magnitude and type of which were closely associated with patient prognosis. Furthermore, we showed that the chromosomal aberrations of the tumors with poor prognosis resembled those of the highly malignant grade III-IV astrocytomas. We found a significant difference in the number of gross genomic changes among the grade 11 astrocytomas with good and poor prognosis. Although the highest number of aberrations was four (median 2, range 0 to 4)

in the seven tumors with good prognosis, tumors of the four patients who died within 2.5 years after the primary operation had 8 to 28 chromosomal aberrations per case (median 15.5). The potential clinical value of the CGH analysis in the characterization of individual grade 11 astrocytomas, on basis of the number of chromosomal alterations, was also shown in another context. Although the two astrocytomas with gemistocytic differentiation recurred as glioblastomas, they differed from each other in terms of the progression rate, ie, the one with many genetic changes by CGH progressed rapidly and the other with only few changes progressed more slowly. In addition, both the pediatric tumors were found to carry only a few chromosomal alterations, two to three respectively. Pediatric grade 11 astrocytomas have usually a less aggressive clinical course than the corresponding tumors of adults.25 The CGH findings support the benign behavior of the two pediatric cases. Interestingly, also the type of genetic aberrations varied among the prognostic subgroups. In the grade 11 astrocytomas with good prognosis, of 16 chromosomal changes only two were recorded as losses, whereas the remaining 14 were gains. In contrast, in tumors of the poor prognostic subgroup, we found losses to be the most frequent changes. The finding is similar to that with breast tumors in which multiple chromosomal losses have been associated with very aggressive tumor behavior.26 Only few cytogenetic or molecular cytogenetic data on diffusely infiltrating grade 11 astrocytomas have been reported in the literature,9-10.27 and none of these have examined the chromosomal aberrations in grade 11 astrocytomas in relation to patient survival. Karyotyping analyses have reported normal stem lines in the majority of grade 11 astrocytomas, sex chromosome losses in ap-

1806

Sallinen et al


proximately 20% of the cases, and rare clonal aberrations such as trisomy of chromosome 7 and structural changes of chromosome 1.27 The two studies using CGH included eight juvenile grade 11 astrocytomas and 20 tumors from adults. Shrock et a19 reported gains on chromosomes 7q, 8q, 10p, and 12p and losses on chromosomes lp, 4q, 9p, 11q, 18, 19, X, and Y in a material of 10 grade II astrocytomas of adults, whereas copy number changes in the eight juvenile tumors were rare, including a gain on chromosome 9, 11, and 19 and loss on chromosome 2 and 22 in single cases. Weber et al10 analyzed 10 grade 11 astrocytomas of adults, which had progressed into more malignant forms, and found the most common aberrations to be gains on chromosomes 8q, 12p, and 19p and losses on chromosomes 5p and Xp. Except for the gain on chromosome 10p, all of the gains previously reported in grade 11 astrocytomas were detected in both prognostic subgroups of the grade 11 astrocytomas of the current study. Interestingly, all of the gains were found also in the grade III-IV astrocytoma group, and the majority of them were in the established high-grade astrocytoma cell lines. However, losses found on chromosomes 1p, 3p, 4, 6q, 9p, 10q, 12q, 13q, and 18 in grade 11 tumors with poor prognosis raised our interest, because these changes were also detected in highly malignant astrocytomas and astrocytoma cell lines. It is also noteworthy that a gain on chromosome 19 and losses on chromosomes 9p, 10q, and 13q have been reported in earlier literature as frequent chromosomal changes in high-grade astrocytomas.5-8, 10 Thus, both the number and type of chromosomal aberrations in the grade 11 astrocytomas with poor prognosis showed a close resemblance to those found in highly malignant grade III-IV tumors. Because chromosomal losses were a rarity in grade 11 astrocytomas with good prognosis, it could be speculated that gains detected by CGH represent activated oncogenes that become significant in astrocytoma progression only after inactivation of critical tumor suppression recorded as chromosomal losses. Also, the order of appearance could reflect the heterogeneity in tumorigenetic pathways of astrocytomas, some of which lead to very aggressive clinical course and early patient death. This study shows that an early accumulation of genetic changes in grade 11 astrocytomas is closely associated with poor patient prognosis. Furthermore, the aggressive tumor behavior seems to be initially driven by chromosomal losses, because these changes were found somewhat equally prevalent both in grade 11 astrocytomas with poor prognosis and highly malignant grade III-IV astrocytomas as well as in established high-grade astrocytoma cell lines. Although we could not point out a single chromosomal aberration that could be directly associated with poor prognosis, we conclude that phenotypically grade 11 astrocytomas with very aggressive clinical course were genotypically closer to high-grade astrocytomas than low-grade tumors. The result opens a possibility for the CGH to be used as a diagnostic and prognostic tool for the characterization of grade 11 astrocytomas in clinical practice. Technically it is already possible because the reagents and image analysis tech-

nology for CGH analysis are commercially available. Large tumor materials can be analyzed with reasonable effort, and the development of techniques to extract DNA has extended the reliable use of the CGH on small tumor samples and paraffin-embedded tissue material.

Acknowledgments We thank Mrs. Minna Ahlstedt-Soini, Mrs. Arja Alkula, Ms. Lila Hakala, Mrs. Sari Toivola, and Mrs. Mariitta Vakkuri for their skillful technical assistance.

References 1. Kleihues P, Burger PC, Scheithauer BW: Histopathological Typing of Tumors of the Central Nervous System. Geneva, World Health Organization, 1993 2. Louis DN, Gusella JF: A tiger behind many doors: multiple genetic pathways to malignant glioma. Trends Genet 1995, 11:412-415 3. Louis DN: A molecular genetic model of astrocytoma histopathology. Brain Pathol 1997, 7:755-764 4. Von Deimling A, von Ammon K, Schoenfeld D, Wiestler OD, Seizinger BR, Louis DN: Subsets of glioblastoma multiforme defined by molecular genetic analysis. Brain Pathol 1993, 3:19-26 5. Schrock E, Thiel G, Lozanova T, du Manoir S, Meffert M-C, Jauch A, Speicher RR, NOrnberg P, Vogel S, Janisch W, Donis-Keller H, Ried T, Witkowski R, Cremer T: Comparative genomic hybridization of human malignant gliomas reveals multiple amplification sites and nonrandom chromosomal gains and losses. Am J Pathol 1994, 144:1203-1218 6. Kim DH, Mohapatra G, Bollen A, Waldman FM, Feuerstein BG: Chromosomal abnormalities in glioblastoma multiforme tumors and glioma cell lines detected by comparative genomic hybridization. Int J Cancer 1995, 60:812-819 7. Schlegel J, Scherthan H, Arens N, Stumm G, Kiessling M: Detection of complex genetic alterations in human glioblastoma multiforme using comparative genomic hybridization. J Neuropathol Exp Neurol 1996, 55:81-87 8. Weber RG, Sommer C, Albert FK, Kiessling M, Cremer T: Clinically distinct subgroups of glioblastoma multiforme studied by comparative genomic hybridization. Lab Invest 1996, 74:108-119 9. Schrock E, Blume C, Meffert M-C, du Manoir S, Bersch W, Kiessling M, Lozanova T, Thiel T, Witkowski R, Ried T, Cremer T: Recurrent gain of chromosome arm 7q in low-grade astrocytic tumors studied by comparative genomic hybridization. Genes Chromosomes Cancer 1996, 15:199-205 10. Weber RG, Sabel M, Reifernberger J, Sommer C, Oberstral3 J, Reifenberger G, Kiessling M, Cremer T: Characterization of genomic alterations associated with glioma progression by comparative genomic hybridization. Oncogene 1996, 13:983-994 11. Philippon JH, Clemenceau SH, Fauchon FH, Foncin JH: Supratentorial low-grade astrocytomas in adults. Neurosurgery 1993, 32:554559 12. McCormack BM, Miller DC, Budzilovich GN, Voorhees GJ, Ransohoff J: Treatment and survival of low-grade astrocytoma in adults - 19771988. Neurosurgery 1992, 31:636-642 13. Schiffer D, Cavalla P, Chio A, Richiardi P, Giordana MT: Proliferative activity and prognosis of low-grade astrocytomas. J Neurooncol 1997, 34:31-35 14. Watanebe K, Tachibana 0, Yonekawa Y, Kleihues P, Ohgaki H: Role of gemistocytes in astrocytoma progression. Lab Invest 1997, 76: 277-284 15. Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, Pinkel D: Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 1992, 258:818-821 16. Haapasalo H, Isola J, Sallinen P, Kalimo H, Helin H, Rantala I: Aberrant p53 expression in astrocytic neoplasms of the brain: association with proliferation. Am J Pathol 1993, 142:1347-1351 17. Sallinen PK, Haapasalo HK, Visakorpi T, Helen PT, Rantala IS, Isola JJ, Helmn HJ: Prognostication of astrocytoma patient survival by Ki-67


1807


18.

19.

20.

21.

22.

(MIB-1), PCNA, and S-phase fraction using archival paraffin-embedded samples. J Pathol 1994, 174:275-282 Isola J, deVries S, Chu L, Ghazvini S, Waldman F: Analysis of changes in DNA sequence copy number by comparative genomic hybridization in archival paraffin-embedded tumor samples. Am J Pathol 1994, 145:1301-1308 Kuukasjarvi T, Tanner M, Pennanen S, Karhu R, Visakorpi T, Isola J: Optimizing DOP-PCR for universal amplification of small DNA samples in comparative genomic hybridization: Genes Chromosomes Cancer 1997, 18:94-101 Kallioniemi O-P, 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 Chromosomes Cancer 1994, 10:231-243 Hyytinen E-R, Thalmann GN, Zhau HE, Karhu R, Kallioniemi O-P, Chung LWK, Visakorpi T: Genetic changes associated with the acquisition of androgen-independent growth, tumorigenicity and metastatic potential in a prostate cancer model. Br J Cancer 1997, 75: 190-195 Kuukasjarvi T, Karhu R, Tanner M, Kahkdnen M, Schaffer A, Nuppo-

23. 24.

25. 26.

27.

nen N, Pennanen S, Kallioniemi A, Kallioniemi O-P, Isola J: Genetic heterogeneity and clonal evolution underlying development of asynchronous metastasis in human breast cancer. Cancer Res 1997, 57:1597-1604 Kuukasjarvi T, Tanner M, Pennanen S, Karhu R, Kallioniemi O-P, Isola J: Genetic changes in intraductal breast cancer detected by comparative genomic hybridization. Am J Pathol 1997, 150:1465-1471 Karhu R, Siitonen S, Tanner M, Keinanen M, Makipernaa A, Lehtinen M, Vilpo AJ, Isola J: Genetic aberrations in pediatric acute lymphoblastic leukemia by comparative genomic hybridization. Cancer Genet Cytogenet 1997, 95:123-129 Pollack IF, Claassen D, al-Shboul Q, Janosky JE, Deutsch M: Lowgrade gliomas of the cerebral hemispheres in children: an analysis of 71 cases. J Neurosurg 1995, 82:536-547 Isola JJ, Kallioniemi O-P, Chu LW, Fuqua SAW, Hilsenbeck SG, Osborne CK, Waldman FM: Genetic aberrations detected by comparative genomic hybridization predict outcome in node-negative breast cancer. Am J Pathol 1995, 147:905-911 Mitelman F: Catalog of Chromosome Aberrations in Cancer. New York, Wiley-Liss, 1994