Premium
Comparative genomic hybridization detects genetic imbalances in primary ovarian carcinomas as correlated with grade of differentiation
Author(s) -
Kiechle Marion,
Jacobsen Anja,
SchwarzBoeger Ulrike,
Hedderich Jürgen,
Pfisterer Jacobus,
Arnold Norbert
Publication year - 2001
Publication title -
cancer
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.052
H-Index - 304
eISSN - 1097-0142
pISSN - 0008-543X
DOI - 10.1002/1097-0142(20010201)91:3<534::aid-cncr1031>3.0.co;2-t
Subject(s) - comparative genomic hybridization , grading (engineering) , ovarian cancer , biology , tumor grade , pathology , oncology , cancer research , medicine , cancer , gene , genome , genetics , ecology
BACKGROUND In the current study the authors attempted to evaluate genetic alterations in a large set of primary ovarian carcinomas and to compare the genetic findings with clinical parameters such as grade of tumor differentiation. This strategy was applied to identify chromosomal regions containing genes associated with tumor progression. METHODS Genetic imbalances were assessed in 106 primary ovarian carcinomas using comparative genomic hybridization (CGH). CGH was applied because it is a powerful tool with which to screen the entire genome of a tumor for genetic changes by highlighting regions of altered DNA sequence copy numbers (deletions and amplifications). Multivariate statistical standard procedures were used to determine an association between tumor grading and genetic alterations. RESULTS One hundred three carcinomas showed aberrant CGH profiles. The most frequent alterations were amplifications of 8q, 1q, 20q, 3q, and 19p, which occurred in 69–53% of tumors, and underrepresentations of 13q, 4q, and 18q, which occurred in 54–50% of tumors. Undifferentiated ovarian carcinomas (World Health Organization Grade 3) were found to be correlated significantly with underrepresentation of 11p and 13q as well as with overrepresentation of 8q and 7p ( P = 0.001, 0.001, 0.01, and 0.027, respectively). However, 12p underrepresentation and 18p overrepresentation were significantly more frequent in well and moderately differentiated tumors ( P = 0.01 and 0.004, respectively). To facilitate the interpretation and clinical application of the results of the current study, the significant aberrations were translated into a score system. This score system can be used easily for the prediction of an undifferentiated phenotype with a specificity and sensitivity of 79% and 86%, respectively. CONCLUSIONS The current study data show that primary ovarian carcinomas are based on consistent genetic alterations that most likely are important for the development of this tumor entity. The correlation between certain aberrations and undifferentiated carcinomas may help to discriminate between primary and secondary genetic events and may indicate the location of those genes involved in cellular functions associated with tumor progression and the development of anaplastic and aggressive phenotypes. Cancer 2001;91:534–40. © 2001 American Cancer Society.