1 Copy Number Aberrations in Benign Serous Ovarian Tumors: a Case for Reclassification?

Author Manuscript Published OnlineFirst on October 5, 2011; DOI: 10.1158/1078-0432.CCR-11-2080 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Copy number aberrations in benign serous ovarian tumors: a case for reclassification? Sally M. Hunter1, Michael S. Anglesio2, Raghwa Sharma3, C. Blake Gilks2,5, Nataliya Melnyk2, Yoke-Eng Chiew4,7, Anna deFazio for the Australian Ovarian Cancer Study Group1, Teri A. Longacre6, Anna deFazio4,7, David G. Huntsman2,5, *Kylie L. Gorringe1, *Ian G. Campbell1. 1Centre for Cancer Genomics and Predictive Medicine, Peter MacCallum Cancer Centre, Melbourne, Australia. 2The Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada. 3Anatomical Pathology, University of Sydney and University of Western Sydney at Westmead Hospital, Australia. 4Department of Gynaecological Oncology, Westmead Hospital, Westmead, Australia. 5Genetic Pathology Evaluation Centre of the Prostate Research Centre and Department of Pathology, Vancouver General Hospital and University of British Columbia, Vancouver BC, Canada. 6Stanford University School of Medicine, Stanford, CA 94305, United States. 7Westmead Institute for Cancer Research, University of Sydney at Westmead Millennium Institute, Westmead Hospital, Westmead, Australia. *Co-senior authors Running title: Copy number aberrations in benign serous ovarian tumors Keywords: ovarian, fibroma, serous, benign, borderline. Financial support: This work was supported by a grant (ID 628630) from the National Health and Medical Research Council of Australia (NHMRC). The AOCS was supported by the U.S. Army Medical Research and Materiel Command under DAMD17-01-1-0729, The Cancer Council Tasmania and The Cancer Foundation of Western Australia and the National Health and Medical Research Council of Australia (NHMRC). Corresponding author: Ian Campbell, VBCRC Cancer Genetics Research Laboratory, Peter MacCallum Cancer Centre, Locked Bag 1, A’Beckett Street, Melbourne, Victoria 8006, Australia. Phone: 613-9656-1803; Fax: 613-9656-1411; E- mail: [email protected]. Conflicts of interest: None. Manuscript notes: Word count: 3390 Figures/tables: 6 Supplementary figures/tables: 11 Abbreviations: BL, borderline; CN, copy number; CNA, copy number aberrations; CN LOH, copy neutral; FISH, fluorescence in situ hybridisation; H&E, haematoxylin and eosin; LGSC, low grade serous carcinoma; LOH, loss of heterozygosity; NOS, not otherwise specified; SBT, serous borderline tumor; SNP, single nucleotide polymorphism; WT, wildtype. 1 Downloaded from clincancerres.aacrjournals.org on October 4, 2021. © 2011 American Association for Cancer Research. Author Manuscript Published OnlineFirst on October 5, 2011; DOI: 10.1158/1078-0432.CCR-11-2080 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Statement of translational relevance Ovarian cancer is a very significant health burden and the seventh leading cause of cancer death in women. At the time of diagnosis, women often have advanced disease and as a consequence their prognosis is extremely poor. Our understanding of the progression of ovarian cancer through precursor stages and the molecular genetic events underlying these changes is very limited. Although a number of candidate precursor lesions have been proposed, the true contribution of these precursor lesions to the onset of ovarian cancer is unresolved. Identifying genuine ovarian cancer precursors and defining key molecular genetic events initiating and promoting tumorigenesis has important implications for early detection and treatment of ovarian cancers. 2 Downloaded from clincancerres.aacrjournals.org on October 4, 2021. © 2011 American Association for Cancer Research. Author Manuscript Published OnlineFirst on October 5, 2011; DOI: 10.1158/1078-0432.CCR-11-2080 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Abstract Purpose: Serous ovarian carcinomas are the predominant epithelial ovarian cancer subtype and it has been widely believed that some or all of these may arise from precursors derived from the ovarian surface epithelium or fimbriae, although direct molecular evidence for this is limited. This study aimed to perform copy number analysis using a series of benign and borderline serous ovarian tumors to identify underlying genomic changes that may be indicative of early events in tumorigenesis. Experimental Design: High resolution copy number (CN) analysis was performed on DNA from the epithelial and fibroblast components of a cohort of benign (N=39) and borderline (N=24) serous tumors using the Affymetrix OncoScan assay and SNP6.0 arrays. Results: CN aberrations were detected in the epithelium of only 2.9% (1/35) of serous cystadenomas and cystadenofibromas. In contrast, CN aberrations were detected in the epithelium of 67% (16/24) of the serous borderline tumors (SBTs). Unexpectedly, CN aberrations were detected in the fibroblasts of 33% (13/39) of the benign serous tumors and in 15% (3/20) of the SBTs. Of the 16 cases with CN aberrations in the fibroblasts, 12 of these carried a gain of chromosome 12. Conclusions: Chromosome 12 trisomy has been previously identified in pure fibromas, supporting the concept that a significant proportion of benign serous tumors are in fact primary fibromas with an associated cystic mass. This is the first high resolution genomic analysis of benign serous ovarian tumors and has demonstrated not only that the majority of benign serous tumors have no genetic evidence of epithelial neoplasia but that a significant proportion may be more accurately classified as primary fibromas. 3 Downloaded from clincancerres.aacrjournals.org on October 4, 2021. © 2011 American Association for Cancer Research. Author Manuscript Published OnlineFirst on October 5, 2011; DOI: 10.1158/1078-0432.CCR-11-2080 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Ovarian cancer is a very significant health burden and the 7th leading cause of cancer death in women worldwide1. At the time of diagnosis, women with epithelial ovarian cancer usually have advanced disease and as a consequence their prognosis is extremely poor (5 year survival for stage III & IV disease is only 25-30%)2, 3. For such a clinically significant disease, remarkably little is known about the molecular events that initiate the disease. While the paradigm that malignancies arise through a stepwise progression from benign precursors has been established for many malignancies, the archetypical example being colorectal carcinogenesis4, it remains unclear if this holds true for ovarian cancer. There is still considerable controversy as to what constitutes a true ovarian cancer precursor; an important definition that needs to be made in order to understand the origins and identify new clinical interventions for this lethal disease. Serous ovarian carcinomas are the predominant clinically important subtype but at present there is little experimental evidence from which to draw convincing conclusions about what constitutes the precursor(s) to this sub-type. It has been widely believed that some or all of these arise from precursors originating from the ovarian surface epithelium, via inclusion cysts or serous benign and borderline tumors5-8. Obvious candidate precursors are serous ovarian cystadenomas and cystadenofibromas, which are benign lesions with a cystic mass ≥1cm in diameter, lined with a single layer of cuboidal to columnar epithelium and commonly associated with a fibromatous stromal mass (Supplementary Figure S1). The serous epithelial layer of these tumors typically displays minimal cellular proliferation and no nuclear atypia. These are relatively common tumors, accounting for 60% of all serous ovarian tumors, while serous borderline tumors (SBTs) and low grade serous carcinomas 4 Downloaded from clincancerres.aacrjournals.org on October 4, 2021. © 2011 American Association for Cancer Research. Author Manuscript Published OnlineFirst on October 5, 2011; DOI: 10.1158/1078-0432.CCR-11-2080 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. (LGSCs) account for 10-15% and 2-9% of all serous ovarian tumors, respectively9-11. Benign serous tumors are presumed by many to be precursors to SBTs based on similarities in the cystic structure of some SBTs and the frequent detection of cases with a benign cyst co-existing with a SBT or cases comprising predominantly benign cysts with regions of atypical proliferation13. Despite the co-occurrence of benign, borderline and low grade carcinoma epithelial components, direct molecular evidence supporting benign lesions as precursors is limited. While some studies have shown the existence of KRAS and BRAF mutations in ovarian serous cystadenomas and cystadenofibromas co-existing with a region of atypical proliferation or adjacent SBT14, mutations in these genes have not been detectable in solitary benign tumors. SBTs have been firmly established as the likely precursor lesions to LGSCs, sharing similar rates of KRAS and BRAF mutation and low levels of genomic instability15-18. This is in contrast to the high rates of TP53 mutation and high levels of genomic alteration observed in high grade serous carcinomas18, 19. Traditionally, these have been believed to arise from ovarian surface epithelium and epithelial inclusion cysts formed from invaginated surface

Load more