Uveal melanoma is the most frequently occurring

Transcription

1 Effect of Heterogeneous Distribution of Monosomy 3 on Prognosis in Uveal Melanoma Inge H. G. Bronkhorst, MD; Willem Maat, MD; Ekaterina S. Jordanova, PhD; Wilma G. M. Kroes, MD; Nicoline E. Schalij-Delfos, MD, PhD; Gregorius P. M. Luyten, MD, PhD; Martine J. Jager, MD, PhD N Context. Fluorescence in situ hybridization (FISH) analyses on tumor sections and on isolated nuclei showed that even low numbers of cells with monosomy of chromosome 3 adversely affected survival. Objective. To determine what percentage of uveal melanoma cells with monosomy of chromosome 3 influences patient mortality. Design. To determine the presence of monosomy 3, karyotyping and FISH on cultured cells and FISH on isolated nuclei were performed on 50 primary uveal melanomas. Clinical and pathologic prognostic factors were assessed and compared with 5-year survival data. Analyses were performed using Cox proportional hazards test, log-rank analysis, sensitivity, specificity, and positive and negative likelihood ratios. Results. Combined karyotyping and FISH on cultured cells showed monosomy 3 in 19 of 50 cases (38%), whereas determination of the monosomy 3 status by FISH on isolated nuclei with a threshold of 5% assigned 31 of 50 cases (62%) to the monosomy-3 category. When monosomy 3 on isolated nuclei with a cutoff value of 5% was used, a significant difference in 5-year survival was present (hazard ratio, 15.5; P =.007), indicating that monosomy 3 in greater than 5% of tumor cells is related to death due to metastases. Conclusion. In uveal melanoma, the presence of greater than 5% of cells with monosomy 3, as determined by FISH on isolated nuclei, is associated with the development of metastases within 5 years after enucleation. (Arch Pathol Lab Med. 2011;135: ) Uveal melanoma is the most frequently occurring primary intraocular tumor in adults and often carries a bad prognosis, with up to 50% of patients dying of metastases. 1 Several histologic prognostic factors have been described for this type of cancer, such as large tumor size, involvement of the ciliary body, extrascleral extension, and an inflammatory phenotype. 2,3 Specific genetic alterations that are significantly associated with the development of metastases have been identified, such as monosomy 3. In spite of the development of very precise techniques for detecting chromosomal alterations, such as spectral karyotyping, fluorescence in situ hybridization (FISH), comparative genomic hybridization, and multiplex ligation-dependent probe amplification (MLPA), loss of a complete chromosome (monosomy 3) is still the most important genetic aberration to predict survival, superior to clinical and pathologic features. Several reports 7 10 have shown that within one tumor, loss of one chromosome 3 can be present in one area, but not in another. Recently, Dopierala et al 11 showed that this heterogeneity not only occurs for chromosome 3 but also for other chromosomes, such as chromosome arm 1p and chromosomes 6 and 8. Accepted for publication December 7, From the Departments of Ophthalmology (Drs Bronkhorst, Maat, Schalij-Delfos, Luyten, and Jager), Pathology (Dr Jordanova), and Clinical Genetics (Dr Kroes), Leiden University Medical Center, Leiden, the Netherlands. The authors have no relevant financial interest in the products or companies described in this article. Reprints: Martine J. Jager, MD, PhD, Department of Ophthalmology, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, Netherlands ( M.J.Jager@lumc.nl). We wondered whether there was a difference in prognosis when only part of the tumor displays monosomy 3, compared with when most tumor cells display this aberration. We had previously performed standard cytogenetic analysis by karyotyping and FISH, on cultured cells obtained from uveal melanomas after enucleation, and by FISH on nuclei isolated from paraffin-embedded tissue 7 from the same tumors. We had quantified the number of nuclei with monosomy 3 in these tumors and were able to determine 5-year survival in patients with tumors with variable frequencies of cells with monosomy 3. MATERIALS AND METHODS Patients and Tissue Samples Fresh material and formalin-fixed, paraffin-embedded specimens from 50 patients with uveal melanoma were obtained from tumors that had been enucleated between 1999 and 2004 at the Leiden University Medical Center (Leiden, the Netherlands), and from which, enough material was present for cytogenetic analysis. After enucleation, tumor material for cytogenetic testing was obtained immediately after the globe had been opened. Each tumor sample was further processed for conventional histopathologic diagnosis. The 5-year follow-up survival data were obtained from charts. In the Netherlands, all causes of death are centrally registered. The database of the Integral Cancer Center West was used, which registers data on metastases and checks the survival status of each patient with uveal melanoma on a yearly basis. In addition, if the cause of death was not known and no metastatic disease was reported, we contacted the general practitioners. Follow-up time was measured in months. During the follow-up period, 22 patients (44%) died, 19 of these 22 from metastatic disease. One patient died from respiratory insufficiency and one patient from another 1042 Arch Pathol Lab Med Vol 135, August 2011 Heterogeneity for Chromosome 3 and Survival Bronkhorst et al

2 primary tumor (esophageal cancer); in one case, the cause of death remained unknown. There were no patients with metastasis who were still alive at the end of the follow-up, and no patients were lost to follow-up. All patients were informed that tumor material could be used for research purposes and signed an informed consent form. The use of tumor material for research has been approved by the Ethical Committee of the Leiden University Medical Center. The research protocol followed the tenets of the Declaration of Helsinki. Histopathologic Examination Histologic sections were prepared from tissues fixed in 4% neutral-buffered formalin for 48 hours and embedded in paraffin. Hematoxylin-eosin stained 4-mm sections were reviewed by one ocular pathologist for confirmation of the diagnosis and evaluated for histologic parameters, which included largest basal diameter (in millimeters), prominence (apical height, in millimeters), cell type according to the modified Callender classification, 12 ciliary body involvement, and intrascleral in-growth. These parameters were used for classification in the TNM (tumor-node-metastasis) category/stage (American Joint Committee on Cancer, 7th edition). 13,14 Chromosome 3 Status Determination of chromosome 3 status had been performed by standard cytogenetic testing, that is, by karyotyping and FISH on cultured cells and by interphase FISH on nuclei isolated from paraffin-embedded tissue, as described previously. 7 When either test on cultured cells showed monosomy 3, the tumor was categorized as having monosomy 3. Briefly, after enucleation, a small part of each tumor was removed and sent out for cell culture. Following mechanical dissection of the tissue, cells were washed and placed into a flask with RPMI 1640 and 15% fetal bovine serum (Invitrogen, Breda, the Netherlands) and into another flask with Amniochrome II (Cambrix Bio Science, Verviers, Belgium). The flasks were cultured at 37uC with 5% carbon dioxide for up to 4 weeks and harvested, according to standard protocols, when at least 75% of the surface was covered with cells. When cell culturing was successful, conventional karyotyping and FISH with a DNA-specific probe (CEN 3; Vysis, Des Plaines, Illinois) were performed to determine the presence of chromosome 3 changes, according to established methods, 15 with a cutoff value of 10% (standard value for the probe). Cytogenetic analysis was performed on GTG-banded (G-banding with Giemsa and trypsin) metaphases. In the case of a healthy karyotype, at least 20 metaphases were analyzed. When 3 cells with loss of one copy of chromosome 3 were observed, monosomy 3 was identified. Interphase FISH was performed on nuclei isolated from 50-mm paraffin-embedded tissue sections. 16,17 After enzymatic digestion with pepsin and with 2 additional washing steps with phosphate-buffered saline, the cells were filtered through a nylon filter with 70-mm pore size (Verseidag-Industrietextilen GmbH, Kempen, Germany). Nuclei were fixed with methanol to acetic acid (3:1). Before making cytospins, nuclei density was determined and adjusted with fixative to ensure that 400 to 500 nuclei would be present on each slide. The slides were air dried and used for hybridization. A DNA probe (CEP3 SpectrumOrange to CEN 3, Vysis, Des Plaines, Illinois), specific for the centromere region of chromosome 3 (band 3p11.1-q11.1), was used for hybridization according to the manufacturer s protocol (Abbott Laboratories, Downers Grove, Illinois). Slides were analyzed by fluorescence microscopy (model DMRXA; Leica Microsystems, Cambridge, United Kingdom). Image capture was performed by a monochrome charge-coupled device camera (Cohu, Inc, San Diego, California) attached to the fluorescence microscope and commercial software (Q-FISH; Leica). Uveal melanomas contain infiltrating cells, such as lymphocytes and macrophages. Thus, our technique could lead to an underestimation of the number of monosomy 3 tumor cells. However, visual inspection of the sections using CD68 staining showed that the mean percentage of macrophage infiltrate in our tumor whole slides was only 6% (range, 1% 15%) of the tumor. 18 Therefore, no antibodies were used to sort the melanoma cells. Three healthy tonsils from different individuals were used as controls. The tonsil sections were treated in exactly the same manner as the tumor samples. The cutoff level was set at the mean of these controls plus 3 times the standard deviation (ie, at 5%) for detecting monosomy 3. 7 Two independent observers assessed all slides, each without knowledge of the results obtained by the other investigator, to ensure the accuracy of the quantification of the slides. The 2 observers each analyzed 100 cells, 200 interphase nuclei in total. Scoring results corresponded in 47 cases (94%); for 3 cases (6%) with values around the cutoff value, consensus was reached during a simultaneous session. Statistical Analysis All statistical analyses were performed with a statistical software program (PASW Statistics 17.0, SPSS Inc, Chicago, Illinois). For the comparisons, a P value,.05 was considered statistically significant. Cross-tabulations were generated and P values for categoric parameters were obtained by Pearson x 2 test, for ordered categories by linear-by-linear association x 2 test and for the numeric data by Student t test. Kaplan-Meier survival curves were generated for analyzing survival by using the logrank test. Two cases with another cause of death and one case with an unknown cause of death were censored. For ordered categories, a log-rank test for trend was used. A univariate and multivariate Cox proportional hazards model was used to determine independent prognostic factors. For Cox multivariate analysis, we used the backward stepwise method based on Wald statistic. Sensitivity, specificity, likelihood ratios, and predictive values were assessed for all clinical and pathologic factors. Death due to metastasis was used as the endpoint. RESULTS As the material studied had been obtained from enucleated eyes, the size of the tumors was relatively large: according to the TNM staging, 5 tumors (10%) belonged to stage I, 10 (20%) to stage IIA, 15 (30%) to stage IIB, 17 (34%) to stage IIIA, and 3 (6%) to stage IIIB. The 5-year survival rates of these stages were 100%, 100%, 67%, 29%, and 33%, respectively, with 19 of the 50 patients (38%) dying from metastases and 3 patients (6%) from other or unknown causes. Chromosome 3 status had been assessed by karyotyping and FISH on cultured cells and by the FISH technique on nuclei isolated from paraffin-embedded tissue. Standard cytogenetic testing had shown the presence of monosomy 3 in 19 cases (38%), whereas 29 cases (58%) were found to be diploid for chromosome 3 and 2 tumors (4% of cases, excluded in statistical analysis) showed 3 copies of chromosome 3. Fluorescence in situ hybridization on isolated nuclei at a cutoff level of 5% (based on healthy controls) had demonstratedmonosomy3in31of50tumors(62%).the presence of monosomy 3 was compared with clinical and histologic parameters. There were significant associations related to ciliary body involvement, the presence of epithelioid cell type, the TNM stages, and the presence of metastasis (Table 1). To determine whether in tumors in which not all cells have monosomy 3, loss of one chromosome 3 in only a fraction of tumor cells was associated with the development of metastases, we compared the percentages of nuclei, displaying one signal, to the development of metastases within 5 years of follow-up (Figure 1). Patients who developed metastases had significantly (P,.001) higher numbers of monosomy 3 positive nuclei. Cox univariate proportional hazards testing showed that the presence of monosomy 3, as determined by either technique, was significantly associated with death due to Arch Pathol Lab Med Vol 135, August 2011 Heterogeneity for Chromosome 3 and Survival Bronkhorst et al 1043

3 Table 1. Distribution of Chromosome 3 Status and Clinical and Pathologic Features of 50 Cases of Uveal Melanoma a Clinicopathologic Factors Total No. % Disomy (n = 29) Standard Cytogenetic Testing b Monosomy Disomy (n = 19) P Value d (n = 19) FISH on Isolated Nuclei c Monosomy (n = 31) P Value d Age at diagnosis, y (SD) (15) 62 (16) (15) 63 (616).09 Sex M F Eye R L Tumor diameter, mm (SD) (3) 13 (3) (3) 14 (3).13 Tumor thickness, mm (SD) 50 7 (3) 8 (2).08 8 (2) 7 (3).70 Histopathologic cell type Spindle Mixed Epithelioid Ciliary body involvement Not present Present Scleral invasion None Superficial Deep Extrascleral TNM stage Stage I Stage IIA Stage IIB Stage IIIA Stage IIIB Metastasis e.005,.001 No Yes Abbreviation: FISH, fluorescence in situ hybridization. a Data are mean (SD) for the numeric parameters. For other variables, data are the number of cases with monosomy 3 or disomy 3. b By combining karyotyping with FISH data on the cultured cells in all 50 cases, 2 tumors (4% of cases) showed 3 copies of chromosome 3 and were excluded in statistical analysis. c Chromosome 3 analysis by FISH on nuclei isolated from paraffin-embedded tissue, using a cutoff value of 5%. d P values for categoric parameters were obtained by Pearson x 2 test, for ordered categories by linear-by-linear association x 2 test, and for the numeric data, by Student t test. P values #.05 are shown in italics. e Two cases with another cause of death and one case with an unknown cause of death were excluded in statistical analysis. Figure 1. Schematic representation of the results of fluorescence in situ hybridization (FISH) analysis of chromosome 3 in a group of 50 patients, divided according to the presence or absence of death due to metastases at 5 years follow-up. metastasis. Of the 19 patients with less than 5% cells with monosomy 3, one had died of metastases, whereas of the 31 patients with greater than 5% monosomic nuclei, 18 died. To determine whether there was a specific number of nuclei carrying monosomy 3 that was indicative of death due to metastases, we compared 3 groups, using a cutoff level of 5%, 30%, and of 50% positive nuclei. A cutoff value of 30% was the best prognostic factor with the highest hazard ratio (hazard ratio, 28.0; P 5.001). Kaplan-Meier analysis showed similar results (Figure 2). When monosomy 3 was determined by FISH on isolated nuclei with a cutoff level of 30% and all significant prognostic variables from Table 2 were entered into a Cox multivariate model, the analysis showed monosomy 3 (hazard ratio, 44.1; P 5.001) and the largest basal diameter (hazard ratio, 1.6; P 5.001) as the most significant prognostic factors. P values and hazard ratios can be found in Tables 2 and 3. Sensitivity, specificity, likelihood ratios, and predictive values were calculated for each clinical, pathologic, and molecular prognostic factor (data not shown). In this analysis, the metastatic group consisted of all 19 patients 1044 Arch Pathol Lab Med Vol 135, August 2011 Heterogeneity for Chromosome 3 and Survival Bronkhorst et al

4 Figure 2. Kaplan-Meier survival curves and log-rank analysis of 50 patients with uveal melanoma at 5 years follow-up of all cases (A), according to TNM (tumor-node-metastasis) stages (B), monosomy 3 as assessed by standard cytogenetic testing on cultured cells (C), and fluorescence in situ hybridization (FISH) on isolated nuclei for the threshold values set at 5%, 30%, and 50% (D through F). The numbers in the tables represent the number of patients at risk. Arch Pathol Lab Med Vol 135, August 2011 Heterogeneity for Chromosome 3 and Survival Bronkhorst et al 1045

5 Table 2. Chromosomal Aberration: Monosomy 3 Summary of Survival With Death Due to Metastasis at 5 Years Follow-up as the Endpoint a Kaplan-Meier (Log-Rank Test) Cox Univariate Categories x 2 P Value Categories P Value HR 95% CI Standard cytogenetic testing Disomy/monosomy Disomy b /monosomy FISH on isolated nuclei Cutoff.5% Disomy/monosomy 13.7,.001 Disomy b /monosomy Cutoff.30% Disomy/monosomy 23.8,.001 Disomy b /monosomy Cutoff.50% Disomy/monosomy 19.9,.001 Disomy b /monosomy, Age, y c For each year Sex M/F M b /F Cell type Spindle/mixed Spindle b /mixed epithelioid epithelioid Tumor thickness, mm c For each mm Largest tumor diameter, mm c For each mm, Ciliary body involvement Not present/present Not present b /present Scleral in-growth Not present/present Not present b /present , Extrascleral growth Not present/present Not present b /present TNM stage d I and IIA/IIB/IIIA/IIIB 14.1,.001 For each category, Abbreviations: 95% CI, 95% confidence intervals; FISH, fluorescence in situ hybridization; HR, hazard ratio. a Chromosome 3 analysis was performed by karyotyping and fluorescence in situ hybridization (FISH) on cultured uveal melanoma cells obtained after enucleation, and by FISH on nuclei isolated from paraffin-embedded tissue; P,.05, shown in italic. b Reference group. c For numeric variables, the HR is the HR for each unit (mm or y) for the parameter. For log-rank testing, these parameters were divided into categoric variables as follows: the medians for tumor prominence and largest tumor diameters were 8.0 and 13.0, respectively. For age stratification, the mean age at enucleation was 60 y. d For TNM (tumor-node-metastasis) stages, log-rank test for trends was used. who had developed metastasis, and the nonmetastatic group contained 28 patients without metastasis, whereas 3 patients were excluded for other and unknown causes of death within 5-year follow-up after enucleation. Monosomy 3 determined on isolated nuclei with a threshold of 30% outperformed all other prognostic variables, with 95% sensitivity, 79% specificity, and positive and negative likelihood ratios of 4.4 and 0.1, respectively. The other accurate predictive factors were the presence of a mixed or epithelioid cell type (89% sensitivity, 43% specificity), ciliary body involvement (68% sensitivity, 75% specificity), and largest tumor diameter (68% sensitivity, 75% specificity). COMMENT We previously showed 7 that uveal melanoma can be heterogeneous for the number of copies of chromosome 3, which was subsequently confirmed by other studies. 8,19,20 Chromosomal heterogeneity can contribute to the metastatic potential of these tumors. We wanted to determine whether it was necessary for all cells to show monosomy 3 or if only a small subset of cells with monosomy 3 was enough for a bad prognosis. We observed that the presence of as few as 5% cells with monosomy 3 puts the patient at risk for the development of metastases. The percentage of nuclei with one chromosome 3 that is set as the threshold for calling the tumor monosomic for chromosome 3 is somewhat arbitrary and may lead to false-negative or false-positive results. We determined the predicative accuracy of this prognostic feature at different thresholds. The monosomy 3 classifier based on FISH on isolated nuclei with a threshold value of 30% predicted metastasis more accurately than clinicopathologic features and monosomy 3 assessed with other threshold values or standard cytogenetic testing. Some groups suggest 21,22 that monosomy 3 is a primary event in the developmental pathway of high-risk uveal melanoma, and it may start as a monoclonal event, whereas over time, more clones with monosomy 3 can develop, or the primary clone may show a growth advantage compared with other cells. Although the sensitivity of the nuclei test at a threshold of as little as 5% was high (95%), the low specificity (64%) can be explained by a small number of cells with monosomy 3 being easily missed by taking a single sample if there is only one small clone present. Many techniques are available for detecting monosomy 3, and therefore, the techniques used may play a role in differentiating patients with uveal melanoma who have favorable and adverse prognoses. Cytogenetic analysis Table 3. Independent Predictive Factors for Metastatic Death, Identified by Cox Multivariate Analysis a Multivariate Analysis B P Value Hazard Ratio 95% CI Largest tumor diameter For each mm Ciliary body involvement Epithelioid cell type TNM stage For each stage Monosomy 3 (fluorescence in situ hybridization nuclei.30%) Abbreviations: 95% CI, 95% confidence intervals; B, regression conflict. a P values,.05 are shown in italics Arch Pathol Lab Med Vol 135, August 2011 Heterogeneity for Chromosome 3 and Survival Bronkhorst et al

6 requires the short-term culture of tumors followed by producing good-quality metaphase spreads, which could also induce in vitro genetic changes, and only a small number of cells can be studied. The development of FISH allows the interphase cytogenetic analysis of either fresh or archival tumor tissue not only quickly but also accurately in a large number of cells. 23 Patel et al 24 showed that FISH analysis is a reliable and efficient technique in a large number of cells of fresh-frozen uveal melanoma and suggested it is important to consider at removal what level of genetic imbalance is important in determining prognosis; it is possible that even patients with minimal genetic imbalance will ultimately develop metastatic disease. The difference in levels of monosomy 3 detected by FISH between isolated nuclei and cultured nuclei may be based on the small cultured part of the tumor we look at. In many cases, the tumor specimen probably did not contain the monosomy 3 clone, which was definitively present in a different part of the tumor as shown by FISH on isolated nuclei. The cutoff rate of 5% as well as 30% for heterogeneity was shown to be a strong single predictor of prognosis at 5 years follow-up. As can be seen in Figure 1, the absence of any cells with monosomy 3 was significantly associated with a favorable prognosis and the low cutoff level of 5% for detecting monosomy 3 described in our previous report showed a significant high hazard ratio (15.5). However, because 6 patients who had at least 30% of monosomy 3 were still alive after 5 years, we have to take into account that metastases can occur many years after enucleation, 25 and we will, therefore, have to evaluate these data again at 10 years follow-up. Besides monosomy 3, other defined abnormalities of chromosomes 1, 6, and 8 have been associated with metastatic death in uveal melanoma. Damato et al 26 showed that the combination of monosomy 3 and polysomy 8 determined by FISH is highly predictive of metastatic uveal melanoma at 7 years follow-up. Dopierala et al 11 determined intratumor heterogeneity of chromosomal abnormalities of 1p, 3, 6, and 8 in uveal melanoma by MLPA and showed that this heterogeneity caused equivocal MLPA results, which could be caused by disomy 3 cells diluting the DNA of the monosomy 3 cell clones. Lake et al 27 also used this MLPA method to detect deletions of chromosome 3 and a single nucleotide polymorphism array to detect specific gene deletions and loss of heterozygosity in rare metastasizing disomy 3 cases of uveal melanoma, which were undetectable by FISH. In such cases, instead of loss of a single copy of chromosome 3, specific genes were found to be deleted on chromosome 3, which are essential for an early progression to metastasis. Therefore, it would be interesting to perform MLPA and single nucleotide polymorphism array on the samples from the only patient with disomy of chromosome 3 who died from metastases in our study and to examine abnormalities in chromosomes 1, 6, and 8 on the isolated nuclei as well. During tumor progression, tumor cells acquire various mutations, which generate clonal outgrowth and heterogeneity. Clonal advantages can be related to as-yet to be determined genes present on chromosome 3, associated with, for example, growth stimulation or escape from immune elimination. Therefore, intratumoral heterogeneity is an important issue, and we show here that tumors with only a low number of cells showing monosomy 3 already put the patient at risk for lethal metastatic disease within 5 years follow-up. This study was supported by the Board of Directors, Leiden University Medical Center (Dr Bronkhorst, PhD, position recipient), Stichting Nederlands Oogheelkundig Onderzoek, Rotterdamse Vereniging Blindenbelangen, Landelijke Stichting voor Blinden en Slechtzienden, and by grant RUL from the Dutch Cancer Society (KWF). We would also like to thank Geert Haasnoot, BSc, statistician, for his help with the statistical analysis. References 1. Kivela T, Eskelin S, Kujala E. Metastatic uveal melanoma. Int Ophthalmol Clin. 2006;46(1): Mooy CM, De Jong PT. Prognostic parameters in uveal melanoma: a review. Surv Ophthalmol. 1996;41(3): Maat W, Ly LV, Jordanova ES, Wolff-Rouendaal D, Schalij-Delfos NE, Jager MJ. Monosomy of chromosome 3 and an inflammatory phenotype occur together in uveal melanoma. Invest Ophthal Vis Sci 2008;49(2): Horsman DE, Sroka H, Rootman J, White VA. Monosomy 3 and isochromosome 8q in a uveal melanoma. Cancer Genet Cytogenet 1990;45(2): Prescher G, Bornfeld N, Becher R. Nonrandom chromosomal abnormalities in primary uveal melanoma. J Natl Cancer Inst. 1990;82(22): Sisley K, Rennie IG, Cottam DW, Potter AM, Potter CW, Rees RC. Cytogenetic findings in six posterior uveal melanomas: involvement of chromosomes 3, 6, and 8. Genes Chromosomes Cancer 1990;2(3): Maat W, Jordanova ES, van Zelderen-Bhola SL, et al. The heterogeneous distribution of monosomy 3 in uveal melanomas: implications for prognostication based on fine-needle aspiration biopsies. Arch Pathol Lab Med. 2007;131(1): Mensink HW, Vaarwater J, Kilic E, et al. Chromosome 3 intratumor heterogeneity in uveal melanoma. Invest Ophthalmol Vis Sci. 2009;50(2): Sandinha T, Farquharson M, McKay I, Roberts F. Correlation of heterogeneity for chromosome 3 copy number with cell type in choroidal melanoma of mixedcell type. Invest Ophthalmol Vis Sci. 2006;47(12): Schoenfield L, Pettay J, Tubbs RR, Singh AD. Variation of monosomy 3 status within uveal melanoma. Arch Pathol Lab Med. 2009;133(8): Dopierala J, Damato BE, Lake SL, Taktak AF, Coupland SE. Genetic heterogeneity in uveal melanoma assessed by multiplex ligation-dependent probe amplification. Invest Ophthalmol Vis Sci. 2010;51(10): McLean IW, Foster WD, Zimmerman LE, Gamel JW. Modifications of Callender s classification of uveal melanoma at the Armed Forces Institute of Pathology. Am J Ophthalmol. 1983;96(4): Finger PT. The 7th edition AJCC staging system for eye cancer: an international language for ophthalmic oncology. Arch Pathol Lab Med. 2009;133(8): Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A. American Joint Committee on Cancer, Cancer Staging Manual. 7th ed. New York, NY: Springer; 2010: Dauwerse JG, Jumelet EA, Wessels JW, et al. Extensive cross-homology between the long and the short arm of chromosome 16 may explain leukemic inversions and translocations. Blood 1992;79(5): Haralambieva E, Kleiverda K, Mason DY, Schuuring E, Kluin PM. Detection of three common translocation breakpoints in non-hodgkin s lymphomas by fluorescence in situ hybridization on routine paraffin-embedded tissue sections. J Pathol. 2002;198(2): Jordanova ES, Riemersma SA, Philippo K, Giphart-Gassler M, Schuuring E, Kluin PM. Hemizygous deletions in the HLA region account for loss of heterozygosity in the majority of diffuse large B-cell lymphomas of the testis and the central nervous system. Genes Chromosomes Cancer. 2002;35(1): Bronkhorst IH, Ly LV, Jordanova ES, et al. Detection of M2 macrophages in uveal melanoma and relation with survival [published online ahead of print September 1, 2010]. Invest Ophthalmol Vis Sci. doi: /iovs Meir T, Zeschnigk M, Masshofer L, Pe er J, Chowers I. The spatial distribution of monosomy 3 and network vasculogenic mimicry patterns in uveal melanoma. Invest Ophthalmol Vis Sci. 2007;48(5): Worley LA, Onken MD, Person E, et al. Transcriptomic versus chromosomal prognostic markers and clinical outcome in uveal melanoma. Clin Cancer Res. 2007;13(5): Prescher G, Bornfeld N, Friedrichs W, Seeber S, Becher R. Cytogenetics of twelve cases of uveal melanoma and patterns of nonrandom anomalies and isochromosome formation. Cancer Genet Cytogenet 1995;80(1): Hoglund M, Gisselsson D, Hansen GB et al. Dissecting karyotypic patterns in malignant melanomas: temporal clustering of losses and gains in melanoma karyotypic evolution. Int J Cancer 2004;108(1): Kallioniemi A, Visakorpi T, Karhu R, Pinkel D, Kallioniemi OP. Gene copy number analysis by fluorescence in situ hybridization and comparative genomic hybridization. Methods 1996;9(1): Patel KA, Edmondson ND, Talbot F, Parsons MA, Rennie IG, Sisley K. Prediction of prognosis in patients with uveal melanoma using fluorescence in situ hybridisation. Br J Ophthalmol. 2001;85(12): Ly LV, Odish OF, Wolff-Rouendaal D, Missotten GS, Luyten GP, Jager MJ. Intravascular presence of tumor cells as prognostic parameter in uveal melanoma: a 35-year survey. Invest Ophthalmol Vis Sci. 2010;51(2): Damato B, Duke C, Coupland SE, et al. Cytogenetics of uveal melanoma: a 7-year clinical experience. Ophthalmology 2007;114(10): Lake SL, Coupland SE, Taktak AF, Damato BE. Whole-genome microarray detects deletions and loss of heterozygosity of chromosome 3 occurring exclusively in metastasizing uveal melanoma. Invest Ophthalmol Vis Sci. 2010;51(10): Arch Pathol Lab Med Vol 135, August 2011 Heterogeneity for Chromosome 3 and Survival Bronkhorst et al 1047