Reduced FANCD2 influences spontaneous SCE and RAD51 foci formation in uveal melanoma and Fanconi anaemia.

Uveal melanoma (UM) is unique among cancers in displaying reduced endogenous levels of sister chromatid exchange (SCE). Here we demonstrate that FANCD2 expression is reduced in UM and that ectopic expression of FANCD2 increased SCE. Similarly, FANCD2-deficient fibroblasts (PD20) derived from Fanconi anaemia patients displayed reduced spontaneous SCE formation relative to their FANCD2-complemented counterparts, suggesting that this observation is not specific to UM. In addition, spontaneous RAD51 foci were reduced in UM and PD20 cells compared with FANCD2-proficient cells. This is consistent with a model where spontaneous SCEs are the end product of endogenous recombination events and implicates FANCD2 in the promotion of recombination-mediated repair of endogenous DNA damage and in SCE formation during normal DNA replication. In both UM and PD20 cells, low SCE was reversed by inhibiting DNA-PKcs (DNA-dependent protein kinase, catalytic subunit). Finally, we demonstrate that both PD20 and UM are sensitive to acetaldehyde, supporting a role for FANCD2 in repair of lesions induced by such endogenous metabolites. Together, these data suggest FANCD2 may promote spontaneous SCE by influencing which double-strand break repair pathway predominates during normal S-phase progression.

Aldehyde dehydrogenase activity selects for the holoclone phenotype in prostate cancer cells.

Aldehyde dehydrogenase 1 (ALDH) activity is considered to be a marker of cancer stem cells (CSCs) in many tumour models, since these cells are more proliferative and tumourigenic than ALDH(Lo) cells in experimental models. However it is unclear whether all CSC-like cells are within the ALDH(Hi) population, or whether all ALDH(Hi) cells are highly proliferative and tumourigenic. The ability to establish a stem cell hierarchy in vitro, whereby sub-populations of cells have differing proliferative and differentiation capacities, is an alternate indication of the presence of stem cell-like populations within cell lines. In this study, we have examined the interaction between ALDH status and the ability to establish a stem cell hierarchy in PC3 prostate cancer cells. We demonstrate that PC3 cells contain a stem cell hierarchy, and isolation of ALDH(Hi) cells enriches for the most primitive holoclone population, however holoclone formation is not restricted to ALDH(Hi) cells. In addition, we show that ALDH activity undergoes phenotypic plasticity, since the ALDH(Lo) population can develop ALDH(Hi) populations comparable to parental cells within 2 weeks in culture. Furthermore, we show that the majority of ALDH(Hi) cells are found within the least primitive paraclone population, which is circumvented by culturing PC3 cells as spheroids in defined medium favouring stem cell characteristics. Although ALDH(Hi) status enriches for holoclone formation, this activity may be mediated by a minority of ALDH(Hi) cells.

Resistance of uveal melanoma to the interstrand cross-linking agent mitomycin C is associated with reduced expression of CYP450R.

BACKGROUND: Uveal melanoma (UM) is the most common primary intraocular tumour of adults, frequently metastasising to the liver. Hepatic metastases are difficult to treat and are mainly unresponsive to chemotherapy. To investigate why UM are so chemo-resistant we explored the effect of interstrand cross-linking agents mitomycin C (MMC) and cisplatin in comparison with hydroxyurea (HU). METHODS: Sensitivity to MMC, cisplatin and HU was tested in established UM cell lines using clonogenic assays. The response of UM to MMC was confirmed in MTT assays using short-term cultures of primary UM. The expression of cytochrome P450 reductase (CYP450R) was analysed by western blotting, and DNA cross-linking was assessed using COMET analysis supported by γ-H2AX foci formation. RESULTS: Both established cell lines and primary cultures of UM were resistant to the cross-linking agent MMC (in each case P<0.001 in Student's t-test compared with controls). In two established UM cell lines, DNA cross-link damage was not induced by MMC (in both cases P<0.05 in Students's t-test compared with damage induced in controls). In all, 6 out of 6 UMs tested displayed reduced expression of the metabolising enzyme CYP450R and transient expression of CYP450R increased MMC sensitivity of UM. CONCLUSION: We suggest that reduced expression of CYP450R is responsible for MMC resistance of UM, through a lack of bioactivation, which can be reversed by complementing UM cell lines with CYP450R.

The in vivo modulatory effects of an anterior-chamber microenvironment on uveal melanoma.

BACKGROUND: Primary melanoma of the iris, for reasons unknown has a lower metastatic rate compared with primary ciliary-body melanoma. Six histology cases of ciliary-body melanoma were identified that had spread onto the iris surface and into the stroma, representing a change in tumour microenvironment from aqueous humour non-exposure (ciliary-body component) to aqueous humour exposure (iris surface component). This provided an ideal paradigm for investigating the effects of different environments on melanoma. METHOD: Conventional light microscopy was performed on stained paraffin sections of the identified cases, followed by immunohistochemistry to cell cycle proteins p27 and Cyclin D1. Fluorescence in situ hybridisation (FISH) analysis was conducted on the paraffin sections for changes of chromosomes 3 and 8, associated with poor uveal melanoma prognosis. RESULTS: Iris surface melanoma cells were smaller compared with the adjacent deeper iris stromal melanoma cells and with those in the ciliary body. Fewer iris surface melanoma cells expressed Cyclin D1 protein, but more expressed p27 protein, compared with the larger iris stromal melanoma cells (paired Wilcoxon signed ranks test: Cyclin D1 p = 0.028; p27 p = 0.046) and with the ciliary-body melanoma cells (paired Wilcoxon signed ranks test: Cyclin D1 p = 0.028; p27 p = 0.028). With FISH, chromosome 3 and 8 alterations were less common among the iris surface melanoma cells than the deeper iris stromal melanoma cells and the ciliary-body melanoma cells, which were consistently characterised by a relative genetic imbalance for chromosomes 3 and 8. CONCLUSIONS: These data suggest that there are tumour-modulatory factors within the anterior chamber environment that probably select populations of ciliary-body melanoma cells, with a less aggressive, better-differentiated status. Furthermore, it may help explain why iris melanomas generally have a less aggressive course than ciliary-body and choroidal melanomas.

What is the significance of vortex vein invasion in uveal melanoma?

PURPOSE: To correlate vortex vein invasion with established prognostic factors for uveal melanoma. METHODS: Enucleated eyes with a confirmed histopathological diagnosis of uveal melanoma with vortex vein invasion were identified, over a 10-year period. Established uveal melanoma prognostic factors, with tumour genetics were correlated with vortex vein invasion and patient survival. RESULTS: Microscopic vortex vein involvement was present in 29 of 244 (11.9%) uveal melanomas. Of 29, 6 (20.7%) tumours had macroscopic evidence of vortex vein invasion. Of 29, 14 (48.3%) tumours also showed evidence of non-vortex vein, 'direct' scleral invasion. 23 (79.3%) of 29 melanomas involved only the choroid. The mean maximum diameter of tumours with vortex vein invasion was 15.8 mm and the mean thickness was 9.7 mm. The uveal melanoma was a discrete nodule in 27 of 29 (93.1%) cases. Histologically, 8 of 29 tumours (27.6%) were spindle cell, 19 of 29 (65.5%) were mixed cell, and 2 of 29 (6.9%) were epithelioid cell type. Of 29, 22 (75.9%) uveal melanomas with vortex vein invasion contained extracellular matrix networks and loops. Genetic abnormalities correlated with poor prognosis were seen in 25 of 29 (86.2%) tumours with vortex vein invasion. Liver metastasis was confirmed in 19 of 29 (65.5%) patients with vortex vein invasion. No patients with uveal melanomas showing vortex vein invasion suffered orbital recurrence of disease following enucleation. CONCLUSIONS: The trends show that vortex vein invasion is associated with a choroidal location, large tumour size, spindle cell bias, presence of extracellular matrix loops/networks and genetic markers. A higher proportion of patients with vortex vein invasion progress to develop liver metastasis compared with the general uveal melanoma population.

Multiple locations on chromosome 3 are the targets of specific deletions in uveal melanoma.

PURPOSE: Loss of chromosome 3 is a frequent event in uveal melanomas, which is associated with hepatic metastases and a poor prognosis. The entire copy of chromosome 3 is usually lost (monosomy 3); however, a small subset of tumours demonstrate partial deletions of chromosome 3. Analysis of these tumours may allow the identification of tumour suppressor genes (TSGs) that are the molecular target of monosomy 3. Therefore, the purpose of this investigation was to determine the location of these partial deletions of chromosome 3 in uveal melanomas. METHODS: Microsatellite analysis and restriction fragment-length polymorphism analysis were performed on 52 primary uveal melanomas using 19 markers located on both arms of chromosome 3. Cytogenetic analysis and fluorescence in situ hybridisation were performed, where possible, to confirm molecular findings. RESULTS: Of 52 tumours studied, five tumours (10%) demonstrated LOH at one or more informative markers, but retention of heterozygosity was observed at other loci on chromosome 3, consistent with the presence of structural abnormalities to chromosome 3. Consistent with previous findings, the pattern of LOH in these tumours indicates the presence of deletions around 3p25-26 and on 3q, and that a new target region at 3p11-14 is preferentially deleted. CONCLUSIONS: These results indicate the presence of several tumour suppressor loci on chromosome 3 and support the notion that the high rate of monosomy 3 in uveal melanoma is driven by disruption of several TSGs located on both arms of chromosome 3.

Invasive and noninvasive uveal melanomas have different adhesive properties.

AIMS: To establish if invasive and noninvasive uveal melanomas have differences in expression of adhesion molecules, and whether their adhesive interactions with the extracellular matrix (ECM) and endothelium vary. METHODS: Cells from an invasive and noninvasive uveal melanoma cell line and hepatic and dermal microvascular endothelial cells were assessed by flow cytometry for adhesion molecule expression. Tumour cell adhesion to ECM substrates (collagens I and IV, fibronectin, laminin, and vitronectin) and endothelial cells was also investigated using a commercially available assay or a fluorescence-based in vitro assay, respectively. The significance of results comparing cell lines was determined using a Student's t-test, whereby P-values of less than 0.05 were taken as significant. RESULTS: alpha1- and alpha4-integrins were not expressed by noninvasive cells, but were detected on invasive cells. The invasive cell line also expressed higher levels of other integrins than the noninvasive line. Correspondingly, invasive cells adhered in higher numbers to ECM substrates and endothelial cells, and for the latter, the difference was highly significant (P<0.001). No preference in adhesion of invasive cells for the hepatic endothelium was observed. CONCLUSIONS: Successful attachment to and migration through the ECM, basement membrane, and endothelium are vital processes involved in malignant progression. Differential expression of alpha1- and alpha4-integrins by invasive and noninvasive cells infers a role for these receptors in invasion, while the ability of invasive cells to adhere more efficiently to the endothelium suggests that this is a critical factor in uveal melanoma invasion.

CD80-mediated induction of immunostimulation in two ocular melanoma cell lines is augmented by interferon-gamma.

Although the transfection of the T-cell costimulatory molecule CD80 cDNA into human tumours can augment their immunogenicity in vitro, its expression alone is ineffective in many tumour systems. We evaluated the influence of CD80 expression on the immunostimulatory activity of ocular melanoma cell lines and determined whether IFN-gamma could enhance the effect. Two ocular melanoma cell lines were transfected with CD80 cDNA. The immunostimulatory capacity of the CD80+ transfectants was determined by their ability to stimulate the proliferation of allogeneic peripheral blood mononuclear cells (PBMC). The influence of additional accessory molecules on PBMC proliferation was assessed by pre-treating the CD80 transfectants with IFN-gamma. The CD80+ transfectants induced proliferation of allogeneic PBMC. IFN-gamma treatment of the tumour cells induced upregulated expression of MHC class I, de novo expression of MHC class II and CD54, and enhanced the ability of the CD80+ transfectants to stimulate PBMC proliferation. CD4+ T cells were not required for the proliferative response against untreated CD80+ tumour cells but were necessary for the augmentation of proliferation observed following IFN-gamma treatment. CD80+ ocular melanoma cells possess immunostimulatory potential which is augmented by IFN-gamma induced upregulation of cell surface molecules. Further studies on the role of costimulatory molecules in inducing anti-tumour immunity in ocular melanoma may help to define new strategies for application of immunotherapeutic approaches to treat this aggressive disease.

Prediction of prognosis in patients with uveal melanoma using fluorescence in situ hybridisation.

BACKGROUND/AIMS: Uveal melanoma is the commonest malignancy of the eye, with a high proportion of patients dying of metastatic disease. Tumours showing a loss of chromosome 3 and gains of chromosome 8 are associated with a worse prognosis. The efficiency of fluorescence in situ hybridisation (FISH) in determining copy numbers of these chromosomes was assessed in individual tumours and related to patient survival. METHODS: 33 fresh frozen samples were analysed with centromeric probes for chromosomes 3 and 8. Patient outcomes were divided into two groups: (1) absence of genetic abnormalities (no genetic imbalance) and (2) presence of genetic abnormalities (genetic imbalance). The log rank test was used to compare survival, which was represented by Kaplan-Meier survival curves. RESULTS: Of the 33 tumours analysed, 16 showed evidence of genetic imbalances. Of these 16 tumours, 14 patients had died by the end of the study, with 10 having died of liver metastases. Of the tumours without evidence of genetic imbalances, five patients had died by the end of the study, although none had died as a result of either liver metastases or from the primary uveal melanoma. The difference in survival between the two groups was highly significant (p<0.0001). CONCLUSION: The authors have shown that FISH analysis for chromosome 3 and 8 is a reliable and efficient technique in the analysis of fresh frozen tumour specimens and is valuable in the prediction of prognosis in individuals with uveal melanomas.

Expression of integrins, degradative enzymes and their inhibitors in uveal melanoma: differences between in vitro and in vivo expression.

Posterior uveal melanoma is the most common intraocular malignancy in adults. Metastasis occurs in approximately 40% of all cases and spread is primarily to the liver. Once secondary hepatic disease has developed the prognosis is poor. Metastasis involves a series of adhesion and de-adhesion events, coupled with regulated tissue degradation to facilitate tumour cell invasion and spread to both local and distant sites. These processes are assisted by the expression of integrins and degradative enzymes by both tumour and host cells. Using a series of 10 uveal melanomas, we investigated the expression of a panel of integrins, degradative enzymes and their inhibitors that have been shown to be associated with metastasis. In addition, we undertook to establish if there might be differential expression in response to growth under artificial conditions. All the tumours expressed matrix metalloproteinases (MMP)-2 and-9, tissue inhibitor of metalloproteases (TIMP)-2, urokinase plasminogen activator (u-PA), plasminogen activator inhibitor (PAI)-1 and PAI-2. Differences in the expression of the integrins alpha1beta1, alpha2beta1 and alpha6beta1 were observed; in particular, these differences appeared to relate to expression as a consequence of growth in culture. In summary, uveal melanoma cells express both degradative enzymes and their respective inhibitors, which are important in metastasis. It would appear that differential expression of integrins is present, probably as a response to in vitro stimulation.

Reduced expression of TAP-1 and TAP-2 in posterior uveal melanoma is associated with progression to metastatic disease.

In order to determine the effects of the loss or reduced expression of molecules associated with antigen presentation (transporter associated with antigen presentation [TAP]-1, TAP-2, low molecular weight protein [LMP]-2 and LMP-7), we examined the expression of these molecules in primary uveal melanoma lesions. Paraffin-embedded sections from 29 primary uveal melanoma lesions were analysed for expression of TAP-1, TAP-2, LMP-2 and LMP-7 using specific primary antibodies followed by a three-stage immunoperoxidase technique. Microscopic examination was undertaken to determine differences in expression of these molecules on the tumour and the surrounding stroma. Overall, 72% (21 out of 29) of the tumours showed some loss or reduced expression of TAP-1, TAP-2, LMP-2 and/or LMP-7. Statistical analysis of these results showed that progression to metastatic disease was strongly associated with reduced expression of TAP-1 (P < 0.05) and TAP-2 (P < 0.01), taking patient age, tumour site and histology into account. We conclude that the reduced expression of molecules important in eliciting an immune response, such as TAP-1 and TAP-2, may facilitate the metastatic spread of uveal melanoma lesions and may have important implications for prospective immunotherapy.

A comparison of ocular melanocyte and uveal melanoma cell invasion and the implication of alpha1beta1, alpha4beta1 and alpha6beta1 integrins.

BACKGROUND/AIMS: Posterior uveal melanoma is the most common intraocular tumour in adults, responsible for the death of approximately 35% of patients. Hepatic metastases are most frequent, and once diagnosed survival is usually less than 1 year. The beta1 family of integrins, alphavbeta3 and MMP-2 and MMP-9 have been implicated in the metastasis of several types of tumour. To study their involvement in uveal melanoma we analysed the expression of the beta1 integrins, alphavbeta3, MMP-2, and MMP-9 in 10 primary posterior uveal melanomas, and correlated expression with invasive potential in vitro. Comparable studies were undertaken on cultures of melanocytes. METHODS: Expression of integrins was studied by immunohistochemistry, secretion of MMP-2 and MMP-9 by zymography, and the invasive potential was assessed using a transwell model. RESULTS: MMP-2 was secreted by all uveal melanomas and seven of 10 secreted MMP-9. Among uveal melanoma, invasion levels of 4-25% were observed and the major integrins expressed were alpha1beta1, alpha2beta1, alpha3beta1, alpha5beta1, and avbeta3. Melanocytes did not express alpha1beta1, alpha4beta1, and alpha6beta1. CONCLUSION: The laminin binding alpha6beta1 integrin was not expressed by either melanocytes or tumours with spindle morphology, which are considered to have a better prognosis. It is possible that expression of the alpha6beta1 integrin may prove useful as a prognostic indicator.

Dead or alive: immunogenicity of human melanoma cells when presented by dendritic cells.

Uveal melanoma is an aggressive malignancy with a poor prognosis despite current therapeutic intervention. These tumors have been shown to be antigenic because they express a number of melanoma-associated antigens and are therefore attractive targets for immunotherapy. Here, we investigated the immunogenicity of uveal melanoma cells that have undergone apoptosis and compared this with their necrotic or live counter-parts. The fate of the tumor antigens in these cells largely depends on their ability to be processed and phagocytosed by dendritic cells (DCs). Flow cytometric analysis shows that human DCs form conjugates more efficiently with dead uveal melanoma cells, and consequently these are effective stimuli of lymphocyte proliferation. However, only DCs pulsed with apoptotic cells were able to induce proliferation of CD8+ cytotoxic T cells and stimulate antigen-specific T cells. This study demonstrates for the first time that DCs derived from melanoma patients process and present antigens derived from both HLA-matched or HLA-mismatched human apoptotic tumor cells stimulating both CD4+ and CD8+ T cells. This approach may be important to the development of DC-based immunotherapies for melanoma.

Alpha-melanocyte stimulating hormone can reduce T-cell interaction with melanoma cells in vitro.

This study was undertaken to investigate whether alpha-melanocyte stimulating hormone (alphaMSH) influences the interaction of melanoma cells with T-lymphocytes in the light of previous work from our laboratories showing that alphaMSH can reduce tumour necrosis factor-alpha (TNFalpha) stimulated ICAM-1 upregulation in both normal and transformed melanocytes. Two cutaneous melanoma cell lines--A375-SM and HBL--were examined initially. A375-SM cells gave only a two-fold increase in T-cell proliferation, which was not much improved by the pretreatment of the melanoma cells with cytokines. HBL cells induced a three-fold increase in T-cell proliferation, which was slightly enhanced by the addition of cytokines. Neither cell line expressed B7(1), HBL cells expressed a low level of B7(2), whereas A375-SM cells had little, if any, B7(2) expression. Addition of alphaMSH reduced the interaction between these cutaneous melanoma cells and T-lymphocytes in some, but not all, conditions. An ocular melanoma cell line transfected with B7 showed a modest interaction with T-cells (in two out of three donors) and this response was reduced by the addition of alphaMSH. Pretreatment of the transfected line with cytokines markedly enhanced stimulation of T-cell proliferation by these tumour cells, and alphaMSH reduced the interaction between melanoma cells and T-cells for two out of three donors. In summary, under experimental conditions where melanoma cell stimulation of T-cells occurred (generally pretreatment of the cells with interferon-gamma gave the most convincing response), alphaMSH reduced this response in the majority of experiments, providing preliminary evidence to confirm the hypothesis that MSH may assist melanoma cells to evade interaction with immune cells.

Association of specific chromosome alterations with tumour phenotype in posterior uveal melanoma.

Posterior uveal melanomas have recurrent alterations of chromosomes 1, 3, 6 and 8. In particular, changes of chromosomes 3 and 8 occur in association, appear to characterize those tumours with a ciliary body component, and have been shown to be of prognostic significance. The relevance of other chromosome alterations is less certain. We have performed cytogenetic analysis on 42 previously untreated primary posterior uveal melanomas. Of interest was the observation that as tumour size increased the involvement of specific chromosome changes, and the amount of chromosome abnormalities likewise increased. Loss, or partial deletions, of the short arm of chromosome 1 were found to associate with larger ciliary body melanomas; typically, loss of the short arm resulted from unbalanced translocations, the partners of which varied. Trisomy of chromosome 21 occurred more often in ciliary body melanomas, whilst rearrangements of chromosomes 6 and 11 were primarily related to choroidal melanomas. Our results imply that alterations of chromosome 1 are important in the progression of some uveal melanomas, and that other chromosome abnormalities, besides those of chromosomes 3 and 8, are associated with ocular tumours of particular locations.

Clinical applications of chromosome analysis, from fine needle aspiration biopsies, of posterior uveal melanomas.

PURPOSE: An accurate assessment of prognosis is essential to the clinical assessment of malignancy. In posterior uveal melanoma specific chromosome alterations have been shown to correlate significantly with prognosis; but the procedure is restricted to patients treated surgically, and in consequence has been limited mainly to large tumours. Fine needle aspiration biopsy (FNAB) may provide sufficient material to perform this technique, and allow its use in the in situ assessment of tumours, including small lesions. To determine the feasibility of this approach we have conducted a pilot study using enucleated tumours. METHODS: Ten cases of posterior uveal melanoma were studied. In each instance both a test FNAB and a standard tissue preparation were conducted, and the results compared. FNABS were obtained from enucleated tumours by aspirating cells using a 5 ml syringe with a .25 gauge needle; cells were injected into phosphate-buffered saline, spun down and established in vitro. Conventional short-term cultures were established from tumour tissue samples, which were minced prior to the establishment of cultures. Cytogenetic analysis was performed following standard protocols. RESULTS: Of the 10 cases examined, full chromosome analysis was obtainable from all standard tissue short-term cultures. Cytogenetics was successful from cultures of 6 FNAB, with 2 further FNAB producing partial analyses. No major clonal differences were determined between the two procedures. CONCLUSIONS: Cytogenetic analysis of FNAB appears to be entirely feasible for posterior uveal melanomas, and may permit an accurate in situ assessment of tumours, including small lesions.

Cytogenetics of iris melanomas: disparity with other uveal tract melanomas.

The chromosomal alterations of iris melanomas are poorly characterized, only one report has been detailed. Cytogenetic analysis was performed on the tumors and heparinized blood samples of three patients with iris melanomas; in one case a primary tumor and its related seedling were examined. On analysis of lymphocytes, two of the patients were found to experience a low level fragility of chromosome 9, in the region of a cutaneous melanoma susceptibility gene. All iris melanoma lesions were karyotyped. Clonal abnormalities of chromosomes 3, 5, 6, 7, 8, 9, 12, 15, 17, 18, 19, and Y were found, and in one case a large number of marker chromosomes were observed. No specific chromosomal change was common to the iris melanomas, but two cases had different abnormalities of chromosomes 5 and 18. Variations between the primary tumor and its related seedling were the acquisition of an additional chromosome 15, and a polyploid form of the cell line in the seedling. This study suggests that the most common chromosomal changes of posterior uveal melanomas are less frequent in iris melanomas. Iris melanomas also appear to experience relatively high levels of chromosomal alterations, including the formation of marker chromosomes, which is perhaps reminiscent of cutaneous melanoma.

Abnormalities of chromosomes 3 and 8 in posterior uveal melanoma correlate with prognosis.

Posterior uveal melanomas have nonrandom alterations affecting chromosomes 3, 6, and 8. Loss of chromosome 3 in uveal melanoma has been shown to act as a predictor of disease-free and overall survival. To confirm the significance of chromosome 3 loss and to extend the observations to include those of the associated alterations of chromosome 8, we have conducted a cytogenetic analysis on a series of 42 tumours from patients with primary uveal melanoma who were followed up for a median of 31 months (range = 8-96 months). Abnormalities of chromosomes 3 and 8 were the commonest changes and were confirmed in 10 tumours using fluorescence in situ hybridization. Monosomy of chromosome 3 was found in 21 (50%) of the tumours, and 23 (54%) tumours had additional copies of 8q. Alterations of chromosomes 3 and 8 were found occurring together in 19 (45%) of the tumours and were significantly associated with a ciliary body component (P < 0.0001). Prognostic indicators and changes of chromosomes 3 and 8 were analysed for correlation with patient survival. Of the chosen parameters, only ciliary body involvement (P = 0.003), monosomy of chromosome 3 (P = 0.0007), and additional copies of 8q (P = 0.003) correlated with reduced survival. Evaluation of the dosage effect of additional copies of chromosome arm 8q showed a significant association with reduced survival (P = 0.0001), which was also predictive of a decreased disease-free interval (P = 0.01). Thus, the cytogenetic analysis of uveal melanoma may provide a valuable predictor of prognosis.