Uveal melanoma - Chromosomal aberrations defining poor prognosis

Forskningsårsrapport

Jacob Kinggaard Lilja-Fischer

Faculty of Health Sciences University of Aarhus 2005 Department of Experimental Clinical Oncology, Aarhus University Hospital

1

Forord ............................................................................................................................................3 Dansk resumé................................................................................................................................4 Introduction ..................................................................................................................................5 Background and status................................................................................................................5 Prognostic predictors of poor survival in uveal melanoma ........................................................6 Cytogenetic prognostic factors ...................................................................................................7 Tumours with and without monosomy 3 are different entities...................................................7 Genetic mechanisms of tumour genesis .....................................................................................7 Purpose ..........................................................................................................................................8 Hypotheses.....................................................................................................................................9 Subjects and methods...................................................................................................................9 Patients .......................................................................................................................................9 Quantitative real-time PCR ......................................................................................................10 Assay design .........................................................................................................................10 DNA extraction ....................................................................................................................11 Quantitative real-time PCR procedure .................................................................................11 Comparative Ct-method........................................................................................................12 Fluorescence in situ hybridization (FISH)................................................................................12 Tissue preparation.................................................................................................................13 In situ hybridisation ..............................................................................................................13 Quantification of FISH signals .............................................................................................13 Validation of FISH method ..................................................................................................14 Results..........................................................................................................................................15 Quantitative real-time PCR ......................................................................................................15 FISH .........................................................................................................................................16 Assessment of chromosome index .......................................................................................16 Assessment of relative chromosome index ..........................................................................17 Assessment of monosomy 3 status .......................................................................................17 Relation between monosomy 3 and survival............................................................................18 Relation between monosomy 3 and other patient characteristics.............................................19 Discussion ....................................................................................................................................20 Selection of methods ................................................................................................................20 What is monosomy 3? ..............................................................................................................21 Prognostic implications of monosomy 3 ..................................................................................22 Relation with other prognostic factors......................................................................................23 Limitations................................................................................................................................23 Conclusions .................................................................................................................................23 Perspectives .................................................................................................................................24 Appendixes ..................................................................................................................................25 Figures. .....................................................................................................................................25 Table: Patient characteristics ....................................................................................................28 FISH protocol ...........................................................................................................................29 References....................................................................................................................................30

2

Forord Utroligt mange har været mig behjælpelige i mit arbejde med dette forskningsårsprojekt. Fra Afdeling for Eksperimentel Klinisk Onkologi, Århus Sygehus, Nørrebrogade, skal der lyde en stor tak til Inger-Marie Thuesen, Bente Kirkegaard, Mogens Johansen, Birthe Hermansen og Brita Singers for en stor hjælp med mange opgaver, og for at have oplært mig i laboratoriearbejdet. Og i øvrigt tak til alle for en god og hyggelig arbejdsplads. Fra Cancercytogenetisk Laboratorium, Århus Sygehus, Tage Hansensgade, er jeg Eigil Kjeldsen, Pia B. Kristensen, Bente Madsen og Kirsten Madsen en stor tak skyldig for oplæring og hjælp med FISH analyse. Jeg er Peter Isager taknemmelig for det flotte og grundige forarbejde der udgør fundamentet for dette arbejde, ligesom patientdata udgår fra hans database. Til Jens Overgaard og Jan Alsner skal der lyde en særlig tak for at have taget mig ind, og for at støtte og vejlede mig i projektarbejdet.

3

Dansk resumé BAGGRUND: Uveale melanomer udgår fra øjets iris, corpus ciliare og choroidea, det sidste det hyppigste. Der er i Danmark ca. 50 tilfælde årligt, med en median alder på 64 år. Diagnosen stilles med oftalmoskopi og ultralydsundersøgelse. Tumoren kan bestråles med en lokal, indopereret strålekilde (brachyterapi), ved store tumorer må hele øjet fjernes. Alligevel udvikler omtrent halvdelen af patienterne metastatisk sygdom, idet sygdommen spredes med blodet til især leveren, før behandlingen af den lokale tumor. Behandlingen af metastatisk sygdom er typisk kemoterapi alene eller i kombination med immunterapi. Faktorer som lokalisation, størrelse og celletype giver information om sandsynligheden for at udvikle metastatisk sygdom. For nylig har det dog vist sig, at tumorer med monosomi 3 har en meget høj sandsynlighed for at metastasere, og at tumorer med monosomi 3 sandsynligvis udgør en særskilt biologisk gruppe. FORMÅL: Formålet med dette studie var derfor at undersøge den prognostiske betydning af monosomi 3 i en kohorte af patienter med tilgængeligt histologisk materiale. Endvidere at udvikle en molekylærbiologisk metode til at undersøge for kromosomale forandringer i en større kohorte. METODER: Der var tilgængeligt histologisk materiale fra 64 patienter. Vi udviklede et PCR assay til bestemmelse af monosomi 3 på DNA-materiale. Desuden undersøgte vi tumorerne med Fluorescens in situ Hybridisering (FISH). RESULTATER: Vort PCR assay gav ikke-reproducerbare resultater, og måtte opgives. Det lykkedes at analysere 37 af 64 tumorer (58 %) med FISH. Af disse havde 19 (51 %) tab af et kromosom 3. De 37 patienter overlevede mellem 3 måneder og næsten 30 år, mediant 4,8 år. Der var en signifikant forskel i både samlet og sygdomsspecifik overlevelse, hvor patienter med monosomi 3 havde en højere dødelighed (p=0,04 for begge). Der var en antydning af sammenhæng mellem monosomi 3, og en tumor involverende corpus ciliare (p=0,07). KONKLUSION: Der fandtes en signifikant sammenhæng mellem monosomi 3 og dårlig overlevelse. Der skal dog tages forbehold for det lille patientmateriale.

4

Introduction Background and status. Uveal melanoma arises from melanocytes of the iris, ciliary body and the choroid, and is the most common primary intraocular malignancy.1 The disease may lead to operation, radiotherapy, blindness, metastatic disease, chemotherapy and death.2 The incidence of ocular melanoma (also including conjunctival melanoma) is remarkably stable, in Denmark estimated at 0,78 per 100.000 person years for men and 0,65 for females, which is at level with international investigations.1;3 This is in contrast to cutaneous melanoma, which has shown a remarkable increase in incidence the last fifty years.1 Approximately 50 persons are annually diagnosed with uveal melanoma in Denmark, accounting for 4–5 % of all melanomas. Median age at diagnosis has increased the last fifty years, and is at present 64 years, with males being slightly older at diagnosis than females.1;4 Uveal melanoma is very rarely a consequence of inherited genetic defects.5 Exposition to ultraviolet light is not thought to be a causal factor as in cutaneous melanoma, and in a recent incidence study, sunny Hawaii had an incidence four times lower than that of the United States as a whole3. On the other hand, welding is reported to increases the risk of uveal melanoma.6 Instead the disease might arise from intraocular naevi, which are benign melanocytic lesions and parallels to cutaneous naevi. This is, however, still controversial, as the tissue is not easily accessible for analysis.7 Symptoms experienced by patients with uveal melanoma are in general vague, if any. A number of cases are found at routine examination. Visual impairment and disturbances, as photopsy, might be the presenting symptom. In later stages, pain and inflammation of the eye may present.2;4 Uveal melanoma is primarily diagnosed through anamnesis, clinical exam, ophthalmoscopy and ultrasonography. Occasionally, scleral transillumination, fluorescein angiography or CT- or MR-scan may support the diagnosis.2 The local disease is primarily treated with local radiotherapy (brachytherapy), resection or enucleation. In episcleral brachytherapy, radiation is delivered to the tumour by a small plaque radioactive substance, usually Ruthenium-106 .2;8 A large randomized study has shown equivalent survival after enucleation or brachytherapy treatment for medium-sized uveal melanomas.9

5

If possible, brachytherapy is preferred, as it may conserve vision. Smaller tumours in the iris or the ciliary body may be treated by local resection.2 In some cases, mainly due to large tumour size, enucleation may be necessary.2 Approximately half the patients develop metastatic disease. Uveal melanoma metastasizes haematogenously, primarily to the liver, which is involved in most patients.10 Lungs, bones, skin and lymph nodes are involved to a lesser extent. Metastatic disease may develop late; in fact 10 % of melanoma-related mortality occurs more than 15 years after treatment of the primary tumour.11 Treatment of metastatic disease is typically chemotherapy, alone or in combination with immunotherapy. Intraarterial chemo- and immunotherapy might treat local hepatic disease, and in some cases a solitary liver metastasis can be surgically resected.8;12 In general, the treatment of metastatic uveal melanoma is disappointing, and at best palliative. It is currently an incurable disease.8 The improved treatment of the primary disease has led to no significant improvement of survival rates for patients with uveal melanoma1;13 This is probably due to the fact that the critical step in uveal melanoma, the haematogenous dissemination, is an early event, and thus is not affected by the primary treatment.14 In consequence, the prognosis has not improved significantly for the last 25 – 50 years,1;13 and approximately one half of the patients will currently die from their disease.11;15

Prognostic predictors of poor survival in uveal melanoma A number of clinical and histopathological factors are known to be prognostic in uveal melanoma (See table 1). Factors as higher age and higher clinical stage has long been known to predict a poor prognosis.4 Also, the histopathological cell-type, classified as suggested by Callender in 1931,

Worse prognosis, if

has long been known to be prognostic. The spindle cell

- anterior location

type is the most benign, where as the epithelioid cell

- larger tumour

type is the most malignant, with the mixed cell type 4;16

being of intermediary malignancy.

Other well-

known prognostic factors include tumour-size, with larger tumours having a worse prognosis, and tumour location; tumours involving the ciliary body having a worse prognosis.1;16

- epithelioid or mixed cell type - invasion - older age - male* Table 1. Some factors found to be prognostic in uveal melanoma. Based on 1;16 and table, app. I in Isager (2004)1. *Decreased overall survival.

6

Cytogenetic prognostic factors Clinical and histopathological parameters such as sex, age, localisation, size, cell type and invasion provide important information on the patients risk of developing metastatic disease.16;17 However, recent studies have revealed a much closer association of prognosis with cytogenetic prognostic markers. It has been shown, that approximately one half of uveal melanomas develop monosomy 3,18 and that this finding may have great prognostic value in uveal melanoma.19;20 One study found, that only patients with monosomy 3 present in the tumour died from liver metastases.21 Also it has been shown that amplification of chromosome 8q, often found together with monosomy 3, correlates with decreased chance of survival.20 The prognostic implications of chromosome 6 copy number alterations are still vague, but they appear correlated with improved survival.22

Tumours with and without monosomy 3 are different entities Studies of gene expression in uveal melanomas have shown, that uveal melanomas may be divided into two completely distinct groups with differential expression, and that these groups correlate strongly with the presence or absence of monosomy 3.23;24 Also, it has been suggested that monosomy 3 and gain of 6p represent two different entry points in the process of tumour genesis, as these seem to be early and mutually exclusive events, with gain of 8q appearing as a possible later event, worsening the prognosis.25 These observations have led to the assumption, that tumours with and without monosomy 3 represent two distinct pathological entities.23;25

Genetic mechanisms of tumour genesis It is well established, that cancers arise and develop in a multi-step process, involving loss of recessive tumour suppressor gene function and gain of dominant oncogene function. These changes lead to capabilities as self-sufficiency in growth signals, insensitivity to anti-growth signals, a limitless replicative potential, evasion of apoptosis, sustained angiogenesis and the ability to invade tissue and metastasize.26 Little is known about the underlying biology of monosomy 3, but chromosome 3 is known to harbour several important tumour suppressor genes,27 and several potential tumour suppressor genes have been shown to have lover expression in tumours with monosomy 3.23;24 The hypothesis is therefore, that the loss of one copy of chromosome 3 is an early event in tumour genesis, that effectively discards important tumour suppressor genes.28 The other allele may be inactivated through point mutation, deletion or epigenetic mechanisms.29 The observation of uveal melanomas with only partial deletions on

7

chromosome 3, occurring in at least 4 % of cases, have provided some clues to the location of these putative tumour suppressor genes, specifically regions 3p25 and 3q24-26.28;30 Almost all tumours with monosomy 3 also show amplification of chromosome 8q,20 probably leading to amplification of the oncogene c-myc located at 8q24.12; an oncogene involved in the control of differentiation and apoptosis.31 DDEF1, a potential oncogene involved in cell motility and located at 8q24.2, have also been reported amplified and overexpressed in uveal melanoma.32 The biologic significance of the alterations observed at chromosome 6 are still unclear, but expression of HLA class 1 antigens, coded by genes located at 6p, are impaired.33 Also, loss of chromosome 1p seems involved in the development of metastatic disease, but the mechanism is uncertain.34 Although this development is only in its beginning, these findings offer new insights into the mechanisms leading from tumour genesis to metastatic uveal melanoma, giving hope to developing new treatments. In short, uveal melanoma has a unique clinical presentation, characterized by the fact that metastases may arise many years after successful treatment of the primary tumour. Approximately half of the patients die from the disease, regardless of treatment. The presence of monosomy 3 in the tumour, reflecting a different tumour class, may predict this.

Purpose Rather disappointing, the prognosis of uveal melanoma patients has only improved very little over the last 25 - 50 years, in spite of advances in both diagnosis and treatment.1;13 Changing the course of this deadly disease requires that new adjuvant systemic chemotherapies or immune therapies are developed. Establishing a significant correlation between a biological marker and poor survival, and secondary developing a reliable method of assessing this in tumours, may set the stage for this development. Firstly, a more accurate prediction of individual prognosis would be very valuable, not only in clinical practice, but also in the conduction of clinical trials, especially of adjuvant therapies. Also it would be beneficial to accurately determine, which patients should be monitored closely for metastatic disease. Secondarily, biologic prognostic markers are important, because they may be informative of the mechanisms, by which the disease develops. These mechanisms might in the future be the target of specific treatments.

8

In consequence, the purposes of this project were: 1. to explore the prognostic significance of chromosomal aberrations, especially loss of chromosome 3, in uveal melanoma. This would be done in a series of uveal melanomas with available histological material at the Institute of Pathology, Aarhus University Hospital, and enucleated 16 to 37 years ago, and 2. to develop a method for easy assessment of loss of chromosome 3 in archival tissues, suitable for application in a larger series.

Hypotheses In accordance with the abovementioned purposes of this study, the following hypotheses were to be tested: I. A method for easily evaluating chromosomal imbalances in archival paraffin-embedded, formalin-fixed tissue-samples of ocular melanomas can be developed. II. Chromosomal imbalances in ocular melanoma, monosomy 3 and gains of chromosome 8 in particular, hold important prognostic information on the risk of metastatic disease and death.

Subjects and methods Patients A cohort of 64 patients with histological material was available for analysis. Patients were treated by enucleation at Aarhus University Hospital in the years 1968 - 1989 and were followed through 2002.

Sex (Male / Female)

20 / 17

Age, median

61 years (range 26 to 88)

Cell type

Spindle: 15 (41%) Mixed: 17 (46 %)

These are a part of a larger

Epithelioid: 5 (14 %)

historical cohort previously

Tumour location

Choroid: 28 (82 %)

described,1 and the gathering

(n= 34)

Ciliary body / choroid: 3 (9 %)

of data was performed by Dr. Peter Isager. Characteristics of the cohort are summarized

Iris / ciliary body: 3 (9 %) Largest basal diameter, median

12 mm (range 4,5 – 23)

(n=17) Table 2. Summary of patient characteristics

in table 2.

9

Quantitative real-time PCR By measuring the amount of fluorescence emitted from a PCR reaction with a fluorescently labelled probe, real-time PCR allows fast quantification of relative copy number changes in specific DNA sequences. Assay design We designed a real-time PCR assay for quantification of DNA copy number changes at chromosome 3. Primer locations were selected as shown in figure 1, A. Primers and probes were designed with the assistance of Primer Express software (Applied Biosystems), and produced by DNATechnology (Århus, DK). The test amplicons were located at 3p25 and 3q24, respectively, in regions previously reported to be consistently deleted in uveal melanomas with partial deletions.28 Control amplicons were positioned at 1q, 2q, 5q and 7q; regions not known to be frequently amplified or deleted in uveal melanoma.35;36 Primers surrounding (CA)n microsattelite-repeat sequences were selected (figure 1, B). This allowed for use of a common probe, homologous to the (CA)n-repeat region, for control amplicons. This permits the creation of a common control “pool” of one probe but several primer-pairs, more robust to random amplifications and deletions.37 Primer and probe sequences are available in appendix.

D3S1537

D7S2516

D1S2707

D2S385 D3S2404 D5S643

A Pooled control with a common probe sequence: GTGGCTGAGGCAGGAGAATCACACACACACACACAACTATCGATTACCCTGCACT CACCGACTCCGTCCTCTTAGTGTGTGTGTGTGTGTTGATAGCTAATGGGACGTGA CAACGTCACTCGGTTCTAGCAGTGTGTGTGTGTGTGTGGTCTCAGTTATCGTGTGGT GTTGCAGTGAGCCAAGATCGTCACACACACACACACACCAGAGTCAATAGCACACCA

B Figure 1. A) Chromosomal location of PCR probes. Red indicates test probes for detection of loss of chromosome 3, blue indicates control probes. B) Control amplicons were selected, so that probes surrounded a CA-repeat region, and a common probe were constructed.

10

DNA extraction Before analysis, an extraction of DNA from formalin-fixed, paraffin-embedded tissue was performed. Briefly, four tissue sections of 10 µm were transferred into a microcentrifuge tube. Before section, a macrodissection was performed to ensure rough exclusion of non-tumour tissue. 200 µL proteinase-K buffer and 10 µL proteinase K (20 mg/mL) was added before incubating for 5 days at 65° C with continuous vibration. After brief centrifugation, 5 minutes at 20.000 g, the DNA purification continued on NucleoSpin Tissue Mini (Macherey-Nagel) spin columns, according to manufacturer’s protocol. Briefly, a binding-buffer B3 was added to the DNA solution, and the mixture was incubated at 70° C for 10 minutes. Before loading to the microcentrifuge spin columns, ethanol was added. After centrifugation, a wash was performed by adding wash buffers to the column and centrifuging. Finally, the purified DNA solution was harvested by adding 100 µL of 70° C hot elution buffer to the spin columns, incubating for 1 minute and centrifuging the solution into a new microcentrifuge tube.

Quantitative real-time PCR procedure PCR was performed on an ABI Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, USA) using TaqMan Universal PCR MasterMix (Applied Biosystems) according to manufacturers protocol. For test amplicons (D3S1537 and D3S2404), 7.5 pmol of each primer, 6.25 pmol probe, 12.5 µL TaqMan Universal PCR MasterMix and water was added to a total of 20 µL. Test amplicons were run together in one pool, and thus 7.5 pmol of each primer (4 pairs) was added, with less water to a total of 20 µL. Finally, 5 µL of purified DNA solution was added to each well. After an initial denaturation for 10 minutes at 95° C, 40 cycles of 15 seconds denaturation at 95° C and 60 seconds elongation at 60° C was run. Each sample was run in triplicates. Positive and negative controls were included in all experiments. Emitted fluorescence after each cycle was registered by the ABI Prism 7000 Sequence Detection System and reported to a computer, using ABI SDS software. Data were then exported to Microsoft Excel for completion of calculations.

11

Comparative Ct-method The comparative Ct-method, as originally described38, and later modified37, allows for calculation of a relative DNA copy number under certain assumptions, using quantitative realtime PCR (Please refer to figure 2). The cycle where the fluorescence emitted from the exponentially increasing PCR-reaction reaches a fixed threshold above background level, is called the threshold cycle, Ct. The Ct for test loci and reference pool were obtained for patients and controls by running reactions in triplicates and averaging results afterwards, excluding obvious errors. A ∆Ct was calculated by subtracting ∆Ct = Ct(test locus) - Ct(control pool). By subtracting values obtained from normal control tissue from values obtained from tumours, a ∆∆Ct is calculated: ∆∆Ct = ∆Ct(tumour DNA) - ∆Ct(control DNA). This allows us to calculate a relative DNA copy number: Relative DNA copy number = 2-∆∆Ct , assuming that the efficiencies of the PCR reactions are equivalent. As control material, we used DNA extracted from formalin-fixed, paraffin-embedded lymph nodes.

Fluorescence in situ hybridization (FISH) FISH is the direct localization of genetic sequences in a nucleus or in tissues, by the hybridization of a fluorescently labelled probe to a complementary DNA sequence. In this study,

Patient

Control

Control pool

Control pool ∆Ct

∆Ct Chromosome 3

Chromosome 3

∆∆Ct = ∆Ct(control) - ∆Ct(patient) Figure 2. Schematic concept of the ∆∆Ct-method. A relative loss of chromosome 3 material in this fictive patient is seen by the later rise of the red curve (left). ∆Ct(control) - ∆Ct(patient) < 0, and therefore 2∆∆Ct < 1.

12

we used probes specific for the centromeric regions of chromosomes 3 and 7, respectively. The centromeric region of chromosome 3 is lost along with the rest of the chromosome in monosomy 3, apart from the rare cases of partial deletions.28 Chromosome 7 was chosen as a control, since this chromosome is known not to be frequently amplified or deleted in uveal melanoma.35;36

Tissue preparation Available material was enucleated eyes and resected tumours, all formalin-fixed and paraffinembedded. Sections in 3 µm thickness was mounted on glass slides (Super Frost Plus) and airdried. Deparaffinization was performed using Dako kit. Briefly, slides were immersed in xylene, and slides were subsequently rehydrated in graded ethanol to Wash Buffer (Dako Cytomation). Sections were then boiled in Pre-Treatment Solution at 95° to 99° C for 10 minutes. After gentle cooling, slides were washed in Wash Buffer and placed on a heat block at 37° C. Tumour was covered with Pepsin and incubated for 3½ minutes before immersion in Wash Buffer. Sections were then dehydrated in graded ethanol series and air dried for 30 minutes.

In situ hybridisation Centromere-specific probes for chromosome 3 (SpectrumOrange) and 7 (SpectrumGreen) were purchased from Vysis, and used according to the manufacturers instructions. A probe mix was prepared, containing 7 µL Hybridisation Buffer (Vysis), 1 µL water and 1 µL of each probe. 2.5 – 5 µL probe mix was applied, covered with coverslip and sealed with vulcanising solution (Tip Top, Germany). Probe and target DNA was then denaturated by placing the slide on a heating block at 85° C for 5 minutes before overnight hybridisation in a humid chamber at 42° C. After removal of coverslip, slides were washed in wash buffer (0.4 x SSC, 0.3% Tween-20) at 72° C for 2 minutes before immersion in wash buffer (2 x SSC, 0.1% Tween-20) at room temperature for 20 seconds. Before covering with cover glass, slides were mounted in antifade (Vectashield) with counterstain (DAPI).

Quantification of FISH signals An epifluorescent microscope (Leica), equipped with appropriate filters, was used to examine the tissue sections. Initially, the quality of each specimen was assessed, and specimens with high background or weak or absent hybridisation signals were disqualified. Only specimens with

13

clear and strong hybridisation signals and intact nuclear morphology, assessed using DAPI filter, were scored. Overlapping or damaged nuclei, nuclei without hybridisation signals or requiring subjective assessment were not scored. For each tissue section, the number of signal spots for chromosome 3 and chromosome 7 was determined in a total of 200 nuclei in 10 clusters, if possible representing the entire tumour. Prior to analysis, nuclei with no control signal (chromosome 7) were excluded. The data were analysed as follows. For each sample the chromosome index (CI), which gives an average chromosome copy number, was calculated for chromosomes 3 and 7, respectively, using the following formula:

Chromosome Index =

number of centromere signals number of nuclei counted

The relative Chromosome Index (rCI), which gives the number of chromosome 3 per chromosome 7, was calculated using this formula:

Relative Chromosome Index =

CI3 CI7

Validation of FISH method To confirm the affinity of the purchased probes to the correct chromosomal regions, we hybridized the probes to metaphase cells of human bone marrow under standard conditions. (See figure 8, appendix). This confirmed that the probes attach themselves to the centromeric regions of chromosomes 3 and 7, respectively. To assess the properties of the probes under optimum conditions, we performed

Mean

SD

Range

Mean ± 3 SD

hybridization to two specimens of normal

CI3

1.61

0.06

1.58 - 1.73

1.30 – 1.93

human bone marrow, counting 200 nuclei in

CI7

1.72

0.10

1.65 – 1.80

1.55 – 1.88

each specimen. A total of 94% of the nuclei

rCI

0.94

0.04

0.88 – 1.00

0.81 – 1.07

showed two hybridization signals for both

Table 3. Mean Chromosome Indexes and mean relative Chromosome Index for normal tissues of five specimens.

chromosomes 3 and 7. Loss of one signal for

chromosome 3 was recorded in 2% of the nuclei, and loss of one signal for chromosome 7 was seen in 2.7%.

14

To obtain a normal reference, we examined the signal distribution of the normal tissues surrounding the tumours of five specimens. The tissues selected were the inner and outer nuclear layer of the retina. (See figure 9, appendix) This tissue was chosen, since it has been treated in the exact same way as the underlying tumour specimen, and since nuclear size and morphology in this tissue is similar to those of the tumour. A total of 47% of the nuclei showed two hybridization signals for both chromosomes 3 and 7. The mean CI for chromosome 3 (CI3) was 1.61, and the 99.8 % prediction interval (mean CI ± 3 standard deviations) was 1.30 – 1.93. The mean rCI was 0.94, and 99.8 % prediction interval was 0.81 – 1.07. From this we defined monosomy 3 as either CI3 < 1.30 or rCI < 0.81. These limits will be considered further in the discussion section. See table 3.

Results Quantitative real-time PCR

3,0

For validation purposes, we tested the real-

D3S1537 D3S2404

2,5

Relative DNA copy number

time assay with DNA extracted from formalin-fixed, paraffin-embedded lymph nodes from 8 patients. These lymph nodes had been found tumour free, and so were not

2,0 1,5 1,0

expected to show any genetic imbalance. In

0,5

these 8 normal controls, calculated relative

0,0

1

DNA copy numbers for either locus ranged

bottom. Ranging between 0.63 and 1.95, the mean relative DNA copy number was 1.1, and the standard deviation was 0.6.

DDCt Relative DNA copy number, 2

uveal melanomas, as shown in figure 3,

4 3 5 Control sample #

6

7

8

2,5

between 0.3 and 2.7 (Figure 3, top). We also tested DNA extracted from five

2

2,0

1,5

1,0

0,5 D3S1537 D3S2404 0,0

We expected the results from the control samples to scatter with minimal variation around 1. Results from the tumour samples

12661/77A 8015/72-73 3109/71-72 Patient ID

4876/78

2126/78

Figure 3. An example of results obtained by quantitative real-time PCR. Top. Calculated relative DNA copy number in two loci for control samples. Bottom. Results obtained from five uveal melanomas.

15

were expected to not exceed 1, with the occasional pair falling significantly below 1. As seen from figure 3, variation was significant, with several values close to 2. Moreover, these figures could not be consistently reproduced. In spite of energetic efforts of optimization and troubleshooting, the results obtained with the quantitative real-time PCR assay were neither dependable nor reproducible. The reasons for this will be discussed later.

FISH Of the 64 available tumours, fluorescence in situ hybridization was successful in only 37 cases (58 %). Analysis was unsuccessful in the remaining 27 cases, largely due to high background intensities or weak or missing signals. In most cases, more than one attempt to produce a proper specimen was made. Assessment of chromosome index In the 37 tumours analyzed, chromosome index for chromosome 7 (CI7) varied uniformly between 1.45 and 1.94 (Please refer to figure 4, and figure 7 in the appendix), with mean and median at 1.65. This is slightly lower than the mean CI7 for the five control samples. The standard deviation was 0.093. In five cases, the CI7 is slightly more than 3 standard deviations below the mean CI7 of the control samples. However, there was no connection with chromosome 3 status, and the finding is probably due chance or to lower tissue quality of the tumours compared with the retina. In one tumour (9092/72-73), we found a CI7 at 1.94, which is more than 3 SD above the mean CI7 of the control samples. In this tumour, we found a moderate proportion of nuclei with three hybridization signals for chromosome 7. CI3 for this tumour was 1.37, which is within the normal area, giving a rCI of 0.70. This tumour was classified as having a relative loss of chromosome 3.

16

The CI3 for the 37 tumours varied between 0.71 and 1.72, which is a much wider distribution than that of CI7 in the tumours. The standard deviation was 0.31, 3½ times larger than that of CI7. This probably implies that some tumours have loss of chromosome 3. Applying a cut-off of 1.30 as described in the methods section, 17 tumours fall below this limit. Assessment of relative chromosome index The relative chromosome index rCI, calculated by dividing CI3 by CI7, is shown for the 37 tumours in figures 4 and 7. The rCI varies between 0.45 and 0.99, and has a median of 0.79. Applying a cut-off of 0.81, 19 tumours have relative loss of chromosome 3. Assessment of monosomy 3 status The two measures for detection of loss of chromosome 3 identified 17 and 19 tumours with loss, respectively. In general, concordance between the two measures was high (95%). In only two cases, the methods were disconcordant. One is discussed above (9092/72-73). The other case (8520/68-69) has a CI3 of 1.33 and an rCI of 0.79. As later discussed, we chose to assess tumours with rCI only, and hence this tumour was also categorized as having loss of chromosome 3. This made it possible to identify 19 cases of uveal melanoma with relative loss of chromosome 3, out of a total of 37 cases. This proportion, 51 %, is in concordance with other studies.19;28

Relative Chromosome Index

1.5 1

Control mean CI3 – 2 SD Control mean CI3 – 3 SD CI7

1,0

0,8

0,6

0,4 Control mean - 2 SD Control mean - 3 SD

.5

Chromosome Index

2

1,2

CI3

0,2

Figure 4. Left, the distribution of chromosome index for chromosomes 3 and 7, respectively, are shown. Right, the distribution of the relative chromosome index is shown. Lines indicate mean – 3 SD (broken line) or mean – 2 SD (dotted line) from the control specimens.

17

Relation between monosomy 3 and survival The overall survival in the 37

Overall survival

1.00

cases ranged between 3 months median survival of 4 years and

0.75

and almost 30 years, with a

within the first five years, with a second peak at 15 years (see figure 5). The mortality peak at 1 to 5 years

0.25

most of the mortality occurs

0.00

interval: 2.5 – 14.2 years). Thus,

0.50

10 months (95 % confidence

0

10

20

Years

30

Figure 5. Overall survival in 37 patients.

is mainly due to melanomarelated mortality, whereas the mortality peak at 15 years mainly reflects other causes of death. Survival data are also shown in table 6, appendix.

median survival was 14 years, compared to only 2.5 years in the

1.00 0.75

loss of chromosome 3, the overall

Overall survival

p = 0,037

0.50

In the group of 18 patients with no

statistically significant (p = 0.037, log-rank test). Please refer to figure

Disomy 3 Monosomy 3

0.00

monosomy 3. This difference was

0.25

group of 19 patients with

0

1

6. The hazard rate ratio in the

Years

3

4

5

4

5

Melanoma-specific survival 1.00

group with monosomy compared

2

1.0 – 4.4).

0.75

to no monosomy was 2.1 (95% CI:

groups (p = 0.036, log-rank test). There were 5 events in the group

0.25

differed significantly in the two

Disomy 3 Monosomy 3

0.00

The melanoma-specific survival

0.50

p = 0,036

0

1

2

Years

3

Figure 6: Overall and melanoma-specific survival by presence of monosomy 3 in 37 patients. Truncated at 5 years.

18

with no loss of chromosome 3, as compared to 10 events in the group with monosomy 3 (See table 4). The hazard rate in the group with monosomy compared to no monosomy was 3.0 (95% CI: 1.0 – 8.8). It follows from this, that monosomy 3, assessed with FISH technology, is able to define two subgroups of patients with uveal melanomas with different survival.

Monosomy 3

Yes No

Melanoma

Other cancer

Other disease

Accident

Censored

Total

10 5

3 4

4 4

1 0

1 5

19 18

1

6

37

Total 15 7 8 Table 4. Cause of death and relation to monosomy 3 in 37 patients.

Relation between monosomy 3 and other patient characteristics. We also investigated the relation between monosomy 3 and other prognostic factors as sex, age, tumour size, location and cell type (Shown in table 5). No significant correlations could be established, although there is evidence of a relationship between monosomy 3 and ciliary body involvement (p=0.07).

Monosomy 3 Yes Age, mean (years) Sex (Male / Female) Spindle Cell type Mixed Epithelioid Location Ciliary body involved (n=34)

Choroid Largest basal diameter, mean (mm) (n=17)

No

62 (26 to 79) 9/10 7 9 3

55 (28 to 88) 11/7 8 8 2

5

1

12 12.4

16 13.0

n.s. (t-test) n.s. (chi2-test) n.s. (chi2-test) p=0.07 for anterior location n.s (t-test)

Table 5. Relation between monosomy 3 and classic prognostic parameters in 37 patients. N.s.: Not significant.

19

Discussion Selection of methods When investigating genetic changes in solid tumours, the selection of proper methods is all important. We selected to analyze the tumours for loss of chromosome 3 with FISH technique, because of the advantages of this method; namely that it is a robust and much used method, offering intuitive results. Other studies have shown that all of chromosome 3, including the centromeric part, is lost in tumours with monosomy 3,21 apart from a minor subcategory of tumours with only partial deletions on chromosome 3, approximately 4 % of all tumours.28 Another advantage is, that the histological properties of the tumour are visible; allowing the investigator to exclude obvious non-tumour nuclei. The method also provides information on a single-cell level, uncovering possible intratumoral heterogeneity. The disadvantages of the FISH method is the laborious work setting and the observer-dependent analysis. Our investigations show, that the FISH method was not as robust as anticipated, and we failed to produce specimens of proper quality in 27 of 64 (42 %) cases, which is a rather high failure rate. While performing interphase FISH on tissue sections has its advantages, namely visible histology and less laborious work setting, it also created new problems, due to nuclear truncation. This had the consequence, that the results obtained with FISH analysis were not as intuitively interpreted as expected, see later discussion. As our second method of investigation, quantitative real-time PCR was not quite the tool we had hoped for. The advantages of this method, which include the recognition of specific DNA sequences and a high throughput, were outweighed by the fact, that the results obtained were inconsistent and irreproducible. The reasons for this are probably in part, that the tissue analyzed was of poor quality. Previous successful studies utilizing this method have mostly been performed on DNA extracted from whole-blood,39 although some have been successful in applying this method on DNA extracted from formalin-fixed paraffin-embedded tissues.40 A second cause of the failure of this method may lie in the design of the oligonucleotide primers and probes. The comparative Ct-method assumes equal efficiency of PCR amplification in control and reference-loci, which might not have been the case. Our initial validation studies on good-quality DNA showed an acceptable relative efficiency, but subsequent analysis on DNA extracted from normal formalin-fixed, paraffin-embedded tissues showed a greater divergence in PCR amplification efficiency in control and reference-loci.

20

What is monosomy 3? The normal cell contains two copies of each of the chromosomes 1 - 22, and ideally monosomy is the pathological condition in which only one copy of a chromosome is present. However, life gets more complicated when investigating tissue slices with interphase FISH technology. Euploid cells may contain only one FISH signal due to a number of reasons. Above all, slicing of the tissue, resulting in nuclear truncation, obviously leads to a loss of nuclear material and hence chromosomes and FISH signals. Two overlapping signals are also interpreted as one. Furthermore, the sensitivity of the FISH probe may be slightly lower than one hundred percent, so that not all centromeres are recognised by the observer as a signal, especially in archival tissues. On the other hand, euploid cells contain four copies of each chromosome during mitosis, and due to nuclear truncation this may be observed as only three signals. Also, one centromeric signal is sometimes observed as two, so called “split signals”. These circumstances have the effect that one may observe between zero and four FISH signals in a normal nucleus. One would expect, that in a disomic cell population, cells with zero copies would be very rare, and contrary cells with two copies would be very rare in a monosomic population, but to what extent? On top of this, one might also observe normal cells within the tumour, e.g. vascular endothelium cells or cells of the immune system, or the tumour may be heterogeneous in its constitution. As a result, objective measures are needed when it comes to distinguishing between monosomy and disomy. For this purpose, we calculated the chromosome index (CI), which is the average number of chromosome copies per nucleus, for chromosomes 3 and 7. From this, we also calculated the relative chromosome index (rCI), which is the ratio between CI3 and CI7. One might argue, that a “natural” definition of monosomy 3 is, when you have 1 or less copy of chromosome 3 per nuclei, i.e. when CI3 ≤1. You could also argue that in monosomy 3, nuclei contains half as many copies of chromosome 3 than of chromosome 7, and therefore monosomy 3 should be defined as rCI ≤ 0.5. But due to the factors mentioned above, these would not seem valid definitions in a population of dividing, possibly heterogeneous cells in a sliced specimen. To obtain a normal reference, we analyzed the normal tissues surrounding five tumour specimens. From this we defined the normal reference range of CIs and rCI in normal tissue, fixed and treated under the exact same conditions as the tumours. From this we define monosomy 3 as a relative loss of chromosomal material, when CI3 falls below 1.30 or rCI falls below 0.81. The two measures were concordant in 35 of 37 cases (95 %). Since rCI is

21

independent of the absolute number of signals in the tissue, and is normalized to a control signal of chromosome 7, this measure is more independent of specimen quality, at least in theory. As a consequence of these considerations, we chose to categorize the tumours according to rCI alone, which seems fair in view of the high concordance between these two measures. This, however, only forebodes new problems. Our introduction of a quantitative parameter to describe a phenomenon which is dichotomous in nature, inevitably leads to the question whether this gives the correct diagnosis? Looking at figure 7, one might doubt that the cut-off points obtained from our studies of normal tissues actually reflects a true distinction of two natural groups. On the other hand, two large studies of enucleated uveal melanomas found frequencies of monosomy 3 at 49% and 55%, respectively.19;28 Our finding of a frequency of monosomy 3 of 51% is fully in line with this. Therefore, we find a severe misclassification to be unlikely.

Prognostic implications of monosomy 3 Other studies have shown an intriguing association of monosomy 3 with metastasis-related death. Two studies, with each 33 and 105 patients found, that only patients with monosomy 3 died form metastatic disease.21;41 In another study with 42 patients, only two patients died from melanoma without monosomy 3.20 Thus, from these three studies only 2 patients, out of a total of 42 (5 %) dead from melanoma, did not show monosomy 3. These observations have led to the general perception, that metastatic disease in the absence of monosomy 3 is rare. In this study, we found the presence of monosomy 3 in the tumour to be significantly associated with increased overall mortality, as well as increased melanoma-specific mortality. This confirms the findings of other studies. This association was however not as strong as seen in other studies. In this cohort, five patients classified as having monosomy 3 died from melanoma, which is unusual. Possible explanations for this include misclassification of the tumour, although this seems an unlikely explanation for all cases. Another thing to consider is that all the patients of this cohort were treated in the years 1968 – 1989, a time where the standard treatment of these tumours was enucleation, and no eye-preserving therapy was performed.1 For that reason, this cohort might be less biased towards large tumours, as compared to recent studies based on tissue available from relatively uncommon enucleations.

22

Relation with other prognostic factors We were not able to define a significant relationship between monosomy 3 and other prognostic factors investigated. This is in concordance with other studies, who have shown only vague associations between monosomy3 and classical prognostic factors.41 However, we did note that tumours involving the iris all showed monosomy 3 (n=3), and that of tumours involving the ciliary body and either the iris or the choroid, 5 out of 6 showed monosomy 3 (not significant).

Limitations As emphasized by several authors, many studies of prognostic factors suffer from similar flaws; small sample sizes, irreproducible assays and inappropriate selection of cut-points when allocating a continuous variable to categories.42 These criticisms also apply to studies of uveal melanoma,43 and most likely the present study as well. The included 37 patients only comprise roughly 4% out of an estimated total of approximately 1000 patients with uveal melanoma in Denmark in the years 1968 – 1989 (Estimate based on Isager, 20041). This is an obvious limitation in the interpretation of the results. Recently, several authors have suggested a special progression of prognostic investigations,42;44 including a second validating, retrospective study to confirm initial findings. This seems a reasonable approach, and we had indeed planned to do so. Unfortunately, our low rate of success in the FISH analysis prevented us from doing so. Also, we would have liked to develop and validate a PCR based assay in this series, but were unsuccessful in doing so. This could have led to investigation of a larger series, and thus accommodate much of the criticism put forth about prognostic studies like this.

Conclusions I. We were not able to construct a reproducible assay for easy evaluation of chromosomal imbalances in DNA extracted from archival formalin-fixed, paraffin-embedded tissues. We were, however, able to assess loss of chromosome 3 in a cohort of patients with available histological material using FISH technique. II. We have shown, that loss of chromosome 3 was significantly associated with a decreased overall and melanoma-specific survival. Hypothesis II was accepted.

23

Perspectives The cumulative evidence of studies, including this, correlating monosomy 3 to a poor prognosis in uveal melanoma is conclusive. Yet, further research is warranted, as most of the studies are rather small, and only include patients with relatively large tumours treated by enucleation. Larger, possibly prospective, studies would be the ideal. A precise, individual estimate of prognosis would be beneficial to patients with uveal melanoma. This would enable selection of high-risk patients for screening for metastatic disease. A minority of patients with solitary metastases might gain prolonged survival after surgical resection or hepatic intra-arterial chemotherapy.45 Early detection of metastatic disease is becoming more relevant, since phase II-trials of effective chemotherapy, improving progression-free survival, are emerging.46 Cytogenetic analysis subsequent to intraocular biopsy is possible,47 allowing a combination of eye-preserving therapy and screening of high-risk individuals. Ideally, these patients should be included in large-scale, multi-centre protocols, effectively testing different screening and treatment strategies.

24

Appendixes Figures.

Chromosome Indexes and relative Chromosome Index 1,4 rCI Control mean - 3 SD Control mean - 2 SD

Relative Chromosome Index

1,2

1,0

0,8

0,6

0,4

0,2

0,0 2,0

Chromosome Index

1,5

1,0

0,5

CI3 CI7 Control mean - 3 SD Control mean -2 SD

0,0

77 78 88 73 88 73 72 71 78 69 69 78 85 70 73 74 79 75 69 70 74 72 77 73 69 79 73 74 69 76 78 74 69 84 74 69 84 8/ 76/ 47/ 72- 06/ 72- 71- 70- 37/ 68- 68- 71/ 24/ 69- 72- 73- 06/ 74- 68- 69- 73- 71- 40/ 72- 68- 44/ 72- 73- 68- 95/ 12/ 73- 68- 28/ 73- 68- 10/ 5 5 48 10 5/ 6 1/ 9/ 0/ 26 7/ 5/ 53 54 1/ 2/ 7/ 1 3/ 0/ 0/ 3/ 1/ 85 2/ 4/ 33 2/ 2/ 2/ 43 3 1/ 2/ 36 0/ 8/ 75 1 8 0 2 9 7 8 7 4 4 2 8 5 7 9 7 1 7 1 43 864 747 893 80 47 31 86 56 74 56 36 65 80 85 13 45 62 90 74 52 95 32

Patient ID

Figure 7. Chromosome indexes for chromosomes 3 and 7 (bottom) and relative chromosome index (top) for 36 patients. The reference lines refer to normal values of CI and rCI found in the normal tissues. SD: Standard deviations. CI3: Chromosome index for chromosome 3. Deaths from melanoma are marked by ● in the top graph.

25

Figure 8. Probes for centromere 3 (red) and centromere 7 (green) hybridized to a metaphase spread of normal chromosomes.

Figure 9. Inner and outer nuclear layer of a detached retina overlying a uveal melanoma (not shown). Left: view through DAPI-filter, right: view through colour filters.

26

Figure 10. FISH on tumour specimens 9092/72-73(top) and 2637/78 (bottom).

27

Clinical

Histological

Sex

Age

Year

Cell type

Tumour location

312/78

Female

82

1977

Mixed

7470/73-74

Male

78

1973

9092/72-73

Male

88

9572/73-74

Female

8520/68-69 8938/68-69 5424/85

FISH LBD

Survival

CI7

CI3

rCI

Monosomy 3

Months

Cause of death

Choroid

1,5

0,8

0,5

Yes

3

Other disease

Mixed

Choroid

1,5

0,7

0,5

Yes

7

Other cancer

1973

Mixed

Iris/Ciliary body

1,9

1,4

0,7

Yes

8

Accident

70

1974

Epithelioid

Choroid

1,7

1

0,6

Yes

12

Melanoma

Male

52

1969

Spindle

1,7

1,3

0,8

Yes

13

Other cancer

Female

68

1969

Epithelioid

1,6

0,8

0,5

Yes

14

Melanoma

Male

43

1985

Mixed

1,7

1,5

0,9

No

16

Melanoma

4395/76

Female

52

1976

Epithelioid

Ciliary body/Choroid

1,7

0,9

0,5

Yes

22

Melanoma

7475/68-69

Female

69

1969

Spindle

Choroid

1,5

1,4

0,9

No

25

Other disease

3212/68-69

Male

73

1968

Spindle

Choroid

1,6

0,9

0,6

Yes

27

Other disease

3672/72-73

Female

47

1972

Mixed

Choroid

1,8

1,5

0,9

No

27

Melanoma

4553/73-74

Female

61

1973

Mixed

Iris/Ciliary body

1,6

1,3

0,8

Yes

29

Melanoma

12

Choroid

20 17

431/73-74

Male

29

1973

Mixed

Ciliary body/Choroid

1,6

0,8

0,5

Yes

30

Melanoma

4781/72-73

Male

75

1972

Mixed

Choroid

16

1,7

1,6

1

No

30

Other disease

6271/71-72

Female

70

1971

Spindle

Iris/Ciliary body /Choroid

17

1,6

1,2

0,8

Yes

32

Melanoma

558/77

Female

79

1976

Epithelioid

Choroid

1,7

1,6

1

No

33

Melanoma

3109/71-72

Female

34

1971

Mixed

Ciliary body/Choroid

1,6

1,5

0,9

No

36

Melanoma

7510/84

Male

81

1984

Spindle

1,7

0,8

0,5

Yes

54

Melanoma

8642/68-69

Female

58

1969

Spindle

Choroid

1,7

0,9

0,5

Yes

58

Melanoma

8540/77

Male

68

1977

Mixed

Choroid

1,7

1,3

0,7

Yes

86

Other cancer

1380/69-70

Female

72

1969

Spindle

Choroid

1,5

1,2

0,8

Yes

89

Melanoma

5681/69-70

Male

61

1969

Mixed

Choroid

1,6

1,4

0,9

No

133

Other cancer

8620/70-71

Male

61

1971

Spindle

Choroid

1,7

1,5

0,9

No

164

Other disease

1047/88

Male

65

1987

Spindle

Choroid

1,8

1,7

1

No

167

Censored

606/88

Male

58

1987

Mixed

Choroid

9

1,8

1,7

1

No

168

Censored

3628/84

Male

28

1984

Mixed

Choroid

7,5

1,6

0,8

0,5

Yes

170

Melanoma

6547/73-74

Female

50

1973

Spindle

Choroid

1,6

1,3

0,9

No

171

Other cancer

8015/72-73

Male

77

1973

Spindle

Choroid

1,7

1,6

1

No

173

Other disease

7474/68-69

Male

65

1966

Mixed

Choroid

1,7

1,1

0,7

Yes

182

Other disease

8043/74-75

Male

26

1974

Spindle

Choroid

18

1,6

1,3

0,8

No

184

Other cancer

4876/78

Female

68

1978

Spindle

Choroid

10

1,7

1,6

1

No

255

Other cancer

3344/79

Female

34

1979

Spindle

Choroid

23

1,6

1

0,7

Yes

273

Censored

106/79

Male

40

1978

Mixed

Choroid

8,8

1,8

1,5

0,8

No

275

Censored

5371/78

Female

54

1978

Mixed

Choroid

6

1,5

1,4

0,9

No

283

Censored

2637/78

Male

41

1977

Epithelioid

Choroid

5,3

1,6

1,5

0,9

No

285

Censored

5697/68-69

Male

38

1966

Spindle

Choroid

1,7

1,5

0,9

No

303

Melanoma

5212/72-73

Female

53

1972

Mixed

Choroid

1,6

1

0,6

Yes

354

Other disease

LBD: Largest basal diameter.

18 14 11

4,5

Table: Patient characteristics

28

Table 6. Summary of clinical and histopathological data for 37 patients analysed with FISH. Ranked by survival.

Patient ID

FISH protocol 1. Afparaffinering og rehydrering (stinkskab). 1. 2. 3. 4. 5.

(Opstart vandbad med Pre-Treatment Solution til pkt. 2.1, kontroller temperaturen) Objektglassene anbringes i xylen-bad i 2 x 5 minutter, omrystes forsigtigt. Badet skiftes én gang. Overskydende xylen bankes af på karret, og objektglassene overføres til 96 % ethanol i 2 minutter, badet skiftes gang, og de står yderligere 2 minutter. Overskydende ethanol bankes af på karret, og objektglassene overføres til 70 % ethanol i 2 minutter, badet skiftes gang, og de står yderligere 2 minutter. Overskydende ethanol bankes af på karret, og objektglassene overføres til Wash Buffer (Dako) i mindst 2 minutter.

2. Varmeforbehandling. 1. 2. 3. 4.

Et glas med Pre-Treatment Solution (Dako) anbringes i varmebadet. Temperaturen indstilles til 98,5° C, således at vandbad, kar og buffer opvarmes samtidig. Det kontrolleres, at bufferen er 95-99° C. Objektglassene anbringes i Pre-Treatment Solution. Når temperaturen igen er 95-99° C inkuberes i 10 minutter. Hele karret tages op af vandbadet og afkøler ved stuetemperatur i 15 minutter uden låg. Objektglassene overføres til Wash Buffer i 2 x 2 minutter. Badet skiftes én gang.

3. Pepsinfordøjelse. 1. 2.

Overskydende buffer bankes forsigtigt af objektglasset, der anbringes på varmeblok v 37° C. Tumor dækkes med 5-8 dråber Pepsin (2-8° C), inkuberes 3½ minutter. Pepsinen bankes forsigtigt af, og objektglasset overføres til Wash Buffer i 2 x 3 minutter. Badet skiftes én gang.

4. Dehydrering. 1. 2. 3. 4. 5.

Objektglassene anbringes i kar med 70 % ethanol, står 2 minutter. Objektglassene overføres til kar med 85 % ethanol, står 2 minutter. Objektglassene overføres til kar med 96 % ethanol, står 2 minutter. Vævssnittene lufttørres helt, > 15 min. Imens forvarmes inkuberingskassen, og probemix klargøres.

5. Co-denaturering og hybridisering. 1.

Probe-mix klargøres. Til 10 µL bruges:

7 µL hybridiseringsbuffer 1 µL nalgene H2O 1 µL af hver CEP probe (i alt 2 µL)

(Til 18x18 mm dækglas bruges 5 µL probemix, til runde 12 mm dækglas bruges 2,5-3 µL) Knipses let og spinnes ned før brug. Husk at beskytte proben mod lys vha. sølvpapir. 2. 3. 4.

Proben påføres over tumor, dækglas lægges på og evt. luft trykkes ud med pincet. Forsegles med cykellim. Co-denaturering på varmeblok ved 85° C i 5 minutter. Overføres til hybridisering forvarmet kasse i fugtigt varmeskab ved 42° C natten over.

6. Stringensvask og montering. 1. 2. 3. 4. 5. 6.

Vaskebuffer (0,4 x SSC, 0,3 % Tween) opvarmes til 72° C, temperaturen kontrolleres. Cykellimen fjernes med pincet, dækglasset skubbes forsigtigt af. Præparatet anbringes i vaskebuffer i 2 minutter. Præparatet anbringes i skyllebuffer (2 x SSC, 0,1% Tween) ved stuetemperatur ca. 20 sek. 30 µL DAPI påsættes over tumor, 24x60 mm dækglas lægges over. Efter 15-30 min. i mørke fjernes overskydende DAPI ved at trykke præparatet let mellem to servietter. Klar til mikroskopi!

29

References (1) Isager P. Ocular malignant melanoma - Epidemiological and clinical aspects. Ph.D. thesis. Faculty of Health Sciences, University of Aarhus; 2004. (2) Damato B. Developments in the management of uveal melanoma. Clin Experiment Ophthalmol 2004; 32(6):639-647. (3) Singh AD, Topham A. Incidence of uveal melanoma in the United States: 1973-1997. Ophthalmology 2003; 110(5):956-961. (4) Jensen OA. Malignant Melanoma of the Uveas in Denmark 1943-1952. A Clinical, Histopathological and Prognostic Study. Thesis. Medical Faculty, University of Copenhagen; 1963. (5) Singh AD, Wang MX, Donoso LA, Shields CL, De Potter P, Shields JA. Genetic aspects of uveal melanoma: a brief review. Semin Oncol 1996; 23(6):768-772. (6) Shah CP, Weis E, Lajous M, Shields JA, Shields CL. Intermittent and chronic ultraviolet light exposure and uveal melanoma: a meta-analysis. Ophthalmology 2005; 112(9):1599-1607. (7) Pe'er J. Uveal Nevi versus Uveal Melanomas. In: Jager MJ, Niederkorn JY, Ksander BR, editors. Uveal melanoma: a model for exploring fundamental cancer biology. 1st ed. Lisse, The Netherlands: Swetz and Zeitlinger (Taylor and Francis); 2004. 31-35. (8) Woll E, Bedikian A, Legha SS. Uveal melanoma: natural history and treatment options for metastatic disease. Melanoma Res 1999; 9(6):575-581. (9) Diener-West M, Earle JD, Fine SL, Hawkins BS, Moy CS, Reynolds SM et al. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma, III: initial mortality findings. COMS Report No. 18. Arch Ophthalmol 2001; 119(7):969-982. (10) Diener-West M, Reynolds SM, Agugliaro DJ, Caldwell R, Cumming K, Earle JD et al. Screening for metastasis from choroidal melanoma: the Collaborative Ocular Melanoma Study Group Report 23. J Clin Oncol 2004; 22(12):2438-2444. (11) Kujala E, Makitie T, Kivela T. Very long-term prognosis of patients with malignant uveal melanoma. Invest Ophthalmol Vis Sci 2003; 44(11):4651-4659. (12) Kodjikian L, Grange JD, Rivoire M. Prolonged survival after resection of liver metastases from uveal melanoma and intra-arterial chemotherapy. Graefes Arch Clin Exp Ophthalmol 2005. (13) Singh AD, Topham A. Survival rates with uveal melanoma in the United States: 1973-1997. Ophthalmology 2003; 110(5):962-965. (14) Shields JA, Shields CL, Donoso LA. Management of posterior uveal melanoma. Surv Ophthalmol 1991; 36(3):161-195. (15) Hawkins BS. The Collaborative Ocular Melanoma Study (COMS) randomized trial of pre-enucleation radiation of large choroidal melanoma: IV. Ten-year mortality findings and prognostic factors. COMS report number 24. Am J Ophthalmol 2004; 138(6):936-951. (16) Singh AD, Shields CL, Shields JA. Prognostic factors in uveal melanoma. Melanoma Res 2001; 11(3):255-263. (17) Isager P, Ehlers N, Overgaard J. Prognostic factors for survival after enucleation for choroidal and ciliary body melanomas. Acta Ophthalmol Scand 2004; 82(5):517-525. (18) Prescher G, Bornfeld N, Becher R. Nonrandom chromosomal abnormalities in primary uveal melanoma. J Natl Cancer Inst 1990; 82(22):1765-1769.

30

(19) Prescher G, Bornfeld N, Hirche H, Horsthemke B, Jockel KH, Becher R. Prognostic implications of monosomy 3 in uveal melanoma. Lancet 1996; 347(9010):1222-1225. (20) Sisley K, Rennie IG, Parsons MA, Jacques R, Hammond DW, Bell SM et al. Abnormalities of chromosomes 3 and 8 in posterior uveal melanoma correlate with prognosis. Genes Chromosomes Cancer 1997; 19(1):22-28. (21) 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):1440-1444. (22) White VA, Chambers JD, Courtright PD, Chang WY, Horsman DE. Correlation of cytogenetic abnormalities with the outcome of patients with uveal melanoma. Cancer 1998; 83(2):354-359. (23) Tschentscher F, Husing J, Holter T, Kruse E, Dresen IG, Jockel KH et al. Tumor classification based on gene expression profiling shows that uveal melanomas with and without monosomy 3 represent two distinct entities. Cancer Res 2003; 63(10):2578-2584. (24) Onken MD, Worley LA, Ehlers JP, Harbour JW. Gene expression profiling in uveal melanoma reveals two molecular classes and predicts metastatic death. Cancer Res 2004; 64(20):7205-7209. (25) Parrella P, Sidransky D, Merbs SL. Allelotype of posterior uveal melanoma: implications for a bifurcated tumor progression pathway. Cancer Res 1999; 59(13):3032-3037. (26) Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100(1):57-70. (27) Braga E, Senchenko V, Bazov I, Loginov W, Liu J, Ermilova V et al. Critical tumor-suppressor gene regions on chromosome 3P in major human epithelial malignancies: allelotyping and quantitative real-time PCR. Int J Cancer 2002; 100(5):534-541. (28) Tschentscher F, Prescher G, Horsman DE, White VA, Rieder H, Anastassiou G et al. Partial deletions of the long and short arm of chromosome 3 point to two tumor suppressor genes in uveal melanoma. Cancer Res 2001; 61(8):3439-3442. (29) Baggetto LG, Gambrelle J, Dayan G, Labialle S, Barakat S, Michaud M et al. Major cytogenetic aberrations and typical multidrug resistance phenotype of uveal melanoma: current views and new therapeutic prospects. Cancer Treat Rev 2005; 31(5):361-379. (30) Parrella P, Fazio VM, Gallo AP, Sidransky D, Merbs SL. Fine mapping of chromosome 3 in uveal melanoma: identification of a minimal region of deletion on chromosomal arm 3p25.1-p25.2. Cancer Res 2003; 63(23):8507-8510. (31) Parrella P, Caballero OL, Sidransky D, Merbs SL. Detection of c-myc amplification in uveal melanoma by fluorescent in situ hybridization. Invest Ophthalmol Vis Sci 2001; 42(8):1679-1684. (32) Ehlers JP, Worley L, Onken MD, Harbour JW. DDEF1 is located in an amplified region of chromosome 8q and is overexpressed in uveal melanoma. Clin Cancer Res 2005; 11(10):3609-3613. (33) Anastassiou G, Rebmann V, Wagner S, Bornfeld N, Grosse-Wilde H. Expression of classic and nonclassic HLA class I antigens in uveal melanoma. Invest Ophthalmol Vis Sci 2003; 44(5):2016-2019. (34) Aalto Y, Eriksson L, Seregard S, Larsson O, Knuutila S. Concomitant loss of chromosome 3 and whole arm losses and gains of chromosome 1, 6, or 8 in metastasizing primary uveal melanoma. Invest Ophthalmol Vis Sci 2001; 42(2):313-317. (35) Ghazvini S, Char DH, Kroll S, Waldman FM, Pinkel D. Comparative genomic hybridization analysis of archival formalin-fixed paraffin-embedded uveal melanomas. Cancer Genet Cytogenet 1996; 90(2):95101.

31

(36) Vajdic CM, Hutchins AM, Kricker A, Aitken JF, Armstrong BK, Hayward NK et al. Chromosomal gains and losses in ocular melanoma detected by comparative genomic hybridization in an Australian population-based study. Cancer Genet Cytogenet 2003; 144(1):12-17. (37) Ginzinger DG, Godfrey TE, Nigro J, Moore DH, Suzuki S, Pallavicini MG et al. Measurement of DNA copy number at microsatellite loci using quantitative PCR analysis. Cancer Res 2000; 60(19):5405-5409. (38) Livak KJ. Comparative Ct method. ABI Prism Sequence Detection System User Bulletin no 2 1997. (39) Aarskog NK, Vedeler CA. Real-time quantitative polymerase chain reaction. A new method that detects both the peripheral myelin protein 22 duplication in Charcot-Marie-Tooth type 1A disease and the peripheral myelin protein 22 deletion in hereditary neuropathy with liability to pressure palsies. Hum Genet 2000; 107(5):494-498. (40) Nigro JM, Takahashi MA, Ginzinger DG, Law M, Passe S, Jenkins RB et al. Detection of 1p and 19q loss in oligodendroglioma by quantitative microsatellite analysis, a real-time quantitative polymerase chain reaction assay. Am J Pathol 2001; 158(4):1253-1262. (41) Scholes AG, Damato BE, Nunn J, Hiscott P, Grierson I, Field JK. Monosomy 3 in uveal melanoma: correlation with clinical and histologic predictors of survival. Invest Ophthalmol Vis Sci 2003; 44(3):1008-1011. (42) Hall PA, Going JJ. Predicting the future: a critical appraisal of cancer prognosis studies. Histopathology 1999; 35(6):489-494. (43) Mudhar HS, Parsons MA, Sisley K, Rundle P, Singh A, Rennie IG. A critical appraisal of the prognostic and predictive factors for uveal malignant melanoma. Histopathology 2004; 45(1):1-12. (44) Altman DG, Riley RD. Primer: an evidence-based approach to prognostic markers. Nat Clin Pract Oncol 2005; 2(9):466-472. (45) Singh AD, Borden EC. Metastatic uveal melanoma. Ophthalmol Clin North Am 2005; 18(1):143-50, ix. (46) Assessment of metastatic disease status at death in 435 patients with large choroidal melanoma in the Collaborative Ocular Melanoma Study (COMS): COMS report no. 15. Arch Ophthalmol 2001; 119(5):670-676. (47) Naus NC, Verhoeven AC, van Drunen E, Slater R, Mooy CM, Paridaens DA et al. Detection of genetic prognostic markers in uveal melanoma biopsies using fluorescence in situ hybridization. Clin Cancer Res 2002; 8(2):534-539.

32