The objectives of the study are the morpho-phenotypic evaluation of uveal melanoma and to identify molecular prognostic factors that may be correlated with disease severity, tumour progression and response to treatment.

These objectives will be achieved through immunohistochemical and genetic analyses.

Tissues from surgical specimens regardless of the stage of the disease, fixed in formalin and included in paraffin from samples already collected for clinical practice and kept in the archives of the UOC of Pathological Anatomy of the Fondazione Policlinico Universitario A. Gemelli IRCCS will be used for the study, subject to the patient's informed consent to the use of these samples for research purposes. The UOC Anatomia Patologica will take care of the recovery of archive material (slides).

The pseudoanonymised slides, which will be prepared by clinical practice, will be sent to the Institute of Pathological Histology and Molecular Diagnostics of the Azienda Ospedaliero- Universitaria Careggi, Florence for immunohistochemical analysis, which will be carried out as reported below. Once the following analysis has been performed, the slides will be destroyed.

Sections of 3 μm thickness derived from FFPE samples will be prepared for immunohistochemical analysis. Sample processing will be performed using an automated immunostainer (Ventana Discovery ULTRA, Ventana Medical Systems, Tucson, AZ). Sections will be deparaffinised in EZ prep (#950-102; Ventana Medical Systems, Tucson, AZ), followed by treatment with Cell Conditioning 1 (CC1) buffer (#950-124; Ventana Medical Systems, Tucson, AZ) in order to promote antigenic recovery. Finally, primary antibodies will be dispensed according to the desired staining and the signal will be developed with Ultramap anti-mouse or anti-rabbit detection kits. For TME characterisation, samples will be stained with the following primary antibodies: CD3, CD4, CD8, CD68, CD163. All sections processed with IHC will be digitally scanned in Whole Slide Images (WSI) using the Aperio AT2 platform (Leica Biosystems, Wetzlar, DE).Digital Pathology, which will be performed at the University of Florence, is an emerging discipline that allows the quantitative analysis of digital images using highly standardised approaches. It will allow quantitative measurements of the interactions between immune cells and characterise their relative spatial distribution. In addition, it will be able to generate further information that will allow the TME of the MU to be studied in depth.

The HALO image analysis platform, present at the University of Florence, will allow: i) cell quantification, evaluation of the intensity of the marking, recognition of cell compartments (nucleus, cytoplasm, membrane); ii) spatial analysis through the identification of the relative distributions of cells in the intra- and peri-tumour portion; iii) sharing of the database generated together with the digital images.

Samples labelled with single or multiple IHC will be evaluated using this automated digital quantification system in order to characterise the various cell subsets that make up the TME. By combining multiple immunohistochemistry with digital analysis, the data will be accurately analysed, thus producing a new analysis work-flow for finding new signatures. All data derived from the digital analysis will be correlated with the progression of the pathology, with the aim of identifying new prognostic factors in the pathology of MU.

The analysis for the identification of cytogenetic alterations, whose data are already available to us since they are regularly performed in clinical practice, was conducted on the entire cohort of patients enrolled for the study at the Genetics Institute of the Fondazione Policlinico Gemelli on fresh tissue samples taken in the operating theatre as per clinical practice. Array-CGH methods on oligonucleotide platforms from Agilent Technologies were used. Tumour DNA from MU samples was extracted using the Gentra Pure Gene commercial kit (QIAGEN) as per clinical practice; once extracted, the quantity and purity of the DNA obtained was evaluated. All samples deemed suitable for quality were processed following the protocol indicated by Agilent Technologies. The data obtained were analysed using Cytogenomics software. These analyses will allow a genotypic characterisation of the tumour to be obtained.

Inclusion Criteria:

1. Age 18 years or older;
2. Clinical and histological diagnosis of uveal melanoma;
3. Ability to obtain formalin-fixed, paraffin-embedded tissue (FFPE).
4. Availability of genetic analysis results from tumour tissue obtained by enucleation or fine-needle biopsy.
5. Follow-up of at least 5 years from enucleation or fine-needle biopsy.
6. Written informed consent to use the data for research purposes.

Exclusion Criteria:

1. Age < 18 years;
2. Lack of sufficient clinical data on selected patients.
3. Follow-up of less than 5 years.
4. Refusal to give informed consent to data processing for research purposes.

Patients diagnosed with uveal melanoma undergoing histological and cytogenetic analysis will be included in the study. The cases will be identified among the patients referred to the Ocular Oncology Unit during the period from January 2008 to December 2019.