ASSESSMENT OF THE PROGNOSTIC POWER OF NEW BIOMARKERS

IN UVEAL MELANOMA 

​

 

Uveal melanoma (MU) is the most common primary intraocular tumour in adulthood, with an incidence of approximately 7 cases per million population per year, and a mortality of 30% at 5 years after diagnosis and treatment (Singh et al., 2005). Despite the advances that have been made in recent decades, especially in forms of local therapy, the vital prognosis of patients with MU has not changed in the last 50 years. Epidemiological studies indicate that MU is more common in older people and Caucasians, especially those with iridescent blue-grey eyes; there is no gender predominance and MU is rarely caused by genetic mutations. Its most frequent location is in the choroid (90% of cases), although it can also occur in the ciliary body (5-10%) and in the iris (3%). The MU with the best prognosis are those located in the iris and those with the worst prognosis are those in the ciliary body. Most of these tumors are initially asymptomatic. With their growth, ophthalmological signs and symptoms appear, ranging from spots on the iris to significant losses of visual acuity.

​

 

Although MU and cutaneous melanoma originate from a common cellular precursor, melanocyte, they have important differences in tumor biology and behavior, including molecular characteristics (Iwamoto et al., 2000). Unlike cutaneous melanoma, MU is primarily blood-borne and its metastases are preferentially located in liver tissue. Systemic-spreading disease is the leading cause of death in MU, and may appear after a long interval of time, suggesting the presence of hidden micrometastases at the time the primary tumor is diagnosed and treated (COMS report No 18, 2001).

​

 

Eyeball enucleation was the first accepted standard treatment for MU and currently remains the treatment of choice for large tumors. However, conservative treatments have emerged to avoid this form of radical therapy. These include: I125 brachytherapy, proton beam, trans-scleral or trans-retinal resection, and combinations of the above with transpupillary thermotherapy. Currently, using any of the conservative treatment options or their combinations, the degree of local control is greater than 90% after 5 years. However, despite this success, metastases occur by hematogenous route in more than 50% of patients after 15 years (Lorigan et al., 1991). 

 

The Adult Intraocular Tumor Group has focused in the last decade on the study of molecular factors associated with the development of MU (Pardo et al., 2004; 2005; 2006; 2007; Bande et al., 2012). We first studied proteins related to the G1/S phases of the cell cycle in primary UM cultures established from enucleated eyes. We assess the proliferation of these cells and the dysregulation of proteins that control the cell cycle such as cyclins, cdks and cdk inhibitors. In addition, we paid special attention to the anti-tumor protein Rb, which did not interact with the transcription factor E2F in certain UM. We found that the inactivation of Rb produced by the interaction with other proteins prevented its anti-tumor effect in the cell cycle of cultured UM cells (Pardo et al., 2004). After this study we decided to apply the new technology of Proteomics to study UM. This allowed us to identify a large number of new proteins not previously related to UM by global and differential proteomic analysis in UM cells. Our goal was to identify specific molecular markers involved in the development and metastasis of UM. More specifically, we initially analyzed the UM proteome from established primary cultures. In this way we were able to identify 683 proteins corresponding to 393 genes; 18% were related to cancer, including those involved in invasion and metastasis (alpha-actin, ezrin, protomyosin, vimentin, MUC-18), cell division and proliferation (CENP-F, Op18, p80), drug resistance (P-glycoprotein, gp94), oncogenes (DJ-1, EMS1, FUS), etc. The publication of these findings constituted for the first time the use of Proteomics for the study of UM (Pardo et al., 2005). After the study of the proteome, we characterized the invasion phenotype by differential proteomic analysis comparing cultures with high and low invasion potential in Matrigel assays. This study showed us that 290 proteins differentiated these cultures; most of them related to cell metabolism, but also to cell adhesion and migration. We have published the list of all these proteins focusing on the validation of the invasion markers HMG-1 and MUC18 and the oncogene DJ-1. In the case of DJ-1, we not only saw that cells release this protein to the extracellular environment, but also that this protein could be detected by Western blot in the serum of some patients (Pardo et al., 2006; 2007). In relation to this oncogene, our group has recently found that this protein could constitute an important blood biomarker as a sign of risk of evolution from nevi to choroidal melanoma, that is, the malignization of a benign tumor (Bande et al., 2012). Thus we have described for the first time how patients with choroidal nevi presented significantly elevated circulating levels of DJ-1 in cases with symptoms (ophthalmological signs), nevus size greater than 1.5mm, a basal diameter greater than 8mm, and presence of intra-occlusive acoustic shadowing.