High expression of glycolytic and pigment proteins is associated with worse clinical outcome in stage III melanoma.

Falkenius J, Lundeberg J, Johansson H, Tuominen R, Frostvik-Stolt M, Hansson J, Egyhazi Brage S

Melanoma Res. 23 (6) 452-460 [2013-12-00; online 2013-10-17]

There are insufficient numbers of prognostic factors available for prediction of clinical outcome in patients with stage III malignant cutaneous melanoma, even when known adverse pathological risk factors, such as macrometastasis, number of lymph node metastases, and ulceration are taken into consideration. The aim of this study was therefore to identify additional prognostic factors to better predict patients with a high risk of relapse, thus enabling us to better determine the need for adjuvant treatment in stage III disease. An RNA oligonucleotide microarray study was performed on first regional lymph node metastases in 42 patients with stage III melanoma: 23 patients with short-term survival (≤ 13 months) and 19 with long-term survival (≥ 60 months), to identify genes associated with clinical outcome. Candidate genes were validated by real-time PCR and immunohistochemical analysis. Several gene ontology (GO) categories were highly significantly differentially expressed including glycolysis (GO: 0006096; P<0.001) and the pigment biosynthetic process (GO: 0046148; P<0.001), in which overexpression was associated with short-disease-specific survival. Three overexpressed glycolytic genes, GAPDHS, GAPDH, and PKM2, and two pigment-related genes, TYRP1 and OCA2, were selected for validation. A significant difference in GAPDHS protein expression between short- and long-term survivors (P=0.021) and a trend for PKM2 (P=0.093) was observed in univariate analysis. Positive expression of at least two of four proteins (GAPDHS, GAPDH, PKM2, TYRP1) in immunohistochemical analysis was found to be an independent adverse prognostic factor for disease-specific survival (P=0.011). Our results indicate that this prognostic panel in combination with established risk factors may contribute to an improved prediction of patients with a high risk of relapse.

NGI Stockholm (Genomics Applications)

NGI Stockholm (Genomics Production)

National Genomics Infrastructure

PubMed 24128789

DOI 10.1097/CMR.0000000000000027

Crossref 10.1097/CMR.0000000000000027