Discovery of reliable biomarkers in melanoma patients is still awaited, as we need to enhance our ability to predict their outcome. Biomarkers can be divided into diagnostic markers, for screening; prognostic markers, which can be used once the cancer has been diagnosed; and predictive markers, which should predict the likely response to a treatment.
Cancer biomarkers include molecular tools, such as proteins, peptides, DNA, mRNA, or processes that can be measured in a given cancer with specific quantitative and qualitative tools .
Actually, no ideal serological biomarker exists today for melanoma patients even if the use of many molecules has been proposed, for example, melanin metabolites, cytokines and cytokine receptors, neuron-specific enolase, melanoma inhibitory activity, cell adhesion molecules, and YKL-40. Despite this, only three molecules currently remain the basis for a comparison of a new biomarker: lactate dehydrogenase (LDH), C-reactive protein (CRP), and S100B .
LDH has been considered as the main serum parameter in metastatic melanoma patients. Some studies have presented LDH as the most important independent prognostic factor, and this led to a stratification in the new American Joint Commission on Cancer (AJCC) staging system . Metastatic melanoma patients with high LDH levels are designated as M1c whatever the site of metastasis. However, Hamberg et al.  recently indicated that in a series of 53 AJCC stage IV melanoma patients, only 38% had high levels of LDH, suggesting that elevated LDH is a late event in the natural course of the disease. Moreover, in a multivariate analysis of 64 AJCC stage IV melanoma patients, Hauschild et al.  failed to demonstrate the independent prognostic value of LDH.
High serum CRP levels have been linked to poor prognosis in various neoplasia. In a recent report, Deichmann et al.  analyzed the prognostic significance of CRP compared with LDH in AJCC stage IV melanoma patients and pointed to the superiority of CRP. With a cut-off point of 3 mg/l, serum analysis discriminated between stage IV and nonstage IV melanoma patients, with a sensitivity of 76.9% and a specificity of 90.4%. CRP is a nonspecific inflammatory parameter, which might have a role in the detection of melanoma progression. This protein is produced by hepatocytes as a nonspecific acute-phase response of inflammation processes.
S100B protein is a 21-kDa dimeric protein, consisting of two β subunits. This protein is a member of a family of 19 proteins, and was first isolated from bovine brain in the mid-1960s. S100B protein is expressed by glial cells and melanocytes and has been shown to be produced in brain tumors and melanoma. Roles for S100B are probably multiple and underestimated. It can interact with the p53 tumor suppressor gene in a calcium-dependent manner.
The serum S100B level is linked to the tumor burden and reflects clinical stage and tumor progression as reported by some. Many reports have shown that S100B levels are correlated with clinical stage (the higher the level, the more advanced the stage) and could be used to monitor the effectiveness of antitumoral treatment whatever the type of treatment (surgical, chemotherapy, immunotherapy) [7,8]. Mohammed et al. [8–10] have even suggested the use of S100B instead of LDH in the AJCC staging system, whereas other authors consider that S100B does not have any added value when comparing its sensitivity and specificity to CRP and LDH. This also underlines the importance of specifying the group of patients for who a biomarker can be used. S100B prognostic value seems limited to advanced stage III and stage IV melanoma patients.
Gal-3 is a galactoside-binding protein, which can be released into the extracellular compartment by both melanoma and inflammatory cells [11–13]. Higher Gal-3 serum levels have recently been shown in advanced melanoma patients but the proof of a usefulness of this protein as a biomarker is still to be determined [14–16]. To clarify the prognostic role of serum Gal-3, we conducted this study and compared the Gal-3 serum levels with the most commonly used biomarkers in melanoma patients.
Materials and methods
Serum samples and measurement of galectin-3
A total of 83 serum samples from locoregional (AJCC stage III) and distant (AJCC stage IV) metastatic melanoma patients were taken . All individuals gave their informed consent for specific blood analysis according to local ethics committee. Serum samples were obtained from clotted blood after centrifugation at 4°C and then stored at −80°C until analysis. Concentrations of Gal-3 were measured by enzyme-linked immunosorbent assay test as explained by the manufacturer's instructions (BMS 279, from Bender MedSystems GmbH, Vienna, Austria). Determinations of CRP, LDH and S100B were assessed in triplicate by the modular system analyser (P module) from Roche Diagnostics GmbH, Mannheim, Germany. Clinical and serum data are summarized in Table 1.
Analysis of data
The spearman test was used to study the relationship between Gal-3 and S100B serum levels. To investigate the impact of Gal-3 serum level on patient prognosis, survival curves were plotted according to Kaplan–Meier, and the Cox proportional hazards model was used to determine the prognostic importance of serum Gal-3, S100B, CRP, and AJCC stage.
Correlation between Gal-3 and S100B serum levels
There was a significant correlation between Gal-3 and S100B levels (r=0.328, P<0.001, Fig. 1) as well as between Gal-3 and CRP, and Gal-3 and LDH (r=0.239, P<0.05 and r=0.505, P<0.0001, respectively), as already shown in a small group of patients.
Kaplan–Meier survival curves according to serum Gal-3levels (Fig. 2)
On the basis of previously published data, three groups of patients were defined according to Gal-3 levels: less than 8 ng/ml (group 0), 8–10 ng/ml (group 2), and more than 10 ng/ml (group 3). The first two groups had a similar overall survival whereas the latter one, corresponding to a high Gal-3 serum level, had the worst outcome: among 17 patients only two survived, and the median survival was 4.1 months.
Results of univariate and multivariate analysis (Table 2)
To evaluate prognostic importance regarding overall survival, two cut-off values for Gal-3 levels were chosen: 8 and 10 ng/ml. For the other markers the cut-off values were the values considered as ‘pathological’, above the upper limit, namely 0.1 μg/ml for S100B, 1 mg/dl for CRP and 480 UI/l for LDH. Because only eight patients had pathological LDH levels, LDH was excluded form the analysis.
In the univariate analysis (cut-off for Gal-3=8 ng/ml), each of these four variables (Stage, S100B, CRP, Gal-3) seemed to be of significant prognostic value for overall survival. However, by using a multivariate Cox proportional hazards model, among these four variables AJCC stage and CRP serum levels seemed to be the most important independent prognostic factors [hazard ratio (HR)=9.60, P=0.0002 and HR=2.75, P=0.002, respectively]. When these two variables and Gal-3 were included in the Cox model, the relative prognostic importance of Gal-3 was only marginally significant (HR=1.74, P=0.07).
In the second analysis, we chose a cut-off value of 10 ng/ml for Gal-3, as suggested by the results of the Kaplan–Meier analysis, and we were able to demonstrate in the quatri-variate analysis that serum Gal-3 had a strong independent prognostic value, superior to the other markers (HR=4.64, P=0.0001).
Biomarker discovery is a complex research process, which involves scientific collaboration and data share . Routine use of tumor markers is an important issue because it would allow early detection and definition of therapeutic strategy. This is particularly important in the management of melanoma patients for whom, at least for a subgroup of them, early treatment has been shown to prolong disease-free and overall survival. The ideal serum biomarker should be a molecule, detection of which in the blood allows diagnosis of a growing tumor in a patient. The biomarker must exhibit sufficient sensitivity and specificity to minimize false-negative and false-positive results. The sensitivity refers to the proportion of patients with a confirmed disease who will have a positive test for a biomarker, whereas the specificity can be defined by the proportion of healthy individuals with a negative test. Former studies have shown the presence of many molecules in the serum of cancer patients, but the clinical/practical significance of these is still questionable. These molecules can be produced and secreted or shed into the bloodstream directly by melanoma cells or indirectly through destruction of melanoma cells by chemotherapy, immunotherapy or combined therapy. The most commonly used biomarker single molecules are a matter of debate, but nowadays no melanoma biomarker has been unanimously accepted for its sensitivity and specificity in predicting the outcome of patients. Thus, biomarker research should be extensively developed in the field of melanoma [2,17].
Gal-3 is a 31-kDa multifaceted protein, which might be important in melanoma progression process [18–21]. Its participation to many steps (cell proliferation, cell apoptosis, cell migration, angiogenesis) of the cancer progress has been established [22–27]. The different roles of this protein are currently investigated. Gal-3 is a member of the family of galectins, which contains β-galactoside-binding lectins characterized by conserved sequence parts in the carbohydrate-binding sites (carbohydrate recognition domain) . It is increasingly accepted that cancer progression is facilitated by change of the expression pattern of the galectins, as this can modify the ability of cancer cells to penetrate into the extracellular matrix, to enter into blood vessels and lymphatics, and to generate secondary sites. Gal-3 has one carbohydrate recognition domain at its C-terminal part, which is linked to R domain, which contains aminoacid repeats and an aromatic structure, and can be degradated by matrix metalloproteinases . The N domain is defined by a short N-terminal sequence of unrelated structure. One recent striking finding was the association of a high serum level of Gal-3 to an advanced stage, and this was an argument to study its prognostic value [13,16].
In this study we evaluated serum Gal-3 levels in a series of 83 AJCC stages III and IV melanoma patients and three markers (LDH, CRP, S100B) were selected for comparison. We previously showed that Gal-3 levels were significantly correlated to LDH and CRP levels in AJCC stage IV melanoma patients, and we confirm this observation in this large group of patients, as well as a significant correlation with S100B . The cut-off values for Gal-3 were chosen according to previously published works and results of a Kaplan–Meier analysis. In the first multivariate Cox proportional hazard model, when the cut-off value of Gal-3 was 8 ng/ml, we found only stage and CRP levels to be independently and significantly associated with the prognosis while in a second multivariate analysis (cut-off value 10 ng/ml), we finally demonstrated a strong independent prognostic value for Gal-3. These results are consistent with the importance to define cautiously the choice of the cut-off value of a protein, which can be found in the serum of nonmetastatic patients and healthy individuals.
In summary, according to our data, one might speculate that Gal-3 serum level could be measured routinely. This study indicates a strong independent prognostic value for serum Gal-3 in advanced melanoma patients, when the chosen cut-off value is 10 ng/ml, and we believe that further study of this protein may be interesting to understand melanoma patients' outcome. Medical literature also provides data that show that, despite a possible usefulness in prognostic assessment of advanced patients, Gal-3 serum measurements seem not to be helpful in the early diagnosis of cancer such as melanoma, nor to predict response to therapy. Indeed, in a very recent study Demotte et al. , from the Ludwig Institute for Cancer Research, demonstrate that Gal-3 interferes with the TCR-CD8 colocalization in cytolytic T-lympocytes, supporting the hypothesis of a role for the protein in conferring anergy in tumors. Melanoma biomarker research is an open field for the comprehension of molecular events in melanoma progression and should, in the future, provide new molecular targets for therapeutic intervention in advanced stages at which treatment options are poor. Further studies will uncover whether Gal-3 will be able to open new therapeutic perspectives.
This study was supported by a grant from the Erasmus Foundation, Brussels, Belgium (PV). The authors are grateful to Dr Meunier JC, CHU-Charleroi (Charleroi, Belgium) and to Mrs Suzanne Gruber, Roche-Diagnostics for providing us the S100B Elisa Kits.
1. Hamdan MH. Cancer biomarkers
: analytical techniques for discovery. Hoboken, New Jersey (USA): John Wiley & Sons, Inc; 2007.
2. Brochez L, Naeyaert JM. Serological markers for melanoma
. Br J Dermatol 2000; 143:256–268.
3. Balch CM, Buzaid AC, Soong SJ, Atkins MB, Cascinelli N, Coit DG, et al. Final version of the American-Joint Committee on cancer staging sytem for cutaneous melanoma
. J Clin Oncol 2001; 19:3635–3648.
4. Hamberg AP, Korse CM, Bonfrer JM, de Gast GC. Serum S100B
is suitable for prediction and monitoring of response to chemoimmunotherapy in metastatic malignant melanoma
Res 2003; 13:45–49.
5. Hauschild A, Michaelsen J, Brenner W, Rudolph P, Gläser R, Henze E, et al. Prognostic significance of serum S100B
detection compared with routine blood parameters in advanced metastatic melanoma
Res 1999; 9:155–161.
6. Deichmann M, Kahle B, Moser K, Wacker J, Wüst K. Diagnosing melanoma
patients entering American Joint Committee on Cancer stage IV, C-reactive protein
in serum is superior to lactate dehydrogenase
. Br J Cancer 2004; 91:699–702.
7. Buer J, Probst M, Franzke A, Duensing S, Haindl J, Volkenandt M, et al. Elevated serum levels of S100 and survival in metastatic malignant melanoma
. Br J Cancer 1997; 75:1373–1376.
8. Deichmann M, Benner A, Bock M, Jackel A, Uhl K, Waldmann V, et al. S100-beta, melanoma
-inhibiting activity and lactate dehydrogenase
discriminate progressive from non-progressive American Joint Committee on Cancer stage IV melanoma
. J Clin Oncol 1999; 17:1891–1896.
9. Mohammed MQ, Abraha HD, Sherwood RA, MacRae K, Retsas S. Serum S100beta protein as a marker of disease activity in patients with malignant melanoma
. Med Oncol 2001; 18:109–120.
10. Banfalvi T, Boldizsar M, Gergye M, Gilde K, Kremmer T, Otto S. Comparison of prognostic significance of serum 5-S-Cysteinyldopa, LDH and S100B
protein in stage III–IV malignant melanoma
. Pathol Oncol Res 2002; 8:183–187.
11. Barondes SH, Castronovo V, Cooper DN, Cummings RD, Drickamer K, Feizi T, et al. Galectins: a family of animal beta-galactoside-binding lectins. Cell 1994; 76:597–598.
12. Hughes RC. Secretion of the galectin family of mammalian carbohydrate-binding proteins. Biochim Biophys Acta 1999; 1473:172–185.
13. Vereecken P, Zouaoui Boudjeltia K, Debray C, Awada A, Legssyer I, Sales F, et al. High serum galectin-3
in advanced melanoma
: preliminary results. Clin Exp Dermatol 2006; 31:105–109.
14. Iurisci I, Tinari N, Natoli C, Angelucci D, Cianchetti E, Iacobelli S. Concentrations of galectin-3
in the sera of normal controls and cancer patients. Clin Cancer Res 2000; 6:1389–1393.
15. Vereecken P, Heenen M. Serum galectin-3
in advanced melanoma
patients: a hypothesis on a possible role in melanoma
progression and inflammation. J Int Med Res 2006; 34:119–120.
16. Vereecken P, Ghanem G, Morandini R, Suciu S, Van Baren N, Heenen M. Defining the prognostic value for galectin-3
in cutaneous melanoma
. J Int Med Res 2007; 35:731–732.
17. Hoon DS, Bostick P, Kuo C, Okamoto T, Wang HJ, Elsahoff R, et al. Molecular markers in blood as surrogate prognostic indicators of melanoma
recurrence. Cancer Res 2000; 60: 2253–2257.
18. Rubinstein N, Ilarregui JM, Toscano MA, Rabinovich GA. The role of galectins in the initiation, amplification and resolution of the inflammatory response. Tissue Antigens 2004; 64:1–12.
19. Almkvist J, Karlsson A. Galectins as inflammatory mediators. Glycoconj J 2004; 19:575–581.
20. Mey A, Berthier-Vergnes O, Apoil PA, Doré JF, Revillard JP. Expression of the galactose binding protein Mac-2 by human melanoma
cell lines. Cancer Lett 1994; 81:155–163.
21. Vereecken P, Debray C, Petein M, Awada A, Lalmand MC, Laporte M, et al. Expression of galectin-3
in primary and metastatic melanoma
: immunohistochemical studies on human lesions and nude mice xenograft tumors. Arch Dermatol Res 2005; 296:353–358.
22. Yang RY, Liu FT. Galectins in cell growth and apoptosis. Cell Mol Life Sci 2003; 60:267–276.
23. Barondes SH. Galectins: a personal overview. TIGG 1997; 9:1–7.
24. Barondes SH, Cooper DNW, Gitt MA, Leffler H. Galectins: structure and function of a large family of animal lectins. J Biol Chem 1994; 269:20807–20810.
25. Debray C, Vereecken P, Belot N, Teillard P, Brion JP, Pandolfo M, et al. Multifaceted role of galectin-3
on human glioblastoma cell motility. Biochem Biophys Res Commun 2004; 325:1393–1398.
26. Califice S, Castronovo V, Van Den Brule F. Galectin-3
and cancer (Review). Int J Oncol 2004; 25:983–992.
27. Takenaka Y, Fukumori T, Raz A. Galectin-3
and metastasis. Glyconj J 2004; 19:543–549.
28. Ochieng J, Green B, Evans S, James O, Warfield P. Modulation of the biological functions of galectin-3
by matrix metalloproteinases. Biochim Biophys Acta 1998; 1379:97–106.
29. Demotte N, Stroobant V, Courtoy PJ, Van Der Smissen P, Colau D, Luescher IF, et al. Restoring the association of the T cell receptor with CD8 reverses anergy in human tumor-infiltrating lymphocytes. Immunity 2008; 28:414–424.