Tumor microenvironment-derived S100A8/A9 is a novel prognostic biomarker for advanced melanoma patients and during immunotherapy with anti-PD-1 antibodies.

BACKGROUND: Predicting metastasis in melanoma patients is important for disease management and could help to identify those who might benefit from adjuvant treatment. The aim of this study was to investigate whether the tumor microenvironment-derived protein S100A8/A9 qualifies as prognostic marker for melanoma patients, also in the setting of immunotherapy. METHODS: S100A8/A9 gene and protein expression were analyzed on melanocytic nevi, primary melanomas and metastases using a cDNA library and three independent tissue-microarrays (TMA). Serum levels of S100A8/A9 were measured using a specific ELISA in two independent cohorts of 354 stage III and stage IV melanoma patients as well as in two independent cohorts of patients treated with the PD-1 antibody pembrolizumab. RESULTS: cDNA analysis revealed an upregulation of S100A8 and S100A9 gene expression in melanoma metastases compared to primary melanomas. Significantly higher numbers of infiltrating S100A8/A9 positive cells were found in tissue samples of metastasizing primary melanomas compared to non-metastasizing melanomas (P < .0001) and in melanomas of short-term survivors compared to long-term survivors (P < .0001). Serum S100A8/A9 levels > 5.5 mg/l were associated with impaired overall survival in two independent cohorts (both P < .0001). Importantly, patients with serum elevated S100A8/A9 treated with pembrolizumab showed significantly impaired survival compared to patients with lower S100A8/A9 levels (cohort 1: P = .0051; cohort 2: P < .0001). CONCLUSIONS: The tumor microenvironment-associated protein S100A8/A9 serves as a novel prognostic marker for metastasis and survival of metastatic melanoma patients and predicts response to immunotherapy with pembrolizumab. These data underscore the significance of tumor microenvironment-derived factors as suitable biomarkers for melanoma.

Biomarker value and pitfalls of serum S100B in the follow-up of high-risk melanoma patients.

BACKGROUND AND OBJECTIVES: Serum levels of S100B are standard in monitoring advanced malignant melanoma patients in order to discriminate progressive from non-progressive disease. False-positive results lead to distress among patients and increase the amount of cost-intensive diagnostics. We therefore analyzed reported comorbid diseases as putative sources of excessive S100B release. PATIENTS AND METHODS: Here, we report a single-center experience on serum S100B levels in 2,664 blood samples from 1,113 stage IB to IV melanoma patients (AJCC) who presented for follow-up examinations over a period of 24 months. RESULTS: Overall, 295 (11%) of patients developed disease progression. In patients with a high tumor load, the rate of false-negative results was 30/185 (16%). The rate of false-positive results was 247/2369 (12%). One hundred and six false-positive results (69%) compared to 46 true-positive results (31%) were found in patients with cardiovascular diseases such as arrhythmia (50/32) or previous myocardial infarction (22/14). Moreover, obesity (85/14), liver cirrhosis (31/10), migraine (18/2), chronic kidney disease (13/2), and previous stroke (11/1) were found to be associated with false-positive S100B levels. CONCLUSIONS: Serum S100B is a useful quantitative biomarker in routine follow-up of high-risk melanoma patients. While false-negative results are frequent in patients with low tumor load, false-positive results are associated with several comorbid diseases and warrant careful reevaluation.

Myeloid Cells and Related Chronic Inflammatory Factors as Novel Predictive Markers in Melanoma Treatment with Ipilimumab.

PURPOSE: Immunotherapy with ipilimumab improves the survival of patients with metastatic melanoma. Because only around 20% of patients experience long-term benefit, reliable markers are needed to predict a clinical response. Therefore, we sought to determine if some myeloid cells and related inflammatory mediators could serve as predictive factors for the patients' response to ipilimumab. EXPERIMENTAL DESIGN: We performed an analysis of myeloid cells in the peripheral blood of 59 stage IV melanoma patients before the treatment and at different time points upon the therapy using a clinical laboratory analysis and multicolor flow cytometry. In addition, the production of related inflammatory factors was evaluated by ELISA or Bio-Plex assays. RESULTS: An early increase in eosinophil count during the treatment with ipilimumab was associated with an improved clinical response. In contrast, elevated amounts of monocytic myeloid-derived suppressor cells (moMDSC), neutrophils, and monocytes were found in nonresponders (n = 36) as compared with basal levels and with responding patients (n = 23). Moreover, in nonresponders, moMDSCs produced significantly more nitric oxide, and granulocytic MDSCs expressed higher levels of PD-L1 than these parameters at baseline and in responders, suggesting their enhanced immunosuppressive capacity. Upon the first ipilimumab infusion, nonresponders displayed elevated serum concentrations of S100A8/A9 and HMGB1 that attract and activate MDSCs. CONCLUSIONS: These findings highlight additional mechanisms of ipilimumab effects and suggest levels of eosinophils, MDSCs, as well as related inflammatory factors S100A8/A9 and HMGB1 as novel complex predictive markers for patients who may benefit from the ipilimumab therapy. Clin Cancer Res; 21(24); 5453-9. ©2015 AACR.

Diminished levels of the soluble form of RAGE are related to poor survival in malignant melanoma.

RAGE is a central driver of tumorigenesis by sustaining an inflammatory tumor microenvironment. This study links the soluble forms of RAGE (sRAGE and esRAGE) with clinical outcome of melanoma patients. Moreover, tissue expression of RAGE was analyzed using immunohistochemistry on two independent tissue microarrays (TMA) containing 35 or 257 primary melanomas, and 41 or 22 benign nevi, respectively. Serum concentrations of sRAGE and esRAGE were measured in 229 Stage III-IV patients using ELISA and plasma concentrations of sRAGE were analyzed in an independent second cohort with 173 samples of Stage I-IV patients. In this cohort, three well-described SNPs in the RAGE gene were analyzed. RAGE protein expression was highly upregulated in primary melanomas compared to benign nevi in the two TMA (p < 0.001 and p = 0.005) as well as in sun-exposed melanomas (p = 0.046). sRAGE and esRAGE were identified as prognostic markers for survival as diminished sRAGE (p = 0.034) and esRAGE (p = 0.012) serum levels correlated with poor overall survival (OS). Multivariate Cox regression analysis showed that diminished serum sRAGE was independently associated with poor survival (p = 0.009). Moreover, diminished sRAGE was strongly associated with impaired OS in the second cohort (p < 0.001). Multivariate Cox regression analysis including the investigated SNPs revealed an independent correlation of the two interacting promoter SNPs with impaired OS. In conclusion, the soluble forms of RAGE and variants in its genetic locus are prognostic markers for survival in melanoma patients with high risk for progression.