The incidence of cutaneous malignant melanoma is rising, and tumour markers are attracting attention as a possible alternative to clinical examination in the follow-up situation. S100 is the preferred marker for malignant melanoma, and correlation between serum S100 and disease relapse and survival has been reported. S100 tests previously used in clinical studies were specified poorly regarding reactivity with S100A1B and S100BB. In this study, a newly designed S100 assay (designed to measure exclusively S100A1B and S100BB) and two newly developed serological assays, S100A1B, and S100BB, were investigated postoperatively in patients undergoing radical surgery for cutaneous malignant melanoma. Additionally, immunohistochemical analysis of S100A4 was performed on the primary malignant melanoma using tissue microarrays. The primary aim of the study was to investigate whether any of these assays, either singly or in combination, can contribute additional information concerning increased risk of relapse and death because of malignant melanoma. In total, 98 patients (54 males, 44 females) with malignant melanoma were included in the study. As a continuous variable, S100BB (P=0.016) was associated statistically with increased risk of relapse; this was not the case for increased values of either S100 (P=0.11) or S100A1B (P=0.92). The Kaplan-Meier overall survival as well as disease specific survival curve for the S100 serum level demonstrated a statistically significant association with better survival if the patient had a S100 level <or=150 ng/l (P<0.001). Survival analyses for S100A1B using a defined cutoff of 50 ng/l showed a statistically significant association concerning overall and disease specific survival (P<0.001). Furthermore, S100BB was associated with overall and disease specific survival using a defined cutoff of 50 ng/l (P<0.001). No statistically significant correlation was found between S100A4 and overall survival (P=0.96) and there was no correlation between elevated levels of S100 and the immunohistochemical staining of S100A4 (P=0.1), nor for serum S100A1B (P=0.1) nor serum S100BB (P=0.17). Circulating S100A1B and S100BB are potential biomarkers in patients with malignant melanoma. S100BB should be considered as the preferred biomarker, showing potential in predicting both relapse and survival, in contrast to both S100 and S100A1B.