We investigated the predictive value of the protein S100 as a tumor-marker in the post-surgical follow-up staging of patients with High Risk melanomas (Clark Levels IV/V, Thickness > 0.75 mm).
Methods: In 51 follow-up studies, performed in 50 patients, serum concentrations of the protein S100 were measured and evaluated with respect to their diagnostic value based on the findings of whole body F-18-FDG-PET studies. The findings of FDG-PET were correlated with the findings of CT, MRI, lymph node sonography, bone scintigraphy, and with histological findings, or a follow-up staging after at least 6 months, respectively. Thus, FDG-PET exhibited a sensitivity of 100% and a specificity of 95%. S100 concentrations were measured by an established RIA (Ria-mat Sangtec 100) and by a novel LIA (Lia-mat Sangtec 100 beta). Subsequently, S100 was compared with thymidine kinase for the predictive values of both tumor-markers in the follow-up of malignant melanoma.
Results: Intraindividual comparison of S100 concentrations measured by LIA with those obtained by RIA showed a calculated correlation coefficient of 0.97 (Cutoff-levels were 0.1 microgram/l [RIA] and 0.2 microgram/l [LIA]). The arrangement of these cutoff-levels leads to a sensitivity of 85% at a specificity of 55% for Protein S100 when measured by RIA, and to a sensitivity of 77% at a specificity of 61% when measured by LIA. The negative predictive values are 91% (RIA) and 88% (LIA), in association with positive predictive values of 39% (RIA) and 40% (LIA). At a cutoff-level of 4.0 micrograms/l, thymidine kinase showed a sensitivity of 70% and a specificity of 41%.
Conclusions: a) The results obtained by LIA are in a very good correlation with the results measured by the well-established RIA. Thus, the LIA is a suitable method for the clinical routine. b) The protein S100 as a tumor-marker of malignant High Risk melanomas has no clinically important sensitivity and specificity as could be proved by a comparison with the predictive value of FDG-PET. c) Whether or not during therapy information concerning progression/remission of the disease can be received out of repetitive measurements of the S100, has to be shown by the results of validated long term follow-up studies.