Abstract
A number of studies have demonstrated that high tumor tissue levels of tissue inhibitor of metalloproteinases-1 (TIMP-1) are associated with a poor prognosis in breast cancer, suggesting that TIMP-1 could be a valid prognostic marker in this disease. Recently, our laboratories have presented results showing that TIMP-1 also carries prognostic information when measured in serum. This is an important finding, since serum is a much more preferable material compared with tumor tissue extracts. The aim of the present study was to validate the previous results concerning the prognostic value of TIMP-1 in serum obtained preoperatively from 68 patients with primary breast cancer. This was done by measuring the same serum samples as in the previous study but in a different laboratory using a different ELISA assay. We confirmed that patients with the highest serum levels of TIMP-1 (> 197.7 ng/ml) had significantly shorter disease-specific survival compared with patients with low serum TIMP-1 levels. In the group of node-negative patients, 53% of the patients with high levels of TIMP-1 survived after 10 years of follow-up compared to 92% of the patients with low levels. This study thus confirms the reproducibility across laboratories of the results concerning the prognostic value of TIMP-1 in serum. We also investigated whether measurements of the specific fraction of uncomplexed TIMP-1 improved the prognostic value of TIMP-1 in serum, as has been shown to be the case for tumor tissue extracts. However, including information of the level of uncomplexed TIMP-1 did not seem to provide additional prognostic information to that already provided by total TIMP-1.