Subantimicrobial-dose doxycycline treatment increases serum cholesterol efflux capacity from macrophages
New potential tumor markers such as matrix metalloproteinases and their inhibitors have been extensively studied during the last decades. The aim is to find prognostic markers that are measurable in easily available samples, such as serum or plasma. The proper sample type to use when measuring the levels of gelatinases and their inhibitors from blood samples is currently under critical evaluation. In this study, the effect of sample type is studied in 26 healthy controls, and the result is confirmed in a series of 80 breast carcinoma patients.Sample type had the most evident effect on the levels of TIMP-1 and MMP-9. Serum samples gave about two-fold levels of TIMP-1 compared to plasma samples (p < 0.001), with a strong linear correlation between these two (r = 0.79). Pro-MMP-9 levels were significantly affected by the presence of a blood coagulation activator in the serum sample, or a different anticoagulant in the plasma sample. The serum and plasma values had only a weak correlation (r = 0.37).In conclusion, sample type should be carefully considered, especially when measuring proMMP-9, and plasma should be preferred for this measurement. For TIMP-1 the correlation of serum and plasma values is good; the use of serum samples can therefore be justified as long as the generally higher levels in the serum are acknowledged.