Abstract
AFP, CEA, CA125 and CA19‐9 are commonly used serum tumour markers (TMs) in clinical practice, although their quantification by immunoassay may be influenced by pre‐analytical sample handling. Though the effect of repetitive freeze–thaw cycles is generally recognized, it is not clear in detail. The present study measured (CLIA) these TMs in serum samples freshly separated and after each of five freeze–thaw cycles, in which the samples were frozen at −40°C for 10 months at cycle 4 and 2 h at other cycles. Statistical analysis with the General Linear Model for Repeated Measures revealed significant decreases in the measurements of the four TMs, with the least decrease of 6.8 % for CA125 and the most decrease of 18.2 % for CA19‐9 after the last cycle, and an overwhelming single cycle decrease of mean 7.7 % at cycle 4 for AFP, CEA and CA125, of 7.5 % and 9.3 % at cycles 4 and 5 for CA19‐9. So it seems that measurements of AFP, CEA and CA125 are more readily affected by long‐term frozen storage compared with frequent freezing–thawing, while CA19‐9 is relatively unstable under both conditions.