Increased expression of the ABCB1 gene in cancer cells is usually connected with occurrence of multidrug resistance (MDR) and poor prognosis. However, the correlation between ABCB1 expression and MDR phenotype is difficult to prove in clinical samples. Most of the researchers believe that these difficulties are due to the poor reliability and sensitivity of assays for detection of ABCB1 expression in clinical samples. However, the complexity of P-gp mediated resistance cannot be reduced to the methodical difficulties only. Here, we addressed the question how widely used methods for detection of ABCB1 expression levels could predict its functional activity and thus its contribution to drug resistance in defined conditions in vitro. The ABCB1 expression was assessed at the mRNA level by quantitative real-time polymerase chain reaction (qRT-PCR), and at the protein level by flow cytometry using UIC2 antibody. The ABCB1 function was monitored using a calcein AM accumulation assay. We observed that K562 cells have approximately 320 times higher level of ABCB1 mRNA than HL-60 cells without detectable function. In addition, resistant K562/Dox cells exhibited significantly higher ABCB1 mRNA expression than resistant K562/HHT cells. However, the functional tests clearly indicated opposite results. Flow cytometric assessment of P-gp, although suggested as a reliable method, contradicted the functional test in K562/ Dox and K562/HHT cells. We further used a set of MDR cells expressing various levels of P-gp. Similarly here, flow cytometry not always corresponded to the functional analysis. Our results strongly suggest that an approach which exclusively relies on a simple correlation between ABCB1 expression, either at the mRNA level or protein level, and overall resistance may fail to predict actual contribution of P-gp to overall resistance as the data indicating transporter expression reflect its function only roughly even in welldefined in vitro conditions.