Pgp and MRP Activities Using Calcein-AM Are Prognostic Factors in Adult Acute Myeloid Leukemia Patients

Thirteen cell lines with different levels of Pgp and MRP expression were used to assess the ability of calcein acetoxymethyl ester (calcein-AM) uptake and calcein efflux to measure Pgp and MRP functions, respectively. There was a good correlation between MRP expression and the modulatory effect of probenecid (a specific modulator of MRP) on the calcein efflux (r = .91, P = .0003) and between Pgp expression and the modulatory effect of CsA on calcein-AM uptake (r = .96, P < .0001). In light of the high correlations for both proteins, we tested calcein-AM uptake and efflux in fresh myeloid leukemic cells. In 53 acute myeloid leukemia (AML) patients, there was also a good correlation between MRP expression (measured by reverse transcription-polymerase chain reaction and by MRPm6 expression by flow cytometry) and the modulatory effect of probenecid on the calcein fluorescence (r = .92, P < .0001) and between Pgp expression as measured by UIC2 antibody binding on flow cytometry and the modulatory effect of cyclosporin A on calcein-AM uptake (r = .83, P < .0001). Pgp activity was higher in CD34+ leukemia than in CD34- leukemia (2.26 +/- 1.50 v 1.46 +/- 1.21, respectively; P = .003), and MRP activity was higher in CD34- leukemia than in CD34+ leukemia (1.77 +/- 0.40 v 1.4 +/- 0. 29, respectively; P = .004). Pgp expression and activity (P = .004 and P = .01, respectively) and MRP activity (P = .03) but not MRP expression were prognostic factors for achievement of complete remission. These results suggest that functional testing (with calcein-AM +/- modulators) for the presence of both MRP and Pgp activities is of prognostic value and that MRP contributes to drug resistance in AML.