Multidrug Resistance Gene (MDR1) Polymorphisms May Serve as Predictors of Resistance to Imatinib in Chronic Phase Chronic Myeloid Leukemia Patients.

The Bcr-Abl kinase inhibitor imatinib mesylate (IM) is the first-choice treatment in chronic myeloid leukemia (CML), but resistance has been observed in a proportion of patients (pts). While acquired resistance is known to be most often associated with Bcr-Abl gene amplification/overexpression and/or with the emergence of mutations in the Bcr-Abl kinase domain abrogating IM binding, the determinants of primary resistance remain largely unknown. Since blood and tissue concentrations of drugs are influenced by interindividual variations (single nucleotide polymorphisms) in genes encoding drug metabolizing enzymes and drug transporters, we hypothesized that polymorphisms influencing the extent to which IM is actually delivered to target cells may account for the lack of response that is observed in some pts. IM is metabolised mainly by Cytochrome P450 family enzymes CYP3A4 and CYP3A5, and, to a lesser extent, by CYP1A1, CYP2D6 and CYP2C19 whereas IM transport in and outside cells is mediated by hOCT1 and MDR1, respectively. One hundred and twenty-one chronic-phase CML pts who achieved a complete cytogenetic response (CCyR) on IM 400 mg/d (responders) and sixty-three pts who have not achieved a CCyR (non-responders) on IM 400 mg/d were genotyped for the following polymorphisms: Cyp3A4*1B; Cyp3A5*2 and *3; Cyp1A1*m1, *m2, and *m4; Cyp2C19*2 and *3; Cyp2D6 *4; MDR1*6 and *8; hOCT1 Pro283Leu and Arg287Gly. Two hundred and seventy healthy individuals were analyzed in parallel to check for polymorphism frequency. MDR1*6 and *8 polymorphisms were under-represented in non-responders as compared to responders; however differences were not statistically significant. Analysis of MDR1 haplotypes revealed that responders had a marginally significant lower frequency of wild-type MDR1*6 and *8 as compared to non-responders (24.8% as against 33.3%, respectively; P=0.049). We did not find any significant differences in genotype frequencies between responders and non-responders (and between either groups and healthy individuals) as far as the other genes were concerned. Wild-type MDR*6 and *8 have both been shown to result in higher pump expression and activity with respect to the polymorphic counterparts and this may result in suboptimal intracellular concentrations of IM. Accordingly, MDR1 overexpression has been shown to be associated with IM resistance in CML cell lines. Analysis of a larger series of pts is now ongoing to further investigate the potential usefulness of MDR1 polymorphisms as predictors of resistance. In conclusion, our preliminary data suggest that genotyping for some MDR1 polymorphisms is a straightforward genetic test that could find an application in identifying individuals at risk of resistance, for whom IM high-dose (800 mg/d) or alternative inhibitors might represent a more suitable therapeutic approach.