Polymorphisms in the Multidrug Resistance Gene MDR1 (ABCB1) Predict for Molecular Resistance in Patients with Newly Diagnosed Chronic Myeloid Leukemia (CML) Receiving High-Dose Imatinib.

Hematologic and cytogenetic responses to first-line imatinib are high in patients with chronic myeloid leukemia (CML), but only a minority of patients proceeds to a complete molecular response (CMR). Persistent residual disease at the molecular level may be designated as molecular resistance and has been suggested to be accounted for by quiescent malignant stem cells. P-glycoprotein (P-gp), which is encoded by the ABCB1 multi drug resistance (MDR1) gene, has been demonstrated to mediate efflux of imatinib. As hematopoietic progenitor cells efficiently express P-gp, it was hypothesized that single nucleotide polymorphisms (SNPs) of the ABCB1 gene that account for differences in functional activity may possibly be involved in the probability of developing a molecular response to imatinib and molecular resistance. Methods: We set out to evaluate whether the 3 most prevalent MDR1 SNPs (C1236T; G2677T/A; C3435T) would be associated with molecular response in a cohort of 46 early chronic phase CML patients who received 800 mg of imatinib. Molecular response was assessed according to the International Scale. Results: Patient characteristics, including age and Sokal risk score, were distributed evenly among groups of patients, classified according to SNP-genotype. The median follow-up was 46 months (range, 32-60 months). Each of the 3 SNPs were in Hardy-Weinberg equilibrium. However, each combination of 2 SNPs was in strong linkage disequilibrium (P<.001). The overall cumulative incidences of a major molecular response (MMR) and CMR were 78% and 41%, respectively, at 2 years. Molecular responses proved to depend strongly on SNP-genotype. A cumulative incidence of MMR of 52% and 50% was observed in patients with genotype 1236CT or TT as compared to 92% in patients homozygous for 1236C (hazard ratio's (HR): 0.32 and 0.33, P=.02) in univariate analysis, which remained significant (P=.03) following multivariate analysis. Patients homozygous for allele 3435T showed a CMR rate of 10% versus a CMR of 50% at 1 year in patients homozygous for allele 3435C (HR: 0.24 (0.07-0.83, P=.04). Hazard ratio's with respect to CMR associated with C1236T and G2677T were: 0.27 (0.08-0.97, P=.01) for the 1236TT genotype versus the 1236CC genotype, and 0.23 (0.06-0.88, P=.05) for the 2677TT genotype versus the 2677GG genotype, indicating a 4-fold reduction of the probability to develop a CMR. Given the strong linkage disequilibrium, the response differences could not unequivocally be attributed to a single SNP. However, given the enhanced P-gp activity earlier associated with the 2677TT genotype, it is suggested that enhanced clearance of imatinib by the 2677TT genotype may account for the lower incidences of a MMR and CMR in those patients. Conclusions: Molecular resistance in CML patients receiving high-dose imatinib appeared strongly associated with ABCB1 genes SNPs, suggesting a role for P-gp mediated drug efflux in malignant hematopoietic progenitor cells, that may possibly account for persistent molecular residual disease in patients favourably responding to imatinib.

No relevant conflicts of interest to declare.