Combined unfavourable polymorphisms ofhOCT1 and ABCG2 are responsible for delayed complete cytogenetic response during imatinib treatment.

In the recent years, imatinib has changed the outcome of patients affected by chronic myeloid leukemia (CML); nevertheless, about one third of patients must stop the treatment for side effects or resistance. In both these contexts, the adherence to treatment and the drug plas- ma levels have been reported to be strictly linked both to the quality of response and to the tolerability. A large work on Imatinib pharmacoki- netics is at present carried out according to the procedures of the TIK- let protocol (ClinicalTrials.gov identifier: NCT 01860456), thus extend- ing the work previously published by our group (Di Paolo et al., Phar- macogenomic J, 2014) on the role of several genetic covariates. As addi- tional side study, we have investigated the possibility that the time to the achievement of the complete cytogenetic response (CCyR) could be influenced by the genetic polymorphisms in the trans-membrane trans- porters, such as SLC22A1, ABCB1, ABCG2, hOCT1, and SLCO1B1. To accomplish with this objective, both logistic regressions and Anova tests were applied in a population composed by 43 adult CML patients on imatinib. The goal of our approach was to investigate the possible role on CCyR of single and combined different genetic polymorphisms. From a first preliminary check we have found that the only significant role is played by the polymorphisms combination of ABCG2 and hOCT1. Through logistic regression we have computed the probabili- ty of obtaining a delayed CCyR (lasting more than 12 months) as a function of ABCG2 and hOCT1. We found that patients with favourable genotype (high-activity wild-type hOCT1 and low-activity polymorphic ABCG2) have a high probability of achieving the CCyR by 6 months of treatment, while patients with unfavourable genotype (both high-activity wild-type hOCT1 and ABCG2, or both low-activi- ty polymorphic hOCT1 and ABCG2) have a significantly higher prob- ability of achieving a late CCyR (>12 months). The logistic regression performed can be considered significant since patients with an unfavourable genotype have an Odd Ratio (O.R.) value equal to 1.15, p=0.002. Subsequently Anova tests have confirmed these results. In conclusion, our study demonstrated that a delayed cytogenetic response is influenced by the combination of the genetic polymorphisms that concomitantly cause a reduced intake (hOCT1) and an increased efflux (ABCG2) of imatinib, whereas there is not any observed effect due to each single polymorphism. This could be relevant, considering that the 2013 ELN guidelines identify as optimal responder patients who achieve the CCyR by 6 months from the beginning of therapy.