Itraconazol Inhibits the P-Glycoprotein Pump and Increases Vincristine Induced Toxicity in Patients with Acute Lymphoblastic Leukemia Carrying a Genetic Variant of the MDR1 Gene

Abstract 231

One of the most well known drug interactions in pediatric oncology concerns the interaction of vincristine, one of the cornerstones in treatment of acute lymphoblastic leukemia( ALL), and the antifungal agent itraconazole. Several studies have shown that the co-administration of itraconazole with vincristine can induce increased neurotoxicity. It is believed that this increased neurotoxicity is mainly caused by the inhibitory effect of itraconazole on the CYP3A subfamily enzymes system (e.g. CYP3A4/5) and that, depending on the genetic variation, some patients are more prone to exhibit this neurotoxicity. However, itraconazole can also inhibit the efflux pump P-glycoprotein (Pg) but so far, data on the role of the Pg pump on vincristine toxicity are lacking. We therefore performed a pharmacogenetic analysis in 103 Caucasian pediatric patients with ALL analyzing both the MDR1 gene and the CYP3A5 gene in the germ-line. CYP3A5 was not associated with vincristine-related toxicity in our cohort. However, one SNP in MDR1 (3435C>T) was significantly associated with developing peripheral neuropathy (Odds Ratio [OR] = 3.8, 95% confidence interval [CI] 1.2 – 11.9, p=0.026). It is known that this genetic variant of the MDR gene (3435T/2677T), leads to a reduced expression of the Pg-pump. Peripheral neurotoxicity in patients carrying this SNP was present even without the co-administration of itraconazole. A subgroup of these patients did receive itraconazole, and the patients with the genetic variant 3435T/2677T of MDR1 showed not only increased peripheral neurotoxicity, but also developed central nervous toxicity (OR=6.4, 95% CI 1.1 – 37.4 p=0.038) upon combined VCR-itraconazole therapy.

To gain more insight in the mechanisms of increased vincristine induced toxicity during azole treatment, in vitro experiments using the LLC-PK1-MDR cell line that over-expresses the MDR1 gene and has no expression of the CYP3A4, were performed. The LLC-PK1-MDR cells were resistant to vincristine exposure in vitro, but became sensitive for VCR after inhibition of the Pg-pump bij PSC833. In contrast, LLC-PK1-MDR1 cells exposed to vincristine in the presence of itraconazole (0.5 and 5 microgram/ml) showed significantly decreased survival in a dose dependent way (LD50 at a dose of 10 ⁻⁴M and 10⁻⁷ M vincristine at dose respectively of 0.5 and 5.0 microgram/ml itraconazole), in contrast to a cell survival of 100% with the same dose of vincristine itraconazole (p< 0.001).

These results indicate that itraconazol indeed inhibits the efflux Pg-pump which may explain the CNS toxicity that occurred in our patient group during the period in which itraconazole was co-administerd during vincristine treatment. Based on these data, we conclude that the Pg- pump plays a dominant role in increased vincristine toxicity during azole treatment. Furthermore, those patients carrying the genetic variant 3435T/2677T of the MDR gene are at increased risk for this toxicity due to reduced expression of the efflux pump Pg.