Methotrexate is an antifolate chemotherapeutic drug commonly used to treat cancer, including acute lymphoblastic leukemia (ALL). Interindividual variation in clearance of methotrexate results in a vast range of exposure to the drug, which affects its clinical effectiveness and risk of toxicity. In a genome-wide association study of children with ALL, we identified the
SLCO1B1 gene, encoding liver transporter OATP1B1, as harboring multiple common polymorphisms associated with methotrexate clearance (Trevino et al. J Clin Oncol, 27:5972-5978, 2009). Because these common polymorphisms could not account for the variation in methotrexate clearance entirely, we hypothesized that there may be rare variants in addition to the common polymorphisms in this gene that affect methotrexate disposition. To test this hypothesis, we sequenced
SLCO1B1 exons in 699 pediatric leukemia patients, and compared the effects of common versus rare variants on methotrexate clearance (adjusting for clinical covariates). We identified 93 SNPs, 15 of which were non-synonymous (NS); 11 of the 15 NS SNPs were rare, with minor allele frequencies < 1%. We used four computational programs (SIFT, PMUT, SNPS3D, and Polyphen2) to predict whether the NS SNPs were damaging to the function of the protein. The 7 NS SNPs predicted to be functionally damaging (common or rare) were more likely to be found among patients with the lowest methotrexate clearance than patients with high clearance (p=2.4×10
−8), including one SNP that was observed only once among nearly 1400 chromosomes. Four
SLCO1B1 haplotypes were associated with reduced methotrexate clearance (p<0.05) and we verified that these haplotypes have lower function with
in vitro studies of cellular methotrexate transport. After adjusting for clinical covariates,
SLCO1B1 variants accounted for a large portion of the population variability in clearance (p<10
−15). Of the variability attributable to
SLCO1B1 variants, common NS variants accounted for the majority, but rare damaging NS variants constituted 17.8% of
SLCO1B1's effects on variability in clearance (p<10
−3), and had larger effect sizes than common NS variants. Our results show that rare variants are likely to have an important effect on antileukemic agent phenotypes and should not be ignored in clinical pharmacogenetic tests.
Figure 1.
Methotrexate clearance and presence of non-synonymous (NS) SNPs among outliers. The average methotrexate clearance (adjusted for age, gender, race, and treatment arm) was used to divide the patients into the lowest, middle, and highest tenth percentile groups (n=70 patients per decile). The number of NS SNPs did not differ among clearance decile groups (p=0.2, data not shown), whereas (A) the number of damaging NS SNPs differed by decile group (p=2.4×10−8) as did (B) the number of rare (MAF < 1%) damaging NS SNPs p=0.049). P values generated by the exact Chi-squared test.
Figure 1.
Methotrexate clearance and presence of non-synonymous (NS) SNPs among outliers. The average methotrexate clearance (adjusted for age, gender, race, and treatment arm) was used to divide the patients into the lowest, middle, and highest tenth percentile groups (n=70 patients per decile). The number of NS SNPs did not differ among clearance decile groups (p=0.2, data not shown), whereas (A) the number of damaging NS SNPs differed by decile group (p=2.4×10−8) as did (B) the number of rare (MAF < 1%) damaging NS SNPs p=0.049). P values generated by the exact Chi-squared test.
