Background Detection of the intracellular levels of polyglutamated MTX (MTX-PGn) can predict response of MTX or possibly its adverse effects. As efficacy and toxicity of MTX differs among individual patients, we had proposed a predictive model for MTX efficacy consisting of 9 SNPs (2015 EULAR). In the present study, we measured erythrocyte MTX-PGn in RA patients with low disease activity for long time receiving a stable dose of MTX, investigated their associations with genetic polymorphisms, and speculated MTX dose required to reach possible effective MTX-PG levels.

Objectives To investigate if erythrocytes MTX-PGn concentration is associated with the 9 SNPs in 7 genes reportedly related to MTX-efficacy in RA patients with low disease activity.

Methods The study was cross-sectional using 121 adult patients with RA in Shinko Hospital. All patients enrolled had received a stable dose of MTX (mean 8.9±0.3 mg/week), and kept lower disease activity (DAS28-CRP <2.7, mean 1.9±0.4) for at least 3 months. Concentrations of MTX-PG with 1 - 5 glutamate residues (PG1- PG5) in erythrocytes were quantitated by LC-ESI-MS/MS as described by den Boer et al. Nine SNPs in 7 genes previously found to associate with efficacy of MTX by us were genotyped by RT-PCR (Applied Biosystems, Inc.). First, association of the 9 SNPs with concentrations on MTX-PGn was analyzed by one-way layout (Wilcoxon signed-rank test). Secondly, multivariate logistic regression analyses were performed to estimate roles of MTX dose and polymorphisms in concentrations of MTX-PGn.

Results Total PGn concentrations were 82.1±31.7 nmol/L (m±SD) and positively correlated with MTX dose (Rs =0.4104, p<0.001), but range of concentration was rather wide even if the dose was same. By one-way layout, GGH c.452C>T was associated with PG1, PG2, and total PG, SLC19A1 c.80G>A with total PG, and SLC28A3_c.267G>A with PG3. Three SNPs (FPGS c. 192T>C, EPHX1_c.357G>A, SLC28A3_c.267G>A) were related to MTX doses. Next, we investigated roles of MTX dose and 9 SNPs in maintenance of MTX-PGn concentrations as by multiple regression analyses. As PG1 levels were related to GGH c.452C>T but not to MTX dose, PG1 levels were 38.2±19.6 and 54.8±21.5 nmol/L in CC and CT genotypes, respectively (P=0.0291). MTX dose was strongly associated with PG3 levels as well as with those of PG4. MTX-PG3 levels were regressed as follows; MTX-PG3 (nmol/L) = -0.035 +[2.04 * MTX dose] + [2.34 * SLC19A1 c.80G>A] + [-2.01 * FPGS c.^*192T>C]. When upper and lower limits of them were defined, we can speculate MTX dose to reach the effective levels for individual patients. However, the ranges of effective levels were rather wide, suggesting that effectiveness of MTX may depend on not only concentrations of MTX-PG but also responsiveness to MTX-PG.

Conclusions We measured erythrocyte MTX-PGn in RA patients with low disease activity and investigated for their associations with MTX dose and genetic polymorphisms. We could speculate MTX dose required to reach possible effective MTX-PG levels in individual patient.

References

1. Den Boer E, et al: Measuring methotrexate polyglutamates in red blood cells: A new LC-MS/MS-based method. Anal Bioanal Chem, 405:1673–1681, 2013.

References

Disclosure of Interest S. Kumagai Consultant for: Sysmex CO, M. Nishida: None declared, Y. Uemura: None declared, M. Izumi: None declared, K. Abe: None declared, K. Yoneda: None declared, Y. Noda: None declared, S. Sendo: None declared, A. Ohishi: None declared, M. Shinohara: None declared, G. Tsuji: None declared