Background Recent research has shown that response to glucocorticoids (GC) predicts the effectiveness of DMARD induction therapy including methotrexate (MTX).[1] Moreover, GC use has been associated with higher intracellular concentrations of methotrexate-polyglutamates (MTX-PG),[2] and higher MTX-PG concentrations were associated with lower disease activity in rheumatoid arthritis (RA).[3] If GC use is associated with higher MTX-PG concentrations, the association between MTX-PG concentrations and non-response may be modified by GC use. Objectives The objectives were to investigate whether MTX concentrations are associated with response and whether this association is modified by CG use. Methods This study included combined data of RA patients treated with MTX from 2 longitudinal cohorts: 285 from the tREACH study and 102 from the MTX-R study. We measured MTX with a tail of 1,2,3,4 and 5 glutamates in erythrocytes after 3 months of therapy using an LC-MS/MS assay. GC use was defined as use of oral or intramuscular GC therapy at start of treatment. As outcome measure for MTX response we defined disease activity score (DAS) 28 at 3 months. GC use, MTX-PG concentrations and their interaction terms (CGuse*MTX-PG concentrations) were tested for associations with DAS28 with an ANCOVA analysis. The analysis was corrected for gender, age, MTX-dose, baseline DAS28, and other DMARD use. Results Mean DAS28 decreased from 4.76 (SD=1.26) to 3.07 (SD=1.20) after 3 months. 49% of patients used glucocorticosteroids at baseline and the median total MTX-PG was 54 nmol/l (IR=42-70). MTX-PG1 (p=0.012), MTX-PG2 (p=0.001), MTX-PG3 (p=0.006) and total MTX-PG were negatively associated with DAS28. GC use was not associated with DAS28 (p=0.450). Patients with GC use had higher concentrations of MTX-PG3 (p=0.005), MTX-PG4 (p<0.001) and MTX-PG5 (P<0.001) than patients without GC treatment. The interactions between GC use and total MTX-PG was significantly associated with DAS28 (p=0.049). When data were stratified by CG use lower MTX-PG1 (p=0.004), MTX-PG2 (p=0.005) and total MTX-PG (p=0.008) were associated with non-response only in patients who did not use corticosteroids at baseline. In persons who used CGs no association was shown. Conclusions This study shows that MTX-PG concentrations were related to a lower DAS28. Furthermore we showed that GC use increases the accumulation speed of MTX-PG concentrations and thereby possibly potentiating the efficacy of MTX, resulting in a lower DAS28. The association between MTX-PGs and DAS28 depends on CG use. References de Jong PH, et al. Ann Rheum Dis. 2012 Oct 31; Stamp LK, et al. Arthritis Rheum. 2009 Aug;60(8):2248-56; Dervieux T, et al. Ann Rheum Dis. 2005 Aug;64(8):1180-5. Acknowledgements tREACH: Unrestricted grant from Pfizer bv. (0881-102217). RDJ: Dutch Arthritis Association (06-02-402 and 09-1-402). Disclosure of Interest None Declared

Background Low-dose methotrexate (MTX) is gold standard in the treatment of rheumatoid arthritis (RA). Response to MTX is very variable and is related to the intracellular MTX-polyglutamate (MTX-PG) levels.1,2 Similar to response, the intracellular MTX-PG concentrations are very variable.3,4 There is little information on the effect of different MTX dosing schemes on the concentration, distribution and accumulationspeed of intracellular MTX-PGs. Objectives The aim of this study was to measure the speed of erythrocyte MTX-PG accumulation as well as the concentration and distribution of MTX-PG species in RA patients on two different MTX dosing regimens. Methods Erythrocyte MTX-PG concentrations were prospectively measured at 3, 6 and 9 months of treatment. Adult RA patients from two longitudinal cohorts were included. Patients in the MTX-R received 15 mg/week MTX, whereas patients from the tREACH received 25 mg/week MTX. Patients were included if they were treated with MTX during the whole observation period and had at least one MTX-PG measurement. A recently developed LC-MS/MS assay was used for the determination of the separate MTX-PG levels.5 Results The largest part of the accumulation of the MTX-PGs occurred during the first 3 months of treatment with only marginal increase in the months thereafter. Steady state levels for all MTX-PGs were reached after 6 months of treatment in both cohorts. Patients from the MTX-R needed a longer time to accumulate similar levels of MTX-PGs than patients from the tREACH and after 9 months of treatment the MTX-R still had lower levels of the long-chain MTX-PGs. Strikingly, during the early phase of treatment, patients from the MTX-R had higher levels of the short-chain MTX-PGs (p=0.02), but lower levels of long-chain MTX-PGs (p=0.006). In the MTX-R the distribution of MTX-PGs was biased towards short-chain MX-PGs whereas in the tREACH the distribution was biased towards long-chain MTX-PGs even though the total MTX-PG levels were not different. Conclusions Higher treatment dose of MTX led to a faster accumulation and higher concentrations of long-chain MTX-PGs resulting in a selective distribution towards long-chain MTX-PGs. As the long-chain MTX-PGs are considered to be the more active MTX-species this study suggests that more intensive treatment may be more efficient. This is supported by a shift towards higher treatment doses and extended co-medication over time in the MTX-R, whereas in the tREACH medication stayed more or less stable. In contrast to previous findings, the main bulk of MTX-PG accumulation takes place during the first three months of treatment and steady states are reached at six months of treatment. References de Rotte MC, den Boer E, de Jong PH, et al., Ann Rheum Dis. Dec 5 2013. Bulatovic Calasan M, den Boer E, de Rotte MC, et al., Ann Rheum Dis. Nov 28 2013. den Boer E, de Rotte MCFJ, Pluijm SMF, Heil SG, Hazes JMW, de Jonge R. Determinants of erythrocyte methotrexate polyglutamate levels in rheumatoid arthritis. Submitted. Stamp ...

BackgroundMethotrexate (MTX) is a first-line therapy in early Rheumatoid Arthritis (eRA). Still, up to 40% of treated patients do not adequately respond to MTX. MTX interferes with the folate cycle, where it indirectly inhibits the global DNA methylation donor S-adenosylmethionine (SAM). Thus, we hypothesised that global DNA methylation changes during MTX use are associated with treatment response.ObjectivesTo examine whether there is a change in global DNA methylation (Δ%Meth) upon MTX use and if this change is associated to MTX response (ΔDAS28) in eRA patients.MethodsDNA was isolated from whole blood (n=120) and Peripheral Blood Mononuclear Cells (PBMCs, n=83) of eRA patients, before and 3 months after MTX use. Samples were collected from the Treatment in the Rotterdam Early Arthritis Cohort (tREACH), a multicenter, stratified single-blind clinical trial of eRA patients. Selected patients received triple (MTX +SSZ + HCQ) or monotherapy (MTX) combined with corticosteroids. 7 CpG sites within Long-Interspersed Nuclear Elements (LINE-1), a proxy for global DNA methylation, were quantified by Sequenom Epityper. Paired t-tests or Wilcoxon Signed Rank tests were conducted to assess a change in methylation. ΔDAS28 score over 3 months was used as a measure for response. Associations between ΔDAS28 and Δ%Meth were corrected for baseline DAS28 in a linear regression model.ResultsIn leukocytes,%Meth did not significantly change over time. However, in PBMCs%Meth in CpG1 (ΔMeth=2.61%, p=0.008), CpG2 (ΔMeth=0.73%, p=0.039) and CpG11.12 (ΔMeth=0.56%, p=0.016) significantly increased over 3 months of MTX use after Bonferroni correction. Δ%Meth in CpG8.9 was significantly associated to the ΔDAS28 (B=0.29, p=0.039), yet this was no longer significant after Bonferroni correction. Δ%Meth was not significantly associated to ΔDAS28 in any of the other LINE1 CpG sites tested.ConclusionsPBMC global DNA methylation in LINE-1 CpG sites increased upon 3 months of MTX use. However, this change in methylation is not associated to MTX response. Further research is needed to investigate the role of global DNA methylation in these patients.Disclosure of InterestNone declared