Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory joint disease. RA affects ~1% of the population worldwide and it is at least three times more common in women compared with men (1). In Colombia, the prevalence of RA was recently estimated to be 1.49% (2). Early diagnosis and effective treatment with disease modifying anti-rheumatic drugs can prevent joint destruction, reduce functional limitation, improve the quality of life, and reduce morbidity and mortality associated with the natural progression of the disease (3-6).

Methotrexate (MTX) has been the first-choice drug for RA treatment since the 1980s (7). The therapeutic effectiveness and safety of MTX has been reported in numerous studies in various populations, and its continuous use has been shown to reduce disease activity and prevent bone destruction (8-10). However, only 45% of patients receiving MTX as a monotherapy achieve significant results in the medium term; and 15-30% of patients with RA stop the treatment after 3 years of use due to gastrointestinal, cutaneous, haematological and hepatic toxicity (11,12).

To explain the variability in the clinical response to MTX, several factors have been studied, including patient-dependent factors (such as, sex, age at RA onset and disease duration), RA-related factors (such as immunological mechanisms conditioning inflammatory response duration and severity) and the presence of autoantibodies [such as rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP)]. In addition, genetic variability of the enzymes involved in the metabolic pathway of MTX has been studied, and this variability may be a factor associated with the therapeutic efficacy and/or safety of MTX (13-18).

MTX is a folic acid analogue with immunomodulatory and anti-inflammatory properties. MTX enters the cell through reduced folate carrier 1, which is encoded by the SLC19A1 gene, and exits the cell via members of the ATP-binding cassette transporter family, also known as multidrug resistance-associated proteins (14). Inside the cell, MTX is converted to methotrexate polyglutamate (MTXPG), which inhibits dihydrofolate reductase (DHFR) (19) and indirectly inhibits the methylenetetrahydrofolate reductase (MTHFR), which is involved in novo purine synthesis. MTXPGs intervene in pyrimidine synthesis through inhibition of thymidylate synthetase enzyme, which converts deoxyuridylate into deoxythymidylate. MTXPGs also inhibits the AICAR transformylase enzyme, which is encoded by the ATIC gene; this leads to the intracellular accumulation of adenosine nucleotides and 5-amino-4-imidazole carboxamide, thereby increasing extracellular adenosine, a potent anti-inflammatory agent when dephosphorylated (20). MTXPGs have been shown to inhibit the proliferation of lymphocytes, monocytes, macrophages and the production of proinflammatory cytokines (such as TNFα, IL-2, IL-4, IL-6, IL8 and IL-10) (21,22).

The aim of the present study was to establish the association between SNPs in the metabolic pathway of MTX and its therapeutic efficacy and safety concerning Colombian patients with RA.

In the present study, an association between the therapeutic response to MTX and MTHFR SNPs (C677T: rs1801133 and A1298C: rs1801131, respectively) was identified in the population studied. There were no associations found for DHFR-CT (rs10072026) and FPGS-AG (rs1544105), in agreement with a previous study (19). To the best of our knowledge, the present study is the first to simultaneously investigate the relationship of SNPs of the primary enzymes involved in the metabolic process of MTX with MTX response and toxicity in Colombian patients with RA. Similar results have been described by Urano et al (27) in a group of 106 Chinese patients with RA, where subjects with a MTHFR 677C-1298C haplotype required a lower dose of MTX to control the inflammatory symptoms, which in-turn resulted in fewer AEs. Similarly, Xiao et al (28) identified MTHFR SNPs in 110 Japanese patients and found that MTHFR C677T (rs1801133) was associated with an improved clinical response, whereas SNPs rs1801133 and rs2274976 were associated with frequent development of primarily gastrointestinal AEs, including anorexia, diarrhoea, dyspepsia, gastrointestinal haemorrhage, hepatitis, nausea or vomiting and pancreatitis.

The aforementioned results are relevant for the information already reported in healthy Colombian populations from the cities of Bogota, Medellin, Barranquilla and Cali, in which comparisons of the ancestral homozygous genotype C/C with mutant heterozygotic C/T and MTHFR T/T demonstrated significant differences. These mutant genotypes may result in an increased risk of toxicity and AEs in individuals that receive MTX for AR treatment (29).

However, there are contradictory reports when comparing with the results from cohorts from other countries. In a Mexican population, 70 patients with RA exhibited greater hepatic toxicity associated with the MTHFR A1298C mutant allele (rs180113) (30). The same association has been observed in a Dutch population of 205 patients with RA in whom MTX treatment was initiated, with a follow-up of 6-months. Among these 205 patients, patients with a mentioned polymorphism more frequently presented with pneumonitis, and skin, gastrointestinal or hepatic toxicity (31).

These findings seem to be consistent with outcomes observed in patients with other rheumatic diseases treated with MTX. In a Slovenian population study performed with a cohort of 119 patients with idiopathic juvenile arthritis (with 6 months follow-up), 55% of the patients had side effects and required a change of pharmacological therapy. These patients had an increased frequency of MTHFR SNP rs1801131, which was associated with hepatic and gastrointestinal toxicity, and in MTHFR rs1801133, which was associated with early discontinuation of MTX treatment (32).

The present study analysed the relationship of SNPs from the enzymes in the MTX metabolic pathway that have been associated with a lack of therapeutic response to the drug previously; these were: ATIC, ATIC C347G and RFC-1-G80A. However, no significant association was identified for any SNPs of these enzymes with response to MTX. These results are in contrast to a study by Dervieux et al (33) in which a group of 108 North American patients with RA were followed-up for 3 months after the initiation MTX therapy. The RFC-1-G80A SNP was analysed and found to be associated with more prominent inflammatory activity, such as increased joint count and increased visual analogue scale pain rating. Similar results were reported for the ATIC C347G SNP (rs237253) in the same study.

The present study reported similar results when comparing the relationship of SNPs associated with MTX toxicity with other populations regarding DHFR-CT (rs10072026) and ATIC C347G (rs2371536) SNPs. A study of 211 central European patients with RA determined the primary polymorphisms from the enzymes involved in the metabolic pathway of DTX, identifying an association for ATIC C347G with a higher risk of toxicity (34). Similar results were reported by Salazar et al (35) in 124 Spanish patients with RA treated with MTX monotherapy; it was identified that the ATIC polymorphism was associated with increased AEs, including hepatotoxicity, gastrointestinal disorders or alopecia. A DHFR polymorphism was also evaluated and was found to be associated with improved treatment response without increased toxicity.

Owen et al (36) studied polymorphisms regarding MTX management in 309 UK patients, and the results showed that the DHFR mutation (rs12517451) was associated with toxicity; however, rs10072026 and rs1643657 were associated with a reduced risk of MTX-related AEs.

Cardiovascular (CV) comorbidities have been observed in 45% of the Colombian population. MTHFR polymorphisms have also been implicated in the onset of CV disease in a healthy population; specifically, patients with the MTHFR C677T polymorphism may have an increased risk of cardiovascular disease. Furthermore, individuals who are homozygotic for the MTHFR C677T polymorphism exhibit increased levels of homocysteine, and high concentrations of this amino acid represent an additional risk of CV and cerebrovascular disease (37). Since CV diseases are the most common cause of morbidity and mortality in patients with RA, it has been proposed that MTHFR polymorphisms represent an additional and non-traditional risk of patients experiencing CV events. Another prospective, observational study involving a Spanish population of 612 patients (mean age, 53 years), who were followed up for 14 years, reported an association between CV events and endothelial dysfunction with a MTHFR A1298C genotype, but not with the C677T genotype (38).

The present study has some limitation. First, there was no follow up of patients to evaluate the evolution of DAS28 over time as well as its toxicity. However, the relatively large sample size decreases biases, which strengthen the results of the present study. Due to the design of the study from the beginning, a COX regression analysis after the univariate analysis was not performed.

In conclusion, the Colombian population shares similarities regarding SNPs associated with MTX efficacy and toxicity; however, the polymorphisms associated with inefficacy according to the literature were not observed in the present study. The variability of the results suggests that the discrepancies between populations may underlie the different outcomes, suggesting these SNPs should be assessed in additional countries. However, the SNPs identified may be used to establish biomarkers for predicting the MTX response in terms of efficacy and toxicity in Colombian patients with RA.