Methotrexate is used as first-line treatment of moderate to severe psoriasis. Despite the marked variability in treatment outcomes, no pharmacogenetic markers are currently used for personalised management of therapy. In this retrospective study, we investigated the effects of genetic predisposition on efficacy and toxicity of low-dose methotrexate in a cohort of 137 patients with moderate to severe plaque psoriasis. We genotyped 16 polymorphisms in genes for enzymes involved in the folate–methionine pathway and in methotrexate transport, and analysed their association with treatment efficacy and toxicity using classification and regression tree analysis and logistic regression. The most pronounced effect observed in this study was for GNMT rs10948059, which was identified as a risk factor for inadequate efficacy leading to treatment discontinuation. Patients carrying at least one variant allele had ~7-fold increased risk of treatment failure compared to patients with the wild-type genotype, as shown by the classification and regression tree analysis and logistic regression (odds ratio [OR], 6.94; p = 0.0004). Another risk factor associated with insufficient treatment responses was DNMT3b rs2424913, where patients carrying at least one variant allele had a 4-fold increased risk of treatment failure compared to patients with the wild-type genotype (OR, 4.10; p = 0.005). Using classification and regression tree analysis, we show that DNMT3b rs2424913 has a more pronounced role in patients with the variant GNMT genotype, and hence we suggest an interaction between these two genes. Further, we show that patients with the BHMT rs3733890 variant allele had increased risk of hepatotoxicity (OR, 3.17; p = 0.022), which is the most prominent reason for methotrexate discontinuation. We also show that variants in the genes for methotrexate transporters OATP1B1 (rs2306283/rs4149056 SLCO1B1 haplotypes) and ABCC2 (rs717620) are associated with increased risk of treatment failure. The associations identified have not been reported previously. These data suggest that polymorphisms in genes for enzymes of the methionine cycle (which affect cell methylation potential) might have significant roles in treatment responses to methotrexate of patients with psoriasis. Further studies are warranted to validate the potential of the pharmacogenetic markers identified.