Psoriasis is a chronic, immune-mediated, multisystemic, inflammatory disease caused by the interaction of multiple susceptibility genes and environmental factors that affects 1–3% of the population worldwide (1). Psoriatic arthritis (PsA) most commonly develops in 20–30% of psoriatic patients with preceding psoriasis or develops concomitantly with psoriasis, and it manifests all features of arthritis, including enthesitis (2). Our previous study has demonstrated that methotrexate (MTX), an efficacious treatment for both psoriasis and PsA but associated with hepatotoxicity, is more effective and safer in patients with psoriasis without arthritis (PsO) than in PsA patients (3). Therefore, it is critical to identify biomarkers for predicting the development of PsA and the efficacy and hepatotoxicity of MTX.

Methylenetetrahydrofolate reductase (MTHFR) is a crucial enzyme in homocysteine/methionine metabolism that catalyzes the formation of 5-methyltetrahydrofolate, which is the methyl donor for the synthesis of methionine from homocysteine (Hcy) (4). The 677 C>T mutation (rs1801133) in MTHFR disrupts its thermostability, leading to defective enzyme activity and elevation of Hcy levels (5). Individuals who are homozygous for the 677T allele have only 32% of the MTHFR enzymatic activity measured in individuals homozygous for the wild-type C allele, whereas heterozygotes retain 64% of the wild-type MTHFR enzymatic activity. In contrast, homozygosity in the 1298 A>C mutation (rs1801131) does not lead to higher Hcy levels (6).

The association between MTHFR polymorphism and the susceptibility to psoriasis has been supported in Caucasians, Turkish, and Chinese populations (7–12). A study carried out in China reported that the rs1801133 T allele and the rs1801131 C allele were all associated with increased psoriasis risk (9). However, no studies have been found that surveyed the frequency of MTHFR polymorphisms in patients with PsA in the Han Chinese populations. In addition, only two studies have attempted to investigate the correlation between MTHFR polymorphisms and MTX efficacy in psoriasis, both suggesting that no significant genotypic associations were found (13, 14). As for the hepatotoxicity of MTX, according to Campalani et al. (14), patients with the rs1801131 C allele were less likely to develop hepatotoxicity, and no association was detected between the rs1801133 and hepatotoxicity of MTX.

In the present study, we sought to evaluate whether MTHFR rs1801131 and rs1801133 SNPs are associated with concomitant PsA, the efficacy, and hepatotoxicity of MTX in patients with psoriasis in the Han Chinese population.

Previous studies have revealed dozens of genes with genomic variations associated with PsA, such as HLA-C, HLA-B, IL12B, IL23R, and REL (15). To our knowledge, this is the first study to assess the potential association between MTHFR polymorphism and the susceptibility to PsA in the Han Chinese population. In the present study, the frequency of the rs1801133C allele in PsA was 64.1%, and the frequency of rs1801133 CC genotype in PsA was 42.3%, which was significantly higher than that in PsO (28.8%) and healthy controls (33.1%). In contrast to the studies on patients with psoriasis (11, 12), we found no difference in the rs1801133 genotype distribution between PsO and healthy controls; this inconsistency may be due to those studies combined PsA and PsO as a whole group when analyzing. It is reported that the association between the SNP rs1801133 and bone phenotypes depends on folate status (16). Since PsA is characterized by both extensive resorption and exuberant new bone formation (17), one possible mechanism by which rs1801133 affect the development of PsA is that SNP rs1801133 modulates serum homocysteine concentrations and leads to subsequent effects on bone. In addition, patients with rheumatoid arthritis (RA) were reported to have a higher frequency of the rs1801133 T allele (18, 19), which is contrary to our observation that the C allele was more frequent among patients with PsA. This finding further provides evidence for different genetics of RA and PsA.

Previous studies reported no associations between SNPs in MTHFR and MTX efficacy in psoriasis (13, 14). In the present study, we found psoriatic patients with the rs1801133 TT genotype had higher PASI 90 response rates to MTX, and patients with the rs1801131 CT had lower PASI 75 response rates to MTX. Furthermore, stepwise multivariate logistic regression analysis confirmed that the odds ratio for patients with the rs1801133 TT genotype to achieve PASI 90 was 2.759, and the odds ratio for those with the rs1801131 CT genotype to achieve PASI 75 was 0.492. One possible explanation might be that the reduction of the MTHFR enzyme activity and subsequently higher Hcy level caused by the 677C>T mutation (rs1801133), not the 1298 A>C mutation (rs1801131), increases MTX sensitivity via the inhibition of S-adenosylmethionine and de novo purine synthesis (20, 21). The decreased DNA methylation exhibited by the rs1801133 TT genotype also increases the anti-proliferative and anti-inflammatory effects of MTX (22). Nevertheless, it should be noted that the efficacy of MTX was reported to be associated with the dose and the length of therapy (23), and therefore all interpretations regarding high-dose MTX must be made with great caution. Also, the effect sizes for PASI 75 and PASI 90 were relatively small, suggesting that the associations between SNPs in MTHFR and MTX efficacy may be weak.

Apart from SNPs in MTHFR, stepwise multivariate logistic regression analysis also showed other independent factors associated with the efficacy and hepatotoxicity of MTX in 309 psoriatic patients. Lower PASI 75 and PASI 90 response rates to MTX were both significantly associated with higher BMI and concomitant PsA; this was in accordance with our previous report showing that MTX was more effective in patients with PsA than in those with PsO (3). In addition, smoking and high BMI are crucial factors increasing MTX-related hepatotoxicity in the present study, which is also in line with our earlier observations (3).

In the present study, in PsA, patients with the rs1801133 CT and TT genotype had a higher risk of abnormal liver function than those with the CC genotype. A similar finding also reported that rs1801133 TT genotype might be associated with MTX-induced hepatotoxicity in PsA (24). However, no associations between rs1801133 genotype and abnormal liver function in PsO were found in the present study. This result may be explained by the different immune responses among patients with PsA and PsO. For example, both TNF and IL-17 inhibitors can significantly relieve joint pain, but IL-17 inhibitors lead to better improvement of psoriatic skin lesions, suggesting the different immune-pathogenesis between PsA and PsO. Hepatic ischemia-reperfusion stimulates Kupffer cells and initiates injury through TNF-α upregulation (25), and anti-TNF-α agents have been reported to increase liver enzyme levels (26). This evidence suggests TNF-α in PsA might be involved in liver injury.

Our study has several limitations. The major limitation is the relatively small sample size and the single-center setting. Because of the nature of our clinic, which is specialized in PsA, the proportion of patients with PsA in the cohort was higher than the reported incidence of PsA (20–30% of patients with psoriasis), and the disease tended to be more severe. Additionally, evidence of juxta-articular new-bone formation was excluded from the assessment of CASPAR and the follow-up was relatively short [psoriasis precedes arthritis by an average of 7 years (27)], leading to a potentially missed diagnosis of PsA. Also, given the small number of TT genotype carriers of rs1801131, results regarding rs1801131 should be interpreted with caution. It is clear that further validation of these preliminary findings is necessary. As a first step, the reproducibility of the results should be substantiated in larger multi-center studies. Then, it would be appropriate to investigate whether this genetic marker could been combined with other predictors. Furthermore, digital medicine such as machine learning algorithms might be applied to these data to provide a molecular fingerprint composed of the differential regulation of multiple markers to predict the risk of PsA and the efficacy and hepatotoxicity of MTX in psoriasis.

In conclusion, although factors including study design, missed diagnosis, and the small number of specific genotype carriers may limit the generalizability of the study, results from this study add useful information to the evidence regarding the efficacy and hepatotoxicity of the low-dose MTX in patients with psoriasis with and without arthritis. Also, this is the first study to assess the potential association between MTHFR polymorphism and the susceptibility to PsA in the Han Chinese population. These observations contribute to our growing knowledge base of pharmacogenetics of psoriasis and PsA and personalized therapies.

The data presented in the study are deposited in the European Variation Archive database, accession number: PRJEB51117.

The studies involving human participants were reviewed and approved by the Medical Ethics Committee of Huashan Hospital, Fudan University. The patients/participants provided their written informed consent to participate in this study.

JZ, ZW, XF, ZZhe, ZZha, and KYan designed the study. JZ and LT collected blood samples. JZ, NY, and KYan wrote the manuscript. JZ, ZW, ZZha, LT, LH, QH, XF, KYang, and GH contributed to the acquisition, analysis, or interpretation of data. ZW, ZZha, ZZhe, and KYan obtained the funding. JZ and ZZha contributed to the revision of the manuscript. KYan and ZZha are available post-publication to respond to any queries or critiques. All authors read and approved the final manuscript.

This work was supported by the National Natural Science Foundation of China (Grant Nos. 81773322, 81673080, 81673073, 81974471, and 82173420), the Key Project in Basic Research Advocated by Shanghai Science and Technology Commission (Grant No. 13JC1402300), and the Clinical Research Plan of SHDC (Grant Nos. SHDC2020CR6022 and SHDC2020CR1014B). Funding sources had no involvement in the study design, the collection, analysis, or interpretation of the data, the writing of the report, and the decision to submit for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.