Abstract
Objectives: Identifying blood-based biomarkers that predict treatment response in rheumatoid arthritis (RA) is a clinical priority. We investigated differential DNA methylation as a candidate biomarker of response for the first-line drug used in RA, methotrexate (MTX).
Methods: DNA methylation was measured in DNA samples from individuals recruited to the Rheumatoid Arthritis Medication Study (RAMS). Differentially methylated positions (DMPs) were compared between whole blood samples collected at baseline and at four weeks from patients who, by six months, had a good (n=34) or poor response (n=34) to MTX using linear modelling, adjusting for gender, age, cell composition, baseline 28-joint disease activity score (DAS28) and smoking status. Analyses also compared methylation with changes in DAS28 and changes in swollen joint (SJC) and tender joint counts (TJC), and changes in C-reactive protein (CRP) over the initial six months after MTX commencement. DMPs showing significant differences with any response parameter were tested using pyrosequencing in an independent group of 100 patients from RAMS.
Results: In the discovery group, two CpG sites showed methylation changes at four weeks associated with clinical EULAR response by six months. Significant changes in methylation for three DMPs associated with change in TJC, three with change in SJC and a further four with change in CRP. Of the 12 CpGs, four showed replicated association in an independent dataset of samples from RAMS.
Conclusion: These data represent an advance on current practice by contributing to a personalised medicine strategy allowing an escalation or change in therapy as early as 4 weeks.
Methods: DNA methylation was measured in DNA samples from individuals recruited to the Rheumatoid Arthritis Medication Study (RAMS). Differentially methylated positions (DMPs) were compared between whole blood samples collected at baseline and at four weeks from patients who, by six months, had a good (n=34) or poor response (n=34) to MTX using linear modelling, adjusting for gender, age, cell composition, baseline 28-joint disease activity score (DAS28) and smoking status. Analyses also compared methylation with changes in DAS28 and changes in swollen joint (SJC) and tender joint counts (TJC), and changes in C-reactive protein (CRP) over the initial six months after MTX commencement. DMPs showing significant differences with any response parameter were tested using pyrosequencing in an independent group of 100 patients from RAMS.
Results: In the discovery group, two CpG sites showed methylation changes at four weeks associated with clinical EULAR response by six months. Significant changes in methylation for three DMPs associated with change in TJC, three with change in SJC and a further four with change in CRP. Of the 12 CpGs, four showed replicated association in an independent dataset of samples from RAMS.
Conclusion: These data represent an advance on current practice by contributing to a personalised medicine strategy allowing an escalation or change in therapy as early as 4 weeks.
Original language | English |
---|---|
Journal | Rheumatology (Oxford) |
Early online date | 10 Oct 2019 |
DOIs | |
Publication status | E-pub ahead of print - 10 Oct 2019 |
Keywords
- Rheumatoid arthritis
- treatment response
- methotrexate
- pharmacogenetics
- DNA methylation
- epigenetics
- DMARD