Long-term durability of atime-limited methotrexate intervention in patients with anti-citrullinated protein antibody-positive and anti-citrullinated protein antibody-negative arthralgia at increased risk for rheumatoid arthritis (TREAT EARLIER) 5-year data from a double-blind, randomised, placebo-controlled trial

Mulligen, E. van; Dumoulin, Q.A.; Steenbergen, H.W. van; Krijbolder, D.I.; Visser, A.W.; Böhringer, S.; Mil, A.H.M.V. (2026)

Article / Letter to editor

Background

The TREAT EARLIER trial showed that a time-limited intervention with 12 months of methotrexate in individuals with arthralgia at risk of rheumatoid arthritis improved inflammatory burden but did not prevent rheumatoid arthritis after 2 years. Long-term follow-up is needed to assess the durability of the effects. In addition, heterogeneity in pathophysiological subtype (with or without anti-citrullinated protein antibodies [ACPA]) and disease risk was not considered in the 2-year analysis. Therefore, we investigated the long-term effect of methotrexate in reducing disease burden and progression to rheumatoid arthritis in ACPA-positive and ACPA-negative individuals at increased risk of rheumatoid arthritis.

Methods

The TREAT EARLIER trial was a randomised, placebo-controlled trial of participants aged 18 years or older, with clinically suspect arthralgia and subclinical joint inflammation. Participants were recruited from 13 rheumatology outpatient clinics in the Netherlands and randomly assigned (1:1) to a single intramuscular glucocorticoid injection (methylprednisolone 120 mg) followed by a 1-year course of methotrexate (up to 25 mg/week), or placebo (single injection and tablets for 1 year). Participants and investigators were masked to group assignment for at least 2 years. Follow-up continued for 5 years. The two primary endpoints in this 5-year follow-up analysis were disease burden (physical disability) and development of rheumatoid arthritis, assessed on an intention-to-treat basis. Patients were stratified for ACPA; only those at increased predicted risk (>10%) were included in the 5-year analysis. People with lived experience of clinically suspect arthralgia or rheumatoid arthritis were involved in the study design. This trial is registered with EudraCT (2014-004472-35) and the Netherlands Trial Register (NTR4853-trial-NL4599), and is complete.

Findings

Between April 16, 2015, and Sept 11, 2019, we enrolled 236 participants; 119 were assigned to active treatment and 117 to placebo. 215 (91%) participants completed the 5-year follow-up. 120 participants at increased predicted risk of rheumatoid arthritis were analysed at 5 years, of whom 66 (55%) were ACPA-negative and 54 (45%) were ACPA-positive. Mean age was 48 (SD 12) years, 72 (60%) of 120 participants were female, and 48 (40%) were male. Over 5 years, ACPA-negative participants in the active treatment group (n=35) had sustained improvement in physical disability compared with ACPA-negative participants in the placebo group (n=31; mean difference in Health Assessment Questionnaire disability index [HAQ] -0 & centerdot;16 [95% CI -0 & centerdot;29 to -0 & centerdot;04], p=0 & centerdot;0082) whereas, in ACPA-positive participants, the previously reported benefit in physical disability at 2 years in the treatment group versus placebo group was not sustained (-0 & centerdot;12 [-0 & centerdot;26 to 0 & centerdot;03], p=0 & centerdot;12). Three (9%) of 35 ACPA-negative participants in the treatment group developed rheumatoid arthritis compared with ten (32%) of 31 in the placebo group over 5 years (hazard ratio [HR] 0 & centerdot;24 [95% CI 0 & centerdot;07 to 0 & centerdot;87], p=0 & centerdot;018), corresponding to a number needed to treat of four. In ACPA-positive participants, 18 (58%) of 31 in the treatment group developed rheumatoid arthritis versus 15 (65%) of 23 in the placebo group (HR 0 & centerdot;75 [0 & centerdot;38-1 & centerdot;49], p=0 & centerdot;41).In the full trial cohort of 236 participants, physical disability in the treatment group improved durably over 5 years compared with the placebo group (mean difference in HAQ -0 & centerdot;06 [95% CI -0 & centerdot;14 to -0 & centerdot;05], p=0 & centerdot;048), though development of rheumatoid arthritis was not prevented, with 26 (22%) of 119 participants developing rheumatoid arthritis in the treatment group and 31 (27%) of 117 in the placebo group (HR 0 & centerdot;79 [95% CI 0 & centerdot;47 to 1 & centerdot;34], p=0 & centerdot;38).

Interpretation

Secondary prevention with a single intramuscular glucocorticoid injection and 1 year of methotrexate has different long-term effects on developing rheumatoid arthritis in ACPA-positive participants and in ACPA-negative participants. ACPA-negative participants had a long-term reduction in inflammatory disease burden and development of rheumatoid arthritis whereas there was no sustained improvement with ACPA-positive participants. These results imply that different treatment strategies are needed for ACPA-positive and ACPA-negative individuals with arthralgia who are at risk of rheumatoid arthritis.

(c) 2026 Elsevier Ltd. All rights reserved, including those for text and data mining, AI training, and similar technologies.

• All authors: Mulligen, E. van; Dumoulin, Q.A.; Steenbergen, H.W. van; Krijbolder, D.I.; Visser, A.W.; Böhringer, S.; Mil, A.H.M.V.
• Date: 2026-08-01
• Journal: The Lancet Rheumatology
• Volume: 8
• Issue: 8
• Pages: 638 - 649
• DOI: doi:10.1016/S2665-9913(26)00109-8