Abatacept Versus Hydroxychloroquine in Palindromic Rheumatism Trial
Key Takeaways
- Over 24 months, abatacept was associated with less progression to rheumatoid arthritis than hydroxychloroquine in this seropositive palindromic rheumatism cohort.
- Remission was more common and attack intensity was lower with abatacept, while attack frequency was similar between groups.
- Most autoantibody measures did not separate clearly overall, both treatments were well tolerated, and interpretation was limited by the open-label small-sample design without central adjudication.
The PALABA trial was a randomized, open-label, multicenter study conducted at 14 rheumatology centers in Spain. Eligible participants were adults aged 18 years or older with palindromic rheumatism lasting at least 3 months but less than 3 years, plus rheumatoid factor and/or anticitrullinated protein antibody positivity. Seventy-three participants were randomized, 37 to abatacept and 36 to hydroxychloroquine, but three assigned to abatacept never received medication, leaving 70 in the modified full analysis and safety population. The primary endpoint was persistent arthritis lasting more than 1 week in the same joint that met rheumatoid arthritis classification criteria during 24 months, as judged by participant clinicians.
In the modified full analysis set with failure imputation over 24 months, 7/34 (20.6%) in the abatacept group and 18/36 (50.0%) in the hydroxychloroquine group developed rheumatoid arthritis, with a risk difference of 29.4% (95% CI 8.2 to 50.7; P=0.010). A first-12-month analysis also favored abatacept, with progression in 3/34 (8.8%) versus 13/36 (36.1%) (P=0.007). The supportive time-to-event analysis also favored abatacept. Remission, defined as no or a single attack during at least four consecutive visits, occurred in 19/34 (55.9%) versus 8/36 (22.9%) during 24 months (P=0.007). Median attack intensity was 4.0 (3.0, 6.0) with abatacept and 7.0 (5.0, 8.0) with hydroxychloroquine.
Most antimodified peptide and/or protein antibody titers did not show clear overall between-group differences. Exploratory associations included lower chimeric fibrin/filaggrin homocitrullinated peptide-IgG at month 24, lower CEP1-IgA at month 3, and lower chimeric fibrin/filaggrin homocitrullinated peptide-IgA at month 24 in participants treated with abatacept. Both drugs were well tolerated, with at least one adverse event in 25/34 (73.5%) and 23/36 (63.9%); total cholesterol and lymphocytes significantly increased with abatacept and decreased with hydroxychloroquine. One serious adverse event with abatacept was melena requiring hospitalization and judged unrelated, one participant discontinued for nonserious toxicoderma, and two hydroxychloroquine participants had five serious adverse events. No deaths were reported. Reported limits included the open-label design, no central adjudication, small sample size, no drug-free follow-up, no imaging assessment, and no patient-reported outcomes, with 5-year follow-up underway.