Efficacy and safety of low-dose interleukin-2 in combination with methotrexate in patients with active rheumatoid arthritis: a randomized, double-blind, placebo-controlled phase 2 trial

Abstract

Rheumatoid arthritis (RA) is an aggressive autoimmune arthritis, and current therapies remain unsatisfactory due to low remission rate and substantially adverse effects. Low-dose interleukin-2 (Ld-IL2) is potentially a therapeutic approach to further improve the disease. This randomized, double-blind, placebo-controlled trial was undertaken to evaluate the efficacy and safety of Ld-IL2 in patients with active RA. Patients were randomly assigned (1:1) to receive Ld-IL2, defined as a dose of 1 million IU, or placebo in a 12-week trial with a 12-week follow-up. Three cycles of Ld-IL2 or placebo were administered subcutaneously every other day for 2 weeks (a total of 7 doses), followed by a 2-week break. All patients received a stable dose of methotrexate (MTX). The primary outcomes were the proportion of patients achieving the ACR20, DAS28-ESR <2.6, and the change from baseline in CDAI or SDAI at week 24. Secondary endpoints included other clinical responses and safety. The primary outcomes were achieved in the per-protocol population. The improvements from baseline in CDAI and SDAI were significantly greater across time points for the Ld-IL2 + MTX group (n = 17) than for the placebo+MTX group (n = 23) (P = 0.018 and P = 0.015, respectively). More patients achieved ACR20 response in the Ld-IL2 + MTX group than those in the placebo+MTX group at week 12 (70.6% vs 43.5%) and at week 24 (76.5% vs 56.5%) (P = 0.014). In addition, low Treg and high IL-21 were associated with good responses to Ld-IL2. Ld-IL-2 treatment was well-tolerated in this study. These results suggested that Ld-IL2 was effective and safe in RA. ClinicalTrials.gov number: NCT02467504.

Fig. 1

Consort flowchart of the study. Of the 47 randomized patients, 23 were exposed to Ld-IL2 (defined as a dose of 1 million IU) and 24 were exposed to placebo. All patients received methotrexate with Ld-IL2 or placebo. Ld-IL2 low-dose interleukin-2, MTX methotrexate

Fig. 2

Clinical responses to Ld-IL2 combined with MTX therapy. The data were presented in a per-protocol (PP) analysis set. The proportion of patients achieving an ACR20/50/70 response by at week 12 and 24 (a). The mean changes from baseline for CDAI and SDAI (b, c). The mean changes from baseline for pain, PtGA and PhGA of disease activity (d). Data in graphs were mean ± SE. *P < 0.05, with analyses with a logistic regression model or a covariance (ANCOVA) model at week 12 or 24. ^#P-value, presenting treatment differences across time points with a mixed model for repeated-measures analysis or Generalized Estimation Equations (GEE) method. ACR20/50/70 the American College of Rheumatology for 20%/50%/70% improvement, CDAI Clinical Disease Activity Index, SDAI Simplified Disease Activity Index, Ld-IL2 low-dose interleukin-2, MTX methotrexate, PtGA patient’s global assessment, PhGA physician’s global assessment

Fig. 3

Predictive biomarkers for potential response to Ld-IL2 treatment in RA. The percentage of regulatory T cells (Tregs) in CD4+ T cells and serum level of IL-21 in responders and non-responders (a, b). The correlation between percentage of Tregs and serum level of IL-2 (c). Heatmap showed a classification of the two groups (d). Each column represented an individual. Colors in the horizontal bar denoted the non-responder group (red) and the responder group (green). Tiles were colored based on clinical features, Tregs and serum cytokine levels, red and blue indicating high and low levels, respectively. Primary composition analysis (e). The non-responder group was shown in red, and the responder group was shown in green. PC1 and PC2 account for 39.8% and 14%, respectively, of the total variance. Panels d and e were performed by R 4.1.0 and R-packages (mixOmics and pheatmap) (http://www.metaboanalyst.ca). Data in graphs were mean ± SE. *P < 0.05

Fig. 4

Ld-IL2 therapy synergized with MTX to expand the population of Tregs and ameliorate inflammation of RA. The proportion of Tregs in CD4+ T cells. Grey areas indicated the periods on Ld-IL2 or placebo therapy (a). The changes of the proportion of Tregs or Th17 in Ld-IL2 combined with MTX group and MTX alone group (b, c). The ratio of Tregs/Th17 (d). The serum levels of IL-17A, IFN-γ, TNF-α, and IL-12 were decreased significantly at week 12 (e). Data in bar graphs were mean ± SE. Ld-IL2 low-dose interleukin-2, MTX methotrexate, Treg regulatory T cell. *P < 0.05