Background: Previously we reported an emerging highly fatal, diffuse lung disease (DLD) in Still’s disease patients, associated with exposure to anakinra, canakinumab, tocilizumab or rilonacept. We discovered that patients met diagnostic criteria for definite or probable drug reaction with eosinophilia and systemic symptoms (DRESS) and that 80% carried DRB1*15 alleles (OR>15) with or without DLD. Here, we compared those stopping to those not stopping the implicated drug for survival probability, resolution of DLD, and treatment intensity (TI) required to subsequently manage Still’s.
Method: Retrospective multicenter Still’s-DRESS cases were collected during Still’s treatment with any interleukin 1 or 6 inhibitor (IL-1i/IL-6i). Resolution of DLD after drug withdrawal was defined as normal oxygen saturation without supplementation, absence of respiratory symptoms, and normal chest HRCT. TI was assessed by the number of immunosuppressive medications (ISM) required to manage Still’s.
Results: 74.4%(58/78) of Still’s-DRESS developed DLD; 48.3%(28/58) stopping IL-1i/IL-6i were followed ≥1 year after drug stop. To date, 61%(17/28) show lung disease resolution vs 0%(0/58) during IL-1i/IL-6i. Additionally, in Still’s cases withdrawing drugs, followed ≥1 year, regardless of DLD, 48.0%(36/75) decreased TI versus 52.0%(39/75) continuing IL-1i/IL-6i. At one year after drug stop, fewer ISMs were needed for disease control: ≤ 1 ISM in 69%(24/36) vs 2%(8/39), p=7x 10-5 stopping versus continuing IL-1i/IL-6i after initial DRESS features, respectively. Survival favored stopping vs continuing any IL-1i/IL-6i [HR=0.24(95% CI: 0.08,0.74)p=0.0127].
Conclusions: Stopping all IL-1i/IL-6i in this condition improves survival, may resolve DLD and simplify subsequent Still’s disease management implicating these drugs in pathophysiology. Specific mechanism remains unknown.
