A multicenter, randomized, controlled, open-label trial to determine the optimal duration of steroid therapy for mild pneumonitis associated with immune checkpoint inhibitors.

Abstract

12006 Background: The optimal duration of corticosteroid therapy for pneumonitis associated with immune checkpoint inhibitors is clinically relevant. Several guidelines recommend a duration of 4 to 6 weeks for mild immune-related pneumonitis. However, evidence from clinical trials is limited. We conducted the first randomized trial to evaluate whether short-term corticosteroid therapy can achieve comparable efficacy. Methods: In this multicenter, open-label, randomized clinical trial at 22 institutions in Japan, we randomly assigned patients with mild immune-related pneumonitis according to the Common Terminology Criteria for Adverse Events grade 1 or 2, in a 1:1 ratio, to receive either 3-week or 6-week corticosteroid treatment. The primary endpoint was the rate of treatment success 8 weeks after the start of steroid administration, with a non-inferiority margin of 16 percentage points. The major secondary endpoints were safety, percentage of participants with treatment failure, quality of life, and overall survival. The primary hypothesis was that a 3-week treatment would be non-inferior to a 6-week treatment in terms of the primary endpoint. Results: Overall, 106 patients were randomized, and after the exclusion of one patient without immune-related pneumonitis, 105 were included in the intention-to-treat (ITT) population: 51 patients in the 3-week group and 54 in the 6-week group. In the ITT population, the patients’ median age was 72 years; 81% of the patients were men, and 73% had grade 2 at baseline. The rate of treatment success was 66.7% in the 3-week group and 85.2% in the 6-week group, which did not demonstrate noninferiority in the overall study population (difference, −18.5% percentage points [80% confidence interval {CI}, −29.0% to -7.9%], p = 0.621), and a predefined exploratory superiority analysis indicated superiority of the 6-week regimen (p = 0.013). Over the entire study period, the relapse or exacerbation rates of pneumonitis were 41.1% in the 3-week group and 24.1% in the 6-week group. Grade 3 or higher adverse events occurred in 12% of patients in the 3-week group and 24% of patients in the 6-week group. The absolute mean change in the total QOL using the K-BILD score from baseline was 4.78 in the 3-week group and 6.28 in the 6-week group (between-group difference, -1.50 points; 95% CI, −5.91 to 2.91). Conclusions: In patients with mild immune-related pneumonitis, non-inferiority of 3-week corticosteroid treatment compared to that of 6 weeks was not confirmed in the overall population, and the relapse or exacerbation rate of pneumonitis was higher in the 3-week group over the entire study period. Corticosteroid therapy shorter than the duration recommended by the guidelines is not supported. Clinical trial information: jRCTs051220082 .