Decline in forced vital capacity as a surrogate for mortality in patients with pulmonary fibrosis.

Abstract

Surrogate endpoints enable determination of meaningful treatment effects more efficiently than applying the endpoint of ultimate interest. We used data from trials of nintedanib in subjects with pulmonary fibrosis to assess decline in forced vital capacity (FVC) as a surrogate for mortality. Data from the nintedanib and placebo groups of trials in subjects with idiopathic pulmonary fibrosis, other forms of progressive pulmonary fibrosis, and pulmonary fibrosis due to systemic sclerosis (NCT00514683, NCT01335464, NCT01335477, NCT01979952, NCT02999178, NCT02597933) were pooled. Using joint models for longitudinal and time-to-event data, we assessed the association between decline in FVC % predicted and time to death over 52 weeks. The rate of change in FVC % predicted and the current value of FVC % predicted were modelled longitudinally and estimates applied as predictors in time-to-event models. Among 2583 subjects with pulmonary fibrosis, both a greater rate of decline in FVC % predicted and a lower current value of FVC % predicted were associated with an increased risk of death over 52 weeks (HR 1.79 [95% CI: 1.57, 2.03] and HR 1.24 [1.17, 1.32] per 5-percentage point decrease, respectively). Associations between the rate of change in FVC % predicted and the risk of death were consistent between patients with IPF and other ILDs. Data from clinical trials in subjects with pulmonary fibrosis of diverse aetiology demonstrate a strong association between decline in FVC % predicted and mortality over 52 weeks, supporting FVC decline as a surrogate for mortality in these patients.