1Division of Pulmonary, Allergy, and Rheumatology Products, Office of New Drugs, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, Maryland; and2Division of Pharmacometrics, Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, Maryland
Checking for direct PDF access through Ovid
Rationale:Exploration of FVC as it relates to mortality in idiopathic pulmonary fibrosis (IPF), a chronic, progressive, and ultimately fatal parenchymal lung disease, is important both clinically and to the current drug development paradigm. We evaluated the association between FVC decline and mortality in what is to our knowledge the largest well-characterized placebo cohort to date. Additionally, we sought to explore the risk of death caused by acute exacerbations and to further validate previously identified baseline predictors of mortality.Objectives:To validate and further characterize FVC decline, acute exacerbations, and previously identified baseline predictors as they relate to risk of death.Methods:A total of 1,132 placebo subjects from six studies used for the clinical development of nintedanib and pirfenidone for the treatment of IPF were included in the present analysis. Deaths were captured as all-cause mortality. A stratified Cox proportional hazards model was used to test the association between baseline predictors, decline in FVC % predicted from baseline, acute exacerbations, and death. Decline in FVC % predicted and exacerbations were treated as time-varying covariates.Results:Subjects were followed for a mean of 60 weeks. At baseline, age, smoking status, lower FVC % predicted, and lower diffusing capacity of the lung for carbon monoxide % predicted were associated with an increased risk of death. The risk of death was also increased for subjects having one or more exacerbations with a hazard ratio (HR) of 10.3 (95% confidence interval [CI], 5.7-18.7). Compared with an FVC % predicted absolute decline from baseline at any time during the study of less than 5%, a decline greater than or equal to 10% to less than 15% was associated with an increased risk of death, with an HR of 2.2 (95% CI, 1.1-4.4), as was a decline greater than or equal to 15%, which was estimated with an HR of 6.1 (95% CI, 3.1-11.8). A decline ranging from greater than or equal to 5% to less than 10% was not associated with increased mortality.Conclusions:Our analyses validate the importance of baseline FVC, diffusing capacity of the lung for carbon monoxide, age, and smoking status as predictors of mortality and strengthen the association between decline in FVC and exacerbations with death, verifying a decline in FVC as an appropriate endpoint in IPF drug development.Clinical trial registered with www.clinicaltrials.gov (NCT00514683, NCT01335464, NCT01335477, NCT00287729, NCT00287716, and NCT01366209).