Long-term tolerability of inhaled human insulin (Exubera) in patients with poorly controlled type 2 diabetes.

OBJECTIVE: Inhaled human insulin (Exubera; EXU) has shown encouraging tolerability in short-term trials. We evaluated the safety profile of EXU after long-term exposure. DESIGN: In two, open-label, 2-year studies patients poorly controlled on a sulphonylurea were randomised to adjunctive EXU or metformin (study 1) and patients poorly controlled on metformin were randomised to adjunctive EXU or the sulphonylurea, glibenclamide (study 2). PATIENTS: The studies included 446 (study 1) and 476 (study 2) patients with type 2 diabetes, no clinically significant respiratory disease and glycosylated haemoglobin (HbA(1c)) levels of 8-12%. MEASUREMENTS: Main outcome measures were pulmonary function tests and insulin antibody assays. RESULTS: A total of 109 patients (study 1) and 195 patients (study 2) completed 104 weeks treatment. In both studies, small treatment group differences in change from baseline forced expiratory volume in 1 s were greatest at 6 months (first time-point measured) and less at later visits, and reversed on treatment discontinuation. At 2 years, differences in mean changes were -0.10 and -0.01 l in studies 1 and 2, respectively, and -0.04 l for the pooled studies. There was no discernable effect of long-term EXU therapy on pulmonary gas exchange. Insulin antibody binding reached a plateau at 6 months and did not correlate with HbA(1c) or lung function changes. Glycaemic control was maintained over 2 years. CONCLUSIONS: Exubera was well tolerated during long-term use. Pulmonary function changes compared with comparator groups were small, non-progressive and reversed upon treatment discontinuation. Importantly, rates of lung function change were indistinguishable between EXU and comparator after 6 months of therapy.