Treatment with an oral prostacyclin analog improved endobronchial dysplasia in former smokers in a randomized pilot study of lung cancer chemoprevention, presented by Robert Keith, MD, Denver VA Medical Center and the University of Colorado, Denver, CO.

All measures of dysplasia (airway damage) improved in patients who were treated with iloprost, whereas the same parameters worsened in patients who were randomized to placebo. Consistent with results from previous studies, only former smokers benefited from treatment, suggesting that smoking cessation is a necessary component of chemoprevention.

Clinical and preclinical evidence has suggested that regulation of prostaglandin production (PGI₂ and PGE₂) plays a major role in the risk, evolution, and progression of lung cancer. For example, manipulation of prostacyclin synthase (PGI) has been shown to prevent tumor formation in mouse models of lung cancer (Keith et al. Cancer Res 2002; Keith et al. Cancer Res 2004).

To test the chemopreventive potential of iloprost, investigators enrolled 152 current and former smokers with a smoking history of at least 20 pack years, no history of cancer, and sputum cytologic atypia. Study participants were randomized to placebo or to oral iloprost (not available in the United States) at a starting dose of 50 μg BID, increasing to a maximum dose of 150 μg BID.

Participants underwent bronchoscopy at baseline and after 6 months of treatment. The primary endpoint was the change from baseline in total dysplasia score at 6 months, based on bronchial biopsy samples that were obtained from six prespecified sites. The dysplasia score was derived from the 8-point World Health Organization classification system for preneoplasia (ranging from normal=1 to invasive cancer=8). Investigators employed three separate histological measures: average histology over all biopsies, worst histology, and dysplasia index, determined by the percentage of biopsies with dysplasia score ≥4 (moderate dysplasia).

The study population comprised 81 current smokers and 71 former smokers. At the 6-month evaluation, all three histological parameters favored iloprost in former smokers. On average, the worst-histology score in the iloprost group decreased from 4.25 at baseline to 3.85 at 6 months, whereas worst histology increased from 3.91 to 4.14 in the placebo group (p=0.038). In addition, twice as many participants in the iloprost group had at least a 1-point improvement in worst histology (40% vs 20%; p=0.014).

Each of the three measures of the primary outcome declined in former smokers who were randomized to iloprost but increased among former smokers in the placebo group. Between-group differences for all three parameters achieved statistical significance (p=0.010 to p=0.002). In contrast, none of the outcome measures differed significantly between current smokers who were treated with iloprost or placebo.

Investigators have begun planning another chemoprevention trial that will be limited to former smokers and will have lung cancer incidence as the primary outcome.