Brass, Eric P.

Background: In peripheral artery disease (PAD), claudication limits exercise performance, functional status and quality of life. Phosphodiesterase III inhibitors improve claudication-limited exercise performance in patients with PAD. K-134, a novel phosphodiesterase III inhibitor, was evaluated in a phase II trial incorporating an adaptive design to assess safety, quality of life and peak walking time (PWT) on a graded treadmill with cilostazol as an active comparator. Design: Patients with PAD were randomized to receive placebo (N = 87), K-134 at a dose of 25 mg (N = 42), 50 mg (N = 85) or 100 mg (N = 84), or cilostazol at a dose of 100 mg (N = 89), each given twice daily for 26 weeks. A Data Safety Monitoring Board (DSMB) implemented adaptive design was used to allow early discontinuation of unsafe or minimally informative K-134 arms. Interim safety assessments at 2 weeks included dosing of study medication at the study site followed by measurements of postural vital signs and symptoms, and at 4 weeks with a treadmill test at peak drug concentration to evaluate for ischemia or arrhythmias. Results: Based on prospectively defined adaptive criteria, the 25 mg K-134 arm was discontinued as the 50 mg and 100 mg doses were adequately tolerated, had no unanticipated safety signals, and were thus most likely to be informative concerning K-134 in the target population. In the intention-to-treat cohorts at 26 weeks PWT increased by 37% in the 100 mg K-134 arm and 23% on placebo (primary analysis p = 0.089). Using a mixed effects analysis in the entire population, 100 mg K-134 and cilostazol were each superior to placebo in improving PWT. In the per protocol population both K-134 100 mg and cilostazol 100 mg increased PWT as compared with placebo at 14 and 26 weeks. K-134 had tolerability and adverse effect profiles similar to that of cilostazol, with gastrointestinal adverse events, headache and dizziness the most common. Both drugs were associated with an increase in withdrawals prior to study completion due to adverse events as compared with placebo. Conclusions: K-134 was generally well tolerated, and at a dose of 100 mg twice daily had a safety profile and effect on treadmill walking performance that was similar to that of cilostazol.

Cilostazol, a phosphodiesterase III inhibitor, is indicated to treat the symptoms of claudication in patients with peripheral arterial disease (PAD). At the time of approval the FDA required an additional long-term safety study to evaluate cilostazol’s effect on mortality. Subjects (n=1899) with a clinical diagnosis of PAD and symptoms of claudication were screened for a randomized, double blinded, placebo-controlled safety study of cilostazol. The modified intent-to-treat (mITT) population (n=1435) included all randomized patients who received at least one dose of study medication and includes follow up of more than 30 days after discontinuation of study drug. A total of 717 patients received cilostazol and 718 received placebo. Cilostazol was administered at a primary dose of 100 mg twice daily. Long-term adherence to study medication was poor with over 60% of participants discontinuing therapy by 36 months. Therefore the mortality analysis focused on deaths during the period on-treatment defined as the period during which study drug was taken plus a 30-day post-dosing follow up period. Total patient-years exposure on-treatment for cilostazol was 1046 and on placebo was 1090. On-treatment there were 18 deaths on cilostazol and 19 deaths on placebo for a hazard ratio of 0.99 (95% confidence interval 0.52 to 1.88). Cardiovascular deaths on-treatment occurred in 14 patients on cilostazol and 14 on placebo. In the full mITT population at 36 months, there were 101 deaths, 49 on cilostazol and 52 on placebo with hazard ratio of 0.94 (0.64 –1.39). Thus most deaths occurred more than 30 days after study drug discontinuation. Serious bleeding events affected 18 patients on cilostazol and 22 on placebo in the on-treatment population. Bleeding events were similar in patients also taking aspirin, aspirin plus clopidogrel or anticoagulants over the course of the study. This long-term study demonstrated no safety signal for cilostazol on all cause or cardiovascular mortality. However the study was under-powered to detect a small adverse impact of cilostazol on mortality (hazard ratio upper bound of the 95% confidence interval 1.88 in the on-treatment population). Serious bleeding events appeared not to be increased by cilostazol.