This was a randomized (2: 1), double-blind, placebo-controlled, event-driven trial enrolling subjects with IPF. Randomization was stratified by presence of pulmonary hypertension at baseline, determined by a right heart catheterization, and IPF diagnosis by surgical lung biopsy. The primary endpoint was time to IPF disease progression, defined as all-cause mortality, adjudicated respiratory hospitalization, or a categorical decrease in lung function (a 10% decrease in forced vital capacity [FVC] with a 5% decrease in the diffusion capacity for carbon monoxide [DLCO] or a 15% decrease in DLCO with a 5% decrease in FVC). At 75%(492 subjects) of the intended total enrollment and after a mean exposure of 34 weeks to the study drug, the Data Monitoring Board terminated the study due to low likelihood of demonstrating efficacy. From 136 clinical sites, 329 and 163 subjects were randomized to receive ambrisentan or placebo respectively. At baseline, 36 (11%) and 18 (11%) subjects had pulmonary hypertension (PH) in the ambrisentan and placebo groups respectively. Ambrisentan treated subjects had more primary events (90 [27.4%] versus 28 [17.2%], see Figure), and a 1.74 fold increase in risk of meeting the primary endpoint (95% confidence interval [CI] 1.14-2.66, p= 0.010). Evaluation of the primary endpoint components indicated that the number of deaths (hazard ratio [HR] 2.08, 95% CI 0.75-5.76, p= 0.100) and subjects with a categorical decrease in lung function (HR 1.53, 95% CI 0.84-2.78, p= 0.109) were not statistically significantly different between the groups. However, ambrisentan treated subjects had more respiratory hospitalizations (44 [13%] versus 9 [6%]), and a 2.59 fold increase in risk of experiencing a respiratory hospitalization (95% CI 1.14-5.89, p= 0.007). Stepwise Cox multivariate analysis revealed that after adjustment for baseline IPF severity, the risk for primary events was reduced and the p-value was> 0.1 (HR 1.42, 95% CI 0.85-2.05, p= 0.108). Although the risk for respiratory hospitalization was also reduced, the p-value remained< 0.05 (HR 2.11, 95% CI 1.03-4.33, p= 0.042). Presence of PH at baseline did not affect these point estimates significantly. The secondary endpoints at 48 weeks and the incidence of liver toxicity were not statistically significantly different between the groups. Ambrisentan was ineffective in treating IPF and was associated with an increased risk of respiratory hospitalizations. ARTEMIS-IPF: A PLACEBO-CONTROLLED TRIAL OF AMBRISENTAN IN IDIOPATHIC PULMONARY FIBROSIS