American Thoracic Society International Conference
American Thoracic Society International Conference
May 29, 2019
4 min read

Combination therapy with treprostinil reduced risk for clinical worsening in PAH: FREEDOM EV

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

DALLAS — Adding treprostinil to monotherapy, as compared with placebo, reduced the risk for clinical worsening events in patients with pulmonary artery hypertension, especially after adjustment for baseline risk, new data from FREEDOM EV indicate.

The global, event-driven FREEDOM EV trial, which was presented at the American Thoracic Society International Conference by R. James White III, MD, PhD, from the University of Rochester Medical Center in New York, involved 690 patients, all of whom were required to enter the trial on only one oral PAH medication. Upon enrollment, patients were randomly assigned to oral treprostinil (Tyvaso, United Therapeutics Corp.) or placebo, with all receiving at least one dose of their assigned therapy.

The primary outcome, which was selected to assess the effect of treprostinil on time to first adjudicated clinical worsening event, included death, hospitalization due to worsening PAH, initiation of inhaled or infused prostacyclin, disease progression — defined as an at least 15% decrease in 6-minute walk distance and an increase in New York Heart Association (NYHA) functional class or worsening heart failure — or an unsatisfactory long-term clinical response.

The researchers also evaluated risk by calculating the number of low-risk criteria — NYHA functional class I or II, 6-minute walk distance greater than 440 m and N-terminal pro-B-type natriuretic peptide (NT-proBNP) less than 300 ng/L — met at baseline and follow-up.

Reduced risk for worsening PAH

The risk for clinical worsening events was significantly lower in the oral treprostinil vs. placebo groups (HR = 0.74; 95% CI, 0.56-0.97), according to White.

At each 12-week follow-up, the researchers analyzed NT-proBNP. The median value at baseline was slightly higher in the oral treprostinil group and slightly lower in the placebo group. However, the values declined quickly at week 12 and continued to decline at week 24 and week 36 such that the median values were below placebo by week 24 (P < .02), according to White.

In terms of NYHA functional class, at week 4, about 5% of patients in the placebo group had deteriorated, whereas 8% of patients in the oral treprostinil group had improved. By week 24, NYHA functional class was significantly improved with oral treprostinil vs. placebo (P < .02).

By week 12, there was some numeric separation between the oral treprostinil and placebo groups with regard to 6-minute walk distance, but it was not statistically significant, according to White. The same was true at week 24, but by week 36, the placebo group had started to deteriorate, whereas the oral treprostinil group maintained its gains. Further, by week 48, there was significant improvement with oral treprostinil vs. placebo (P < .01).


“I would point out that the mixed model repeated measurement did not require imputation, so imputation is not influencing the results at all,” White said.

Influence of risk

Notably, the placebo group was at significantly lower risk at baseline, according to White. Eleven percent of patients in the oral treprostinil group vs. 21% in the placebo group had all three low-risk factors and 25% of the oral treprostinil group vs. only 18% of the placebo group had no low-risk factors, he noted.

“The noninvasive risk strategy was what we anticipated using to evaluate improvement. What we did not expect to see was a remarkable, highly significant difference in the risk for mortality at the time of baseline randomization, meaning that our randomization strategy that focused on 6-minute walk distance and background therapy was insufficient in defining risk in this event-driven trial,” White said.

After accounting for the imbalance in risk at baseline, the HR for the primary outcome with oral treprostinil vs. placebo declined to 0.61 (95% CI, 0.46-0.81).

“Looking at it a bit differently, we rather arbitrarily assigned those with no or one low-risk criteria as high risk and those with two or three risk factors as low risk,” White said.

The researchers found that patients on treprostinil had a lower risk for the primary outcome than those on placebo in both high- (HR = 0.64; 95% CI, 0.46-0.88) and low-risk groups (HR = 0.64; 95% CI, 0.36-1.13), but the statistical significance was greater in the higher-risk group.

Patients on oral treprostinil also shifted to lower-risk criteria, as compared with placebo, from week 12 to week 60 (Fisher’s exact P < .002 for all).

“I would point out that this risk calculation uses observed cases only and doesn’t do any data imputation, and as was shown, the placebo patients were more likely to have worsening risk while the oral treprostinil patients had successive increases in improvements in their risk profile,” White said.

However, adverse events were likely, he noted, with 18% of the oral treprostinil group vs. 4% of the placebo group discontinuing treatment.

“We had a risk reduction that greatly increased once we adjusted for our initial imbalance in baseline risk. There was significant improvement in the prespecified, externally validated noninvasive risk score without imputation,” White said, adding that adverse effects were a problem and must be actively managed.


“Future studies should randomize participants based on a risk assessment with optimal discrimination,” White said.

Patient demographics

The study, White said, was truly a global one.

“About half of the patients came from Asia and Australia; there was a very nice contribution from our colleagues in Mexico and South America; and there was a reasonable contribution from North America and Europe,” he said.

At baseline, most patients were women with idiopathic or heritable PAH. Additionally, White noted that the population differed somewhat from those of previous studies. For instance, most were younger than 65 years, with a mean age of 45 years; 63% were NYHA functional class II; and patients’ median 6-minute walk distance was 405 m, with most patients having a 6-minute walk distance greater than 340 m.

The researchers also enrolled most patients within 6 months of their initial therapy for PAH.

“We had a group of patients with relatively early symptoms, good exercise tolerance and a recent diagnosis,” he said.

For the study, dosing proceeded over time without a scheduled protocol. Investigators were asked to balance adverse effects with the expected improvements in symptoms and signs of PAH. At week 12, White said, patients were receiving 2.5 mg three times daily and at week 24, they were receiving slightly more than 3.5 mg three times daily, which continued up until about week 60 before plateauing. – by Melissa Foster


White RJ. Abstract 5587. Presented at: American Thoracic Society International Conference; May 17-22, 2019; Dallas.

Disclosure: White reports he has served as steering committee leadership in drug development projects for United Therapeutics and PhaseBio; he is a consultant for Bayer, Celtaxsys, Gilead and Reata; and he is the chair of the data and safety monitoring board for a randomized, placebo-controlled trial of genetically modified, endothelial progenitor cells in Canada sponsored by Northern Therapeutics.